- Mar 12, 2025
-
-
Update all version indicators to 1.5.0. Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
Fix for reading LHS scale values from in kai_matmul_clamp_f32_qsi8d32p1vlx4_qsi4c32p4vlx4_1vlx4vl_sme2_mopa Fix the out-of-bounds read while loading the scale values from LHS packed matrix in \`kai_matmul_clamp_f32_qsi8d32p1vlx4_qsi4c32p4vlx4_1vlx4vl_sme2_mopa\` by updating the predicate Resolves: KLEIDIAI-507 Signed-off-by:
Anitha Raj <anitha.raj@arm.com> Reviewed-by:
Viet-Hoa Do <viet-hoa.do@arm.com> Reviewed-by:
Emil Ohlsson <emil.ohlsson@arm.com> Approved-by:
Emil Ohlsson <emil.ohlsson@arm.com>
-
- Mar 11, 2025
-
-
Build system robustness improved by several methods: * Mark standard 'build' folder as ignored. This helps when doing different builds from a same folder * Combine source files for assembler kernels in same targets * Add sorting for new kernel lists * Relax clean step in CI for faster builds Signed-off-by:
Anton Bondarenko <anton.bondarenko@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
Jens Elofsson authored
Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Emil Ohlsson <emil.ohlsson@arm.com>
-
- Mar 07, 2025
-
-
Anton Bondarenko authored
Analyzing skip test w/o a proper report message is hard. Providing more details helps with that. Signed-off-by:
Anton Bondarenko <anton.bondarenko@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
Emil Ohlsson authored
A couple of cleanups were done while adding support for QAI8 GEMV, these have been moved out to this patch * Sorts file lists in `CMakeLists.txt` * Add additional test shapes * Minor readability tweaks Signed-off-by:
Emil Ohlsson <emil.ohlsson@arm.com> Approved-by:
Anton Bondarenko <anton.bondarenko@arm.com>
-
- Mar 05, 2025
-
-
Jens Elofsson authored
This flag have been removed from CMakeLists.txt, but accidentally left in kai_defs.bzl. Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Anton Bondarenko <anton.bondarenko@arm.com>
-
- Feb 27, 2025
-
-
Jens Elofsson authored
Change type of rhs_zero_point to uint8_t to match the data type in the kai_rhs_pack_qs4cxs1s0_param-struct. Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Anton Bondarenko <anton.bondarenko@arm.com>
-
- Feb 26, 2025
-
-
Jens Elofsson authored
The argument to std::mt19937:s constructor is uint32_t, but the supplied value (the variable "seed") was uint64_t. This has been changed to uint32_t. Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Anton Bondarenko <anton.bondarenko@arm.com>
-
- Feb 24, 2025
-
-
* Refactor the benchmark tool to create a generic abstraction that allows for running matrix multiplication micro-kernels with different interfaces. * Extend benchmark support to all matrix multiplication micro-kernels in the library. Signed-off-by:
Jakub Sujak <jakub.sujak@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
This flag is a stylistic option in GCC and does not add to security hardening. Signed-off-by:
Jakub Sujak <jakub.sujak@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
- Feb 20, 2025
-
-
- Add new assembly ukernel optimized with FEAT_I8MM for matrix multiplication with 4x8 block size. - Update build script. - Add to unit test. Signed-off-by:
Michael Kozlov <michael.kozlov@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
- Feb 18, 2025
-
-
Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
Jens Elofsson authored
Update all version indicators to 1.4.0. Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Reviewed-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
Felix Johnny Thomasmathibalan authored
lhs stride is removed from kai_run_matmul_clamp_qai8_qai8_qsi8cxp2vlx4sb_1x16vl_sme2_dot Signed-off-by:
Felix Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
- Feb 17, 2025
-
-
- Add new assembly ukernel optimized with FEAT_DOTPROD for matrix multiplication with 1x8 block size. - Update build script. - Add to unit test. Signed-off-by:
Michael Kozlov <michael.kozlov@arm.com> Reviewed-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
Add support for GEMV like kernel for producing QAI8 from QAI8 LHS and QSI8CXP packed RHS. Update unit tests to include support for new kernel Signed-off-by:
Emil Ohlsson <emil.ohlsson@arm.com> Reviewed-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com> Reviewed-by:
Anton Bondarenko <anton.bondarenko@arm.com> Reviewed-by:
Emil Ohlsson <emil.ohlsson@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
When adding negative tests a potential crash from positive one got masked. Enforce exit from a test script for positive test crash. Signed-off-by:
Anton Bondarenko <anton.bondarenko@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
Add negative test for CPU features to make sure actual binaries using non-supported instructions would crash with illegal instruction signal. Signed-off-by:
Anton Bondarenko <anton.bondarenko@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
- Feb 13, 2025
-
-
Testing with bare minimum of CPU features allows to verify that optional features are properly guarded with feature checks. Signed-off-by:
Anton Bondarenko <anton.bondarenko@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
- Feb 12, 2025
-
-
Anton Bondarenko authored
Number of parallel jobs suitable for current CI runner configuration not always could be retrieved by standard system utilities. The value need to be hardcoded to avoid unexpected memory usage and could be aligned with current CI resources and configuration. Signed-off-by:
Anton Bondarenko <anton.bondarenko@arm.com> Reviewed-by:
Emil Ohlsson <emil.ohlsson@arm.com> Approved-by:
Emil Ohlsson <emil.ohlsson@arm.com>
-
- Update build script - Add to unit test Signed-off-by:
Michael Kozlov <michael.kozlov@arm.com> Reviewed-by:
Anton Bondarenko <anton.bondarenko@arm.com> Approved-by:
Anton Bondarenko <anton.bondarenko@arm.com>
-
- Feb 11, 2025
-
-
To test a potential ukernel use case and to test offset helper functions. Resolves: #COMPMID-7887 Signed-off-by:
Michael Kozlov <michael.kozlov@arm.com> Reviewed-by:
Anitha Raj <anitha.raj@arm.com> Reviewed-by:
Anton Bondarenko <anton.bondarenko@arm.com> Approved-by:
Anton Bondarenko <anton.bondarenko@arm.com>
-
Resolves: #KLEIDIAI-485, #COMPMID-8024 Signed-off-by:
Anitha Raj <anitha.raj@arm.com> Reviewed-by:
Anitha Raj <anitha.raj@arm.com> Approved-by:
Anton Bondarenko <anton.bondarenko@arm.com>
-
- Feb 07, 2025
-
-
Anton Bondarenko authored
Multi-platform images simplifies usage as there is no need to use dedicated tag for different platform. But a special care need to be taken when building the image. Current version expect native, or emulated, support from builder. Signed-off-by:
Anton Bondarenko <anton.bondarenko@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
- Feb 06, 2025
-
-
Signed-off-by:
Gian Marco Iodice <gianmarco.iodice@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
Jakub Sujak authored
Communicate build requirements to targets linking to KleidiAI by adding the absolute path rather than relative path in the header search path. Signed-off-by:
Jakub Sujak <jakub.sujak@arm.com> Approved-by:
Emil Ohlsson <emil.ohlsson@arm.com>
-
Jens Elofsson authored
- Remove designated initializers in matmul_clamp_f32_qai8dxp_qsi4c32p_test - Fix number format in testcase instantiation - Enable matmul_clamp_f32_qai8dxp_qsi4c32p_test for all platforms Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
- Update build script - Add to unit test Signed-off-by:
Michael Kozlov <michael.kozlov@arm.com> Approved-by:
Anitha Raj <anitha.raj@arm.com>
-
- Feb 04, 2025
-
-
Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
- Jan 31, 2025
-
-
Suhail M authored
- r27C is necessary for MTE Builds Signed-off-by:
Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> Approved-by:
Anton Bondarenko <anton.bondarenko@arm.com>
-
- Added demonstration of klediai integration as external dependency Signed-off-by:
Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> Reviewed-by:
Mohammed Suhail Munshi <mohammedsuhail.munshi@arm.com> Reviewed-by:
Emil Ohlsson <emil.ohlsson@arm.com> Approved-by:
Emil Ohlsson <emil.ohlsson@arm.com>
-
- Jan 30, 2025
-
-
Signed-off-by:
Anton Bondarenko <anton.bondarenko@arm.com> Approved-by:
Felix Johnny Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
-
Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
- Jan 29, 2025
-
-
Update all version indications Signed-off-by:
Emil Ohlsson <emil.ohlsson@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
- Jan 28, 2025
-
-
Remove fp32 and fp16 scale datatypes from - kai_rhs_pack_kxn_qsi4c32p_qsu4c32s1s0 - kai_rhs_pack_nxk_qsi4c32p_qsu4c32s1s0 since they aren't being used and just creates more maintenance work. Signed-off-by:
Jens Elofsson <jens.elofsson@arm.com> Approved-by:
Anitha Raj <anitha.raj@arm.com>
-
Anton Bondarenko authored
Version 6.5 does not support conlyopts for cc_library and C only options need to be handled in a different way Signed-off-by:
Anton Bondarenko <anton.bondarenko@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
Signed-off-by:
Michael Kozlov <michael.kozlov@arm.com> Signed-off-by:
Anitha Raj <anitha.raj@arm.com> Reviewed-by:
Viet-Hoa Do <viet-hoa.do@arm.com> Reviewed-by:
Jakub Sujak <jakub.sujak@arm.com> Approved-by:
Jakub Sujak <jakub.sujak@arm.com>
-
Jakub Sujak authored
All licenses in the LICENSES directory apply equally to the distribution of KleidiAI. Signed-off-by:
Jakub Sujak <jakub.sujak@arm.com> Approved-by:
Emil Ohlsson <emil.ohlsson@arm.com>
-