Isaac Perry authored Feb 14, 2024 and Diego Sueiro committed Mar 18, 2024

Kronos pipelines with meta-arm sibling pipelines were formerly not
checking meta-arm success, resulting in success shown in the FVP report
when the Kronos pipeline passed and meta-arm failed. This commit adds a
check of the meta-arm pipeline (in the case that one exists). The row
in the FVP report will now correctly show as red in the case that
either Kronos or meta-arm failed. There is a race condition introduced
between the meta-arm pipeline and the report being rendered. If the
meta-arm pipeline has not completed by the time the FVP report is
rendered, the row for the pipeline will be gray.

Older builds will never have the parent project ID variable to pass to
the pipeline so will ignore meta-arm's success in the report, however
these builds will be removed from Artifactory and the report in future
as per the retention policy and this change will apply to all future
builds.

Signed-off-by: Isaac Perry <isaac.perry@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Debbie Martin authored Mar 18, 2024 and Diego Sueiro committed Mar 18, 2024

Include a notice at the top of the FVP report that it might be out
of date. This is because GitLab badges are only updated upon a
successful pipeline, so it won't show results for any pipelines since
the latest successful one.

Signed-off-by: Debbie Martin <Debbie.Martin@arm.com>

Debbie Martin authored Mar 18, 2024 and Diego Sueiro committed Mar 18, 2024

The virtualization test suite inherits from meta-arm and was
overriding the run command utility to run and check commands on DomU.
This resulted in a transient error in reading back the echo of the
command since this is sensitive to timing. The utility also had to
filter out messages from Xen hypervisor which are observed on the
telnet session.

To fix both issues, use the Linux terminal utilities, which SSH to
Dom0 and mean that the Xen hypervisor messages are not seen. This
also increases code commonality.

Signed-off-by: Debbie Martin <Debbie.Martin@arm.com>

Bump critical-application-monitoring to the latest version
available that includes fixing UUID url in documentation.

Signed-off-by: Rahul Singh <rahul.singh@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

1- Add note to remove any non-relevant configuration options.
2- Rephrase text of Xen `unhandled word write` messages.
3- Remove unnecessary text in HIPC reproduce subsection.
4- Mention the safety island cluster on running ping command.
5- Update libcam version in the logs to v1.0.
6- Remove using `send esc` for Domu2 in Critical Application
   Monitoring Virtualization use case, because DomU2 is logged
   into using ssh which does not need to enter `send esc` after
   `Ctrl-]`.
7- Add generic known issue for automated validation timeouts.
8- Inform the user that the `ssl_client1` results will appear
   after a few seconds.
9- Fix invalid TF-M links.

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

When the virtualization architecture is used, the DomU1 and DomU2
clocks need to be synchronized with the Dom0 clock before running
CAM, currently there is no check for this condition.

Add a check in the OEQA test that ensure the clock is synchronized
and update the documentation to add this step before running CAM.

Add a missing word to the dictionary to please the spellchecker.

Signed-off-by: Luca Fancellu <luca.fancellu@arm.com>

Increasing the timeout for exit_guest_to_dom0 test due to occasional
timeout on CI

Signed-off-by: Divin Raj <divin.raj@arm.com>

Signed-off-by: Diego Sueiro <diego.sueiro@arm.com>

Update the CAM SHA to include:
 * an implementation of a retry mechanism for the accept() call
 * a fix for a semaphore destruction on an error path
 * documentation improvements

Signed-off-by: Luca Fancellu <luca.fancellu@arm.com>

Add the encoding issue to the list of known issues.

Signed-off-by: Divin Raj <divin.raj@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

All the known issues have been resolved and not reproducible
anymore, therefore remove the list of known issues.

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

The FVP build number has been updated to 11.25.15, Update
meta-arm commit to match the latest.

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

In Secure Firmware Update, if the user runs the demo and tries
to run the validation tests or vice versa, they fail because
the Firmware has already been updated.

Add instructions to first build a clean flash before running the
demo or the validation tests.

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Diego Sueiro <diego.sueiro@arm.com>

Signed-off-by: Diego Sueiro <diego.sueiro@arm.com>

The Cassini v1.1.0 release is formally done from the nanbield branch
instead of nanbield-dev. Hence, set its repo to the nanbield default
setting.

Signed-off-by: Diego Sueiro <diego.sueiro@arm.com>

Debbie Martin authored Mar 05, 2024 and Debbie Martin committed Mar 06, 2024

Add a section to the Critical Application Monitoring section
of the design documentation which details the virtualization
architecture running CAM on multiple VMs.

Signed-off-by: Debbie Martin <Debbie.Martin@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Step 7 in fault management, says "Inject the same fault into the
System FMU again", but the next instruction doesn't inject the
same fault as the previous. It should say "Inject another
Lockstep error into the System FMU".

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

The Linux boot time in the PSA Crypto APIs use case takes
a considerable amount of time because the FVP is busy with the
RSE Crypto operations, add a note to mention that it is not
mandatory to wait for the Primary Compute to fully boot.

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

In "Primary Compute PSA Protected Storage and Crypto APIs
Architecture Test Suite" use case, extend the text before the
test suite results so it gives the user an indication that this
log would appear once the tests have completed.

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>

The screenshots for actuation use case of baremetal and
virtualization are outdated, Replace them with up to
date screenshots.

Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>