Now that there's a FVP for aarch64, we don't need to pin the CI pipeline
to x86-64.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
To allow running the TF-A TFTF tests we need to ensure the images for
N1SDP and Corstone-1000 MPS3 boards build
Signed-off-by: Drew Reed <Drew.Reed@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Just execute all tests with the meta-arm tag, instead of hardcoding the
list of tests.
Also run two tests in parallel as there's no reason to limit it to one.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The Kas binary is identical, but the container has been rebuilt using
Debian 12 (Bookworm).
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The Kas container in version 4 onwards is based on Debian 12, which
forbids pip from installing files into /usr or ~/.local/.
We want to install the arbitrary dependencies for the documentation
build, so these should be installed in a venv.
The kas container doesn't currently install python3-venv, so we have to
install that manually (patch sent upstream).
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The fvp-baser-aemv8r64 machine will not be actively maintained.
Signed-off-by: Divin Raj <divin.raj@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The update-repos script currently exits immediately if one of the
underlying Git commands fails (e.g. because of a network issue). If the
repo already exists, then catch this error inside the loop and
carrying on attempting to update other repos, as the network error may
be upstream.
KAS_REPO_REF_DIR is ultimately an optimization and subsequent build
stages should be able to continue if one of the updates fail. Therefore,
ensure the script returns a special error code if at least of the Git
commands fail, and use this to set the allow_failure property of the
job.
If a repo does not exist, fail immediately as before.
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
To make the pipeline slightly more resilient to external networking
issues, allow a local container registry mirror to be specified in the
GitLab settings. If not specified, the upstream container registry is
used automatically.
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Build a custom N1SDP image with only optee-xtests as part of the image. A fresh custom image build is necessary to include the relevant test suite for running the xtests.
Signed-off-by: Anusmita Dutta Mazumder <anusmita.duttamazumder@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
In the update-repos job, generate a Kas lock file to pin the SHAs of the
repositories being build during the build. This ensures that commits to
the repositories _during_ the build doesn't cause differing builds. All
of the kas calls use this lock file to ensure that their builds are
identical.
This lockfile is also added to the artifacts, so that it can be reused
afterwards to replicate the build: either as a known good base or to
replicate failures.
This lock file is only generated if it doesn't exist, so that
development branches can temporarily add a lockfile.yml if for example
master is too unstable to develop on.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Kas 3.3 has the following new features:
- kas: Introduce commit and branch as alternative to refspec key
- kas: Warn if a repo uses legacy refspec
- kas: add support for lock files via dump plugin
- kas: track root repo dir config files of menu plugin
- kas: add support for --log-level argument
- kas: add GIT_CREDENTIAL_USEHTTPPATH environment variable
- kas: improve error reporting
- kas: drop support for Python 3.5
- kas-container: fix invocations with --isar for some layers
- kas-container: Purge tmp* on clean
- kas-container: enable colored logging
4.0 is basically the same but the Kas container uses Debian 12, which is
not yet supported as a tested platform. Until Debian 12 is tested we
should continue to use 3.3.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
It's useful to have known good logs when debugging problems, so always
preserve the logs.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
This recipe is a rebuild from source of the Arm GCC release, with
patches from oe-core added to make it work well in our environment.
Most people are happy with the GCC in oe-core, and this release is often
behind: at the time of writing oe-core has mainline GCC 13.1, but Arm
GCC is 12.2. Users who actually want the improvements in Arm GCC will
likely want to use the binary toolchain so that they can have support
from Arm, and they're welcome to do so via the "external-arm" binary
toolchain.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Acked-by: Sumit Garg <sumit.garg@linaro.org>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Split trusted-services.xml into qemuarm64-secureboot-ts.yml and
n1sdp-ts.yml as collection of Trusted Services which can be tested on
each platform has diverged.
Signed-off-by: Adam Johnston <adam.johnston@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Add musl testing to have qemuarm-secureboot match qemuarm64-secureboot.
Since the Arm GCC binary toolchain cannot work with musl, move that out
into it's own testing.
Signed-off-by: Jon Mason <jon.mason@arm.com>
Add tc1 ecosystem FVP and bits to enable in the tc1 machine config file
Also, do some hacks to speed things up.
Signed-off-by: Jon Mason <jon.mason@arm.com>
The dev kernel can frequently fail, and is not anything that is used in
production. Allow failure to prevent CI issues but still notify that
there are potential issues.
Signed-off-by: Jon Mason <jon.mason@arm.com>
If the repository reference directory gets corrupted it's not easy to
wipe it, so add a variable CI_CLEAN_REPOS that if set in the pipeline
will clean the clones and re-fetch them.
Also, stop the fetch from detaching during the garbage collection, just
in case it was a long-running GC that got killed that caused the
corruption in the first place.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The infrastructure for edk2 and fvp-base is already present, but not
being used. Make the changes to get it compiling cleanly, and add it to
CI.
Note: testing is not passing because edk2 isn't booting an image
Signed-off-by: Jon Mason <jon.mason@arm.com>
With the removal of fvp-base-arm32, we no longer have test coverage for
the external Arm toolchain. Add this to qemuarm-secureboot CI so that
there is coverage again. Note: it must be a 32bit machine, since there
are currently no aarch64 host toolchains for aarch64
Signed-off-by: Jon Mason <jon.mason@arm.com>
fvp-base-arm32 isn't a real machine and supporting it has become hacky.
Drop support and remove from meta-arm-bsp
Signed-off-by: Jon Mason <jon.mason@arm.com>
Add the various kernels available in oe-core, as well as the poky-tiny
minimal distribution (which has a minimal kernel config). This
necessitated combining some kernel bbappends to have patching coverage
for all the variants.
Signed-off-by: Jon Mason <jon.mason@arm.com>
Make things more obvious by adding yml files for the poky defaults
instead of disregarding them in the jobs-to-kas script
Signed-off-by: Jon Mason <jon.mason@arm.com>
Mirrors of meta-arm may focus their development on a small subset of
MACHINEs so provide the option to restrict the boards that are built on
CI using the variable BUILD_ENABLE_REGEX. If set, it conditionally
enables builds; if unset there is no change in behavior.
This variable could be overridden in a scheduled build, to e.g. build
all the MACHINEs weekly.
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Mirrors of meta-arm may have the persistent cache directory mounted in a
different place. To make it easier to configure, define this location
using a single $CACHE_DIR variable.
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
For some reason the kas 3.2.1 container fails:
No such file or directory: '/builds/engineering/yocto/meta-arm/ci/ci/base.yml'
Note the repeated /ci/, which is wrong.
Pin the kas container to 3.2 for now until this is resolved.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Kas 3.2 ships python3-subunit, so we don't need to try to install that
anymore.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Kas 3.2 has a 'dump' plugin, so use that instead of cat in a shell.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
This job doesn't use the standard helpers, so needs to pass these
explicitly otherwise it can pick up an old SHA.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Now that the FVP is available for both aarch64 and x86-64, don't set a
tag so this can run on both architectures.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The meta-gem5 layer is unmaintained and gem5 is incompatible with Python
3.11, so won't work with master without work that is still ongoing.
Signed-off-by: Ross Burton <ross.burton@arm.com>
The meta-atp layer is unmaintained and gem5 is incompatible with Python
3.11, so won't work with master without work that is still ongoing.
Signed-off-by: Ross Burton <ross.burton@arm.com>
DEFAULT_TAG and CPU_REQUEST are being used to help with internal Gitlab
pipeline setups know which type of machines to run on, but has no value
outside of Arm Corp. Gitlab CI allows for variables to be overridden
by default. So, we can give it a default value of NULL/empty and have
everything work internally and externally by default.
Signed-off-by: Jon Mason <jon.mason@arm.com>
Tag all jobs with the DEFAULT_TAG variable so each instance can control
what tags the jobs have, whilst still explicitly tagging the jobs which
need specific tags (such as x86_64 for jobs which need to run x86-only
binaries)
Signed-off-by: Ross Burton <ross.burton@arm.com>
The Kas container needs to use the entrypoint as that is where the user
changes from root to a normal user.
Also set the KUBERNETES_CPU_REQUEST to the variable CPU_REQUEST as this
needs to be tuned per-deployment.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Add support for n1sdp to trusted-services bbappends and rework some
things to make it easier to add more in the future.
Signed-off-by: Jon Mason <jon.mason@arm.com>
The Total Compute 2020 BSP is obsolete and unsupported, so remove it
from meta-arm. The Total Compute team would like TC1 to be available in
langdale, but removed in mickledore.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Remove microbit-v1, qemu-cortex-a53, qemu-cortex-m3, and qemu-cortex-r5
from CI (and the tree in general). These machines are part of the
meta-zephyr CI now and keeping them here is redundant. However, keeping
zephyr builds for machines that also have TF-M.
Signed-off-by: Jon Mason <jon.mason@arm.com>
Ross Burton
Drew Reed
Divin Raj
Peter Hoyes
Anusmita Dutta Mazumder
Jon Mason
Adam Johnston
Adrián Herrera Arcila
Anton Antonov