The primary motivation for having a sandbox without LTO build in CI is
to ensure that we don't have that option break. We now have the ability
to run tests of specific options being enabled/disabled, so drop the
parts of CI that build and test that configuration specifically and add
a build test instead. We still test that "NO_LTO=1" rather than editing
the config file works via the ftrace tests.
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Use the most recent upstream release of OpenSBI for CI testing.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
At this point we have all of the defconfigs maintained again, so
re-enable the check to prevent further regressions.
Signed-off-by: Tom Rini <trini@konsulko.com>
This add some basic functions to create images, and a test for said
functions. This is not intended to be a test of the image parsing
functions, but rather a framework for creating minimal images for testing
load methods. That said, it does do an OK job at finding bugs in the image
parsing directly.
Since we have two methods for loading/parsing FIT images, add LOAD_FIT_FULL
as a separate CI run.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Now that sandbox64 can run and pass the regular test.py suite, add it
here as well.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Currently, most sandbox runs take a long time (due to running so many
tests) while QEMu based test.py runs are fairly short. Split the
pipeline here so that we get more consistent average run times.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Both to aide in debugging of any test.py issues as well as to make it
easier to split the current matrix in two, have a new job that creates
and publishes the current wrapper script we use for test.py jobs.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
- We have added more TODO/etc comments since this task was created and
never focused on removing them.
- The output of sloccount isn't preserved or looked at, and if desired
should be in the release stats pages instead somehow.
- The results of cppcheck aren't investigated and require modeling work
to be useful to start with.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
These jobs are to confirm specific build targets, on a Linux host. We
can safely combine these two build tests, with a make mrproper in
between.
Signed-off-by: Tom Rini <trini@konsulko.com>
Now that we have 3600 minutes per build job, condense and rework things
such that our overall time largely doesn't change, but we can also
largely avoid having to re-tweak this job to avoid timeouts. Given that
we have 10 threads, we also move a few of the specific sandbox test
builds to a prior stage.
Note that while sandbox builds with address sanitization enabled (ASAN)
not all tests pass, so we limit ourselves to just checking that the
version test passes for now.
Link: https://learn.microsoft.com/en-us/azure/devops/pipelines/process/phases?view=azure-devops&tabs=yaml#timeouts
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
This platform is behind on migrations (it is the sole user of the oldest
legacy version of the USB gadget stack and is long overdue for
migration) and with Pali no longer being a maintainer, we remove this
platform.
Signed-off-by: Tom Rini <trini@konsulko.com>
When running the trace test on the sandbox platform, the current size
of 16MiB is no longer large enough for capturing the entire trace
history, and results in truncation. Use a size of 32MiB for the trace
buffer on the sandbox platform while running the trace test.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
As per current Azure Pipelines documentation we qualify for 3600 minutes
per job, if specified, as the timeout. The default unspecified timeout
is 60 minutes. Rework things to specify 0 as the timeout (and so maximum
allowed) so that we don't have failures due to running slightly past 60
minutes total.
Link: https://learn.microsoft.com/en-us/azure/devops/pipelines/process/phases?view=azure-devops&tabs=yaml#timeouts
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The latest kernel.org toolchains for gcc are now 13.2.0, so upgrade to
that.
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
When running tools for various tests use the tools-only build rather
than sandbox_spl. We used sandbox_spl here for historical reasons that
are no longer true.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Use a recent coreboot build for this test.
The coreboot commit is:
6f5ead14b4 mb/google/nissa/var/joxer: Update eMMC DLL settings
This is build with default settings, i.e. QEMU x86 i440fx/piix4
Add some documentation as to how to update it next time.
Signed-off-by: Simon Glass <sjg@chromium.org>
To reduce overall job time, move a number of smaller jobs together.
These should still be safely under 1 hour total time, but reducing the
overall number of jobs should help with the queue slightly.
Signed-off-by: Tom Rini <trini@konsulko.com>
The job for rockchip vendor platforms has again gotten close to or
exceeded one hour. Rework things such that we move the 32bit platforms
back to the general 32bit ARM job (as there's time there) and make these
build only the 64bit platforms.
Signed-off-by: Tom Rini <trini@konsulko.com>
Rather than using the -R option to get this report as a side effect, add
a dedicated option for it.
Disable CI for now as there are some missing maintainers, unfortunately.
Signed-off-by: Simon Glass <sjg@chromium.org>
Now that buildman has a requirements.txt file we need to make use of it.
Reviewed-by: Simon Glass <sjg@chromium.org>
[n-francis@ti.com: Adding missing command from .azure-pipelines.yml]
Signed-off-by: Neha Malcom Francis <n-francis@ti.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
It is not uncommon for some of the QEMU-based jobs to fail not because
of a code issue but rather because of a timing issue or similar problem
that is out of our control. Make use of the keywords that Azure and
GitLab provide so that we will automatically re-run these when they fail
2 times. If they fail that often it is likely we have found a real issue
to investigate.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The job to build all imx8 and imx9 platforms is currently close to, or
sometimes exceeding the allowed build time. Exclude some platforms that
are already being built under their vendor-specific job as well to
reduce the time.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Currently the 64bit "rk" job is close to and sometimes goes over the job
time limit. Let us rework this in to one job for "rk" and "rv" (which
are the SoC prefixes) jobs which include or exclude "rockchip" the board
vendor. This gives us two jobs of similar numbers of platforms to build
now instead.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
As this is the current version of the public cross toolchains we use,
upgrade to this now.
Suggested-by: Alexey Brodkin <Alexey.Brodkin@synopsys.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
Acked-by: Alexey Brodkin <abrodkin@synopsys.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Use the latest OpenSBI v1.2 release binaries for the RISC-V CI.
Signed-off-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Currently the PowerPC build job in Azure will hit the maximum time limit
for a build and stop. Looking at the job, the easiest path to reducing
it is to move Keymile vendor boards to their own job and exclude them
from the PowerPC one (and while at this, the ls102 job).
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
We do not want to merge documentation that produces Sphinx warnings.
scripts/kernel-doc uses environment variable KDOC_WERROR to determine
if warnings should be treated as errors.
Reported-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Add M5208EVBE board to CI. This does not use default config due to
limitations of QEMU emulation, instead the timer is switched from
DMA timer to PIT timer and RAMBAR accesses are inhibited.
Local QEMU launch command is as follows:
$ qemu-system-m68k -nographic -machine mcf5208evb -cpu m5208 -bios u-boot.bin
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Acked-by: Angelo Dureghello <angelo@kernel-space.org>
As this is now the stable release, move to using that now for our tests.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The default pytest cache directory is in a read-only directory in Azure,
which results in a warning on the build page. Use the pytest command
line option to set the cache dir to somewhere writable.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Enable use of the python-azurepipelines package which provides automatic
formatting and uploading of the pytest output.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Now that the Dockerfile creates images which have the binaries we
require included, have CI make symlinks for them and update the existing
script to support this.
Signed-off-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
These should not be used anymore. Add a check to ensure they don't creek
back into U-Boot. Use bootph-... instead.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add a test which runs sandbox, collects a trace and makes sure it can
be processed by trace-cmd. This should ensure that this feature continues
to work as U-Boot and trace-cmd evolve.
Signed-off-by: Simon Glass <sjg@chromium.org>
Now that all remaining in-tree cases where we define or undef a CONFIG
symbol have been migrated to Kconfig or renamed to CFG we can make the
CI check more robust. We will exclude the doc, tools and arch/arm/dts
directories from this check as they are special cases. Further, we can
exclude the scripts/kconfig/lkc.h and include/linux/kconfig.h files as
the CONFIG values they define are special tooling cases and not real
symbols.
In the case of docs, the only places that currently fail this test are
old documentation that should be rewritten so that we can remove this
special case.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Add infodocs target to CI testing.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
Update to the latest "jammy" tag. This requires us to list libc6-i386 as
a required package to install (for nokia_rx51 tests) that was previously
implicit.
Signed-off-by: Tom Rini <trini@konsulko.com>
The rockchip job is getting close to the hard time limit in Azure for
the free tier. Split this in to 32bit and 64bit board jobs.
Signed-off-by: Tom Rini <trini@konsulko.com>
Now that all symbols have been migrated to Kconfig, or are part of the
CFG namespace we do not need a complex check for unmigrated CONFIG
symbols. Any instance of #define (or #undef) or a CONFIG value is wrong,
so cause CI to fail.
This test is not as strict as possible yet as we have more symbols that
were not previously caught to deal with.
Signed-off-by: Tom Rini <trini@konsulko.com>
