* add meta-qcom in repo-defaults, since it's used for the first time
* add meta-qcom in check-layer-nightly
Running yocto-check-layer right now, shows:
INFO: Getting initial bitbake variables ...
INFO: Getting initial signatures ...
INFO: Adding layer meta-qcom
INFO: Starting to analyze: meta-qcom
INFO: ----------------------------------------------------------------------
INFO: test_bsp_defines_machines (bsp.BSPCheckLayer)
INFO: ... ok
INFO: test_bsp_no_set_machine (bsp.BSPCheckLayer)
INFO: ... ok
INFO: test_machine_signatures (bsp.BSPCheckLayer)
INFO: ... skipped 'No machines set with --machines.'
INFO: No machines set with --machines.
INFO: test_machine_world (bsp.BSPCheckLayer)
INFO: ... skipped 'No machines set with --machines.'
INFO: No machines set with --machines.
INFO: test_layerseries_compat (common.CommonCheckLayer)
INFO: ... ok
INFO: test_parse (common.CommonCheckLayer)
INFO: ... ok
INFO: test_patches_upstream_status (common.CommonCheckLayer)
INFO: ... ok
INFO: test_readme (common.CommonCheckLayer)
INFO: ... ok
INFO: test_security (common.CommonCheckLayer)
INFO: ... ok
INFO: test_show_environment (common.CommonCheckLayer)
INFO: ... ok
INFO: test_signatures (common.CommonCheckLayer)
INFO: ... ok
INFO: test_world (common.CommonCheckLayer)
INFO: ... ok
INFO: test_world_inherit_class (common.CommonCheckLayer)
INFO: ... ok
INFO: skipped "DistroCheckLayer: Layer meta-qcom isn't Distro one."
INFO: ----------------------------------------------------------------------
INFO: Ran 13 tests in 167.729s
INFO: OK
INFO: (skipped=3)
INFO:
INFO: Summary of results:
INFO:
INFO: meta-qcom ... PASS
Signed-off-by: Nicolas Dechesne <nicolas.dechesne@oss.qualcomm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Since meta-openembdedded commit bf400cd73fad ("reproducibility: move
known non-repro list to layer.conf")[0],
OEQA_REPRODUCIBLE_EXCLUDED_PACKAGES is directly defined in each
layer.conf. Remove the redundant definition and simplify the code a
little.
[0]: https://git.openembedded.org/meta-openembedded/commit/?h=master&id=bf400cd73fad89aff85017dfc4c169e7a39fdc4a
Signed-off-by: Yoann Congal <yoann.congal@smile.fr>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There is a bug which has the Target Milestone of 5.0.15 and appears in
the Old Milestone list despite the latest 5.0.x release being 5.0.10.
This happened because we only considered the next three point releases,
so extend the list of future point releases to ten to ensure that we
include all realistic versions.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If the git clone operations don't succeed the mirror_processor thread
will spin updating nothing and index builds will fail. Instead the
thread will attempt the clones as many times as needed logging each
failure.
These are bare mirrors so look for the refs dir instead of .git.
Signed-off-by: Michael Halstead <mhalstead@linuxfoundation.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Drop core-image-sato from the x86-64-x32 tune test suite and falls back
to core-image-full-cmdline for verification. This change is supported by
previous issues reported in openembedded-core related to graphics and
multimedia, and a dependency issue caused by lack of support of this ABI
in languages like Go [1].
[1] https://lists.openembedded.org/g/openembedded-core/message/216111
Signed-off-by: Randolph Sapp <rs@ti.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We have a 6.0 target in bugilla now, so consider that active.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Since we're not actively testing mips, we should drop this specific
multilib test too.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Switch the autobuilder to use the config fragements in OE-Core, maintained
there so people can easily see and test with them if needed.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This is identical to the musl-qemux86 builds, but for qemuarm64.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Testing builds often run on branches <someone>/master[-next]. In these
cases we are really testing "master", so return that. Instead of
exiting with: "Release <future release> does not exist".
Signed-off-by: Antonin Godard <antonin.godard@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some autobuilder workers don't have IPv4 addresses and were failing some selftest
with problems connecting to downloads.yoctoproject.org.
Using a buildtools tarball with IPv6 support fixes this.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We considered the next release to be active, but failed to include the
next release's milestones.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
g++ on Ubuntu 20.04 is now too old so switch to using buildtools there.
This also lets us switch to a minimum of python 3.9.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Very useful for eg Old Milstone where you can then sort it by owner.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This is a a little script I have that should be somewhere public. It
simply lists all of the connected workers, but is very useful when used
in conjunction with parallel-ssh to run commands on every worker at once.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Somehow, meta-initramfs was missing from the
commit 91c85d95bb ("config.json: move repro excludes from AB config.json to meta-oe")
Signed-off-by: Yoann Congal <yoann.congal@smile.fr>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We need buildtools (and the poky repo) for resulttool in some indexing steps. We don't
want buildtools in the dashboard step as we need semver from the host.
Rework the config to allow this and fix failures on the infrastructure.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
METADATA_BRANCH is the only way for us to check whether we're on master.
Check this variable first, and return the value of
LAYERSERIES_COMPAT_core if we're not on master.
Signed-off-by: Antonin Godard <antonin.godard@bootlin.com>
Using METADATA_BRANCH is less accurate than getting the current release
from LAYERSERIES_COMPAT_core, which should always be equal to the
release we want to compare for.
The code the exit if the current release was not found is left there
just in case.
Signed-off-by: Antonin Godard <antonin.godard@bootlin.com>
Add a replacement for the the existing Bug Triage page on the wiki[1].
This uses client-side JavaScript to parse the releases.json and query
Bugzilla, so shouldn't need updating very often.
[1] https://wiki.yoctoproject.org/wiki/Bug_Triage
Signed-off-by: Ross Burton <ross.burton@arm.com>
Add a stable_release boolean argument to _compare(). If True just
ignore missing workers on the AB, because we want to keep the full list
of tested releases in poky.conf for stable branches.
Signed-off-by: Antonin Godard <antonin.godard@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
