| Age | Commit message (Collapse) | Author |
|---|
|
We don't use the CD/DVD ROM drive in any of our tests, but it
periodically fails discovery and that leads to a QA error:
[ 6.403477] ata3.00: failed to IDENTIFY (I/O error, err_mask=0x4)
The only way to disable the optical ROM drive in qemu is to use
the '-nodefaults' option, which disables the CDROM (among other things).
We can't be sure that none of our tests, or extended users are relying
on default devices, so using that option is more of a risk than adding
the message to our ignore list.
To date, no one has sent a patch to just disable the optical drive
(either in qemu or the BIOS), but that is something we could consider
in the future.
[YOCTO #14528]
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 143fc5504539c69752ca87717507c197a8920ce5)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
The test uses the broken out time and can only handle about 59s of delay,
use a UNIX timestamp to allow for up to a 300s delay.
[YOCTO #14463]
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit b705e9373acd4119da75af4eb96ec92cc964aa86)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
Some tests such as lttng-tools are marginal and timing out on the autobuilder
with the current 300s default. Increase to avoid this noise in the ptest
failures list.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 5fb902a52e35130af6b0735a087c709daa35655f)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
All of the errors being masked off for qemuarm are legacy from before
the migration of qemuarm to qemuarmv5. Rename the machine to that to
allow for qemuarmv5 to pass parselog test. Light testing shows no
errors in dmesg for qemuarm.
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 701a58504de15b244b970908f2de0971a35b5a09)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
PACKAGE_CLASES should be PACKAGE_CLASSES.
Signed-off-by: Ralph Siemsen <ralph.siemsen@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 7a96f3bceb2c857f841b1dbeb4587a8aaace529e)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
Having two possible failures in multiprocesslauch creates a race where one failure
may occur and stop processes being lanuched meaning the second failure may not
be seen. Rather than having periodic races appearing on the autobuilder, only
have one failure, making the test much more deterministic.
[YOCTO #13054]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 31e9dcda40aae3ce0801580c838928956e1455e3)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
If we tweak sstate to not remove empty directories under conditions
where a race could occur, we see failures from:
"oe-selftest -r archiver.Archiver.test_archiver_filters_by_type archiver.Archiver.test_archiver_filters_by_type_and_name"
since an empty directory is left behind. Update the tests to ignore
empty directories.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 10cda713faea9a348fd278137ac75e4a6d76a71c)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
Load on the autobuilder meant we see occasionaly timeout issues with these tests.
Slightly increase the test timeouts to better reflect the real world timings we
see.
[YOCTO #14262]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit fccd2ade0e345625ed9a4b74a7431b000ce2214f)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
choice
The temp file name may contain "_" characters. Switch to a temporary directory
and a fixed filename to avoid this to avoid errors like:
bb.data_smart.ExpansionError: Failure expanding variable PN, expression was
${@bb.parse.vars_from_file(d.getVar('FILE', False),d)[0] or 'defaultpkgname'}
which triggered exception ParseError:
ParseError in /tmp/tmpd_f2__to.bb: Unable to generate default variables from
filename (too many underscores)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 086e2ae7b2b7496b4f3ae01436b4049d7f2ff8c4)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
This test is flawed since multiple parts of the system can write to the log
and we obtain different numbers of log messages depending on factors we
can't control.
Drop the log testing component of the test.
[YOCTO #12465]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
When running a shutdown command, the serial port can close without the
command returning. This is seen as the socket being readable but having
no data. Change the way this case is handled in the code to avoid
tracebacks.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 396a3ba884820d040c91f7592daf20ac28c49b5d)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
The recent logging changes for qemurunner showed up as errors on the
autobuilder where decode couldn't be called on the returned string.
Since the code returns binary data, return b'' instead of '' to match
to avoid tracebacks.
One of these cases was newly added, copied from the other which has
been there for a long time, always broken.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit b8995b27db265b0a0b2d2ca595915f70f9f96e07)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
Rather than totally disabling the logging, inform it we're about to exit
so we can log messages over the exit cleanly too. This aids debugging. It
also avoids a race where the logging handler could still error whilst
shutting down.
Also remove a race window by notificing the handler of the shutdown
first, before triggering it. This removes a race window I watched in
local testing.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 0e19f31a1005f94105e1cef252abfffcef2aafad)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
LD_LIBRARY_PATH leaks into host executables too, and breaks them
as they are not uninative-enabled. E.g. on ubuntu 18.04 trying
to run host bash with a sysroot that was built on Fedora 33:
akanavin@ubuntu1804-ty-3:/home/pokybuild/yocto-worker/oe-selftest-ubuntu/build/build-st-24341/tmp/work/x86_64-linux/gnupg-native/2.3.1-r0/recipe-sysroot-native$ LD_LIBRARY_PATH=./usr/lib /bin/bash
/bin/bash: ./usr/lib/libtinfo.so.5: no version information available (required by /bin/bash)
/bin/bash: /lib/x86_64-linux-gnu/libc.so.6: version `GLIBC_2.33' not found (required by ./usr/lib/libtinfo.so.5)
This was seen e.g. here:
https://autobuilder.yoctoproject.org/typhoon/#/builders/87/builds/2090/steps/14/logs/stdio
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 0e9850486b74a3de934527ca1077df001d3a8d22)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
This is the sequence that didn't properly operate:
- a test case that skips and isn't executed
- a second test case that is skipped via a dependency decorator, and sets a timeout
- a third test case that takes longer than the timeout from the second
test case
Without the fix, the timeout is not cleared, and the third test case is
erroneously aborted. With the fix, the timeout is cleared and the third
test case is able to complete.
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 54ef07a9aa1af8f41cfb9a4802929c918efc43c8)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
Some of the decorators need proper cleanup, such as OETimeout
which sets a signal handler that needs to be cleared via teardown.
If this is not done then the signal gets called later with unpredictable effects.
This can be seen if there's a test that is skipped via a decorator and sets a timeout
at the same time: the timeout isn't cleared, and is invoked later in a
completely unrelated context. The test case for this is added in the
next commit.
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit f42a08e1aabf1ca57e0c09d69fb69cc717c7f156)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
Messages are currently being printed as:
Test requires dropbear, oropenssh-sshd to be installed
but should be
Test requires dropbear, or openssh-sshd to be installed
Adding the space after the 'or' corrects this.
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 51596e0f8cebe1607ab64ffb018d51e815c0ee4b)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
We can disable floppy drive by BIOS on a hardware, but an empty floppy
drive is connected by default on qemu-system-x86. Linux usually detect
the device and modprode the matched floppy.ko at the boot stage. Due to
we don't specify a floppy deivce in qemu boot arguments, then the errors
about floppy reading comes out.
It is harmless and normal, so we could ignore this error message on
qemux86.
Seen if kernel-modules is included in the image which pulls in the
relavent kernel module.
https://lists.gnu.org/archive/html/qemu-devel/2021-04/msg01402.html
Signed-off-by: Yanfei Xu <yanfei.xu@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 3359f23ee9351c70997d5e0a17d17d1e47d59623)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
Previously doing a stop/start worked, but using a disable/enable does
not work on a read-only rootfs. Add a --runtime flag to systemctl so
that systemd only modifies the current boot and does not attempt to
write to the filesystem.
This also keeps the test from making a permanent (one could argue
policy) change to the running system being tested. i.e. What if the
image being tested had intentionally disabled the timesyncd service in
preference to using chrony or ntpd? The test shouldn't assume that the
user wants the timesyncd service enabled.
Signed-off-by: Wes Lindauer <wesley.lindauer@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 43dd83b6a325589368c980a3f17cab90935aaeb0)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
The dnf packages aren't parsed if rpm isn't in PACKAGE_CLASSES which means
the aintainers test failes for OE-Core (where ipk is the default) but not
for poky (where the default is rpm).
Ensure PACKAGE_CLASSES is set so it works in all cases.
[YOCTO #14277]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 842b11107363357ed933cfcf619f1cf23f0d841e)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
|
|
Setting a value of 10 for heartbeat events causes the test to fail. Hardcode
a value to ensure it works correctly even if the default is changed.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add tests to verify that SRC_URI dependency sniffing works correctly.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* The kernel-fitimage class adds a do_assemble_fitimage_initramfs task
regardless of INITRAMFS_IMAGE_BUNDLE setting, which in some cases can
result in that task running after do_uboot_assemble_fitimage and
overwriting the u-boot-spl.dtb file with the pristine version (without
public key). Fix this by making do_uboot_assemble_fitimage dependant
on both do_assemble_fitimage_* tasks, regardless of the aforementioned
setting.
* Adjust 'type' and 'os' on the U-boot fitimage its script so that
mkimage/dumpimage can recognize them.
* Move the deployment of the u-boot-spl-nodtb files outside of
concat_spl_dtb_helper(), so that we can better isolate the scenarios
of creating an (unsigned) U-Boot fitimage versus also signing it. This
prevents some stale files from being deployed in the images directory.
* Remove any u-boot-fitImage and u-boot-its files from build tree, in
case the build tree is being reused across bitbake calls.
Signed-off-by: Klaus Heinrich Kiwi <klaus@linux.vnet.ibm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Derived from the similar kernel fitImage sign testcase, the U-Boot
fitImage testcases exercises the following fitimage.FitImageTest
scenarios:
* test_uboot_fit_image - create unsigned U-Boot fitImage
* test_uboot_sign_fit_image - create unsigned U-Boot fitImage in
addition to signed Kernel fitImage
* test_sign_standalone_uboot_fit_image - Create signed U-Boot fitImage
without a Kernel fitImage
* test_sign_cascaded_uboot_fit_image - Create and sign U-Boot and
Kernel fitImages
Signed-off-by: Klaus Heinrich Kiwi <klaus@linux.vnet.ibm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We have a working theory that IO queues on the autobuilder are impacting
runtime testing under qemu, particularly async writes which inice does not
influence. We already pass the snapshot option to qemu which copies the
image and runs out of the copy. Add in the ability to copy the image to
a specificed location which can be a tmpfs. This means that writes to the
image would no longer be blocked by other writes to disk in the system.
Preliminary tests show that this does improve the qemu errors at the expense
of sometimes showing qemu startup timeouts as on a loaded system with a large
test image, it can take longer than 120s to copy the image to tmpfs. Having
a most consistent failure mode for loaded tests is probably desireable though.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If oe-selftest is run with -j, the output to stdout/stderr is being
lost at present. Capture this and display it upon test failure. We
have code that previously tried to enable this but it wasn't functioning
correctly. This should give more usable error reports on the autobuilder.
This code will mix stdout and stderr as the output is streamed from the test
server without markup. This is most in keeping with subunit/testools though
and the easiest way to handle the various challenges here as far as I can
see.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Each time I look at this code I get confused about what the different
variables represent. Rename a few of them to better indicate what they
represent.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The current test code contains a subtle race. It is building an ext4,
then trying to reuse that image in a wic build for the same recipe.
Whilst this happened to work, the code is within its rights to clean
up the output before the wic code runs.
Avoid this by using separate image targets instead.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
CVE_VERSION_SUFFIX in "patch" to treat version string with suffix "pX"
or "patchX" as patched release.
also update testcases to cover this changes and set CVE_VERSION_SUFFIX
for sudo.
Signed-off-by: Lee Chee Yang <chee.yang.lee@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The hardcoded path issue was fixed, drop the exclusion.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The meson determinism issue is the same frozenset issue we encountered with
python itself. Remove the problematic pyc file until upstream work out the
best way to address the issue (as with core python).
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This test is to keep track of recipes that do not have a HOMEPAGE or DESCRIPTION,
with exceptions to recipes whose SRC_URI are not linked to external sources.
Results of this test: it take about 2 to 3 minutes for this test to complete execution
depending on the laptop used, Recipes that didn't have a homepage or description were found,
and using list data structure instead of set() returned an ordered list of recipes that
didn't have a homepage. Hence it's easier to view recipes with missing homepage and description.
[YOCTO #13471]
Signed-off-by: Dorinda Bassey <dorindabassey@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
go-dep was an effort for dependency management before go modules, which
since 2020 has been deprecated in favor of go modules. Since its not
developed any longer and go mdules is officially supported, this should
be retired from OE-core as well.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Cc: Otavio Salvador <otavio.salvador@ossystems.com.br>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We have fixes for perf reproducibility, so we can drop it from the
exclusion list.
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Some platforms do not use ttyS* for their serial consoles (e.g., qemuarm
and qemuarm64). The hardcoding of this can cause issues. Modify
runqemu to use the serial consoles defined in SERIAL_CONSOLES instead of
hardcoding.
Signed-off-by: Jon Mason <jon.mason@arm.com>
Change-Id: I746d56de5669c955c5e29d3ded70c0a4d3171f17
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
pseries machine emulation ends up with these three messages in kernel
logs which are harmless, so ignore them for now
https://www.mail-archive.com/ubuntu-bugs@lists.ubuntu.com/msg5600561.html
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1827162
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add a recipe that explicitly searches /usr/include, and use that in
oe-selftest to verify that host include paths are correctly causing
build failures.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Under systemd weston's socket is in /run, not in /run/user/0.
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There are build paths which are in generated sources injected
into ${PN}-src. These are in generated files and don't influence the
binary output. As such we can correct them at do_package time by
tweaking in an appropriately injected function. This fixes
the reproducbility of the recipe.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add a patch adding sorting to a couple of points in the Makefiles
which removes most of the determinism issues in ltp.
Build swapon before the main build to ensure libswapon.o is built
deterministically as it races with swapoff.
All issues reported on the upstream mailing list.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This change adds core-image-weston to the reproducible build test.
For this we also need to update the static-group and static-passwd.
Besides the world package build which includes the wayland package,
this adds a wayland/weston-based image to the reproducible build test.
Signed-off-by: Jan-Simon Moeller <jsmoeller@linuxfoundation.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add a hack to hardcode in specific rpaths which we then remove,
allowing the build to be reproducible.
Strip build patches out of one of the test scripts too.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add a patch to address two determinism issues and allow reproducible
builds.
Also strip full paths we don't need out of some ptest files,
we can use the installed binaries.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Refresh xxx-config->pkg-config patches.
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
When debugging why a single recipe doesn't reproduce, its a pain
to wait for the world to rebuild from scratch. Update the selftest
to allow this to be configured, for example you could set
targets as ['perf'] and sstate_targets as ['virtual/kernel']
and then it should only be rebuilding perf in the test rather than
things like the toolchain (parts of the kernel may be unavoiable
as they're not in sstate).
Can be run as:
OEQA_DEBUGGING_SAVED_OUTPUT=/tmp/perf-diffoscope oe-selftest -r reproducible.ReproducibleTests.test_reproducible_builds
to save diffoscope output.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
LTO likely doesn't buy us much here, disable it to allow the binaries
to be reproducible.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add sorting to the globbing within the Makefile to make the output
reproducible.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Remove hardcoding the build configuration into the help/version output
from swig to make the binaries reproducible.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add sorting to wildcard expansion in the makefile to make builds
reproducible.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The library hardcodes paths to the python library internally and currently
these are build paths. Fix this to use the correct target path and
fix reproducibility in the process.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
Bruce Ashfield
Alexandre Belloni
Richard Purdie
Jon Mason
Ralph Siemsen
Alexander Kanavin
Yanfei Xu
Wes Lindauer
Ross Burton
Klaus Heinrich Kiwi
Lee Chee Yang
Dorinda Bassey
Khem Raj
Jan-Simon Moeller