A basic Fedora 42 container does not have 'awk' installed, but it is
needed by Buildroot. First by check-host-python3.sh:
support/dependencies/check-host-python3.sh: line 6: awk: command not found
support/dependencies/check-host-python3.sh: line 19: awk: command not found
support/dependencies/check-host-python3.sh: line 19: awk: command not found
but then even building host-expat assumes awk is available:
config.status: creating Makefile
./config.status: line 1404: awk: command not found
config.status: creating expat.pc
./config.status: line 1404: awk: command not found
Since it's a pretty basic tool, make it part of the tools checked by
dependencies.sh. One minor annoyance is that check-host-python3.sh is
executed *before* dependencies.sh does its thing, so when 'awk' is not
available, we end up seeing:
support/dependencies/check-host-python3.sh: line 6: awk: command not found
support/dependencies/check-host-python3.sh: line 19: awk: command not found
support/dependencies/check-host-python3.sh: line 19: awk: command not found
which: no awk in (/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin)
You must install 'awk' on your build machine
make: *** [support/dependencies/dependencies.mk:27: dependencies] Error 1
It would be nice to have the awk check *before* it gets used in
check-host-python3.sh, but that's a topic for another patch.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 448ceefa78)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Annoyingly, using "--disable warning" does not disable the warnings
checks.
It turns out that we look for "warnings" (i.e. with an 's') to know if
we should disable the warnings check, so update the help text
accordingly.
Signed-off-by: Raphaël Mélotte <raphael.melotte@mind.be>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 92e7ab78d6)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Add a runtime test for the 'dust' package to verify that the binary
executes correctly in a minimal buildroot rootfs. The test checks that:
- 'dust --version' runs without error
- 'dust' can analyze a directory structure with files
- The output includes the expected directory names
Signed-off-by: El Mehdi YOUNES <elmehdi.younes@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 5bca9d741d)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Add a runtime test for the 'bat' package to verify that the binary executes
correctly in a minimal Buildroot rootfs.The test cheks that:
- 'bat --version' runs without error
- 'bat' can read and display a text file
- the displayed content matches the expected string
Signed-off-by: El Mehdi YOUNES <elmehdi.younes@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit dacf8e3c39)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Importing urllib3 already allows us to check that for example zlib can
be loaded at runtime.
For good measure, also create the PoolManager object mentioned in the
user guide ([1]), and check that we can normalize a URL like it is
done in urllib3's 'test/test_util.py'.
[1]: https://urllib3.readthedocs.io/en/stable/user-guide.html
Signed-off-by: Raphaël Mélotte <raphael.melotte@mind.be>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 3e931caf84)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Commit [1] "support/testing: improve weston test reliability" moved
out the wait time from the emulator (to run on the test controller).
While doing so, the sleep time which was initially _after_ the
"killall weston" invocation to in stop_weston() was incorrectly
moved before the command invocation. In this state, the test can
succeed on fast host computer running the test. But it will most
likely fail on an average computer.
This commit fixes this issue by moving the sleep time after
the command invocation.
[1] 6561a5d773
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit bedc44c073)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Some EDK II configurations have complex dependencies on several packages
and additional build options; build tests help keeping track of those
more easily.
Factorize some code common to all the build tests into a new
TestEdk2BuildBase class, which defines a base configuration and a method
to assert that binaries do indeed exist after the build.
While at it, add myself in DEVELOPERS.
Signed-off-by: Vincent Stehlé <vincent.stehle@arm.com>
Cc: Dick Olsson <hi@senzilla.io>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 4f7e3f0bdd)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
The manual states that FOO_DL_OPTS are valid for all the different
download backends, but that is not the case: at least the git backend
does not use them (it does not fail, it just ignores them).
Accept FOO_DL_OPTS in the git backend, and pass them to 'git fetch'.
There is no way that we can pass such options to submodules or lfs,
though.
Update the manual accordingly.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit a1012b363a)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
The ARMV7-A toolchains are capable to compile binaries for ARMv8-A CPU
in AArch32 execution state.
This adds the BR2_ARM_CPU_ARMV8A option in the 'conditions' to allow
ARMV8-A CPU such as Cortex-A53 or Cortex-A72 to use ARMV7-A toolchains.
Signed-off-by: Gaël PORTAY <gael.portay+rtone@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 8fd537ae05)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Running pkg-stats is currently quite verbose, as it shows one line per
package when checking for the upstream URL, and another one line per
package when checking for the latest version on
release-monitoring.org.
This noisy output is a bit annoying when pkg-stats is run in a
cronjob, like we do to update https://autobuild.buildroot.net/stats/
every day. This commit adds a -v/--verbose option, off by default, to
have a less noisy output.
Suggested-by: Peter Korsgaard <peter@korsgaard.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 203e9def71)
Signed-off-by: Arnout Vandecappelle <arnout@rnout.be>
Use the 'C' locale when retrieving the date of the last change using
'svn info' since the svn download helper script expect
"Last Changed Date" string.
If another locale is used, the 'date' is empty so the generated
archive (by mk_tar_gz) will not match the expected hash since
the file timestamp is not set properly.
If LANG=fr_FR.UTF-8 is defined in the host system, svn print some
"French encrypted" text:
eval svn --non-interactive --config-option servers:global:http-timeout=10 info ''\''https://svn.code.sf.net/p/xmlrpc-c/code/advanced@r3176'\'''
...
Date de la dernière modification: 2023-09-02 19:13:35 +0200 (sam. 02 sept. 2023)
diffoscope confirm that the file timestamp is not set correctly
in the generated archive:
$ diffoscope NOK/libxmlrpc-r3176-svn5.tar.gz OK/libxmlrpc-r3176-svn5.tar.gz
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 8d3b1781f6)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Since Buildroot commit [1], the test_gstreamer1 runtime test is
failing to build the tesseract-ocr package. The root cause is that
the test uses the default arm external toolchain, which is the Linaro
ARM 2018.05 based on gcc 7.3.1.
Since [1], tesseract-orc no longer compiles with gcc 7 (it requires at
least gcc 8).
This commit fixes the issue by switching the toolchain to the
bootlin versions (based on gcc 14 at the time of this commit).
Also, changing the compiler version slightly slowed down the video
encoding. This commit also increases the encoding command timeout.
Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/9407846232
[1] f32da8b984
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
(cherry picked from commit 356c70677c)
The busybox date applet accepts the following:
date @1234567
but this confuses the coreutils version which doesn't implicitly set
time. As some tests might need coreutils binaries we should ensure the
emulator login will work with both. Fix this by passing the -s (set)
option to the command.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit bfb490799e)
When time comes to check certificates, the date and time in the emulated
machine should be close enough to the actual values, so that certificate
validity can be checked.
Some Qemu machines have an RTC (e.g. arm vexpress-a9 has a pl031), and
the kernel needs a driver for those RTC. It is not guaranteed that the
machine used for a test meets those two conditions; in such a case, the
time in the machine starts way back in the past (1970-01-01T00:00:00Z on
sysv, or the release date of systemd). This is the case with the default
kernel, so such tests do not have the proper time.
Set the date to the date of the host system. This is going to be accurate
to the second, which is, by far, enough for our purpose.
To avoid having to consider what combination of emulated machine and kernel
configuration are being used, we always set the date, as this is a
generic step that should be done by the infra (like login in as root is).
The Emulator() class doesn't inherit from unittest.TestCase, so we can't
call any of the usual self.assertXXX() methods; instead, we just raise
a standard exception, like is done a few lines above to detect the login
prompt.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Ricardo Martincoski <ricardo.martincoski@datacom.com.br>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit cf8641b73e)
This is in the mainline kernel since v5.1-rc3:
9c38f1f04408 ("kconfig/[mn]conf: handle backspace (^H) key")
Quoting the commit's log:
"
Backspace is not working on some terminal emulators which do not send the
key code defined by terminfo. Terminals either send '^H' (8) or '^?' (127).
But currently only '^?' is handled. Let's also handle '^H' for those
terminals.
"
Signed-off-by: Sébastien Szymanski <sebastien.szymanski@armadeus.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit d65c10c20a)
With the TARGETPLATFORM fix we need to rebuild the container so move to the
latest snapshot to get updated (security) fixes.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit 12e9fc1da5 ("support/docker: add tar 1.35") added logic to download
and build tar for the host, but mistyped a configure option:
configure: WARNING: unrecognized options: --disable-year2028
Fix that.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Previously, these tests had the potential for timing out with the
default 5 second timeout value if initializing /dev/urandom took too
long.
Now the tests use a 10 second timeout value.
Signed-off-by: Vincent Fazio <vfazio@xes-inc.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Prior to b7d251293a, txaio would test both asyncio and twisted.
Add back the twisted sample and include the twisted package in the
config so both modes of the package are tested.
Fixes: b7d251293a ("package/python-txaio: drop python 2 support")
Signed-off-by: Vincent Fazio <vfazio@xes-inc.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
buildx correctly complains about our ENV lines:
3 warnings found (use docker --debug to expand):
- LegacyKeyValueFormat: "ENV key=value" should be used instead of legacy "ENV key value" format (line 21)
- LegacyKeyValueFormat: "ENV key=value" should be used instead of legacy "ENV key value" format (line 94)
- LegacyKeyValueFormat: "ENV key=value" should be used instead of legacy "ENV key value" format (line 95)
The ENV syntax is defined as ENV FOO=BAR, not ENV FOO BAR, so adjust the
Dockerfile to match to get rid of this warning.
https://docs.docker.com/reference/dockerfile/#env
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
With the new 'buildx' command using 'BuildKit' since Docker 23.0 [1],
it become easy to build the same image for several architectures [2].
The only requirement is to use a Dockerfile without any architecture
specific definition.
Since our current Dockefile contains already some i386 specific
packages (g++-multilib,libc6:i386) to provide x86 32bits support,
we have to install them conditionally.
Update the build process described in the Dockerfile accordingly.
For now, Aarch64 hosts can't be used by the Buildroot testsuite (yet)
since the Bootlin external toolchain (used by default) currently only
support x86_64 hosts.
[1] https://docs.docker.com/engine/release-notes/23.0/
[2] https://www.docker.com/blog/multi-arch-build-and-images-the-simple-way/
[3] https://www.docker.com/blog/multi-arch-build-what-about-gitlab-ci/
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since commit b11956fb66 ("support/dependencies: require tar >= 1.35"),
Buildroot will build host-tar if the host does not have >= 1.35 available,
so add it to the container to save build time.
Debian 12 only provides tar 1.34. Tar 1.35 is available in testing - But
that then depends on a newer glibc, so instead build it from source using
the --disable-year2038 flag like we do for host-tar and install it into
/usr/local/bin so it shadows the Debian one.
This step runs as root (and needs to for make install), so add
FORCE_UNSAFE_CONFIGURE=1 as otherwise the configure script errors out.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit afece24a72 ("support/download: introduce curl backend for FTP
transfers") added curl download support, so add it to the container as well.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
SC2086 is now reported for auto_id since shellcheck 0.9.0:
In support/scripts/mkusers line 453:
add_one_group "${g}" ${auto_id}
^--------^ SC2086 (info): Double quote to prevent globbing and word splitting.
So quote it to get rid of this error.
Signed-off-by: Romain Naour <romain.naour@smile.fr>
[Peter: quote variable instead of disabling check]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As for SC1090 and SC2016 [1], disable SC1091 and SC2294 too since they
are now reported by shellcheck 0.9.0:
In support/download/curl line 42:
eval ${CURL} "${@}"
^--^ SC2294 (warning): eval negates the benefit of arrays. Drop eval to preserve whitespace/symbols (or eval as string).
[1] bcee3ca6d6
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
flake8 5.0.4 provided by Debian 12 (bookworm) now detect a missing
whitespace after 'assert' keyword:
E275 missing whitespace after keyword
Signed-off-by: Romain Naour <romain.naour@smile.fr>
[Peter: fix subject, drop extra paranthesises]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
When check-package fail due to flake8 coding style checks, it suggest
to use flake8 command line:
run 'flake8' and fix the warnings
But flake8 is actually missing from the container used by
utils/docker-run, so add it to the container as well.
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since commit [1], test_atf is failing. See [2].
Passing -no-pie to ld does not produce the bl31.elf file (and does not
generate an error). TF-A 2.11 reworked its makefiles and switched to
gcc to link files. This commit fixes this failing test by updating to
this newer TF-A version.
Fixes: [2].
[1] 248c2d045e
[2] https://gitlab.com/buildroot.org/buildroot/-/jobs/8609307891
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Like Yann E. Morin did for
support/testing/tests/package/test_python_django.py, adjust the logic
to avoid the weird "self.assertTrue(False, ...)" construct, and
instead test exit_code after the loop.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
The test runs the flask sample app through the waitress wsgi server
instead of the flask development server.
Signed-off-by: Marcus Hoffmann <buildroot@bubu1.eu>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Since the bump of NetworkManager to 1.50.0 in Buildroot commit
489d1f9249, the
tests.package.test_python_sdbus_networkmanager.TestPythonPy3SdbusNetworkmanager
is failing to build as NM >= 1.50.0 needs headers >= 4.20.
We fix this test by using a newer toolchain. This means we can't use
anymore the basic toolchain config, so we use a fully specific
defconfig for this test.
Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/9019173666
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Updating directly to 4.x is too difficult and complex, as mender 4.x is
a complete rewrite in C++, with several new dependencies and changes.
As such, update to the last version that was written in Go, which is
3.5.3. The following changes are necessary:
The artifact_info file is no longer supported. Instead, mender now expects a
bootstrap.mender artifact created by mender-artifact /var/lib/mender.
See the following for more information:
https://northerntech.atlassian.net/browse/MEN-2585https://northerntech.atlassian.net/browse/MEN-2583https://docs.mender.io/release-information/release-notes-changelog/mender-client#mender-3-5-0-1https://github.com/mendersoftware/mender/blob/3.5.3/Documentation/automatic-bootstrap-artifact.md
- Remove all instances of creating or copying an artifact_info file
from board examples and package/mender.
- Add a generate_mender_bootstrap_artifact method to
board/mender/x86_64/post-image-efi.sh which creates a proper
bootstrap.mender file and places it in the data partition.
- Add a post-build.sh script to support/testing/tests/package/test_mender
with the same generate_mender_bootstrap_artifact method with one change:
As the rootfs.ext4 file is not yet created; we omit the optional
--provides "rootfs-image.checksum:${img_checksum}" argument when
generating the bootstrap.mender file.
Mender expects the device_type file to exist in /var/lib/mender/device_type.
This is further supported by the following line in tests/Dockerfile.daemon:
`echo device_type=docker-client > /var/lib/mender/device_type`
Add a migration section in docs/manual/migrating.adoc
Update the package/mender/readme.txt to include a quick mention of the
bootstrap.mender artifact change.
License changes:
New:
vendor/github.com/klauspost/compress/internal/snapref/LICENSE (BSD-3-Clause)
vendor/github.com/klauspost/compress/zstd/internal/xxhash/LICENSE.txt (MIT)
Removed:
vendor/github.com/klauspost/compress/internal/snapref/LICENSE
vendor/github.com/klauspost/compress/zstd/internal/xxhash/LICENSE.txt
Modified:
LICENSE: Update year from 2022 to 2024
LIC_FILES_CHKSUM.sha256: Sum of all Licenses changed
vendor/github.com/mendersoftware/mender-artifact/LICENSE
- Update year from 2022 to 2023
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
[Arnout:
- fix LIC_FILES_CHKSUM.sha256 hash
- Better URL for bootstrap artifact doc
- Migrate to 2025.02 instead of 2024.11
- Improve migrating text (editorial changes)
]
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
- Move the check for /var/run/mender.pid to before self.run_mender_test()
for the two tests using sysvinit.
- Instead of checking for /var/run/mender.pid during the systemd test,
check to ensure "systemctl is-active mender-client" returns "active"
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
[Arnout: fix flake8 issue]
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
