Dependencies only make sense when opensbi is being used to deliver
payload which maybe an artifact of kernel or u-boot, otherwise it should
be not added. This avoids circular dependencies when RISCV machines do
not define RISCV_SBI_PAYLOAD but do define RISCV_SBI_FDT
(From OE-Core rev: 83a986fcac535415108caf70a9fdee8edc59f7d5)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Upstream docs currently have a typo showing the changes as 8.14.0
89d58c2ec0
9.1.0
-----
* New functions
- iter_index() (from the Python itertools docs)
- transpose() (from the Python itertools docs)
- matmul() (from the Python itertools docs)
- factor() (from the Python itertools docs)
- gray_product() (thanks to haukex)
* Changes to existing functions
- sieve() was updated to match the Python itertools docs
- maxsplit() was updated to fix a bug (thanks to abingham)
- sliced() had its type hint updated (thanks to ad-chaos)
* Other changes
- The batched function is marked as deprecated and will be removed in a
future major release. For Python 3.12 and above, use itertools.batched
instead. (thanks to neutrinoceros)
- The type hints now used postponed evaluation of annotations from PEP 563
(thanks to Isira-Seneviratne)
- Some documentation issues were fixed (thanks to Voskov and jdkandersson)
(From OE-Core rev: 436b92846406273fb3651f4f026e9fd3871ee71c)
Signed-off-by: Tim Orling <tim.orling@konsulko.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This helps in switching toolchains cleanly for kernel build
between gcc and clang
Currently, some kernels allow building with clang but not all
the distro might use clang as default system compiler but kernel
may demand gcc which is provided via KERNEL_* variables, however
kernel does use OBJCOPY at places during build and it maybe set
to use llvm objcopy when using clang. That should be a deliberate
setting when clang is used for kernel as well, otherwise it should
use binutils provided objcopy
(From OE-Core rev: 17b409f2fd97894e0943d13c2cb0d52abde647e3)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Cc: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The python3-gpg's version is now '1.18.0-unknown'. Such version
has been treated as invalid for setuptools >=66.0.0. The error
message is as below:
pkg_resources.extern.packaging.version.InvalidVersion: Invalid version: '1.18.0-unknown'
(package: gpg)
Fix this issue by removing the '-unknown' suffix.
(From OE-Core rev: 3a8bf81c5fb6111eab9784192754271efc85ee5c)
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
By default there is nothing in nghttp2-client and nghttp2-server ,nghttp2-client
and nghttp2-server aren't created. So there are dependences error if install
main package.
Problem: conflicting requests
- nothing provides nghttp2-client >= 1.52.0 needed by nghttp2-1.52.0-r0.core2_64
- nothing provides nghttp2-server >= 1.52.0 needed by nghttp2-1.52.0-r0.core2_64
(From OE-Core rev: d2cbe060955c598bd81923ecd554fbe82c17af99)
Signed-off-by: Lei Maohui <leimaohui@fujitsu.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Instead of creating empty /etc/machine-id file using touch, write
text "uninitialized" into it. Systemd requires "uninitialized" in
the /etc/machine-id file to trigger systemd-firstboot .
(From OE-Core rev: 263c852269027b1bd311e116021ac49d9001e1b0)
Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Multiple filters have been added to resulttool to ensure that test
results are compared only to relevant older tests results
Add some unit tests about added filters:
- tests should only be compared when machine matches
- tests run as oeselftest should only be compared when
OESELFTEST_METADATA matches
- many ptests should match despite non-static names
- runtime tests should match with matching tests regarding ltp tests
content
(From OE-Core rev: b84302ef56f2516742a496aef43b89d4c3decd37)
Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Remove some obsolete CVE ignores now that releases have been made, CPEs
updated, or upgrades done:
CVE-2020-10029 is marked as fixed in 2.32.
CVE-2021-27645 is marked as fixed in 2.34.
CVE-2022-39046 is marked as fixed in 2.37.
(From OE-Core rev: a383d0e726011007419dd102cd6484733d80d9f6)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This CVE has been fixed in the 2.37 release branch (07b9521fc6) and
we're now using a SHA that incorporates that commit, so manually mark it
as ignored.
(From OE-Core rev: b143e355ef362dac89f8b125904d2e47c74e597e)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Multiple CVE are patched in kernel but appears as active because the NVD
database is not up to date.
CVE are ignored if and only if all versions of kernel used by master are patched.
Also ignore CVEs with wrong CPE (applied to kernel but actually are for
another package)
(From OE-Core rev: 92770a08c04a6c1eb351231d937b16e76558f013)
Signed-off-by: Geoffrey GIRY <geoffrey.giry@smile.fr>
Reviewed-by: Yoann Congal <yoann.congal@smile.fr>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Provide a mechanism to allow users to choose whether the /tmp directory
is on persistent storage (non-volatile) or a RAM-based tmpfs (volatile).
The default is volatile.
Works for both sysvinit-based and systemd-based systems.
(From OE-Core rev: 8b76c0637eaeaf5bd5e696680cd74b7a642f4157)
Signed-off-by: Trevor Woerner <twoerner@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The tests need to run as a non-root user, but that user also needs to be
able to write into the tests directory to create the work directories.
Use the new ptest user that ptest-runner creates to avoid having to add
a new user, and ensure that the test/ directory is owned by the ptest
group and that group has write access.
This solves the ptest regression where since oe-core 9a287b4 no quilt
tests were actually running.
A better solution would be to allow the tests to create a work directory
in another directory, but that would need work upstream which can happen
later.
(From OE-Core rev: 70dc9c114283cde18771287fb17fcebc537e7387)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some tests really don't want to be run as root, either because they
expect to get "permission denied" errors occasionally, or have explicit
checks.
Instead of numerous recipes re-inventing user creation in run-ptest,
create a user in the ptest-runner recipe that they can all use.
(From OE-Core rev: d58cf08421e0f89840820370dd3c5b6e153efcbb)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This will produce a more useful hint for those setting
up or testing virgl headless:
runqemu - ERROR - No render nodes found in /dev/dri: ['by-path', 'card0']. If /dev/dri/renderD* is absent due to lack of suitable GPU, 'modprobe vgem' will create one suitable for mesa llvmpipe software renderer.
as qemu itself isn't helpful:
alex@Zen2:/srv/storage/alex/yocto/build-64-alt$ qemu-system-x86_64 -display egl-headless
qemu-system-x86_64: egl: no drm render node available
qemu-system-x86_64: egl: render node init failed
(From OE-Core rev: cbbada6a6c9b0a2e97f7395117dad986555f2db9)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
runqemu (which the test also runs) already performs this check,
and does it better, as it additionally unsets a number of environment
variables that can confuse and misdirect host's pkg-config.
(From OE-Core rev: ad179728a6978d929e2a876abe7e1cbd6b76a502)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add some tests for new yocto_testresults_query.py helper. First test is taken
from yocto-autobuilder-helper feature which has moved in yocto_testresults_query
(From OE-Core rev: 9ce28f685e4282d81f179877cbafd0a52fa887bd)
Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Many stored results TEST_TYPE are set to "oeselftest", however all those
tests are not run with the same sets of parameters, so those tests results may
not be comparable.
Attach relevant parameters as tests metadata to allow identifying tests
configuration so we can compare tests only when they are run with the same
parameters.
(From OE-Core rev: 406ec0a87e80d1ee0dd1d246adfe9507c10450ff)
Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
>From NEWS for v250:
* Device nodes for the Software Guard eXtension enclaves (sgx_vepc) are
now also owned by the system group "sgx".
>From NEWS for v248:
* Intel SGX enclave device nodes (which expose a security feature of
newer Intel CPUs) will now be owned by a new system group "sgx".
Fixes following journal error entry during startup:
/lib/udev/rules.d/50-udev-default.rules:43 Unknown group 'sgx', ignoring
This is seen already on kirkstone.
(From OE-Core rev: bab455cd9b1b82e778f8523a767eb281edf6689e)
Signed-off-by: Peter Marko <peter.marko@siemens.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This brings in following changes
* 7d756e1c (upstream/master, origin/master) dns: prefer monotonic clock for timeouts
* 07616721 fix return value of wcs{,n}cmp for extreme wchar_t values
* 35fdfe62 math: fix undefined shift in logf
* 7e13e5ae inet_pton: fix uninitialized memory use for IPv4-mapped IPv6 addresses
* 7e6da7ac hsearch: fix null pointer arithmetic UB
* f79b973d increase sendmsg internal buffer to support SCM_MAX_FD
* 0ab97350 mq_notify: block all (application) signals in the worker thread
* 711673ee mq_notify: join worker thread before returning in error path
* 8c0c9c69 mq_notify: rework to fix use-after-close/double-close bugs
* fde6891e mq_notify: use semaphore instead of barrier to sync args consumption
* c3cd04fa fix pthread_detach inadvertently acting as cancellation point in race case
* 115149c0 powerpc-sf longjmp clobbering of val argument
* 5763f003 riscv64: add vfork
* 269d1938 fix wrong sigaction syscall ABI on mips*, or1k, microblaze, riscv64
* ea3b40a3 fix integer overflow in WIFSTOPPED macro
* f897461d fix debugger tracking of shared libraries on mips with PIE main program
* a4b0a665 expose memmem under baseline POSIX feature profile
* 9532ae13 use libc-internal malloc for pthread_atfork
* 7d358599 prevent invalid reads of nl_arg in printf_core
* c5f4b2df elf.h: add ELFCOMPRESS_ZSTD
* 159d1f6c semaphores: fix missed wakes from ABA bug in waiter count logic
(From OE-Core rev: 7379412a2ce7fe24943c858755f058dc2a15a081)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
systemd-oomd.service has:
User=systemd-oom
So we need to create a static user for it, if its packageconfig
is enabled.
(From OE-Core rev: 1586edc919326cdbb412feafb9cde2cb15e419cb)
Signed-off-by: Luca Boccassi <luca.boccassi@microsoft.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The patch fixes the following errors observed when building ffmpeg in
vulkan-enabled distros:
| src/libavutil/hwcontext_vulkan.c:363:7: error: 'VK_EXT_VIDEO_DECODE_H264_EXTENSION_NAME' undeclared here (not in a function); did you mean 'VK_EXT_VIDEO_ENCODE_H264_EXTENSION_NAME'?
| 363 | { VK_EXT_VIDEO_DECODE_H264_EXTENSION_NAME, FF_VK_EXT_NO_FLAG },
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | VK_EXT_VIDEO_ENCODE_H264_EXTENSION_NAME
| src/libavutil/hwcontext_vulkan.c:364:7: error: 'VK_EXT_VIDEO_DECODE_H265_EXTENSION_NAME' undeclared here (not in a function); did you mean 'VK_EXT_VIDEO_ENCODE_H265_EXTENSION_NAME'?
| 364 | { VK_EXT_VIDEO_DECODE_H265_EXTENSION_NAME, FF_VK_EXT_NO_FLAG },
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | VK_EXT_VIDEO_ENCODE_H265_EXTENSION_NAME
(From OE-Core rev: b16c8696be9d56edb5ff77210abfff9a784fad89)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This patch changes the docbook-xml distribution package source from
debian to docbook which aligns the recipe with the source
originator. Other changes made to the recipe include:
This patch resolves [YOCTO #13270].
- removal of version 4.0 given docbook no longer distributes version
4.0
- simplification of the header description.
- inline comments to communicate the rationale behind gathering
multiple versions and LICENSE file handling.
- minor ordering changes to align with styling conventions and better
context to better communicate use
- Added comment about magic XMLCATALOGS variable. This variable and
subsequent inherit might be better placed in recipes that depend on
docbook-xml but leave it for now to not cause a patch landslide.
- Modify the related docbook-xml patch to remove docbook-xml-4.0 since
it is no longer provided.
- Put BBCLASSEXTEND to end of recipe per RP note on convention
(From OE-Core rev: fc4471019313d981f0c951220cc65e8374a7e8e6)
Signed-off-by: Richard Elberger <rich@richelberger.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
GI_DATA_ENABLED is the right source for making that decision
as it additionally factors in presence of qemu-usermode in
MACHINE_FEATURES.
(From OE-Core rev: b32bb078d73ab9fd636d059cf70b5530a2f94cb4)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Enabling the network uncondictional is not need for some use cases.
Such use case is usefull to reuse the sstate-cache of the build
and it requires the icecc inherit in all of the builds.
The real control control of the icecc is in the variable ICECC_DISABLED
so this patch change the logic to enable the network when the icecc is in use.
(From OE-Core rev: 0fd3a9c13a30a67ccef6619627efd9613755a0c3)
Signed-off-by: Jose Quaresma <jose.quaresma@foundries.io>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We keep seeing this failure on the autobuilder but the output amounts
to "False is not True". Improve the debug message on the chance it may
make the issue clearer.
(From OE-Core rev: d03f4cf19c2cc96e9d942252a451521dfec42ebc)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
After mc upgrade to 4.8.29, it introduces PERL_FOR_BUILD [1] and will
result in the below build failure on some system when we build under
a project with long path.
| make[4]: Entering directory '/buildarea2/WTEST_Regression/Rerun/build_dir/02161802-multilib_pkgs_conflict/qemux86-64-standard-std-OE/build1/tmp-glibc/work/core2-64-wrs-linux/mc/4.8.29-r0/build/doc/hlp/es'
make[4]: Nothing to be done for 'install-exec-am'.
| ../../../src/man2hlp/man2hlp ../../../doc/man/es/mc.1 ../../../../mc-4.8.29/doc/hlp/es/xnc.hlp mc.hlp.es
| ../../../src/man2hlp/man2hlp: line 32: use: command not found
| ../../../src/man2hlp/man2hlp: line 33: use: command not found
| ../../../src/man2hlp/man2hlp: line 36: syntax error near unexpected token `('
| ../../../src/man2hlp/man2hlp: line 36: `my %static = ('
| make[4]: [Makefile:632: mc.hlp.es] Error 2 (ignored)
$ head -n 5 tmp-glibc/work/core2-64-wrs-linux/mc/4.8.29-r0/build/src/man2hlp/man2hlp
#! /buildarea2/WTEST_Regression/Rerun/build_dir/02161802-multilib_pkgs_conflict/qemux86-64-standard-std-OE/build1/tmp-glibc/hosttools/perl
#
# Man page to help file converter
# Copyright (C) 1994, 1995, 1998, 2000, 2001, 2002, 2003, 2004, 2005,
# 2007, 2010, 2011
It's beacause the shebang line exceeds 128 which is usually the max
length.
So set ac_cv_path_PERL_FOR_BUILD to fix the above issue and altogether
to remove the build path issue.
[1] 6b67d231a2
(From OE-Core rev: 33d91a2434eeb20076c8226340b224b1d606c40f)
Signed-off-by: Mingli Yu <mingli.yu@windriver.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
While arm64 is a valid UBOOT_ARCH (according to mkimage -A) u-boot
keeps arm64 specific dts under 'arch/arm' directory.
As the result the recipe tries to install arch/arm64 (if UBOOT_DTB
was specified) and fails with [1]. Remapping "arm64" to "arm" to fix this
issue.
[1]
| install: cannot stat '.../u-boot/1_2023.01-r0/build/arch/arm64/dts/u-boot.dtb': No such file or directory
(From OE-Core rev: 3ca99403d5f320c6d7ae59b107f3b3bf183b4089)
Signed-off-by: Pavel Zhukov <pavel@zhukoff.net>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Adds support for the kiosk shell to weston. The Kiosk shell is designed
to run a fullscreen application in a "kiosk" mode, but unlike the
fullscreen shell the kiosk mode still provides traditional desktop
protocols.
(From OE-Core rev: da4d6547f312285b5e246d8f974d2de76294fef3)
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changelog:
=========
Added experimental support for suppressing subtest output dots in non-verbose
mode with --no-subtests-shortletter -- this allows the native pytest column
calculations to not be disrupted and minimizes unneeded output for large CI systems.
(From OE-Core rev: 486c627d6851079c49e18d56d7052d3842af3206)
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changelog:
============
Features
--------
- Ignore PIP_REQUIRE_VIRTUALENV for "pip index"
- Implement "--break-system-packages" to permit installing packages into
"EXTERNALLY-MANAGED" Python installations.
Bug Fixes
---------
- Improve handling of isolated build environments on platforms that
customize the Python's installation schemes, such as Debian and
Homebrew.
- Do not crash in presence of misformatted hash field in "direct_url.json".
(From OE-Core rev: 2cab887da4618a23c1fc52d27e1061102857f470)
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mesa contains several optional tools that can be useful for application
and driver developers. Add a packageconfig option to build a mesa-tools
package containing relevant programs.
Note, the fdperf tool for the freedreno depends on libconfig, which is
not a part of oe-core, thus this tool is not built by default by this
recipe even when freedreno driver is selected. The fdperf tool needs to
be explicitly enabled by adding 'freedreno-fdperf' to the PACKAGECONFIG.
(From OE-Core rev: 95eb4008350424f25d08da96bc6d7d56bf4cc161)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Now that we have the dependencies we can set the RDEPENDS for all builds,
not just target.
(From OE-Core rev: e7980308d78e7fb8ea1568b80363c809c91c2363)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
