Existing implementation required to list both specific problematic apis, and files that
use them: neither is necessary as both are seen in package_qa error messages, and
can cause excessive amount of exception lines, if there are too many files, or
they are installed in arch-specific locations. Also, the value of INSANE_SKIP
should be the test that needs to be skipped, and in this case it wasn't.
Also, all problematic recipes are now correctly listed.
(From OE-Core rev: e6ebd0c556dfc576a59f5755d97089a2a241f698)
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>
Otherwise assimp will silently fall back to a vendored copy of zlib
which will fail with -D_TIME_BITS=64 due to https://github.com/madler/zlib/pull/764
This was exposed by multilib mips core-image-minimal SDKs, where the default
64 bit sysroot has zlib, but 32 bit sysroot does not.
(From OE-Core rev: c0fb603c9e26e91388320c02842b42cc7b091d6c)
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>
In particular this enables a number of useful features in glibc
(which utilize newer kernel APIs), such as actually using 64 bit
time_t versions of kernel syscalls:
https://sourceware.org/git/?p=glibc.git;a=blob;f=sysdeps/unix/sysv/linux/kernel-features.h;h=07b440f4eea364b05fa49bf71ceebf78f80efe13;hb=HEAD#l164
In general, OLDEST_KERNEL setting is used in these two places:
- kernel.bbclass compares it with the target kernel version being built.
If a vendor BSP still offers an older kernel, OLDEST_KERNEL should be set to match.
- glibc recipe passes it as a parameter to the build so that additional features
and optimized paths that kernels older than OLDEST_KERNEL are enabled.
Note that there is a related setting, SDK_OLDEST_KERNEL, which remains as
it was (at 3.2.0) to ensure maximum compatibility with kernels on SDK host
machines; that setting is used to build nativesdk-glibc and verify the kernel
version when the SDK is being installed.
Build host kernel versions are not checked directly; compatible distros
are listed instead.
(From OE-Core rev: feb8e3fb71131a414a2a9271832b4e16860301ea)
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>
It builds glibc source like other glibc recipes do,
and so the same problems occur.
(From OE-Core rev: 68b50d362ec61f27be818e40fcbb281d9bacf756)
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>
Otherwise xwayland.pc would not be present in sysroot, this leads to
some xwayland configs missing like have_listenfd, have_glamor.
(From OE-Core rev: 7f1932cb5a408320a5b542e20ba2807718349e8f)
Signed-off-by: Ming Liu <liu.ming50@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fix issue of the below instance template systemd service dependency
[Install]
WantedBy=svc-wants@%i.service
creating the symlink (instance "a" example)
/etc/systemd/system/svc-wants@%i.service.wants/svc-wanted-by@a.service
which should be
/etc/systemd/system/svc-wants@a.service.wants/svc-wanted-by@a.service
as implemented by this change.
The functionality appears regressed just after "thud" baseline when the
logic was refactored from shell script into python (commit
925e30cb10)
(From OE-Core rev: 308397f0bb3d6f3d4e9ec2c6a10823184049c9b5)
Signed-off-by: Martin Siegumfeldt <mns@gomspace.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Below is the output of run-ptest script under qemu
root@qemux86-64:~# /usr/lib/zvariant/ptest/run-ptest
running 37 tests
test framing_offset_size::tests::framing_offset_size_bump ... ok$<2>
test owned_value::tests::from_value ... ok$<2>
test object_path::unit::owned_from_reader ... ok$<2>
test str::tests::from_string ... ok$<2>
test signature::tests::signature_slicing ... ok$<2>
test str::tests::test_ordering ... ok$<2>
test owned_value::tests::map_conversion ... ok$<2>
test owned_value::tests::serde ... ok$<2>
test tests::enums ... ok$<2>
test tests::derive ... ok$<2>
test tests::f64_value ... ok$<2>
test tests::i16_value ... ok$<2>
test tests::fd_value ... ok$<2>
test tests::i32_value ... ok$<2>
test tests::i8_value ... ok$<2>
test tests::i64_value ... ok$<2>
test tests::ip_addr ... ok$<2>
test tests::issue_59 ... ok$<2>
test tests::issue_99 ... ok$<2>
test tests::array_value ... ok$<2>
test tests::object_path_value ... ok$<2>
test tests::dict_value ... ok$<2>
test tests::signature_value ... ok$<2>
test tests::serialized_size ... ok$<2>
test tests::struct_byte_array ... ok$<2>
test tests::struct_ref ... ok$<2>
test tests::str_value ... ok$<2>
test tests::option_value ... ok$<2>
test tests::struct_value ... ok$<2>
test tests::struct_with_hashmap ... ok$<2>
test tests::u16_value ... ok$<2>
test tests::u32_value ... ok$<2>
test tests::unit ... ok$<2>
test tests::u8_value ... ok$<2>
test tests::unit_fds ... ok$<2>
test tests::value_value ... ok$<2>
test tests::recursion_limits ... ok$<2>
test result: ok$<2>. 37 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.21s
root@qemux86-64:~#
(From OE-Core rev: 912bbec9fe44f22ab70c3553af6cb699543b8411)
Signed-off-by: Frederic Martinsons <frederic.martinsons@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The source code of bcrypt extension doesn't define any tests
but it is to show the ptest-cargo usage
Below is the updated test suite output for the recipe (cut):
PASS: tests/test_bcrypt.py:test_invalid_params[password-$2b$04$cVWp4XaNU8a4v1uMRum2SO-513-10-ValueError]
PASS: tests/test_bcrypt.py:test_invalid_params[password-$2b$04$cVWp4XaNU8a4v1uMRum2SO-20-0-ValueError]
PASS: tests/test_bcrypt.py:test_2a_wraparound_bug
============================================================================
Testsuite summary
running 0 tests
test result: ok$<2>. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.01s
root@qemux86-64:/usr/lib/python3-bcrypt/ptest#
(From OE-Core rev: 7388fec005b94d8e3827afcdf0de1e5b05ea1346)
Signed-off-by: Frederic Martinsons <frederic.martinsons@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This new class offers the possibility to build rust unit tests
(and integration tests) and find them correctly.
Due to non deterministic names of generated binaries, a custom
parsing of build result must be performed.
See https://github.com/rust-lang/cargo/issues/1924
All rust projects will generate a test binary with "cargo build --tests"
command, even if there are no test defined in source code.
The binary will just output that it ran 0 tests.
(From OE-Core rev: dad9bad239d757ae0b159fe5f1276b6856547b4c)
Signed-off-by: Frederic Martinsons <frederic.martinsons@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Specifically:
- add missing maintainer.inc entries for initramfs-module-*, systemd-machine-units and
target-sdk-provides-dummy and drop them from exception list.
- remove rust from exception list for unbuildable-by-default recipes as it is now buildable.
- add missing maintainer.inc entry for libx11-compose-data and cve-update-nvd2-native;
as they are also unbuildable by default, they needs to be in exception list as well.
(From OE-Core rev: e9158b191c1cfc16f97abed6c05891aa84fe9463)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
"sysconfdir" contains "/" by definition and thus using os.path.join()
leads to self.target_rootfs being always ignored (and thus attempting to
generate paths in host's /etc).
Use oe.path.join() instead which was made for this purpose.
(From OE-Core rev: 8414c504138f6de663f5130c6b4a6ede5605d88b)
Signed-off-by: Enrico Jorns <ejo@pengutronix.de>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
[YOCTO #15108] -- https://bugzilla.yoctoproject.org/show_bug.cgi?id=15108
Since the latest change, the PYTHONPATH is overwritten instead of extended.
This leads to changed behavior and build errors of recipes where the PYTHONPATH
is set before setup_target_config is run.
Fixes: c9617c03bcee ("python3targetconfig.bbclass: use PYTHONPATH to point to the target config")
(From OE-Core rev: 2442ab92f8610784d28d8d83056b643bd95b0b4e)
Signed-off-by: Johannes Schrimpf <dev@loewen-email.de>
[Luca: add Fixes: tag]
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This is consistently seen with musl and grep from busybox
Therefore backport a patch from upstream to fix it
(From OE-Core rev: 511bcd965af658e6bb0c61d9f2adb1af75af773b)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add back the dropped 0001-attr-fix-utime-for-symlink.patch
to fix symlink time setting issue on NFS.
The problem could be reproduced by runing the following command
on nfs booted qemu:
ln -s dest src && touch -h src
Apart from the rpm operations mentioned in the original patch,
'docker pull' also fails with a 'stale file' error. The common
pattern here is extracting files from a bundle and setting times
for them.
(From OE-Core rev: fe35a2c11ba6f87735bccae244817016f9c1b5db)
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This includes a fix for CVE-2023-30630.
Remove the patch changing prefix and instead pass the paths to the make
install command.
Passing -e to make is no longer needed, the Makefile respects variables
now.
(From OE-Core rev: 86f2f9cfdc23cc9a2579d968cb9bb0fc61608b62)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The recommendation from server maintainers is that the https protocol
is both faster and more reliable than the dedicated git protocol at this point.
Switch to it where possible.
(From OE-Core rev: 139102a73d4151f4748b4a861bd4ab28dda7dab7)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
As it is mainly used for internal purposes, so not necessarily useful
for regular users.
OE-core "nodistro" uses it as a fallback to the default setting.
(From yocto-docs rev: aa4d4034707d7c74488678e77d4f45d65782b2a7)
Signed-off-by: Michael Opdenacker <michael.opdenacker@bootlin.com>
CC: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The default value of address-cells is "1", so the generated "its" file
only support 32bits address for uboot FIT image.
However, some platforms may want to support 64bits address of
UBOOT_LOADADDRESS and UBOOT_ENTRYPOINT.
Therefore, add variables to support both 64bits and 32bits address.
By default, the address-cell is 1 which is used for 32bits load address.
If users would like to use 64bits load address,
users are able to set as following for "0x400000000" 64bits load address.
1. FIT_ADDRESS_CELLS = "2"
2. UBOOT_LOADADDRESS= "0x04 0x00000000"
(From OE-Core rev: b4e46cdbd0727d97d13c2b8e9e4ce19c5c693f51)
Signed-off-by: Jamin Lin <jamin_lin@aspeedtech.com>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The default value of address-cells is "1", so the generated "its" file
only supports 32bits address for kernel FIT image.
However, some platforms may want to support 64bits address
of UBOOT_LOADADDRESS and UBOOT_ENTRYPOINT.
Therefore, adds a variable to support both 64bits and 32bits address cells.
By default, the address-cell is "1" which is used for 32bits load address.
If users would like to use 64bits load address, users are able to
set as following for "0x400000000" 64bits load address.
1. FIT_ADDRESS_CELLS = "2"
2. UBOOT_LOADADDRESS= "0x04 0x00000000"
(From OE-Core rev: 8ce15c76c50d5d61524fea585d19989baefb5df2)
Signed-off-by: Jamin Lin <jamin_lin@aspeedtech.com>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* this in the end doesn't help much, I was debugging warning (about base-files.do_install
signature being different than expected) from:
python3 $target_sdk_dir/ext-sdk-prepare.py $LOGFILE '${SDK_INSTALL_TARGETS}'
this shows the warning on console, but it doesn't end in $LOGFILE, because it
writes only contents of cooker log into the $LOGFILE with:
with open(logfile, 'a') as logf:
logf.write('Preparing SDK for %s...\n' % ', '.join(sdk_targets))
ret = run_command_interruptible('BB_SETSCENE_ENFORCE=1 bitbake --quiet %s' % ' '.join(sdk_targets))
if not ret:
ret = run_command_interruptible('bitbake --quiet build-sysroots')
lastlog = get_last_consolelog()
if lastlog:
with open(lastlog, 'r') as f:
for line in f:
logf.write(line)
if ret:
print('ERROR: SDK preparation failed: error log written to %s' % logfile)
return ret
maybe we could remove whole support for $LOGFILE parameter and just redirect
the output like other commands on this line
(From OE-Core rev: 719f22df160ebde303274ccfc04049cffdb51577)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Use appropriate kernel image name instead of hard-coded vmlinuz for bootimg-pcbios plugin.
(From OE-Core rev: 8d9f00ba456fe76e0f4ef0e68ec64fc538c90d89)
Signed-off-by: Sergei Zhmylev <s.zhmylev@yadro.com>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add DT_FILES variable to allow the user of the class to select specific
dts files to build. This is useful for packages featuring dts files
for multiple machines.
Since many machine configs contain a list of dtb files
(e.g. KERNEL_DEVICETREE), DT_FILES works with both dts and dtb files.
(From OE-Core rev: a6164c384a5bf3980a7a6c7f23927af9aca93b85)
Signed-off-by: Petr Kubizňák <kubiznak@2n.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The perl-version.bbclass executes functions which can depend on
variables potentially populated by native, such as `libdir`. The
sanity `native-last` suggests that recipes should `inherit native`
last, but when that is done the variables like PERLVERSION end up
as `None`, since `${STAGING_LIBDIR}` needs `${libdir}` which is not
yet populated (by native).
All recipes in poky and widely used meta-layers have already been
updated to use the functions directly instead of relying on these
problematic variables. Delete the variables so that future recipes
do not make the mistake of using them.
Related: openbmc/openbmc#3770
(From OE-Core rev: 9351b6e7fab5669340bb062b9c84fb4faa3dce0b)
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Drop backported/merged patches:
Revert-linux-user-add-more-compat-ioctl-definitions.patch
configure-Fix-check-tcg-not-executing-any-tests.patch
contrib-vhost-user-blk-Replace-lseek64-with-lseek.patch
Revert-linux-user-fix-compat-with-glibc-2.36-sys-mou.patch
Drop socket chardev patch with conflicts:
chardev-connect-socket-to-a-spawned-command.patch
This last patch was added in support of swtpm however it isn't clear if anyone
is still using that workflow. The patch uses API calls such as as qemu_fork()
which were removed in 8.0.0 and replaced with gspawn calls. If anyone needs the
patch, it will be better for them to forward port it, test it and reinstate it,
preferably with a discussion with upstream about it too.
(From OE-Core rev: fe8125565af07b73f9b29db2188ecb6e884bcc70)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Upstream now only generates CMake files on Windows, so remove all
references to CMake.
A zsh completion is now installed, remove this for now as we don't really
use zsh.
(From OE-Core rev: 94cf6ef11bba381ab6f65b03ed1ed14022438151)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Remove determinism.patch, this is obsolete now that we set
GIT_CEILING_DIRECTORIES to stop git climbing the tree outside of the
work directory.
(From OE-Core rev: e652d9bee3940e8ee7c3346c801deef4801dbaeb)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
