Qemu 7.2 finally allows us to move beyond building for original Core 2/Core i7 era hardware,
and this patch adds support for the newer generations. But first, a bit of
background:
Recently toolchains gained support for specifying x86-64 'levels' of
instruction set support; v3 corresponds to 2013-era Haswell CPUs
(and later), with AVX, AVX2 and a few other instructions that
were introduced in that generation. I believe this is preferrable
to picking a specific CPU model as the baseline.
Here's Phoronix's feature article that explains the feature and the available levels:
"Both LLVM Clang 12 and GCC 11 are ready to go in offering the new x86-64-v2, x86-64-v3, and x86-64-v4 targets.
These x86_64 micro-architecture feature levels have been about coming up with a few "classes" of Intel/AMD CPU processor support rather than continuing to rely on just the x86_64 baseline or targeting a
specific CPU family for optimizations. These new levels make it easier to raise the base requirements around Linux x86-64 whether it be for a Linux distribution or a particular software application where
the developer/ISV may be wanting to compile with greater instruction set extensions enabled in catering to more recent Intel/AMD CPUs."
Here's gcc docs for it:
https://gcc.gnu.org/onlinedocs/gcc/x86-Options.html
And here's the formal specification (click on the pdf link):
https://gitlab.com/x86-psABIs/x86-64-ABI
The actual tune file was created by copying corei7 tunes and doing
search/replace on them. Qemu options were dropped as unnecessary.
32 bit tune was dropped as well, as there is no 32 bit only CPU
that also supports these new instructions; all of the v3 capable
chips are 64 bit.
Signed-off-by: Alexander Kanavin <alex@linutronix.de> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Martin Jansa [Thu, 29 Dec 2022 16:21:31 +0000 (17:21 +0100)]
nativesdk-rpm: export RPM_ETCCONFIGDIR and MAGIC in environment like RPM_CONFIGDIR
* the paths in wrapper don't work for rpmdeps which is installed in
${libdir}/rpm unlike other wrapped bins from ${bindir} these relative
paths don't work there
* replace environment.d-rpm.sh with here-doc so that we can use
OE variables
* in the end it might be better to just get rid of the wrappers at
this point and depend on environment.d to always set right values
* the wrappers were added in:
commit 760103cdaed3e820888d8984ec0b76cfc831d534
Author: Ovidiu Panait <ovidiu.panait@windriver.com>
Date: Fri May 25 10:48:29 2018 +0800
nativesdk-rpm: Add wrappers for nativesdk support
When installing the SDK to a non-default path, running "rpm --showrc" from the
sdk will produce the following error:
error: Unable to open /opt/windriver/wrlinux-small/10.17.41/sysroots/x86_64-wrlinuxsdk-linux/usr/lib/rpm/rpmrc for reading: No such file or directory.
Fix this by adding wrappers that dynamically export the RPM_CONFIGDIR,
RPM_ETCCONFIGDIR and MAGIC environment variables, pointing to the proper
sdk locations.
* the rpm.sh in environment.d a bit later:
commit 5f16fd0bf774314c79572daf4ba7e4a8ae209ba1
Author: hongxu <hongxu.jia@windriver.com>
Date: Wed Jul 29 01:22:06 2020 -0700
While installing/extracting SDK to a non-default dir(not /opt),
run rpm failed:
$ python3 -c "import rpm"
|error: Unable to open /opt/windriver/wrlinux-graphics/20.31/sysroots/
x86_64-wrlinuxsdk-linux/usr/lib/rpm/rpmrc for reading: No such file or
directory.
This patch adds a flexible way to configure RPM_CONFIGDIR in SDK.
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Kai Kang [Thu, 29 Dec 2022 13:11:46 +0000 (21:11 +0800)]
libx11: 1.8.1 -> 1.8.3
Update libx11 from 1.8.1 to 1.8.3.
* add LICENSE 'ISC' which is added by
https://github.com/freedesktop/xorg-libX11/commit/a04b84f0
* adjust the ordering according to
https://www.openembedded.org/wiki/Styleguide#Ordering_and_grouping
Signed-off-by: Kai Kang <kai.kang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Kai Kang [Thu, 29 Dec 2022 13:11:45 +0000 (21:11 +0800)]
libx11-compose-data: 1.6.8 -> 1.8.3
Update libx11-compose-data from 1.6.8 to 1.8.3
* add 'ISC' to LICENSE because of commit
https://github.com/freedesktop/xorg-libX11/commit/a04b84f0
* adjust ordering according to
https://www.openembedded.org/wiki/Styleguide#Ordering_and_grouping
Signed-off-by: Kai Kang <kai.kang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Kai Kang [Thu, 29 Dec 2022 13:11:44 +0000 (21:11 +0800)]
xorg-lib-common.inc: set default value of XORG_EXT
It replaced tar.bz2 tar archieves with tar.xz from the middle of 2021 on
page https://www.x.org/releases/individual/lib/. So set the default
value of XORG_EXT with 'tar.xz', and update related recipes accordingly.
Signed-off-by: Kai Kang <kai.kang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Wang Mingyu [Tue, 27 Dec 2022 06:35:06 +0000 (14:35 +0800)]
bind: upgrade 9.18.9 -> 9.18.10
Changelog:
==========
The key file IO locks objects would never get deleted from the hashtable due to
off-by-one error.
ANY responses could sometimes have the wrong TTL.
Speed up the named shutdown time by explicitly canceling all recursing ns_client
objects for
Removing a catalog zone from catalog-zones without also removing the referenced
zone could leave a dangling pointer. [GL #3683]
nslookup and host were not honoring the selected port in TCP mode. [GL #3721]
Deprecate alt-transfer-source, alt-transfer-source-v6 and
use-alt-transfer-source. [GL #3694]
Move the "final reference detached" log message from dns_zone unit to the
DEBUG(1) log level.
Fix assertion failure in isc_http API used by statschannel if the read callback
would be called on HTTP request that has been already closed.
Deduplicate time unit conversion factors.
Copy TLS identifier when setting up primaries for catalog member zones.
Deprecate 'auto-dnssec'. [GL #3667]
The decompression implementation in dns_name_fromwire() is now smaller and
faster. [GL #3655]
Use the current domain name when checking answers from a dual-stack-server.
Ensure 'named-checkconf -z' respects the check-wildcard option when loading a
zone. [GL #1905]
Deprecate 'coresize', 'datasize', 'files', and 'stacksize' named.conf options.
The view's zone table was not locked when it should have been leading to race
conditions when external extensions that manipulate the zone table where in use.
Some browsers (Firefox) send more than 10 HTTP headers. Bump the number of
allowed HTTP headers to 100. [GL #3670]
NXDOMAIN cache records are no longer retained in the cache after expiry,
even when serve-stale is in use. [GL #3386]
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Revert "python3-native: further tweak to sysconfig.py to find python includes correctly"
Unfortunately, sysconfig is used by:
- setuptools to find out where to install new headers
- autoconf-archive to find where Python.h is
In the first case we want the target location, in the second case we want
the sysroot location. The second case is not yet used in the wild
(as projects such as gpgpme are using outdated distutils-based m4 macros),
but it will become a problem once python 3.12 is out, and
said projects are forced to migrate to sysconfig. Let's deal with it
then in a more careful manner, and hopefully convince upstream that
there is a general problem too.
Signed-off-by: Alexander Kanavin <alex@linutronix.de> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Drop 0001-bpo-36852-proper-detection-of-mips-architecture-for-.patch:
the patch has been rejected upstream (see the bug), and does quite a
bit more than just fix the mips softfloat problem. If the problem
still exists, then the fix needs to be re-done, and re-submitted.
The other two patches have been merged upstream.
License-Update: http->https
Signed-off-by: Alexander Kanavin <alex@linutronix.de> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Xiangyu Chen [Mon, 26 Dec 2022 07:16:19 +0000 (15:16 +0800)]
grub2: backport patch to fix CVE-2022-2601 CVE-2022-3775
Backport patch from upstream to solve CVE-2022-2601 CVE-2022-3775 dependency:
font: Fix size overflow in grub_font_get_glyph_internal()
(https://git.savannah.gnu.org/cgit/grub.git/commit/?id=9c76ec09ae08155df27cd237eaea150b4f02f532)
Backport patch from upstream to fix following CVEs:
CVE-2022-2601: font: Fix several integer overflows in grub_font_construct_glyph()
(https://git.savannah.gnu.org/cgit/grub.git/commit/?id=768e1ef2fc159f6e14e7246e4be09363708ac39e)
CVE-2022-3775: font: Fix an integer underflow in blit_comb()
(https://git.savannah.gnu.org/cgit/grub.git/commit/?id=992c06191babc1e109caf40d6a07ec6fdef427af)
Signed-off-by: Xiangyu Chen <xiangyu.chen@windriver.com> Signed-off-by: Steve Sakoman <steve@sakoman.com> Signed-off-by: Liwei Song <liwei.song@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Pavel Zhukov [Sun, 25 Dec 2022 12:46:58 +0000 (13:46 +0100)]
oeqa/rpm.py: Increase timeout and add debug output
[Yocto #14346]
Systemd may be slow in killing pam session sometimes [1][2]. It may cause rpm
test to fail because there's process (sd_pam) running and own by "test1" user
after timeout.
Increasing timeout to 2 mins and assert earlier with debug output if
there's such process(es). If increasing of timeout doesn't help we may
want to force deletion of the user as [2] suggests.
qemu no longer carries libslirp in-tree,
so enabling slirp requires providing external libslirp.
Another noteworthy change is:
x86: TCG support for AVX, AVX2, F16C, FMA3 and VAES instructions
... which means both meta-intel and qemu x86 targets can
now fully utilize Haswell-and-later instruction set with benefits
for performance in emulation and on silicon.
Changelog:
https://wiki.qemu.org/ChangeLog/7.2
Signed-off-by: Alexander Kanavin <alex@linutronix.de> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
python3-native: further tweak to sysconfig.py to find python includes correctly
Without this, sysconfig would report the target location, which is not
what we want in cross-builds. This is used in particular by
latest autoconf-archive m4 macros, which have migrated from
(removed in py 3.12) distutils to setuptools.
Signed-off-by: Alexander Kanavin <alex@linutronix.de> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
tcmode-default.inc: set LLVMVERSION to a major version wildcard
There is no longer need to be specific: none of the recipes
using this variable in core need to know the particular version
of llvm exactly when using the default one set via PREFERRED_VERSION
from LLVMVERSION.
This as well enables automated version updates of llvm
for minor releases.
Signed-off-by: Alexander Kanavin <alex@linutronix.de> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* and switch from tar.gz to tar, because the tar.gz archives upstream are regular tar as well now
https://www.w3.org/XML/Test/ still has 3 separate URLs for .zip, .tar
and .tar.gz, but both tar links return the same file:
xmlts20130923.tar: POSIX tar archive (GNU)
xmlts20130923.tar.gz: POSIX tar archive (GNU)
xmlts20130923.zip: Zip archive data, at least v1.0 to extract, compression method=store
-rw-r--r-- 1 martin martin 5.7M Sep 23 2013 xmlts20130923.tar
-rw-r--r-- 1 martin martin 5.7M Sep 23 2013 xmlts20130923.tar.gz
-rw-r--r-- 1 martin martin 1.6M Sep 23 2013 xmlts20130923.zip
Luis [Sat, 24 Dec 2022 01:49:46 +0000 (01:49 +0000)]
rm_work.bbclass: use HOSTTOOLS 'rm' binary exclusively
The do_rm_work() task is using the first available 'rm' binary
available in PATH to remove files and folders.
However, depending on the PATH setup and RECIPE_SYSROOT_NATIVE
contents, the function can be using the 'rm' binary available
in RECIPE_SYSROOT_NATIVE, a folder that will get removed.
This causes a sporadic race-condition when trying to access the
'rm' binary of a folder already deleted.
Solve this by exclusively using the HOSTTOOLS 'rm' binary, as
this folder will not get removed.
Signed-off-by: Luis Martins <luis.pinto.martins@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Chen Qi [Thu, 24 Nov 2022 02:55:28 +0000 (18:55 -0800)]
selftest: allow '-R' and '-r' be used together
For now, there's no easy way to skip some specified tests when
using 'oe-selftest -r'. This is because '-R' and '-r' are in the
mutually exclusive group. Looking at the codes, the codes have already
allowed running specified tests and skipping specified tests at
the same time. So move '-R' out of the mutually exclusive group
and change its help message.
After this change, the following command could be used.
oe-selftest -r A -R A.B.C
This does introduce a behavior regression.
Previous 'oe-selftest -R xxx' needs to now be changed to
'oe-selftest -a -R xxx'.
Signed-off-by: Chen Qi <Qi.Chen@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Alex Kiernan [Wed, 21 Dec 2022 12:52:58 +0000 (12:52 +0000)]
libstd-rs: Move source directory to library/test
Building libstd-rs from library/std omits proc_macro from the sysroot.
Using library/test causes that to be installed which then allows cargo
to build (https://github.com/meta-rust/meta-rust/issues/266)
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changqing Li [Fri, 23 Dec 2022 03:38:40 +0000 (11:38 +0800)]
base.bbclass: Fix way to check ccache path
The previous code had 2 issues:
1. make hosttools/ccache always link to host's ccache (/usr/bin/ccache)
even we have one buildtools
2. make hosttools/gcc etc, link to host's gcc event we have one
buildtools when keyword ccache in buildtools's path, eg:
/mnt/ccache/bin/buildtools
This patch is for fix above issues.
Signed-off-by: Changqing Li <changqing.li@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 21 Dec 2022 23:20:23 +0000 (23:20 +0000)]
oeqa/concurrencytest: Add number of failures to summary output
When running oe-selftest and seeing the end of a running log, it is
extremely helpful to know if there have been failures or not to save
looking at the rest of the log. Add the number of failures to the summary
line so that people monitoring builds have an easier time before the end
totals are printed.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Randy MacLeod [Thu, 22 Dec 2022 19:17:53 +0000 (14:17 -0500)]
valgrind: skip the boost_thread test on arm
This test is failing on the arm workers only so skip there until the issue
can be worked on and resolved. The bug #14311 will remain open for tracking.
Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Wang Mingyu [Wed, 21 Dec 2022 11:05:56 +0000 (19:05 +0800)]
gstreamer1.0: upgrade 1.20.4 -> 1.20.5
Changelog:
===========
systemclock waiting fixes for certain 32-bit platforms/libcs
alphacombine: robustness improvements for corner case scenarios
avfvideosrc: Report latency when doing screen capture
d3d11videosink: various thread-safety and stability fixes
decklink: fix performance issue when HDMI signal has been lost for a long time
flacparse: Fix handling of headers advertising 32 bits per sample
mpegts: Handle when iconv doesn't support ISO 6937 (e.g. musl libc)
opengl: fix automatic dispmanx detection for rpi4 and fix usage of eglCreate/DestroyImage
opusdec: Various channel-related fixes
textrender: event handling fixes, esp. for GAP event
subparse: Fix non-closed tag handling
videoscale: fix handling of unknown buffer metas
videosink: reverse playback handling fixes
qtmux: Prefill mode fixes, especially for raw audio
multiudpsink: allow binding to IPv6 address
rtspsrc: Fix usage of IPv6 connections in SETUP
rtspsrc: Only EOS on timeout if all streams are timed out/EOS
splitmuxsrc: fix playback stall if there are unlinked pads
v4l2: Fix SIGSEGV on state change during format changes
wavparse robustness fixes
Fix static linking on macOS (opengl, vulkan)
gstreamer-vaapi: fix headless build against mesa >= 22.3.0
GStreamer Editing Services library: Fix build with tools disabled
webrtc example/demo fixes
unit test fixes for aesdec and rtpjitterbuffer
Cerbero: Fix ios cross-compile with cmake on M1; some recipe updates and other build fixes
Binary packages: pkg-config file fixes for various recipes (ffmpeg, taglib, gstreamer)
Binary packages: Enable high bitdepth support for libvpx (VP8/VP9 encoding/decoding)
Binary packages: ship aes plugin
Miscellaneous bug fixes, memory leak fixes, and other stability and reliability improvements
Performance improvements
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Wang Mingyu [Wed, 21 Dec 2022 03:01:01 +0000 (11:01 +0800)]
psmisc: upgrade 23.5 -> 23.6
Changelog:
=========
* buildsys: Fix DEJAGNU work-around Debian #1015089
* killall: Use kill if pidfd_send_signal fails Debian #1015228
* fuser: Do not mention nonexistent - reset option #42
* fuser: Use modern statn where possible
* pstree: Better AppArmor support !30
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Ross Burton [Tue, 20 Dec 2022 12:55:41 +0000 (12:55 +0000)]
gtk+3: fix reproducible builds
We can't drop the missing meson.build file into the build tree via SRC_URI
as then the timestamp of that file becomes the timestamp of the build,
which results in repeated builds having different timestamps.
Instead patch the file into the build, which doesn't change the SDE
every build.
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>
recipes: Enable nativesdk for gperf, unifdef, gi-docgen and its dependencies
I'm building a SDK (bitbake task: populate_sdk) for WPE/WebKit
development [1] and I need the tools below to be extended with the class
nativesdk. They work fine on the SDK after this change.
The needs are because:
1) gperf and unifdef are required by the WebKit build system at build
time.
2) gi-docgen is required by the WebKit build system when CMake option
'ENABLE_DOCUMENTATION' is enabled, and gi-docgen itself depends on
the following python modules that also need to enable nativesdk:
python3-markdown, python3-smartypants and python3-typogrify
Signed-off-by: Carlos Alberto Lopez Perez <clopez@igalia.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
rust: Enable building rust from stable, beta and nightly channels
Rust follows the train release model via the stable, beta and nightly channels,
by default we build rust from the stable channel, however there are certain
features which are only available in the beta or nightly channels.
Make these channels available by setting a RUST_CHANNEL variable which defaults
to stable making this change transparent to the user.
The snapshot version used by rust during its compilation wont necessarily match
the version being built, specially if were building from an unstable channel,
to avoid confusion rename this to SNAPSHOT_VERSION and use RUST_VERSION for the
version to be built, which is automatically defined to PV.
Append -beta or -nightly to rusts PV for signature awareness.
It is important to note that this does not build rust from the beta/nightly
published tarball (which today build rust v1.67.0 and v1.68.0 respectively),
instead this builds rust from the current selected version (1.66.0) and enables
the beta/nightly features for that version.
Setting the variable RUST_CHANNEL=nightly results in the following:
projects / thirdparty / openembedded / openembedded-core.git / log
