Martin Jansa [Mon, 15 May 2023 17:37:40 +0000 (19:37 +0200)]
image-live.bbclass: respect IMAGE_MACHINE_SUFFIX
* fixes:
ERROR: core-image-minimal-1.0-r0 do_bootimg: /OE/build/poky/build/tmp/deploy/images/qemux86-64/core-image-minimal-initramfs-qemux86-64.cpio.gz is invalid. initrd image creation failed.
ERROR: core-image-minimal-1.0-r0 do_bootimg: ExecutionError('/OE/build/poky/build/tmp/work/qemux86_64-poky-linux/core-image-minimal/1.0-r0/temp/run.build_hddimg.2001892', 1, None, None)
when IMAGE_MACHINE_SUFFIX is set to empty in local.conf
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Add a PACKAGECONFIG for the environment variable proxy configuration
(http_proxy, etc), and enable by default as this is the fallback
provider if you're not using libproxy or gnomeproxy.
Signed-off-by: Ross Burton <ross.burton@arm.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Ross Burton [Mon, 15 May 2023 13:29:49 +0000 (14:29 +0100)]
gnutls: add PACKAGECONFIG options for certification compression
GnuTLS has optional support for certificate compression. These are not
widely enabled in other distributions and were previously disabled in
standard builds, so don't enable them by default.
Signed-off-by: Ross Burton <ross.burton@arm.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Qiu Tingting [Mon, 15 May 2023 02:45:41 +0000 (10:45 +0800)]
e2fsprogs: Fix error SRCDIR when using usrmerge DISTRO_FEATURES
When build e2fsprogs ptest with usrmerge DISTRO_FEATURES,
in test_script and test_one scripts, value of SRCDIR has problem.
SRCDIR=/usr/usr/lib/e2fsprogs/ptest/test
ptest log
# ptest-runner e2fsprogs
START: ptest-runner
2022-12-17T11:08
BEGIN: /usr/lib/e2fsprogs/ptest
ls: cannot access '/usr/usr/lib/e2fsprogs/ptest/test/[a-zA-Z]_*': No such file or directory
./test_script: line 54: /usr/usr/lib/e2fsprogs/ptest/test/test_post: No such file or directory
DURATION: 0
END: /usr/lib/e2fsprogs/ptest
2022-12-17T11:08
STOP: ptest-runner
TOTAL: 1 FAIL: 0
Reason
In Makefile.in, SRCDIR is set by prefix and libdir.
@echo "SRCDIR=${prefix}${libdir}/e2fsprogs/ptest/test" >> test_one
@echo "SRCDIR=${prefix}${libdir}/e2fsprogs/ptest/test" >> test_script
prefix=/usr
libdir=/usr/lib (when usrmerge is set)
Chen Qi [Mon, 15 May 2023 04:53:35 +0000 (12:53 +0800)]
oe-buildenv-internal: remove path from previous project
Util now, only paths added for current project are removed
to avoid PATH growing unnecessarily. This is to handle the case
of sourcing the init script into different build directories.
However, if we source the init script from different projects into
different build directories, the paths added by previous projects
are not cleaned up.
To avoid this, we record the paths added into OE_ADDED_PATHS, and
remove it in the next sourcing.
The OE_ADDED_PATHS variable is exported mainly for the two reasons:
1. users can clearly see it in the `env' command's output.
2. if the prefixed PATH is carried into the subprocess (e.g., a subshell),
so should this OE_ADDED_PATHS variable that prefixes it.
Note that the paths, "$OEROOT/scripts:$BITBAKEDIR/bin:", are added
as a whole. A previous commit, "oe-buildenv-internal: Add paths to $PATH individually",
made the change to treat these two paths separately, the reason was
not "assuming the path to the scripts directory always is in $PATH
directly before the bitbake directory". But this is exactly the effect
of the codes. I see no reason why we should complicate things.
Max Krummenacher [Thu, 11 May 2023 23:22:14 +0000 (01:22 +0200)]
perf: add libtraceevent packageconfig
With kernel 6.4-rc1 commit 6898e60f709b ("perf build: If libtraceevent
isn't present error the build") having libtraceevent is a required
dependency unless building it with 'NO_LIBTRACEEVENT=1'.
Older kernels did build libtraceevent as provided by the kernel sources.
(Unless LIBTRACEEVENT_DYNAMIC=1 was defined, then it tried to pickup
an external provided libtraceevent and printed a warning if not found.)
Add a libtraceevent packageconfig which adds the dependency on
libtraceevent if set, otherwise builds with NO_LIBTRACEEVENT=1.
For older kernels the old behaviour of building libtraceevent from
the kernel sources isn't changed.
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Max Krummenacher [Thu, 11 May 2023 23:22:13 +0000 (01:22 +0200)]
perf: make pkg-config available
Fixup the to be called pkg-config as $(CROSS_COMPILE)pkg-config doesn't
exist.
With kernels < 6.4 pkg-config is only used for the gtk+-2.0 integration
which is disabled in the OE config. Thus the issue wasn't exposed so
far. With kernel 6.4 pkg-config is additionally used for libtraceevent
and libtracefs.
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Max Krummenacher [Thu, 11 May 2023 23:22:11 +0000 (01:22 +0200)]
libtraceevent: import from meta-openembedded
Import from meta-openembedded at commit d250a0dc0 ("libtraceevent:
upgrade 1.7.1 -> 1.7.2").
Starting with kernel 6.4 libtraceevent is a dependency of perf. While
one can still build perf without it by opting out one would loose its
functionality compared with building perf from kernels before 6.4
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Mikko Rapeli [Wed, 10 May 2023 12:59:33 +0000 (15:59 +0300)]
qemurunner: support serial console login via qemu stdout
runqemu script works with qemu machines which provide login
and serial console to the qemu process stdout. Add the same support
to qemurunner so that testing with testimage.bbclass is possible.
Default qemu machines provide serial console boot logs and login via
socket to qemu process but I don't see a reason why qemu process stdout
should not be supported too since they work with runqemu as well.
Signed-off-by: Mikko Rapeli <mikko.rapeli@linaro.org> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Refactor relocate-modules.patch; upstream has implemented the same
logic, but made it os x only - we can simply drop those ifdef guards:
https://gitlab.gnome.org/GNOME/glib/-/commit/af83c6571e47d06468b8162dd29ad57bcf4967a8
Signed-off-by: Alexander Kanavin <alex@linutronix.de> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Martin Jansa [Wed, 10 May 2023 14:21:14 +0000 (16:21 +0200)]
populate_sdk_base.bbclass: respect MLPREFIX for ptest-pkgs's ptest-runner
* since "populate_sdk_base: ensure ptest-pkgs pulls in ptest-runner" change:
https://git.openembedded.org/openembedded-core/commit/?id=ecff0642be5781f7f6cca617158b04ac9a0e85f0
in multilib build when building lib32-foo-image it can pick wrong
ptest-runner package if it was built in the same TMPDIR before the
image, do_rootfs then fails to find it, if the package manager config
doesn't have 64-bit feed enabled:
opkg_prepare_url_for_install: Couldn't find anything to satisfy 'ptest-runner'
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Richard Purdie [Wed, 10 May 2023 11:23:45 +0000 (12:23 +0100)]
qemu: Further updates to the ppc patch after upstream discussion
After someone tested real hardware, the patch needs tweaks to match the 3.0 ISA
behaviour. It won't change much from our perspective but may as well keep the
patch in sync.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Trevor Gamblin [Tue, 9 May 2023 15:55:26 +0000 (11:55 -0400)]
python3-calver: Add recipe
calver is "a setuptools extension for automatically defining your Python
package version as a calendar version." It is required for
python3-trove-classifiers (another new recipe), which in turn is
required for the upgrade of python3-hatchling from 1.13.0 to work.
Trevor Gamblin [Fri, 5 May 2023 18:51:21 +0000 (14:51 -0400)]
python3-trove-classifiers: Add recipe
python3-trove-classifiers is "Canonical source for classifiers on
PyPI.". It is required to update python3-hatchling from the current
version (1.13.0) in oe-core, and depends on python3-calver (another new
recipe). Also add ptests.
Signed-off-by: Trevor Gamblin <tgamblin@baylibre.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Zang Ruochen [Mon, 8 May 2023 02:12:56 +0000 (10:12 +0800)]
elfutils: upgrade 0.188 -> 0.189
Refresh the following patch.
0015-config-eu.am-do-not-use-Werror.patch
Remove the following patches as they have been fixed in the new version.
0001-PR29926-debuginfod-Fix-usage-of-deprecated-CURLINFO_.patch
0002-debuginfod-client-Use-CURLOPT_PROTOCOLS_STR-for-libc.patch
changelog:
Version 0.189 "Don't deflate!"
configure: eu-nm, eu-addr2line and eu-stack can provide demangled symbols
when linked with libstdc++. Use --disable-demangler to disable.
A new option --enable-sanitize-memory has been added for msan
sanitizer support.
libelf: elf_compress now supports ELFCOMPRESS_ZSTD when build against
libzstd
libdwfl: dwfl_module_return_value_location now returns 0 (no return type)
for DIEs that point to a DW_TAG_unspecified_type.
elfcompress: -t, --type= now support zstd if libelf has been build with
ELFCOMPRESS_ZSTD support.
backends: Add support for LoongArch and Synopsys ARCv2 processors.
Andrew Jeffery [Fri, 5 May 2023 13:19:54 +0000 (22:49 +0930)]
Revert "ipk: Decode byte data to string in manifest handling"
cf9df9e8d89f ("ipk: Decode byte data to string in manifest handling")
did a bit of least-effort fix to a string vs byte sequence issue in the
manifest handling. The approach was chosen as it localised the fix,
rather than having to analyse further call sites.
However since then f2167ae80258 ("package_manager/ipk: do not pipe
stderr to stdout") was applied, reworking the output handling from the
subcommand. dummy_bytes() now returns a string, so stop trying to decode
it.
Fixes: f2167ae80258 ("package_manager/ipk: do not pipe stderr to stdout") Cc: Curtis Meier <cmeier@us.ibm.com> Cc: Pam Eggler <eggler@us.ibm.com>
(From OE-Core rev: b61739554780d70307d2b6b37d2b3b1c7df93c77)
Signed-off-by: Andrew Jeffery <andrew@aj.id.au> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Ross Burton [Thu, 4 May 2023 13:10:09 +0000 (14:10 +0100)]
Revert "ffmpeg: move ffmpeg config into packageconfig"
There's very little reason to expose "build shared libraries", "build
position-independent code", or "enable threads" as recipe-specific
packageconfig options. Revert the commit which did this and explicitly
set the relevant options in EXTRA_OECONF.
This was causing:
ERROR: Nothing PROVIDES 'virtual/lib32-arm-distromllib32-linux-gnueabi-binutils' (but virtual:multilib:lib32:oe-core/meta/recipes-core/images/core-image-minimal.bb DEPENDS on or otherwise requires it). Close matches:
virtual/arm-distromllib32-linux-gnueabi-binutils
virtual/lib32-arm-distromllib32-linux-gnueabi-compilerlibs
virtual/lib32-arm-distromllib32-linux-gnueabi-go-runtime
ERROR: Required build target 'lib32-core-image-minimal' has no buildable providers.
Missing or unbuildable dependency chain was: ['lib32-core-image-minimal', 'virtual/lib32-arm-distromllib32-linux-gnueabi-binutils']
in cases where multilib and WIC were enabled at the same time
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Tim Orling [Wed, 3 May 2023 15:39:14 +0000 (08:39 -0700)]
libmodule-build-perl: upgrade 0.4232 -> 0.4234
* Skip t/pod_parser.t ptest since it is for testing
documentation and requires Pod::Parser for which
we do not have a recipe. Adding one does not make
much sense since Pod::Parser was dropped from
Perl > 5.31.1 in favor of Pod::Simple
https://metacpan.org/pod/Pod::Parser#DESCRIPTION
Changes:
https://metacpan.org/dist/Module-Build/changes
0.4234 - Fri 28 Apr 2023 10:46:26 CEST
- PodParser now respects =encoding directives
- Don't use libraries in t/bundled for Build.PL
- make_tarball: workaround for broken tar on Darwin
- Respect $Config{man1ext}/$Config{man3ext}
- Do not require a compiler if c_source is an empty list
Signed-off-by: Tim Orling <tim.orling@konsulko.com> Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Armin Kuster [Wed, 3 May 2023 14:31:18 +0000 (10:31 -0400)]
os-release: Add CPE_NAME
Its time we add the CPE_NAME to os-release.
The vendor field is hardcoded to "openembedded" as it is the base
framework. We will use "DISTRO" to identify which variation of
openembedded is being used.
BSD-4-Clause is only applicable to the {PN}-doc package as when I
check for the source code I find below files which only uses the
license BSD-4-Clause
~/sources/libbsd$ grep -rl "All advertising materials mentioning features or use of this software" *|grep -v \.1|grep -v \.5|grep -v \.8 | sort
COPYING
man/arc4random.3bsd
~/sources/libbsd$ grep -rnB5 "BSD-4"
COPYING-9-Files:
COPYING-10- man/arc4random.3bsd
COPYING-11-Copyright:
COPYING-12- Copyright 1997 Niels Provos <provos@physnet.uni-hamburg.de>
COPYING-13- All rights reserved.
COPYING:14:License: BSD-4-clause-Niels-Provos
So other all PACKAGES should not contain the BSD-4-Clause
Randolph Sapp [Tue, 2 May 2023 16:56:12 +0000 (11:56 -0500)]
kernel-devicetree: allow specification of dtb directory
Fedora/Redhat and Arch are somewhat standardized on their dtb directory
structure. Let's add some flags to configure yocto to mimic that
behavior.
Add the following variables to the kernel class:
- KERNEL_DTBDEST (controls the destination directory for dtbs)
- KERNEL_DTBVENDORED (controls if vendor subdirectories are to
be respected)
Currently KERNEL_DTBDEST is expected to be a subdir of KERNEL_IMAGEDEST
and KERNEL_DTBVENDORED is expected to be "true"/"false". This only
applies to the package directory structure. The deploydir structure is
purposely left untouched for compatibility with existing recipes.
By default this is configured to behave the same as the current recipe
and produce a flat dtb directory at KERNEL_IMAGEDEST.
Markus Volk [Tue, 2 May 2023 15:56:57 +0000 (17:56 +0200)]
gtk4: update 4.10.0 -> 4.10.3
Overview of Changes in 4.10.3, 22-04-2023
=========================================
* Fix a popover positioning regression in 4.10.2
* Fix issues with slow loading files in the file chooser
Overview of Changes in 4.10.2, 21-04-2023
=========================================
* Fixed issues:
- Holding control to select multiple files broken in filechooser (#5669)
- Inspector crash (#5681)
- Listbase doesn't account for bottom padding in size_allocate_child (#5380)
- Leaking AT contexts (#5690)
- OpenGL / Windows: Crash when closing gtk4-widget-factory (#5685)
- GTK apps crash on startup when setting cursor-size to 0 on Wayland (#5700)
- Segmentation fault: gdk_wayland_toplevel_set_startup_id() needs to null-check
display->xdg_activation before using it (#5701)
- Possible use-after-free under gtk_scrolled_window_update_use_indicators() (#5684)
- Wrong error message in `gtk_init` (#5704)
- Segfault when scrolling after changing ListView model (#5763)
- Bluetooth panel from the Settings app: clicking in the "Downloads" link
no longer opens Nautilus (#5671)
- Broadway docs or code is broken (#5662)
- Disabled GtkPicture's are not properly themed (#5683)
- Setting CSS padding to a GtkTextView gives the context menu an offset (#5695)
- A11y: the Showing state is used only for windows (#5194)
- Gtk4 expander: CSS nodes mismatch code vs. documentation (#5723)
- Invoking gtk inspector on a folder results in a crash (#5729)
- Double tap requires very precise touch input (#5580)
- Name autocompletion dropdown in the GTK4 FileChooser's Save dialog gets
stuck, creates artifacts, jumps around (#5743)
- Links are not opened when xdg-desktop-portal OpenURI is not available (#5733)
- GtkSnapshot generates no nodes appending whitespace-only layouts (#5747)
* Translation updates
British English
Bulgarian
Chinese (China)
French
Indonesian
Korean
Russian
Serbian
Slovenian
Turkish
Overview of Changes in 4.10.1, 14-03-2023
=========================================
* GtkFileChooser
- Improve search performance
- Be safe against pathless files
- Fix memory leaks
- Only show local files in recent files
- Show most recent files first
- Make files non-selectable in selet_folder mode
insane.bbclass: simplify exceptions for 32 bit time API check
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.
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.
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>
bitbake.conf: set minimum required target kernel to 5.15
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:
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.
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>
projects / thirdparty / openembedded / openembedded-core-contrib.git / log
