systemd: change pni-names logic to something less confusing
Using bb.utils.contains in negative logic leads to either overly verbose
or plain confusing code, so harness the power of Ross and Antonin to
make this less verbose and slightly clearer.
This moves the negative logic from the inline python ('true if not') to
bash ('if !').
Robert Yang [Thu, 4 Sep 2025 03:40:38 +0000 (20:40 -0700)]
automake: 1.17 -> 1.18.1
* Drop 0001-configure-don-t-report-error-when-the-path-of-Perl-c.patch, it is a
warning not an error any more, so just drop it.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
- gettext-minimal-native recipe adapted to the new macro name (aclocal_DATA -> macros_DATA)
- install Makevars.template to gettext-minimal-native to be able to update missing variables
The file is renamed in the desitanion to Makevars.template.minimal, to avoid clashing
with the full gettext package (some recipes have both as dependency).
- autotools.bbclass: search for all po folders, and update the Makefile.in.in file
in each of them. This is required, because without this the old Makefile.in.in
files in these non ${S}/po folders are not updated with the new version of gettext, and
the compilation routinely fails due to this.
As part of the same change, also concatenate the source's Makevars files with the gettext
Makevars.template: this allows all variables to be defined, even the new ones that the
source didn't contain originally. First set the gettext template, and then append
the source's original Makevars to it, so it should override the variables that it was
setting originally, and keep the new variables at their default value.
- Add new PACKAGECONFIG to use libselinux, and also fix libselinux discovery in autotools
(use pkg-config instead of autotools discovery, because autotools' library search doesn't
respect custom sysroot)
Ptest results:
=======================
All 626 tests passed
(40 tests were not run)
=======================
DURATION: 36
END: /usr/lib/gettext/ptest
2025-08-30T17:49
STOP: ptest-runner
TOTAL: 1 FAIL: 0
With the previous version:
All 545 tests passed
(34 tests were not run)
rust-target-config.bbclass: move target generation before do_configure
The rust_get_targets task from the rust-target-config class generates some
Rust compiler configurations for different the target and host machine, which
is used by Rust during compiling a recipe. These configurations supposed to
be available by the time the Rust compiler is used for the first time.
By default, this task is executed "before do_compile" - it assumes that
this is the first task that would use this configuration.
However this is not always the case, especially with projects which are not
pure Cargo projects, rather Cargo/Rust is called by a different build system.
As an example librsvg uses meson, and during the do_configure step Meson probes
the Rust compiler, trying to gather some library info from Rust. In case the
rust_gen_targets task was executed before the do_configure task, then
everything works. However this is not always the case - sometimes the job is
executed between the configure and compile steps, in which case the configuration
fails.
To avoid such problems, generate these targets before the do_configure step.
Markus Volk [Fri, 5 Sep 2025 18:15:38 +0000 (20:15 +0200)]
glib-2.0: update 2.84.4 -> 2.86.0
Overview of changes in GLib 2.86.0, 2025-09-05
==============================================
* Rework how platform-specific introspected GIO APIs have to be imported to fix
problems with backwards-compatibility provision for it, by removing duplicate
platform-specific symbols from `Gio-2.0`. Users of platform-specific GIO APIs
should be unaffected, as `GIRepository` will now automatically import
`GioWin32-2.0` or `GioUnix-2.0` when asked to import `Gio-2.0`. However,
projects generating introspection data which depends on types from either of
those platform-specific GIRs must make sure they depend on those GIRs
explicitly, rather than just transitively depending on them through `Gio-2.0`
(#3744, work by Emmanuele Bassi, Marco Trevisan, Florian Müllner, and others)
* Fix file existence queries on Solaris, broken due to unexpected flags handling
within `faccessat()` (#3770, work by Niveditha Rau)
* Bugs fixed:
- #3744 GDesktopAppInfo API disappeared after girepository-2.0 port (Emmanuele
Bassi)
- #3768 g_test_trap_subprocess does not check G_TEST_SUBPROCESS_INHERIT_STDIN
(Philip Withnall)
- !4751 gtestutils: Fix a slightly broken example in a doc comment
- !4754 Update Polish translation 250825
- !4758 Update Swedish translation
- !4762 gio: gmemorymonitorpsi: Replace GRegex with g_str_has_prefix()
- !4765 girepository: Add an assertion to help scan-build
- !4767 glocalfile: Disable faccessat()-based query_exists on Solaris
- !4768 gmessages: Fix win32_keep_fatal_message regression
- !4769 docs: Fix typos
- !4770 Update Chinese translation
- !4771 Update Georgian translation
- !4772 po: Update Persian translation.
* Translation updates:
- Chinese (China) (lumingzh)
- Georgian (NorwayFun)
- Persian (Danial Behzadi)
- Polish (Piotr Drąg)
- Swedish (Anders Jonsson)
Overview of changes in GLib 2.85.4, 2025-08-22
==============================================
* Follow symlink (instead of overwriting it) when updating `mimeapps.list`
(#3579, work by Rafael Girão)
* Bugs fixed:
- #3579 mimeapps.list is overwritten if it is a symlink (Rafael Girão)
- #3724 Crash in g_hash_table_add after 252645135 elements (Tobias Stoeckmann)
- #3743 g_utf8_validate out parameter has wrong type (two)
- #3751 meta: clang-format refers to a broken link (Rafael Girão)
- #3758 Out-of-bounds read in GMemoryMonitorPoll (Philip Withnall)
- #3760 Stack overflow when recursing within g_log_structured() with
`G_LOG_FLAG_RECURSION` (Tobias Stoeckmann)
- #3761 Regression in g_printf() - can no longer output formatted values
containing NUL bytes (Luca Bacci)
- #3766 Update sl.po (Slovenian) (Martin)
- !4714 gmain: Reformat docs to fully use gi-docgen and match style guide
- !4720 Disable GMemoryMonitorPsi on Solaris
- !4727 garray: Improve and migrate documentation to gi-docgen
- !4735 build: Fix stp files for development versions
- !4736 systemtap: Use correct formatters/types
- !4738 docs: Add Thomas Haller as a co-maintainer of GObject
- !4739 Annotate ref/unref functions as transfer full
- !4740 gstrfuncs: Check parameter validity
- !4742 garray: Fix g_array_binary_search description
- !4743 Update Russian translation
- !4744 tests/gio: skip Unix socket-mock tests on Windows
- !4747 tests/printf: Use proper compare helper for unsigned types
- !4748 gconstructor: Add attribute used for TLS callback pointer
* Translation updates:
- Russian (jtux270)
- Slovenian (Martin)
Overview of changes in GLib 2.85.3, 2025-08-08
==============================================
* Fix encoding of output from `g_print()` and `g_printerr()` when locale is set
to `.utf8` on Windows (#3341, work by Luca Bacci)
* Bugs fixed:
- #3341 `g_print` and `g_printerr` will cause encoding errors on Windows when
locale is set to `.utf8` (Luca Bacci)
- #3739 Crash in accept_ready() of GThreadedSocketService Under High Load
(Philip Withnall)
- #3740 Documentation of g_win32_error_message does not contain information
about the behaviour when FormatMessageW failed (Philip Withnall)
- #3755 AIX: Unwanted symbol needs to be removed for AIX platform: getpwnam_r,
getpwuid_r (Parth Patel)
- !4706 gthreadpool: Clean up when g_thread_pool_new fails
- !4707 tests: Skip slow mainloop test on valgrind
- !4708 gfilenamecompleter: Fix g_object_unref() of undefined value
- !4709 tests: Connect to GMemoryMonitor signals earlier
- !4712 tests/thread-pool: Add a thread-pool fail test
- !4713 Fix test error for GMemoryMonitor
- !4715 gdbuserror: Reformat docs to fully use gi-docgen and match style guide
- !4722 tests: Add missing unistd.h header to thread-pool test
- !4723 tests: Add a missing poll condition to socket-listener test
- !4724 garray: Pass errors through GByteArray functions
- !4725 garray: Add checks to g_ptr_array_extend_and_steal
- !4726 Add a basic GFilenameCompleter test
- !4728 gbitlock: Fix documentation issues
- !4729 [RFC] Tests: do not set a timeout in Python tests
- !4730 gstrfuncs: Always treat G_MININT64 in g_ascii_strtoll
- !4731 glocalfile: Disable faccessat()-based query_exists on OpenBSD
- !4733 gvalue: Reformat docs to fully use gi-docgen and match style guide
- !4734 gspawn: Improve docstring for g_spawn_async()
Overview of changes in GLib 2.85.2, 2025-07-21
==============================================
* New Linux PSI based backend for `GMemoryMonitor` as an option to use instead
of the existing Low Memory Monitor daemon backend (!4481, work by Kate Hsuan)
* Bugs fixed:
- #1443 Deadlock between g_module_open() and dlopen() when called from a
constructor
- #2848 Doc: clarification request regarding g_match_info_fetch_pos return
value (Mark Lautman)
- #3712 Crash in g_thread_pool_new_full
- #3713 call g_file_enumerator_close in g_file_enumerator_finalize is not safe
(fbrouille)
- #3716 (CVE-2025-7039) (#YWH-PGM9867-104) Buffer Under-read on GLib through
glib/gfileutils.c via get_tmp_file() (Michael Catanzaro)
- #3721 GFile leak in g_local_file_set_display_name during error handling
(Philip Withnall, Michael Catanzaro)
- #3725 Deadlock on source_destroy_lock inside g_main_context_unref() and
g_source_destroy() (with child sources) (Matthew Waters)
- #3726 GApplication sometimes fails to call before_emit (Matthias Clasen)
- !4481 gio: gmemorymonitorpsi: Replace GMemoryMonitor backend with kernel PSI
event
- !4665 gio: enums: Fix GBusNameOwnerFlags's annotation
- !4667 Incorrect output parameter handling in closure helper of
g_settings_bind_with_mapping_closures
- !4669 Add missing `(array zero-terminated=1)` annotations
- !4676 Fix IPv6 scope-id from DNS responses being lost
- !4680 gbacktrace: Correctly wait for children on Unix
- !4681 (CVE-2025-6052) gstring: Improve
g_string_expand/g_string_append_len_inline checks
- !4682 gio-tool-launch: fix %k field code expansion
- !4683 gio-tool-launch: Fix mismatched curly quotes in translatable strings
- !4684 garray: Support unallocated zero terminated arrays
- !4685 garray: Use g_array_elt_len/pos where appropriate
- !4687 gstring: Fix g_string_append_vprintf overflow
- !4690 garray: Fix out of boundary write in g_ptr_array_copy
- !4692 tests: Fix a minor leak in array-test
- !4693 tests: Loosen string comparison assertion in gio-tool.py
- !4694 tests: Do not always skip array overflow checks
- !4695 garray: Add more element_size > 0 checks
- !4698 garray: Avoid exponential growth in g_array_copy
- !4699 garray: Set capacity in terminated take functions
- !4700 gfileutils: Fix OOB read in g_build_path(name)_va
- !4701 gbacktrace: Fix OOB write in stack_trace
- !4702 gio/filenamecompleter: Fix leaks
- !4703 application: NULL check for options
- !4704 tests: Add a regression test for GApplication command line handling
Overview of changes in GLib 2.85.1, 2025-06-13
==============================================
* Re-add the option of a singleton to `GIRepository` (#3664, work by
Christian Hergert)
* Add support for the `e` flag (O_CLOEXEC) to `g_fopen()` (!4564, work by
Luca Bacci and Philip Withnall)
* Make the `sysprof` Meson option yield when using GLib as a subproject (!4659,
work by Matthias Clasen)
* Use the Meson built-in `localedir` option (!4661, work by
Kleis Auke Wolthuizen)
* Bugs fixed:
- #1665 g_file_trash() should return PERMISSION_DENIED if files can't be
deleted (Ignacy Kuchciński)
- #3664 Lack of g_irepository_get_default() equivalent makes cross-library
integration extremely difficult (Christian Hergert)
- #3698 Misleading autogenerated hints in the documentation of
g_async_queue_pop() (Alicia Boya García)
- !4560 glib/gnulib/printf.c: Sync with gnulib
- !4564 gstdio: Add support for the `e` flag (O_CLOEXEC) to g_fopen()
- !4637 Rework Windows implementation of g_getenv()
- !4641 [th/gobj-drop-bit-lock] gobject: drop object_bit_lock() functions
- !4642 [th/gobj-empty-notify-queue] gobject: optimize notify-queue handling
for a single freeze
- !4643 GRegex: apply monospace typeface in description
- !4644 gio: add annotations on parameters of 'g_file_monitor_emit_event' and
of 'g_vfs_get_file_for_path'
- !4645 gregex: Clarify docs for end_pos
- !4646 GRegex: update class description
- !4649 GAsyncQueue: assert non-null data in push_sorted()
- !4650 tests: Add atomics to asyncqueue test global variables
- !4651 Meson: Add libglib_static dependency for use in tests
- !4652 gobject: clarify in documentation that g_value_set_boxed copies
- !4654 Fix buffer overflow in string-test
- !4655 gstring: Fix overflow check when expanding the string
- !4657 docs: Stop hiding the Unix-like APIs which are in Gio-2.0.gir
- !4658 gmarkup: make documentation more discoverable
- !4659 Make the sysprof feature yield
- !4661 meson: Use the appropriate localedir option
Overview of changes in GLib 2.85.0, 2025-05-20
==============================================
* Preserve mode for existing file when creating a temporary file for atomic
updates with g_file_set_contents() (dconf#76, work by Wesley Hershberger)
* Fix race conditions between g_main_context_unref() and g_source_*() methods
(#803, work by Matthew Waters)
* Allow file handles inside nested containers when using the `gdbus call`
command (#3624, work by Julian Sparber)
* Fix DNS resolution of local addresses in offline mode (#3641, work by
Patrick Griffis)
* Various performance improvements to GObject locking (various MRs by
Thomas Haller)
* Prefer matches occurring earlier in the string when searching
`GDesktopAppInfo`s, improving search for apps in gnome-shell (!4369, work by
Fina Wilke)
* Fix thread safety of `GClosure` flags (!4575, !4577, work by Sam James and
Philip Withnall)
* Bugs fixed:
- GNOME/dconf#76 dconf update can set incorrect permissions to dconf system db
(Wesley Hershberger)
- #490 Not clearly documented behavior of g_key_file_set_comment function.
(marklkram)
- #803 g_main_context_unref() versus g_source_*() race (Matthew Waters)
- #1002 GObject doesn't support removing a weak reference in a GWeakNotify for
the same object
- #1250 gsocketlistener: Fix IPv4 listen() error-handling resulting in use-
after-free
- #2377 Document that `g_socket_address_get_native_size()` can return `-1` on
errors
- #2544 Consider `g_log_always_fatal` for aborting in
`g_log_structured_array()` (sid)
- #3405 Enable -Wconversion warnings by default (progress towards this, but it
is not complete)
- #3616 docs: Broken link in GioActionEntry (Philip Withnall)
- #3617 Add generalised version of g_date_get_monday_week_of_year() (Philip
Withnall)
- #3624 `gdbus call` should look for file handles inside nested containers
(Julian Sparber)
- #3630 2.84.0 build failure on Linux: ../gio/gnetworkmonitornetlink.c:47:10:
fatal error: netlink/netlink_route.h: No such file or directory (Philip
Withnall)
- #3634 test failure with gobject-introspection 1.83.4: warning: element
doc:format from state 3 is unknown, ignoring (Philip Withnall)
- #3636 gio/trash does not handle special characters well
- #3641 GResolver: Local DNS resolution failure in offline mode (Patrick
Griffis)
- #3642 `g_cancellable_connect()` documentation incorrect (Marco Trevisan
(Treviño))
- #3643 g_cancellable_connect(): is it safe to unref cancellable from
callback? (Marco Trevisan (Treviño))
- #3649 Crash with some registry key values in GWin32AppInfo (Philip Withnall)
- #3656 Set SYSLOG_IDENTIFIER when logging to journald (Axel Karjalainen)
- #3657 girepository: Wrong typelib path on Windows
- #3663 Cannot use GZlibCompressor in GTK testsuite (Benjamin Otte)
- #3684 UAF in GSignalGroup weak notify callbacks (Thomas Haller)
- #3686 docs.gtk.org doesn't mention that GSourceFuncs.finalize may be NULL
(BZZZZ)
- #3693 Random failures in debian-i386-stable
- !4185 [th/gobject-no-object-locks-pt1-notify] use
`g_datalist_id_update_atomic()` instead of OPTIONAL_BIT_LOCK_NOTIFY
- !4247 mappedfile: Avoid some allocations
- !4369 gdesktopappinfo: Prefer matches that occur earlier in the match string
- !4387 Fix various -Wshorten-64-to-32 warnings
- !4484 Memory sanitizer fixes
- !4489 gobject: Be consistent in using atomic logic to handle the
GParamSpecPool
- !4520 [th/gdataset-cleanup] minor cleanups of gdataset
- !4536 [th/gobj-closure-array-atomic] use g_datalist_id_update_atomic() for
array of closure watches
- !4541 gsettings: Port docs to gi-docgen format, add missing annotations and
make various improvements
- !4544 tests: Don't install runner scripts without installed_tests
- !4545 Update French translation
- !4547 Update Catalan translation
- !4548 Update Turkish translation
- !4551 Updated Danish translation
- !4552 Update Persian translation
- !4553 docs: Document GSignalFlags members added after 2.0
- !4554 Update Indonesian translation
- !4555 tests: Add a test for g_object_freeze_notify() being called too often
- !4557 gfileinfo: Slightly expand docs for
g_file_info_get_attribute_as_string()
- !4558 gi: Dynamically set doc-format
- !4561 tests: Various fixes to create temporary files in /tmp rather than the
build directory
- !4562 gdbusnameowning: Convert docs to gi-docgen linking syntax
- !4563 giounix-private: Fix macro for checking for epoll_create1()
- !4565 Fix LGPL in header
- !4567 gutils: make documentation of g_set_prgname() clearer
- !4568 docs: Add some detail
- !4569 Update Romanian translation
- !4570 gspawn-win32: Fix potential integer overflows in argv handling
- !4571 gvarianttype: Improve docs on type validation
- !4575 gclosure: fix ATOMIC_CHANGE_FIELD to read vint atomically
- !4577 gclosure: Allow full set of closure flags to be queried atomically
- !4578 [th/bit-lock-and-set] bitlock: add g_bit_lock_and_get() and
g_bit_unlock_and_set() API
- !4579 tests: Add missing unistd.h include to scannerapi.c
- !4581 [th/gobj-no-weak-ref-lock] drop OPTIONAL_BIT_LOCK_WEAK_REFS object
lock for `g_object_weak_{ref,unref}()`
- !4583 thread: fix Linux detection
- !4585 gfile: Expand documentation around file info for inaccessible files
- !4586 [th/gobj-doc-weakref] clear #GWeakRef earlier in
g_object_run_dispose() and reword docs about #GWeakRef
- !4588 gstring: carefully handle gssize parameters
- !4590 Various -Wsign-conversion warning fixes
- !4591 gthreadedresolver: fix crash in loopback interface check
- !4592 gstring: Make len_unsigned unsigned
- !4594 Enable -Wsign-conversion for girepository, gthread, gmodule
- !4596 docs: Mention how to run the test suite in CONTRIBUTING.md
- !4598 gtlsconnection: Fix annotation
- !4599 Mark pointer as (type gpointer)
- !4601 garray: Fix annotations
- !4602 docs: fix typo glong: ULONG_MAX -> LONG_MAX
- !4603 Fix GNetworkMonitorNetlink operation under a FreeBSD jail with shared
network stack
- !4604 cocoa: add support for GBytesIcon in notification backend
- !4605 gparamspecs: Use standard min/max constants rather than literals
- !4606 gobject, girepository: Fix several -Wsign-conversion warnings on macOS
- !4609 Update Portuguese translation
- !4610 Update Ukrainian translation
- !4613 Update macOS job for new CI runner
- !4615 shell: Handle empty comment gracefully
- !4619 gslist: Improve documentation for append / prepend / insert methods
- !4620 glocalfile: Disable faccessat()-based query_exists on Android
- !4621 gallocator: mark as deprecated
- !4627 [th/gsignalgroup-dispose] gsignalgroup: make GSignalGroup.dispose() a
bit more reentrant
- !4628 [th/gdataset-fix-zero-key] fix and cleanup related to using a zero
GQuark for keys in GData
- !4631 Update German translation
- !4632 win32: Only print one OS version
- !4633 gzlibcompressor: Convert docs to gi-docgen linking syntax
- !4638 docs: Fix formatting of definition lists
* Translation updates:
- Catalan (Jordi Mas)
- Danish (Ask Hjorth Larsen)
- French (Vincent Chatelain)
- German (Philipp Kiemle)
- Indonesian (Andika Triwidada)
- Persian (Danial Behzadi)
- Portuguese (Hugo Carvalho)
- Romanian (Antonio Marin)
- Turkish (Sabri Ünal)
- Ukrainian (Yuri Chornoivan)
- remove backport patches
Signed-off-by: Markus Volk <f_l_k@t-online.de> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
ptest-cargo.bbclass: sort binaries for reproducibility
When compiling tests with cargo, the produced binaries are created
in a non-deterministic order. The list of binaries themselves are taken from
some of the log info produced by cargo, which contains them in the order
as they were created. The class later writes this list of binaries
in the run-ptest script in the order that it found them.
In case the test suite contains more than 1 or 2 binaries, then the
order of these binaries is different almost each run, making the
resulting ptest package non-reproducible.
To avoid this, sort the list of test binaries before storing them.
BB_CONF_FRAGMENT_SUMMARY = "Use prebuilt sstate artifacts for standard Yocto build configurations."
BB_CONF_FRAGMENT_DESCRIPTION = "The Yocto Project has prebuilt artefacts available for standard build configurations. \
...
GPE_MIRROR is not used by any recipes. Beside this it looks like
the url (http://gpe.linuxtogo.org/download/source) is unavailable,
and the main url (linuxtogo.org) is using an invalid certificate
(issued for another website).
It looks this mirror isn't useful nor valid anymore - so drop it.
grub-efi.bbclass: fix racing issue on grub-efi image
In recipe grub-efi, the deploy task installed grub-efi image to deploy dir.
In image-live.bbclass, the do_bootimg task called build_iso -> efi_iso_populate
-> efi_populate.
In image-live.bbclass, the do_bootimg task called build_hddimg -> efi_hddimg_populate
-> efi_populate.
In grub-efi.bbclass, efi_populate called efi_populate_common for grub-efi to
installed grub-efi image from deploy dir to ISO/HDDDIR.
Explicitly make do_bootimg depend on grub-efi:deploy otherwise a racing issue
on grub-efi image in deploy dir
...
|DEBUG: Executing shell function build_iso
|NOTE: Trying to install tmp/deploy/images/qemux86-64/bzImage as tmp/work/qemux86_64-wrs-linux/
wrlinux-image-installer/1.0/sources/wrlinux-image-installer-1.0/iso/bzImage
|install: cannot stat 'tmp/deploy/images/qemux86-64/grub-efi-bootx64.efi': No such file or directory
|WARNING: tmp/work/qemux86_64-wrs-linux/wrlinux-image-installer/1.0/temp/run.build_iso.954273:419 exit 1 from
'install -m 0644 tmp/deploy/images/qemux86-64/$2-bootx64.efi ${DEST}/EFI/BOOT/bootx64.efi'
...
Yi Zhao [Fri, 5 Sep 2025 06:14:22 +0000 (14:14 +0800)]
libsdl2: upgrade 2.32.8 -> 2.32.10
This is a stable bugfix release, with the following changes:
Fixed the dock occasionally showing on macOS when launching an SDL app
Report the correct device audio format when using Pipewire
Fixed signed 8-bit audio on Android
Support the "ambient" value for SDL_HINT_AUDIO_CATEGORY on iOS
Improved thumbstick range for Nintendo Switch Pro Controllers
Fixed loading user calibration for Nintendo Switch Pro Controllers
Signed-off-by: Yi Zhao <yi.zhao@windriver.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Markus Kurz [Thu, 4 Sep 2025 13:28:20 +0000 (15:28 +0200)]
harfbuzz: upgrade 11.4.1 -> 11.4.5
mainly bugfixes introduced.
for release notes see:
https://github.com/harfbuzz/harfbuzz/releases
Signed-off-by: Markus Kurz <m.kurz92@gmail.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 3 Sep 2025 16:24:11 +0000 (17:24 +0100)]
perf: enable coresight if enabled in MACHINE_FEATURES
Enable the CoreSight[1] functionality in perf[2] if 'coresight' is in the
MACHINE_FEATURES.
Note that enabling this will mean a dependency on the opencsd recipe,
which is currently part of meta-arm[3]. I don't think this is a problem
as this recipe is Arm-specific, and as this is a new feature the
behaviour is opt-in so shouldn't break any existing machines.
The automake before 1.17's mdate-sh couldn't update libcheck's
doc/version.texi, so the doc/check.info wasn't re-generated, then we couldn't
see the build error.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
-qt: Replace icons for dark mode with 16x16 icons
-qt,qt5: Ensure that input field is large enough for generated password
-qt5: Add dark mode icon variants and detection
-Make Show/Hide Password functionality accessible
-qt5: Make Show/Hide Password functionality accessible
-m4: Update libassuan.m4.
-build: Remove defining GPG_ERR_ENABLE_ERRNO_MACROS.
-build: Fix warning about obsolete pinentry-emacs.
-curses: Factor out dialog release to separate function.
-qt: Add dark mode icon variants and detection
-m4: Update gpg-error.m4 and libassuan.m4.
Also: Simplify patch to find gpg-error with pkg-config
Signed-off-by: Gyorgy Sarvari <skandigraun@gmail.com> Cc: Ross Burton <ross.burton@arm.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Dropped patches which are included in this release, or became obsolete:
0001-Fix-compilation-error-with-clang-libc-18.patch - included in this release
0001-Fix-musl-build.patch - included in this release
0001-Raise-cmake_minimum_required-to-3.13-to-avoid-warnin.patch - included in this release
0001-Remove-using-std-binary_function.patch - became obsolete, fixed upstream
0001-aptwebserver.cc-Include-array.patch - became obsolete, fixed upstream
0001-strutl-Add-missing-include-cstdint-gcc-15.patch - included in this release
Added a new patch to avoid compilation error with musl:
error: 'basename' was not declared in this scope; did you mean 'rename'?
Adapted DEPENDS list - gnutls and gcrypt dependencies were dropped in favor of openssl
in version 2.9.19.
Added a new PACKAGECONFIG, 'usrmerge', which displays a gentle warning if the system
isn't a usrmerge system during package installation.
Added new COMMON_ARCH CMake argument - if it is not defined, CMake is trying to
determine the value of this variable by running dpkg, which is usually a futile
endeavour. It is used in config creation, and to print some system info.
Also adapt a self test: the apt-key command has been deprecated since a while,
and in this release it was completely removed. Instead sources.list file
contains the signature data, on a per-repository basis.
wic: add --extra-partition-space option to set unused space
By default, the content of the partition is filled by the filesystem
without leaving any unused free space. The --extra-space flag adds
extra space to the filesystem size, not to the partition.
Unused free space after the filesystem can be useful for some cases,
such as encrypting a partition at runtime.
With --extra-partition-space 32M, we ensure that the last 32M of the
partition is unused: this space does not contain filesystem data and
can store the LUKS2 header.
The implementation sets a difference between the partition and
filesystem size:
- With --fixed-size, the extra part space is removed from the
filesystem size.
- Otherwise (with or without --size flag), the extra part space is
added to the partition size.
Signed-off-by: Pierre-Loup GOSSE <pierre-loup.gosse@smile.fr> CC: Alexander Kanavin <alex.kanavin@gmail.com> CC: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* Drop backported patch and refresh remaining
* Add fixes to build with boost 1.89.0
* Switch to meson build system
* Delete meson-info files
There’s no Meson built-in switch to stop creating intro-*.json
in the build dir (that’s by design), only to stop installing them is
possible. for runtime, Kea doesn’t read it, this is just build metadata
oeqa/selftest/cases/gcc.py: add libitm to default installed packages.
When running oe-selftest few libitm tests were getting failed due
to below error:
/tmp/runtest.923959/alloc-1.exe: error while loading shared libraries:
libitm.so.1: cannot open shared object file: No such file or directory
Adding libitm to default installed packages to fix this issue.
Robert Yang [Wed, 3 Sep 2025 04:16:57 +0000 (21:16 -0700)]
git: 2.50.1 -> 2.51.0
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
- Add site-packages path to pdm info and print interpreter URL in
pdm python install -v.
- Support explicit free-threaded Python selection via version suffix t.
- Multiple lockfile/pylock improvements and hash handling fixes.
- Windows path resolution fixes for uv; CLI help formatting preserved.
- Dependencies: require packaging>22.0; bump truststore to 0.10.4.
- Misc: fix AI coding fast apply error
Tim Orling [Tue, 2 Sep 2025 21:46:28 +0000 (14:46 -0700)]
python3-requests: upgrade 2.32.4 -> 2.32.5
2.32.5 (2025-08-18)
* Bugfixes
- The SSLContext caching feature originally introduced in 2.32.0 has
created a new class of issues in Requests that have had negative
impact across a number of use cases. The Requests team has decided
to revert this feature as long term maintenance of it is proving to
be unsustainable in its current iteration.
* Deprecations
- Added support for Python 3.14.
- Dropped support for Python 3.8 following its end of support.
Release 4.15.0rc1 (August 18, 2025)
* Add the @typing_extensions.disjoint_base decorator, as specified in
PEP 800. Patch by Jelle Zijlstra.
* Add typing_extensions.type_repr, a backport of annotationlib.type_repr,
introduced in Python 3.14 (CPython PR #124551, originally by Jelle
Zijlstra). Patch by Semyon Moroz.
* Fix behavior of type params in typing_extensions.evaluate_forward_ref.
Backport of CPython PR #137227 by Jelle Zijlstra.
Tim Orling [Tue, 2 Sep 2025 21:46:25 +0000 (14:46 -0700)]
python3-maturin: upgrade 1.9.3 -> 1.9.4
1.9.4:
* Add a use-base-python option to pyproject.toml with the same behaviour
as MATURIN_PEP517_USE_BASE_PYTHON.
* Add builtin sysconfigs for GraalPy
* Fix calculation of platform tag for FreeBSD
systemd: stop enabling non-standard MAC policy when using the 'pni-names' DISTRO_FEATURE
This recipe enabled an additional, non-standard policy for predictable
network names: the MAC policy. This is a deviation from upstream
policies and in cases where no MAC address is available, it makes it
actively worse by changing the name(s) every boot.
Change the logic to disable predictable naming when the DISTRO_FEATURE
isn't enabled and only use upstream logic with it *is* enabled.
For DISTROs that wish to use non-standard policies, a DISTRO layer is the correct place.
E.g. in DISTRO_LAYER/recipes-core/systemd/systemd_%.bbappend:
do_install:append:DISTRO() {
if ${@bb.utils.contains('PACKAGECONFIG', 'pni-names', 'true', 'false', d)}; then
if ! grep -q '^NamePolicy=.*mac' ${D}${nonarch_libdir}/systemd/network/99-default.link; then
sed -i '/^NamePolicy=/s/$/ mac/' ${D}${nonarch_libdir}/systemd/network/99-default.link
fi
if ! grep -q 'AlternativeNamesPolicy=.*mac' ${D}${nonarch_libdir}/systemd/network/99-default.link; then
sed -i '/AlternativeNamesPolicy=/s/$/ mac/' ${D}${nonarch_libdir}/systemd/network/99-default.link
fi
}
tiff-native may pick up liblerc from the host using the .so
symlink, in which case it's added as a private requirement
in libtiff-4.pc. That in turn causes a build error in
gtk4-native in Yocto because Lerc.pc does not exist.
This was observed on Fedora 42 with gdk-pixbuf2-devel installed
on the host where libtiff is a dependency of gdk-pixbuf2 and
libtiff 4.7.0 is built with liblerc enabled.
Add a new PACKAGECONFIG setting for liblerc but keep it disabled
since liblerc does not exist in Yocto at the moment.
Signed-off-by: Zoltán Böszörményi <zboszor@gmail.com> Backport-to: walnascar Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
A vulnerability classified as problematic was found in libtiff
4.6.0. This vulnerability affects the function PS_Lvl2page of
the file tools/tiff2ps.c of the component tiff2ps. The
manipulation leads to null pointer dereference. It is possible
to launch the attack on the local host. The complexity of an
attack is rather high. The exploitation appears to be difficult.
The exploit has been disclosed to the public and may be used.
The name of the patch is 6ba36f159fd396ad11bf6b7874554197736ecc8b.
It is recommended to apply a patch to fix this issue. One of the
maintainers explains, that "[t]his error only occurs if
DEFER_STRILE_LOAD (defer-strile-load:BOOL=ON) or TIFFOpen( .. "rD")
option is used."
Markus Kurz [Mon, 1 Sep 2025 21:09:09 +0000 (23:09 +0200)]
libpcre2: upgrade 10.45 -> 10.46
For the changelog see:
https://github.com/PCRE2Project/pcre2/releases/tag/pcre2-10.46
This is a security-only release, to address CVE-2025-58050:
https://nvd.nist.gov/vuln/detail/CVE-2025-58050
Signed-off-by: Markus Kurz <m.kurz92@gmail.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Peter Tatrai [Thu, 28 Aug 2025 10:40:47 +0000 (12:40 +0200)]
connman: Revert removal of no-version-scripts patch
This commit restores the no-version-scripts.patch for ConnMan,
which was removed in the commit (5c269ba9a396832af3f8139ecaa0cfd9f7d4d1b5)
as part of cleaning up gold and MIPS binutils workarounds.
The removal of this patch caused runtime crashes (segfaults)
on the powerpc target due to symbol visibility issues with glib
and stdio. Re-adding the patch resolves these crashes by disabling
version scripts in the linker flags.
Signed-off-by: Peter Tatrai <peter.tatrai.ext@siemens.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
rust-target-config: fix nativesdk-libstd-rs build with baremetal
If TCLIBC='baremetal' is set in local.conf, nativesdk-libstd-rs build fails
with:
| error[E0412]: cannot find type `c_char` in the crate root
| --> /usr/src/debug/libstd-rs/1.75.0/rustc-1.75.0-src/vendor/libc/src/unix/mod.rs:56:29
| |
| 6 | pub type c_schar = i8;
| | ---------------------- similarly named type alias `c_schar` defined here
| ...
| 56 | pub gr_name: *mut ::c_char,
| | ^^^^^^
This happens because rust_gen_target() sets os="none" when TCLIBC is
'baremetal' - even for nativesdk targets. However, nativesdk packages are
built against glibc, so the correct 'os' value should be "linux".
Fix this by setting the os field based on {TARGET,HOST,BUILD}_OS variables,
as it is already done in rust_base_triple(), instead of relying on TCLIBC.
coreutils_9.7.bb: fix update-alternative support for base32
base32 is listed as update-alternative binary but does not
set the alternatie variables for it to take effect causing
the original ELF file to be installed in the destination
instead of the symlink. Fix this by setting the alternative
variables and by moving the original binary into subdir.
Markus Kurz [Mon, 1 Sep 2025 00:13:24 +0000 (02:13 +0200)]
repo: upgrade 2.56 -> 2.58
For changelog, check https://gerrit.googlesource.com/git-repo
Signed-off-by: Markus Kurz <m.kurz92@gmail.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Re-enable DEBUG_BUILDs since the upstream defect:
https://github.com/ColinIanKing/stress-ng/issues/315
is resolved by: 6e62dc1fb 2023-08-31 Makefile: force vnni to be built with at least -O2
Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Gyorgy Sarvari [Fri, 29 Aug 2025 09:18:16 +0000 (11:18 +0200)]
vte: upgrade 0.78.2 -> 0.80.3
0004-fast_float-Add-single-header-library-for-from_char-i.patch
and 0005-color-parser-Use-fast_float-implementation-for-from_.patch
patches dropped: upstream has adopted the changes, and oe-core also
provides now fastfloat, no need to vendor it with a patch.
0002-lib-Typo-fix.patch is dropped, because it was a backport, and
it is included in this release.
Martin Jansa [Fri, 29 Aug 2025 09:04:48 +0000 (11:04 +0200)]
license.py: drop visit_Str
In https://git.openembedded.org/openembedded-core/commit/?h=kirkstone&id=abc93390a3f19bc4cc159c5690a478b9e2270906
visit_Constant was added for compatibility with Python 3.8 and newer, but
visit_Str was kept, because at that time bitbake did require Python 3.6.
Now latest bitbake requires 3.9 and even version 2.4 used in mickledore
builds requires 3.8 since:
https://git.openembedded.org/bitbake/commit/?h=2.4&id=744310f360d2288ac2ef07745abc86852126b5b9
so we can safely remove these.
Signed-off-by: Martin Jansa <martin.jansa@gmail.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Thu, 28 Aug 2025 19:37:50 +0000 (12:37 -0700)]
babeltrace2: Disable ccache explicitly and inherit setuptools3-base
ccache does not work with this package
python plugins use setuptools/distutils to build and it tries to compute
LDCXXSHARED as described in [1] in /src/bindings/python/bt2/setup.py.in
It gets confused and ends up specifying compiler twice and the result is
that linker fails with cryptic errors e.g.
| /mnt/b/yoe/master/sources/poky/build/tmp/work/x86-64-v3-poky-linux-musl/babeltrace2/2.1.1/recipe-sysroot-native/usr/bin/x86_64-poky-linux-musl/../../libexec/x86_64-poky-linux-musl/gcc/x86_64-poky-linux-musl/15.2.0/ld: cannot find x86_64-poky-linux-musl-g++: No such file or directory
| collect2: error: ld returned 1 exit status
| error: command '/mnt/b/yoe/master/sources/poky/build/tmp/hosttools/ccache' failed with exit code 1
| make[3]: *** [Makefile:741: build-python-bindings.stamp] Error 1
| make[3]: Leaving directory '/mnt/b/yoe/master/sources/poky/build/tmp/work/x86-64-v3-poky-linux-musl/babeltrace2/2.1.1/build/src/bindings/python/bt2'
This is because first occurance of linker is encountering second linker on its
cmdline and tries to treat it as an input object to linker
It seems LDCXXSHARED computation gets confused when CXX contains ccache
Gyorgy Sarvari [Thu, 28 Aug 2025 18:17:48 +0000 (20:17 +0200)]
xf86-input-vmmouse, xf86-input-mouse: drop recipes
xf86-input-mouse has dropped Linux support[1], so drop the obsolete
recipe. Only xf86-input-vmmouse is dependent on this recipe, so drop
that also. Could not find recipes depending on on xf86-input-vmmouse.
Yoann Congal [Thu, 28 Aug 2025 18:13:24 +0000 (20:13 +0200)]
rpm-sequoia: remove rc initialisation from run-ptest
With the previous patch "ptest-cargo: move run-ptest rc variable
initialisation", rc is initialised by the ptest-cargo class.
Remove the (now) redundant initialisation and the related comment.
ptest-cargo run-ptest can be generated in two fashions: generated from
scratch or appended to an exiting run-ptest file. The rc variable used
to track tests failure was only initialized in "generated from scratch"
case. Which lead to errors in the "appended" case.
Move the rc variable initialisation to the common code of both case to
fix this problem.
Only initialize rc if it was not already affected in the recipe provided
run-ptest.
|The NumPy 2.3.2 release is a patch release with a number of bug fixes
|and maintenance updates. The highlights are:
|
| Wheels for Python 3.14.0rc1
| PyPy updated to the latest stable release
| OpenBLAS updated to 0.3.30
|
|This release supports Python versions 3.11-3.14
Reproducibility OK:
|2025-08-28 05:50:44,703 - oe-selftest - INFO - ----------------------------------------------------------------------
|2025-08-28 05:50:44,703 - oe-selftest - INFO - Ran 1 test in 1440.103s
|2025-08-28 05:50:44,703 - oe-selftest - INFO - OK
|2025-08-28 05:50:49,531 - oe-selftest - INFO - RESULTS:
|2025-08-28 05:50:49,531 - oe-selftest - INFO - RESULTS - reproducible.ReproducibleTests.test_reproducible_builds: PASSED (1400.47s)
|2025-08-28 05:50:49,724 - oe-selftest - INFO - SUMMARY:
|2025-08-28 05:50:49,724 - oe-selftest - INFO - oe-selftest () - Ran 1 test in 1440.103s
|2025-08-28 05:50:49,724 - oe-selftest - INFO - oe-selftest - OK - All required tests passed (successes=1, skipped=0, failures=0, errors=0)
Khem Raj [Thu, 28 Aug 2025 02:51:34 +0000 (19:51 -0700)]
lttng-tools: Filter out regression testsuite on musl systems
regression testsuite does not run to finish on musl systems, some of
the issues seen
* python is OOM'ing which is fixed by asking for 8G memory for qemu
* It hits the timeout limits of ptest-runner even -t 2400 does not finish it
* Use make -j1 and running run-ptest script directly ( outside
ptest-runner to avoid timeout limit) causes it to run forever
Lets disable this testsuite for musl systems for now.
Khem Raj [Thu, 28 Aug 2025 02:50:11 +0000 (19:50 -0700)]
kea: Upgrade to 2.6.4 release
The following bug fixes have been implemented since the previous stable
release, version 2.6.3:
1. **Syslog logging fix**: We corrected an issue in logging
configuration when parsing "syslog:<facility name>" [#3921, #3992].
2. **Fix for misconfigured global reservations**: Earlier Kea versions
could crash when handling misconfigured global reservations. The problem
is now fixed [#3910, #3911, #4001].
3. **Recent Sphinx support**: Support for recent versions of Sphinx, a
tool used to generate documentation, was added. This allows Kea
documentation and packages to be built on recent systems, while
maintaining compatibility with older versions. This fix was tested with
Sphinx 1.8.4 on Debian 10, 8.1.3 on Fedora 42, and 8.2.3 on ArchLinux
[#3800].
The default CONNECTIVITY_CHECK_URIS uses "https://yoctoproject.org/connectivity.html"
which redirect to "https://www.yoctoproject.org/connectivity.html".
Some network configurations with proxies or restricted internet access
don't handle HTTP redirects properly during the sanity check phase,
causing build failures with:
ERROR: OE-core's config sanity checker detected a potential misconfiguration.
Either fix the cause of this error or at your own risk disable the checker (see sanity.conf).
Following is the list of potential problems / advisories:
Fetcher failure for URL: 'https://yoctoproject.org/connectivity.html'. URL doesn't work.
Updated the default URL to use the final destination directly to avoid
redirect-related connectivity check failures.
Also updated SDK test cases in https.py to use the corrected URL for
consistency.
Signed-off-by: Deepak Rathore <deeratho@cisco.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Peter Marko [Thu, 21 Aug 2025 12:26:55 +0000 (14:26 +0200)]
rust: handle softfp fpu
With TARGET_FPU set to "softfp" stdlib-rs fails with:
error: Error loading target specification: ARM targets must specify their float ABI. Run `rustc --print target-list` for a list of built-in targets
This fpu type is set for example for arm in file
meta/conf/machine/include/arm/feature-arm-vfp.inc
when TUNE_FEATURES do not contain "callconvention-hard",
e.g. when using tune
DEFAULTTUNE = "cortexa8"
Signed-off-by: Peter Marko <peter.marko@siemens.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 27 Aug 2025 06:43:32 +0000 (23:43 -0700)]
libnss-nis: Do not export _nss_nis_getsecretkey when libc does not have xdecrypt
Fixes build errors ( seen with lld particularly )
| x86_64-yoesdk-linux-ld.lld: error: version script assignment of 'NSS_NIS_PRIVATE_2.0' to symbol '_nss_nis_getsecretkey' failed: symbol not defined
| x86_64-yoesdk-linux-clang: error: linker command failed with exit code 1 (use -v to see invocation)
| make[3]: *** [Makefile:642: libnss_nis.la] Error 1
| make[3]: Leaving directory '/mnt/b/yoe/master/build/tmp/work/x86_64-nativesdk-yoesdk-linux/nativesdk-libnss-nis/3.2/build/src'
projects / thirdparty / openembedded / openembedded-core.git / log
