Patrick Vogelaar [Sun, 25 Jan 2026 09:15:40 +0000 (10:15 +0100)]
openssh: add variable for key path
This patch adds a variable for the key directory path. This is especially
useful when working with a read-only file system where you want to
specify the location e.g. on a r/w partition. To be consistent, the
change was also done for the read write path.
For changing the path simply create a bbappend and override the
variable.
Signed-off-by: Patrick Vogelaar <patrick.vogelaar.dev@mailbox.org> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Dmitry Baryshkov [Fri, 23 Jan 2026 16:33:05 +0000 (18:33 +0200)]
linux-firmware: correct wil6210 dependency
The LICENSE.QualcommAtheros_ath10k is already packages as
linux-firmware-ath10k-license package, making OE skip
linux-firmware-qualcommatheros-ath10k-license (because it's now empty)
and thus making linux-firmware-wil6210 uninstallable.
Drop the duplicate licence package and use linux-firmware-ath10k-license
instead.
Trevor Gamblin [Fri, 23 Jan 2026 16:28:58 +0000 (11:28 -0500)]
qemurunner.py: replace 'codecs.open()' with 'open()'
With newer Python versions, codecs.open() is deprecated, leading to the
following warning:
|Stderr:
|/srv/pokybuild/tgamblin-qemux86/openembedded-core/meta/lib/oeqa/utils/qemurunner.py:133: DeprecationWarning: codecs.open() is deprecated. Use open() instead.
| with codecs.open(self.logfile + extension, "ab") as f:
Note that if we try to be explicit and make the 'errors' keyword
'strict' (as the codecs.open() call defaulted to), we see other
warnings:
|ValueError: binary mode doesn't take an errors argument
Theo GAIGE [Fri, 23 Jan 2026 10:46:49 +0000 (11:46 +0100)]
python3-pyelftools: add python3-logging to RDEPENDS
python3-logging is needed as a runtime dependency of python3-pyelftools
else the readelf.py script run into the following error :
```
Traceback (most recent call last):
File "/usr/bin/readelf.py", line 31, in <module>
from elftools.elf.elffile import ELFFile
File "/usr/lib/python3.12/site-packages/elftools/elf/elffile.py", line 29, in <module>
from ..dwarf.dwarfinfo import DWARFInfo, DebugSectionDescriptor, DwarfConfig
File "/usr/lib/python3.12/site-packages/elftools/dwarf/dwarfinfo.py", line 17, in <module>
from .structs import DWARFStructs
File "/usr/lib/python3.12/site-packages/elftools/dwarf/structs.py", line 10, in <module>
from logging.config import valid_ident
ModuleNotFoundError: No module named 'logging'
```
Signed-off-by: Theo GAIGE <tgaige.opensource@witekio.com> Reviewed-by: Louis Rannou <louis.rannou@non.se.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Randolph Sapp [Thu, 22 Jan 2026 19:50:00 +0000 (13:50 -0600)]
bitbake.conf: remove DEBUG_PREFIX_MAP from TARGET_LDFLAGS
Now that the previous bug affecting binary reproducibility has been
addressed [1], we can revert this patch. This will resolve issues with
cgo applications becoming unreprodcible.
Currently go considers link arguments to be sacred, meaning any change
should produce a different binary output. They ensure this by baking
link arguments into the intermediary output, changing the content ID of
that step. As such, the marco prefixes inadvertently end up adding build
paths to the output binary instead of removing them if they are passed
as link arguments to cgo applications.
These paths are later stripped out again, but at this point the content
ID of the dependency has changed and thus the build ID of the end
application will be affected by the cascade of hash changes. See the
upstream bug for more information [2].
Bruce Ashfield [Mon, 26 Jan 2026 15:23:49 +0000 (10:23 -0500)]
linux-yocto/6.12: update to v6.12.66
Updating linux-yocto/6.12 to the latest korg -stable release that comprises
the following commits:
f6044d1fd846 Linux 6.12.66 9e3f8fa53348 bpf: test_run: Fix ctx leak in bpf_prog_test_run_xdp error path f8b406198778 ALSA: hda: intel-dsp-config: Prefer legacy driver as fallback 47e676ce4d68 tpm2-sessions: Fix out of range indexing in name_size 42440155fe27 spi: cadence-quadspi: Prevent lost complete() call during indirect read 3762535fbbc0 scsi: sg: Fix occasional bogus elapsed time that exceeds timeout 0810c8e94d6b ASoC: fsl_sai: Add missing registers to cache default 1d2a10913089 ALSA: hda/realtek: enable woofer speakers on Medion NM14LNL 8072299bf13f ASoC: amd: yc: Add quirk for Honor MagicBook X16 2025 44ed8fae346f ALSA: usb-audio: Update for native DSD support quirks ba6f0d1832ee can: j1939: make j1939_session_activate() fail if device is no longer registered 47206d70d1fb drm/amdkfd: Fix improper NULL termination of queue restore SMI event string 1c06d85c3d51 spi: mt65xx: Use IRQF_ONESHOT with threaded IRQ f609041424d5 drm/amd/display: Fix DP no audio issue 9f6cf07687be ata: libata-core: Disable LPM on ST2000DM008-2FR102 14fa3d1927f1 netfilter: nf_tables: avoid chain re-validation if possible feb28b6827ec powercap: fix sscanf() error return value handling 3835d59f6999 powercap: fix race condition in register_control_type() 7b60aed82db1 net: sfp: extend Potron XGSPON quirk to cover additional EEPROM variant fb9ef40cccdb bpf: Fix reference count leak in bpf_prog_test_run_xdp() 6447e697cfa8 bpf, test_run: Subtract size of xdp_frame from allowed metadata size 6611a73b2991 bpf: Support specifying linear xdp packet data size for BPF_PROG_TEST_RUN 0eb6e9d3b724 bpf: Make variables in bpf_prog_test_run_xdp less confusing 202c5b915e22 bpf: Fix an issue in bpf_prog_test_run_xdp when page size greater than 4K 5c647749bce3 btrfs: fix beyond-EOF write handling afbb57899612 btrfs: use variable for end offset in extent_writepage_io() a915072e5ac3 btrfs: truncate ordered extent when skipping writeback past i_size 7216d78ca34f btrfs: remove btrfs_fs_info::sectors_per_page ced5459df05a btrfs: add extra error messages for delalloc range related errors fb4fa3f9012b btrfs: subpage: dump the involved bitmap when ASSERT() failed 9ed14c3b787b btrfs: fix error handling of submit_uncompressed_range() fcc04c92cbb5 ALSA: ac97: fix a double free in snd_ac97_controller_register() 71138011dc01 ALSA: ac97bus: Use guard() for mutex locks 7388ba6e5ccd erofs: fix file-backed mounts no longer working on EROFS partitions d47b03775d55 erofs: don't bother with s_stack_depth increasing for now dd6ccec088ad arp: do not assume dev_hard_header() does not change skb->head de77d2cd178a net: enetc: fix build warning when PAGE_SIZE is greater than 128K 471dfb97599e net: usb: pegasus: fix memory leak in update_eth_regs_async() 43497313d0da net/sched: sch_qfq: Fix NULL deref when deactivating inactive aggregate in qfq_reset 3d67e8c22685 HID: quirks: work around VID/PID conflict for appledisplay 4969d6fa61af net: netdevsim: fix inconsistent carrier state after link/unlink d916df47a031 idpf: cap maximum Rx buffer size a4212d6732e3 idpf: fix memory leak in idpf_vport_rel() be3d31276199 idpf: keep the netdev when a reset fails 3264881431e3 net: fix memory leak in skb_segment_list for GRO packets 92ff65c660eb riscv: pgtable: Cleanup useless VA_USER_XXX definitions 1eeaaeceafcf btrfs: only enforce free space tree if v1 cache is required for bs < ps cases 6762937a8b45 vsock: Make accept()ed sockets use custom setsockopt() f1029391e604 bnxt_en: Fix potential data corruption with HW GRO/LRO 57f1dd8fa966 net: wwan: iosm: Fix memory leak in ipc_mux_deinit() 9e0f54294fae net/mlx5e: Don't print error message due to invalid module b71d08b96864 netdev: preserve NETIF_F_ALL_FOR_ALL across TSO updates 005671c60fcf net: sock: fix hardened usercopy panic in sock_recv_errqueue 50f65526b33d inet: ping: Fix icmp out counting 03fb1708b7d1 net: mscc: ocelot: Fix crash when adding interface under a lag 269c9283ff7f bridge: fix C-VLAN preservation in 802.1ad vlan_tunnel egress 3950054c9512 net: marvell: prestera: fix NULL dereference on devlink_alloc() failure 26a82dce2bee netfilter: nf_conncount: update last_gc only when GC has been performed e1a436981ac9 netfilter: nf_tables: fix memory leak in nf_tables_newrule() cdafa52ad39b gpio: pca953x: handle short interrupt pulses on PCAL devices d4f333a0155d gpio: pca953x: Add support for level-triggered interrupts dd42e23b6d9e netfilter: nft_synproxy: avoid possible data-race on update operation 16b4508e8717 netfilter: nft_set_pipapo: fix range overlap detection b397bb9c34ac arm64: dts: mba8mx: Fix Ethernet PHY IRQ support 97fdde3189b6 arm64: dts: imx8qm-ss-dma: correct the dma channels of lpuart 35b38dd6a792 arm64: dts: imx8mp: Fix LAN8740Ai PHY reference clock on DH electronics i.MX8M Plus DHCOM 88d60cff3000 ARM: dts: imx6q-ba16: fix RTC interrupt level 3e458210ee2c arm64: dts: add off-on-delay-us for usdhc2 regulator 06b1dfa40090 crypto: qat - fix duplicate restarting msg during AER error 8f6afb166402 arm64: dts: ti: k3-am62-lp-sk-nand: Rename pinctrls to fix schema warnings 75e2bc2985f4 drm/amd/display: Apply e4479aecf658 to dml 163df8d79a0d drm/amd/display: Respect user's CONFIG_FRAME_WARN more for dml files 582ba48e4a4c btrfs: fix NULL dereference on root when tracing inode eviction ec3695dd0acd btrfs: tracepoints: use btrfs_root_id() to get the id of a root 53df7a4c3a56 btrfs: qgroup: update all parent qgroups when doing quick inherit b9b19fecad7d btrfs: fix qgroup_snapshot_quick_inherit() squota bug dedec6e6b421 scsi: Revert "scsi: libsas: Fix exp-attached device scan after probe failure scanned in again after probe failed" 496ca70a15f4 scsi: ufs: core: Fix EH failure after W-LUN resume error f373695d62e0 scsi: ipr: Enable/disable IRQD_NO_BALANCING during reset 261233f765d2 smb/client: fix NT_STATUS_NO_DATA_DETECTED value 596d1b968660 smb/client: fix NT_STATUS_DEVICE_DOOR_OPEN value b1dd68601676 smb/client: fix NT_STATUS_UNABLE_TO_FREE_VM value e83af97d5c39 drm/amd/display: shrink struct members a8559efcd576 NFS: Fix up the automount fs_context to use the correct cred 71029266093b ASoC: rockchip: Fix Wvoid-pointer-to-enum-cast warning (again) e1df03e293a0 NFSv4: ensure the open stateid seqid doesn't go backwards 6a1099604b0c dm-snapshot: fix 'scheduling while atomic' on real-time kernels f002df3a3305 alpha: don't reference obsolete termio struct for TC* constants 1b645cd729ef ARM: 9461/1: Disable HIGHPTE on PREEMPT_RT kernels 7010683101b5 csky: fix csky_cmpxchg_fixup not working 32dc49f49ea0 drm/xe: Ensure GT is in C0 during resumes e724d0261b7c drm/xe: make xe_gt_idle_disable_c6() handle the forcewake internally f09cd209359a tls: Use __sk_dst_get() and dst_dev_rcu() in get_netdev_for_sock(). 6f468f6ff233 libceph: make calc_target() set t->paused, not just clear it e94075e950a6 libceph: reset sparse-read state in osd_fault() d2c4a5f69966 libceph: return the handler error from mon_handle_auth_done() c4c2152a858c libceph: make free_choose_arg_map() resilient to partial allocation 6c6cec3db3b4 libceph: replace overzealous BUG_ON in osdmap_apply_incremental() 2802ef3380fa libceph: prevent potential out-of-bounds reads in handle_auth_done() f94f95b81736 wifi: mac80211: restore non-chanctx injection behaviour 024f71a57d56 wifi: avoid kernel-infoleak from struct iw_point fcb7500bfa24 pinctrl: qcom: lpass-lpi: mark the GPIO controller as sleeping 321e17ff3142 gpio: rockchip: mark the GPIO controller as sleeping 7500ab83bad2 drm/radeon: Remove __counted_by from ClockInfoArray.clockInfo[] bc96db7051cc drm/pl111: Fix error handling in pl111_amba_probe 90b4b130a20d drm/amdgpu: Fix query for VPE block_type and ip_count 49a66829dd36 counter: interrupt-cnt: Drop IRQF_NO_THREAD flag c61440f1e741 counter: 104-quad-8: Fix incorrect return value in IRQ handler 196e8fd7424b lib/crypto: aes: Fix missing MMU protection for AES S-box 97130283b83f mei: me: add nova lake point S DID 0c2413c69129 btrfs: always detect conflicting inodes when logging inode refs 8b402146e3a8 arm64: Fix cleared E0POE bit after cpu_suspend()/resume() 2f05f7737e16 net: 3com: 3c59x: fix possible null dereference in vortex_probe1() 1320d94a4df1 atm: Fix dma_free_coherent() size 3f5d7f3865c6 NFSD: Remove NFSERR_EAGAIN 8c1cf63ed465 NFSD: net ref data still needs to be freed even if net hasn't startup d95499900fe5 nfsd: check that server is running in unlock_filesystem 03c68f94fad1 nfsd: use correct loop termination in nfsd4_revoke_states() ba4811c8b433 nfsd: provide locking for v4_end_grace 6b7ad17f4dd5 NFSD: Fix permission check for read access to executable-only files
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Thu, 22 Jan 2026 15:34:38 +0000 (15:34 +0000)]
gpgme: ensure manpage generation is deterministic
In a similar way to the previous commits with pod2man, gpgme will not
generate manpages on a clean build but will on rebuilds due to
do_package -> rpm-native -> ... -> libgpg-error-native, where the
libgpg-error-native recipe provides the yat2m tool.
To ensure that we generate manpages deterministicly, depend on this
recipe.
Signed-off-by: Ross Burton <ross.burton@arm.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Thu, 22 Jan 2026 15:34:34 +0000 (15:34 +0000)]
perl: provide pod2man
Many recipes uses pod2man to generate manpages from an almost
human-readable source format, which is part of the perl recipe.
This means that we have recipes that don't install manpages, or more
accurately don't install manpages if built from clean but _do_ if they
are a rebuild (because do_package -> rpm-native -> perl-native means the
sysroot now has pod2man in).
The obvious fix here is to DEPEND on perl-native but that an often look
like a redundant dependency that can be removed as removing it doesn't
cause problems (I'm fairly confident I'm responsible for patches like
this).
So, add a PROVIDES of pod2man to perl, so that recipes can DEPEND on
pod2man-native and this dependency is both obvious as to it's purpose
and easily removed if the manpages change source format.
Signed-off-by: Ross Burton <ross.burton@arm.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
453e6b8dba (HEAD, origin/release/2.42/master) resolv: Fix NSS DNS backend for getnetbyaddr (CVE-2026-0915) b0ec8fb689 memalign: reinstate alignment overflow check (CVE-2026-0861) f122d0b4d1 nptl: Optimize trylock for high cache contention workloads (BZ #33704) a1d3294a5b support: Exit on consistency check failure in resolv_response_add_name 8dfb84ad4e support: Fix FILE * leak in check_for_unshare_hints in test-container 2a0873aa81 sprof: fix -Wformat warnings on 32-bit hosts efdf4c0c87 sprof: check pread size and offset for overflow b11411fe2e posix: Fix invalid flags test for p{write,read}v2 8aaf4b732d ppc64le: Power 10 rawmemchr clobbers v20 (bug #33091) 2dbf973fe0 ppc64le: Restore optimized strncmp for power10 6b2957cfe8 ppc64le: Restore optimized strcmp for power10 828b8d23f3 AArch64: Fix and improve SVE pow(f) special cases 710d7a2e83 AArch64: fix SVE tanpi(f) [BZ #33642] 0c9430ed97 AArch64: Fix instability in AdvSIMD sinh ec041b1f53 AArch64: Fix instability in AdvSIMD tan 97297120ce AArch64: Optimise SVE scalar callbacks 17c3eab387 aarch64: fix includes in SME tests de1fe81f47 aarch64: fix cfi directives around __libc_arm_za_disable bf499c2a49 x86: fix wmemset ifunc stray '!' (bug 33542) 71874f167a aarch64: tests for SME 256030b984 aarch64: clear ZA state of SME before clone and clone3 syscalls 6de12fc9ad aarch64: define macro for calling __libc_arm_za_disable ab8c1b5d62 x86: Detect Intel Nova Lake Processor bf48b17a28 x86: Detect Intel Wildcat Lake Processor 18fd689cdc nptl: Fix MADV_GUARD_INSTALL logic for thread without guard page (BZ 33356) 46b4e37c9e nss: Group merge does not react to ERANGE during merge (bug 33361) 1166170d95 libio: Define AT_RENAME_* with the same tokens as Linux
Signed-off-by: Hemanth Kumar M D <Hemanth.KumarMD@windriver.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
mark.yang [Thu, 22 Jan 2026 05:51:43 +0000 (14:51 +0900)]
lto.inc: add clang specific LTO configuration
This commit adds support for Clang-specific LTO options.
1. -fuse-linker-plugin is not supported by Clang.
2. -ffat-lto-objects is used to ensure that object files (.o) are generated as ELF
even when LTO is applied. During the dwarfsrcfiles process, if LTO is used
with the Clang toolchain, static library object files are skipped if they are
LLVM bitcode instead of ELF. Therefore, it is recommended to add this to
LTOEXTRA if necessary.
3. Adding 'thin-lto' to DISTRO_FEATURES enables -flto=thin.
Tim Orling [Wed, 21 Jan 2026 21:07:19 +0000 (13:07 -0800)]
oeqa/runtime: Update tests for maturin
* The output from maturin has changed in newer releases.
* Bump guessing-game version to 0.3.0
* Update to abi3 Python3 >= 3.9 support.
NOTE: The "maturin develop" step builds around 45 crates
and needs enough RAM to run. You will also probably want
the performance of KVM. For QEMU/testimage, you will
want the following in local.conf (or a similar .conf file):
QEMU_USE_KVM = 'True'
QB_MEM = '-m 2048'
Signed-off-by: Tim Orling <tim.orling@konsulko.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
pyproject.toml:
* restrict maturin to >=1.7 to ensure PyO3 27.0+ and python 3.13+ support
src/lib.rs:
* rand 0.9 API change: rand::thread_rng() was removed. Use rand::rng() or
the convenience function rand::random_range().
* PyO3 0.27 has breaking API changes from 0.21. The #[pymodule] function
signature changed from fn module_name(py: Python, m: &PyModule) to
fn module_name(m: &Bound<'_, PyModule>).
* 1.11.5
- Allow combining --compatibility pypi with other --compatibility values (#2928)
* 1.11.4
- Support armv6l and armv7l in pypi compatibility (#2926)
- Improve the reliability of maturin's own CI
* 1.11.3
- Fix manylinux2014 compliance check (#2922)
* 1.11.2
- Fix failed release
* 1.11.1
- Fix compiled artifacts being excluded by source path matching (#2910)
- Better error reporting for missing interpreters (#2918)
- Ignore unreadable excluded directories (#2916)
* 1.11.0 - Yanked
- Note: This release was yanked to a regression: #2909
- Refactor ModuleWriter to be easier to implement and use
- Add Android cross compilation support, fix wheel tags for Android
- Update generate-ci to macos-15-intel and add windows arm support
- Deprecate 'upload' and 'publish' CLI commands
Ulrich Ölmann [Fri, 16 Jan 2026 11:56:19 +0000 (12:56 +0100)]
initramfs-framework: add handover of PID 1's arguments
Although many PID 1 programs parse /proc/cmdline, let's follow the standard
convention and forward the command-line arguments received from the kernel to
the next PID 1 program.
Ricardo Ungerer [Tue, 20 Jan 2026 19:47:21 +0000 (19:47 +0000)]
yocto-check-layer: Fix README email check
So far the test_readme have been use of re.match to find an email
address in the README file. This only matches if the email address
is at the start of the file. This commit changes this to re.search to
find email addresses anywhere in the README file.
Signed-off-by: Ricardo Ungerer <ungerer.ricardo@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ricardo Ungerer [Tue, 20 Jan 2026 19:47:20 +0000 (19:47 +0000)]
yocto-check-layer: Add messages in test_readme assertions
Assertions in test_readme does not provide context message when they
fail. Which leads to errors like:
File "/media/workspace/yocto_master/openembedded-core/scripts/lib/checklayer/cases/common.py", line 41, in test_readme
self.assertTrue(email_regex.match(data))
AssertionError: None is not true
This patch adds context messages to the assertions to help identify the
issue when they fail.
Signed-off-by: Ricardo Ungerer <ungerer.ricardo@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Chen Qi [Thu, 15 Jan 2026 02:19:17 +0000 (10:19 +0800)]
systemd: do not let do_fetch depend on PACKAGECONFIG
It's unnecessary to make SRC_URI contains something like:
${@bb.utils.contains('PACKAGECONFIG', 'xxx', 'file://xxx', '', d)}
This does not give us any benefit and it makes do_fetch depend
on PACKAGECONFIG, which means changing of PACKAGECONFIG will result
in rerunn of do_fetch.
Besides, the related codes in do_install already does the necessary
checks.
Peter Marko [Thu, 15 Jan 2026 00:24:20 +0000 (01:24 +0100)]
libpng: upgrade 1.6.53 -> 1.6.54
Handles CVE-2026-22695 and CVE-2026-22801.
License-Update: copyright years refreshed
Signed-off-by: Peter Marko <peter.marko@siemens.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Peter Marko [Thu, 15 Jan 2026 23:50:40 +0000 (00:50 +0100)]
go: upgrade 1.25.5 -> 1.25.6
Upgrade to latest 1.25.x release [1]:
$ git --no-pager log --oneline go1.25.5..go1.25.6 69801b25b9 (tag: go1.25.6) [release-branch.go1.25] go1.25.6 9d497df196 [release-branch.go1.25] archive/zip: reduce CPU usage in index construction afa9b66ac0 [release-branch.go1.25] net/url: add urlmaxqueryparams GODEBUG to limit the number of query parameters 2526187481 [release-branch.go1.25] cmd/go/internal/work: sanitize flags before invoking 'pkg-config' 082365aa55 [release-branch.go1.25] cmd/go: update VCS commands to use safer flag/argument syntax 4be38528a6 [release-branch.go1.25] crypto/tls: don't copy auto-rotated session ticket keys in Config.Clone 525dd85363 [release-branch.go1.25] crypto/tls: reject trailing messages after client/server hello ddcf27fc8c [release-branch.go1.25] Revert "errors: optimize errors.Join for single unwrappable errors" 14f50f6e3e [release-branch.go1.25] cmd/compile: handle propagating an out-of-range jump table index 4e531b2f14 [release-branch.go1.25] runtime: mark getfp as nosplit 6f07a57145 [release-branch.go1.25] runtime/race: set missing argument frame for ppc64x atomic And/Or wrappers ea603eea37 [release-branch.go1.25] os: allow direntries to have zero inodes on Linux 93f5d1c27e [release-branch.go1.25] os,internal/poll: don't call IsNonblock for consoles and Stdin d5bfdcbc47 [release-branch.go1.25] crypto/tls: use inner hello for earlyData when using QUIC and ECH
Fixes CVE-2025-61728, CVE-2025-61726, CVE-2025-68121, CVE-2025-61731,
CVE-2025-68119 and CVE-2025-61730.
Signed-off-by: Peter Marko <peter.marko@siemens.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Dmitry Baryshkov [Thu, 15 Jan 2026 21:26:40 +0000 (23:26 +0200)]
mesa: drop VDPAU remnants
Mesa 25.3.x dropped support for VDPAU. Commit 88e26a937549 ("mesa:
upgrade 25.2.5 -> 25.3.1") has dropped most of VDPAU-related items,
but it didn't drop mesa-vdpau-drivers package (nor did it drop VDPAU
bits from mesa-dev). Drop those remnants.
Kai Kang [Thu, 15 Jan 2026 07:54:36 +0000 (15:54 +0800)]
qemu: 10.1.3 -> 10.2.0
Upgrade qemu from 10.1.3 to 10.2.0:
* remove backported 0012 patch
* update context for patches 0002 and 0010
Signed-off-by: Kai Kang <kai.kang@windriver.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The ' is stripped from the ID later in distro_identifier with:
# Filter out any non-alphanumerics and convert to lowercase
distro_id = re.sub(r'\W', '', distro_id).lower()
but not from version which results in a weird NATIVELSBSTRING like:
NATIVELSBSTRING = "gentoo-'2.18'"
And similarly the directory name in sstate-cache:
oe-core $ ls -d sstate-cache/gentoo-*
"sstate-cache/gentoo-'2.18'" sstate-cache/gentoo-2.18
Signed-off-by: Martin Jansa <martin.jansa@gmail.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Thomas Perrot [Fri, 16 Jan 2026 14:31:21 +0000 (15:31 +0100)]
opensbi: bump to 1.8.1
This intermediate release has following fixes:
- Typo fix in comments of sbi_hartindex_to_hartid()
- Fix hart protection abstraction for platforms without PMP
Allow the platform to boot without PMP, but it is still necessary to
enable PMP on QEMU; otherwise, there will be nothing to protect the
M-mode firmware (OpenSBI) from lower privilege modes.
Signed-off-by: Thomas Perrot <thomas.perrot@bootlin.com> Tested-by: Michael Opdenacker <michael.opdenacker@rootcommit.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Fri, 16 Jan 2026 17:41:28 +0000 (17:41 +0000)]
alsa-utils: fix packaging
In the previous refactoring of this package[1] I forgot to add provides
when merging some packages, which meant that installation of alsa-utils
wouldn't succeed:
- nothing provides alsa-utils-aplaymidi needed by alsa-utils-1.2.15.1-r0.cortexa57 from oe-repo
- nothing provides alsa-utils-aplaymidi2 needed by alsa-utils-1.2.15.1-r0.cortexa57 from oe-repo
- nothing provides alsa-utils-arecordmidi needed by alsa-utils-1.2.15.1-r0.cortexa57 from oe-repo
- nothing provides alsa-utils-arecordmidi2 needed by alsa-utils-1.2.15.1-r0.cortexa57 from oe-repo
- nothing provides alsa-utils-axfer needed by alsa-utils-1.2.15.1-r0.cortexa57 from oe-repo
This is because the auto-splitting generates a package for each of the
binaries, but there are explicit FILES that take priority which result
in the binaries being in another package. However, the dependencies
still exist. We could filter the package list after the split, but it is
easier to add RPROVIDES for arecord and a*midi to the relevant packages,
so that the dependencies are satisfied.
Also axfer is a separate tool, so put that in its own package instead of
merging with aplay.
Finally, add a PR, as it appears that dynamically-generated packaging
isn't enough to change the hash, and the wrong packages were being pulled
from sstate.
Signed-off-by: Ross Burton <ross.burton@arm.com> Tested-by: Ryan Eatmon <reatmon@ti.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Peter Marko [Fri, 16 Jan 2026 19:39:21 +0000 (20:39 +0100)]
zlib: ignore CVE-2026-22184
This is CVE for example tool contrib/untgz.
This is not compiled in Yocto zlib recipe.
This CVE has controversial CVSS3 score of 9.8.
Signed-off-by: Peter Marko <peter.marko@siemens.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Fri, 16 Jan 2026 18:05:01 +0000 (18:05 +0000)]
fontconfig: backport a patch to fix a build race
There's a build race where some objects don't explicitly depend on the
generated header ftalias.h being created:
In file included from ../sources/fontconfig-2.17.1/src/fcpat.c:23:
../sources/fontconfig-2.17.1/src/fcint.h:129:12: fatal error: fcalias.h: No such file or directory
129 | # include "fcalias.h"
Backport a patch from upstream to add the missing dependencies.
[ YOCTO #16134 ]
Signed-off-by: Ross Burton <ross.burton@arm.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This corresponds to an invalid argument passed to lseek() in the
create_hole() function in src/copy-file-data.c in coreutils. This in
turn is being given bad arguments due to incorrect handling of loop
variables in lseek_copy().
The upstream bug report [1] refers to an infinite loop, which isn't
quite what we're seeing, but the cause is likely the same. So, we're
applying the upstream fixes.
Richard Purdie [Tue, 20 Jan 2026 08:52:27 +0000 (08:52 +0000)]
pseudo: Update to 1.9.3 release
Pulls in the following changes:
Makefile.in: Bump version to 1.9.3
configure: Minor code quality changes
pseudo: code quality scan - resolved various potential issues
makewrappers: improve error handling and robustness
Update COPYRIGHT files
ports/linux/pseudo_wrappers.c: Call the wrappers where possible
ports/linux/pseudo_wrappers.c: Workaround compile error on Debian 11
ports/linux/pseudo_wrappers.c: Reorder the syscall operations
ports/unix/guts/realpath.c: Fix indents
pseudo_util.c: Skip realpath like expansion for /proc on Linux
test/test-proc-pipe.sh: Add test case for proc pipes
ports/unix/guts/realpath.c: realpath fails if the resolved path doesn't exist
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Hatle [Sat, 17 Jan 2026 20:57:11 +0000 (14:57 -0600)]
dpkg: Fix ADMINDIR
dpkg has a hard coded path (from build time) for the ADMINDIR, for some
reason the "set_root" function was using this hard coded value instead
of the value from apt.conf or the environment.
Follow the example of db_dir.c and use the environment if set.
Adjust the matching oe package_manager functions to set the ADMINDIR,
even though the apt.conf sets --admindir. Note it's unclear if the
--admindir value that is set is reasonable or not.
Signed-off-by: Mark Hatle <mark.hatle@kernel.crashing.org>
[AG: add Upstream-Status - Submitted to patch] Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ryan Eatmon [Thu, 8 Jan 2026 20:22:38 +0000 (14:22 -0600)]
u-boot: Add support for UBOOT_CONFIG_FRAGMENTS and UBOOT_FRAGMENTS
There are config fragments that live in the u-boot source tree.
UBOOT_FRAGMENTS will globally add a set of fragments to the configure.
UBOOT_CONFIG_FRAGMENTS makes use of the overhauled config flow to add
support for specifying alternative fragments for each variation build.
Signed-off-by: Ryan Eatmon <reatmon@ti.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ryan Eatmon [Thu, 8 Jan 2026 20:22:37 +0000 (14:22 -0600)]
u-boot: Overhaul UBOOT_CONFIG flow
The existing flow for setting UBOOT_CONFIG options is a little
confusing. The comma separated list is difficult to remember which
slot is used for which option, and the order of the options cannot
be changed just added on to.
This commit overhauls the UBOOT_CONFIG flow while maintaining the
backwards compatibility with the existing settings. Rather than use a
comma separated list, a new set of placeholder variables have been
defined where the same config flags can be used to specify all of the
needed changes:
Adam Blank [Wed, 31 Dec 2025 11:21:01 +0000 (12:21 +0100)]
buildhistory: fix handling of RDEPENDS style strings
RDEPENDS style strings are regex searched for entries matching
the usual, allowed package names (plus possible version flags).
RPROVIDES however, is also allowed to contain actual file paths
like '/bin/sh', which get broken by the current implementation.
Use bb.utils to handle RDEPENDS style stings.
Signed-off-by: Adam Blank <adam.blank.g@gmail.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Dragomir, Daniel [Wed, 17 Dec 2025 14:27:26 +0000 (16:27 +0200)]
oeqa/selftest/wic: test recursive dir copy on ext partitions
Extend the wic selftests to cover recursive directory copying
into ext partitions.
Previously, copying a directory into an ext partition could
appear to succeed, but attempting to access the directory
contents would fail with:
-l: Ext2 inode is not a directory
This was fixed in commit 4fc3b42774 ("wic/engine: fix copying
directories into wic image with ext* partition").
This test now verifies that directories copied with "wic cp"
into an ext4 partition:
- are created with correct inode types
- can be listed recursively with "wic ls"
- preserve files and subdirectories
- can be copied back out of the image without data loss
A simple directory structure is used in this test:
wic/engine: fix copying directories into wic image with ext* partition
wic uses debugfs to write on ext* partitions, but debugfs can only
write to the current working directory and it cannot copy complete
directory trees. Running 'wic ls' on a copied directory show this:
-l: Ext2 inode is not a directory
Fix this by creating a command list for debugfs (-f parameter) when
recursive parsing the host directory in order to create a similar
directory structure (mkdir) and copy files (write) on each level
into the destination directory from the wic's ext* partition.
Signed-off-by: Daniel Dragomir <daniel.dragomir@windriver.com> Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Alex Bradbury [Wed, 14 Jan 2026 10:26:11 +0000 (10:26 +0000)]
runqemu-extract-sdk: support tar.zst format
Building `core-image-minimal-dev` with Poky Whinlatter, the rootfs
tarball is compressed with zstd by default, meaning this script won't
work out of the box with it.
Signed-off-by: Alex Bradbury <asb@igalia.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Tue, 13 Jan 2026 18:46:32 +0000 (18:46 +0000)]
systemd-compat-units: generate masks at build time
Originally, this recipe masked the units at rootfs-time by looking to see
if the init script to be masked existed, and masking it if so.
That however suffers from a race condition where the postinst script may
be executed before the package that provides a script has been installed
so this was changed to be an on-target postinst[1], so we know that all
of the packages have been installed.
However, this now means that you can't have an image that install this
package _and_ is read-only, because we have an on-target postinst to
execute.
There's also another problem that with on-target package management, it
would be possible to install a package that should be masked after the
postinst has been executed, so the mask won't exist.
Taking a step back, a systemd unit mask is just a symlink to /dev/null
with the right name in the right path. So instead of masking in a post-
install script, we can generate all the masks at build-time.
This change reduces this package to just a number of symlinks, with no
post-install scripts.
Trevor Gamblin [Tue, 13 Jan 2026 17:57:58 +0000 (12:57 -0500)]
python3: add new files to compression
As of Python 3.14, there is a new top-level compression module that
contains the gzip and tarfile modules from previous versions, along with
a new _common module. However, for compatibility reasons the project is
also leaving a copy of the gzip and tarfile modules in the original
positions (to be handled in a future deprecation cycle).
Trevor Gamblin [Tue, 13 Jan 2026 17:57:57 +0000 (12:57 -0500)]
python3: add _ast_unparse to core
Without this, we will see import errors for modules that use
annotationlib like so:
|../../python3.14/annotationlib.py:487: in __repr__\n'
| return ast.unparse(self.__ast_node__)\n'
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n'
|../../python3.14/ast.py:622: in unparse\n'
| from _ast_unparse import Unparser as _Unparser\n'
|E ModuleNotFoundError: No module named '_ast_unparse'\n"
Trevor Gamblin [Tue, 13 Jan 2026 17:57:56 +0000 (12:57 -0500)]
python3: add _py_warnings, annotationlib to core
Without adding these, there will be numerous packages that will have to
add python3-misc to their RDEPENDS just to dispel some warnings, like:
|root@qemux86-64:~# python3
|Python 3.14.2 (main, Dec 5 2025, 16:49:16) [GCC 15.2.0] on linux
|Type "help", "copyright", "credits" or "license" for more information.
|warning: can't use pyrepl: No module named 'annotationlib'
and errors like:
|>>> import logging
|Traceback (most recent call last):
| File "<stdin-1>", line 1, in <module>
| File "/usr/lib/python3.14/logging/__init__.py", line 26, in <module>
| import sys, os, time, io, re, traceback, warnings, weakref, collections.abc
| File "/usr/lib/python3.14/traceback.py", line 8, in <module>
| import warnings
| File "/usr/lib/python3.14/warnings.py", line 15, in <module>
| from _py_warnings import (
|ModuleNotFoundError: No module named '_py_warnings'
Add them to python3-core rather than requiring another split.
Trevor Gamblin [Tue, 13 Jan 2026 17:57:54 +0000 (12:57 -0500)]
python_pep517: add --prefix to nativepython3 call
With Python 3.14, the default assumptions about install path in
python_pep517 no longer seem to apply, and as a result we see QA errors
during Python package builds like:
|ERROR: python3-unittest-automake-output-0.3-r0 do_package: QA Issue: python3-unittest-automake-output: Files/directories were installed but not shipped in any package:
| /home
| /home/tgamblin
| /home/tgamblin/workspace
| /home/tgamblin/workspace/yocto
| /home/tgamblin/workspace/yocto/openembedded-core
| /home/tgamblin/workspace/yocto/openembedded-core/build
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao-0.3.dist-info
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/__init__.py
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/pytest.py
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/unittest
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/__pycache__
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/unittest/__init__.py
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/unittest/__main__.py
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/unittest/__pycache__
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/unittest/__pycache__/__init__.cpython-314.pyc
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/unittest/__pycache__/__main__.cpython-314.pyc
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/__pycache__/__init__.cpython-314.pyc
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao/__pycache__/pytest.cpython-314.pyc
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao-0.3.dist-info/entry_points.txt
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao-0.3.dist-info/WHEEL
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao-0.3.dist-info/METADATA
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao-0.3.dist-info/licenses
| /home/tgamblin/workspace/yocto/openembedded-core/build/tmp/work/x86-64-v3-poky-linux/python3-unittest-automake-output/0.3/recipe-sysroot-native/usr/lib/python3.14/site-packages/putao-0.3.dist-info/licenses/LICENSE
|Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them within do_install.
|python3-unittest-automake-output: 34 installed and not shipped files. [installed-vs-shipped]
We need to add the --prefix flag to the nativepython3 call so that it
avoids using the native Python's sysconfig prefix.
As of 3.14.0, Python's sysconfig module looks for _PYTHON_PROJECT_BASE
to be set when cross-compiling, or otherwise it'll find and use the
native interpreter's paths when calling get_path() and filling out
_CONFIG_VARS. Set _PYTHON_PROJECT_BASE to "${B}" in
python3targetconfig's setup_target_config() to help avoid issues with
incorrect paths in rpm and other recipes that depend on this class.
- Remove '0001-Generate-data-for-OpenSSL-3.4-and-add-it-to-multissl.patch'
as this fix was submitted upstream and is included in 3.14.0.
- Modify three existing patches to apply on 3.14.0:
- 0001-Lib-sysconfig.py-use-prefix-value-from-build-configu.patch
- 0001-Skip-failing-tests-due-to-load-variability-on-YP-AB.patch
- 0001-test_active_children-skip-problematic-test.patch
- Add three new ptest skip patches:
- 0001-test_cmd-skip-bang-completion-test.patch - editline issues
with tab completion off of bangs (!).
- 0001-test_pyrepl-skip-test_unix_console.test_cursor_back_.patch -
change in curses dependency in PyREPL leads to issues with a test
that expects the ICH1 terminal capability.
- 0001-test_sysconfig-skip-test_sysconfig.test_sysconfigdat.patch -
OE builds customize prefixes and compiler flags in a way that the
test doesn't expect.
- Add zstd to the DEPENDS since Python3 now provides support for the
Zstandard compression standard.
- In py_package_preprocess(), handle reproducibility issues with the
_sysconfig_vars*.json files that the build generates by stripping
paths from them.
Modify python3-manifest.json to reflect the removal/replacement of
stringold with a new string module directory.
Also update python3-dir Python version to 3.14.
Reproducibility looks OK:
|2025-12-03 17:23:39,179 - oe-selftest - INFO - Reproducibility summary for deb: same=2261 different=0 different_excluded=0 missing=0 total=2261
|unused_exclusions=[]
|2025-12-03 17:23:39,179 - oe-selftest - INFO - Checking ipk packages for differences...
|2025-12-03 17:23:41,631 - oe-selftest - INFO - Reproducibility summary for ipk: same=2261 different=0 different_excluded=0 missing=0 total=2261
|unused_exclusions=[]
|2025-12-03 17:23:41,631 - oe-selftest - INFO - Checking rpm packages for differences...
|2025-12-03 17:23:43,952 - oe-selftest - INFO - Reproducibility summary for rpm: same=2261 different=0 different_excluded=0 missing=0 total=2261
|unused_exclusions=[]
|2025-12-03 17:23:43,952 - oe-selftest - INFO - ... ok
|2025-12-03 17:24:32,829 - oe-selftest - INFO - ----------------------------------------------------------------------
|2025-12-03 17:24:32,830 - oe-selftest - INFO - Ran 1 test in 1616.811s
|2025-12-03 17:24:32,830 - oe-selftest - INFO - OK
|2025-12-03 17:24:36,170 - oe-selftest - INFO - RESULTS:
|2025-12-03 17:24:36,170 - oe-selftest - INFO - RESULTS - reproducible.ReproducibleTests.test_reproducible_builds: PASSED (1567.51s)
|2025-12-03 17:24:38,017 - oe-selftest - INFO - SUMMARY:
|2025-12-03 17:24:38,017 - oe-selftest - INFO - oe-selftest () - Ran 1 test in 1616.811s
|2025-12-03 17:24:38,018 - oe-selftest - INFO - oe-selftest - OK - All required tests passed (successes=1, skipped=0, failures=0, errors=0)
1.Changelog
- Initial 64-bit RISC-V support (Maksymilian Knust, Filip Wasil)
- Add 64-bit LoongArch support (Jin Bo, Hecai Yuan)
- Implement release and reuse of temporary registers for some targets (Jorge Zapata)
- x86: Implement EVEX encoding and an opcode validation system (Jorge Zapata)
- x86: Opcode refactor, improved constant handling and various other fixes (Jorge Zapata)
- x86: add missing rounding operands for AVX and SSE (L. E. Segovia)
- x86: Implement 64-bit single move constant load (L. E. Segovia)
- includes: stop exporting the private compiler and OrcTarget definitions (Jorge Zapata)
- Use hotdoc instead of gtk-doc to generate the documentation (Jorge Zapata)
- ORC_DEBUG_FATAL environment variable allows abort on log messages of a certain level (Carlos Falgueras García)
- Error message improvements and NEON backend clean-ups (Alexander Slobodeniuk)
- Fix a few valgrind issues (Doug Nazar)
- Build: enable tools such as orcc and orc-bugreport by default (Nirbheek Chauhan)
- Various build fixes (L. E. Segovia, Sebastian Dröge)
2.Documentation are no longer built with gtk_doc, it was replaced by hotdoc.
https://gitlab.freedesktop.org/gstreamer/orc/-/commit/0ad1f5884c654b614cf1ddd114a7ec9140311a3a
This means documentation will no longer be built in the recipe or packaged in
the output packages.
Signed-off-by: Liu Yiding <liuyd.fnst@fujitsu.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Drop backport fix-musl-compilation.patch which already exists
in 2.50.4
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
- Fix compile failure on qemux86 [1] to
drop fix_op_instanceof_handler_for_32-bit_C-loop_build.patch,
which conflicts with upstream solution [2]
2. Drop file://0001-Fix-python-path-when-invoking-py-compile-54.patch as it was merged upstream.
3. Add pkgconfig to solve following configure error:
../sources/libcap-ng-0.9/configure: line 14862: syntax error near unexpected token `LIBBPF,libbpf,'
../sources/libcap-ng-0.9/configure: line 14862: `PKG_CHECK_MODULES(LIBBPF,libbpf,'
Signed-off-by: Liu Yiding <liuyd.fnst@fujitsu.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Wang Mingyu [Tue, 13 Jan 2026 10:25:04 +0000 (18:25 +0800)]
python3-urllib3: upgrade 2.6.2 -> 2.6.3
Changelog:
==========
- Fixed a high-severity security issue where decompression-bomb safeguards of
the streaming API were bypassed when HTTP redirects were followed.
- Started treating Retry-After times greater than 6 hours as 6 hours by
default.
- Fixed urllib3.connection.VerifiedHTTPSConnection on Emscripten.
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Wang Mingyu [Tue, 13 Jan 2026 10:25:03 +0000 (18:25 +0800)]
python3-pdm: upgrade 2.26.2 -> 2.26.4
Changelog:
===========
- Make sure cursor closing for fixing PyPy different gc mode also add PyPy in CI.
- Fix a bug that old HTTP cache directories cause PDM to crash when trying to clear them.
- Port to hishel 1.0.0.
- Update .gitignore file in the default template.
- Correct the sysconfig variables for Python standalone build installations.
- Ignore packages.vcs.requested-revision if it's None when formatting pylock.toml.
- Fix test failures with uv test cases using non-venv Python interpreters.
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Wang Mingyu [Tue, 13 Jan 2026 10:25:00 +0000 (18:25 +0800)]
python3-build: upgrade 1.3.0 -> 1.4.0
Changelog:
===========
- Add --quiet flag
- Add option to dump PEP 517 metadata with --metadata
- Support UV environment variable
- Remove a workaround for 3.14b1
- In 3.14 final release, color defaults to True already
- Pass sp-repo-review
- In pytest configuration, log_level is better than log_cli_level
- Split up typing and mypy
- Use types-colorama
- In docs, first argument for _has_dependency is a name
- Fix test failure when flit-core is installed
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Wang Mingyu [Tue, 13 Jan 2026 10:24:59 +0000 (18:24 +0800)]
libtasn1: upgrade 4.20.0 -> 4.21.0
Changelog:
=============
- Undocumented asn1Decoding --debug flag removed
- Code coverage for src/ went from 35% to 82%
- Fix of ASN.1 typo in manual
- NEWS renamed to NEWS.md and uses markdown syntax.
- Update gnulib files and various build/maintenance fixes.
- Fix for vulnerability CVE-2025-13151 Stack-based buffer overflow
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com> Signed-off-by: Antonin Godard <antonin.godard@bootlin.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
projects / thirdparty / openembedded / openembedded-core-contrib.git / log
