linux/poky.git, branch scarthgap

poky: Fix CentOS Stream distro names

2026-05-04T12:57:33+00:00

CentOS Stream 8 and 9 report as `centos-8/9`, not `centosstream-8.9`. On our stream9-vk-1 autobuilder worker: $ ( source /etc/os-release && echo $ID-$VERSION_ID ) centos-9 We don't have running CentOS Stream 8 workers anymore, but from a local docker container: $ docker run --rm -it quay.io/centos/centos:stream8 bash [root@c33732bb804a /]# cat /etc/os-release .... ID="centos" VERSION_ID="8" (From meta-yocto rev: 0845965f1d81d22d47dcf28c525c7463fafe5245) Signed-off-by: Antonin Godard Signed-off-by: Yoann Congal Signed-off-by: Paul Barker

liburcu: upgrade 0.14.0 -> 0.14.2

2026-05-04T12:57:33+00:00

Changelog for liburcu: 0.14.0 -> 0.14.2 ============================================================ 2026-01-26 Userspace RCU 0.14.2 * Fix: Only include linux/time_types.h when __NR_futex_time64 is defined * Use __NR_futex_time64 in futex syscall wrapper * Cleanup: Remove useless declarations from urcu-qsbr * src/urcu-bp.c: assert => urcu_posix_assert 2024-08-28 Userspace RCU 0.14.1 * Fix: missing typename in URCU_FORCE_CAST * Allow building with GCC >= 13.3 on RISC-V * pointer.h: Fix the rcu_cmpxchg_pointer documentation * Adjust shell script to allow Bash in other locations * fix: handle EINTR correctly in get_cpu_mask_from_sysfs * Relicense src/compat-smp.h to MIT * ppc.h: use mftb on ppc * Fix: allow clang to build liburcu on RISC-V * Fix -Walloc-size * urcu/uatomic/riscv: Mark RISC-V as broken * Fix: urcu-bp: misaligned reader accesses * LoongArch: Document that byte and short atomics are implemented with LL/SC * Add LoongArch support * tests/regression/rcutorture: Add wait state * urcu-wait: Initialize node in URCU_WAIT_NODE_INIT * Fix: urcu-wait: add missing futex.h include * Adjust shell scripts to allow Bash in other locations * Add support for OpenBSD * Revert compiler.h: Introduce caa_unqual_scalar_typeof * rculfhash: Use caa_container_of_check_null in cds_lfht_entry * compiler.h: Introduce caa_container_of_check_null * compiler.h: Introduce caa_unqual_scalar_typeof * Avoid calling caa_container_of on NULL pointer in cds_lfht macros * Fix: revise urcu_read_lock_update() comment * Fix: uatomic powerpc comment about lwsync * fix: aarch64: allow RHEL7 gcc 4.8.5-11 * fix: warning 'noreturn' function does return on ppc * Fix: use __noreturn__ for C11-compatibility (From OE-Core rev: dc2df90b1d4f71023169d492f3819326e0e6c055) Signed-off-by: Daniel Turull Signed-off-by: Yoann Congal Signed-off-by: Paul Barker

libatomic-ops: upgrade 7.8.2 -> 7.8.4

2026-05-04T12:57:33+00:00

Changelog for libatomic-ops: 7.8.2 -> 7.8.4 ============================================================ == [7.8.4] 2025-09-05 == * Document Win32-specific AO_NO_ASM_XADD macro in README * Eliminate 'atomic_thread_fence is unsupported with tsan' gcc-11 warning * Eliminate 'munmap argument 1 from incompatible pointer type' gcc warning * Eliminate 'uninitialized memory use' gcc-13 warning in test_atomic * Fix all broken URLs in comments * Fix atomic_ops_sysdeps.S compilation for SunPro compiler in CMake script * Update URLs after repository transfer to bdwgc organization * Update CMake minimum required version to 3.10 (From OE-Core rev: 189ecdff01a9c2c47b809d153b04ea6080f74e4f) Signed-off-by: Daniel Turull (cherry picked from commit 02d66279071cbd043474dc44f44d348ec3228583) Signed-off-by: Yoann Congal Signed-off-by: Paul Barker

lttng-ust: upgrade 2.13.8 -> 2.13.10

2026-05-04T12:57:33+00:00

Changelog for lttng-ust: 2.13.8 -> 2.13.10 ============================================================ 2026-02-13 lttng-ust 2.13.10 * Fix: ustctl: sigbus handling: statement with no effect 2025-04-14 (National Gardening Day) lttng-ust 2.13.9 * Fix: Use UATOMIC_HAS_ATOMIC_{BYTE,SHORT} for counter atomics * Fix: lttng-ust-tp regex warnings * lttng-ust-java: Load lttng-ust-context-jni before other JNI libraries * doc/examples: set minimal CMake version to 3.5.0 * Fix: Update get_mempolicy check to handle EPERM * Fix: Correct numa_available return code checks * Fix: test_benchmark: do not match CPU(s) scaling MHz: * Tests: Fix abi conflict test when building with clang * Fix: Build examples when rpath is stripped from in-build-tree libs * ust-fd: Add close_range declaration * docs: Correct GitHub URLs in lttng-ust.3 * fix: handle EINTR correctly in get_cpu_mask_from_sysfs (From OE-Core rev: 90f529115ed175652d2ab8c538d1c213550056f8) Signed-off-by: Daniel Turull Signed-off-by: Yoann Congal Signed-off-by: Paul Barker

sudo: upgrade 1.9.17p1 -> 1.9.17p2

2026-05-04T12:57:33+00:00

Changelog for sudo: 1.9.17p1 -> 1.9.17p2 ============================================================ Merge sudo 1.9.17p2 from branch 'main' into sudo-1.9 [d1b48c651] * configure, configure.ac: Fix check for which man page type to use with nroff Fixes a bug where configure would use *.man instead of *.mdoc on systems without mandoc. Bug #1077. [aa2498e46] * plugins/sudoers/log_client.c: client_msg_cb: make warning match the function that failed [f73162df3] 2025-07-23 Todd C. Miller * NEWS, configure, configure.ac: Sudo 1.9.17p2 [f0e1a5ca3] * plugins/sudoers/match_command.c, plugins/sudoers/match_digest.c: digest_matches: plug fd leak on snprinf() failure [26a1a7529] 2025-07-21 Todd C. Miller * etc/sudo-logsrvd.pp, etc/sudo-python.pp, etc/sudo.pp, scripts/mkpkg: Add a way to override pp_rpm_arch when building rpms This will be used to build x86_64_v2 packages for Alma Linux. [55d3c99c4] * configure, configure.ac: Fix check for which man page type to use with nroff Fixes a bug where configure would use *.man instead of *.mdoc on systems without mandoc. [2dc10cfbd] * plugins/sudoers/timestamp.c: ts_write: call lseek after fruncate on short write We need to make sure the file position is reset to the old EOF on error. [8e7e0e23f] 2025-07-20 Todd C. Miller * src/exec_ptrace.c: ptrace_readv_string: quiet sign-compare warning [fac2a49e7] * src/exec_ptrace.c: ptrace_readv_string: properly handle reads of more than one page When the intercept and intercept_verify options are enabled and either argv[] or envp[] contains a string larger than the page size (usually 4096), ptrace_readv_string() would fill the buffer with mutiple copies of the same string. Fixes GitHub issue #453. [2e93eabed] 2025-07-14 Todd C. Miller * src/exec_pty.c: revoke_pty: use killpg() not kill() to send HUP to the process group Also make sure we never call killpg(-1, SIGHUP), which would send SIGHUP to process 1 (init). It is possible for cmnd_pid to be -1 in certain error conditions where sudo killed the command itself. This may explain GitHub issue #458. [fb208d383] 2025-07-08 Todd C. Miller * etc/sudo-logsrvd.pp, etc/sudo-python.pp, etc/sudo.pp, scripts/pp: Don't assume RHEL major version is only a single digit Fixes handling of RHEL 10 and higher. [e5d953f33] * plugins/sudoers/visudo.c: visudo: create temporary file as mod 0600 not 0700 This was due to a typo in the mode field when the temporary file was created. Noticed by Bjorn Baron of the sudo-rs project. [1c254b330] 2025-06-30 Todd C. Miller * Makefile.in: We now build sudo releases from git, not mercurial [cb4e26734] 2025-06-28 Todd C. Miller * NEWS, configure, configure.ac: (From OE-Core rev: 76b98657e3dc9da01a746deb7b9d08cb84ba44b6) Signed-off-by: Daniel Turull (cherry picked from commit 12e9ba655153a9cb7c2b79cf52a2300e19634dcf) Signed-off-by: Yoann Congal Signed-off-by: Paul Barker

license.py: Drop visit_Str from SeenVisitor in selftest

2026-05-04T12:57:33+00:00

ast.Str was deprecated in Python 3.8 and removed in Python 3.14. In [1], visit_Str was already removed from the LicenseVisitor subclasses (FlattenVisitor, ListVisitor) in oe/license.py since bitbake now requires Python 3.8+. However, the test-only SeenVisitor class in the selftest was missed at that time and still uses visit_Str/node.s. On Python 3.14, ast.Str is fully removed so visit_Str is never called, causing test_single_licenses to return [] instead of the expected license list. Replace visit_Str/node.s with visit_Constant/node.value in SeenVisitor. [1] https://git.openembedded.org/openembedded-core/commit/meta/lib/oe/license.py?id=6d3da37adbcaf5a7a3dade08f9d052571b195249 Fixes [YOCTO #16220] (From OE-Core rev: f8ac1b567d7c09ee45cb0d23fc9dd25987a76129) Signed-off-by: Sunil Dora Signed-off-by: Richard Purdie (cherry picked from commit 8a5b019eec72676893507d018e7609745d2e3f49) Signed-off-by: Yoann Congal Signed-off-by: Paul Barker

selftest/gdbserver: replace shutil.unpack_archive with tarfile extract

2026-05-04T12:57:33+00:00

This is a followup to https://lists.openembedded.org/g/openembedded-core/message/233609 as the same issue happened in a different selftest. [YOCTO #16195] (From OE-Core rev: 948028ca3466fb7071ee49000e847d6e886e349c) Signed-off-by: Alexander Kanavin Signed-off-by: Richard Purdie (cherry picked from commit c7468e70c238b056acbe06ef722b62b02626db8f) Signed-off-by: Yoann Congal Signed-off-by: Paul Barker

selftest/minidebuginfo: extract files from tar archive using tarfile module

2026-05-04T12:57:33+00:00

Python 3.14 added security checks around archive extraction, and by default will refuse to handle symlinks with absolute paths. It's possible to handle this using 'filter' argument, but it is not always available in older Python versions on various host distributions we need to support, so let's extract only the needed files directly using tarfile module. busybox is itself a symlink to busybox.nosuid, so both are extracted. [YOCTO #16195] (From OE-Core rev: 799e2124cdcb27bd924598619da61cbd11730b56) Signed-off-by: Alexander Kanavin Signed-off-by: Richard Purdie (cherry picked from commit d52d00a3bb4a1ba93e88f1d24d8bb99d6aa321eb) Signed-off-by: Yoann Congal Signed-off-by: Paul Barker

archiver: Don't try to preserve all attributes when copying files

2026-05-04T12:57:33+00:00

Similar to https://git.openembedded.org/bitbake/commit/?id=2f35dac0c821ab231459922ed98e1b2cc599ca9a there is a problem in this code when copying from an NFS mount. We currently use cp -p, which is a shortcut for --preserve=mode,ownership,timestamps. We shouldn't need to preserve mode/ownership, only timestamps. Update the code in the same way the bitbake fetcher was fixed for consistency. This fixes build failures on OpenSUSE 16.0. (From OE-Core rev: 6ffd2c303d23b51a0a11f56e2e89c5a19596d228) Signed-off-by: Richard Purdie (cherry picked from commit 6e8313688fa994c82e4c846993ed8da0d1f4db0e) Signed-off-by: Yoann Congal Signed-off-by: Paul Barker

lib/oeqa/utils/httpserver: use multiprocessing from bb

2026-05-04T12:57:33+00:00

Fixes build with python-3.14 (and hence on Fedora 43) It was added to bitbake in 62be9113d98fccb347c6aa0a10d5c4ee2857f8b6 and oe-core now requires latest bitbake already, so we can use this. (From OE-Core rev: 2273e9ba2ffeea57620af576e3cd1e523e4d3fc3) Signed-off-by: Richard Purdie (cherry picked from commit 3ed1cef082b8ef0fa05684f7a41f0842150811bd) [YC: bitbake commit was backported as 15d7448e04aa ("Use a "fork" multiprocessing context") in the 2.8(scarthgap) branch] Signed-off-by: Yoann Congal Signed-off-by: Paul Barker