Age | Commit message (Collapse) | Author |
|
qemumips and qemuppc were leaving stale processes behind after
running glibc oe-selftest. During analysis, it was found that
it was due to "tst-scm_rights" and "tst-scm_rights-time64" tests.
Disable them so that there are no stale processes left behind.
[YOCTO #15423]
https://bugzilla.yoctoproject.org/show_bug.cgi?id=15423
Signed-off-by: Yash Shinde <Yash.Shinde@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
despite it being an issue in gcc and still being open
glibc has fixed this problem upstream regardless, therefore
apply the backport instead.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Brings
* 1b9c1a0047 Use gcc __builtin_stdc_* builtins in stdbit.h if possible
* e0910f1d32 S390: Do not clobber r7 in clone [BZ #31402]
* d0724994de math: Update mips64 ulps
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Until now, ${datadir}/locale/locale.alias file were automatically added
to a weird glibc-locale-locale.alias package by split_locales() during
do_package task.
Create an explicit package name in recipe for this file.
Signed-off-by: Jonathan GUILLOT <jonathan@joggee.fr>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
When ldconfig-native reads an ELF file, it computes an offset from a LOAD
segment, to point on DT NEEDED entries of dynstr section.
Without this patch, ldconfig-native uses only the first LOAD segment, even if
the offset is incorrect.
This patch adds conditions to compute the offset by parsing all LOAD segments,
one by one.
This is a backport from [0], ported to support endianness and 32/64 bits.
[0]: https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=58e8f5fd2ba47b6dc47fd4d0a35e4175c7c87aaa
Signed-off-by: Fabien Mahot <fabien.mahot@external.desouttertools.com>
Reviewed-by: Yoann Congal <yoann.congal@smile.fr>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This patch is the only change applied with this update
* 312e159626 mips: FIx clone3 implementation (BZ 31325)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Brings this bugfix for arm
63295e4fda arm: Remove wrong ldr from _dl_start_user (BZ 31339)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
- drop irrelevant CVEs for 2.39 release
Signed-off-by: Valek Andrej <andrej.v@skyrain.eu>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
License-Update: Relicenses the IBM portions of resolv/base64.c and
resolv/res_debug.c to a new license that does not have use-limited
patent language [1]
Upgrade localdef to get glibc 2.39 build fixes
Details of release [2]
Add fix for mips clone3 crash
[1] https://sourceware.org/git/?p=glibc.git;a=commit;h=ae49a7b29acc184b03c2a6bd6ac01b5e08efd54f
[2] https://sourceware.org/glibc/wiki/Release/2.39
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Support has been removed from 2.39+ see [1]
[1] https://sourceware.org/git/?p=glibc.git;a=commit;h=25f1e16ef03a6a8fb1701c4647d46c564480d88c
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Set `CVE_STATUS`for those CVEs, they have already been fixed with the latest
pull for stable branch fixes done in rev
e444d2bed0ea140a574414fcd5a689867e8ba312. Hence the issues are fixed
already.
Signed-off-by: Simone Weiß <simone.p.weiss@posteo.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
I'm not sure why this was included and enabled to begin with: the tests
predictably mass-fail if system time is set to after y2038.
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Below commits on glibc-2.38 stable branch are updated.
1e04dcec49 Revert "elf: Move l_init_called_next to old place of l_text_end in link map"
719866ab2f Revert "elf: Always call destructors in reverse constructor order (bug 30785)"
e0b6c9706c Revert "elf: Remove unused l_text_end field from struct link_map"
Signed-off-by: Sundeep KOKKONDA <sundeep.kokkonda@windriver.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
The documentation of systemd states that /etc/tmpfiles.d should be
reserved for the local administrator and packages should put their files
in /usr/lib/tmpfiles.d [1].
[1] https://www.freedesktop.org/software/systemd/man/tmpfiles.d.html
Signed-off-by: Stefan Herbrechtsmeier <stefan.herbrechtsmeier@weidmueller.com>
Signed-off-by: Lukas Funke <lukas.funke@weidmueller.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
Pull in stable branch fixes including:
* tunables: Terminate if end of input is reached (CVE-2023-4911)
* Propagate GLIBC_TUNABLES in setxid binaries
* Document CVE-2023-4806 and CVE-2023-5156 in NEWS
* Fix leak in getaddrinfo introduced by the fix for CVE-2023-4806 [BZ #30843]
Also set CVE_STATUS accordingly for the fixes pulled in.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Below commits on glibc-2.38 stable branch are updated.
0e1ef6779a (HEAD -> release/2.38/master, origin/release/2.38/master) manual/jobs.texi: Add missing @item EPERM for getpgid
d94461bb86 string: Fix tester build with fortify enable with gcc < 12
63250e9c57 iconv: restore verbosity with unrecognized encoding names (bug 30694)
00ae4f10b5 getaddrinfo: Fix use after free in getcanonname (CVE-2023-4806)
b25508dd77 CVE-2023-4527: Stack read overflow with large TCP responses in no-aaaa mode
89da8bc588 NEWS: Add the 2.38.1 bug list
d3ba6c1333 elf: Move l_init_called_next to old place of l_text_end in link map
750f19526a elf: Remove unused l_text_end field from struct link_map
a3189f66a5 elf: Always call destructors in reverse constructor order (bug 30785)
7ae211a01b elf: Do not run constructors for proxy objects
92201f16cb libio: Fix oversized __io_vtables
5bdef6f27c io: Fix record locking contants for powerpc64 with __USE_FILE_OFFSET64
0024-CVE-2023-4527.patch is dropped
Signed-off-by: Deepthi Hemraj <Deepthi.Hemraj@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Upstream-Status: Backport[https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=4ea972b7edd7e36610e8cde18bf7a8149d7bac4f]
Signed-off-by: Yash Shinde <Yash.Shinde@windriver.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We've been removing PR values from recipes at upgrade time for a while. In general
anyone maintaining a binary distro would end up having to curate these themselves
so the values in OE-Core aren't really that useful anymore. In many ways it makes
sense to clear out the remaining ones (which are mostly for 'config' recipes that
are unlikely to increase in PV) and leave a clean slate for anyone implementing
a binary distro config.
References are left in meta-selftest since the tests there do involve them and
their removal upon upgrade.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The recipe originates from meta-y2038 where the name was not
confusing, but in oe-core it is.
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
file /usr/include/bits/math-vector.h from install of lib32-libc6-dev-2.38-r0.armv7ahf_neon conflicts with file from package libc6-dev-2.38-r0.aarch64
Reference to the git log of glibc, upstream modified math-vector.h for
aarch64, so this file has many differences from aarch32.
For detailed modifications, please refer to these two commit log of
glibc:
commit 4a9392ffc27ad280f84779eea3ba01f2c134d1d8
commit 78c01a5cbeb6717ffa2d4d66bb90ac5c39bd81a9
Signed-off-by: Lei Maohui <leimaohui@fujitsu.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
Pull in the following stable branch updates:
1aed90c9c8f8be9f68b58e96b6e4cd0fc08eb2b1 sysdeps: tst-bz21269: fix -Wreturn-type
ad9b8399537670a990572c4b0c4da5411e3b68cf sysdeps: tst-bz21269: handle ENOSYS & skip appropriately
c8ecda6251dd4a0dfe074e0a6011211cadeef742 sysdeps: tst-bz21269: fix test parameter
2af141bda3cd407abd4bedf615f9e45fe79518e2 malloc: Remove bin scanning from memalign (bug 30723)
98c293c61f770b6b7a22f89a6ea81b711ecb1952 malloc: Enable merging of remainders in memalign (bug 30723)
7ac405a74c6069b0627dc2d8449a82a621f8ff06 i686: Fix build with --disable-multiarch
6135d50e44233d8c89ca788f78c669941ad09fb9 x86_64: Fix build with --disable-multiarch (BZ 30721)
5ea70cc02626d9b85f1570153873d8648a47bf95 x86: Fix incorrect scope of setting `shared_per_thread` [BZ# 30745]
6b99458d197ab779ebb6ff632c168e2cbfa4f543 nscd: Do not rebuild getaddrinfo (bug 30709)
ced101ed9d3b7cfd12d97ef24940cb00b8658c81 x86: Fix for cache computation on AMD legacy cpus.
d97cca1e5df812be0e4de1e38091f02bb1e7ec4e stdlib: Improve tst-realpath compatibility with source fortification
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Older glibc on host might be missing __glibc_has_attribute added in glibc-2.34 with:
https://sourceware.org/git/?p=glibc.git;a=blobdiff;f=misc/sys/cdefs.h;h=8e244a77cf6271f09cbd26d18b1e07b1d1641404;hp=57ca262bdfb642bf4a945645532f2319ec1ff437;hb=c8ba52ab3350c334d6e34b1439a4c0c1431351f3;hpb=7dd416491e080456fc7742d884c520526509413e
Fixes:
In file included from ../git/localedef/include/programs/xmalloc.h:21,
from glibc/locale/programs/localedef.h:124,
from glibc/locale/programs/localedef.c:36:
../git/localedef/include/sys/cdefs.h:85:51: error: missing binary operator before token "("
85 | #if __GNUC_PREREQ (4, 3) || __glibc_has_attribute (__cold__)
| ^
Only commit included:
e0eca29 cdefs.h: define __glibc_has_attribute
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Some tests e.g. tst-y2039-time64 from glibc 2.38+ needs it.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Its removed and is implicit default in 2.38+ [1]
[1] https://mail.gnu.org/archive/html/info-gnu/2023-07/msg00010.html
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Announcement - https://mail.gnu.org/archive/html/info-gnu/2023-07/msg00010.html
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This resolves two issues:
1. metapackages were depending on themselves (except -binaries which wouldn't match against 'glibc-binary').
2. for the nativesdk variant, due to a non-empty dependency list at parsing time caused by
issue 1, map_depends_variable() from meta/lib/oe/classextend.py was forcibly setting PACKAGES
to the initial parse-time value (e.g. missing the dynamically created packages). This meant that
three out of four nativesdk- metapackages were entireyly missing the dependencies on the
respective dynamic package sets.
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Dont fill up the test log with ssh warning about having added the host
to list of known hosts.
Also helps fix a test case failure where stderr log was being compared
to a known value.
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
- Try to add convert and apply statuses for old CVEs
- Drop some obsolete ignores, while they are not relevant for current
version
Signed-off-by: Andrej Valek <andrej.valek@siemens.com>
Reviewed-by: Peter Marko <peter.marko@siemens.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
The check target for the glibc testsuite uses networking to access
a QEMU in system mode. This was resulting in large numbers of test
failures for the x86 targets that use it.
Enable network access to resolve this.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
glibc configury tries to detect ld version and assumes BFD or gold
linker but when system ld is pointing to lld or mold it might fail the
linker check, therefore pass LD variable to explicitly point at ld.bfd
we are using BFD linker only to link glibc after all.
Second problem in such a case is that some partial objects are linked
with CC -r which will fail if we do not inform the compiler to use BFD
linker thusly pass it via appending to CC variable
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
The recommendation from server maintainers is that the https protocol
is both faster and more reliable than the dedicated git protocol at this point.
Switch to it where possible.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
handling
There is little point in having "crosssdk" suffex added to the virtual provider within
binutils since the TARGET_PREFIX or SDK_PREFIX already encapsulates this. Remove it
allowing some of the special case overriding to be removed.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
A recent package.py change has highlighted some corruption issues with -dev
pkgconfig package dependencies. Bump the output versions to trigger a rebuild
and ensure everything is consistent.
Take the opportunity to also drop all HASHEQUIV_HASH_VERSION entries since the
main version is changing.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
glibc gcc vulkan add support for loongarch64
Signed-off-by: Jialing Zhang <zhangjialing@loongson.cn>
Signed-off-by: Qizheng Zhu <zhuqizheng@loongson.cn>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
As distro builders this option helps using different compilers to build
glibc
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The script has a bashism and needs bash to execute correctly. Mark it
as such and add the missing bash dependency so it executes in minimal
images.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There are cached reproducibility issues on the autobuilder due to the PRServ
sstate checksum issues, flush the bad data out the system by bumping the
versions.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
glibc has it's dependencies handled more manually due to it's place
in the toolchain bootstrap. It depends upon the compiler and indirectly
through that to binutils. This did mean that if binutils changes and the
compiler does not, sstate and hash equivalence could mean that glibc
wouldn't rebuild.
Add a direct dependency on binutils that if it changes, it forces glibc
to rebuild, as it should.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Remove some obsolete CVE ignores now that releases have been made, CPEs
updated, or upgrades done:
CVE-2020-10029 is marked as fixed in 2.32.
CVE-2021-27645 is marked as fixed in 2.34.
CVE-2022-39046 is marked as fixed in 2.37.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This CVE has been fixed in the 2.37 release branch (07b9521fc6) and
we're now using a SHA that incorporates that commit, so manually mark it
as ignored.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Below commits on glibc-2.37 stable branch are updated.
d8e1a7590d elf: Smoke-test ldconfig -p against system /etc/ld.so.cache
6fe86ecd78 NEWS: Document CVE-2023-25139.
07b9521fc6 Account for grouping in printf width (bug 30068)
fb7b95dc47 Use 64-bit time_t interfaces in strftime and strptime (bug 30053)
9f8513dc64 LoongArch: Add new relocation types.
020b43544a cdefs: Limit definition of fortification macros
Signed-off-by: Yash Shinde <Yash.Shinde@windriver.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There is a redundant change in
0016-wordsize.h-Unify-the-header-between-arm-and-aarch64.patch that
causes wordsize.h to be different in arm and aarch64. This causes the
build error when enable multilib:
Error: Transaction test error:
file /usr/include/bits/wordsize.h conflicts between attempted installs of lib32-libc6-dev-2.37-r0.armv7vet2hf_vfp and libc6-dev-2.37-r0.cortexa57
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
|
|
Details are here [1] and summary of changes is here [2]
[1] https://sourceware.org/glibc/wiki/Release/2.37
[2] https://sourceware.org/pipermail/libc-announce/2023/000035.html
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Patch so ppoll is properly redirected to it's 64bit time variant even
when source fortification is active.
Signed-off-by: Ola x Nilsson <olani@axis.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
The tests are packaged into the main glibc-tests package which is fine,
but then glibc-tests-ptest package needs to depend on that.
Which is what this commit addresses.
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
The glibc-locale recipe already partially depends on the base depends in
order to satisfy the do_package dependency on binutils. However since
commit d6ffd683bf6 NM has defaulted to gcc-nm, meaning do_package
depends on gcc (for minidebuginfo).
Whilst the do_package task could handle having the dependencies
explicitly defined (either in glibc-locale or in package.bbclass),
setting these would require some amount of conditional dependency
configuration (cross/crosssdk/etc.). Since both binutils and gcc are
already dependencies of virtual/libc (although compilerlibs is not),
having glibc-locale not inhibit the default depends simplifies the
handling of this situation for both glibc-locale and package.bbclass.
Signed-off-by: Nathan Rossi <nathan.rossi@digi.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
glibc-locale defaults to ENABLE_BINARY_LOCALE_GENERATION ?= "0", but gets
changed to "1" in the default-distrovars.inc
When it is explicitly set back to "0", it fails with this error:
ERROR: glibc-locale-2.36-r0 do_package: QA Issue: glibc-locale: Files/directories were installed but not shipped in any package:
/usr/lib/locale
Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them within do_install.
glibc-locale: 1 installed and not shipped files. [installed-vs-shipped]
ERROR: glibc-locale-2.36-r0 do_package: Fatal QA errors were found, failing task.
There's a code to remove empty dirs in ${libdir}, but it's keyed off of
PACKAGE_NO_GCONV to clean up $[libdir]/gconv, just extend it to also cover
other empty dirs, including ${libdir}/locale.
Signed-off-by: Denys Dmytriyenko <denis@denix.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The fix for this CVE has been merged into the 2.36 branch and fixed in
oe-core as of 055ce284.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|