Fixes the following security issue:
CVE-2025-4207: PostgreSQL GB18030 encoding validation can read one byte past
end of allocation for text that fails validation
A buffer over-read in PostgreSQL GB18030 encoding validation allows a
database input provider to achieve temporary denial of service on platforms
where a 1-byte over-read can elicit process termination. This affects the
database server and also libpq.
https://www.postgresql.org/about/news/postgresql-175-169-1513-1418-and-1321-released-3072/
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit a8f53a907b)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Since commit [1], the host-nodejs package was turned into a virtual
package. However, the target nodejs package was intentionally not
turned into a virtual package and became an empty package.
The reason for this is:
- No alternatives providers were introduced for the target nodejs
package.
- The Config.in remained unchaged for the target package.
As a result, running `make show-info` with a config that includes the
target nodejs package, outputs an entry for the empty package:
```
"nodejs": {
"type": "target",
"name": "nodejs",
"virtual": false,
"version": "",
...
"cpe-id": "cpe:2.3:a:nodejs:node.js::*:*:*:*:*:*:*"
},
```
This can be an issue because the CPE ID of the empty nodejs package
is the following `cpe:2.3:a:nodejs:node.js::*:*:*:*:*:*:*`.
Reporting such a CPE ID can be an issue for certain software that consume
the SBOM and could be interpreted as CPE that matches with every versions
of the package.
This patch converts the target nodejs package into a virtual package to
prevents the empty package from being included in the SBOM.
[1] 4cbc2af604 package/nodejs: rename to nodejs-src and convert to virtual package
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Tested-by: johan.derycke@barco.com
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit ccf3536fcb)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Version 6.1.2 of ffmpeg fails to build with GCC 14.x due to the V4L2
ioctl code:
libavdevice/v4l2.c:137:17: error: assignment to ‘int (*)(int, long unsigned int, ...)’ from incompatible pointer type ‘int (*)(int, int, ...)’ [-W
incompatible-pointer-types]
137 | s->ioctl_f = prefix ## ioctl; \
| ^
libavdevice/v4l2.c:151:9: note: in expansion of macro ‘SET_WRAPPERS’
151 | SET_WRAPPERS();
| ^~~~~~~~~~~~
This has been fixed upstream in the release/6.1 branch, which has 27
fixes on top of 6.1.2. The commits necessary to fix our issue are:
f71076c009f84917e7a0f2f1ece86b718de2d8d3 configure: improve check for POSIX ioctl
60593d6c06c9b610359bd6af26a268feff1293eb configure: restore autodetection of v4l2 and fbdev
However, since all other commits are fixes, we believe bumping to the
latest commit in the release/6.1 branch is a better idea.
This allows to drop
0008-libavcodec-arm-mlpdsp_armv5te-fix-label-format-to-wo.patch, which
is upstream as of:
4c688845a50f7dce3af9afebe60f0f7a493c4f07 libavcodec/arm/mlpdsp_armv5te: fix label format to work with binutils 2.43
Note that we set FFMPEG_CPE_ID_VERSION to get proper CVE matching even
with FFMPEG_VERSION being set to n6.1.2-27-ge16ff06adb. One who have
ideally set FFMPEG_VERSION to n$(FFMPEG_CPE_ID_VERSION)-ge16ff06adb,
but that makes check-package unhappy with:
WARNING: package/ffmpeg/ffmpeg.mk:7: expecting package version to be set before CPE_ID_VERSION
Fixes:
https://autobuild.buildroot.net/results/fe1574443acd50ca7e576bb4beb24467be1713e3/
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Arnout Vandecappelle <arnout@rnout.be>
(cherry picked from commit 32df543fa3)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Fixes:
Build with gcc-15 was fixed upstream in 10.11.11:
https://github.com/MariaDB/server/commit/1d6f857
Remove patch, since the problem was fixed upstream:
https://github.com/MariaDB/server/commit/4375245
After bumping the version to 10.11.11 the configuration step failed for target:
CMake Error: try_run() invoked in cross-compiling mode, please set the following cache variables appropriately:
HAVE_SYSTEM_LIBFMT_EXITCODE (advanced)
To fx this problem we set HAVE_SYSTEM_LIBFMT_EXITCODE=0.
Fix legal-info after changes to README.md.
Signed-off-by: Roy Kollen Svendsen <roykollensvendsen@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 6cd8f95346)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
dbus has a session socket directory configuration setting,
that, if not set, will be autodeducted based on env vars
like TMPDIR during configuration time.
Becuse of that, the builder's environment variables will
lead to an image with a broken session bus while
leaking builder's details to the image.
Add an explicit setting of session-socket-dir to /tmp dir.
Fixes: https://gitlab.com/buildroot.org/buildroot/-/issues/67
Signed-off-by: Nikita Kiryushin <kiryushin@ancud.ru>
Reviewed-by: Fiona Klute <fiona.klute@gmx.de>
Signed-off-by: Arnout Vandecappelle <arnout@rnout.be>
(cherry picked from commit f777c79912)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
A set of `unterminated-string-initialization` errors appeared when
building the micropython package with GCC15 on the host.
The autobuilder failed to build the package micropython with the
following error:
```
CC ../py/emitinlinethumb.c
../py/emitinlinethumb.c:153:9: error: initializer-string for array of 'unsigned char' truncates NUL terminator but destination lacks 'nonstring' attribute (4 chars into 3 available) [-Werror=unterminated-string-initialization]
153 | {0, "r0\0"},
| ^~~~~~
../py/emitinlinethumb.c:154:9: error: initializer-string for array of 'unsigned char' truncates NUL terminator but destination lacks 'nonstring' attribute (4 chars into 3 available) [-Werror=unterminated-string-initialization]
154 | {1, "r1\0"},
| ^~~~~~
...
```
This patch adds the set of upstream commits to fix the compatbility with
GCC15 (see [1]).
The patches were backported to micropython v1.22.2. The main difference
with the upstream version is that since the v1.23, the project removed the
use of the `STATIC` macro (see [2]).
Also, in the codebase of v1.22.2 the 'unterminated-string-initialization'
error occured in another file that was reworked in the patch [3] and
included in v1.25. This patch is included as well to remove the error in
v1.22.2.
[1] package/micropython/0003-Fixes-for-GCC-15-1-unterminated-string-literal-warning.patch
[2] decf8e6a8b
[3] package/micropython/0002-py-emitinlinextensa-Simplify-register-name-lookup.patch
Fixes: https://autobuild.buildroot.org/results/fdf/fdf1d7c3e3a51e6fc7fa5abea57de6c9ce792015
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 0814b614c2)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Inspired by
c250c61cc3
"GCC 15 defaults to C23. The last release of this package was over a
decade ago, and it is no longer maintained, therefore it should not be
expected to compile to the latest standards."
Signed-off-by: Bernd Kuhls <bernd@kuhls.net>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit c1d422edde)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
This commit adds the same patch that was already added to GDB 14.x,
15.x and 16.x to fix a GCC 15.x build issue.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit dc0691f038)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
This commit backports an upstream patch fixing the bundled readline
library so that it builds with GCC 15.x.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 56c834400e)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
bash uses K&R function declarations which have been removed in C23.
Since part of the build process (like the mkbuiltins helper) is written
in C, building bash now fails on hosts with GCC 15 (which defaults to
C23).
Since properly fixing this on the source code level is a larger
endeavor, just set the C standard to an old enough version for now.
Signed-off-by: Florian Larysch <fl@n621.de>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 6d09b25d08)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
On systems running GCC 15, host-cpio will fail to build with errors like
copyout.c:646:12: error: too many arguments to function 'xstat'; expected 0, have 2
and
main.c:407:13: error: assignment to 'int (*)(void)' from incompatible pointer type 'int (*)(const char * restrict, struct stat * restrict)' [-Wincompatible-pointer-types]
This was reported[1] and fixed upstream, but there is no new release
yet. Import the upstream patch for now.
Fixes:
https://autobuild.buildroot.org/results/a10c5f2b0f9cb05b2550fe97f1133deaaac97277/
(and many more)
[1] https://lists.gnu.org/archive/html/bug-cpio/2025-05/msg00000.html
Signed-off-by: Florian Larysch <fl@n621.de>
[Julien:
- add missing "Signed-off-by:" in patch to fix check-package error
- change "Upstream:" link to use the commitdiff in patch
- add "Fixes:" in commit log
]
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit e3cae9e1ca)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
When using host gcc 15, host-gmp fails at configure time with
error message:
configure: error: could not find a working compiler, see config.log for details
The error is due to the macro GMP_PROG_CC_WORKS in the file
acinclude.m4 containing C-code declaring functions without declaring
its parameters. This construct is now an error in C23, which is the
new default in gcc 15. See:
https://gcc.gnu.org/gcc-15/porting_to.html#c23
This commit fixes the issue by adding a package patch from
upstream changesets, not yet published in a release.
Even if gcc 15 is not yet included in Buildroot, this commit also
sets _AUTORECONF = YES for the target package in order to have this
package already fixed.
Fixes:
https://autobuild.buildroot.org/results/623634fa7bbeceeb6d90b15ce0abb1b9b4b24045/
Tested-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 31569bcc1b)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
When compiling host-m4 1.4.19 with a host gcc 15 (which is the version
included in Fedora 42, released on 2025-04-15), compilation fails with
error:
In file included from gl_avltree_oset.h:21,
from gl_avltree_oset.c:21:
gl_oset.h:275:1: warning: 'nodiscard' attribute ignored [-Wattributes]
275 | GL_OSET_INLINE _GL_ATTRIBUTE_NODISCARD int
| ^~~~~~~~~~~~~~
gl_oset.h:275:40: error: expected identifier or '(' before 'int'
275 | GL_OSET_INLINE _GL_ATTRIBUTE_NODISCARD int
| ^~~
This error is due to the gnulib copy included in m4 1.4.19, which does
not detect properly the default C language standard of gcc 15 which
has been changed from "gnu17" to "gnu23". See [1]. Note that m4 1.4.19
is the latest version available at the time of this commit, and was
released in May 2021. The issue is tracked upstream in [2].
Upcoming m4 release is expected to fix this issue, by updating its
gnulib copy. See [3], which states: "Update to comply with newer C
standards, and inherit portability improvements from gnulib".
Until this new m4 version is released, this commit fixes the issue by
forcing the C langage standard to "-std=gnu17" (the previous gcc
default) when host-gcc 15 is detected.
Note that the "-std=gnu17" option was introduced in gcc 8. See [4].
This is the reason why this patch adds this option only when the
problematic gcc 15 version is detected.
See also the discussions around this patch at [5].
Fixes:
https://autobuild.buildroot.org/results/1c33ef0a710cfae13e496485787b351c8f951217/
(and many, many others)
[1] https://gcc.gnu.org/gcc-15/changes.html#c
[2] https://savannah.gnu.org/support/?111150
[3] https://git.savannah.gnu.org/cgit/m4.git/commit/?h=branch-1.4&id=a22c9802dd7e724eaefb21dc21d84ac2d3a49c89
[4] https://gcc.gnu.org/gcc-8/changes.html#c
[5] https://lore.kernel.org/buildroot/CAPWx8vsoJUt8YMJG1aUqFRK1=yizNbgjVjGL1Q1+9ygjJGnZLA@mail.gmail.com/
Signed-off-by: Joseph Zikusooka (ZIK) <zik@jambula.net>
Tested-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
[Julien:
- change mail url to lore.kernel.org for stable link
- reword, reflow and add extra info in the commit log
- force -std=gnu17 only when host gcc-15 is detected
- add a comment in .mk to remove the workaround at next bump
]
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 7a07a9d155)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
