Fix the following build failure with aarch64 and BR2_OPTIMIZE_0 raised
since commit 0512910e49:
src/convert-neon-body-bayer8.inc.h: In function 'bayer2rgb_convert_neon':
src/convert-neon.c:233:17: error: third argument to '__builtin_prefetch' must be a constant
233 | __builtin_prefetch(((void const *)(_addr)) + 64, 0, (_lvl)); \
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
src/convert-neon-body-bayer8.inc.h:8:9: note: in expansion of macro 'do_prefetch'
8 | do_prefetch(&in[+1*stride - 1], (rows_per_loop) < 3 ? 3 : 0);
| ^~~~~~~~~~~
Fixes:
- http://autobuild.buildroot.org/results/307c1c448becd99dfba3a17b3ef816db0bb5cad6
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 13f8df8ed6)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit 6b2329bb80 ("configs/ti_am64x_sk: new defconfig") forgot to
specify the Linux kernel version, so do that now.
When the defconfig was added, the default version was 6.4.x, so use
version 6.4.16.
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 820db039f2)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit a7736afaca (package/opencv4-contrib: properly note dependencies
between modules) forgot to propagate the dependencies of selected
symbols, causing unmet dependencies warnings (unfortunately, not
errors), such as:
$ KCONFIG_SEED=0x9F66F854 make randconfig
WARNING: unmet direct dependencies detected for BR2_PACKAGE_OPENCV4_LIB_DNN
Depends on [n]: BR2_PACKAGE_OPENCV4 [=y] && !BR2_TOOLCHAIN_USES_UCLIBC [=y] && BR2_PACKAGE_PROTOBUF_ARCH_SUPPORTS [=y]
Selected by [y]:
- BR2_PACKAGE_OPENCV4_CONTRIB_LIB_DNN_OBJDETECT [=y] && BR2_PACKAGE_OPENCV4_CONTRIB [=y]
Propagate the dependencies and add associated comments.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Woodrow Douglass <wdouglass@carnegierobotics.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit a2e01b23fc)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit 719921bebe (package/kexec: bump to version 2.0.27) forgot to
propagate the new dependency on kernel headers to petitboot, which may
select kexec, causing unmet dependencies warnings (unfortunately, not
errors), such as:
$ KCONFIG_SEED=0x84462FFC make randconfig
WARNING: unmet direct dependencies detected for BR2_PACKAGE_KEXEC
Depends on [n]: BR2_PACKAGE_KEXEC_ARCH_SUPPORTS [=y] && BR2_TOOLCHAIN_HEADERS_AT_LEAST_3_17 [=n]
Selected by [y]:
- BR2_PACKAGE_PETITBOOT [=y] && (BR2_PACKAGE_KEXEC_ARCH_SUPPORTS [=y] || BR2_PACKAGE_KEXEC_LITE_ARCH_SUPPORTS [=n]) && BR2_USE_MMU [=y] && BR2_USE_WCHAR [=y] && !BR2_STATIC_LIBS [=n] && BR2_TOOLCHAIN_HAS_THREADS [=y] && BR2_PACKAGE_HAS_UDEV [=y] && !BR2_PACKAGE_KEXEC_LITE_ARCH_SUPPORTS [=n]
Propagate the dependency from kexec, and add comments about dependencies
also inherited from kexec-lite.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Laurent Vivier <laurent@vivier.eu>
Cc: Julien Olivain <ju.o@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit c721a58379)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Currently, the comment about availability is between the package main
symbol and its suboptions, breaking the threading in menuconfig.
Invert the order, and move the comment last. Also enclose the suboptions
between an if-block, as it is more customary.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 06dcbdf8ad)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
libressl dropped engine support since version 3.8.1 resulting in the
following build failure since bump of libressl to version 3.8.2 in
commit 21eca49ed5:
eng_front.c: In function 'IMPLEMENT_DYNAMIC_CHECK_FN':
eng_front.c:262:1: error: expected declaration specifiers before 'IMPLEMENT_DYNAMIC_BIND_FN'
262 | IMPLEMENT_DYNAMIC_BIND_FN(bind_fn)
| ^~~~~~~~~~~~~~~~~~~~~~~~~
Fixes:
- http://autobuild.buildroot.org/results/389ed112f082a4917f777f3e236211059c4c6d6e
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit f622944bc8)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Security fixes:
- CVE-2023-52425: Fix quadratic runtime issues with big tokens that can
cause denial of service, in partial where dealing with compressed XML
input. Applications that parsed a document in one go -- a single call
to functions XML_Parse or XML_ParseBuffer -- were not affected. The
smaller the chunks/buffers you use for parsing previously, the bigger
the problem prior to the fix.
- CVE-2023-52426: Fix billion laughs attacks for users compiling
*without* XML_DTD defined (which is not common). Users with XML_DTD
defined have been protected since Expat >=2.4.0 (and that was
CVE-2013-0340 back then).
https://blog.hartwork.org/posts/expat-2-6-0-released/
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 9dad5e7d7f)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The section of the manual describing the makedev syntax is not
up-to-date with the current features, and does not properly describe
existing ones.
- extend the list of types with the requirements on the existence of
the target file or directory; for 'c', 'b', and 'p', the existence
requirement is inherited from mknod(2):
ERRORS
...
ENOENT A directory component in pathname does not exist or is a
dangling symbolic link.
for the other types, the existence requirements are extracted from
the source of makedev.c;
- format the types flags, so they are rendered in monospace;
- extend the 'mode' description, as it can be set to -1 for 'f', 'd',
or 'r', so that only the uid and gid are set. This is most useful
for 'r', where setting the same mode recursively for all the
sub-directories and files alike does not really make sense; indeed
in this case, the modes are usually set correctly when the package
(or rootfs overlay) installs the files, and only the uid and gid are
interesting to set;
- extend and update the examples to show-case the -1 mode use-case.
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 738fb6dfa4)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
cpe:2.3:a:green_end:sftpserver has been deprecated in favor of
cpe:2.3:a:greenend:sftpserver in December 2022:
<cpe-item name="cpe:/a:green_end:sftpserver:0.2.2" deprecated="true" deprecation_date="2022-12-30T17:53:22.797Z">
<reference href="https://github.com/ewxrjk/sftpserver/releases">Version</reference>
<reference href="http://www.greenend.org.uk/rjk/sftpserver/">Product</reference>
<cpe-23:cpe23-item name="cpe:2.3:a:green_end:sftpserver:0.2.2:*:*:*:*:*:*:*">
<cpe-23:deprecated-by name="cpe:2.3:a:greenend:sftpserver:0.2.2:*:*:*:*:*:*:*" type="NAME_CORRECTION"/>
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 4adc8b3d24)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
While the list of licenses is present in the show-info output, the list
of license files and the redistributable status were missing. Add them.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 5647d5c190)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Disable lua bindings to avoid the following build failure if host-lua
and lua are built before brltty raised since bump to version 6.5 in
commit 82c6e87d5e and
https://github.com/brltty/brltty/commit/42bf48ca010d9163652cb0a014662fa243132736:
configure: lua package: lua
configure: lua includes:
configure: lua libs: -L/home/autobuild/autobuild/instance-0/output-1/host/bin/../sparc64-buildroot-linux-gnu/sysroot/usr/lib -llua -lm -ldl
checking for lua5.4... no
checking for lua... /home/autobuild/autobuild/instance-0/output-1/host/bin/lua
configure: Lua shell: /home/autobuild/autobuild/instance-0/output-1/host/bin/lua
configure: Lua library directory: /home/autobuild/autobuild/instance-0/output-1/host/lib/lua/5.3
[...]
/usr/bin/install -c -d /home/autobuild/autobuild/instance-0/output-1/target/home/autobuild/autobuild/instance-0/output-1/host/lib/lua/5.3
/usr/bin/install -c -s --strip-program /bin/true brlapi.so /home/autobuild/autobuild/instance-0/output-1/target/home/autobuild/autobuild/instance-0/output-1/host/lib/lua/5.3
make[3]: Leaving directory '/home/autobuild/autobuild/instance-0/output-1/build/brltty-6.6/Bindings/Lua'
[...]
brltty: installs files in /home/autobuild/autobuild/instance-0/output-1/target//home/autobuild/autobuild/instance-0/output-1
While at it, disable emacs bindings which have also been added in
version 6.5 with
https://github.com/brltty/brltty/commit/64a487e29941df54796645229a872ecbe2a06bfe
Fixes:
- http://autobuild.buildroot.org/results/f424e6727bbe61ea4b7703e93aae6fcab7506898
- http://autobuild.buildroot.org/results/bf4a161fc66fb99a24f08bea3436b13738f2e383
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 19f06441e8)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
package/wlroots/0001-Add-feature-macros-to-more-C-files-.patch has been
rejected upstream, and the initial pull request was moved to the
freedesktop gitlab. Update the patch to reflect these changes.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 840849afff)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Without the vulkan-loader package present, the configure step fails with the
following error message:
```
Run-time dependency vulkan found: NO (tried pkgconfig and system)
Message: Install "vulkan" or pass "-Dvulkan=disabled" to disable it.
```
The above error is due to the lack of the vulkan.pc file provided by the
vulkan-loader package. A search of autobuild failures containing
BR2_PACKAGE_WLROOTS=y and BR2_PACKAGE_MESA3D_VULKAN_DRIVER=y shows two build
failures. However, the failure reasons happened before wlroots could compile.
One for gerbera-1.10.0 and another for host-rust-1.64.0.
Add a dependency on the vulkan-loader package to resolve the above issue.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit f27b0322a5)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Before commit ca63464e37 (package/python-numpy: bump to version
1.25.0), numpy was using python distuils. Then, the default for BLAS and
LAPACK detection was to disable support if they were not found, even
when OPENBLAS is installed on the host system.
Commit ca63464e37 switched python-numpy to use the meson-package
infrastructure, where the default has changed, and now fails when the
host system does have OPENBLAS installed:
buildroot$ make
>>> host-python-numpy 1.25.0 Configuring
[...]
Run-time dependency openblas found: NO (tried pkgconfig and cmake)
WARNING: CMake Toolchain: Failed to determine CMake compilers state
Run-time dependency openblas found: YES 0.3.21
Dependency openblas found: YES 0.3.21 (cached)
Program _build_utils/process_src_template.py found: YES (/path/to/buildroot/output/host/bin/python3 /path/to/buildroot/output/build/host-python-numpy-1.25.0/numpy/_build_utils/process_src_template.py)
Program _build_utils/tempita.py found: YES (/path/to/buildroot/output/build/host-python-numpy-1.25.0/numpy/_build_utils/tempita.py)
output/build/host-python-numpy-1.25.0/numpy/meson.build:201:44: ERROR: Could not get cmake variable and no default provided for <CMakeDependency OpenBLAS: True None>
[...]
Two obvious ways to correct this:
- Add some kind of host-lapack / host-openblas dependency to
host-python-numpy, or
- Tell host-python-numpy not to look for (or rely on) external code.
Because the host numpy environment shouldn't be doing any numerical
heavy lifting, we opt for the latter here.
Signed-off-by: Graeme Smecher <gsmecher@threespeedlogic.com>
[yann.morin.1998@free.fr:
- explain why the issue was not found before
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit f6c33f3e89)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Use pkg-config to find libnet and avoid the following build failure
raised since commit 100ba8351c if
libnet-config is found on the host:
/home/autobuild/autobuild/instance-4/output-1/host/bin/mips64el-buildroot-linux-gnu-gcc -s -o ngrep ngrep.o tcpkill.o -lpcap -lnet -lpcre -L/usr/lib -lnet
mips64el-buildroot-linux-gnu-gcc: ERROR: unsafe header/library path used in cross-compilation: '-L/usr/lib'
While at it, format Upstream tag of other patches
Fixes:
- http://autobuild.buildroot.org/results/05c916162b9c1027cf53f1699a1f5653151098ed
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 14c0c37eaf)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Calling `setup.py` build builds and installs an empty package
after upstream changed their packaging. Calling setuptools via
`python -m build` builds the package correctly.
Signed-off-by: Marcus Hoffmann <bubu@bubu1.eu>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit d7e45215a4)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The http://pagesperso-orange.fr service being down,
http://pagesperso-orange.fr/sebastien.godard/ is no longer
accessible (or more precisely serves some HTML junk), meaning that we
currently cannot download the tarball of sysstat from the official
upstream location, and we always fallback to sources.buildroot.net.
This commit therefore updates to use the new upstream location on
github.io. The 12.6.1 release is available there, with the same hash,
so we can just point to the new upstream location.
Fixes:
ERROR: while checking hashes from package/sysstat//sysstat.hash
ERROR: sysstat-12.6.1.tar.xz has wrong sha1 hash:
ERROR: expected: a730982e0c2d4964a0022c1509f3ea0a345402bc
ERROR: got : d64a079729f2ff5b304294d68fbc041a46c4966d
ERROR: Incomplete download, or man-in-the-middle (MITM) attack
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 35a2cb0227)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Setting HAPROXY_CFLAGS on the haproxy build command line overrides CFLAGS
which were internally set by the haproxy Makefile.
Among those omitted CFLAGS is -fwrapv. Compiling haproxy without it and
and then running the program results in runtime error:
$ haproxy
FATAL ERROR: invalid code detected -- cannot go further, please recompile!
...
To address this issue, include HAPROXY_CFLAGS in the DEFINE variable instead
of CFLAGS in haproxy.mk.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 9b90ef6178)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit 363b4ebf1a set DBUS_CPE_ID_PRODUCT
to dbus but this is not needed as DBUS_CPE_ID_PRODUCT will be set to the
package name (i.e. dbus) by default
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit b878e7bbbc)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
cpe:2.3:a:flex_project:flex has been deprecated in favor of
cpe:2.3:a:westes:flex in October 2023:
<cpe-item name="cpe:/a:flex_project:flex:2.6.4" deprecated="true" deprecation_date="2023-10-06T17:12:20.950Z">
<title xml:lang="en-US">Flex Fast Lexical Analyser 2.6.4</title>
<reference href="https://github.com/westes/flex/releases">Version</reference>
<reference href="https://github.com/westes/flex/">Product</reference>
<cpe-23:cpe23-item name="cpe:2.3:a:flex_project:flex:2.6.4:*:*:*:*:*:*:*">
<cpe-23:deprecated-by name="cpe:2.3:a:westes:flex:2.6.4:*:*:*:*:*:*:*" type="NAME_CORRECTION"/>
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit c438422d49)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
cpe:2.3:a:json-c_project:json-c has been deprecated in favor of
cpe:2.3:a:json-c:json-c in September 2023:
<cpe-item name="cpe:/a:json-c_project:json-c:0.14" deprecated="true" deprecation_date="2023-09-16T13:05:58.643Z">
<title xml:lang="en-US">JSON-C Project JSON-C 0.14</title>
<reference href="https://github.com/json-c/json-c/releases">Version</reference>
<cpe-23:cpe23-item name="cpe:2.3:a:json-c_project:json-c:0.14:*:*:*:*:*:*:*">
<cpe-23:deprecated-by name="cpe:2.3:a:json-c:json-c:0.14-20200419:*:*:*:*:*:*:*" type="NAME_CORRECTION"/>
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 63118fbb55)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
cpe:2.3:a:libxpm_project:libxpm has been deprecated in favor of
cpe:2.3:a:x.org:libxpm in October 2023:
<cpe-item name="cpe:/a:libxpm_project:libxpm:3.5.15" deprecated="true" deprecation_date="2023-10-17T15:55:35.850Z">
<title xml:lang="en-US">Libxpm Project Libxpm 3.5.15</title>
<reference href="https://gitlab.freedesktop.org/xorg/lib/libxpm/-/tags">Version</reference>
<cpe-23:cpe23-item name="cpe:2.3:a:libxpm_project:libxpm:3.5.15:*:*:*:*:*:*:*">
<cpe-23:deprecated-by name="cpe:2.3:a:x.org:libxpm:3.5.15:*:*:*:*:*:*:*" type="NAME_CORRECTION"/>
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 03a9171cce)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
cpe:2.3:a:netatalk_project:netatalk has been deprecated in favor of
cpe:2.3:a:netatalk:netatalk in September 2023:
<cpe-item name="cpe:/a:netatalk_project:netatalk:3.1.13" deprecated="true" deprecation_date="2023-09-29T11:15:00.340Z">
<reference href="https://sourceforge.net/projects/netatalk/files/netatalk/">Version</reference>
<cpe-23:cpe23-item name="cpe:2.3:a:netatalk_project:netatalk:3.1.13:*:*:*:*:*:*:*">
<cpe-23:deprecated-by name="cpe:2.3:a:netatalk:netatalk:3.1.13:*:*:*:*:*:*:*" type="NAME_CORRECTION"/>
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 5ae38ec621)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit 8519de517e (package/{glibc, localedef}: security bump to version
glibc-2.38-27-g750a45a783906a19591fb8ff6b7841470f1f5701) correctly mentioned
CVE-2023-4806 in the commit message, but forgot to add an ignore for it.
Fix that.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 62b767fd3e)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit c07aafa087 (package/Makefile.in: set GIT_DIR=. in {HOST,
TARGET}_MAKE_ENV) added GIT_DIR=. to TARGET_MAKE_ENV (which is included in
TARGET_CONFIGURE_OPTS) to work around issues with packages getting confused
when building in a subdir of the Buildroot git repo.
This unfortunately also causes git commands to fail when
output/host/environment-setup is sourced:
git status
fatal: not a git repository: '.'
So strip GIT_DIR= from TARGET_CONFIGURE_OPTS when generating
environment-setup.
Reported-by: Mircea Gliga <gliga.mircea@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 48874afb9d)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following build failure without gbm raised since commit
534c22dd60:
Message: dmabuf-feedback requires gbm which was not found. If you rather not build this, drop "dmabuf-feedback" from simple-clients option.
Move the option assignment further down, below all the simple-clients
lists; in Makefile, and because we are usign simply expanded variables,
this is not necessary, but it is easier on us humans when we review the
code.
Also add a comment explaining why the initial list is incomplete.
Fixes:
- http://autobuild.buildroot.org/results/ebbba1d73ceeaacee17fde0c6c853415cd316091
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 611c0cb198)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The NuGet packaging description file is installed as:
$(DEST_DIR)/build/native/hiredis.targets
This is a sprurious file that has nothing to do on a Linux system,
whether that be in host/, staging/, or target/.
Backport an upstream patch to get rid of it.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Fabrice Fontaine <fontaine.fabrice@gmail.com>
(cherry picked from commit 52f3793d46)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes the following security issues:
1) CVE-2023-6816 can be triggered by passing an invalid array index to
DeviceFocusEvent or ProcXIQueryPointer.
2) CVE-2024-0229 can be triggered if a device has both a button and a
key class and zero buttons.
3) CVE-2024-21885 can be triggered if a device with a given ID was
removed and a new device with the same ID added both in the same
operation.
4) CVE-2024-21886 can be triggered by disabling a master device with
disabled slave devices.
5) CVE-2024-0409 can be triggered by enabling SELinux
xserver_object_manager and running a client.
6) CVE-2024-0408 can be triggered by enabling SELinux
xserver_object_manager and creating a GLX PBuffer.
For details, see the advisory:
https://lists.x.org/archives/xorg-announce/2024-January/003444.html
Switch to .tar.gz as the announcement mail only contained hashes for that:
https://lists.x.org/archives/xorg-announce/2024-January/003442.html
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 219178ef3e)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes the following security issues:
1) CVE-2023-6816 can be triggered by passing an invalid array index to
DeviceFocusEvent or ProcXIQueryPointer.
2) CVE-2024-0229 can be triggered if a device has both a button and a
key class and zero buttons.
3) CVE-2024-21885 can be triggered if a device with a given ID was
removed and a new device with the same ID added both in the same
operation.
4) CVE-2024-21886 can be triggered by disabling a master device with
disabled slave devices.
5) CVE-2024-0409 can be triggered by enabling SELinux
xserver_object_manager and running a client.
6) CVE-2024-0408 can be triggered by enabling SELinux
xserver_object_manager and creating a GLX PBuffer.
For details, see the advisory:
https://lists.x.org/archives/xorg-announce/2024-January/003444.html
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit b8d9e75eb8)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit e88823d667 (package/refpolicy: fix build with smartmontools) added
a 0001-policy-modules-services-smartmon.te-make-fstools-opt.patch patch, but
forgot to put it in the version specific sub directory - Breaking builds
using BR2_PACKAGE_REFPOLICY_CUSTOM_GIT as shown by the TestSELinuxCustomGit
test:
>>> refpolicy RELEASE_2_20200818 Extracting
gzip -d -c /builds/buildroot.org/buildroot/test-dl/refpolicy/refpolicy-RELEASE_2_20200818-br1.tar.gz | tar --strip-components=1 -C /builds/buildroot.org/buildroot/test-output/TestSELinuxCustomGit/build/refpolicy-RELEASE_2_20200818 -xf -
>>> refpolicy RELEASE_2_20200818 Patching
Applying 0001-policy-modules-services-smartmon.te-make-fstools-opt.patch using patch:
patching file policy/modules/services/smartmon.te
Hunk #1 FAILED at 143.
1 out of 1 hunk FAILED -- saving rejects to file policy/modules/services/smartmon.te.rej
make[1]: *** [package/pkg-generic.mk:241: /builds/buildroot.org/buildroot/test-output/TestSELinuxCustomGit/build/refpolicy-RELEASE_2_20200818/.stamp_patched] Error 1
https://gitlab.com/buildroot.org/buildroot/-/jobs/5929796183
Fix it by moving the patch to a versioned sub directory.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit bde468127c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
see CVE-2024-0553: Fix more timing side-channel inside RSA-PSK key exchange
see CVE-2024-0567: Fix assertion failure when verifying a certificate chain with a cycle of cross signatures
Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit b136bed2fd)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The help text is currently copy and pasted from the gstreamer video
player plugin help text. Change it to reflect the text from the
CMakeLists.txt file.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 7437cad018)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit 99a50a8c98 (package/flutter-pi: new package) erroneously made
the gstreamer-based audio plugin depend on GLES, although there is no
such requirement defined in the CMakeLists. This error was likely due to
a copy/paste mistake.
Remove the requirement.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
[yann.morin.1998@free.fr: reword commit log]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 07c1329814)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following build failure raised since the addition of the package
in commit 6aa1bc3167:
In file included from /home/buildroot/autobuild/run/instance-3/output-1/build/vulkan-loader-1.3.262/loader/extension_manual.h:24,
from /home/buildroot/autobuild/run/instance-3/output-1/build/vulkan-loader-1.3.262/loader/extension_manual.c:23:
/home/buildroot/autobuild/run/instance-3/output-1/host/aarch64-buildroot-linux-gnu/sysroot/usr/include/vulkan/vulkan.h:71:10: fatal error: X11/extensions/Xrandr.h: No such file or directory
71 | #include <X11/extensions/Xrandr.h>
| ^~~~~~~~~~~~~~~~~~~~~~~~~
Fixes:
- http://autobuild.buildroot.org/results/55ddfd44393e3bcc2f25bad2f9ecb7e1b142a985
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Reviewed-by: Adam Duskett <adam.duskett@amarulasolutions.com>
Tested-by: Adam Duskett <adam.duskett@amarulasolutions.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 3b8b1125ad)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The variable should be a YES/NO value, FALSE is not a valid value.
E.g. the yesno-to-bool cmd does not translate a FALSE value and therefore returns invalid JSON.
Signed-off-by: Maximilian Senftleben <maximilian.senftleben@frogblue-tec.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 9fa01e3097)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The MICROPYTHON_MAKE_ENV variable contained two things;
- the comon target environment variables CC, CFLAGS et al. defined in
TARGET_MAKE_ENV,
- the GIT_DIR workaround
Commit 9024e18665 (package/micropython: drop GIT_DIR=. workaround)
totally dropped the assignment to MICROPYTHON_MAKE_ENV, but did not
replace its expansin with TARGET_MAKE_ENV.
This yields build error like:
LINK build-standard/micropython
arm-linux-gcc: ERROR: unsafe header/library path used in cross-compilation: '-L/usr/lib64/../lib64'
Fix this by expanding TARGET_MAKE_ENV in lieu of MICROPYTHON_MAKE_ENV.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 6c9be611a8)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
See release notes (https://github.com/redis/redis/blob/7.2.4/00-RELEASENOTES):
================================================================================
Redis 7.2.4 Released Tue 09 Jan 2024 10:45:52 IST
================================================================================
Upgrade urgency SECURITY: See security fixes below.
Security fixes
==============
* (CVE-2023-41056) In some cases, Redis may incorrectly handle resizing of memory
buffers which can result in incorrect accounting of buffer sizes and lead to
heap overflow and potential remote code execution.
Bug fixes
=========
* Fix crashes of cluster commands clusters with mixed versions of 7.0 and 7.2 (#12805, #12832)
* Fix slot ownership not being properly handled when deleting a slot from a node (#12564)
* Fix atomicity issues with the RedisModuleEvent_Key module API event (#12733)
Signed-off-by: Titouan Christophe <titouanchristophe@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 5f59c175a4)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Remove and create the kernel_blob.bin file in the flutter_assets directory:
When Flutter compiles the source code, it creates the kernel_blob.bin file,
a Dart Kernel Bytecode of the application.
If kernel_blob.bin exists inside the flutter_assets directory, then the
application source code can be extracted with nothing more than apktool.
Not only is this a security risk, it's also redundant and a waste of space.
Removing the kernel_blob.bin file generated for flutter-gallery saves 86M!
Because this package is a reference package, and some applications may check
if the kernel_blob.bin file exists, we also touch a blank kernel_blob.bin
file.
Do not use relative symlinks:
Users may install their applications in any arbitaryt location, not
necessarily in the /usr/share/flutter/${package_name} directory. Because
flutter-gallery is a reference application, using exact symlinks to
icudtl.dat and libflutter_engine.so is preferable.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 0805b8048f)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As the flutter-gallery package is a reference package for users wishing to use
Flutter for their UX with Buildroot, this package must have the correct build
options. Indeed, this package currently starts and runs, but only because of
the 0001-remove-GetStorage.patch. Through testing, flutter-gallery fails to
run during the following scenario:
- The xdg-user-dirs package is ported and present.
- flutter-gallery depends on xdg-user-dirs.
- The 0001-remove-GetStorage.patch file is removed.
After extensive testing and comparing the current build arguments against what
the meta-flutter repository for Yocto passes to all of the applications that
inherit flutter-app, it is clear that handling the dart_plugin_registrant.dart
file is missing from the dart arguments in the flutter-gallery build step.
As the documentation for the dart_plugin_registrant.dart file is nonexistent
in any official documentation. However, there is a comment from an issue on
the official dart-lang/sdk page on Github that explains what this file is
(and refers to the Dark SDK source code instead of official documentation.)
From https://github.com/dart-lang/sdk/issues/52506#issuecomment-1562806787:
```
The dart_plugin_registrant.dart is a very special file. It's neither included
in the Dart app nor any dependent packages. Rather it's an artificially
created file by the flutter tools. It contains logic to run plugin
registration logic.
A flutter build will eventually compile the Dart application where it will add
<dir>/.dart_tool/flutter_build/dart_plugin_registrant.dart as an extra source
file (see here). Additionally it will also inject that uri as a constant into
Dart source code via a -Dflutter.dart_plugin_registrant=<uri>.
Once the app runs it will access the
package:flutter/src/dart_plugin_registrant.dart:dartPluginRegistrantLibrary
constant and use it to look up the library object and then invoke the plugin
registration logic.
```
Now that what the dart_plugin_registrant.dart does is understood, we need to
pass the following to the dart binary during the flutter-gallery build step:
-Dflutter.dart_plugin_registrant=file://[...]/dart_plugin_registrant.dart:
Injects a file containing the logic to run the plugin registration logic as
a constant into the flutter-application source code.
--source file://$(@D)/.dart_tool/flutter_build/dart_plugin_registrant.dart:
Adds the dart_plugin_registrant.dart file as a source file to compile.
--source package:flutter/src/dart_plugin_registrant.dart:
Binds the plugin implementation to the platform interface based on the
configuration of the app's pubpec.yaml, and the plugin's pubspec.yaml.
The native_assets.yaml file provides the native-assets mapping for
@Native external functions. The flutter-gallery package has no functions
marked as @Native; however, calling "flutter build bundle" creates a blank
template "native_assets.yaml" file, which is safe to include in the build.
This line, while not necessary for flutter-gallery, may be helpful for other
users who use @Native external functions in their applications, and this
example makes porting other applications quicker and easier.
Finally, there is a known issue when using the dart_plugin_registrant.dart
file outlined here: https://github.com/flutter/flutter/issues/137972.
To summarize: If a user fails to pass the --obfuscate flag to gen_snapshsot
when using the dart_plugin_registrant.dart file, their application may fail
to start. One such application is Gallery, which I have independently verified.
As such, pass the --obfuscate flag to gen_snapshot to ensure that
flutter-gallery properly starts when building with the additional
dart_plugin_registrant.dart arguments above.
However, I acknowledge that the obfuscate flag hides function and class names
in compiled Dart code, and there are some cases when a user should avoid using
the flag. For example, when using the runtimeType API:
https://api.flutter.dev/flutter/dart-core/Object/runtimeType.html. However,
this is not the case with flutter-gallery, and the --obfuscate flag is needed.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
[yann.morin.1998@free.fr: restore FLUTTER_RUNTIME_MODES]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit a821aee2d2)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
To make the next patch easier to understand, add a configure step to the
flutter-gallery.mk file. The `$(HOST_FLUTTER_SDK_BIN_FLUTTER) build bundle`
command is placed in the configure step because it generates several files
needed for the flutter_gen_snapshot tool to generate the target .so file.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 562ed3dfd4)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The canonical name for the 'pub' action is 'pub', while 'packages' is
just an alias. Switch to using the canonical name.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
[yann.morin.1998@free.fr: split off to its own commit]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 1cc2c4bb86)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As Yann mentioned in commit 9d8497e79d, a
dependency on a glibc toolchain implies a dependency on
BR2_TOOLCHAIN_HAS_THREADS_NPTL. As such, remove the dependency from
package/flutter-engine/Config.in.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 019cfcfc4a)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The current depot-tools version is from Jul 13 2023, and there is a bug
where when building with PPD enabled, building a Flutter application on
my laptop running Fedora 39 or even in the Debian 11 container fails due
to permission issues when running rsync such as:
/usr/bin/sed: couldn't open temporary file [...]/output/per-package/flutter-gallery/host/share/flutter/sdk/.vpython-root/0e1e32/lib/python3.8/sed763MrF: Permission denied
make: *** [package/pkg-generic.mk:267: [...]/output/build/flutter-gallery-2.10.2/.stamp_configured] Error 123
After several hours of attempting to find the exact cause of the issue
by parsing git logs of the depot_tools repository and cross-referencing
bugs on https://bugs.chromium.org/p/chromium/issues I am unable to
determine the exact reason why the problem occurs or the exact commit
that fixes the issue.
However, updating depot-tools does indeed fix the issue, and looking at
the source code between 4e87f5bf and 8d14454b shows improvements to
several bash scripts, such as gclient, vpython3, and spid. All of which
could be culprits of this bug. As flutter-engine is the only package
that uses host-depot-tools, it is safe to update.
This permission bug is present on PPD and normal builds, but causes a
build to crash only on PPD because of rsyncing. On non-ppd builds, the
bug only causes some minor annoyance when trying to run `make clean`
because permission denials crop up when trying to remove the
.vpython-root directory.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
[yann.morin.1998@free.fr:
- copy the cover-letter description as commit log
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 2b7b68cfe5)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
When running the command "flutter pub get," the plugins are stored in the
pub-cache directory along with their sha256sum hashes. The default location of
the pub-cache directory is current $(HOST_DIR)/share/flutter/sdk/.pub-cache,
which is not an acceptable choice by default because every plugin is
re-downloaded during every build of a flutter application either during a new
build or when building with the per-package-directory option enabled.
Furthermore, keeping the pub-cache in its current location prevents users from
committing the pub-cache directory to git for faster rebuilds of a
Buildroot-based system, as users cannot store the pub-cache for later use.
To fix the above issue completely, the following two changes must occur:
- Change the hard-coded Flutter pub-cache location to
$(DL_DIR)/br-flutter-pub-cache.
- Remove the `rm -rf $(HOST_FLUTTER_SDK_BIN_SDK)/.pub-cache` and the
associated comment about why the build system removes the .pub-cache
directory. After further research, the help text of the precache command
reads, "Populate the Flutter tool's cache of binary artifacts."
The current reasoning listed in the comments is not accurate for a
the following reasons:
1. We do not want to remove their directory if users already have a pub
cache they have symlinked to.
2. If the flutter-sdk-bin package previously set up the pub-cache, then
the pub-cache directory is set up with the options we want, and there
is no reason to remove the pub-cache directory.
Note that upstream considers it safe to have multiple instances of
readers/writers to the pub cache concurently, which is a situation that
can happen when two flutter-based pacakges are going to be built in
parallel. There have been reports upstream [0] [1] [2] where concurrency
was an issue, and they have always been fixed [3] [4] (or considered
fixed already). So we can assune that, if the conncurrent ccess to the
shared pub-cache causes issues, that will be an upstream bug that will
get solved.
If that turns out to be an unsolvable problem, we'll still have the
option to run the pub-get commands under flock.
[0] https://github.com/dart-lang/pub/issues/1178
[1] https://github.com/dart-lang/pub/issues/3404
[2] https://github.com/dart-lang/pub/issues/3420
[3] https://github.com/dart-lang/pub/issues/1178#issuecomment-1006489414
[4] https://github.com/dart-lang/pub/issues/1178#issuecomment-1007273739
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
[yann.morin.1998@free.fr: add blurb about concurrent access]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 37809258fc)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As found out by Yann E. Morin in [1], apcupsd configure script is ugly,
and uses gcc to do the link line-wrapping which will raise the following
build failure with gcc 13:
/home/buildroot/autobuild/instance-1/output-1/host/opt/ext-toolchain/bin/../lib/gcc/powerpc64le-buildroot-linux-gnu/13.2.0/../../../../powerpc64le-buildroot-linux-gnu/bin/ld: /home/buildroot/autobuild/instance-1/output-1/host/opt/ext-toolchain/bin/../lib/gcc/powerpc64le-buildroot-linux-gnu/13.2.0/../../../../powerpc64le-buildroot-linux-gnu/lib/../lib64/libsupc++.a(eh_alloc.o): in function `std::basic_string_view<char, std::char_traits<char> >::compare(unsigned long, unsigned long, char const*, unsigned long) const':
eh_alloc.cc:(.text._ZNKSt17basic_string_viewIcSt11char_traitsIcEE7compareEmmPKcm[_ZNKSt17basic_string_viewIcSt11char_traitsIcEE7compareEmmPKcm]+0x44): undefined reference to `std::__throw_out_of_range_fmt(char const*, ...)'
It will also raise the following build failure on sparc/arc:
/home/autobuild/autobuild/instance-3/output-1/host/opt/ext-toolchain/bin/../lib/gcc/arceb-snps-linux-uclibc/9.2.1/../../../../arceb-snps-linux-uclibc/bin/ld: /home/autobuild/autobuild/instance-3/output-1/host/arceb-buildroot-linux-uclibc/sysroot/lib/libsupc++.a(eh_throw.o): in function `__exchange_and_add_dispatch':
/SCRATCH/arcjenkins2/slaves/ru20-custom-arcgnu2/workspace/arcoss_verification/arc_gnu_toolchain_release/arc_gnu_toolchain_release/bd-uclibceb/gcc-stage2/arceb-snps-linux-uclibc/libstdc++-v3/include/ext/atomicity.h:82: undefined reference to `__gnu_cxx::__exchange_and_add(int volatile*, int)'
/home/autobuild/autobuild/instance-3/output-1/host/opt/ext-toolchain/bin/../lib/gcc/arceb-snps-linux-uclibc/9.2.1/../../../../arceb-snps-linux-uclibc/bin/ld: /SCRATCH/arcjenkins2/slaves/ru20-custom-arcgnu2/workspace/arcoss_verification/arc_gnu_toolchain_release/arc_gnu_toolchain_release/bd-uclibceb/gcc-stage2/arceb-snps-linux-uclibc/libstdc++-v3/include/ext/atomicity.h:82: undefined reference to `__gnu_cxx::__exchange_and_add(int volatile*, int)'
Instead of trying to patch the configure script as advocated by
Yann E. Morin, set LD to TARGET_CXX as:
- this solution is quicker
- usptream is dead (last release in 2016)
- this solution has already been used in other packages (nodejs, zmqpp)
[1]: https://patchwork.ozlabs.org/project/buildroot/patch/20200812171821.2517-1-Evgeniy.Didin@synopsys.com/
Fixes:
- http://autobuild.buildroot.org/results/6096c3ddc5edf3204635c2c90246c2e8c8e074e7
- http://autobuild.buildroot.org/results/d8a/d8a3ab31c5b86871c7e1117f4ffa7b6cedfcb7e0/build-end.log
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit cd2dcaa6c6)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The condition checking whether the webrtc-audio-processing package is
enabled, added in commit
3ccd3b4c38 ("package/pipewire: bump to
version 0.3.32") is obviously incorrect, and can never be true.
Fix the condition to use the correct variable instead.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 52f8db409f)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
A number of packages try to detect if they are running in a git repo and run
git describe at build time instead of using the hard coded version number if
it succeed, leading to odd version numbers as they end up picking up the
Buildroot git version if building inside a Buildroot git checkout, E.G.:
rauc --version
rauc 2023.11-562-g9c954953b4+
This is because rauc builds with meson and uses vcs_tag:
https://github.com/rauc/rauc/blob/v1.11/meson.build#L168-L171https://mesonbuild.com/Reference-manual_functions.html#vcs_tag
Another example is micropython, where we already work around it by passing
GIT_DIR=.
In the context of Buildroot the packages are never built in their own git
checkout, so pass GIT_DIR=. to ensure git doesn't walk back up the
directory tree and finds the Buildroot git repo, which fixes the rauc (and
similar) issues.
>>> rauc 1.11 Building
..
ninja: Entering directory `/home/peko/source/buildroot/output-rauc/build/rauc-1.11//build'
[1/29] Generating version.h with a custom command
fatal: not a git repository: '.'
cat output-rauc/build/rauc-1.11/build/version.h
#define PACKAGE_STRING "rauc 1.11"
#define PACKAGE_VERSION "1.11"
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit c07aafa087)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Building for i386 raises the following build failure since the additon
of the package in commit 3e4b479f39:
Makefile:23: *** "The architecture i386 isn't supported". Stop.
Setting ARCH=x86 won't work either as it results in the following build
failure:
compel/arch/x86/plugins/std/memcpy.S: Assembler messages:
compel/arch/x86/plugins/std/memcpy.S:20: Error: bad register name `%rdi'
compel/arch/x86/plugins/std/memcpy.S:21: Error: bad register name `%rdx'
compel/arch/x86/plugins/std/memcpy.S:22: Error: `shrq' is only supported in 64-bit mode
compel/arch/x86/plugins/std/memcpy.S:24: Error: `movsq' is only supported in 64-bit mode
compel/arch/x86/plugins/std/syscalls/syscall-common-x86-64.S: Assembler messages:
compel/arch/x86/plugins/std/syscalls/syscall-common-x86-64.S:13: Error: bad register name `%rcx'
compel/arch/x86/plugins/std/syscalls/syscall-common-x86-64.S:19: Error: bad register name `%rax'
Fixes:
- http://autobuild.buildroot.org/results/94cc463762b57efacf743d107a8dda7660a995a3
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit bb3ede3b36)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit 768f9f80f6 (support/download: generate even more reproducible
tarballs) causes non-reproducibility in tarballs we previousy
generated, especially the archives for two cargo-vendored packages,
ripgrep and sentry-cli.
The cause is that those two pakcages eventually vendor a file that has
the u+x bit set, but is otehrwise go-x. With 768f9f80f6, the files are
now go+x, so the hash for those generated archives has changed.
Besides, that commit was wrong: it did not account for the 'r' bit for
go part, leaving some non-reproducibility still unaccounted for.
So, to generate really reproducible archives, we would need to fix that
read bit as well, and that has the potential to affect all the archives
we generated so far. If we wanted to do so, we'd need a way to version
all generated archives, like we do for git and svn, but now for all the
different CVSes, as well as for all the vendoring post-processes.
For 768f9f80f6, all that was of conern was the working copies of CVSes
(i.e. git, svn, cvs...) that we cache in the Buildroot download dir, not
the temporary files during post-processing. Indeed, in that latter case,
the user has virtually no way to mangle with the mode of the
intermediate extract before repack.
And we do have a big fat warning that users should not attempt to meddle
with the git tree that Buildroot caches.
As 768f9f80f6 however demonstrates, is that it took quite a long time
between the introduction of the git caching, and the time someone
eventually discovered they could meddle in there. This shows that the
issue it not actually critical in most setups.
Also, the tar manual [0] hints at a better solution to handle
reproducibility, which even avoids touching the files on disk which is
even nicer:
‘--mode='go+u,go-w'’
Omit irrelevant information about file permissions.
If we were to actually handle the mode bit for reproducibility, we'd
need to:
- introduce archive versioning for all download backends and
prost-processing
- use the tar officially suggested method
So, revert that change, as it was incomplete, was not really fixing much
issues, and causes actual issues.
This reverts commit 768f9f80f6.
[0] https://www.gnu.org/software/tar/manual/tar.html#Reproducibility
Thanks to Vincent and Arnout for pointing at the tar manual.
Reported-by: Antoine Coutant <antoine.coutant@smile.fr>
Reported-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Vincent Fazio <vfazio@xes-inc.com>
Cc: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Tested-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Antoine Coutant <antoine.coutant@smile.fr>
(cherry picked from commit 9fbd3d8574)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit 6d163e12a4 (package/udev: move render and sgx to
package/systemd) moved the sgx group creation to the systemd package because
eudev at that time did not reference it. This changed in eudev 3.1.12 with
commit a8ffcd1b985fb4 (rules/50-udev-default.rules: fix issue 160) so move
it back to get rid of a warning from udevd:
udevd[303]: specified group 'sgx' unknown
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit edfa9ea45c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
flutter-gallery was erroneously placed under the "Graphic libraries"
section of the menu "Graphic libraries and applications (graphic/text)"
menu. However, as flutter-gallery is a flutter-based graphical user
interface (GUI) application, it is better suited to be placed under the
"Graphic applications" section.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 75d78e4225)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
flutter-pi was erroneously placed under the "Graphic libraries" section
of the menu "Graphic libraries and applications (graphic/text)" menu.
However, as flutter-pi is an application that runs graphic applicaitons
it is better suited to be placed under the "Graphic applications"
section.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 1a2ae469d0)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following build failure on sparc64 raised since bump to version
0.85 in commit 470f0fb1ec:
error[E0308]: mismatched types
--> /home/autobuild/autobuild/instance-7/output-1/build/nushell-0.85.0/VENDOR/uucore/src/lib/features/fs.rs:121:16
|
111 | pub fn number_of_links(&self) -> u64 {
| --- expected `u64` because of return type
...
121 | return self.0.st_nlink;
| ^^^^^^^^^^^^^^^ expected `u64`, found `u32`
|
help: you can convert a `u32` to a `u64`
|
121 | return self.0.st_nlink.into();
| +++++++
For more information about this error, try `rustc --explain E0308`.
error: could not compile `uucore` (lib) due to previous error
Fixes:
- http://autobuild.buildroot.org/results/f9f0287a8e39c65895014ca513ed25071f020add
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit b7c163f190)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- Use official tarball
- Update hash of license file (some packages have been added or removed
but the list of licenses is the same)
- Fix CVE-2023-7158: A vulnerability was found in MicroPython up to
1.21.0. It has been classified as critical. Affected is the function
slice_indices of the file objslice.c. The manipulation leads to
heap-based buffer overflow. It is possible to launch the attack
remotely. The exploit has been disclosed to the public and may be
used. Upgrading to version 1.22.0 is able to address this issue. It is
recommended to upgrade the affected component. The identifier of this
vulnerability is VDB-249180.
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 1e12b7dd49)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Update site to avoid redirections (HSTS, etc.)
Version 5.0.3 - 12/17/2023
A memory leak fix in the prior version wasn't applied correctly, resulting
in an invalid memory access causing a crash. Bug fixed.
Version 5.0.2 - 11/8/2023
Fixed bug that caused crash when a CLIENT_KEY arrived out of order
Fixed option handling on Windows when an argument is missing
https://sourceforge.net/projects/uftp-multicast/files/Changes.txt/download
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit bfe2fe2269)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This usage of <PKG>_NAME was introduced in commit f9e9c6349a
("package/rng-tools: bump to 6.7"). No other package uses <PKG>_NAME
this way.
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit a2b8596873)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following build failure without neon raised since bump to
version 1.4.0 in commit 2f7f8f3813 and
https://github.com/jgaeddert/liquid-dsp/commit/c821187dd9af2758c8e3c82e95eaf041b8ac81a0:
/home/peko/autobuild/instance-0/output-1/host/bin/arm-none-linux-gnueabi-gcc -std=gnu11 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -g0 -D_FORTIFY_SOURCE=2 -ffast-math -mcpu=cortex-a7 -mfloat-abi=hard -mfpu=neon-vfpv4 -Wall -fPIC -Wno-deprecated -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -I. -Iinclude -c -o src/audio/src/cvsd.o src/audio/src/cvsd.c
In file included from /home/peko/autobuild/instance-0/output-1/host/arm-buildroot-linux-gnueabi/sysroot/usr/include/features.h:388:0,
from /home/peko/autobuild/instance-0/output-1/host/arm-buildroot-linux-gnueabi/sysroot/usr/include/stdlib.h:24,
from src/libliquid.c:25:
/home/peko/autobuild/instance-0/output-1/host/arm-buildroot-linux-gnueabi/sysroot/usr/include/gnu/stubs.h:10:29: fatal error: gnu/stubs-hard.h: No such file or directory
# include <gnu/stubs-hard.h>
^
Indeed, upstream considers that NEON is available on all ARM platforms,
and their configure.ac contains that code snippet:
239 arm|armv7*|armv8*)
240 # assume neon instructions are available
241 # TODO: check for Neon availability
242
243 # ARM architecture : use neon extensions
Fixes:
- http://autobuild.buildroot.org/results/36b3c2220c462e7a20262fd1b9064d9aeb6c9ec4
- http://autobuild.buildroot.org/results/881826b4b6c141e59a0da2d7d1ad55d3709fdb95
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
[yann.morin.1998@free.fr:
- refactor with LIQUID_DSP_SIMDOVERRIDE
- add comment about --disable-simdoverride
- extend commit log with upstream code snippet
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 9501bc80f5)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
libaio is only needed for standard install
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
[yann.morin.1998@free.fr: fix check-package]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit ee9c92e4a4)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
--{dis,en}able-avahi is unrecognized since bump to version 2.3.3op2 in
commit 8cf034ab0f (which switched upstream
location from apple to openprinting):
configure: WARNING: unrecognized options: --disable-gtk-doc, --disable-gtk-doc-html, --disable-doc, --disable-docs, --disable-documentation, --with-xmlto, --with-fop, --disable-dependency-tracking, --enable-ipv6, --disable-nls, --disable-systemd, --disable-avahi
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 36743d6175)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
With [1], [2] & [3] we made sure Buildroot packages get built with
proper MMU page size assumed. This was done nicely through insertion of
required flags into the toolchain wrapper so that there's no need to
pass these flags to each and every package separately - toolchain
wrapper used for real building has all set internally and so proper
flags are implicitly used.
But there's yet another corner case which is not handled that way -
these are binaries or rather libraries which are being used as a part of
GCC compilation: libgcc_s.so.1 and libstdc++.so.
And so to make sure both the libraries get built properly we need to
set TARGET_CFLAGS (cures libgcc_s.so) & TARGET_LDFLAGS (cures
libstdc++.so).
In case of ARM by defaut 64 KiB page size seems to be used, as w/o
that patch we see the following for BR2_ARM64_PAGE_SIZE_4K=y:
--------------------------->8----------------------------
$ ./output/host/bin/aarch64-linux-readelf -l ./output/target/lib/libgcc_s.so.1
Elf file type is DYN (Shared object file)
Entry point 0x0
There are 6 program headers, starting at offset 64
Program Headers:
Type Offset VirtAddr PhysAddr
FileSiz MemSiz Flags Align
LOAD 0x0000000000000000 0x0000000000000000 0x0000000000000000
0x0000000000013d1c 0x0000000000013d1c R E 0x10000
LOAD 0x000000000001fd98 0x000000000002fd98 0x000000000002fd98
0x0000000000000438 0x00000000000005c8 RW 0x10000
DYNAMIC 0x000000000001fdb8 0x000000000002fdb8 0x000000000002fdb8
0x0000000000000200 0x0000000000000200 RW 0x8
$ ./output/host/bin/aarch64-linux-readelf -l ./output/target/usr/lib/libstdc++.so.6.0.32
Elf file type is DYN (Shared object file)
Entry point 0x0
There are 7 program headers, starting at offset 64
Program Headers:
Type Offset VirtAddr PhysAddr
FileSiz MemSiz Flags Align
LOAD 0x0000000000000000 0x0000000000000000 0x0000000000000000
0x00000000001d3462 0x00000000001d3462 R E 0x10000
LOAD 0x00000000001d5760 0x00000000001e5760 0x00000000001e5760
0x000000000000e528 0x0000000000012de8 RW 0x10000
DYNAMIC 0x00000000001deef0 0x00000000001eeef0 0x00000000001eeef0
0x0000000000000240 0x0000000000000240 RW 0x8
--------------------------->8----------------------------
Note alignment of 0x10000 in sections marked for loading.
And with the patch applied we get expected alignment of 0x1000 (4
KiB):
--------------------------->8----------------------------
$ ./output/host/bin/aarch64-linux-readelf -l ./output/target/lib/libgcc_s.so.1
Elf file type is DYN (Shared object file)
Entry point 0x0
There are 6 program headers, starting at offset 64
Program Headers:
Type Offset VirtAddr PhysAddr
FileSiz MemSiz Flags Align
LOAD 0x0000000000000000 0x0000000000000000 0x0000000000000000
0x0000000000013d1c 0x0000000000013d1c R E 0x1000
LOAD 0x0000000000013d98 0x0000000000014d98 0x0000000000014d98
0x0000000000000438 0x00000000000005c8 RW 0x1000
DYNAMIC 0x0000000000013db8 0x0000000000014db8 0x0000000000014db8
0x0000000000000200 0x0000000000000200 RW 0x8
$ ./output/host/bin/aarch64-linux-readelf -l ./output/target/usr/lib/libstdc++.so.6.0.32
Elf file type is DYN (Shared object file)
Entry point 0x0
There are 7 program headers, starting at offset 64
Program Headers:
Type Offset VirtAddr PhysAddr
FileSiz MemSiz Flags Align
LOAD 0x0000000000000000 0x0000000000000000 0x0000000000000000
0x00000000001d3462 0x00000000001d3462 R E 0x1000
LOAD 0x00000000001d3760 0x00000000001d4760 0x00000000001d4760
0x000000000000e528 0x0000000000012de8 RW 0x1000
DYNAMIC 0x00000000001dcef0 0x00000000001ddef0 0x00000000001ddef0
0x0000000000000240 0x0000000000000240 RW 0x8
--------------------------->8----------------------------
A nice side effect is that we can get rid of the special handling of
"-matomic" as it's already part of ARCH_TOOLCHAIN_WRAPPER_OPTS.
[1] https://git.buildroot.net/buildroot/commit/?id=3cc2c6d19ab2e1bb4634f26f9318da9b07df5fff
[2] https://git.buildroot.net/buildroot/commit/?id=dcb74db89e74e512e36b32cea6f574a1a1ca84c4
[3] https://git.buildroot.net/buildroot/commit/?id=5e52c28397b79f8c4c99552217cbe95202166626
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Signed-off-by: Vladimir Isaev <VVIsaev@gmail.com>
Signed-off-by: Pavel Kozlov <kozlov@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 747dff5a36)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- Drop all patches except first one (already in version)
- This bump will fix the following build failure with kernel >= 6.6:
/home/autobuild/autobuild/instance-2/output-1/build/dahdi-linux-3.2.0/drivers/dahdi/wct4xxp/base.c: In function ‘free_wc’:
./include/linux/workqueue.h:639:9: error: call to ‘__warn_flushing_systemwide_wq’ declared with attribute warning: Please avoid flushing system-wide workqueues. [-Werror=attribute-warning]
639 | __warn_flushing_systemwide_wq(); \
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/home/autobuild/autobuild/instance-2/output-1/build/dahdi-linux-3.2.0/drivers/dahdi/wct4xxp/base.c:2025:9: note: in expansion of macro ‘flush_scheduled_work’
2025 | flush_scheduled_work();
| ^~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
https://github.com/asterisk/dahdi-linux/releases/tag/v3.3.0
Fixes:
- http://autobuild.buildroot.org/results/e9755e1f4814b6b0c151c590b5c34acfd89556ad
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit a608e519a0)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following musl build failure with powerpc raised since bump to
version 2.14 in commit c6addf4606:
In file included from fault.h:36,
from handler-unix.c:77,
from handler.c:19:
handler-unix.c: In function 'sigsegv_handler':
fault-linux-powerpc.h:35:73: error: 'mcontext_t' has no member named 'uc_regs'; did you mean 'gregs'?
35 | # define SIGSEGV_FAULT_STACKPOINTER ((ucontext_t *) ucp)->uc_mcontext.uc_regs->gregs[1]
| ^~~~~~~
handler-unix.c:157:43: note: in expansion of macro 'SIGSEGV_FAULT_STACKPOINTER'
157 | uintptr_t old_sp = (uintptr_t) (SIGSEGV_FAULT_STACKPOINTER);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~
Fixes:
- http://autobuild.buildroot.org/results/77b600071f07605be3ec28e2da46d6938e240087
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 74f401025d)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This reverts commit 245b13a077 as docker
selinux module is for docker-engine, not for "a system tray dock for X"
Moreover, it raises the following build failure:
Compiling targeted policy.33
env LD_LIBRARY_PATH="/home/buildroot/autobuild/instance-0/output-1/per-package/refpolicy/host/lib:/home/buildroot/autobuild/instance-0/output-1/per-package/refpolicy/host/usr/lib" /home/buildroot/autobuild/instance-0/output-1/per-package/refpolicy/host/usr/bin/checkpolicy -c 33 -U deny -S -O -E policy.conf -o policy.33
policy.conf:1912:ERROR 'attribute container_engine_domain is not declared' at token ';' on line 1912:
type dockerd_t, container_engine_domain;
type dockerd_exec_t;
Fixes:
- http://autobuild.buildroot.org/results/87d78b6f15875f0fa3e6fc85e352db14ab0383bb
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 3e91de6428)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Yann reported in [1] that edk2 build could sometimes fail. The issue
can be reproduced when per-package directories is enabled, or also
when building on a system with GNU Make >= 4.4 using the
"--shuffle=reverse" option (such as Fedora 39). Those are pointing
toward a Makefile dependency issue.
The issue can be reproduced with commands:
cat > .config <<EOF
BR2_riscv=y
BR2_RISCV_64=y
BR2_TOOLCHAIN_EXTERNAL=y
BR2_TARGET_EDK2=y
EOF
make olddefconfig
Then, building either with:
make --shuffle=reverse
Or:
utils/config -e BR2_PER_PACKAGE_DIRECTORIES
make olddefconfig
make -j$(nproc)
It is interesting to mention that when using "make --shuffle=reverse"
to build, the build can be completed if restarted only with "make". It
will not pull any other Buildroot package. This fact hints toward a
Makefile dependency issue internal to the EDK2 build system, rather
than in the Buildroot recipe.
The EDK2 build system is quite unique. See [2]. It generates files,
makefiles and internally uses GNU Make to compile code. This system is
likely not tested as being a sub-Make process in a complex Makefile
such as Buildroot.
In order to prevent Buildroot to pass unexpected Make flags to the
EDK2 sub-Make, this commit unset the MAKEFLAGS variable in the EDK2
build environment. This will put the EDK2 build script in a more
common and tested state. See GNU Make documentation about recursive use
of Make, more specifically [3].
Note: as mentioned, the build failure is likely due to an internal
issue of the EDK2 build system. The failure points to a missing
dependency in the EDK2 generator itself. This commit does not fix this
issue, but rather put the EDK2 build system in a normalized
environment, avoiding Buildroot flags being passed to the internal
EDK2 sub-Make invocation. The upstream EDK2 build system most likely
need a fix too.
Fixes:
make[2]: *** No rule to make target '/buildroot/output/build/edk2-edk2-stable202308/Build/RiscVVirtQemu/RELEASE_GCC5/RISCV64/MdeModulePkg/Application/UiApp/UiApp/DEBUG/UiApp.efi', needed by '/buildroot/output/build/edk2-edk2-stable202308/Build/RiscVVirtQemu/RELEASE_GCC5/FV/Ffs/462CAA21-7614-4503-836E-8AB6F4662331UiApp/UiApp.offset'. Stop.
build.py...
: error 7000: Failed to execute command
make tbuild [/buildroot/output/build/edk2-edk2-stable202308/Build/RiscVVirtQemu/RELEASE_GCC5/RISCV64/MdeModulePkg/Application/UiApp/UiApp]
build.py...
: error F002: Failed to build module
/buildroot/output/build/edk2-edk2-stable202308/MdeModulePkg/Application/UiApp/UiApp.inf [RISCV64, GCC5, RELEASE]
[1] https://lists.buildroot.org/pipermail/buildroot/2023-December/681507.html
[2] https://tianocore-docs.github.io/edk2-BuildSpecification/draft/4_edk_ii_build_process_overview/42_build_process_overview.html
[3] https://www.gnu.org/software/make/manual/make.html#Options_002fRecursion
Reported-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 44af6938fb)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes the following security issues:
1) CVE-2023-6377: X.Org server: Out-of-bounds memory write in XKB button actions
A device has XKB button actions for each button on the device. When a
logical device switch happens (e.g. moving from a touchpad to a mouse), the
server re-calculates the information available on the respective master
device (typically the Virtual Core Pointer). This re-calculation only
allocated enough memory for a single XKB action rather instead of enough for
the newly active physical device's number of button. As a result, querying
or changing the XKB button actions results in out-of-bounds memory reads and
writes.
This may lead to local privilege escalation if the server is run as root or
remote code execution (e.g. x11 over ssh).
2) CVE-2023-6478: X.Org server: Out-of-bounds memory read in
RRChangeOutputProperty and RRChangeProviderProperty
This fixes an OOB read and the resulting information disclosure.
Length calculation for the request was clipped to a 32-bit integer. With
the correct stuff->nUnits value the expected request size was truncated,
passing the REQUEST_FIXED_SIZE check.
The server then proceeded with reading at least stuff->nUnits bytes
(depending on stuff->format) from the request and stuffing whatever it finds
into the property. In the process it would also allocate at least
stuff->nUnits bytes, i.e. 4GB.
See also CVE-2022-46344 where this issue was fixed for other requests.
For more details, see the advisory:
https://lists.x.org/archives/xorg-announce/2023-December/003435.html
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 9b62f5905e)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since commit 5770a645a3 ("package/qt5:
bump packages to latest kde submodule versions"), the
QT_HEADERS_SYNC_HOOK hook no longer calls the syncqt.pl script, so
host-perl is no longer needed as a dependency of running this
hook (and as a dependency of building Qt).
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit b678091a1c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
BR2_DOWNLOAD_FORCE_CHECK_HASHES currently has the following
dependency:
depends on BR2_GLOBAL_PATCH_DIR != ""
However, strictly speaking checking all hashes does not necessarily
require using BR2_GLOBAL_PATCH_DIR, as long as you don't use custom
versions.
But more importantly:
- Having this dependency means that this options is hidden when people
don't use BR2_GLOBAL_PATCH_DIR. Instead the option should always be
made visible, encouraging people to turn it on.
- The Config.in comment was there to mitigate this previous argument,
but this comment then shows up all the time when you have an empty
global patch dir.
This seems over-complicated, and it sounds much easier to have the
option unconditionally available, and visible, and clarify in its help
text that in order to this to work fully with custom package versions,
BR2_GLOBAL_PATCH_DIR can be used to provide extra hash files.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[yann.morin.1998@free.fr:
- fix typo noticed by Peter K.
- reword kast sentence after review by Peter K.
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 5b0c02a77a)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
With upstram commit [1] (since version v0.1.0) the pipeline option 'raspberrypi'
was renamed to 'rpi/vc4'.
Change the buildroot option name from BR2_PACKAGE_LIBCAMERA_PIPELINE_RASPBERRYPI
to BR2_PACKAGE_LIBCAMERA_PIPELINE_RPI_VC4 (and add Config.in.legacy entry
accordingly) and move handling in Config.in/libcamer.mk to follow alphabetic
ordering.
Fixes:
.../build/libcamera-v0.1.0/meson.build:3:0: ERROR: Options "raspberrypi" are not in allowed choices: "all, auto, imx8-isi, ipu3, rkisp1, rpi/vc4, simple, uvcvideo, vimc"
[1] https://git.libcamera.org/libcamera/libcamera.git/commit/?id=726e9274ea95fa46352556d340c5793a8da51fcd
Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 782d268aba)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This reverts commit c9645fd29b.
Building libcamera-apps 1.3.0 with current libcamera 0.1.0 fails because
some of the symbols like controls::AeFlickerMode are not recognized.
According to my research, they have been introduced after libcamera 0.1.0
but there is no release version of libcamera newer than 0.1.0 available
to which we could bump.
Signed-off-by: Sebastian Bauer <mail@sebastianbauer.info>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit bf7a1f10dd)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since GDB 13.x and upstream commit
b686ecb5b10be9a33ab8f1bfdcff22eef920d1a5 ("gdb: link executables with
libtool"), gdb will be linked against the shared variants of libbfd
and libopcodes if they exist. However, this causes host gdb and target
gdb to not work, because our gdb package does not install libbfd and
libopcodes (to not clash with the ones potentially installed by
binutils).
In order to get around this, this commit proposes to get back to the
situation we had before GDB 13.x: libbfd and libopcodes are only
compiled as static libraries, so that they are linked directly inside
the gdb binary, avoiding the problem entirely.
This resolves:
# gdb --version
gdb: error while loading shared libraries: libopcodes-2.39.50.so: cannot open shared object file: No such file or directory
for target gdb, and:
$ ./host/bin/arm-linux-gdb --version
./host/bin/arm-linux-gdb: error while loading shared libraries: libopcodes-2.39.50.so: cannot open shared object file: No such file or directory
for host gdb.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit e5729d3008)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- Drop patches (already in version) and so drop autoreconf
- Fix the following security issues:
- CVE-2023-40660: Fix Potential PIN bypass
- CVE-2023-40661: Important dynamic analyzers reports
- CVE-2023-4535: Out-of-bounds read in MyEID driver handling
encryption using symmetric keys
https://github.com/OpenSC/OpenSC/releases/tag/0.24.0
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 37eb68c9fb)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit d344ffe624 (configs/rock5b: add hash for custom uboot)
explicitly noted that the kernel was retrieved from a git-clone, so the
sha1 of the commit was enough to get what we expect.
However, that does not account for the fact that the upstream repository
can disapear or be temporarily unavailable (maliciously or not). In that
case, the kernel archive will be looked up on the backup mirror.
In that case, the download is via wget over https, which protects the
transport, but does not guarantee that the remote server serves the
expected archive.
The hash file was dropped when d344ffe624 was applied; restore it.
Since the defconfig now has hashes for all its downloads, enforce
checking hashes.
Signed-off-by: Kilian Zinnecker <kilian.zinnecker@mail.de>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 9ebbfeff38)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The defconfig uses a custom uboot version, downloaded with wget, so we
weant to be sure that it does not get modified on the server, so we add
a hash for it.
The kernel we get from a git clone, so the sha1 of the commit is enough
to be sure that what we get is what we expect (because we do a local
tarball out of a git clone).
Since we only get a hash for uboot and not for the kernel, we don't
enable BR2_DOWNLOAD_FORCE_CHECK_HASHES.
Signed-off-by: Kilian Zinnecker <kilian.zinnecker@mail.de>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit d344ffe624)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since upstream commit
eec95e3d5e1a4f2e13b1f6b34cc287475ca57daf ("backend/drm: use pnp.ids to
fetch EDID data"), the pnp.ids file from hwdata is parsed at build
time to generate a C source file. As per backend/drm/meson.build:
hwdata = dependency('hwdata', required: false, native: true)
if hwdata.found()
hwdata_dir = hwdata.get_variable(pkgconfig: 'pkgdatadir')
pnp_ids = files(hwdata_dir / 'pnp.ids')
else
pnp_ids = files('/usr/share/hwdata/pnp.ids')
endif
This is only needed when the DRM backend of wlroots is enabled, but
currently, Buildroot enables this backend unconditionally.
This failure can be reproduced using the following defconfig:
BR2_x86_64=y
BR2_TOOLCHAIN_EXTERNAL=y
BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_EUDEV=y
BR2_PACKAGE_MESA3D=y
BR2_PACKAGE_MESA3D_OSMESA_GALLIUM=y
BR2_PACKAGE_MESA3D_OPENGL_EGL=y
BR2_PACKAGE_MESA3D_OPENGL_ES=y
BR2_PACKAGE_WLROOTS=y
The issue was not caught in the autobuilders because the last
successful build of a configuration that includes wlroots dates back
from 2022-05-05, at which time Buildroot had wlroots 0.15.1.
This change in wlroots was introduced in wlroots 0.16.0, which means
that it's only since Buildroot bumped from 0.15.1 to 0.16.2 in
d6279bc82c ("package/wlroots: bump to
version 0.16.2") that the issue occurs. This commit is not yet in any
tagged release, so there is no need to backport this fix.
It should be noted that the proposed patch also installs pnp.ids to
the target filesystem, while it is in practice not needed at runtime
by wlroots. However, our current hwdata packaging doesn't allow
installing it only in staging, and since wlroots anyway implies we're
building a fairly heavy graphics stack, the size overhead of hwdata is
deemed to be an acceptable trade-off.
Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
[Thomas: further extend the commit log, with details gathered by Yann
and myself.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 50eed2060a)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As described in the announcement, this fixes a security issue:
There is one security fix in this release:
- Fix for a newly discovered security issue known as the 'Terrapin'
attack, also numbered CVE-2023-48795. The issue affects widely-used
OpenSSH extensions to the SSH protocol: the ChaCha20+Poly1305
cipher system, and 'encrypt-then-MAC' mode.
In order to benefit from the fix, you must be using a fixed version
of PuTTY _and_ a server with the fix, so that they can agree to
adopt a modified version of the protocol. Alternatively, you may be
able to reconfigure PuTTY to avoid selecting any of the affected
modes.
If PuTTY 0.80 connects to an SSH server without the fix, it will
warn you if the initial protocol negotiation chooses an insecure
mode to run the connection in, so that you can abandon the
connection. If it's possible to alter PuTTY's configuration to
avoid the problem, then the warning message will tell you how to do
it.
https://lists.tartarus.org/pipermail/putty-announce/2023/000037.html
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 922132c39e)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
5 CVEs affecting glibc according to the NVD database are considered as
not being security issues by upstream glibc developers:
* CVE-2010-4756: The glob implementation in the GNU C Library (aka
glibc or libc6) allows remote authenticated users to cause a denial
of service (CPU and memory consumption) via crafted glob expressions
that do not match any pathnames. glibc maintainers position: "That's
standard POSIX behaviour implemented by (e)glibc. Applications using
glob need to impose limits for themselves"
* CVE-2019-1010022: GNU Libc current is affected by: Mitigation
bypass. The impact is: Attacker may bypass stack guard
protection. The component is: nptl. The attack vector is: Exploit
stack buffer overflow vulnerability and use this bypass
vulnerability to bypass stack guard. NOTE: Upstream comments
indicate "this is being treated as a non-security bug and no real
threat. glibc maintainers position: "Not treated as a security issue
by upstream https://sourceware.org/bugzilla/show_bug.cgi?id=22850"
* CVE-2019-1010023: GNU Libc current is affected by: Re-mapping
current loaded library with malicious ELF file. The impact is: In
worst case attacker may evaluate privileges. The component is:
libld. The attack vector is: Attacker sends 2 ELF files to victim
and asks to run ldd on it. ldd execute code. NOTE: Upstream comments
indicate "this is being treated as a non-security bug and no real
threat. glibc maintainers position: "Not treated as a security issue
by upstream https://sourceware.org/bugzilla/show_bug.cgi?id=22851"
* CVE-2019-1010024: GNU Libc current is affected by: Mitigation
bypass. The impact is: Attacker may bypass ASLR using cache of
thread stack and heap. The component is: glibc. NOTE: Upstream
comments indicate "this is being treated as a non-security bug and
no real threat. glibc maintainers position: "Not treated as a
security issue by upstream
https://sourceware.org/bugzilla/show_bug.cgi?id=22852"
* CVE-2019-1010025: GNU Libc current is affected by: Mitigation
bypass. The impact is: Attacker may guess the heap addresses of
pthread_created thread. The component is: glibc. NOTE: the vendor's
position is "ASLR bypass itself is not a vulnerability. Glibc
maintainers position: "Not treated as a security issue by upstream
https://sourceware.org/bugzilla/show_bug.cgi?id=22853"
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit adaae82c58)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As reported in bug 15895, the GLIBC_VERSION field having a value
looking like 2.38-27-g750a45a783906a19591fb8ff6b7841470f1f5701, it
prevents the CPE/CVE matching with the NVD database to work correctly.
This commit fixes that by defining GLIBC_CPE_ID_VERSION, derived from
GLIBC_VERSION, by extracting the base version.
Also, we update GLIBC_IGNORE_CVES to account for the CVEs that have
clearly been fixed between 2.38 and
2.38-27-g750a45a783906a19591fb8ff6b7841470f1f5701. There are a number
of other CVEs still affecting the glibc package, but they are not
related to this
2.38...2.38-27-g750a45a783906a19591fb8ff6b7841470f1f5701 range.
Fixes: #15895
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit af8c0e5c74)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
It turns out that wildcard expansion, * and ?, is not performed in
matching lists {...}, at least in the vim plugin. The spec is not clear
about that, but refer to "pattern matching through Unix shell-style
wildcards" [0].
So, let's consider that this is not supported. Expand the patterns into
one section each, rather than use a list.
[0] https://spec.editorconfig.org/
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit ceb678ca19)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
OpenSSH 9.6 was released on 2023-12-18.
This release contains fixes for a newly-discovered weakness in the
SSH transport protocol (the "Terrapin" attack), a logic error relating
to constrained PKCS#11 keys in ssh-agent(1) and countermeasures for
programs that invoke ssh(1) with user or hostnames containing invalid
characters.
https://www.openssh.com/txt/release-9.6
Signed-off-by: Christian Stewart <christian@aperture.us>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 3c047ea463)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since the update of Python to version 3.11 in commit
738500c296 ("package/python3: bump to
version 3.11.0"), python-sip fails to compile with:
siplib.c: In function ‘sip_api_get_frame’:
siplib.c:13750:22: error: invalid use of undefined type ‘struct _frame’
13750 | frame = frame->f_back;
This is due to a change in the Python C API, which is fixed by a new
patch. The patch can't be upstreamed, as SIP 4.x is no longer
maintained upstream.
Fixes:
http://autobuild.buildroot.net/results/7b01739e7514e48c06182bc1804b32497ce2e414/
Signed-off-by: Ralf Dragon <hypnotoad@lindra.de>
[Thomas: improved commit log, reformatted patch using Git]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 3ef6884e6d)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes the following security issues:
- CVE-2023-5367 X.Org server: OOB write in
XIChangeDeviceProperty/RRChangeOutputProperty
- CVE-2023-5380: Use-after-free bug in DestroyWindow
- CVE-2023-5574: Use-after-free bug in DamageDestroy
For details, see the advisory:
https://lists.x.org/archives/xorg-announce/2023-October/003430.html
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 36a9ec8921)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
rsync is used in the infrastructure, mostly for the per-package infra,
and for the override-srcdir mechanism, but also to build the manual.
As such, it is not optional but mandatory, and already listed so.
Drop the reference to rsync from the list of optional packages.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit b79fb3c224)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes the following security issues:
- CVE-2023-46218: cookie mixed case PSL bypass
This flaw allows a malicious HTTP server to set "super cookies" in curl
that are then passed back to more origins than what is otherwise allowed
or possible. This allows a site to set cookies that then would get sent
to different and unrelated sites and domains.
https://curl.se/docs/CVE-2023-46218.html
- CVE-2023-46219: HSTS long file name clears contents
When saving HSTS data to an excessively long file name, curl could end up
removing all contents, making subsequent requests using that file unaware
of the HSTS status they should otherwise use.
https://curl.se/docs/CVE-2023-46219.html
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit aaa9438b96)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The prebuilt kernel has been updated to 5.10.202, sync the kernel
built by InitSystemSystemdBaseOverlayfs.
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit f6254689f8)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/5834758777
Commit e7d16c35a (boot/arm-trusted-firmware: fix the RPATH of fiptool) tried
to fix the build of host-fiptool, but forgot to pass HOST_CFLAGS.
On hosts without (compatible) openssl development headers, this breaks
the build when it cannot find the openssl headers:
fiptool_platform.h:19:11: fatal error: openssl/sha.h: No such file or directory
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit e6ef64d955)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Update the zynq readme.txt to add documentation for the zc702 and correct
documentation that was no longer up to date.
Signed-off-by: Neal Frager <neal.frager@amd.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 9675f6150c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The arm-trusted-firmware package builds a host tool called "fiptool",
which is used during the build process of arm-trusted-firmware
itself. This tool links against the OpenSSL host library, and
therefore needs to be built with the correct RPATH pointing to
$HOST_DIR/lib.
This is why commit a957d9a90a
("boot/arm-trusted-firmware: build fiptool separately with dependency
o n host-openssl") added the ARM_TRUSTED_FIRMWARE_BUILD_FIPTOOL
variable, which builds the fiptool tool first, with the right
variables set, before invoking the full build of TF-A. This ensured
that fiptool was built with the correct RPATH.
However, more recent versions of TF-A have modified their Makefile
machinery, and fiptool is being rebuilt even if it was built
before. Unfortunately, this rebuild is no longer done with the right
flags, so we end up with a fiptool binary that no longer has the right
RPATH, and fiptool fails to find the OpenSSL libraries from
$HOST_DIR/lib.
In order to fix this, we take a different approach: we do not build
fiptool separately first, but we inject the necessary flags through
the HOSTCC variable. Indeed, there's no HOST_LDFLAGS or HOST_LDLIBS
variable or similar that would allow us to pass the -Wl,-rpath flag
that is needed. Shoe-horning this flag into HOSTCC gets the job done,
and actually simplifies our arm-trusted-firmware.mk.
This patch break the compatibility with version prior to 1.4 (upstream
commit 72610c4102990 ("build: Introduce HOSTCC flag")). v1.4 is very old
(July 2017), not used anymore in-tree and probably not used anymore
outside the tree.
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
Co-authored-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit e7d16c35ae)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Configure Buildroot: (use the command for the specific board)
Configure Buildroot:
$ make zynqmp_zcu102_defconfig
$ make zynqmp_zcu104_defconfig
$ make zynqmp_zcu106_defconfig
Compile everything and build the rootfs image:
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Peter Korsgaard