]> Git Repo - buildroot-mgba.git/log
buildroot-mgba.git
2 years agoconfigs/roc_pc_rk3399: remove defconfig
Thomas Petazzoni [Fri, 25 Nov 2022 18:02:31 +0000 (19:02 +0100)]
configs/roc_pc_rk3399: remove defconfig

Commit 5370ec74516495a4ac6c0bc9780b8e92a2f1e6b1 was supposed to remove
the roc_pc_rk3399 defconfig. It actually removed everything related to
this defconfig, but not the defconfig itself.

The build failure this commit was supposed to fix is therefore still
happening. We fix it up by finally removing the defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3372859807

Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/gcc: ensure __register_frame is optimized out for glibc
James Hilliard [Fri, 28 Oct 2022 22:38:44 +0000 (16:38 -0600)]
package/gcc: ensure __register_frame is optimized out for glibc

On some architectures when building with -O0 the __register_frame
symbol fails to get optimized out which can cause linking failures
when building glibc.

To fix this set -O1 for GCC target libs when building with glibc
and BR2_OPTIMIZE_0 on the problematic target architectures.

This was reported both to GCC [1] and glibc [2] upstream. It is not
entirely clear yet where the bug lies exactly. At the moment the
assumption is that it's GCC, so create a symbol
BR2_TOOLCHAIN_HAS_GCC_BUG_107728.

This issue only seems to occur when linking glibc, not with anything
else, so only compile libgcc from host-gcc-initial with -O1.

Fixes:
 - http://autobuild.buildroot.net/results/89b/89b6c6924240b7cf82035a844f3573673e91b364
 - http://autobuild.buildroot.net/results/46f/46f4ec99d2b23d354a4bb5e92123d64f0da6ed27
 - http://autobuild.buildroot.net/results/839/839f929f700cf181ebdf34389c7806a96f55813e
 - http://autobuild.buildroot.net/results/0e2/0e202bf53a683930f3cad6edef2a4dea629eaecb
 - http://autobuild.buildroot.net/results/8a8/8a8c917f597fdcca744e696e19e9300b64004335
 - http://autobuild.buildroot.net/results/c05/c058b27ed2834dfa633b63ec6c3639ab1e8bf412
 - http://autobuild.buildroot.net/results/8ba/8ba8882a861cf7df359c23969c09b2be0725b2e5

[1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107728
[2] https://sourceware.org/bugzilla/show_bug.cgi?id=29621

Signed-off-by: James Hilliard <[email protected]>
Signed-off-by: Arnout Vandecappelle <[email protected]>
2 years agopackage/rtl8821au: move upstream and fix missing linux options
Giulio Benetti [Wed, 21 Sep 2022 20:58:15 +0000 (22:58 +0200)]
package/rtl8821au: move upstream and fix missing linux options

This package builds to fail with Linux > 5.15 and abperiasamy's
rtl8812AU_8821AU_linux repository is not maintained since 2 years and
there is now a fork where all pending patches have been upstreamed, so
let's switch to lwfinger's rtl8812au repository that is well
maintained with Linux up to version 5.18 supported. While switching
let's drop all local patches. Also add me as maintainer for this
package in DEVELOPERS file.

Fixes:

  http://autobuild.buildroot.net/results/a3db3a6540b67a1f1fe31d61fe1d6824d43f59f0/

Signed-off-by: Giulio Benetti <[email protected]>
Reviewed-by: Christian Stewart<[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/wilc-driver: fix build failure due to missing Linux options
Giulio Benetti [Tue, 20 Sep 2022 13:32:24 +0000 (15:32 +0200)]
package/wilc-driver: fix build failure due to missing Linux options

Enable Linux options depending on the bus has been chosen, so:
1) enable by default common Linux options:
CONFIG_NET
CONFIG_WIRELESS
CONFIG_CFG80211
CONFIG_CRC_ITU_T
CONFIG_CRC7
2) enable for SDIO bus:
CONFIG_MMC
3) enable for SPI bus:
CONFIG_SPI

Fixes:
http://autobuild.buildroot.net/results/d8c4f0f959dd2ec110db8a75980f13172c3c116c/

Signed-off-by: Giulio Benetti <[email protected]>
Reviewed-by: Kris Bahnsen <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoUpdate for 2022.11-rc3
Peter Korsgaard [Thu, 24 Nov 2022 16:47:01 +0000 (17:47 +0100)]
Update for 2022.11-rc3

Signed-off-by: Peter Korsgaard <[email protected]>
2 years agopackage/libkrb5: security bump to version 1.20.1
Peter Korsgaard [Thu, 24 Nov 2022 13:53:34 +0000 (14:53 +0100)]
package/libkrb5: security bump to version 1.20.1

Fixes the following security issue:

CVE-2022-42898: In MIT krb5 releases 1.8 and later, an authenticated
attacker may be able to cause a KDC or kadmind process to crash by reading
beyond the bounds of allocated memory, creating a denial of service.  A
privileged attacker may similarly be able to cause a Kerberos or GSS
application service to crash.  On 32-bit platforms, an attacker can also
cause insufficient memory to be allocated for the result, potentially
leading to remote code execution in a KDC, kadmind, or GSS or Kerberos
application server process.  An attacker with the privileges of a
cross-realm KDC may be able to extract secrets from a KDC process's memory
by having them copied into the PAC of a new ticket.

Bugfix tarballs are located in the same directory as the base version, so
introduce LIBKRB5_VERSION_MAJOR.

Signed-off-by: Peter Korsgaard <[email protected]>
Signed-off-by: Arnout Vandecappelle <[email protected]>
2 years agopackage.libopenssl: fix enabling/disabling mdc2
Yann E. MORIN [Sat, 5 Nov 2022 16:55:24 +0000 (17:55 +0100)]
package.libopenssl: fix enabling/disabling mdc2

Commit 3dbc86f09897 (openssl: bump version, enable mdc2+camellia+tlsext)
form 2010-06-03, forced the build of mdc2. Commit a83d41867c8d
(package/libopenssl: add option to enable some features) added an option
to explicitly disable mdc2, but forgot to amend the existing enabling
option.

It appears that, like most (all?) openssl config options, mdc2 ends up
enabled unless explicitly disabled.

Additionally, mdc2 depends on DES, so without DES, mdc2 gets disabled.

So, drop the explicit enabling option, and make mdc2 select DES.

Signed-off-by: Yann E. MORIN <[email protected]>
Cc: "GAUTRON, Erwan" <[email protected]>
Cc: "Weber, Matthew L Collins" <[email protected]>
Signed-off-by: Arnout Vandecappelle <[email protected]>
2 years agopackage/libopenssl: drop useless option for rc5
Yann E. MORIN [Sat, 5 Nov 2022 16:55:23 +0000 (17:55 +0100)]
package/libopenssl: drop useless option for rc5

Commit a83d41867c8d (package/libopenssl: add option to enable some
features) added an option to enable rc5. However, since commit
1fff94121936 (Fixup non-x86 openssl build), dated 2002-12-30, rc5
has always been forcibly disabled in Buildroot.

Given that it was unconditionally disabled all this time, and no
one complained, it means there is virtually no-one using rc5, so we
can just drop the option.

Signed-off-by: Yann E. MORIN <[email protected]>
Cc: "GAUTRON, Erwan" <[email protected]>
Cc: "Weber, Matthew L Collins" <[email protected]>
Signed-off-by: Arnout Vandecappelle <[email protected]>
2 years agopackage/swupdate: add libubootenv as optional dependency
Michael Nosthoff [Thu, 10 Nov 2022 10:23:12 +0000 (11:23 +0100)]
package/swupdate: add libubootenv as optional dependency

If the swupdate configuration contains CONFIG_UBOOT=y it uses
libubootenv to access the U-Boot environment.

We don't have Buildroot config options for all the different optional
dependencies of swupdate, instead we rely on the user to select the
appropriate packages and simply add the dependency in the .mk file. Do
this for libubootenv as well. swupdate doesn't have anything like
HAVE_LIBUBOOTENV, it just assumes libubootenv is available.

Fixes:
bootloader/uboot.c:23:10: fatal error: libuboot.h: No such file or directory
   23 | #include <libuboot.h>

Note that libubootenv is normally built before swupdate (alphabetical
ordering), so the error only occrus with BR2_PER_PACKAGE_DIRECTORIES or
when building swupdate directly.

Note that the autobuilders don't have this error, because they only
build swupdate with a default configuration that doesn't have U-Boot
support.

Signed-off-by: Michael Nosthoff <[email protected]>
Signed-off-by: Arnout Vandecappelle <[email protected]>
2 years agopackage/heimdal: security bump to version 7.7.1
Fabrice Fontaine [Wed, 23 Nov 2022 22:24:01 +0000 (23:24 +0100)]
package/heimdal: security bump to version 7.7.1

This release fixes the following Security Vulnerabilities:

- CVE-2022-42898 PAC parse integer overflows
- CVE-2022-3437 Overflows and non-constant time leaks in DES{,3} and
  arcfour
- CVE-2022-41916 Fix Unicode normalization read of 1 bytes past end of
  array
- CVE-2021-44758 NULL dereference DoS in SPNEGO acceptors
- CVE-2021-3671 A null pointer de-reference when handling missing sname
  in TGS-REQ
- CVE-2022-44640 Heimdal KDC: invalid free in ASN.1 codec

  Note that CVE-2022-44640 is a severe vulnerability, possibly a 10.0
  on the Common Vulnerability Scoring System (CVSS) v3, as we believe
  it should be possible to get an RCE on a KDC, which means that
  credentials can be compromised that can be used to impersonate
  anyone in a realm or forest of realms.

  Heimdal's ASN.1 compiler generates code that allows specially
  crafted DER encodings of CHOICEs to invoke the wrong free function
  on the decoded structure upon decode error. This is known to impact
  the Heimdal KDC, leading to an invalid free() of an address partly
  or wholly under the control of the attacker, in turn leading to a
  potential remote code execution (RCE) vulnerability.

  This error affects the DER codec for all extensible CHOICE types
  used in Heimdal, though not all cases will be exploitable. We have
  not completed a thorough analysis of all the Heimdal components
  affected, thus the Kerberos client, the X.509 library, and other
  parts, may be affected as well.

  This bug has been in Heimdal's ASN.1 compiler since 2005, but it may
  only affect Heimdal 1.6 and up. It was first reported by Douglas
  Bagnall, though it had been found independently by the Heimdal
  maintainers via fuzzing a few weeks earlier.

  While no zero-day exploit is known, such an exploit will likely be
  available soon after public disclosure.

- CVE-2019-14870: Validate client attributes in protocol-transition

- CVE-2019-14870: Apply forwardable policy in protocol-transition
- CVE-2019-14870: Always lookup impersonate client in DB

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoutils/scanpypi: add LICENCE.TXT to the list of the license files
Yegor Yefremov [Tue, 22 Nov 2022 14:50:17 +0000 (15:50 +0100)]
utils/scanpypi: add LICENCE.TXT to the list of the license files

Signed-off-by: Yegor Yefremov <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoconfigs/qemu_aarch64_ebbr: add host-qemu
Vincent Stehlé [Wed, 23 Nov 2022 15:08:59 +0000 (16:08 +0100)]
configs/qemu_aarch64_ebbr: add host-qemu

Add the host-qemu package to enable testing on gitlab.

Signed-off-by: Vincent Stehlé <[email protected]>
Cc: Romain Naour <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoboot/edk2: refine license
Vincent Stehlé [Wed, 23 Nov 2022 10:20:35 +0000 (11:20 +0100)]
boot/edk2: refine license

The edk2 project is licensed under the BSD-2-Clause license with a patent
grant, as per commit 304bff7223a8 ("edk2: Change License.txt from 2-Clause
BSD to BSD+Patent").

There is a BSD-2-Clause-Patent SPDX license identifier[1] for this case,
therefore refine the edk2 package to use this more specific identifier.

[1]: https://spdx.org/licenses/BSD-2-Clause-Patent.html

Signed-off-by: Vincent Stehlé <[email protected]>
Cc: Dick Olsson <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/edk2-platforms: refine license
Vincent Stehlé [Wed, 23 Nov 2022 10:20:34 +0000 (11:20 +0100)]
package/edk2-platforms: refine license

The edk2-platforms project is licensed under the BSD-2-Clause license with
a patent grant, as per commit ae604e4ffe8f ("edk2-platforms: Change
License.txt from 2-Clause BSD to BSD+Patent").

There is a BSD-2-Clause-Patent SPDX license identifier[1] for this case,
therefore refine the edk2-platforms package to use this more specific
identifier.

[1]: https://spdx.org/licenses/BSD-2-Clause-Patent.html

Signed-off-by: Vincent Stehlé <[email protected]>
Cc: Dick Olsson <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/python3: add upstream security fix for CVE-2022-45061
Peter Korsgaard [Tue, 22 Nov 2022 20:18:25 +0000 (21:18 +0100)]
package/python3: add upstream security fix for CVE-2022-45061

Fixes the following security issue:

CVE-2022-45061: An issue was discovered in Python before 3.11.1.  An
unnecessary quadratic algorithm exists in one path when processing some
inputs to the IDNA (RFC 3490) decoder, such that a crafted, unreasonably
long name being presented to the decoder could lead to a CPU denial of
service.  Hostnames are often supplied by remote servers that could be
controlled by a malicious actor; in such a scenario, they could trigger
excessive CPU consumption on the client attempting to make use of an
attacker-supplied supposed hostname.  For example, the attack payload could
be placed in the Location header of an HTTP response with status code 302.

Signed-off-by: Peter Korsgaard <[email protected]>
2 years agoboot/uboot/uboot.mk: fix zynqmp without pmufw
Brandon Maier [Tue, 22 Nov 2022 20:17:01 +0000 (14:17 -0600)]
boot/uboot/uboot.mk: fix zynqmp without pmufw

Commit d07e6b70 (boot/uboot/uboot.mk: add pmufw.elf support) broke
configurations where the UBOOT_ZYNQMP_PMUFW was blank. Previously it
would set the U-Boot CONFIG_PMUFW_INIT_FILE to the blank string, but now
it will set it to ".bin" which causes U-Boot to fail to build.

Signed-off-by: Brandon Maier <[email protected]>
Reviewed-by: Luca Ceresoli <[email protected]>
Signed-off-by: Brandon Maier <[email protected]>
Reviewed-by: Neal Frager <[email protected]>
Signed-off-by: Peter Korsgaard <[email protected]>
2 years ago{linux, linux-headers}: bump 4.{9, 14, 19}.x / 5.{4, 10, 15, 19}.x / 6.0.x series
Peter Korsgaard [Tue, 22 Nov 2022 21:30:46 +0000 (22:30 +0100)]
{linux, linux-headers}: bump 4.{9, 14, 19}.x / 5.{4, 10, 15, 19}.x / 6.0.x series

Signed-off-by: Peter Korsgaard <[email protected]>
2 years agopackage/gnupg2: bump version to 2.3.8
Michael Fischer [Wed, 16 Nov 2022 10:09:13 +0000 (11:09 +0100)]
package/gnupg2: bump version to 2.3.8

Brings a number of fixes: https://dev.gnupg.org/T6106

Add patch 0001 to fix undefined reference to `ks_ldap_free_state'
backported from commit 7011286ce6e1fb56c2989fdafbd11b931c489faa

Signed-off-by: Michael Fischer <[email protected]>
[Peter: add changelog info]
Signed-off-by: Peter Korsgaard <[email protected]>
2 years agopackage/gitlab-runner: fix inconsistency of systemd and sysv daemons
Marek Metelski [Mon, 7 Nov 2022 09:16:58 +0000 (10:16 +0100)]
package/gitlab-runner: fix inconsistency of systemd and sysv daemons

Copy default $DAEMON_ARGS from systemd service to sysv init script.

Make GITLAB_RUNNER_USER home directory the same as default
--work-directory (-d) flag.

Run sysv daemon process using root user (remove -c option)
This is needed to correctly access config files as specified.
System access can still be limited with gitlab-runner `--user` flag.

Use same $DAEMON_ARGS variable name so it can be overwritten in
/etc/default/gitlab-runner environment file in both cases.

Signed-off-by: Marek Metelski <[email protected]>
Reviewed-by: Marcin Niestroj <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/iwd: add dbus compile time dependency
James Hilliard [Mon, 31 Oct 2022 17:49:59 +0000 (11:49 -0600)]
package/iwd: add dbus compile time dependency

In 5b3b2d80f4cf586d360ff696c3dacbd4cb48fdc4 we dropped dbus as a build
dependency, however we still need it when building with systemd so
that the service directory is available via pkg-config.

In addition we can drop --with-dbus-datadir by unconditionally
requiring dbus as the datadir will then be fetched from pkg-config.

Fixes:
checking D-Bus bus services directory... configure: error: D-Bus bus services directory is required

  http://autobuild.buildroot.net/results/4a48676460e6ce588897598f0022ec840b4b4b8d/

Signed-off-by: James Hilliard <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoboot/edk2: fix the build for arm sgi575
Vincent Stehlé [Tue, 22 Nov 2022 16:05:44 +0000 (17:05 +0100)]
boot/edk2: fix the build for arm sgi575

The edk2 package can be configured for platform Arm Sgi575 but this
does not build correctly:

Usage: build.exe [options] [all|fds|genc|genmake|clean|cleanall|cleanlib|modules|libraries|run]

build.exe: error: option -a: invalid choice: '-b' (choose from 'IA32', 'X64', 'EBC', 'ARM', 'AARCH64', 'RISCV64')
make[1]: *** [package/pkg-generic.mk:293: /home/thomas/buildroot/buildroot/output/build/edk2-edk2-stable202102/.stamp_built] Error 2
make: *** [Makefile:84: _all] Error 2

Add the necessary definitions to fix the build.

Signed-off-by: Vincent Stehlé <[email protected]>
Cc: Dick Olsson <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/rtl8189es: bump to latest version to fix build failure with Linux >= 6.0
Giulio Benetti [Wed, 16 Nov 2022 00:40:55 +0000 (01:40 +0100)]
package/rtl8189es: bump to latest version to fix build failure with Linux >= 6.0

Drop local patch that has been upstreamed[0] and drop the endianness
handling too since from this commit[1] on it's handled by using Linux
macro __LITTLE_ENDIAN.

[0]: https://github.com/jwrdegoede/rtl8189ES_linux/commit/4a555ffb77a5947814b6c7f330968318e265c496
[1]: https://github.com/jwrdegoede/rtl8189ES_linux/commit/b3da33576da5359d85c7cfe82716c9a08a592aff

Fixes:
http://autobuild.buildroot.net/results/6178fbfbe9fe762645b1907c4ceb032a00e75a89/

Signed-off-by: Giulio Benetti <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/mali-driver: remove Miquèl from the DEVELOPERS list
Miquel Raynal [Wed, 16 Nov 2022 08:43:13 +0000 (09:43 +0100)]
package/mali-driver: remove Miquèl from the DEVELOPERS list

I am not really maintaining these packages, I don't follow closely
enough nor use them to take the time to make the necessary changes.
Giulio has been much more reactive than me to fix issues and he is
already listed for them anyway.

Cc: Giulio Benetti <[email protected]>
Signed-off-by: Miquel Raynal <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/rtl8723ds: fix build failure due to endianness and Linux version 6.0
Giulio Benetti [Wed, 16 Nov 2022 15:22:36 +0000 (16:22 +0100)]
package/rtl8723ds: fix build failure due to endianness and Linux version 6.0

Add local patch pending upstream[0] to override CFLAGS to set endianness
according to BR2_ENDIAN. Let's also bump version to latest to support up to
Linux 6.1.

[0]: https://github.com/lwfinger/rtl8723ds/pull/29

Fixes:
http://autobuild.buildroot.net/results/2646ec0512f867e20c25c1d0a6417826218942d6/

Signed-off-by: Giulio Benetti <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/mesa3d: fix uClibc build
Bernd Kuhls [Wed, 16 Nov 2022 18:04:52 +0000 (19:04 +0100)]
package/mesa3d: fix uClibc build

Moved the util/compiler.h include to util/macros.h due to upstream
commit which added static_assert() to src/util/macros.h
https://cgit.freedesktop.org/mesa/mesa/commit/src/util/macros.h?h=22.2&id=f1023571e8ce7ccb6ec7bc115240cb76aef3e5e5

Please note that this patch can be removed when buildroot toolchains
are updated to uClibc 1.0.42:
https://cgit.uclibc-ng.org/cgi/cgit/uclibc-ng.git/commit/?h=v1.0.42&id=03fbd941e943976bb92cb392882c2ff7ec218704

Fixes:
http://autobuild.buildroot.net/results/a55/a55d6980faad8b5063f8f4f8b89467061d44a2ae/

Signed-off-by: Bernd Kuhls <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoutils/genrandconfig: don't build igh-ethercat drivers
Thomas Petazzoni [Wed, 16 Nov 2022 21:21:05 +0000 (22:21 +0100)]
utils/genrandconfig: don't build igh-ethercat drivers

igh-ethercat comes with a small number of patched Linux kernel network
drivers, which aim at replacing the ones available in upstream Linux
kernel. All those drivers are provided only for specific kernel
releases. For example:

r8169-2.6.24-ethercat.c
r8169-2.6.24-orig.c
r8169-2.6.27-ethercat.c
r8169-2.6.27-orig.c
r8169-2.6.28-ethercat.c
r8169-2.6.28-orig.c
r8169-2.6.29-ethercat.c
r8169-2.6.29-orig.c
r8169-2.6.31-ethercat.c
r8169-2.6.31-orig.c
r8169-2.6.32-ethercat.c
r8169-2.6.32-orig.c
r8169-2.6.33-ethercat.c
r8169-2.6.33-orig.c
r8169-2.6.35-ethercat.c
r8169-2.6.35-orig.c
r8169-2.6.36-ethercat.c
r8169-2.6.36-orig.c
r8169-2.6.37-ethercat.c
r8169-2.6.37-orig.c
r8169-3.10-ethercat.c
r8169-3.10-orig.c
r8169-3.12-ethercat.c
r8169-3.12-orig.c
r8169-3.14-ethercat.c
r8169-3.14-orig.c
r8169-3.16-ethercat.c
r8169-3.16-orig.c
r8169-3.2-ethercat.c
r8169-3.2-orig.c
r8169-3.4-ethercat.c
r8169-3.4-orig.c
r8169-3.6-ethercat.c
r8169-3.6-orig.c
r8169-3.8-ethercat.c
r8169-3.8-orig.c
r8169-4.4-ethercat.c
r8169-4.4-orig.c

Obviously, this doesn't play well with the random configuration
testing done by utils/genrandconfig. This commit avoids this issue by
making sure we never build any of those drivers as part of the
genrandconfig generated configurations.

Fixes:

  http://autobuild.buildroot.net/results/07b7475d780c067d99ee5618a5fd2bb024a5b4e7/

Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/igh-ethercat: bump to latest Git commit
Thomas Petazzoni [Wed, 16 Nov 2022 21:21:04 +0000 (22:21 +0100)]
package/igh-ethercat: bump to latest Git commit

The current version 1.5.2 dates back from 2013, so it is extremely
old. The latest master branch of igh-ethercat contains numerous fixes,
including fixes to ensure that it builds with recent Linux kernel
releases. Backporting the individual patches fixing those issues on a
9 year old release would be too much effort, so we propose to simply
bump the version to the latest available in the Git master branch.

Fixes:

  http://autobuild.buildroot.net/results/4dc9b71c805a8156bcf8f398edd3a30f2b6ac6da/

Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/libnss: fix build failure with make 4.3.91
Giulio Benetti [Thu, 17 Nov 2022 19:45:05 +0000 (20:45 +0100)]
package/libnss: fix build failure with make 4.3.91

Make 4.3.91 doesn't allow to safely override Simple Expanded Variables, so
let's add a patch pending upstream[0] to make those variable Conditional
Expanded.

[0]: https://bugzilla.mozilla.org/show_bug.cgi?id=1801182

Fixes:
http://autobuild.buildroot.net/results/1074143dbea60567cd83be0a23f7c0214d470de9/

Signed-off-by: Giulio Benetti <[email protected]>
Tested-by: James Hilliard <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/sdl: add upstream security fix for CVE-2022-34568
Peter Korsgaard [Mon, 21 Nov 2022 07:24:43 +0000 (08:24 +0100)]
package/sdl: add upstream security fix for CVE-2022-34568

SDL v1.2 was discovered to contain a use-after-free via the XFree function
at /src/video/x11/SDL_x11yuv.c.

https://github.com/advisories/GHSA-wr7h-5wm3-p3h4

Signed-off-by: Peter Korsgaard <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/sdl2: fix sdl_init() error with kernel 5.15
Michael Fischer [Mon, 21 Nov 2022 12:40:01 +0000 (13:40 +0100)]
package/sdl2: fix sdl_init() error with kernel 5.15

Fixes #6421
Backport from: da9ba3a2a1536017e4ce1ee0f4276578d1ce6e29

Signed-off-by: Michael Fischer <[email protected]>
[[email protected]: make it an actual backport]
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/gptfdisk: fix runtime failure with popt 1.19
Fabrice Fontaine [Wed, 16 Nov 2022 17:11:56 +0000 (18:11 +0100)]
package/gptfdisk: fix runtime failure with popt 1.19

Fix the following runtime failure raised since bump of popt to version
1.19 in commit 895bfba93f6e5535f2132aeea144d2cd87ebc71b:

Problem opening  for reading! Error is 2.
The specified file does not exist!

Fixes:
 - No autobuilder failure

Signed-off-by: Fabrice Fontaine <[email protected]>
Reported-by: Florian Fainelli <[email protected]>
Tested-by: Florian Fainelli <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/samba4: security bump to version 4.15.12
Peter Korsgaard [Sun, 20 Nov 2022 17:13:00 +0000 (18:13 +0100)]
package/samba4: security bump to version 4.15.12

Fixes the following security issue:

- CVE-2022-42898: Samba buffer overflow vulnerabilities on 32-bit systems
  https://www.samba.org/samba/security/CVE-2022-42898.html

Signed-off-by: Peter Korsgaard <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/asterisk: security bump to version 16.28.0
Peter Korsgaard [Sun, 20 Nov 2022 14:17:20 +0000 (15:17 +0100)]
package/asterisk: security bump to version 16.28.0

Asterisk 16.26.0 fixed the following security issues:

- [ASTERISK-29476] – res_stir_shaken: Blind SSRF vulnerabilities
  https://issues.asterisk.org/jira/browse/ASTERISK-29476

- [ASTERISK-29838] – ${SQL_ESC()} not correctly escaping a terminating \
  https://issues.asterisk.org/jira/browse/ASTERISK-29838

- [ASTERISK-29872] – res_stir_shaken: Resource exhaustion with large files
  https://issues.asterisk.org/jira/browse/ASTERISK-29872

https://www.asterisk.org/asterisk-news/asterisk-16-26-0-now-available/

It unfortunately also introduced a change to chan_iax2, breaking builds
without OpenSSL:
https://github.com/asterisk/asterisk/commit/59a8cdaca2dbb5eeb7382dfbe78c0c1cbed8ce6d

Which was again fixed in 16.28.0:
https://github.com/asterisk/asterisk/commit/f812dfb68c6ed7ae55b4c163716fd1ddc063ff54

So bump to 16.28.0:
https://www.asterisk.org/asterisk-news/asterisk-16-28-0-now-available/

The libxml2 support now uses pkg-config, so drop the libxml2-config handling:
https://github.com/asterisk/asterisk/commit/bf9dafa7c22302b2f1a12b8216da63102116d9c9

Signed-off-by: Peter Korsgaard <[email protected]>
[[email protected]:
  - add host-pkgconf dep, don't rely on implicit dep from host-asterisk
]
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/systemd: security bump to version v250.8
Peter Korsgaard [Sun, 20 Nov 2022 16:26:22 +0000 (17:26 +0100)]
package/systemd: security bump to version v250.8

Fixes the following security issue:

- CVE-2022-3821: An off-by-one Error issue was discovered in Systemd in
  format_timespan() function of time-util.c.  An attacker could supply
  specific values for time and accuracy that leads to buffer overrun in
  format_timespan(), leading to a Denial of Service.
  https://github.com/systemd/systemd/issues/23928

Drop now upstream 0001-missing-syscall-define-MOVE_MOUNT_T_EMPTY_PATH-if-mi.patch

Signed-off-by: Peter Korsgaard <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/sysstat: security bump to version 12.6.1
Fabrice Fontaine [Sun, 20 Nov 2022 10:25:31 +0000 (11:25 +0100)]
package/sysstat: security bump to version 12.6.1

Fix CVE-2022-39377: sysstat is a set of system performance tools for the
Linux operating system. On 32 bit systems, in versions 9.1.16 and newer
but prior to 12.7.1, allocate_structures contains a size_t overflow in
sa_common.c. The allocate_structures function insufficiently checks
bounds before arithmetic multiplication, allowing for an overflow in the
size allocated for the buffer representing system activities. This issue
may lead to Remote Code Execution (RCE).

Despite what is written above in the CVE announcement, and as written in
the Changelog, the fix is also included in version 12.6.1 (12.7.1 is a
development version):
    https://github.com/sysstat/sysstat/commit/c1e631eddc50c04e4dcea169ba396bee2bd6b0ab

As a consequence, 12.6.1 is still reported as being affected. Until the
NVD is updated appropriately, we mark the CVE as ignored with a comment
that explains why.

Note: that commit is not reachable from any branch in the sysstat
repository, and Github warns about that, but the commit does belong to
the upstream repository and is reachable from the 12.6.1 tag (it looks
like sysstat only pushes tags-with-history for fix releases).

https://github.com/sysstat/sysstat/security/advisories/GHSA-q8r6-g56f-9w7x
https://github.com/sysstat/sysstat/blob/v12.6.1/CHANGES

Signed-off-by: Fabrice Fontaine <[email protected]>
[[email protected]:
  - ignore the CVE, explain why
  - explain why github warns about the fix commit
]
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/openpgm: fix build for non-x86 targets
Alexander Lukichev [Sun, 20 Nov 2022 09:06:54 +0000 (11:06 +0200)]
package/openpgm: fix build for non-x86 targets

openpgm-5-3-128 has assembly code for x86 that is not guarded by
architecture defines. A patch to fix that has been merged upstream
some time ago, and the next release will have it. This includes
that patch for the time being.

Fixes: http://autobuild.buildroot.net/results/338291e5bf0671cb7ed7a32cc10e546c7a521acc
Fixes: http://autobuild.buildroot.net/results/3ab6d7f9ee841fa18c1c220d722b1c06ca1fff30
Fixes: http://autobuild.buildroot.net/results/68e840b1fec8f14775cef0b6a14d9b847337324b
Signed-off-by: Alexander Lukichev <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/xterm: security bump to patch 376
Peter Korsgaard [Sat, 19 Nov 2022 13:45:10 +0000 (14:45 +0100)]
package/xterm: security bump to patch 376

Fixes the following security issue:

CVE-2022-45063: xterm before 375 allows code execution via font ops, e.g.,
because an OSC 50 response may have Ctrl-g and therefore lead to command
execution within the vi line-editing mode of Zsh:

https://www.openwall.com/lists/oss-security/2022/11/10/1

Additionally, patch 376 fixes a null pointer access issue:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1022942

Signed-off-by: Peter Korsgaard <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/xen: security bump to version 4.14.5
Peter Korsgaard [Sat, 19 Nov 2022 10:50:34 +0000 (11:50 +0100)]
package/xen: security bump to version 4.14.5

Includes a number of bugfixes and the security fixes up to xsa-400:
https://xenproject.org/downloads/xen-project-archives/xen-project-4-14-series/xen-project-4-14-5/

Signed-off-by: Peter Korsgaard <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/nodejs: security bump to version 16.18.1
Peter Korsgaard [Sat, 19 Nov 2022 10:19:03 +0000 (11:19 +0100)]
package/nodejs: security bump to version 16.18.1

Fixes the following security issue:

DNS rebinding in --inspect via invalid octal IP address (Medium) (CVE-2022-43548)

The Node.js rebinding protector for --inspect still allows invalid IP
address, specifically, the octal format.  An example of an octal IP address
is 1.09.0.0, the 09 octet is invalid because 9 is not a number in the base 8
number system.  Browsers such as Firefox (tested on latest version m105)
will still attempt to resolve this invalid octal address via DNS.  When
combined with an active --inspect session, such as when using VSCode, an
attacker can perform DNS rebinding and execute arbitrary code

Update license hash for an update of base64 (MIT license) and a change in
copyright year:

https://github.com/nodejs/node/commit/8ea9a71b15a953cd0936f7e6aae84c62873a77b5
https://github.com/nodejs/node/commit/9f14dc1a8f43a9f3755c673009378b798cbdd73b

Signed-off-by: Peter Korsgaard <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/libksba: security bump to version 1.6.2
Michael Fischer [Tue, 15 Nov 2022 16:27:05 +0000 (17:27 +0100)]
package/libksba: security bump to version 1.6.2

A severe bug has been found in Libksba , the library used by GnuPG for parsing
the ASN.1 structures as used by S/MIME. The bug affects all versions of Libksba
before 1.6.2 and may be used for remote code execution.

Fix CVE-2022-3515

Signed-off-by: Michael Fischer <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/nginx: add upstream security fix for CVE-2022-4174{1, 2}
Peter Korsgaard [Sat, 19 Nov 2022 22:22:34 +0000 (23:22 +0100)]
package/nginx: add upstream security fix for CVE-2022-4174{1, 2}

Fixes the following security issues:

- CVE-2022-41741: Memory corruption in the ngx_http_mp4_module
- CVE-2022-41742: Memory disclosure in the ngx_http_mp4_module

https://mailman.nginx.org/archives/list/[email protected]/message/RBRRON6PYBJJM2XIAPQBFBVLR4Q6IHRA/

Signed-off-by: Peter Korsgaard <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/rsync: fix configure options
Fabrice Fontaine [Wed, 16 Nov 2022 22:29:08 +0000 (23:29 +0100)]
package/rsync: fix configure options

Rename configure options to avoid the following build failure raised
since bump to version 3.2.5 in commit
ae2807821d897c16141d003f646475fee9e77cba:

./simd-checksum-x86_64.cpp: In function 'uint32_t get_checksum1_cpp(char*, int32_t)':
./simd-checksum-x86_64.cpp:89:52: error: multiversioning needs 'ifunc' which is not supported on this target
   89 | __attribute__ ((target("default"))) MVSTATIC int32 get_checksum1_avx2_64(schar* buf, int32 len, int32 i, uint32* ps1, uint32* ps2) { return i; }
      |                                                    ^~~~~~~~~~~~~~~~~~~~~
./simd-checksum-x86_64.cpp:480:1: error: use of multiversioned function without a default
  480 | }
      | ^
If you can't fix the issue, re-run ./configure with --disable-roll-simd.

Fixes:
 - http://autobuild.buildroot.org/results/069da8e585da2e51bfd4f475cc12b9a134954b08

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Peter Korsgaard <[email protected]>
2 years agopackage/freerdp: security bump to version 2.9.0
Fabrice Fontaine [Wed, 16 Nov 2022 22:07:34 +0000 (23:07 +0100)]
package/freerdp: security bump to version 2.9.0

Backported #8403: Fixed multiple client side input validation issues
(CVE-2022-39316, CVE-2022-39317, CVE-2022-39318, CVE-2022-39319,
CVE-2022-39320, CVE-2022-41877, CVE-2022-39347)

https://github.com/FreeRDP/FreeRDP/releases/tag/2.9.0

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Peter Korsgaard <[email protected]>
2 years agodocs/website: update for 2022.02.7
Peter Korsgaard [Thu, 17 Nov 2022 14:18:33 +0000 (15:18 +0100)]
docs/website: update for 2022.02.7

Signed-off-by: Peter Korsgaard <[email protected]>
2 years agoUpdate for 2022.02.7
Peter Korsgaard [Thu, 17 Nov 2022 09:44:53 +0000 (10:44 +0100)]
Update for 2022.02.7

Signed-off-by: Peter Korsgaard <[email protected]>
(cherry picked from commit d036dc6ec873f50a24e73d8dbdafecb3b4098909)
[Peter: drop Makefile/Vagrantfile changes]
Signed-off-by: Peter Korsgaard <[email protected]>
2 years agodocs/website: update for 2022.08.2
Peter Korsgaard [Thu, 17 Nov 2022 07:29:20 +0000 (08:29 +0100)]
docs/website: update for 2022.08.2

Signed-off-by: Peter Korsgaard <[email protected]>
2 years agoUpdate for 2022.08.2
Peter Korsgaard [Wed, 16 Nov 2022 17:08:23 +0000 (18:08 +0100)]
Update for 2022.08.2

Signed-off-by: Peter Korsgaard <[email protected]>
(cherry picked from commit 25680e6aa823bdeabb51d0768f12cebcd09c6d68)
[Peter: drop Makefile changes]
Signed-off-by: Peter Korsgaard <[email protected]>
2 years agoUpdate for 2022.11-rc2
Peter Korsgaard [Wed, 16 Nov 2022 12:56:56 +0000 (13:56 +0100)]
Update for 2022.11-rc2

Signed-off-by: Peter Korsgaard <[email protected]>
2 years agoboard/zynqmp/kria/kv260/kv260.sh: fix u-boot.itb without CONFIG_MULTI_DTB_FIT option
Neal Frager [Mon, 14 Nov 2022 13:51:26 +0000 (06:51 -0700)]
board/zynqmp/kria/kv260/kv260.sh: fix u-boot.itb without CONFIG_MULTI_DTB_FIT option

This patch fixes the kv260.sh to generate a working u-boot.itb
now that the CONFIG_MULTI_DTB_FIT u-boot option is no longer used.

This is a follow-up fix of
515319b86f17ea5a2ce3e51c9063bac9f9e00a01 ("board/zynqmp/kria/kv260/uboot.fragment:
remove unnecessary CONFIG_MULTI_DTB_FIT option") to fix the build of:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3310463281

Signed-off-by: Neal Frager <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/libvncserver: fix CVE-2020-29260
Fabrice Fontaine [Sun, 13 Nov 2022 21:55:58 +0000 (22:55 +0100)]
package/libvncserver: fix CVE-2020-29260

libvncclient v0.9.13 was discovered to contain a memory leak via the
function rfbClientCleanup().

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/perl-net-ssleay: bump to version 1.93_01
Fabrice Fontaine [Tue, 8 Nov 2022 17:51:29 +0000 (18:51 +0100)]
package/perl-net-ssleay: bump to version 1.93_01

- Refresh patch
- Drop -lz from Makefile with libressl as this is the only solution for
  now: https://github.com/radiator-software/p5-net-ssleay/issues/399
- License has been clarified to be Artistic-2.0 since version 1.86.11:
  https://github.com/radiator-software/p5-net-ssleay/commit/aa4a0206d6d2a5ac2998dd9d6a8c5b88902c04de
- This bump will fix the following build failure with libressl:

  In file included from /home/autobuild/autobuild/instance-11/output-1/host/armeb-buildroot-linux-uclibcgnueabi/sysroot/usr/lib/perl5/5.34.1/armeb-linux/CORE/perl.h:5748,
                   from SSLeay.xs:141:
  SSLeay.xs: In function 'XS_Net__SSLeay_SESSION_get_master_key':
  SSLeay.xs:5569:37: error: invalid use of incomplete typedef 'SSL_SESSION' {aka 'struct ssl_session_st'}
   5569 |      sv_setpvn(ST(0), (const char*)s->master_key, s->master_key_length);
        |                                     ^~

https://metacpan.org/release/CHRISN/Net-SSLeay-1.93_01/changes

Fixes:
 - http://autobuild.buildroot.org/results/71337cc496727f2b1173c055d706c5bfc2f5d2bc

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/vlc: fix opengl library check
Bernd Kuhls [Sun, 13 Nov 2022 10:00:57 +0000 (11:00 +0100)]
package/vlc: fix opengl library check

Fixes:
http://autobuild.buildroot.net/results/9710753984a38b8c6f83a136b39c3bc320ba558b/

Signed-off-by: Bernd Kuhls <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoutils/genrandconfig: disallow configs with BR2_XTENSA_CUSTOM=y
Thomas Petazzoni [Fri, 11 Nov 2022 21:57:59 +0000 (22:57 +0100)]
utils/genrandconfig: disallow configs with BR2_XTENSA_CUSTOM=y

When BR2_XTENSA_CUSTOM=y is used with the internal toolchain, an
overlay file is mandatory, which genrandconfig can't provide. So we
simply disallow such configurations.

Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoarch/arch.mk.xtensa: relax check on overlay file to apply only to internal toolchains
Thomas Petazzoni [Fri, 11 Nov 2022 21:57:58 +0000 (22:57 +0100)]
arch/arch.mk.xtensa: relax check on overlay file to apply only to internal toolchains

Commit 4cbf7336914f25478aea943456ba7dc3c892c21a ("arch/xtensa: custom
configuration requires an overlay") added a check in
arch/arch.mk.xtensa to bail out if a custom Xtensa core is selected
but not overlay file is provided. While this is indeed a perfectly
valid check to make when building an internal toolchain, with an
external toolchain it's entirely possible to build with no overlay
file: the toolchain already exists, and there's no overlay to be
applied in the context of the Buildroot build.

And indeed commit 4cbf7336914f25478aea943456ba7dc3c892c21a broke some
of the runtime test cases that use a custom Xtensa core configuration,
with no overlay, to test the toolchains.bootlin.com Xtensa external
toolchain. By relaxing the check to only apply to internal toolchain
configurations, we fix those test cases.

It is to be noted that this still allows a configuration where gdb gets
built for a custom core, but with no overlay, so basically that means
the fsf variant, which can lead to build or run failures that
4cbf7336914f attempted to fix to begin with. This still covers the
most common cases.

Finally, it also means being able to build a kernel with no overlay, but
this is offset by the fact that the kernel may be already patched with
an overlay (as it is possible to specify a custom kernel), which is most
probably what people using a custom core would have.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3282261966
  https://gitlab.com/buildroot.org/buildroot/-/jobs/3282261963

Signed-off-by: Thomas Petazzoni <[email protected]>
[[email protected]: extend commit log with last two paragraphs]
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/ntfs-3g: security bump to version 2022.10.3
Fabrice Fontaine [Sat, 12 Nov 2022 21:36:55 +0000 (22:36 +0100)]
package/ntfs-3g: security bump to version 2022.10.3

Fix CVE-2022-40284: A buffer overflow was discovered in NTFS-3G before
2022.10.3. Crafted metadata in an NTFS image can cause code execution. A
local attacker can exploit this if the ntfs-3g binary is setuid root. A
physically proximate attacker can exploit this if NTFS-3G software is
configured to execute upon attachment of an external storage device.

https://github.com/tuxera/ntfs-3g/security/advisories/GHSA-v4w8-jv3w-7prm
https://github.com/tuxera/ntfs-3g/releases/tag/2022.10.3

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/dbus-broker: audit support needs libcap-ng
Yann E. MORIN [Sat, 12 Nov 2022 12:34:32 +0000 (13:34 +0100)]
package/dbus-broker: audit support needs libcap-ng

Since v14 in 2018, audit support has needed libcap-ng.

Fixes:
    http://autobuild.buildroot.org/results/43abdb85cc2f386d427cec1cfa876e20e3509cb8/

Signed-off-by: Yann E. MORIN <[email protected]>
Cc: Norbert Lange <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/dracut: workaround breakage on non-merged-usr hosts
Yann E. MORIN [Sat, 12 Nov 2022 13:37:07 +0000 (14:37 +0100)]
package/dracut: workaround breakage on non-merged-usr hosts

dracut is not really ready to be installed with a non-/ prefix, and it
has a lot of hard-coded assumptions that it is going to run on the host
for which it is goign to generate an initramfs; for example, it
hard-codes calls to /lib/dracut/some-file in some of its modules. It
also uses the host system layout to decide whether it needs a
merged-usr or not.

Furthermore, dracut populates the temporary directory which content will
be used to generate the cpio, with a bunch of files, even before calling
any of the dracut modules.

The name for that temporary directory is not predictable (looks like the
output of 'mktemp -d dracut.XXXXXX', with names like dracut.1Vfn9F seen
while debugging).

As a consequence, we can't prepare the temporary directory with the
proper symlinks beforehand.

So, we provide a very-early module of our own, that will (hopefully) run
before any other module, to fixup the messed-up layout prepared by
dracut. This module moves the content of /lib, /bin, and /sbin, out and
into their counterparts in /usr, and creates the usual symlinks.

When we do not require a merged-usr, then we have nothing to do, so the
module checks for /lib being a symlink, as the hint that we want a
merged-usr or not.

Note: currently, we've seen nothing that dracut installed in /bin or
/sbin, but for trying to be future-proof, we also handle them; this
causes a spurious warning:
    mv: cannot stat '..../build/buildroot-fs/cpio/tmp/dracut.YQnzNP/initramfs/bin/*': No such file or directory

Since there are already quite a bunch of similar failures in the
official modules bundled in dracut, an extra such issue or two should
not be too scary...

Fixes:
    https://gitlab.com/buildroot.org/buildroot/-/jobs/3282261241
    https://gitlab.com/buildroot.org/buildroot/-/jobs/3282261239
    https://gitlab.com/buildroot.org/buildroot/-/jobs/3282261236

Signed-off-by: Yann E. MORIN <[email protected]>
Cc: Thierry Bultel <[email protected]>
Cc: Adam Duskett <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/systemd: fix build with -Ofast
Yann E. MORIN [Sun, 13 Nov 2022 11:09:01 +0000 (12:09 +0100)]
package/systemd: fix build with -Ofast

systemd does not build with -Ofast (at least with gcc-12), leading to
build errors like:

    ../src/shared/condition.c: In function ‘condition_dump_list’:
    ../src/shared/condition.c:1227:33: error: ‘%s’ directive argument is null [-Werror=format-overflow=]
     1227 |                 "%s\t%s: %s%s%s %s\n",
          |                                 ^~
    cc1: some warnings being treated as errors

It is not really clear what the reason is, but it smells like a compiler
error.

Indeed, the failing format is passed to an fprintf, and the parameter
corresponding to the failing %s directive is a call to a function
which prototype is defined but the implementation only comes later in
the same compilation unit, but is the result of macro expansion, which
yields a function definition like:

    const char foo_to_string(foo_type i) {
        if (i < 0 || i >= (foo_type) ELEMENTSOF(foo_table))
            return NULL;
        return foo_table[i]
    }

(where ELEMENTSOF(x) is a macros arounf sizeof(x) to determine the
number of elements in the array foo_table).

However, in the failing case, foo_table is a static const array indexed
with constants from an enum, and foo_to_string() is only ever called
with variables that are only ever set to one of those enum values.

Since -Ofast is also explicitly documented as breaking otehrwise
conformant programs, we're not going to debug further the reason for the
build failure.

Instead, just revert to the best alternate optimisation level. We chose
-O3, as -Ofast is based on -O3 with breaking optimisation flags.

With -O3, the build succeeds.

Fixes:
    http://autobuild.buildroot.org/results/3ffaa9b3ecacc6ac326be78196af1ad613f195ed/ (sparc64)
    http://autobuild.buildroot.org/results/3f6ae2e503dd1539e4240f344865da4881879204/ (arm)
    http://autobuild.buildroot.org/results/68c17056490d441c7f862349e9c7e471b4570162/ (ppc64)
    ...

Signed-off-by: Yann E. MORIN <[email protected]>
Cc: Norbert Lange <[email protected]>
Cc: Sen Hastings <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/matchbox-starup-monitor: fix build without C++
Yann E. MORIN [Sat, 12 Nov 2022 22:05:37 +0000 (23:05 +0100)]
package/matchbox-starup-monitor: fix build without C++

matchbox-startup-monitor is an ageing package, and uses an old
configure.ac with archaic constructs. This had generated a configure
script that incorrectly tries to look for and validate a C++ compiler:

    checking for powerpc64le-buildroot-linux-gnu-g++... no
    checking whether we are using the GNU C++ compiler... no
    checking whether no accepts -g... no
    checking dependency style of no... none
    checking how to run the C++ preprocessor... /lib/cpp
    configure: error: C++ preprocessor "/lib/cpp" fails sanity check

Calling autoreconf fixes the issue, as the generated configure no longer
tries to look for a C++ compiler at all anymore. Running autoreconf does
not add any new dependency, as they are already in the dependency chain
via other packages.

Fixes:
    http://autobuild.buildroot.org/results/223/223f43dd76ee907c5f25c4fee94a0f5d75614dd5/

See also similar changes:
    9993a36f5e12 package/pamtester: fix build without C++
    c05cc5de868c package/madplay: needs autoreconf
    eae18d01abc7 libmad: needs autoreconf
    43274dd3e0da package/libid3tag: needs autoreconf

Signed-off-by: Yann E. MORIN <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoboard/zynqmp/kria/kv260/uboot.fragment: remove unnecessary CONFIG_MULTI_DTB_FIT option
Neal Frager [Sun, 13 Nov 2022 18:49:33 +0000 (11:49 -0700)]
board/zynqmp/kria/kv260/uboot.fragment: remove unnecessary CONFIG_MULTI_DTB_FIT option

This patch removes the CONFIG_MULTI_DTB_FIT u-boot option for the
zynqmp_kria_kv260_defconfig as it is not necessary.  The post build
kv260.sh creates the proper u-boot.itb without needing this option.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3310463281

Signed-off-by: Neal Frager <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoboot/arm-trusted-firmware: fix SSP disable in v2.2
Baruch Siach [Sun, 13 Nov 2022 19:09:54 +0000 (21:09 +0200)]
boot/arm-trusted-firmware: fix SSP disable in v2.2

ATF version 2.2 and older does not disable SSP when
ENABLE_STACK_PROTECTOR is not set. This is because the compiler enables
SSP by default, and ATF does not pass -fno-stack-protector to the
compiler. Upstream commit 7af195e29a42 ("Disable stack protection
explicitly") fixed the issue for v2.3 and newer.

Add -fno-stack-protector in CFLAGS when
BR2_TARGET_ARM_TRUSTED_FIRMWARE_SSP is not set to fix older ATF
versions.

Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/3301821171

Cc: Dick Olsson <[email protected]>
Cc: Sergey Matyukevich <[email protected]>
Signed-off-by: Baruch Siach <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/mdio-tools: needs fork
Fabrice Fontaine [Sat, 12 Nov 2022 21:20:34 +0000 (22:20 +0100)]
package/mdio-tools: needs fork

Fix the following build failure raised since the addition of the package
in commit 8fdf8731e727dc5c2d4080d7a56e1d71dc25e533:

mdio.c: In function 'mdio_modprobe':
mdio.c:738:15: error: implicit declaration of function 'fork' [-Werror=implicit-function-declaration]
  738 |         pid = fork();
      |               ^~~~

Fixes:
 - http://autobuild.buildroot.org/results/c53aaeaa34dd4d6d9a57da196687beecaeed9fe2

Signed-off-by: Fabrice Fontaine <[email protected]>
Reviewed-by: Joachim Wiberg <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/imagemagick: utilities now need C++ support
Thomas Petazzoni [Fri, 11 Nov 2022 22:40:10 +0000 (23:40 +0100)]
package/imagemagick: utilities now need C++ support

Since upstream commit
https://github.com/ImageMagick/ImageMagick/commit/07f3b487f9860fd4eb9422f1a906d0fe83b6fd1c
(which first appeared in version 7.1.0-47), ImageMagick forces the
need of a C++ compiler to build its utilities. Despite the request of
Bernd Kuhls to revert this change, upstream declined.

Since this change is causing build failures in our autobuilders, our
only choice is to follow the choice of upstream, and disable building
the utilities when C++ support is not available.

Fixes:

  http://autobuild.buildroot.net/results/4283235d697408cf2e70be5e3769dbe6ebb9ddae/

Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackages/sudo: explicitly set enable-tmpfiles.d
Nuno Gonçalves [Thu, 10 Nov 2022 14:00:54 +0000 (14:00 +0000)]
packages/sudo: explicitly set enable-tmpfiles.d

sudo's configure script looks up on the host to determine the path where
to install its systemd tmpfiles. That is incorrect in cross-compilation.

We can explicitly tell sudo where to install its tmpfiles, which we do
when systemd is enabled (in Buildroot, systemd-tmpfiles is always
enabled when systemd is), or we can tell it not to install tmpfiles at
all, which we do otherwise.

Signed-off-by: Nuno Gonçalves <[email protected]>
[[email protected]: reword and extend commit log]
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackages/sudo: explicitly set with-tzdir
Nuno Gonçalves [Thu, 10 Nov 2022 14:00:53 +0000 (14:00 +0000)]
packages/sudo: explicitly set with-tzdir

sudo's configure script looks up on the host to determine the path to
the timezone data location. That fails in cross-compilation.

This is used to sanitise the TZ envirnment variable at runtime, and is
not used at buildtime (except to be stored as a string in the program).

We can tell sudo where the tz data will be, which we do when the tzdata
package is enabled, and we can tell it not to use it at all (to not pass
TZ down to sudo-ed executions) othwerwise.

Signed-off-by: Nuno Gonçalves <[email protected]>
[[email protected]: rewrite and extend commit log]
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/kontron_bl_imx8mm_defconfig: bump U-boot to 2022.10
Heiko Thiery [Tue, 8 Nov 2022 07:36:07 +0000 (08:36 +0100)]
configs/kontron_bl_imx8mm_defconfig: bump U-boot to 2022.10

Commit 223516b51e1a (configs/kontron_bl_imx8mm: U-Boot needs util-linux)
added the needed dependency against host-util-linux, but missed an
earlier comment about u-boot still failing [0]

The U-Boot makefile for the host tools does not handle the
compiler/linker options properly. There are some patches [1][2] that fixes
that issue already applied in the newer U-Boot version 2022.10. So we have to
bump U-Boot to fix an autobuilder failure.

Fixes: https://gitlab.com/buildroot.org/buildroot/-/jobs/3267233833
[0] https://lore.kernel.org/buildroot/CAEyMn7Y3UgT-8dYY5rbnzcPfbGmqRVXG=joWx1fSSCC=WiFzbg@mail.gmail.com/
[1] U-Boot: a638bd349ea43825 (kbuild: add KBUILD_HOSTLDFLAGS to cmd_host-csingle)
[2] U-Boot: 31a7688cbe0ed5ed (tools: mkeficapsule: use pkg-config to get -luuid and -lgnutls)

Cc: Yann E. MORIN <[email protected]>
Signed-off-by: Heiko Thiery <[email protected]>
[[email protected]:
  - update commit log with reference to [0]
  - slightly tweak commit log
]
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/efivar: fix build with musl libc
Erico Nunes [Tue, 11 Oct 2022 08:49:46 +0000 (10:49 +0200)]
package/efivar: fix build with musl libc

Backport upstream patch to fix build with musl libc.
This patch is only a requirement since efivar 38 and was applied
upstream shortly after the 38 version tag.

Fixes:
http://autobuild.buildroot.net/results/c49d894b109d68e2624074eab8b939fefa3b42ef/

Signed-off-by: Erico Nunes <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agopackage/wavemon: fix kernel header collision
Fabrice Fontaine [Thu, 10 Nov 2022 22:06:20 +0000 (23:06 +0100)]
package/wavemon: fix kernel header collision

Fix the following build failure raised since bump to version 0.9.4 in
commit 5cae1a0d675eca41125b6430157d3d0d78d1220c:

In file included from iw_if.h:26:0,
                 from conf.c:19:
/home/buildroot/autobuild/instance-1/output-1/host/mips-buildroot-linux-gnu/sysroot/usr/include/linux/if.h:71:2: error: redeclaration of enumerator 'IFF_UP'
  IFF_UP    = 1<<0,  /* sysfs */
  ^
/home/buildroot/autobuild/instance-1/output-1/host/mips-buildroot-linux-gnu/sysroot/usr/include/net/if.h:44:5: note: previous definition of 'IFF_UP' was here
     IFF_UP = 0x1,  /* Interface is up.  */
     ^

Fixes:
 - http://autobuild.buildroot.org/results/cbdf3e0cf0bee8f1b076581768c24155afc320d9

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/alsa-utils: update dependency for topology support
Bernd Kuhls [Fri, 11 Nov 2022 20:29:21 +0000 (21:29 +0100)]
package/alsa-utils: update dependency for topology support

This is a follow-up patch for
https://git.busybox.net/buildroot/commit/?id=28497102e1788df3628bd3324a3304a03c7942d0
which created a new option for topology support in alsa-lib, this
dependency was not ported over to alsa-utils.

Fixes:
http://autobuild.buildroot.net/results/a9c9f5157365efe8271bf69dab0d5a5532b196ea/

Signed-off-by: Bernd Kuhls <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/lxc: fix systemd build
Fabrice Fontaine [Fri, 11 Nov 2022 22:42:18 +0000 (23:42 +0100)]
package/lxc: fix systemd build

Fix the following systemd build failure raised since bump to version
5.0.1 in commit db19998035701d08bf1b5fd366e0b5aa9adb8e57:

../src/lxc/cgroups/cgfsng.c: In function 'unpriv_systemd_create_scope':
../src/lxc/cgroups/cgfsng.c:1234:104: error: incompatible type for argument 10 of 'sd_bus_call_method_asyncv'
 r = sd_bus_call_method_asyncv(bus, NULL, DESTINATION, PATH, INTERFACE, "Subscribe", NULL, NULL, NULL, NULL);
                                                                                                       ^~~~

Fixes:
 - https://gitlab.com/buildroot.org/buildroot/-/jobs/3282261449

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoDEVELOPERS: drop entry related to inexisting directory
Thomas Petazzoni [Fri, 11 Nov 2022 21:58:42 +0000 (22:58 +0100)]
DEVELOPERS: drop entry related to inexisting directory

Commit
4e7dfe20bb8cf0cccfbd177321eff73e82c8d940 ("configs/friendlyarm_nanopi_m4:
remove defconfig") forgot to fully reflect the removal of the
defconfig in the DEVELOPERS file, causing a get-developers warning:

WARNING: 'board/friendlyarm/nanopi-m4' doesn't match any file

Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoconfigs/kontron_bl_imx8mm: U-Boot needs util-linux
Thomas Petazzoni [Fri, 11 Nov 2022 20:19:10 +0000 (21:19 +0100)]
configs/kontron_bl_imx8mm: U-Boot needs util-linux

Fixes:

/usr/bin/ld: cannot find -lgnutls
/usr/bin/ld: cannot find -luuid
collect2: error: ld returned 1 exit status

when building U-Boot.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3301821264

Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoconfigs/rock_pi_n10: remove defconfig
Thomas Petazzoni [Wed, 2 Nov 2022 20:18:57 +0000 (21:18 +0100)]
configs/rock_pi_n10: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3259671399

Cc: Suniel Mahesh <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoconfigs/rock_pi_4: remove defconfig
Thomas Petazzoni [Wed, 2 Nov 2022 20:18:56 +0000 (21:18 +0100)]
configs/rock_pi_4: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3259671395

Cc: Suniel Mahesh <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoconfigs/roc_pc_rk3399: remove defconfig
Thomas Petazzoni [Wed, 2 Nov 2022 20:18:55 +0000 (21:18 +0100)]
configs/roc_pc_rk3399: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3259671226

Cc: Suniel Mahesh <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/udisks: fix the tool name in the config help
Wolfgang Grandegger [Tue, 8 Nov 2022 17:19:04 +0000 (18:19 +0100)]
package/udisks: fix the tool name in the config help

The name of the tool in udisks2 is udisksctl.

Signed-off-by: Wolfgang Grandegger <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/boost: fix patch formatting issue
Thomas Petazzoni [Tue, 8 Nov 2022 07:40:59 +0000 (08:40 +0100)]
package/boost: fix patch formatting issue

Commit ea38acd17d88b9eaf853313398c772b94338ad47 ("package/boost:
backport development branch changes to fix MIPS64 build failure")
introduced a patch formatting issue detected by check-package. This
commit fixes this issue.

Fixes:

  package/boost/0001-Improve-modfunc-performance.patch:4: generate your patches with 'git format-patch -N'

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3287511158

Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/libkcapi: disable -Werror
Fabrice Fontaine [Mon, 7 Nov 2022 22:41:05 +0000 (23:41 +0100)]
package/libkcapi: disable -Werror

Fix the following musl build failure (disabling -Werror seems to be the
only "solution": https://github.com/smuellerDD/libkcapi/issues/136):

lib/kcapi-kernel-if.c: In function '_kcapi_common_send_meta':
lib/kcapi-kernel-if.c:212:12: error: unsigned conversion from 'long int' to 'long unsigned int' changes value from '-4' to '4294967292' [-Werror=sign-conversion]
  212 |   header = CMSG_NXTHDR(&msg, header);
      |            ^~~~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/c789d43e71e6075e297bef58d888228a89055bbc

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/s6-linux-init: bump to version 1.0.8.0
Fabrice Fontaine [Sun, 6 Nov 2022 15:48:31 +0000 (16:48 +0100)]
package/s6-linux-init: bump to version 1.0.8.0

This bump will fix the following build failure raised since bump of
skalibs to version 2.12.0.1 in commit
0f7bfc80814e90918b9527cab4cbf6afedd22195 thanks to
https://github.com/skarnet/s6-linux-init/commit/6696cee3bd309f61fc6c945ef747262a845e9267

In file included from src/shutdown/hpr_shutdown.c:8:
src/shutdown/hpr.h:20:40: error: unknown type name 'tain_t'; did you mean 'tain'?
   20 | extern int hpr_shutdown (unsigned int, tain_t const *, unsigned int) ;
      |                                        ^~~~~~
      |                                        tain

Update hash of COPYING (update in year with
https://github.com/skarnet/s6-linux-init/commit/42a91a98b780759257862642c764c2102237fea6)

https://github.com/skarnet/s6-linux-init/blob/v1.0.8.0/NEWS

Fixes:
 - http://autobuild.buildroot.org/results/4382beeb0f94d583ba605c70a47f34b45c76e311

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/lxc: add systemd optional dependency
Fabrice Fontaine [Sun, 6 Nov 2022 15:38:03 +0000 (16:38 +0100)]
package/lxc: add systemd optional dependency

Fix the following build failure raised since bump to version 5.0.1 in
commit db19998035701d08bf1b5fd366e0b5aa9adb8e57:

Run-time dependency systemd found: NO (tried pkgconfig and cmake)

../output-1/build/lxc-5.0.1/config/init/systemd/meson.build:11:4: ERROR: Dependency "systemd" not found, tried pkgconfig and cmake

Fixes:
 - http://autobuild.buildroot.org/results/495175c367084093bc01c40290736c738d876894

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/libkcapi: fix uclibc build
Fabrice Fontaine [Mon, 7 Nov 2022 17:27:18 +0000 (18:27 +0100)]
package/libkcapi: fix uclibc build

Fix the following uclibc build failure raised since bump to version
1.4.0 in commit 06a9dc3528847ec6c12a9b6188d4106e086eabe5 and
https://github.com/smuellerDD/libkcapi/commit/12f19b9a1dd308117f83e8cb33e28e3c040710a0:

lib/kcapi-kernel-if.c: In function '_kcapi_common_send_meta':
lib/kcapi-kernel-if.c:196:26: error: conversion to 'int' from 'size_t' {aka 'unsigned int'} may change the sign of the result [-Werror=sign-conversion]
  196 |         msg.msg_iovlen = kcapi_downcast_int(iovlen);
      |                          ^~~~~~~~~~~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/eccf4b84670b5ef0fdd68b46338edf5043c7cc0d

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/ipmitool: fix per-package build
Fabrice Fontaine [Mon, 7 Nov 2022 17:25:30 +0000 (18:25 +0100)]
package/ipmitool: fix per-package build

Fix the following per-package build failure raised since commit
82480ee22bdfae6f7d399975722b5c125749d8c7:

configure.ac:44: error: possibly undefined macro: AC_SEARCH_LIBS
      If this token and others are legitimate, please use m4_pattern_allow.
      See the Autoconf documentation.

Fixes:
 - http://autobuild.buildroot.org/results/ecf72cd1ebd53f3b3b10f0a24082f9c0ab5d4d12

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/linux-tools: fix static build
Fabrice Fontaine [Sun, 6 Nov 2022 22:20:42 +0000 (23:20 +0100)]
package/linux-tools: fix static build

Pass TARGET_LDFLAGS (which contains -static) to fix the following static
build failures with gpio, iio and pci:

  LINK     lsgpio
/home/autobuild/autobuild/instance-8/output-1/host/lib/gcc/arm-buildroot-linux-musleabi/10.4.0/../../../../arm-buildroot-linux-musleabi/bin/ld: /home/autobuild/autobuild/instance-8/output-1/host/lib/gcc/arm-buildroot-linux-musleabi/10.4.0/libgcc.a(_dvmd_lnx.o): in function `__aeabi_ldiv0':
/home/autobuild/autobuild/instance-8/output-1/build/host-gcc-final-10.4.0/build/arm-buildroot-linux-musleabi/libgcc/../../../libgcc/config/arm/lib1funcs.S:1499: undefined reference to `raise'

[...]

  LINK    iio_event_monitor
/home/thomas/autobuild/instance-2/output-1/host/lib/gcc/microblazeel-buildroot-linux-musl/11.3.0/../../../../microblazeel-buildroot-linux-musl/bin/ld: /home/thomas/autobuild/instance-2/output-1/host/lib/gcc/microblazeel-buildroot-linux-musl/11.3.0/libgcc.a(unwind-dw2.o): in function `size_of_encoded_value':
/home/thomas/autobuild/instance-2/output-1/build/host-gcc-final-11.3.0/build/microblazeel-buildroot-linux-musl/libgcc/../../../libgcc/unwind-pe.h:88: undefined reference to `abort'

Fixes:
 - http://autobuild.buildroot.org/results/f202eb843ef331939f5f12325bdbf2d0d664a7ce
 - http://autobuild.buildroot.org/results/504ec8be1ebd15c5da09f8b90bb723a58d4c58c5

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/gptfdisk: fix popt static build
Fabrice Fontaine [Sun, 6 Nov 2022 23:04:50 +0000 (00:04 +0100)]
package/gptfdisk: fix popt static build

Fix the following static build failure with popt and iconv raised since
bump to version 1.0.9 in commit 69015ce94ac3ffe7f349ec37868eea0d0ffec90e
and
https://sourceforge.net/p/gptfdisk/code/ci/122b58ad82f1a144226d262c87241ee035ed1aff
(which added an unified Makefile):

/home/autobuild/autobuild/instance-0/output-1/host/bin/mips64el-buildroot-linux-uclibc-g++ crc32.o support.o guid.o gptpart.o mbrpart.o basicmbr.o mbr.o gpt.o bsd.o parttypes.o attributes.o diskio.o diskio-unix.o sgdisk.o gptcl.o -static -liconv -lpopt  -o sgdisk
/home/autobuild/autobuild/instance-0/output-1/host/lib/gcc/mips64el-buildroot-linux-uclibc/11.3.0/../../../../mips64el-buildroot-linux-uclibc/bin/ld: /home/autobuild/autobuild/instance-0/output-1/host/mips64el-buildroot-linux-uclibc/sysroot/usr/lib64/../lib64/libpopt.a(poptint.o): in function `strdup_locale_from_utf8':
poptint.c:(.text+0x113c): undefined reference to `libiconv_open'

As can be seen above, this build failure is raised because -liconv is
added before -lpopt so use pkgconfig and SGDISK_LDLIBS

The addition of -liconv in LDLIBS could probably be removed in a
follow-up patch for next branch

Fixes:
 - http://autobuild.buildroot.org/results/c9f2c9e737c2dd1cd4c1a08a5e8a48165179282d

Signed-off-by: Fabrice Fontaine <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agopackage/boost: backport development branch changes to fix MIPS64 build failure
Andreas Ziegler [Mon, 7 Nov 2022 10:05:27 +0000 (11:05 +0100)]
package/boost: backport development branch changes to fix MIPS64 build failure

Backport from Boost/intrusive development branch. Fixes issue introduced with
version 1.80.

Issue: https://github.com/boostorg/intrusive/issues/79
Fixes: http://autobuild.buildroot.net/results/6cb/6cbc0672f3c5046405793d8cdc8f961d2ffb1d3e
Signed-off-by: Andreas Ziegler <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
2 years agoconfigs/orangepi_win: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:20:27 +0000 (23:20 +0100)]
configs/orangepi_win: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499145

Cc: Jagan Teki <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/orangepi_rk3399: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:20:21 +0000 (23:20 +0100)]
configs/orangepi_rk3399: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499144

Cc: Suniel Mahesh <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/orangepi_prime: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:20:15 +0000 (23:20 +0100)]
configs/orangepi_prime: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499141

Cc: Jagan Teki <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/orangepi_plus: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:20:11 +0000 (23:20 +0100)]
configs/orangepi_plus: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

This defconfig did not even have an entry in the DEVELOPERS file.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499140

Cc: Shyam Saini <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/orangepi_pc2: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:20:06 +0000 (23:20 +0100)]
configs/orangepi_pc2: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499136

Cc: Jagan Teki <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/friendlyarm_nanopi_r1: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:20:00 +0000 (23:20 +0100)]
configs/friendlyarm_nanopi_r1: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499056

Cc: Davide Viti <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/friendlyarm_nanopi_neo_plus2: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:19:55 +0000 (23:19 +0100)]
configs/friendlyarm_nanopi_neo_plus2: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499055

Cc: Louis Aussedat <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/friendlyarm_nanopi_neo4: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:19:50 +0000 (23:19 +0100)]
configs/friendlyarm_nanopi_neo4: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499054

Cc: Marek Belisko <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/friendlyarm_nanopi_neo2: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:19:43 +0000 (23:19 +0100)]
configs/friendlyarm_nanopi_neo2: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499053

Cc: Jagan Teki <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/friendlyarm_nanopi_m4: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:19:38 +0000 (23:19 +0100)]
configs/friendlyarm_nanopi_m4: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499052

Cc: Suniel Mahesh <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/friendlyarm_nanopi_m1: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:19:30 +0000 (23:19 +0100)]
configs/friendlyarm_nanopi_m1: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

This defconfig did not even have an entry in the DEVELOPERS file.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499051

Cc: Filip Skoneczny <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/friendlyarm_nanopi_m1: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:18:49 +0000 (23:18 +0100)]
configs/friendlyarm_nanopi_m1: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

This defconfig did not even have an entry in the DEVELOPERS file.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499050

Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/friendlyarm_nanopi_a64: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:03:40 +0000 (23:03 +0100)]
configs/friendlyarm_nanopi_a64: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499049

Cc: Jagan Teki <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/friendlyarm_nanopc_t4: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:03:39 +0000 (23:03 +0100)]
configs/friendlyarm_nanopc_t4: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499048

Cc: Suniel Mahesh <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
2 years agoconfigs/bananapi_m64: remove defconfig
Thomas Petazzoni [Tue, 1 Nov 2022 22:03:38 +0000 (23:03 +0100)]
configs/bananapi_m64: remove defconfig

This defconfig has been failing to build for several months, with
nobody stepping up to fix it. It's time to drop it.

See
https://lore.kernel.org/buildroot/20220806224338.0159e15c@windsurf/
for a notification on August 6, 2022 about the issues with this
defconfig.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/3234499008

Cc: Jagan Teki <[email protected]>
Signed-off-by: Thomas Petazzoni <[email protected]>
Signed-off-by: Yann E. MORIN <[email protected]>
This page took 0.107916 seconds and 4 git commands to generate.