Commit [1] "support/testing: improve weston test reliability" moved
out the wait time from the emulator (to run on the test controller).
While doing so, the sleep time which was initially _after_ the
"killall weston" invocation to in stop_weston() was incorrectly
moved before the command invocation. In this state, the test can
succeed on fast host computer running the test. But it will most
likely fail on an average computer.
This commit fixes this issue by moving the sleep time after
the command invocation.
[1] 6561a5d773
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit bedc44c073)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
The patch bumps the Linux kernel to version 6.1.133. The size of xipImage
has increased by only 22 bytes (1671804 bytes compared to 1671782 in
version 6.1.126).
Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit cb6729d214)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
lmbench dependency discovery relies on implicit return types for
main(). This causes compiler errors when -Wimplicit-int is enabled,
which has become the default with recent gcc versions. The failure of
the dependcy discovery causes lmbench to redifine typedefs the
compiler already has in an incompatible manner. Add a patch to fix the
discovery.
Fixes:
bench.h:81:13: error: conflicting types for ‘socklen_t’; have ‘int’
81 | typedef int socklen_t;
bench.h:85:15: error: conflicting types for ‘off64_t’; have ‘int64’ {aka ‘long long int’}
85 | typedef int64 off64_t;
Fixes:
http://autobuild.buildroot.net/results/33cf97a79125c20f67f620eb6a7b5ad2206b2503/
Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit fd914e9e4c)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
libcamera has some optional dependencies that automatically detected at
build time.
To improve reproducibility, we should add a dependency on those optional
dependencies if the symbols that build them are enabled so that the
order in which packages are built does not influence the libcamera
package.
Note that the optional libyuv dependency isn't added as:
1) it is only used for the virtual pipeline and the android feature,
both of which are disabled/not supported right now,
2) libcamera has it in a submodule if missing (though if that works with
Buildroot is to be determined),
3) adding the dependency isn't enough as meson somehow doesn't find the
dependency,
Signed-off-by: Quentin Schulz <quentin.schulz@cherry.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 028bdac7ed)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
meson allows to force all features set to "auto" to default to disabled
except if explicitly enabled via the auto_features argument, c.f.
https://mesonbuild.com/Build-options.html#features
"""
If the value of a feature option is set to auto, that value is
overridden by the global auto_features option (which defaults to auto).
This is intended to be used by packagers who want to have full control
on which dependencies are required and which are disabled, and not rely
on build-deps being installed (at the right version) to get a feature
enabled.
"""
The only auto feature that we hadn't disabled explicitly is the
gstreamer plugin. It is however expected that this wasn't a mistake as
the dependencies for the gstreamer pluging wouldn't have been met
thanks to the explicit LIBCAMERA_DEPENDENCIES we have based on the
presence of the gstreamer symbols in the global config.
This should make it less likely for future releases of libcamera to
regress in terms of reproducibility because of "auto" features (though
changes from "auto" to "enabled" wouldn't be caught and would be
susceptible to race conditions with their dependencies in case they
aren't properly specified and built after libcamera is).
Signed-off-by: Quentin Schulz <quentin.schulz@cherry.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit ef3c3bc9b1)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
We currently disable building the documentation via the "documentation"
feature, but it is enabled nonetheless by meson if sphinx-build-3 is
found on the host.
This makes sure it doesn't happen by making the sphinx-build-3 check
only happen when the "documentation" feature is "auto" or "enabled",
which isn't the case for Buildroot.
The bug seems to have been introduced in v0.0.1 release.
Signed-off-by: Quentin Schulz <quentin.schulz@cherry.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 569272be91)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Commit c1597f989654 ("ipa: raspberrypi: Use YamlParser to replace
dependency on boost"), part of the v0.0.1 release, removed the
dependency on boost, so let's remove it from the pipeline option and
dependency for Buildroot.
Signed-off-by: Quentin Schulz <quentin.schulz@cherry.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 65721c6e0a)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Some EDK II configurations have complex dependencies on several packages
and additional build options; build tests help keeping track of those
more easily.
Factorize some code common to all the build tests into a new
TestEdk2BuildBase class, which defines a base configuration and a method
to assert that binaries do indeed exist after the build.
While at it, add myself in DEVELOPERS.
Signed-off-by: Vincent Stehlé <vincent.stehle@arm.com>
Cc: Dick Olsson <hi@senzilla.io>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 4f7e3f0bdd)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
The manual states that FOO_DL_OPTS are valid for all the different
download backends, but that is not the case: at least the git backend
does not use them (it does not fail, it just ignores them).
Accept FOO_DL_OPTS in the git backend, and pass them to 'git fetch'.
There is no way that we can pass such options to submodules or lfs,
though.
Update the manual accordingly.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit a1012b363a)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
The dependencies for pulseview are not entirely clean:
- arch deps are not first;
- second-level inherited deps are listed;
- deps are not alphabetically ordered (for deps on packages);
- the comment is hidden even when the arch deps are met, because of an
incorrect dependency on Qt5;
- qt5 is a depends-on when it could be a select.
Update the dependencies to fix all the points above.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 70bcb7655b)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Some protocol decoders in libsigrokdecode need the zlib module:
srd: ModuleNotFoundError: Failed to load decoder usb_power_delivery: import by name failed: No module named 'zlib'
srd: Traceback (most recent call last):
File "/usr/share/libsigrokdecode/decoders/usb_power_delivery/__init__.py", line 24, in <module>
from .pd import *
File "/usr/share/libsigrokdecode/decoders/usb_power_delivery/pd.py", line 24, in <module>
import zlib # for crc32
^^^^^^^^^^^
ModuleNotFoundError: No module named 'zlib'
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 9d6ab1244a)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Since iptables v1.8.11, `iptables -C` commands return unexpected zero status
even for cases when the rules don't exist. This breaks e.g. standard Docker
operation, where checking for existing rules is used extensively when creating
networks.
The patch fixing the behavior is available upstream - apply it to v1.8.11
before a newer version is available.
Signed-off-by: Jan Čermák <sairon@sairon.cz>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit f5e7cefe77)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
The ARMV7-A toolchains are capable to compile binaries for ARMv8-A CPU
in AArch32 execution state.
This adds the BR2_ARM_CPU_ARMV8A option in the 'conditions' to allow
ARMV8-A CPU such as Cortex-A53 or Cortex-A72 to use ARMV7-A toolchains.
Signed-off-by: Gaël PORTAY <gael.portay+rtone@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 8fd537ae05)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Without this, packages using libopenmpt are not able to find
the dependency.
Signed-off-by: J. Neuschäfer <j.neuschaefer@gmx.net>
[Julien: reword commit log]
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 24a41c8fb5)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
The VideoCore blobs for the Raspberry Pi 1, 2, 3, Zero, Zero W and Zero
2 W are differents from the ones for the Raspberry Pi 4 and the Compute
Module 4.
The VideoCore blobs for the Raspberry Pi 4 are named with a 4[1] (i.e.
start4.elf, start4x.elf, start4db.elf and start4cd.elf).
This precises which VideoCore firmware blob files are present on the
Raspberry Pi 1, 2, 3, Zero W, Zero 2 W and which files are present on
the the Rapsberry Pi 4 and the Compute Module 4.
Note: The VideoCore blobs for the Raspberry Pi 5 are self-contained in
its bootloader EEPROM[1][2]; there is no additional files for pi5.
[1]: https://www.raspberrypi.com/documentation/computers/configuration.html#start-elf
[2]: https://www.raspberrypi.com/documentation/computers/config_txt.html#start_file-fixup_file
Signed-off-by: Gaël PORTAY <gael.portay+rtone@gmail.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 7ebcfe3cb9)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Any Raspberry Pi may use device-tree overlays; it is not restricted to
Raspberry Pi 3 and 4.
The defconfigs for the Raspberry Pi 3, 4, 400, Zero W, Zero 2 W, Compute
Module 4 and 4s use the miniuart-bt dtoverlay to switch the Bluetooth to
the mini-UART (ttyS0) and restore UART0 (ttyAMA0) over GPIOs 14 and 15
to enable the serial console (like the Raspberry Pi 1 and 2 that have no
Bluetooth)[1][2].
As a consequence, the Raspberry Pi Zero W, Zero 2 W, 400, Compute Module
4 and 4s install the device-tree overlays as well.
This rewords the note by adding the three missing hardwares installing
the images/rpi-firmware/overlays directory. It rewords the miniuart note
at the same time.
[1]: https://www.raspberrypi.com/documentation/computers/configuration.html#uarts-and-device-tree
[2]: https://github.com/raspberrypi/linux/blob/rpi-6.6.y/arch/arm/boot/dts/overlays/miniuart-bt-overlay.dts#L4-L6
Signed-off-by: Gaël PORTAY <gael.portay+rtone@gmail.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit a642bed09a)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Changelog:
* v2.4.3 - 02/28/2025
* Fix memory safety with some older `strerror_r()` implementations in error
formatters for all modules.
* Build
* Fix character device GPIO support tests in Makefile for alternate
shells and older versions of make.
* Fix cross-compilation in Makefile from Windows.
* Fix CMake minimum required version.
* Add CMake package generation.
* Contributors
* Ryan Barnett, @rjbarnet - ec31b39
* javalikescript, @javalikescript - 024a25d
* HopeCollector, @HopeCollector - aca6815, b5e53e6
The hash of the license file has changed due to a copyright year
change:
- Copyright (c) 2014-2023 vsergeev / Ivan (Vanya) A. Sergeev
+ Copyright (c) 2014-2025 vsergeev / Ivan (Vanya) A. Sergeev
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 638fe82e8b)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Upstream changelog from
https://github.com/google/double-conversion/releases/tag/v3.3.1:
Hash pin Github workflows by @joycebrum in #198
Create dependabot.yml by @joycebrum in #199
Add _ITERATOR_DEBUG_LEVEL=2 and _DEBUG defines by @ffa-csturdy in #202
Add bzlmod MODULE.bazel file by @jsharpe in #205
Add CIFuzz Github Action by @DavidKorczynski in #203
Update MODULE.bazel and README.md. by @BYVoid in #234
Remove the explicit dependency on rules_cc. by @BYVoid in #235
Add missing headers by @BhavikaSharma in #239
Tested with the following configuration, which includes the only two
reverse dependencies of double-conversion.
BR2_arm=y
BR2_cortex_a9=y
BR2_ARM_ENABLE_VFP=y
BR2_TOOLCHAIN_EXTERNAL=y
BR2_TOOLCHAIN_EXTERNAL_BOOTLIN=y
BR2_INIT_NONE=y
BR2_SYSTEM_BIN_SH_NONE=y
# BR2_PACKAGE_BUSYBOX is not set
BR2_PACKAGE_QT6=y
BR2_PACKAGE_PYTHON3=y
BR2_PACKAGE_PYTHON_UJSON=y
# BR2_TARGET_ROOTFS_TAR is not set
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 7ffdf75311)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
When building a firmware for the MACCHIATObin with edk2 and
arm-trusted-firmware, the build can randomly fail with the
following make error:
make[1]: Circular output/build/edk2-edk2-stable202411/.stamp_configured <- arm-trusted-firmware dependency dropped.
The message appears also when the build is not failing, depending on
the number of parallel jobs and the build order.
The issue can be observed with the following commands:
cat >.config <<EOF
BR2_aarch64=y
BR2_TOOLCHAIN_EXTERNAL=y
BR2_TARGET_ARM_TRUSTED_FIRMWARE=y
BR2_TARGET_ARM_TRUSTED_FIRMWARE_PLATFORM="a80x0_mcbin"
BR2_TARGET_ARM_TRUSTED_FIRMWARE_EDK2_AS_BL33=y
BR2_TARGET_BINARIES_MARVELL=y
BR2_TARGET_EDK2=y
BR2_TARGET_EDK2_PLATFORM_SOLIDRUN_ARMADA80X0MCBIN=y
BR2_TARGET_MV_DDR_MARVELL=y
EOF
make olddefconfig
utils/brmake
grep -FC5 'dependency dropped' br.log
The circular dependency happen due to [1] and [2].
In fact, only TF-A depends on EDK II (passed as BL33) for building and
not vice versa. See [3]. The EDK II "SolidRun MacchiatoBin" platform
build does not need any TF-A image, compared to some other platforms
such as "Socionext DeveloperBox" or "QEMU SBSA" which are referencing
TF-A images in a hook added in EDK2_PRE_BUILD_HOOKS.
Drop the false dependency on TF-A to fix the build.
This issue has been present since the EDK2 introduction in commit [4].
[1] https://gitlab.com/buildroot.org/buildroot/-/blob/2025.02/boot/arm-trusted-firmware/arm-trusted-firmware.mk#L121
[2] https://gitlab.com/buildroot.org/buildroot/-/blob/2025.02/boot/edk2/edk2.mk#L118
[3] https://github.com/Semihalf/edk2-platforms/wiki/Build_firmware
[4] 1074a37e78
Signed-off-by: Vincent Stehlé <vincent.stehle@arm.com>
Cc: Dick Olsson <hi@senzilla.io>
[Julien: add extra info in commit log]
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 7361a155ef)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
When Building arm-trusted-firmware for the Macchiatobin platform
(a80x0_mcbin), which depends on the mv-ddr-marvell package, the build fails
complaining that this package's folder "does not contain valid
mv-ddr-marvell git repository".
This is expected under Buildroot, where we use intermediate archives.
The issue can be reproduced with the commands:
cat >.config <<EOF
BR2_aarch64=y
BR2_TOOLCHAIN_EXTERNAL=y
BR2_TARGET_ARM_TRUSTED_FIRMWARE=y
BR2_TARGET_ARM_TRUSTED_FIRMWARE_PLATFORM="a80x0_mcbin"
BR2_TARGET_ARM_TRUSTED_FIRMWARE_EDK2_AS_BL33=y
BR2_TARGET_BINARIES_MARVELL=y
BR2_TARGET_EDK2=y
BR2_TARGET_EDK2_PLATFORM_SOLIDRUN_ARMADA80X0MCBIN=y
BR2_TARGET_MV_DDR_MARVELL=y
EOF
make olddefconfig
make
The build is failing with the error message:
plat/marvell/armada/a8k/common/ble/ble.mk:34: *** "'MV_DDR_PATH=/buildroot/output/build/mv-ddr-marvell-d5acc10c287e40cc2feeb28710b92e45c93c702c' was specified, but '/buildroot/output/build/mv-ddr-marvell-d5acc10c287e40cc2feeb28710b92e45c93c702c' does not contain valid mv-ddr-marvell git repository". Stop.
Add patches to fix the build for this platform, for a few versions of TF-A
(v2.6, v2.7, v2.8, lts-v2.8.20, v2.9, v2.10, lts-v2.10.5, v2.11, v2.12 and
lts-v2.12.1).
Signed-off-by: Vincent Stehlé <vincent.stehle@arm.com>
Cc: Dick Olsson <hi@senzilla.io>
Cc: Sergey Matyukevich <geomatsi@gmail.com>
[Julien: add commands to reproduce the issue]
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit fd02add21b)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
libv4l has some really special code that is built explicitly with
_FILE_OFFSET_BITS=32, which doesn't work with _TIME_BITS=64, causing
build failures when BR2_TIME_BITS_64=y. This build issue has been
fixed upstream. While how the upstream patch exactly works is unclear,
it's the patch that upstream has decided to implement to resolve the
build issue, so we simply backport it.
The issue exists since at least upstream commit
99f245f5e2826c7ae3ac8de530bc2fbd906eb62b, which was merged in
v4l-utils 1.26.0.
Fixes:
https://autobuild.buildroot.org/results/616608ef2a44efff67fa21b3263b341da82744c4/
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[Julien: add upstream commit url in patch]
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit bdb5809adf)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
