imx6ul has currently an issue on kernel 4.7 that causes a stall when running
the "reboot" command.
This issue has been reported in the linux-arm-kernel mailing list, but we
don't have a proper fix at the moment.
This problem is not seen when SMP is disabled, so let's disable it for now
until a proper fix becomes available.
Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Following reports from me of build failures of the GDB simulator for the
Blackfin architecture, Waldemar cooked a
patch (0005-fix-sim-compile.patch) that removes the typedef of SIM_CPU,
because there was a redefinition of this typedef for Blackfin. This was
not causing an issue with recent compilers as redefining the same
typedef is valid with recent compilers, but was causing build failures
with gcc 4.4.x.
However, by removing the common definition of SIM_CPU, this patch broke
the build of the GDB simulator on other architectures, which did not had
an architecture-specific redefinition of SIM_CPU (unlike Blackfin).
The crux of the problem is in a commit from Mike Frysinger, that tries
to refactor the SIM_CPU definition into a common one. Except that it
leaves a redefinition of it for Blackfin. Removing this second
definition however doesn't easily work, due to include ordering
issues. The easiest solution is to simply revert the patch from Mike
Frysinger. This allows to fix the build for all architectures and all
compiler versions.
Fixes:
http://autobuild.buildroot.net/results/3b82c44ee853fab0e0c63881f0705bb659412917/http://autobuild.buildroot.net/results/dafbb93ab38a4285ce42436219d552cceb14828b/
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When the kernel headers in the toolchain don't provide __NR_getrandom,
eudev has a predefined set of values. However, the provided value for
ARM64 is incorrect, and causes a runtime problem on this architecture
when old kernel headers (not providing __NR_getrandom) are used.
This commit adds a patch to eudev to fix this issue.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: Kevin Hilman <khilman@baylibre.com>
Two patches are removed, as they have been upstreamed:
- 130-fix_build_with_gcc-6.patch (svn commit 233721, Git commit
8c3fa311caa86f61b4e28d1563d1110b44340fb2)
- 920-libgcc-remove-unistd-header.patch (svn commit 226092, Git commit
e940d7953f06af11d09229a29ecbcc1ba25b378d)
All other patches have simply been refreshed, with no manual edit
needed.
A build+runtime test has been done with an ARM, Cortex-A8, EABIhf, musl
configuration, booted under Qemu.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This commit fixes the build of shellinabox with the musl C library,
which requires two changes:
- A patch to include <sys/ttydefaults.h>, which is needed to get the
definitions of TTYDEF_*
- A hack to workaround what seems to be a problem in musl itself (musl
does #define utmp utmpx, which causes some symbol conflicts down the
road). Since anyway the utmpx implementation is just a set of stubs
in musl, we simply make shellinabox believe that <utmpx.h> is not
available by passing the appropriate variable. The musl issue has
been reported at http://www.openwall.com/lists/musl/2016/08/04/8.
Fixes:
http://autobuild.buildroot.net/results/1847cab964957da3c9bf4911a5ad3602b3c82431/
Signed-off-by: Olivier Singla <olivier.singla@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This patch bumps mono to version 4.4.2.11
Signed-off-by: Angelo Compagnucci <angelo.compagnucci@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This reverts commit 5ca9b7cf66. Installing
dtc headers in $(HOST_DIR) causes some build failures of
host-uboot-tools due to a conflicting libfdt.h header.
The possible solution to solve this problem, based on -isystem, has been
reverted has it was causing too many issues that we didn't want to solve
so close to cutting 2016.08-rc1.
Since the headers of libfdt for the host are currently not needed for
any package in the Buildroot tree, it is easier and safer to just revert
this patch.
Fixes:
http://autobuild.buildroot.net/results/9fba77f4080bf6bc26e341f3f383cdba26c0e14f/
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The main benefit of using make foreach loops is that they will abort
if one of the iteration of the loop fails. The current for loops will
continue, and only report a failure if the last iteration was a
failure, but will silently ignore other errors.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Matt Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This long list of commands to copy the terminfo files from staging to
target is really silly and can be factorized using a
NCURSES_TERMINFO_FILES variable, which is then iterated through.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Matt Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Also adding the '-m4' device tree blobs which allow to disable the
peripherals that are used by the Cortex-M4.
The 6x_bootscript is already setup to load the '-m4' dtb if m4enabled
u-boot variable is set to 1.
The FreeRTOS BSP for both Nitrogen6SX and Nitrogen7 can be found here:
https://github.com/boundarydevices/freertos-boundary
Signed-off-by: Gary Bisson <gary.bisson@boundarydevices.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Changelog:
* imxv4l2videosrc:
* Stricter checks for video device capabilities
* Add support for more video formats like 8/16-bit grayscale
* pxp: Fix incorrect return code which broke blitter with GStreamer 1.8
* blitter:
* Improvements in error logging
* Fixes for RGBx framebuffer format detection
* Fix for subtle tearing problem by implementing triple buffering
* audio:
* Replace objdump-based scan during build configuration
Codec filenames are instead moved to the C code; the libdir
path of the libfslaudiocodec.pc file is used to know where on
the target the files are placed
* Explicitely add the libfslaudiocodec.pc libdir to the MP3 encoder
Fixes build errors in buildroot
* Minor logging fixes
* misc:
* Proper check for region copies in phys_mem_meta
* Add missing G_END_DECLS lines in headers
Tested with the following commands on i.MX6Q (IPU):
# gst-launch-1.0 playbin uri=file:///root/tears_of_steel_1080p.webm
# gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
matroskademux ! imxvpudec ! imxipuvideosink
# gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
matroskademux ! imxvpudec ! imxeglvivsink
Tested with the following commands on i.MX7 (PXP):
# gst-launch-1.0 imxv4l2videosrc device=/dev/video1 ! imxpxpvideosink
Signed-off-by: Gary Bisson <gary.bisson@boundarydevices.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
By default, all the libraries are installed under /usr/lib/imx-mm
which causes problems at runtime.
[Peter: drop -mindepth/-maxdepth, add comment explaining why this is done]
Signed-off-by: Gary Bisson <gary.bisson@boundarydevices.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
By default, all the libraries are installed under /usr/lib/imx-mm
which causes problems at runtime.
The hooks are inspired from the mechanism used in the Yocto recipe:
https://github.com/Freescale/meta-fsl-arm/blob/krogoth/recipes-multimedia/imx-codec/imx-codec.inc
Tested with the following commands:
# gst-launch-0.10 audiotestsrc ! mfw_mp3encoder ! fakesink
# gst-launch-1.0 audiotestsrc ! imxmp3audioenc ! fakesink
[Peter: drop -mindepth/-maxdepth, add comment explaining why this is done]
Signed-off-by: Gary Bisson <gary.bisson@boundarydevices.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The options to purge locales and to generate locale data are currently
located in the toolchain menu. However, these options are not really
related to the toolchain per-se, they are more system-level
configuration options, much like the timezone selection option we
already have in the "System configuration" menu.
Therefore, it makes more sense to have the locale-related options in
the "System configuration" menu as well.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Our current list of locales to keep by default is "C en_US de fr". It
doesn't make much sense to keep "de" and "fr" more than any other
language. So let's keep only the "C" and "en_US" locales by default,
and leave it to the user to specify other locales to keep if needed.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Our current default is to keep all locales installed in
/usr/share/locale/. However, in practice, those locales take up a
significant amount of space, and most users do not need
locales. Therefore, it makes more sense to default to purging locales,
in order to keep only a few useful ones rather than keeping them all.
It helps in providing a small filesystem size by default, and still
allows advanced users who really need locales to tune their
configuration.
As an example, a very basic system with just util-linux enabled (not
even Busybox) weights 11 MB, including 6.4 MB of locales. With this new
default, the generated system is only 4.2 MB.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
With this commit we update ARC defconfigs with the following:
- "snps_axs101_defconfig" and "snps_axs103_defconfig":
- bump linux kernel version to 4.7
- set up host linux headers to 4.7
- bump u-boot version to 2016.07
- "snps_hs38_smp_vdk_defconfig":
- bump linux kernel version to 4.7
- set up host linux headers to 4.7
Also we delete "snps_hs38_vdk_defconfig" as we doesn't support
uni-processor HS38 configuration for vdk any longer.
Signed-off-by: Vlad Zakharov <vzakhar@synopsys.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The output of "/etc/init.d/S40network start/stop" was lacking a
newline due to the usage of printf. Fix it by echoing the status, like
we do in other init scripts.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The cairo package fails to build on some architectures:
BR2_TOOLCHAIN_HAS_LIBATOMIC is enabled, but libatomic is in fact not
available.
This happens because the gcc logic in libatomic/configure.tgt does not
recognize "uclinux" as a valid OS part of the target tuple, and
therefore it does not build libatomic.
The "uclinux" part of the tuple is used by Buildroot when
BR2_BINFMT_FLAT=y, so we make BR2_TOOLCHAIN_HAS_LIBATOMIC enabled only
if !BR2_BINFMT_FLAT.
It is worth mentioning that support for the uclinux tuple could most
likely very easily be added to gcc: it could rely on the generic
"posix" implementation of libatomic, which uses pthread locks,
available on all architectures where thread support is available.
Fixes:
[arm] http://autobuild.buildroot.net/results/3d8dc45e41a043d2c2c26bfb26c3617499fbe671
[m68k] http://autobuild.buildroot.net/results/318e01406e3e92eb589ee5b2231c671a4dbb6da4
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
[Thomas: adjust dependency after analysis of the gcc code.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The boost-context module contains ARM assembly code, which uses ARM
instructions. Therefore, on Thumb-2 only platforms (such as Cortex-M),
the following build error occurs:
libs/context/src/asm/make_arm_aapcs_elf_gas.S: Assembler messages:
libs/context/src/asm/make_arm_aapcs_elf_gas.S:33: Error: unshifted register required -- `bic a1,a1,#15'
libs/context/src/asm/make_arm_aapcs_elf_gas.S:42: Error: immediate value out of range
To avoid this error, we make the boost-context module depend on
BR2_ARM_CPU_HAS_ARM.
Fixes:
http://autobuild.buildroot.net/results/8f6770b59a343b9f710e9363b43227ee9f026660
Signed-off-by: Romain Naour <romain.naour@gmail.com>
[Thomas: use BR2_ARM_CPU_HAS_ARM instead of a dependency on
!BR2_ARM_CPU_ARMV7M.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: David Glessner <david.glessner@rockwellcollins.com>
Signed-off-by: Matt Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
With this new version, the generation of atom-glue fails with the
following error:
Making all in src/lib
make[3]: Entering directory '/mnt/data/src/sfl/zodiac/buildroot/output/build/lldpd-0.9.4/src/lib'
GEN atom-glue.c
arm-buildroot-linux-uclibcgnueabi-cpp.br_real: fatal error: too many input files
compilation terminated.
arm-buildroot-linux-uclibcgnueabi-cpp.br_real: fatal error: too many input files
compilation terminated.
make[3]: *** [Makefile:899: atom-glue.c] Error 1
There is an upstream patch pushed after 0.9.4 which fixes the
corresponding Makefile.am, included in this commit.
Note that since the provided tarball ships the related Makefile.in file,
we need to tell Buildroot to autoreconfigure the package.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
[Thomas: add comment about AUTORECONF=YES in .mk file.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Add a simple sysv init script to start the lldpd daemon at boot.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
[Thomas: cosmetic improvements.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Bump libseccomp version to 2.3.1 to enable power architectures.
Signed-off-by: Judd Meinders <judd.meinders@rockwellcollins.com>
Signed-off-by: Matt Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Upstream i2c-tools executables are installed to /usr/sbin by default.
Make buildroot match this behavior.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Before changing the install directory of i2c-tools to /usr/sbin/ make
sure busybox is built first (if selected) in order to ensure that the
symlinks to the busybox-provided versions are overwritten by the
i2c-tools' install commands.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Upstream has this patch a while and other projects are using
it already. So better switch to this.
Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
As of version 1.25.0 busybox by default includes a nearly complete
implementation of i2c-tools. Hide the upstream version unless the
BR2_PACKAGE_BUSYBOX_SHOW_OTHERS option is selected.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
