The manual is a bit vague about whether Config.in and external.mk
are mandatory or optional.
Make it explicit in the manual that they are mandatory.
Reported-by: Jérémy Rosen <jeremy.rosen@openwide.fr>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
All PowerPC CPUs do not support altivec instruction set; so enable its
support only when the sub-architecture support it.
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Add info about how to install a custom uboot on the zedboard.
Signed-off-by: Spenser Gilliland <spenser@gillilanding.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Forcibly disable the vorbis encoder support when the vorbis option isn't
enabled to avoid autodetection.
Said autodetection has a build bug that it won't properly set libvorbis
build options when the decoder is disabled. Fixes:
http://autobuild.buildroot.net/results/a46/a46b4c97fe126b23a96051a8bf1243a7f4935434/
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
Reviewed-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
[Thomas: update to recent Buildroot which had renamed several
Config.in options, lock down kernel headers version]
Signed-off-by: Spenser Gilliland <spenser@gillilanding.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The rpi-compositor was broken with the 1.4.0 release,
but we now have a fix from upstream.
Add this patch, and remove the 'depends on BROKEN'.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
With toolchains that include linux kernel headers before the 3.4 kernel,
an issue exists with linux/mmc/ioctl.h in that it doesn't include
linux/types.h.
Merged the patch with an existing patch for mmc-utils as it modifies
the same file and fixes similar issues. Also updated the patch to
conform to the buildroot patch naming standard.
Fixes:
http://autobuild.buildroot.net/results/828/828c7a686736710c4fcb152230d0fa6b5f4d6ad3
Signed-off-by: Ryan Barnett <rjbarnet@rockwellcollins.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Cleanup mixed indents and remove commented lines.
Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
If the grub package is selected it also selects
BR2_HOSTARCH_NEEDS_IA32_COMPILER. This triggers a test in dependencies.sh
to verify the host compiler can build 32 bit executables. Currently this
test does not set any output for the compiler which causes a stray a.out
to be create outside the output directory. This patch sets the compiler
output to /dev/null so no a.out is created but the test is still performed.
Signed-off-by: Martin Bark <martin@barkynet.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since the opencv package globally depends on BR2_TOOLCHAIN_HAS_THREADS
and BR2_USE_WCHAR, then cleanup these redundant (and useless)
dependencies in the opencv's sub-options.
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Thanks to the previous patches, no-one is using pkgparentdir anymore, so
remove it.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The fourth parameter to inner-generic-package is no longer used. Removing
this parameters requires renaming all usages of $(5) to $(4), and updating
the calls to inner-generic-package (and equivalent for the other package
infrastructures).
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
In preparation of the removal of the 4th parameter to inner-generic-package
and the pkgparentdir helper function, this patch removes the direct usage of
this 4th parameter. The remaining usage
ifeq ($(4),boot/)
can become
$(filter boot/%,$(pkgdir))
instead (and similar for toolchain).
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Variable FOO_DIR_PREFIX in inner-generic-package isn't really needed. The
contents of this variable are 'package' for normal packages, 'boot' for
bootloaders, and 'linux' for the linux kernel.
When patching a package, all you need to know is the directory where
patches can reside, which is already returned by $(pkgdir). In order to be
able to use this variable outside of inner-generic-package, we introduce a
target-specific variable PKGDIR that equals to this $(pkgdir).
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
When calling make 'functions', the $(call) keyword is only needed if the
function takes arguments. For pkgdir, pkgname and pkgparentdir this is not
the case, so we can remove the call to make things more readable.
Suggested-by: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
To make the naming consistent (qstripped variant of a config option
should be named BR_XXX).
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This patch removes the sh2, sh3 and sh3eb support, because the user
base is inexistent, and the Linux support for these architectures is
poor. The sh2a support is preserved, because at least one user
expressed interest in this architecture, and is actually using it:
http://lists.busybox.net/pipermail/buildroot/2013-April/070399.html
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The default of choosing the fixed-point implementation if
the target hardware does not have floating-point unit is not
always the best choice. Modern ARM hardware has floating
point support but most likely the fixed-point implementation
in Opus will still perform better. In particular if your CPU
supports the NEON extension.
This commit allows to select the fixed-point implementation
manually and sets the default based on the availability of
a floating-point unit.
[Peter: force option for sw float, slightly reword]
Signed-off-by: Sven Neumann <neumann@teufel.de>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The pread/pwrite handling on the 0.9.33 branch doesn't take into
consideration architectures with an ABI requiring natural alignment for
64bit arguments (afaik ARM EABI, MIPS O32, PPC, SH and Xtensa), causing
havoc.
The issue has been brought with the uClibc developers, but so far a fix is
not available - So lets revert the patches for now as they cause real
issues.
For further details, see:
http://lists.busybox.net/pipermail/buildroot/2014-January/087050.html
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Our current stripping strategy requires that shared libraries have the
executable permission. However, this is by far not something
recognized as a standard behavior: Debian/Ubuntu distributions for
example do not have executable permissions on their
libraries. Therefore, pushing to upstream packages fixes that add the
executable permissions is not easy.
As a result, this commit improves the stripping logic so that it not
only strips the files that are executable, but also the ones that
match '*.so*', which should match both the shared libraries and the
dlopen()'able plugins, as long as they have a .so extension.
Thanks to this addition, a number of manual "chmod +x" done by various
packages can be removed.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The configure script finds libz in the distribution libraries, which causes
zlib support to be dropped from the cross-compiled GnuTLS.
Signed-off-by: Antoine Pierlot-Garcin <antoine@bokbox.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
If two files with the same relative paths exist in multiple overlay
skeletons, and they have the same modification time and size, then rsync
might not copy the later file on top of the earlier file. This patch fixes
this by adding the -I option to the rsync commands used in the overlay
skeleton file installations. ("man rsync" indicates that this option turns
off the file-size/mod-date "quick check" behavior, causing all files to be
updated - more like the cp commands that we had originally.)
[Peter: use --ignore-times to make it obvious what the option does]
Signed-off-by: Danomi Manchego <danomimanchego123@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
iucode_tool is a program to manipulate Intel® X86 and X86-64 processor
microcode collections, and to use the kernel facilities to upgrade the
microcode on Intel system processors.
It was originally sent by Richard Braun <rbraun@sceen.net>
[Peter: drop 'sid' from upstream URL]
Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Cc: Richard Braun <rbraun@sceen.net>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Some compilers, such as the Blackfin GNU compiler, prefix a character to any
C symbol in generated assembly code. If any assembly symbol is invoked
from C code, it needs to be prefixed as well.
Note: since autoreconf doesn't work with this package because automake
isn't used.
Fixes:
http://autobuild.buildroot.net/results/c40a22814b405de2c5c75139cfc393c863ed4b81/
[Ryan: add information about why patching configure is ok]
[Ryan: add renaming of patches to be consistent with standard]
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Ryan Barnett <rjbarnet@rockwellcollins.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes a couple of buffer overflows so it should be considered as a
security bump (no CVEs though).
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The uClibc options for sh2a is SH2A, not SHA2, and this value should
be used for sh2a, not sh2.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The current git download helper creates the tarball by doing:
git archive <version> | gzip -c > <tarball>
Unfortunately, even if "git archive" fails and returns a non-zero
error code, gzip ignores that, compresses nothing, and returns success
(zero error code). The consequence of this behavior is that when the
git version provided in the package is incorrect, we are not failing
during the download step, but later on when trying to extract the
tarball (which was incorrectly created as a result of the failing git
archive).
To fix this, we change the tarball creation logic to:
git archive -o <tarball>.tmp <version> &&
gzip -c <tarball>.tmp > <tarball> &&
rm -f <tarball>.tmp
If the build is interrupted during the "gzip" command, we may leave
the .tmp file behind us, but this also happens with wget downloads,
and is generally not considered a problem, since this temporary file
will be overwritten next time we attempt to do download this package.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Probably due to some copy/paste mistake, Config.in.sh was defining
BR2_ENDIAN to "LITTLE" when the architecture is BR2_x86_64. Due to the
fact that Config.in.sh is never included when we are building for
x86_64, this is a no-op, but is certainly good to clean up.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
