Internal xz backend is statically linked and need a static
libintl library when BR2_NEEDS_GETTEXT_IF_LOCALE is set.
But for a shared only build, there is no static libintl available
in the sysroot.
Use the external xz library provided by Buildroot instead of
using the internal xz backend.
Fixes:
http://autobuild.buildroot.net/results/2c6/2c61562008418ff58fa3085a3342b0e304923073/
Signed-off-by: Romain Naour <romain.naour@openwide.fr>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Allow enabling features depends on the relying packages.
Add the old sub-options to Config.in.legacy.
Signed-off-by: Romain Naour <romain.naour@openwide.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
rpm check for libintl library if BR2_NEEDS_GETTEXT_IF_LOCALE is set.
checking for GNU gettext in libintl... yes
checking whether to use NLS... yes
checking where the gettext function comes from... external libintl
checking how to link with libintl... -lintl
But the dependency on gettext package is missing to ensures
reproducible builds.
Signed-off-by: Romain Naour <romain.naour@openwide.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
If the libtool used by the package is 2.4 (i.e. with no patchlevel), we
end up with a confusing message:
/bin/sh: line 0: test: -gt: unary operator expected
That's because patchlevel is empty, and thus there is nothing to comapre
to 2, and test complains.
Fake a patchlevel of 0 (which is really what a missing patchlevel means
anyway) in that case.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Peter Korsgaard <jacmet@uclibc.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
[Thomas: add dependency on either readline or libedit, which is needed
for new versions of neardal.]
Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
i.MX6 SoloLite SoC has a different external memory base address
than the other i.MX6 family SoCs:
- 0x80000000 for i.MX6 SoloLite (see i.MX6 SoloLite Reference Manual [1])
- 0x10000000 for the others (see i.MX6Quad/Dual Reference Manual [2])
[1] http://cache.freescale.com/files/32bit/doc/ref_manual/IMX6SLRM.pdf?fasp=1
Table 2-1 System Memory Map, MMDC - DDR Controller
[2] http://cache.freescale.com/files/32bit/doc/ref_manual/IMX6DQRM.pdf?fasp=1
Table 2-1 System Memory Map, MMDC - DDR Controller
Signed-off-by: Gilles Talis <gilles.talis@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Memtest86+ is a bootable standalone memory test program.
Please note that this is the forked memtest86+ program and not
the original memtest86 which has different licensing. Buildroot
does not support packages with a '+' sign in their name.
Memtest86+ is a utility designed to test whether your memory
is in working order. It repeatedly writes an enormous amount
of different patterns to all memory locations and reads them
back again and verifies whether the result of the read is the
same as what was written to memory.
Memtest86+ will only work on 32-bit or 64-bit x86 targets.
It boots as an i486 program and autodetects hardware.
[Peter: tweak help text as suggested by Thomas]
Signed-off-by Stephen M. Kenton <skenton@ou.edu>
Reviewed-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix:
WARNING: No hash found for perl-5.20.1-cross-0.9.4.tar.gz
Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The protoc stuff changed a bit since 0.9.3. When 0.9.4 was released,
support for cross-compiling was broken because of it. A patch was merged
upstream shortly after that fixes the problem, so I included it here.
When the next version is released, the patch should not be necessary
anymore.
[Peter: actually change _VERSION]
Signed-off-by: Simon Marchi <simon.marchi@polymtl.ca>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Though U-Boot introduced a Kconfig and Kbuild build system with version
2014.10 the old build commands with <board>_config are still supported.
Tested with BR2_TARGET_UBOOT_BOARDNAME="m28evk"
Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: David Bender <codehero@gmail.com>
[yann.morin.1998@free.fr:
- remove ldaps config option, use build-time dependencies
- add dependencies against openssl/gnutls/libnss as appropriate
- fix mis-detected tls support when openssl is enabled
- fix strip at install time
- cleanup configure arguments
- add hash file
- fix help entry
]
[Peter: needs wchar]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Jinja2 is a general purpose templating language for python and can be
required during build time for python scripts that generate code.
While currently there are no in-tree Buildroot packages which depend on
host-python-jinja2, I'm currently having to deal with a proprietary
software build system that requires it.
I have tested that it builds and works correctly as a host python
package for a Buildroot host-python and someone else might require it,
so I'm proposing its inclusion.
python-jinja2 for target builds fine without python-markupsafe as a
build dependency, but when building host-python-jinja2 without
host-python-markupsafe installed, the python package manager tries to
download and install it.
To avoid that and install host-python-markupsafe properly through
Buildroot, the host build dependency is needed. Host support for
python-markupsafe was submitted in a previous patch from this patchset.
Signed-off-by: Erico Nunes <erico.nunes@datacom.ind.br>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
python-markupsafe is a dependency for python-jinja2, which can be used
during build time as a host tool to generate code based on templates.
This package is being added host package support mostly to allow
host-python-jinja2 to build.
This has been tested to build and work correctly as a host tool as well.
Signed-off-by: Erico Nunes <erico.nunes@datacom.ind.br>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The adaptation from commit 74dd54bf is incomplete/bad causing segfaults when
using cryptodev for digest offload, examples: openssh, openssl speed, others.
Tested on real hardware (talitos).
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Switch sed options around since defaults have changed.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: David Bender <codehero@gmail.com>
[yann.morin.1998@free.fr:
- add optional dependency to nettle
- fix build against nettle
- switch to using the ftp for release, instead of github
- add missing dependency to host-pkgconf
- remove extraneous configure options, we already pass them
- drop extraneous ac_cv_func_memcmp_working, we already pass it
- drop extraneous ac_cv_func_uname, it is properly found
- add hash file
- tweak help entry
]
[Peter: license is BSD-2c]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The new c_rehash from openssl 1.0.2 can't take a minus in the directory
string since the regex for matching commands checks for - in any
position instead of just the beginning to trigger the command parser. Fixes:
http://autobuild.buildroot.net/results/ee6/ee683569350d5deaf0ccc603ed7066bffb83cbe3/
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Squid 3.5.x deprecated the use of HOSTCXX & friends in favour of
BUILDCXX and BUILDCXXFLAGS.
When they are not specified BUILDCXXFLAGS takes flags from target
CXXFLAGS which specifies C11 standard and which older vesions of gcc
don't understand, thus causing breakage.
cf_gen hasn't got any C11 features so it's not required. Fixes:
http://autobuild.buildroot.net/results/816/8162e4ec941e7642248373be47cca99113a648e8/
Also drop ACLOCAL and AUTOMAKE trickery from e27ccbab since it's no
longer required.
And fix a typo in the hash file pointing to an improper hash file
upstream.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Configure ffmpeg to use freetype and fontconfig
if those packages are selected.
ffmpeg freetype support require fenv.h, which is not available in uClibc or
musl, except for certain archs / depending on specific configuration
options, so only enable it for glibc.
[Peter: only enable on glibc, simplify logic]
Signed-off-by Stephen M. Kenton <skenton@ou.edu>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
