Ancient distros (especially enterprise-grade still in use) do not accept
ln --relative.
Revert the upstream commit, since the problem it was trying to fix is
not applicable in the context of Buildroot (even with a merged /usr).
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Adam Duskett <Aduskett@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
>From the documentation on BR2_ENABLE_DEBUG, the intention is to get
a build with debug symbols and not a "debug build" since that can have
the unintended consequence of being a different code path then a
release build type definition.
Switch the "Debug" to "RelWithDebInfo" in the cmake package support
to accomodate getting the debug symbols and still be a release
build.
Signed-off-by: Charles Hardin <ckhardin@exablox.com>
Reviewed-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
For fortran detection, some projects check for fortran availability
using the FC/FCFLAGS variables, and others for the legacy F77/FFLAGS
ones.
So, make sure the legacy fortran F77 and FFLAGS variables are set in
TARGET_CONFIGURE_OPTS.
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
TARGET_FCFLAGS is already added to TARGET_CONFIGURE_OPTS, but was not
defined so far. This change fixes this.
Cc: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Benjamin Kamath <bkamath@spaceflight.com>
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
gfortran supports all options supported by gcc, so it can and should be called
via the toolchain wrapper.
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Some of our systems are using ttyUSB0 devices as a serial console.
Add them to mdev.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Joris Lijssens <joris.lijssens@gmail.com>
[Romain:
- fix autoreconf issue by using GETTEXTIZE and AUTORECONF
- fix issue with missing makeinfo
- move version number comment in .mk]
Signed-off-by: Romain Naour <romain.naour@smile.fr>
[Thomas: add GPLv2 license, used for the programs.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently, for a custom headers version, or for the same headers as the
kernel, wedefault to a "very old" version (i.e. 2.6.x in practice).
However, as Vivien explained, when using the same headers as the kernel,
and the kernel is set to use the default version (aka latest version
known to Buildroot) of the kernel, one would expect the headers are
automatically tracking the latest version. Off course, that expectation
is broken because of the above.
However, whatever version we default to, it will probably not be
correct, whether we default to the latest version or to the "very old"
version.
So, simply drop the specific default version, so the default is now the
latest version.
Note: that has the potential to break existing defconfig files that
relied on the "very old" version to be the default. Well, whatever,
they'll get a build failre quite early, and it is easy to fix.
Reported-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
BR2_TOOLCHAIN_BUILDROOT_WCHAR is only defined when uclibc is selected,
whereas BR2_USE_WCHAR is always defined. Due to this, we were disabling
quadmath support even with glibc or musl.
So, use BR2_USE_WCHAR to drive the gcc libquadmath option.
In addition, invert the logic of the condition to use positive logic,
and rework the comment to no longer mention gcc 4.6: libquadmath still
exists, and it still requires wchar support.
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The way the help text and prompt for the kernel header version choice is
written can be misleading to some users.
Rewite them so it is more explicit.
Reported-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Jaap Crezee <jaap@jcz.nl>
[Thomas: minor tweaks in commit log and patch.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The efivar internal build system uses flags such as -std=gnu11 ,
-Wmaybe-uninitialized , -flto which are available in gcc >= 4.7 only.
These flags are passed both to target build and to build a host tool.
For the host part, this has been worked around by overriding
'gcc_flags', but doing that for the target would remove all flags that
upstream intended for the target build.
Buildroot doesn't support building gcc 4.6 in its own toolchain anymore
but it's possible to use an external toolchain with gcc 4.6 which would
be unable to build this package.
This patch adds a limitation on toolchains with target gcc >= 4.7 to
make sure that the flags are available in the chosen toolchain.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
OpenBLAS is optimised for specific CPU models, which don't fully match
with the GCC code generation options. Therefore, we can't automatically
select BR2_PACKAGE_OPENBLAS_TARGET based on the CPU choice. Instead, let
the user select the TARGET name, but offer a sensible default. Other
possible solutions were deemed too complicated: adding choice options in
the ambiguous cases, or only making the option user-visible when there
is ambiguity.
Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
efivar for Nios II also has to be linked with shared libgcc to avoid an
error due to FDE encoding in static libgcc when linking libefiboot.so.
Fixes:
http://autobuild.buildroot.net/results/0c9/0c90e3e7ad41d21dd832f6f266af35fc19185170/
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Disable the -Werror gcc option because is causes the following error
when using gcc 6.x on the host.
p_vmlinx.cpp💯5: error: this ‘if’ clause does not guard...
[-Werror=misleading-indentation] if (a->p_paddr > b->p_paddr) return 1;
Signed-off-by: Martin Bark <martin@barkynet.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
ucl fails with the following error when using gcc 6.x on the host:
checking whether your compiler passes the ACC conformance test... FAILED
configure:
configure: Your compiler failed the ACC conformance test - for details see
configure: `config.log'. Please check that log file and consider sending
configure: a patch or bug-report to <markus@oberhumer.com>.
configure: Thanks for your support.
configure:
configure: error: ACC conformance test failed. Stop.
Fix the issue by using the ISO C90 standard.
Signed-off-by: Martin Bark <martin@barkynet.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
According to setup.py tornado requires backports.ssl_match_hostname
for Python < 3.2 and certifi for Python < 3.4.
Actually both requirements are optional and tornado alone can work
without them, but Python packages like python-circus check requirements
at runtime and though they don't use this funtionality, they insist on
having these packages installed.
An upstream patch fixes backports.ssl_match_hostname dependency and
selecting python-certifi for Python 2 fixes certifi dependency till
this is fixed upstream.
Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This commit reworks the output of the start(), stop() and reload()
functions. The return values of start-stop-daemon are now checked and a
OK or FAIL message is printed out.
Signed-off-by: Andreas Ehmanns <universeII@gmx.de>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
In case of a reload command the snmpdtrapd wasn't restarted.
Signed-off-by: Andreas Ehmanns <universeII@gmx.de>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This patch removes 'set -e' from the netsnmp init script since it causes
the init script to terminate if an error occurs. This prevents the
script to create an "FAIL" printout in case of an error.
Signed-off-by: Andreas Ehmanns <universeII@gmx.de>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Change the init script to get rid of double code. Start, stop and reload
code has been put into separate functions and 'restart' will just call
stop() and start() with a delay in between.
Signed-off-by: Andreas Ehmanns <universeII@gmx.de>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
- remove patch already upstream 0001-remove-use-of-sys-cdefs.h.patch
7b085136a7
- add new dependency on BR2_USE_WCHAR (version 2.2 uses mbtowc())
- add new dependency on locale support
- add note to help text about the need of a working UTF-8 locale
- also rewrap help text to 72 characters
Signed-off-by: Ricardo Martincoski <ricardo.martincoski@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Setools 3.3.8 won't cleanly compile against newer versions
of the new selinux libraries. This patch fixes these errors.
Signed-off-by: Adam Duskett <Aduskett@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
An additional patch is needed to fix the build with uClibc.
Signed-off-by: Adam Duskett <Aduskett@gmail.com>
[Thomas: add patch to fix build with uClibc.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Adam Duskett <Aduskett@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The 0001-execption-lib-path-fix.patch is also no longer needed, as the
new version of libsemanage includes the fix provided by the patch.
As such, the patch was removed, and 0002 was renamed to 0001.
Audit was added as a dependency as the new version of libsemanage
will fail to compile searching for audit.h
Signed-off-by: Adam Duskett <Aduskett@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Thomas: add the select on audit + propagate the dependencies.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
In addition, if a user is using glibc 2.22, the default CFLAG
D_FILE_OFFSET_BITS=64 will cause a compile error. This flag is now
removed from the CFLAGS in the make file to ensure that toolchains
compiled against glibc 2.22 will build the new version of the package
properly.
In addition, libselinux now uses fts(), which is not available on musl,
and not provided by our default uClibc configuration. Therefore,
libselinux now depends on glibc, as well as all its reverse
dependencies.
Signed-off-by: Adam Duskett <Aduskett@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Thomas: add glibc dependency for fts().]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Adam Duskett <Aduskett@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Erlang works fine on mipsel platforms so add it to the supported list.
Signed-off-by: Frank Hunleth <fhunleth@troodon-software.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
MIPS R6 support in Go has not yet been developed.
Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
In most cases Python's package dependencies found in setup.py are
runtime dependencies and hence don't need to be mentioned in *.mk
file.
Also add '# runtime' tag to select statements in Config.in.
__create_mk_requirements() itself is left for future uses (cffi backend
handling etc.).
Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
