Add patch to kill kernel checks since we don't build the kernel
modules anyway.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Commit 009d8fceab introduced
--enable-shared --disable-static options for the host autotools packages,
ultimately causing a regression on the host-python build, leading to
a number of critical modules not being built on the target python on
64 bits system. Introduce a quick fix for the release and before a deeper
fix.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
The prefix is not set when selecting this toolchain, resulting in a
build failure.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Bundled config files have architecture-specific values in them,
which may break if buildroot is configured with another architecture
that does not forcibly set these option.
For example, the bundled config files are for x86_64, and define:
CT_ARCH_TUNE="generic"
This comes from the BR2_GCC_TARGET_TUNE config option (in buildroot) that
is set accordingly to the selected (arch,sub-arch).
But if someone configures buildroot for, say, generic ARM, then the
BR2_GCC_TARGET_TUNE config option is not set, and the crosstool-NG backend
Makefile believes it should not be pushed down to the crosstool-NG config
file.
BUT...
The crosstool-NG backend Makefile can not forcibly push BR2_GCC_TARGET_TUNE
down to the CT-NG config file. If BR2_GCC_TARGET_TUNE is empty, the user can
still set CT_ARCH_TUNE by running ctng-menuconfig.
The backend Makefile already passes such values only if they are set.
In the end, we can't push options as-is to the crosstool-NG config, but we
MUST provide sane bundled config files, which this patch does.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Closes#4093
AUTOTARGETS doesn't handle compressed patches, so the '400' patch did
not get applied.
Fix it by extracting the patch. With this, the '500' patch is no longer
needed.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
[Peter: only activate if barebox is enabled, fix deps]
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
These options should be in sync, both for consistency and because udev
needs devtmpfs enabled in the kernel.
We unfortunately cannot just select BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_UDEV
here, as that one selects udev - leading to recursion erors, so instead
we use depends on.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
This patch makes libxml-parser-perl behave on my system - though it looks
like it isn't necessary on all host OSses (on gentoo and opensuse it works
without, but on ubuntu lucid and Scientific 6 it fails without the patch).
Signed-off-by: Heiko Helmle <helman@gmx.net>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
When compiling Python on a host running Linux 3.0, the sys.platform
constant is set at "linux3". A lot of code (inside the interpreter
itself, its build system and third party apps and libraries) relies on
it to be linux2 on a Linux system.
This leads to the build of the target python package to break.
This behaviour has been reported on the Python bugtracker but is not
fixed yet.
http://bugs.python.org/issue12326
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
There's an even newer stable release available (1.4.14), but it
has problems with cross-compilation that we are still resolving
on the directfb-dev mailing-list.
[Peter: rename ppc patch to match]
Signed-off-by: Sven Neumann <s.neumann@raumfeld.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
util-linux uses mkinstalldirs to install .po files when locales are
enabled, but the definition of MKINSTALLDIRS comes from a tweaked
nls.m4 (based on the one from gettext).
When we autoreconf the package, we end up using the system (staging)
version of nls.m4, so MKINSTALLDIRS doesn't get defined.
Fix it by passing a definition of MKINSTALLDIRS to make during the
install steps.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
The configure script correctly detects presence of libintl, but it forgets
to link against it, breaking the build.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Update the crosstool-NG libc-specific config files to match the
new set of options coming with the updated crosstool-NG.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Buildroot should use a stable release, not use an unstable development
version. Updating to the current stable branch (2.34) is not trivial,
so at least use the latest release from the stable 2.32 branch.
Signed-off-by: Sven Neumann <s.neumann@raumfeld.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Closes#3583, #3649
Fixes xfonts_font-adobe build failure.
Signed-off-by: Michael J. Hammel <mjhammel@graphics-muse.org>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
As the kernel doesn't automatically mount devtmpfs when an initramfs
is used, commit 424888e474 has
introduced a small wrapper script that mounts devtmpfs before starting
the real init.
Unfortunately, the problem is that in this case, the init process runs
without any 0, 1 and 2 file descriptors, so none of the
messages/errors printed by the various initialization scripts can be
seen. This is due to the fact the init process relies on 0, 1 and 2
being opened by the kernel before init is started. However, as
/dev/console isn't present on the filesystem at the time the kernel
tries to open the console to create the 0, 1 and 2 file descriptors,
the kernel fails on this and prints the famous "Warning: unable to
open an initial console".
The proposed workaround is to actually open 0, 1 and 2 to /dev/console
in the wrapper script, right after mounting the devtmpfs filesystem,
and before starting the real init. The "Warning" from the kernel is
still shown, but at least the messages from the init scripts are
visible.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
993e51bc22 introduced a <pkg>_DL_VERSION
in addition to <pkg>_VERSION. This variable, purely internal to the
package infrastructure, allows to store the *raw* version name, as
defined in the package .mk file. However, as this version string can
contain slashes, it is not appropriate to be part of the build
directory name of the package. This is why <pkg>_VERSION is defined to
be the same thing as <pkg>_DL_VERSION, except that slashes are
replaced by underscores.
However, the initial implementation didn't take into account the case
of host packages. So 763822e874 was
committed to fix this. But unfortunately, this commit got it wrong
again: it forgot the case where <pkg>_VERSION is set, but not
HOST_<pkg>_VERSION. In this case, HOST_<pkg>_DL_VERSION remained
empty. This can be seen for example by building host-sstrip.
This problem has been noticed by Julien Boibessot
<julien.boibessot@free.fr>, who also helped in fixing the
issue. Thanks!
The fix is simply to ensure that <pkg>_DL_VERSION is properly set in
all cases: when HOST_<pkg>_VERSION is defined, when only <pkg>_VERSION
is defined, and when no version is defined.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
