Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com> Reviewed-by: Samuel Martin <s.martin49@gmail.com> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Luca Ceresoli [Thu, 2 Oct 2014 10:11:01 +0000 (12:11 +0200)]
ffmpeg: fix license file
Somewhere between 1.2.7 and 2.3.1 (version bump in bf76f4319131a107)
the LICENSE file has been renamed to LICENSE.md, so `make legal-info`
is currently broken. Update to the current file name.
Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The gnupg2 configure script checks whether <stdint.h> complies with
C99. When doing this, it expects a number of WCHAR_* definitions to be
present, which is not the case on non-wchar capable toolchains. The
gnupg2 configure script then concludes that <stdint.h> is not
C99-compliant and generates its own, which causes some build failures
related to intmax_t being not defined.
Since wchar is not actually used in gnupg2, this commit fixes this
problem by forcing gnupg2 to think that our <stdint.h> is
C99-compliant.
Luca Ceresoli [Wed, 1 Oct 2014 15:03:45 +0000 (17:03 +0200)]
freescale/imx6-sabresd: lock down kernel headers to the kernel version
The current defconfigs for iMX6 SABRESD kernel 3.10.17 from the Freescale
git repository. This kernel is older than the toolchain kernel headers
used in Buildroot (currently 3.16.3).
Explicitly lock down the kernel headers to match the version of the kernel
that is built.
Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net> Tested-by: Vincent Stehlé <vincent.stehle@freescale.com> Cc: Gilles Talis <gilles.talis@gmail.com> Cc: Gary Bisson <bisson.gary@gmail.com> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The stable-1.1 branch of freerdp fails to build when using a uClibc
toolchain because it's using functions which are not implemented in
uClibc, like eventfd_read, eventfd_write and futimes. That is causing
build failures like these ones:
../../libwinpr/synch/libwinpr-synch.so.0.1.0: undefined reference to
`eventfd_read'
../../libwinpr/synch/libwinpr-synch.so.0.1.0: undefined reference to
`eventfd_write'
../../common/libfreerdp-client.so.1.1.0: undefined reference to
`futimes'
Backport upstream patch, with non-applicable parts (new files from
master, but not in the stable-1.1 branch) trimmed down.
Mike Williams [Wed, 1 Oct 2014 15:15:49 +0000 (11:15 -0400)]
valgrind: allow building on more ARM systems
Signed-off-by: Mike Williams <mike@mikebwilliams.com> Reviewed-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Fixes CVE-2014-3634 - potential abort when a message with PRI > 191 was
processed if the "pri-text" property was used in active templates, this
could be abused to a remote denial of service from permitted senders.
Fixes:
CVE-2014-6051 and
CVE-2014-6052 denial of service and possible code execution via
integer overflow and lack of malloc error handling in
MallocFrameBuffer()
CVE-2014-6053 denial of service via large ClientCutText message.
CVE-2014-6054 denial of service via zero scaling factor.
CVE-2014-6055 denial of service and possible code execution via
stack overflows in File Transfer feature.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Currently, the MESSAGE macro double-quotes the message. But a few call
sites also double-quote the message, resulting in a borked call to echo,
like:
$(call MESSAGE,"user text")
would expand to:
echo "Leading text "user message" trailign text"
Just use qstrip on the user message, so we get protected from either a
comma in the user text (so make does not interpret it), or a semi-colon
(so the shell does not interpret it.)
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr> Cc: Peter Korsgaard <jacmet@uclibc.org> Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
docs/manual: add musl to toolchain dependency options
Currently, we only document (e)glibc and uClibc as C libraries on which
we can depend. However, we've supported musl for quite some time now.
So, also refference it in the manual, when dealing with toolchain
options a package may have to depend on.
[Peter: order musl before uClibc] Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- Bump version to 1.5.9
- Remove patches already included in this version
- Use make parameters instead of environment variables
- Add an upstream patch to fix building statically
- Adapt old patches to the new version
- Add a hash file
The Makefile of dosfstools passes -D_GNU_SOURCE to get access to
non-standard definitions such as 'loff_t'. But this flag gets dropped
by the CFLAGS override, so explicitly add it.
[Peter: reword commit message] Signed-off-by: Maarten ter Huurne <maarten@treewalker.org> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
When libcurl-0001-build-link-curl-to-NSS-libraries-when-NSS-support.patch
was removed the corresponding autoreconf was left behind:
http://git.buildroot.net/buildroot/commit/?id=9185b64ed5599622cb89ca4ee6ee29440b02ec8a
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
package/mtd: install libubi.a and libmtd.a to staging
Currently, the mtd package only installs the mtd utilities into the
target directory. However, the mtd-utils also provide two
static libraries (libmtd.a and libubi.a) that can be used by other
packages or tools.
Therefore, this commit extends the mtd package by adding the
installation to the staging directory of those two libraries.
Signed-off-by: Romain Naour <romain.naour@openwide.fr> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The VideoCore file server daemon startup script installed from this package is
not compatible with BuildRoot (because of its naming and other Debian
dependencies), which prevented vcfiled from starting. Hence, prevent this
package from installing its vcfiled startup script, and add a configuration
option to install a vcfiled SysV init script suitable for BuildRoot.
[Peter: tweak help text as suggested by Yann] Signed-off-by: Benoît Thébaudeau <benoit.thebaudeau@advansee.com> Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr> Reviewed-by: Samuel Martin <s.martin49@gmail.com> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Samuel Martin [Thu, 21 Aug 2014 20:45:20 +0000 (22:45 +0200)]
toolchainfile.cmake.in: do not force the CMAKE_{C, CXX}_FLAGS to the cache
Fix #7280 [1]
When the FORCE option is passed to the set command, the variable is
added/updated in the CMake cache every single time CMake processes this
command.
Because the toolchainfile.cmake prepends architecture/toolchain flags
to the CMAKE_{C,CXX}_FLAGS, this makes the CFLAGS being updated in the
generated Makefiles each time one reconfigures its project. So it
forces the compilation of everything, even when nothing has changed.
[1] https://bugs.busybox.net/show_bug.cgi?id=7280
Signed-off-by: Samuel Martin <s.martin49@gmail.com> Tested-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
(tested the SimpleApp reproduction scenario described in the bug report) Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Samuel Martin [Thu, 21 Aug 2014 20:45:18 +0000 (22:45 +0200)]
toolchainfile.cmake.in: set linker flags
The linker flags are part of the toolchain configuration, so set them for
the CMake-based packages.
Signed-off-by: Samuel Martin <s.martin49@gmail.com> Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Samuel Martin [Thu, 21 Aug 2014 20:45:16 +0000 (22:45 +0200)]
package/pkg-cmake.mk: remove the USE_CCACHE CMake flag for host-package
The USE_CCACHE CMake flag is only useful when building target package
(only the toolchainfile.cmake uses it).
Signed-off-by: Samuel Martin <s.martin49@gmail.com> Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
check-kernel-headers: mktemp --tmpdir not available on RedHat RHEL5
mktemp --tmpdir is not available on older Redhat RHEL5 machines. The
alternative that has the same behavior is 'mktemp -t'.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com> Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
jamvm: bump to 2.0.0 (including mips big-endian support)
This commit bumps jamvm to version 2.0.0. From the release notes:
I'm pleased to announce a new release of JamVM. JamVM 2.0.0 is the
first release of JamVM with support for OpenJDK (in addition to GNU
Classpath). Although IcedTea already includes JamVM with OpenJDK
support, this has been based on periodic snapshots of the development
tree.
JamVM 2.0.0 supports OpenJDK 6, 7 and 8 (the latest). With OpenJDK 7 and
8 this includes full support for JSR 292 (invokedynamic). JamVM 2.0.0
with OpenJDK 8 also includes full support for Lambda expressions (JSR
335), type annotations (JSR 308) and method parameter reflection.
In addition to OpenJDK support, JamVM 2.0.0 also includes many
bug-fixes, performance improvements and improved compatibility (from
running the OpenJDK jtreg tests).
As this release now also supports MIPS big-endian targets, this commit
fixes bug #7010 (https://bugs.busybox.net/show_bug.cgi?id=7010)
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Not all cdrkit variables are prefixed with CDRKIT_, so they leak.
Most importantly, they look like they belong to another package's
namespace, cmake, as they start with CMAKE_ (but fortunately, they
have no impact on cmake, as they are not used by our infras.)
Fix that by removing the intermediate variables, and directly set
the CDRKIT_CONF_OPT variable.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr> Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
OpenCV's highgui module needs to be auto-selected when the ts module is
selected, because it depends on it. Otherwise it would fail with an
error like this one:
[ 96%] Building CXX object
modules/ts/CMakeFiles/opencv_ts.dir/src/gpu_perf.cpp.o
In file included from
/home/test/autobuild/instance-1/output/build/opencv-2.4.8/modules/ts/src/gpu_perf.cpp:43:0:
/home/test/autobuild/instance-1/output/build/opencv-2.4.8/modules/ts/include/opencv2/ts/gpu_perf.hpp:47:39:
fatal error: opencv2/highgui/highgui.hpp: No such file or directory
compilation terminated.
The gnuplot 4.6.6 tarball was updated because of this:
(based on a diff from the original release and the updated one)
http://thread.gmane.org/gmane.comp.graphics.gnuplot.user/7856
Some packages use -lcurses when they are linking, so they try to link to
a library called libcurses. The library provided by our ncurses package
is called libncurses, so those packages fail to link with a message like
this one:
/bin/ld: cannot find -lcurses
Installing a libcurses symlink to libncurses fixes the problem.
projects / coffee / buildroot.git / log