- Feb 25, 2014
-
-
Peter Korsgaard authored
Signed-off-by:Peter Korsgaard <peter@korsgaard.com>
-
- Feb 24, 2014
-
-
Peter Seiderer authored
Fixes runtime failure of gst-fsl-plugins, e.g. (gst-plugin-scanner:1700): GStreamer-WARNING **: Failed to load plugin '/usr/lib/gstreamer-0.10/libmfw_vpu.so': /usr/lib/libgstfsl-0.10.so.0: undefined symbol: shm_open) when compiled with buildroot/glibc toolchain. See also old buildroot mailing list request [1]. [1] http://lists.busybox.net/pipermail/buildroot/2013-June/073910.html Signed-off-by:Peter Seiderer <ps.report@gmx.net> Signed-off-by:
-
Thomas Petazzoni authored
Fixes: http://autobuild.buildroot.org/results/5e2/5e2c627cfcb2699b16a82eb9554f2c9dfabd5483/ Signed-off-by:
Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by:
-
Thomas Petazzoni authored
Fixes: http://autobuild.buildroot.org/results/d46/d46f9097780bdde68ec15f693281ee0b04667f98/ Signed-off-by:
-
Peter Korsgaard authored
Signed-off-by: -
Yann E. MORIN authored
Rename the GRAPH_OUT and GRAPH_ALT variables according to our recently-agreed naming scheme for user-facing variables: - GRAPH_OUT -> BR2_GRAPH_OUT - GRAPH_ALT -> BR2_GRAPH_ALT The documentation part of the rename is handled by Thomas as part of his manual fixing spree. ;-) Signed-off-by:
"Yann E. MORIN" <yann.morin.1998@free.fr> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by:
-
Yann E. MORIN authored
There was one instance of DL_DIR that was left out. Fix that. Signed-off-by:
-
Thomas De Schampheleire authored
This commit fixes a few minor spelling or grammar mistakes since the recent additions to the manual (commits 0b100de2 to 7cbb4766). Signed-off-by:
Thomas De Schampheleire <thomas.de.schampheleire@gmail.com> Signed-off-by:
-
Arnout Vandecappelle authored
package/libtool/libtool-0001-mips64-n64-linking.patch touches an m4 file, so applying it triggers a run of autoconf, automake, etc. This sometimes leads to build failures due to incompatible system autotools. This was observed on Fedora 10, which has automake 1.11 while the generated files that are included in the tarball require automake 1.11.1. We cannot simply set HOST_LIBTOOL_AUTORECONF = YES because that would create a circular dependency on host-libtool. Therefore, the patch patch is applied on the installed libtool.m4 instead of in the source tree. Then, it also needs to be renamed to something that doesn't match buildroot's patch convention so it doesn't get picked up automatically. Signed-off-by:
Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be> Signed-off-by:
-
Thomas Petazzoni authored
libsecret does built on AVR32 due an internal compiler error, and since it's unlikely that gcc will ever be upgraded on AVR32, we simply mark this package as not available. The only reverse dependency of libsecret is webkit, which already cannot be selected on AVR32. Fixes: http://autobuild.buildroot.org/results/2d4/2d41f97378f06d0f3922e078ec0d02549690d4c1/ Signed-off-by:
-
Thomas Petazzoni authored
The BR2_PACKAGE_SCONESERVER_HTTP_SCONESITE_IMAGE option of sconeserver selects imagemagick, but forgets to depends on BR2_USE_MMU, which imagemagick depends on. Signed-off-by:
-
Thomas Petazzoni authored
Signed-off-by:
-
Thomas Petazzoni authored
A number of packages that select Pango (which depends on BR2_USE_MMU) forget to also depends on BR2_USE_MMU. This commit fixes that. Signed-off-by:
-
Thomas Petazzoni authored
libgtk2 depends on !MMU, so when OpenCV selects Gtk, it should propagate this dependency. Signed-off-by:
-
Thomas Petazzoni authored
The libblkid library in util-linux uses fork. While it seems potentially possible to disable the parts of the library that use fork (the only parts using fork are parts related to LVM and Device Mapper, which are said to be legacy as they are replaced by sysfs based scanning, instead of having to fork to run a separate process). However, since libblkid is used by: - e2fsprogs, which uses lots of fork all over the place - udev, which most likely will also need fork There is not much point in making libblkid fork-less. As a consequence, this commit makes util-linux/libblkid unavailable on non-MMU architectures, and propagates the relevant dependency to the reverse dependencies of libblkid. This fixes the e2fsprogs build failure seen on Blackfin. The failure was due to a configure test of e2fsprogs which was trying to link a small test program against libblkid, which failed because there was an undefined reference to fork() in the libblkid code. Fixes: http://autobuild.buildroot.net/results/ee2c1568d16ac040011dd4d6d8b543ff9e9e2622/ Signed-off-by:
-
Thomas Petazzoni authored
Patch cairo-001-blackfin-build-fix.patch was added in commit d372d0b6 ('cairo: add patch to fix Blackfin build failure') to take into account the fact that Blackfin has a non-empty USER_LABEL_PREFIX: C symbol 'foo' corresponds to assembly symbol '_foo', contrary to most other architectures where they are the same. However, the patch was only fixing the relevant macros in src/cairo-compiler-private.h and did not take into account the duplicated version of these macros in util/cairo-script/cairo-script-private.h. This commit fixes that. This commit is fixing a build failure on Blackfin, for which there was no recorded autobuilder result because this problem was hidden by the problem fixed by the previous commit ('cairo: fix build on Blackfin in test/cairo-test-runner.c'). Signed-off-by:
-
Thomas Petazzoni authored
Add a patch for Cairo that ensures <unistd.h> is properly included to get the prototype of readlink and getppid even on non-MMU platforms. Fixes: http://autobuild.buildroot.net/results/93a97f90345d193b52035b2dc7559a306e11098e/ Signed-off-by:
-
- Feb 23, 2014
-
-
Arnout Vandecappelle authored
Fixes http://autobuild.buildroot.org/results/1ea/1ea98985ce06dc1b7569ef5abe2fc13090fb5f3a The upstream patch patches both configure.ac and configure, so we keep that. It is numbered 001 because it is an upstream patch, so it should come before the buildroot patches. Signed-off-by:
-
Thomas Petazzoni authored
Signed-off-by:
-
Thomas Petazzoni authored
[Peter: adjust wording as suggested by Yann] Signed-off-by:
-
Thomas Petazzoni authored
Signed-off-by:
-
Thomas Petazzoni authored
The ELDK now relies on Yocto, so it no longer has the funky non-standard way of doing multilib. Also, we didn't mention that Yocto toolchains/SDK are not suitable for usage with Buildroot. Therefore, this commit rewords this part of the documentation to explain that Yocto/OpenEmbedded SDKs cannot be used as external toolchains in Buildroot, and why. Signed-off-by:
-
Thomas Petazzoni authored
Signed-off-by:
-
Thomas Petazzoni authored
Replace ADI by Analog Devices, which is clearer for most people, and remove the Xilinx external toolchains since we have deprecated them. Signed-off-by:
-
Thomas Petazzoni authored
Now that we have eglibc and glibc support in the internal backend, and no longer marked as experimental, a little bit of rewording is needed. It is no longer necessary to indicate that uClibc was historically supported as the only C library, and that the glibc support is experimental. We also update the rest of the description to be less uClibc specific. Signed-off-by:
-
Thomas Petazzoni authored
Signed-off-by:
-
Thomas Petazzoni authored
Signed-off-by:
-
Thomas Petazzoni authored
[Peter: adjust wording as suggested by Yann] Signed-off-by:
-
Thomas Petazzoni authored
Signed-off-by:
-
Thomas Petazzoni authored
There is no need to tell people that they should remove stamp files: they should use the make <pkg>-reconfigure and make <pkg>-rebuild make targets instead. We still keep an explanation about stamp files, just to give the user an insight on how Buildroot works internally. Signed-off-by:
-
Thomas Petazzoni authored
Signed-off-by:
-
Peter Korsgaard authored
Signed-off-by: -
Thomas De Schampheleire authored
Fixes http://autobuild.buildroot.net/results/4f7/4f744595efceba2d5a54151bf5f1802eb1268fac/ Signed-off-by:
-
Peter Korsgaard authored
The conditional (suggested by me) introduced in 108952 (rt-tests: disable for uclibc mips) isn't actually valid kconfig syntax, causing menuconfig to error out. Rewrite it to use proper syntax. Signed-off-by: -
Ryan Barnett authored
rt-tests depends on NPTL which is not available on the bfin architecture. Fixes: http://autobuild.buildroot.net/results/ac9/ac95f73b2351d46d9a8c48caae7f6c3ea124e8bc Signed-off-by:
Ryan Barnett <rjbarnet@rockwellcollins.com> Signed-off-by:
-
Ryan Barnett authored
rt-tests is not supported by a uclibc toolchain that does not implement _pid in struct sigevent. Currently this is all MIPS architectures in uclibc. Fixes: http://autobuild.buildroot.net/results/074/074265602bec4aba6c82d1aee389045e8ad33d4b Signed-off-by:
-
Yann E. MORIN authored
For some architectures (eg. Arc, Cris, Hexagon, ia64, Parisc, Score and Xtensa), the Linux buildsystem tries to call the cross-compiler when installing the headers. This is a spurious call, since a cross-compiler is not needed at all to install the headers. As some users have reported the issue, just add a comment in linux-headers.mk directing the user to ignore those errors. Reported-by:
Noam Camus <noamc@ezchip.com> Signed-off-by:
-
Thomas De Schampheleire authored
webkit does not build correctly on powerpc. There is an upstream bug report with a partial fix, but with remaining problems. https://bugs.webkit.org/show_bug.cgi?id=113638 Until this problem is fixed, we disable webkit on powerpc. Fixes http://autobuild.buildroot.org/results/3ce/3ce992a663ddf1030a987bb778f8275a8e5fdef0/ Signed-off-by:
-
Gustavo Zacarias authored
Signed-off-by:
Gustavo Zacarias <gustavo@zacarias.com.ar> Signed-off-by:
-
Gustavo Zacarias authored
Signed-off-by:
-
