Dexuan Cui [Tue, 28 Jun 2011 05:18:59 +0000 (13:18 +0800)]
base.bbclass: show layer's branches/revisions in the banner info
The patch removes METADATA_BRANCH and METADATA_REVISION, and treats the meta/
in the same way as other layers.
In the case some layers belonging to the same repo, the branch and revision
are only printed once, but all the layer names are still printed.
gettext: fix ERROR: QA Issue: gettext rdepends on gettext-dev
This was fixed in 2008 already, see http://dominion.thruhere.net/koen/cms/the-testlab-strikes-again but regressed during the 0.16.x -> 0.18.x update. Strangely enough the comment was kept.
Signed-off-by: Koen Kooi <koen@dominion.thruhere.net> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Fri, 17 Jun 2011 16:27:25 +0000 (17:27 +0100)]
gcc-package-cross: Switch to using pattern matching to detect when to stash libgcc into the sysroot
Currently, we stash libgcc if PN is gcc-cross or gcc-crosssdk. This patch
changes it to work for *gcc-cross and *-gcc-crosssdk which means it
will patch for multilib extended toolchains.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Mon, 4 Jul 2011 14:57:00 +0000 (15:57 +0100)]
gettext: Disable both git and cvs for autopoint's archive format.
The git-native dependency is between a 2-10% build performance hit
for a small reduction in autopoint's archive size (500kb vs 3.5MB).
We could always enable this for the target version only I guess.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
It was not setting COMPILE_IN_THUMB_MODE in .config
when ARM_INSTRUCTION_MODE = thumb. Reason was that
this entry has to exist in merged defconfig
so we add it disabled which means no harm to other
architectures
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Anders Darander [Fri, 1 Jul 2011 13:07:58 +0000 (15:07 +0200)]
image|kernel.bblass|module-init-tools: do not use depmod-2.6
Change to only depend on virtual/*/depmod.
Change all calls to only use depmod.
Do not install depmod as depmod-2.6
Bump PR in module-init-tools-cross.
Signed-off-by: Anders Darander <anders@chargestorm.se> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Fri, 1 Jul 2011 14:47:12 +0000 (15:47 +0100)]
insane.bbclass: Start to rework this so specific checks can be easily made warnings/errors
This patch means the warning/error handling can be controlled from local.conf
and/or from the distro level and no longer uses numbers but strings instead.
The system becomes extensible so that other classes can extend the path QA
checks at least.
It also removes all th duplicate error message code, we should have *one*
good error message.
Much work remains including making INSANE_SKIP take the classes of QA test
to skip but its a start.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Fri, 1 Jul 2011 09:06:02 +0000 (10:06 +0100)]
cairo: Clean up packaging and fix warnings
There were a number of QA errors being reported when packaging cairo. This patch
ensures the debug files go in the -dbg package, the .so development links go into
the -dev package, .a files into -staticdev.
It also switches back to relying on debian package naming for the library name.
Since this is working correctly now the files are in the correct packages, we
can simplify the recipe and there is no change in the resulting package names
from a package feed perspective.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fix these kinds of Package QA warnings before they are converted into fetal errors:
WARNING: QA Issue: package perl-module-compress contains bad RPATH /build_disk/poky_build/build0/tmp/sysroots/qemux86/usr/lib in file /build_disk/poky_build/build0/tmp/work/i586-poky-linux/perl-5.12.3-r1/packages-split/perl-module-compress/usr/lib/perl/5.12.3/auto/Compress/Raw/Zlib/Zlib.so
This fixes this warning for perl recipe as well as libxml-parser-perl recipe.
It is a fix to MakeMaker within perl, so all such perl recipes will get
fixed with this perl fix.
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Otherwise it ends up with this error
| configure: error: Package requirements (dbus-glib-1 >= 0.82
| gobject-2.0 >= 2.14
| gthread-2.0 >= 2.14
| ) were not met:
|
| No package 'gobject-2.0' found
| No package 'gthread-2.0' found
|
| Consider adjusting the PKG_CONFIG_PATH environment variable if you
| installed software in a non-standard prefix.
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Michael Lippautz [Thu, 30 Jun 2011 18:09:11 +0000 (20:09 +0200)]
linux-firmware: Add Realtex rt8192* firmwares
This patch adds the firmware files for:
* rt8192cu
* rt8192ce
* rt8192su
[RP: Changed file modes to 0644] Signed-off-by: Michael Lippautz <michael.lippautz@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Thu, 30 Jun 2011 08:07:15 +0000 (01:07 -0700)]
uclibc: Fix bug exposed by udev 168+ for mips architecture
newer udev uses sgnalfd + epoll and this exposed a bug
in uclibc for mips where SFD_NONBLOCK should be defined
with 0200 for mips but was using 04000. This would cause
random segfaults in udev during boot process
Tested on qemumips/angstrom console-image since angstrom
uses udev 171. It worked well when we did not use meta-oe
layer because udev in oe-core will then be used which is
at version 164.
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Hatle [Mon, 20 Jun 2011 15:57:49 +0000 (10:57 -0500)]
classes/package.bbclass: Add fixup_perms
Add a new function that is responsible for fixing directory and file
permissions, owners and groups during the packaging process. This will fix
various issues where two packages may create the same directory and end up
with different permissions, owner and/or group.
The issue being resolved is that if two packages conflict in their ownership
of a directory, the first installed into the rootfs sets the permissions.
This leads to a least potentially non-deterministic filesystems, at worst
security defects.
The user can specify their own settings via the configuration files
specified in FILESYSTEM_PERMS_TABLES. If this is not defined, it will
fall back to loading files/fs-perms.txt from BBPATH. The format of this
file is documented within the file.
By default all of the system directories, specified in bitbake.conf, will
be fixed to be 0755, root, root.
The fs-perms.txt contains a few default entries to correct documentation,
locale, headers and debug sources. It was discovered these are often
incorrect due to being directly copied from the build user environment.
The entries needed to match the base-files package have also been added.
Also tweak a couple of warnings to provide more diagnostic information.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Khem Raj [Tue, 28 Jun 2011 01:40:49 +0000 (18:40 -0700)]
uclibc: Remove redundant machine/arch configs
Big endian counterparts are not needed since
we already add the proper endianness to final
config as seen in uclibc-config.inc
The difference between arm and armv5te configs
were also superficial since all it did was select
arm926t which made sure that right mtune and march
flags were passed to gcc when building uclibc
We already do that via passing them in
UCLIBC_EXTRA_CFLAGS.
Similarly for other architectures the features
are really taken care of in config mangler and
machine config fragments are no longer needed
Scott Garman [Tue, 28 Jun 2011 22:15:39 +0000 (15:15 -0700)]
shadow-native: fix creation of home directories
Pseudo was recently changed so that when system() calls are
made after a chroot(), the host binaries can no longer be found,
breaking the system("mkdir -p") approach when useradd creates
home directories.
Instead, use mkdir(2) to create home directories with a helper
function to ensure parent directories get created.
Signed-off-by: Scott Garman <scott.a.garman@intel.com>
Wenzong Fan [Fri, 24 Jun 2011 07:42:14 +0000 (15:42 +0800)]
ccache: Set CCACHE on a per recipe basis
Set 'CCACHE_DIR' in 'bitbake.conf' and create the dirs for every
package before task 'do_configure' started.
[RP: Merge dirs variables into one] Signed-off-by: Wenzong Fan <wenzong.fan@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
do_configure and do_compile need a umask of 022 set because -many- recipes
directly copy generated files out of recipe's build directory. Instead of
fixing each existing and future recipe, it was shown to be much easier to
just set the umask.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
sanity.bbclass: pass the data object to the less frequent test harnesses
By passing the data object to the less frequently run test harnesses
(check_sanity_tmpdir_change(), check_sanity_sstate_dir_change() and
check_sanity_version_change()) we can run tests against BitBake data here
too.
Darren Hart [Wed, 29 Jun 2011 19:16:07 +0000 (12:16 -0700)]
kernel: move menuconfig task after configure
Fixes [YOCTO 1136]
linux-yocto adds some configure steps that are necessary to prepare the source
tree after the do_patch task. This causes a "-c menuconfig" to fail in a clean
build tree. Typical use of menuconfig should be to modify the config provided
by the recipe being built. It therefor makes sense for the menuconfig task to
come after the configure task. This also happens to fix the issue seen with the
linux-yocto kernel recipe.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Robert Yang [Tue, 28 Jun 2011 20:31:04 +0000 (14:31 -0600)]
Share gcc work directories
This patched is derived from Richard, make gcc use the shared source
directory during the different building:
1) Make gcc-cross, gcc-cross-initial, gcc-cross-intermediate and
gcc-runtime share the same source directory.
2) The source directory is ${TMPDIR}/work-shared/gcc-${PV}, for example:
tmp/work-shared/gcc-4.5.1
3) Fix do_clean to clean the shared source directory and stamps
4) gcc uses sed and creates config files against ${S} which means the
directory should not be shared. Change the way to make it work:
* The configure option --with-headers=${STAGING_DIR_TARGET}${SYSTEMHEADERS}
can replace the sed command, see the code in configure:
if test "x$with_headers" != x; then
glibc_header_dir=$with_headers
This has the same effect as the sed command:
sed -i 's:^\([ ]*\)glibc_header_dir=\"${with_build_sysroot}/usr/include\": ...
so add the --with-headers=${STAGING_DIR_TARGET}${SYSTEMHEADERS} to
gcc-configure-cross.inc( not add to gcc-configure-common.inc, since
not all the gcc building need this, the one which has its own do_configure
doesn't need it).
* Move t-oe from ${T} to ${B}/gcc, so that the patched Makefile.in
can read it easily, please see the commit for gcc-4.5.1 and
gcc-4.6.0.
* Use the defaults.h in ${B}/gcc instead of ${S}/gcc, and the patched
configure.ac(configure) can read it correctly, please see the
commit for gcc-4.5.1 and gcc-4.6.0.
* The gcc-crosssdk.inc used sed to edit ${S}/config/*/linux*.h
to change the GLIBC_DYNAMIC_LINKER, which made the source
incompatible. To make the source compatible:
- Use:
sed -i ${S}/gcc/config/*/linux*.h -e \
's#\(GLIBC_DYNAMIC_LINKER[^ ]*\)\( *"/lib.*\)#\1 SYSTEMLIBS_DIR\2#'
so entries in the files that look like:
#define GLIBC_DYNAMIC_LINKER64 "/lib64/ld-linux-x86-64.so.2"
would become
#define GLIBC_DYNAMIC_LINKER64 SYSTEMLIBS_DIR"/ld-linux-x86-64.so.2"
and we define SYSTEMLIBS_DIR in defaults.h.
NOTE:
#define GLIBC_DYNAMIC_LINKER64 (SYSTEMLIBS_DIR "/ld-linux-x86-64.so.2")
doesn't work in in the following define:
#define LINUX_DYNAMIC_LINKER \
CHOOSE_DYNAMIC_LINKER (GLIBC_DYNAMIC_LINKER, UCLIBC_DYNAMIC_LINKER)
so use
#define GLIBC_DYNAMIC_LINKER64 SYSTEMLIBS_DIR"/ld-linux-x86-64.so.2"
5) Add do_configure_prepend to gcc-configure-common.inc and remove the
one in gcc-crosssdk.inc, this makes it easy to share the source,
otherwise we need do extra changes in gcc-configure-sdk.inc.
6) Use "cat > file <_EOF" to replace the "echo > file"
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Commit 477ede7472db0bacd5daacb96e97f849d1be84ee accidentally reverted
some code changes it shouldn't have done leading to continued eglibc
packaging issues. This patch corrects that damage.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
