Adrian Alonso [Thu, 5 May 2011 16:29:33 +0000 (11:29 -0500)]
siteinfo: add microblaze target info
* Add microblaze target info
* Microblaze soft CPU can be configured as big-endian/little-endian
* Currently target info support for microblaze big endian, using prebuilt
toolchains.
Signed-off-by: Adrian Alonso <aalonso@secretlab.ca>
Khem Raj [Fri, 6 May 2011 05:24:40 +0000 (22:24 -0700)]
liburcu_0.5.4.bb: Let it build on uclibc targets
uclibc requires -D_GNU_SOURCE to be defined for it
to enable GNU extensions which CPU_SET/CPU_ZERO are
used by this package. So we add -D_GNU_SOURCE to
CFLAGS when compiling for uclibc.
Some distributions e.g. angstrom do not support gconf-dbus
anymore since its now provided by gconf and in oe-core we
have PREFERRED_PROVIDER_gconf tunable, here we replace all
dependencies on gconf-dbus with gconf, which should work
in both cases where poky defines
PREFERRED_PROVIDER-gconf = "gconf-dbus" and for angstrom
it is simply gconf
Without this meta-toolchain-sdk ends up with conflicts in
runqueue and does not build
Khem Raj [Thu, 5 May 2011 16:57:56 +0000 (09:57 -0700)]
gettext.bbclass, bitbake.conf: Append nls options to EXTRA_OECONF instead of +=
Some recipes do not defined EXTRA_OECONF in such cases += drops
the --enable|--disable-nls options. In another case where recipe
defines EXTRA_OECONF instead of adding/appending to it then
--enable|--disable-nls options are lost from EXTRA_OECONF
We define EXTRA_OECONF = "" in bitbake.conf so the variable exists
always.
We use _append instead of += so the option is added at very end
and not lost.
We only return empty gettext dependencies if its a target recipe
in case when USE_NLS is not set because the native/cross/nativesdk recipes still
need the gettext dependencies
Tom Rini [Wed, 4 May 2011 19:28:25 +0000 (12:28 -0700)]
binutils.inc: Switch to TARGET_PREFIX not TARGET_SYS
When TARGET_PREFIX and TARGET_SYS didn't match up, the symlinks and
update-alternatives weren't working. TARGET_PREFIX is what we use when
configuring so it's what we should be using here.
Tom Rini [Wed, 4 May 2011 19:26:12 +0000 (12:26 -0700)]
gcc-package-target: Switch to TARGET_PREFIX in symlinks
When TARGET_PREFIX and TARGET_SYS didn't match up, the symlinks we made
were invalid. TARGET_PREFIX is what we use when configuring so it's what we
should be using here.
Packaging is synced with OE .dev, but without the -large, thanks to pseudo we retain all (hard)links properly:
* git-perltools holds all the utils requiring perl and adds appropriate RDEPENDS
* git-tk holds the tcl/tk utils once they get activate
* git now packages /usr/libexec/git-core
GCC 4.6.0 in Fedora rawhide turned up some compile errors in tools/perf
due to the -Werror=unused-but-set-variable flag.
I've gone through and annotated some of the assignments that had side
effects (ie: return value from a function) with the __used annotation,
and in some cases, just removed unused code.
In a few cases, we were assigning something useful, but not using it in
later parts of the function.
kyle@dreadnought:~/src% gcc --version
gcc (GCC) 4.6.0 20110122 (Red Hat 4.6.0-0.3)
Cc: Ingo Molnar <mingo@redhat.com>
LKML-Reference: <20110124161304.GK27353@bombadil.infradead.org> Signed-off-by: Kyle McMartin <kyle@redhat.com>
[ committer note: Fixed up the annotation fixes, as that code moved recently ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
[Backported to 2.6.38.2 by deleting unused but set variables] Signed-off-by: Thomas Meyer <thomas@m3y3r.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
[Backported to linux-yocto kernel git version] Signed-off-by: Khem Raj <raj.khem@gmail.com>
]
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Bruce Ashfield [Fri, 8 Apr 2011 04:01:35 +0000 (00:01 -0400)]
linux-yocto: safely process unbranched repositories
The BSP bootstrap and -dev use cases can be applied against
unbranched or repos without meta data. To allow the proper
and safe processing of those repositories, slight modifications
to the tools are required to pass the branch on the command
line (rather than detecting it always) and to only checkout
branches that exist.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Bruce Ashfield [Mon, 4 Apr 2011 03:50:19 +0000 (23:50 -0400)]
linux-yocto: apply meta data to external repos
To support quick uprev and testing, it is desireable to build
repositories that do not have embedded meta data. In this scenario
the meta data can be automatically created or provided externally.
This commit supports the first situation by detecting the lack
of meta data and then automatically creating a base set of meta
files.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
This is set of bugfixes that has been done on
FSF gcc-4_2-branch since 4.6.0 was released
They will roll into 4.6.1 release once that
happens in coming approx 6 months time then
we can simply remove them thats the reason
so use a separate .inc file to define the
SRC_URI additions
Paul Eggleton [Wed, 27 Apr 2011 16:34:17 +0000 (17:34 +0100)]
initscripts: remove -i from halt/reboot arguments and allow override
Introduces a variable HALTARGS which specifies the arguments sent to
halt and reboot, and sets the default value to "-d -f", dropping the
previous -i (shut down all network interfaces before halt/reboot, which
causes a freeze with NFS root.)
Fixes [YOCTO #997].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
This patch fixes compilation/linking of python third-party extensions, i.e.
Extensions that ship with C code.
Problem:
Python uses distutils(-native) to compile third-party extensions. distutils
uses its own sysconfig module to get the options for compiling and linking.
Since third-party extensions have to be linked against this libpython it
important that -L points into staging. This is not the case because
distutils.sysconfig uses a special Makefile that is shipped with python
determine the paths. The Makefile is the same that would be used on the
target to build third-party extensions. It therefore points into /usr/lib
instead of staging.
Solution:
Stage a modified version of the Makefile where the paths (incdir, libdir) have
been replaced by ones that point into staging.
Side-problem:
The recipe actually should not stage files itself in do_compile, but rather
handle everything that needs to be staged in do_install. This is currently not
possible because python compiles itself using distutils-native. Distutils on
the other hand does only allow to add a path, but not to substitute it,
requiring a staged Makefile and libpython.so before the actual python
compilation is triggered.
The second step to solve this would be to either patch distutils, or split
python into python-initial and python. The -initial part could create the
Makefile and the library, while the main part focuses on the target.
For further references see:
http://lists.linuxtogo.org/pipermail/openembedded-core/2011-May/001752.html
Signed-off-by: Michael Lippautz <michael.lippautz@gmail.com> Acked-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jiajun Xu [Wed, 4 May 2011 01:25:49 +0000 (09:25 +0800)]
qemuimagetest: Add executable permission to test scripts for toolchain
The file mode bits of toolchain test scripts is 644, which could not be
executed by user. Fix the issue by adding executable permission(755) to
all test scripts for toolchain test.
Signed-off-by Jiajun Xu <jiajun.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jiajun Xu [Tue, 3 May 2011 02:07:30 +0000 (10:07 +0800)]
qemuimagetest: Enable toolchain automation tests in qemuimagetest
Enable toolchain automation tests in qemuimagetest framework. 3 C/C++ test
projects are added to test toolchain - cvs, iptables and sudoku-savant. User
needs to set TEST_SCEN to "toolchain" in local.conf to enable tests. Test case
will check if toolchain tarball exists under "${DEPLOY_DIR}/sdk". And it will
extract toolchain tarball into /opt. It requires user to chown /opt to non-root
user, who will run qemuimagetest.
Khem Raj [Fri, 18 Mar 2011 05:40:15 +0000 (22:40 -0700)]
perl-native_5.12.2.bb: Fix compliation on ubuntu 11.04-alpha
Ubuntu has moved eglibc to /usr/lib/${arch}-linux-gnu and
/lib/${arch}-linux-gnu so we need that to be added to glibpth in
Configure.
Currently we set LD=ld in environment for recipes inheriting native
class. This overrides the LD settings in the Makefiles of perl and
it tries to link by calling ld which does not work since its using
-l<x> on commandline and ubuntu linker seems not to look into
the new location for these libraries. Its better to use gcc for linking
here anyway
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Further tweaks to Configure-multilib.patch for x86_64 vs x86
from Gary Thomas
Saul Wold [Tue, 3 May 2011 17:13:35 +0000 (10:13 -0700)]
linuxdoc-tools-native: add groff-native to DEPENDS
groff-native is needed to ensure that configure finds
the groff-native binary instead of the host's groff,
this is to ensure the correct macros are used (-ms vs -mgs)
Martin Jansa [Fri, 29 Apr 2011 10:13:46 +0000 (12:13 +0200)]
git: use DESTDIR=$D instead prefixing all variables by $D
* with git-native and rm_work enabled I've noticed git fetcher errors like:
warning: templates not found /OE/shr-core/tmp/work/x86_64-linux/git-native-1.7.3.4-r0/image/OE/shr-core/tmp/sysroots/x86_64-linux/usr/share/git-core/templates
fatal: Unable to find remote helper for 'http'
for every recipe using http:// for git repo
* after this change template_dir points to
/OE/shr-core/tmp/sysroots/x86_64-linux/usr/share/git-core/templates
without that workdir prefix
* haven't tested target recipe, but I guess it needs different fix or
maybe it worked before and gets broken by this change
[sgw: removed RFC comment, target patch to follow] Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Koen Kooi <koen@dominion.thruhere.net> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Paul Eggleton [Fri, 29 Apr 2011 09:33:32 +0000 (10:33 +0100)]
netbase: fix check for nfsroot in if-pre-up.d script
The if-pre-up.d/nfsroot script was no longer functioning correctly due to
a change in iproute2's installation of the "ip" executable. The script will
now function correctly regardless of whether ip is in /bin or /sbin.
Fixes [YOCTO #1006].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Darren Hart [Fri, 29 Apr 2011 04:52:50 +0000 (21:52 -0700)]
logging: delete the oe(note|warn|fatal|debug) functions
The new logging.bbclass replaces the oe* logging functions with bb* equivalents.
There are no longer any users of the oe* API within oe-core. Remove the oe*
functions.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Darren Hart [Fri, 29 Apr 2011 04:49:31 +0000 (21:49 -0700)]
logging: update existing oe* logging users to the bb* interface
The new bash logging class provides bbnote, bbwarn, bbfatal, and bbdebug
replacements (as well as bbplain and bberror) for the oe* equivalents. Use the
new bb* API in preparation to delete the oe* logging API.
This patch was automatically generated by a sed script. The result has been
visually inspected and used to build core-image-sato for qemux86.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Mei Lei [Thu, 28 Apr 2011 14:14:17 +0000 (22:14 +0800)]
distro_check.py: Fix some bugs in distro_check.py
The recipe meta-ide-support's alias is Poky, but not defined in distro_exceptions, this will lead an error.
The compare_in_distro_packages_list function omit a split process, this will lead to a form issue.
Change the log file generate way, use symbolic links link to the latest log file, instead of appending every log file in old log.
Gary Thomas [Thu, 28 Apr 2011 17:05:26 +0000 (11:05 -0600)]
Fix matchbox-panel-2 compile error
This patch is needed when building with GCC-4.6.0
--
------------------------------------------------------------
Gary Thomas | Consulting for the
MLB Associates | Embedded world
------------------------------------------------------------
>From 7337ecd4970f7d94080ad70b912bdb2742d2d7aa Mon Sep 17 00:00:00 2001
From: Gary Thomas <gary@mlbassoc.com>
Date: Thu, 28 Apr 2011 11:03:44 -0600
Subject: [PATCH] matchbox-panel: fix compile error with GCC-4.6.0
Upstream-Status: Pending
[sgw: added Upstream-Status] Signed-off-by: Gary Thomas <gary@mlbassoc.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Kang Kai [Wed, 27 Apr 2011 08:16:46 +0000 (16:16 +0800)]
lsbsetup: add commands install_initd and remove_initd
Add commands install_initd and remove_initd link to chkconfig, which
can handler LSB format init.d scripts. Needed by LSB commands check.
Bump PR to r2
Part fixes [Yocto 520]
