Richard Purdie [Wed, 14 Nov 2012 14:31:24 +0000 (14:31 +0000)]
python: Resolve intermediate staging issues
Its bad practise to poke into the sysroot without knowledge of sstate.
This adds a patch to python allowing us to account for cross compiling
and allow it to find the Makefile/pyconfig.h files without needing them
in the sysroot for do_compile/do_install to complete.
Tested on two architectures and compared with buildhistory with no
significant delta.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Hatle [Wed, 31 Oct 2012 19:43:48 +0000 (14:43 -0500)]
python-smartpm: Add basic knowledge of RPMSENSE_MISSINGOK
Currently smart does not support recommend dependencies. Add the first set
of 'support' for RPMSENSE_MISSINGOK (the flag that makes something a
recommend). This initial support ends up ignoring the recommendation, but is
written in a way that it will be the basis of eventual support.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Mark Hatle [Thu, 4 Oct 2012 18:57:00 +0000 (13:57 -0500)]
python-smartpm: Add smartpm recipe
This is the initial integration, basic functionality such as 'smart query'
has been tested. Active use of remote feeds and such has not yet been
verified.
Thanks to Paul Eggleton for corrections and bug fixes for the initial
integration.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Richard Purdie [Wed, 14 Nov 2012 23:32:56 +0000 (23:32 +0000)]
sstate: Bump version number to deal with layout fixes
The recent preveeding sstate directory layout fixes made the code do what it
was originally intended to do, as can be clearly seen from the code.
Unfortunately this changed the contents and layout of the sstate files
themselves since the bug was leading to a directory prefix being missing.
This is now resulting in chaotic messages on the console since things
are getting confused with the two different layouts. The simplest way to
resolve this is to bump the version number, hence moving the new layout
into its own new namespace.
Its worth noting that whilst the failure messages are scary, the failure
mode is relatively harmless since it will just fall back to building the
data rather than installing from sstate.
Usually I'd give more notice of a change like this but under the
circumstances, I'm just going to push this in to resolve the failures
people are seeing. Initially I thought the problem was limited to
some of the -cross packages and therefore of low impact but that is
clearly not the case.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
On hosts with FORTIFY_SOURCES, stringize support is required, as it's used by
the macros to wrap functions (e.g. read and open in unistd.h). Those wrappers
use the STRING() macro from unistd.h. A header in the bash sources overrides
the unistd.h macro to 'x' when HAVE_STRINGIZE is not defined, causing the
wrappers to generate calls to 'xread' and 'xopen', which do not exist,
resulting in a failure to link.
Assume we have stringize support when cross-compiling, which works around the
issue.
It may be best for upstream to either give up on supporting compilers without
stringize support, or to not define STRING() at all when FORTIFY_SOURCES is
defined, letting the unistd.h one be used, instead.
Signed-off-by: Christopher Larson <chris_larson@mentor.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 14 Nov 2012 15:52:45 +0000 (15:52 +0000)]
sstate: Fix various path manipulation issues
Fix missing parameter to endswith and pass paths through normpath to remove
any duplicate "/" characters which would corrupt other calls like basename.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sat, 3 Nov 2012 17:01:32 +0000 (17:01 +0000)]
classes: Be consistent about sstate-inputdirs/outputdirs ending with '/'
If sstate-inputdirs and sstate-outputdirs don't match with ending '/'
characters, the manifest file can end up corrupted. This change
ensures the metadata is consistent in ending do_populate_root tasks
with this character to avoid manifest file corruption.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sat, 3 Nov 2012 17:03:14 +0000 (17:03 +0000)]
sstate: Be consistent about sstate-inputdirs/outputdirs ending with '/'
The manifest file can become corrupted if sstate-inputdirs and sstate-outputdirs
don't have matching endings. This patch ensures that even if set incorrectly,
the code functions as intended, thereby handling manifest corruption safely.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Mon, 12 Nov 2012 16:18:35 +0000 (16:18 +0000)]
initrdscripts: fix udevd in the live boot init scripts
udevd moved location and isn't in $PATH anymore, so use an absolute path to
start it.
The control socket path moved too, so mkdir the directory it's in.
Mounts the new devtmpfs on /dev device tree.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Alexandru Damian <alexandru.damian@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Mon, 12 Nov 2012 17:52:36 +0000 (17:52 +0000)]
libunique: fix compilation with GLib 2.34
The new GLib has deprecated some functions, and libunique was building with
-Werror. Take a patch from upstream to update the build system and rationalise
the warning flags.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Laurentiu Palcu [Mon, 5 Nov 2012 13:06:59 +0000 (15:06 +0200)]
populate_sdk_base.bbclass: use SDK_ARCH instead of SDKMACHINE
If SDKMACHINE is not set, the toolchain will be built but the tarball
installer will not run. A better choice is to use SDK_ARCH because, even
if SDKMACHINE is not set, SDK_ARCH is set, by default, to BUILD_HOST.
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Mon, 5 Nov 2012 16:22:48 +0000 (16:22 +0000)]
glib-2.0: upgrade to latest stable, 2.34.1.
Also explicitly disable the test suite (as we can't run it), subsequently
dropping 60_wait-longer-for-threads-to-die.patch and nodbus.patch.
nolibelf.patch has been merged upstream, drop.
Upstream has dropped the pre-generated man pages, to generate them again we'd
need libxslt and the DocBook infrastructure. We can live without the man pages
as those build-dependencies are non-trivial.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Martin Jansa [Mon, 5 Nov 2012 23:36:47 +0000 (00:36 +0100)]
qt4: remove -lGLU from QMAKE_LIBS_OPENGL in our linux.conf
upstream does not need GLU since:
commit e7eed096a0c33607a7a37baaf06e5952dc9d556b
Author: Bj<C3><B8>rn Erik Nilsen <bjorn.nilsen@nokia.com>
Date: Mon Aug 9 14:07:01 2010 +0200
Remove dependency of OpenGL Utility Library (GLU).
GLU is not part of standard OpenGL and is not used internally in Qt,
so we should not depend on it.
Task-number: QT-12227 Reviewed-by: kim Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The drivers don't generally use AC_CHECK_FILE anymore, so remove this
brute-force kludge. Any subsequent breakage can be worked-around in the recipe
and fixes submitted upstream.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Tue, 30 Oct 2012 11:30:26 +0000 (11:30 +0000)]
gnomebase: depend on gnome-common-native
gnome-common is a build-only dependency so we should depend on the native
variant. This also resolves an (incorrect) GPLv3 license issue in gnome-common
at build-time.
This will also remove the pointless gnome-common-dev RRECOMMENDS in any -dev
package that uses gnomebase.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Ross Burton [Tue, 30 Oct 2012 12:02:45 +0000 (12:02 +0000)]
gnome-common: Fix license
gnome-common 2.28 is GPLv2+. From Christian Persch, upstream:
The licence is presumed GPL2+, although it's not there explicitly. GPL2+
because as far as I could figure out when I tried to, gnome-autogen started in
gnome-core which had a GPL2 COPYING file.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Khem Raj [Wed, 31 Oct 2012 06:45:47 +0000 (23:45 -0700)]
bitbake.conf: Add udev rules into ${PN} files by default
As we move to systemd, udev is not provided by systemd where the arch
independent files are stored in /lib and /usr/lib and not in
${base_libdir} and ${libdir} which means the files like udev rules
go into /lib/udev or /usr/lib/udev. This patch adds these paths
to be packaged into default ${PN} output package from a recipe
Kang Kai [Thu, 1 Nov 2012 02:44:56 +0000 (10:44 +0800)]
create-recipe: update re pattern and output
In the URL, there may be more than just digits in the version section,
something like xz 5.1.2alpha. Update RE pattern to catch all the string
after package name and before '.tar' in URL as package version.
And error message which has been sent to /dev/null still shows on Ubuntu
12.10 with perl 5.14. Update the way to find source tar file to eraser
the error message.
configure files may rewrite the version section, and that is not necessary.
Test when version section has been set, omit the version value from
configure files.
And tweak for output to bb file.
Signed-off-by: Kang Kai <kai.kang@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Chen Qi [Thu, 1 Nov 2012 06:47:46 +0000 (14:47 +0800)]
hwclock.sh: improve hwclock.sh script to use UTC variable
Make UTC variable in /etc/default/rcS has effect on hwclock.sh.
This variable declares whether the Hardware Clock is kept in UTC
or local time. Default its value to "yes" and change the comment.
Robert Yang [Sat, 27 Oct 2012 08:48:17 +0000 (16:48 +0800)]
recipes-support: replace virtclass-native(sdk) with class-native(sdk)
The overrides virtclass-native and virtclass-nativesdk are deprecated,
which should be replaced by class-native and class-nativesdk.
NOTE:
There were 2 errors in libcap.inc, the BUILD_LDFLAGS_virtclass_native
should be BUILD_LDFLAGS_virtclass-native (the "_" should be "-"),
otherwise it doesn't work, and the value was: "-Wl,rpath=...", this is
incorrect, it shoudl be: "-Wl,-rpath=..." (lacked a - ), but we don't
need this line, since it is already in the default BUILD_LDFLAGS. Remove
it and we don't need to bump the PR since we just removed a unused line.
[YOCTO #3297]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Daniel Stone [Fri, 26 Oct 2012 15:00:09 +0000 (16:00 +0100)]
mesa: Build separate GLU library
Mesa has removed GLU from the core tree upstream, so remove it from the
Mesa build and add the separate tarball as a new recipe.
Signed-off-by: Daniel Stone <daniel@fooishbar.org> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Fri, 2 Nov 2012 11:23:33 +0000 (11:23 +0000)]
terminal: Ensure existing environment exports are preserved in devshell
After recent changes to terminal.bbclass, variables like PATH were no longer
preserved within the devshell. This change ensures they are inherited into
the environment of devshell and PATH for example has the correct values.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Tue, 30 Oct 2012 16:36:50 +0000 (16:36 +0000)]
terminal.bbclass: Ensure parent environment is set
If this isn't done, various terminals fail to launch correctly
with "No such file or directory" errors. This adds back the environment
manipulation removed in the addition of "custom" terminal command
support but shouldn't regress that additional functionality
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Phil Blundell [Mon, 29 Oct 2012 15:20:19 +0000 (15:20 +0000)]
openssl: Use ${CFLAGS} not ${FULL_OPTIMIZATION}
The latter variable is only applicable for target builds and could
result in passing incompatible options (and/or failing to pass
required options) to ${BUILD_CC} for a virtclass-native build.
Signed-off-by: Phil Blundell <philb@gnu.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Mon, 29 Oct 2012 15:16:17 +0000 (15:16 +0000)]
udev: Ensure tmpfs are mounted and volatile/run exists
There is a race with udev where eiher the run directory can get replaced
during bootup leading to ude errors, or if the tmpfs was mounted and
populate-volatiles hasn't run, udev won't start at all.
This ensures that any tmpfs get mounted before udev starts and that the
default volatiles/run directory at least exists, fixing the races
and boot time errors caused after the recent udev upgrade.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Phil Blundell [Mon, 29 Oct 2012 17:36:37 +0000 (17:36 +0000)]
polkit: remove license.html from LIC_FILES_CHKSUM
This is a generated file and gets removed by "make clean" which then
causes subsequent rebuilds to fail. Also, the content in this file
is taken verbatim from COPYING (which is already in LIC_FILES_CHKSUM)
so checking it for a second time doesn't accomplish much.
Signed-off-by: Phil Blundell <philb@gnu.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Andy Ross [Tue, 18 Sep 2012 21:38:45 +0000 (14:38 -0700)]
udev-extraconf: Don't mount root filesystem under /media
The mount.sh handler attempts to prevent already-mounted filesystems
from being mounted as dynamic/removable "/media". But it misses the
case where the kernel has mounted the root filesystem (e.g. with
"root=/dev/sda1"). In that situation, /proc/mounts has a device name
of "/dev/root" instead of the proper $DEVNAME string exposed by udev.
So we must also test the root filesystem device number vs. the
$MAJOR/$MINOR udev tells us.
Signed-off-by: Andy Ross <andy.ross@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Andy Ross [Tue, 18 Sep 2012 21:38:44 +0000 (14:38 -0700)]
busybox: add /usr/bin/stat applet
The busybox defconfig lacks a stat tool, the functionality of which
cannot be reproduced in a way accessible to a shell script running in
a minimal configuration. Enable, and modify the installation path to
/usr/bin/stat to match the coreutils tool for proper alternatives
handling.
Signed-off-by: Andy Ross <andy.ross@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Tom Zanussi [Fri, 26 Oct 2012 16:21:16 +0000 (11:21 -0500)]
lttng-modules: remove unused lttng-syscalls patch
commit b7e184508 (lttng-2.0: fix srcrev/pv to match the recipe
filenames) removed the
lttng-sycalls-protect-is_compat_task-from-redefiniti.patch from the
SRC_URI but forgot to remove the patch itself.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Fri, 26 Oct 2012 07:03:49 +0000 (00:03 -0700)]
bitbake.conf: Change systemd_unitdir definition
systemd_unitdir indicates the arch independent
files which are basically scripts and unit files
and systemd wants then to be in /lib always even
when base_libdir is /lib64, hence we have to reflect
that and not use base_libdir to define it. Otherwise
on architectures where base_libdir is lib64 e.g. ppc64
or multilibbed x86_64 this wont work
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mihai Lindner [Fri, 26 Oct 2012 14:37:18 +0000 (17:37 +0300)]
distro_identifier: replace slash with hyphen
Use "-" instead of "/" in "n/a" strings ("Distributor ID" and/or
"Release"), provided by `lsb_release`.
This leads to directories and subdirectories created in ./sstate-cache/
e.g. Distro-n/a/ where "Distro-n" is dir and "a" is subdir.
Signed-off-by: Mihai Lindner <mihaix.lindner@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fabien Proriol [Tue, 16 Oct 2012 14:32:30 +0000 (14:32 +0000)]
populate_sdk_base: allow SDK path of various level
In the previous version, tar extraction use the --strip-component
option with "4" hard coded value.
If we set another SDKPATH, with a different depth, the sdk installation
fails.
This patch computes the level from the SDKPATH value.
Signed-off-by: Fabien Proriol <fabien.proriol@jdsu.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Saul Wold [Thu, 18 Oct 2012 19:25:05 +0000 (12:25 -0700)]
sstate: add manifest info for shared file matches
Present the manifest file that contains the matches for
files being installed to a location that already contains
that file. This will help to determine which is the correct
recipe to fix when this occurs.
Ross Burton [Wed, 24 Oct 2012 16:37:34 +0000 (17:37 +0100)]
pulseaudio: move helper binaries into the relevant module packages
proximity-helper is only used by the bluetooth-proximity module, and
gconf-helper is only used by the gconf module. Clarify the packaging and clean
up dependencies by shipping the helper binaries with the modules that spawn
them.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Peter Seebach [Tue, 23 Oct 2012 22:56:00 +0000 (17:56 -0500)]
insane.bbclass and friends: Fix sanity checks and multlib headers for n32
The n32 architecture is odd, in that it's a mips64 ABI which happens
to be 32-bit. To handle this, we need something in the environment
which can be used to distinguish it. The obvious place to stash this
is the ABI suffix, so we use "n32" as an ABI suffix. This allows
a couple of improved checks:
1. In insane.bbclass, we can use "linux-gnun32" to discern that it's
okay for a mips64 binary to be a 32-bit binary in some cases.
2. In multilib_header, we can check for the n32 ABI, and use a distinct
value.
3. In siteinfo, add linux-gnun32 as a synonym for linux, similar to
what's done for linux-gnux32, and tell the mips*-linux-gnun32 variants
to pick up the corresponding mips-linux site configs.
Note that the multilib header wrapper already has n32 hooks in it, there
was just nothing creating -n32 header variants.
Signed-off-by: Peter Seebach <peter.seebach@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Mark Hatle [Mon, 22 Oct 2012 14:55:47 +0000 (09:55 -0500)]
populate_sdk_base: Ensure that the multilib cross-canadian tools are used
Update the host toolchain list, for cross-canadian toolchains, to ensure
that all of the supported multilibs are built and installed. This
dynamically generates the dependnecy set based on the current multilib
configuration.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Mark Hatle [Wed, 26 Sep 2012 23:02:06 +0000 (18:02 -0500)]
populate_sdk_base: Update extraction script for multilibs
When multilibs are enabled, there will be more then one environment
file created. We need to be sure to process each environment file.
The next function can simply use the last environment file processed
to get the magic value(s) that it requires.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Mark Hatle [Wed, 26 Sep 2012 23:01:04 +0000 (18:01 -0500)]
multilib - crosssdk: Stop building multilib for crosssdk packages
Crosssdk packages are not actually multilib packages, so treat them
the same as other nativesdk packages in the multilib, base, and
classextend components.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Mark Hatle [Mon, 24 Sep 2012 21:25:37 +0000 (16:25 -0500)]
multilib: Add support for cross-canadian multilib packages
Add support for the generation of cross-canadian packages.
Each cross-canadian package has:
PN = "pkg-cross-canadian-${TRANSLATED_TARGET_ARCH}"
in order for that to be evaluated properly with multilibs enabled, it was
necessary to detect both the presence of the cross-canadian packages and
then update the vars using the OVERRIDE for the multilib. Additional checks
were made to ensure that any dependency that sais "cross-canadian" did not
get prefixed with the MLPREFIX.
Also, make sure that even when building multilib cross-canadian packages,
we only use the single SDK PACKAGE_ARCH, we don't want or need variants.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
