]> code.ossystems Code Review - openembedded-core.git/log
openembedded-core.git
12 years agodistro_tracking_fields: update information for ltp
Jiajun Xu [Thu, 24 May 2012 09:10:42 +0000 (17:10 +0800)]
distro_tracking_fields: update information for ltp

Update information for recipe ltp 20120401

Signed-off-by: Jiajun Xu <jiajun.xu@intel.com>
12 years agoltp: Upgrade LTP to version 20120401
Jiajun Xu [Thu, 24 May 2012 09:08:36 +0000 (17:08 +0800)]
ltp: Upgrade LTP to version 20120401

Signed-off-by: Jiajun Xu <jiajun.xu@intel.com>
12 years agofontconfig-native: Be clear about FILE search path
Richard Purdie [Thu, 24 May 2012 10:05:46 +0000 (11:05 +0100)]
fontconfig-native: Be clear about FILE search path

This avoids warnings from the new file checksum code.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agowpa-supplicant: add wpa_supplicant.conf to CONFFILES and don't install it twice
Martin Jansa [Thu, 24 May 2012 14:09:11 +0000 (16:09 +0200)]
wpa-supplicant: add wpa_supplicant.conf to CONFFILES and don't install it twice

* first we install ${WORKDIR}/wpa_supplicant.conf-sane
  install -m 600 ${WORKDIR}/wpa_supplicant.conf-sane ${D}${sysconfdir}/wpa_supplicant.conf
  and a bit later in do_install we were overwritting it with ${WORKDIR}/wpa_supplicant.conf
  install -m 644 ${WORKDIR}/wpa_supplicant.conf ${D}${sysconfdir}
* notice that this patch also changes .conf permissions from 644 back to 600

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
12 years agodropbear: fix include dependent PR generation
Steffen Sledz [Tue, 22 May 2012 14:16:00 +0000 (16:16 +0200)]
dropbear: fix include dependent PR generation

Signed-off-by: Steffen Sledz <sledz@dresearch-fe.de>
12 years agooe/process.py: remove it since it is unused and duplicated
Robert Yang [Thu, 24 May 2012 09:08:53 +0000 (17:08 +0800)]
oe/process.py: remove it since it is unused and duplicated

The meta/lib/oe/process.py is only used by oe_run and oe_popen in
meta/classes/utils.bbclass, and they will be removed, we have a better
one: bitbake/lib/bb/process.py, which can replace of it.

[YOCTO #2489]

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
12 years agoutils.bbclass: remove the unused oe_system
Robert Yang [Thu, 24 May 2012 08:03:26 +0000 (16:03 +0800)]
utils.bbclass: remove the unused oe_system

The following functions in meta/classes/utils.bbclass were not
used by anyone, and they didn't work when I tried to use them:

def oe_popen_env(d):
def oe_run(d, cmd, **kwargs):
def oe_popen(d, cmd, **kwargs):
def oe_system(d, cmd, **kwargs):

There error was:
AttributeError: type object 'str' has no attribute 'getVar'

We have bitbake/lib/bb/process.py to instead, so remove them.

[YOCTO #2489]

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
12 years agoopkg-nogpg: add file path
Paul Eggleton [Thu, 24 May 2012 16:06:12 +0000 (17:06 +0100)]
opkg-nogpg: add file path

There was no explicit path in FILESPATH for this recipe to allow it to
find files referred to in SRC_URI (files shared with the standard opkg
recipe). This problem may have been masked by the fetcher behaviour of
looking for files last in DL_DIR, thus if opkg has been fetched
beforehand there would have been no error.

Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
12 years agosudo: fixed wrong `chmod` path
Mihai Lindner [Sun, 20 May 2012 13:17:32 +0000 (13:17 +0000)]
sudo: fixed wrong `chmod` path

Placed $D between braces ${D} to be correctly expanded to the
workdir path, instead of a path relative to host rootfs.
Currently, bitbake sudo fails on host systems where sudo is not
installed.

Signed-off-by: Mihai Lindner <mihaix.lindner@linux.intel.com>
12 years agobitbake.conf: add EXTERNAL_TOOLCHAIN to BB_HASHBASE_WHITELIST
Christopher Larson [Wed, 16 May 2012 01:28:00 +0000 (20:28 -0500)]
bitbake.conf: add EXTERNAL_TOOLCHAIN to BB_HASHBASE_WHITELIST

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake.conf: Improve wget timeouts
Richard Purdie [Thu, 24 May 2012 10:02:53 +0000 (11:02 +0100)]
bitbake.conf: Improve wget timeouts

The wget default is a 900 second timeout and 20 retries. This is way too long
for most of our usecases so this patch changes it to a 30 second timeout and
reduces retries from 5 to 2. We have good mirror infrastructure, this will
let us fall back to it easier.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoqemu: disable vnc-jpeg compression
Saul Wold [Tue, 22 May 2012 22:24:13 +0000 (15:24 -0700)]
qemu: disable vnc-jpeg compression

This is disabled by default to ensure a deterministic build, qemu
configure would sometimes detect the libjpeg and then enable this.
Since mostly qemu is used locally disable by default.

Signed-off-by: Saul Wold <sgw@linux.intel.com>
12 years agoqemu.inc: Remove mesa-xlib as PREFERRED_PROVIDER
Saul Wold [Fri, 11 May 2012 17:13:06 +0000 (10:13 -0700)]
qemu.inc: Remove mesa-xlib as PREFERRED_PROVIDER

The xserver-xorg uses and depends on mesa-dri, so we should
use the default PREFERRED_PROVIDER of libgl as mesa-dri.

This resolves the following:
ERROR: Multiple .bb files are due to be built which each provide virtual/libgl (/intel/poky/distro/meta/recipes-graphics/mesa/mesa-dri_7.11.bb /intel/poky/distro/meta/recipes-graphics/mesa/mesa-xlib_7.11.bb).
 This usually means one provides something the other doesn't and should.

Signed-off-by: Saul Wold <sgw@linux.intel.com>
12 years agoxserver-xf86-config: Add xorg.conf for arm,mips,ppc,sh qemu machines
Khem Raj [Thu, 3 May 2012 03:19:50 +0000 (20:19 -0700)]
xserver-xf86-config: Add xorg.conf for arm,mips,ppc,sh qemu machines

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agoqemumachines: Enable xserver-xorg as default xserver
Khem Raj [Tue, 1 May 2012 05:01:24 +0000 (22:01 -0700)]
qemumachines: Enable xserver-xorg as default xserver

For qemux86 and qemux86-64 include qemu.inc after defining XSERVER

XSERVER variable is also weakly defined in task-core-x11.bb
which means we can not use ??= otherwise when building any qemu image
that uses task-core-x11.bb will get the wrong definition

So we define the XSERVER common set for qemu in qemu.inc
and as we know x86 and x86-64 qemu overrides the default
we include qemu.inc after that definition which means that
qemux86 and qemux86-64 get their own definitions and other
qemus get the definitions from qemu.inc. other non-qemu machine
will get their defintion from task which points to kdrive
as of now.

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agoUpdate to distro_tracking_fields.inc
Laurentiu Palcu [Wed, 23 May 2012 11:54:14 +0000 (14:54 +0300)]
Update to distro_tracking_fields.inc

The fields for the following packages were updated:

libdrm
xf86-video-vmware
xf86-input-mouse
xf86-video-vesa
xf86-video-intel
xf86-input-synaptics
mdadm
xextproto
xcmiscproto
scrnsaverproto
recordproto
xproto
kbproto
fontsproto
libxt
xtrans
libpciaccess
fontconfig
xcb-proto

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agoxf86-input-synaptics: upgrade to 1.6.1
Laurentiu Palcu [Wed, 23 May 2012 08:36:12 +0000 (11:36 +0300)]
xf86-input-synaptics: upgrade to 1.6.1

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agoxf86-video-intel: upgrade to 2.19.0
Laurentiu Palcu [Wed, 23 May 2012 08:30:49 +0000 (11:30 +0300)]
xf86-video-intel: upgrade to 2.19.0

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agoxf86-video-vesa: upgrade to 2.3.1
Laurentiu Palcu [Wed, 23 May 2012 08:24:32 +0000 (11:24 +0300)]
xf86-video-vesa: upgrade to 2.3.1

Also, licence checksum has changed because a new copyright line from
Red Hat has been added. The rest of licence file, however, is unchanged.

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agoxf86-input-mouse: upgrade to 1.7.2
Laurentiu Palcu [Wed, 23 May 2012 07:55:17 +0000 (10:55 +0300)]
xf86-input-mouse: upgrade to 1.7.2

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agoxtrans: upgrade to 1.2.7
Laurentiu Palcu [Tue, 22 May 2012 13:33:15 +0000 (16:33 +0300)]
xtrans: upgrade to 1.2.7

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agolibxt: upgrade to 1.1.3
Laurentiu Palcu [Tue, 22 May 2012 13:04:49 +0000 (16:04 +0300)]
libxt: upgrade to 1.1.3

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agoxcb-proto: upgrade to 1.7.1
Laurentiu Palcu [Tue, 22 May 2012 12:41:12 +0000 (15:41 +0300)]
xcb-proto: upgrade to 1.7.1

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agorecordproto: upgrade to 1.14.2
Laurentiu Palcu [Tue, 22 May 2012 12:17:51 +0000 (15:17 +0300)]
recordproto: upgrade to 1.14.2

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agoscrnsaverproto: upgrade to 1.2.2
Laurentiu Palcu [Tue, 22 May 2012 12:14:54 +0000 (15:14 +0300)]
scrnsaverproto: upgrade to 1.2.2

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agoxcmiscproto: upgrade to 1.2.2
Laurentiu Palcu [Tue, 22 May 2012 12:11:03 +0000 (15:11 +0300)]
xcmiscproto: upgrade to 1.2.2

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agoxextproto: upgrade to 7.2.1
Laurentiu Palcu [Tue, 22 May 2012 12:07:19 +0000 (15:07 +0300)]
xextproto: upgrade to 7.2.1

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agokbproto: upgrade to 1.0.6
Laurentiu Palcu [Tue, 22 May 2012 11:49:05 +0000 (14:49 +0300)]
kbproto: upgrade to 1.0.6

Also, shifted up one line the start and end lines for licence checksum
computation.

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agoxproto: upgrade to 7.0.23
Laurentiu Palcu [Tue, 22 May 2012 11:15:38 +0000 (14:15 +0300)]
xproto: upgrade to 7.0.23

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agolibdrm: upgrade to 2.4.34
Laurentiu Palcu [Mon, 21 May 2012 14:58:52 +0000 (17:58 +0300)]
libdrm: upgrade to 2.4.34

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agomdadm: upgrade to 3.2.5
Laurentiu Palcu [Tue, 22 May 2012 07:06:18 +0000 (10:06 +0300)]
mdadm: upgrade to 3.2.5

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agovalgrind: Enable valgrind for armv7
Samuel Stirtzel [Wed, 23 May 2012 13:46:28 +0000 (15:46 +0200)]
valgrind: Enable valgrind for armv7

Valgrind supports the armv7 architecture, this patch allows armv7 users to build and use Valgrind

This patch was run-tested on a Gumstix Overo (armv7a cortex-a8)
* The test consisted of running valgrinds memcheck (memory leakage detection),
* and callgrind (profiling) on a Qt 4 application

Signed-off-by: Samuel Stirtzel <s.stirtzel@googlemail.com>
12 years agogawk: update to 4.0.1
Radu Moisan [Wed, 23 May 2012 07:20:50 +0000 (10:20 +0300)]
gawk: update to 4.0.1

Signed-off-by: Radu Moisan <radu.moisan@intel.com>
12 years agojpeg: update to upstream version 8d
Marko Lindqvist [Sun, 13 May 2012 18:37:03 +0000 (21:37 +0300)]
jpeg: update to upstream version 8d

* fix_for_automake_1.11.2.patch removed as issue resolved upstream
* Have README as license file instead of several source files that just
  refer to it

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
12 years agoglib-2.0: upgrade to 2.32.3
Martin Jansa [Thu, 17 May 2012 07:02:51 +0000 (09:02 +0200)]
glib-2.0: upgrade to 2.32.3

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
12 years agosanity.bbclass: check sanity at BuildStarted rather than ConfigParsed
Joshua Lock [Wed, 23 May 2012 23:36:42 +0000 (16:36 -0700)]
sanity.bbclass: check sanity at BuildStarted rather than ConfigParsed

This enables a user to use bitbake -e even when the sanity checks are
failing.

Signed-off-by: Joshua Lock <josh@linux.intel.com>
12 years agosanity.bbclass: catch an extra exception in check_create_long_filename
Joshua Lock [Wed, 23 May 2012 23:28:59 +0000 (16:28 -0700)]
sanity.bbclass: catch an extra exception in check_create_long_filename

The call to bb.mkdirhier() in check_create_long_filename() can fail with an
OSError, explicitly catch this and report something useful to the user.

Signed-off-by: Joshua Lock <josh@linux.intel.com>
12 years agosanity.bbclass: add extra information when SSTATE_CACHE unusable
Joshua Lock [Wed, 23 May 2012 23:22:12 +0000 (16:22 -0700)]
sanity.bbclass: add extra information when SSTATE_CACHE unusable

If the user does not have write permissions to SSTATE_CACHE, detected by
the check_create_long_filename() test failing with a "Permission denied"
value in strerror, then suggest they might want to use the location as
an entry in SSTATE_MIRRORS.

Signed-off-by: Joshua Lock <josh@linux.intel.com>
12 years agosanity.bbclass: add newline to check_create_long_filename failure message
Joshua Lock [Wed, 23 May 2012 23:19:23 +0000 (16:19 -0700)]
sanity.bbclass: add newline to check_create_long_filename failure message

Each failure in the sanity message should be reported on a new line.

Signed-off-by: Joshua Lock <josh@linux.intel.com>
12 years agosanity.bbclass: data.getVar(VAR, obj, exp) -> obj.getVar(VAR, exp)
Joshua Lock [Wed, 23 May 2012 22:42:24 +0000 (15:42 -0700)]
sanity.bbclass: data.getVar(VAR, obj, exp) -> obj.getVar(VAR, exp)

Replace calls to data.getVar(VARIABLE, data_object, expand) to
direct calls to the getVar method the the data_object.

Signed-off-by: Joshua Lock <josh@linux.intel.com>
12 years agosanity.bbclass: copy the data store and finalise before running checks
Joshua Lock [Wed, 23 May 2012 22:16:19 +0000 (15:16 -0700)]
sanity.bbclass: copy the data store and finalise before running checks

At the ConfigParsed event the datastore has yet to be finalised and thus
appends and overrides have not been set.
To ensure the sanity check is being run against the configuration values
the user has set call finalize() on a copy of the datastore and pass that
for all sanity checks.

Signed-off-by: Joshua Lock <josh@linux.intel.com>
12 years agoRevert "sanity.bbclass: check user can read and write to SSTATE_DIR"
Joshua Lock [Wed, 23 May 2012 20:50:38 +0000 (13:50 -0700)]
Revert "sanity.bbclass: check user can read and write to SSTATE_DIR"

This has caused problems for several users, including the Yocto Project
autobuilder. Since the message was added in order to be more user friendly
revert the change.

This reverts commit 0c0c4efbf92bcf0f8942f17c18525a4b4ed1798c.

12 years agopython: bump PR of packages after update of distutils.bbclass
Andreas Oberritter [Mon, 21 May 2012 20:54:24 +0000 (22:54 +0200)]
python: bump PR of packages after update of distutils.bbclass

* Bump every recipe inheriting distutils or setuptools and
  not overriding do_install without calling distutils_do_install.

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
12 years agodistutils.bbclass: don't delete .pyo files
Andreas Oberritter [Mon, 21 May 2012 20:54:23 +0000 (22:54 +0200)]
distutils.bbclass: don't delete .pyo files

* Deleting .pyo files causes them to get compiled on the target.
* First boot gets *really* slow for python based projects.
* No space gets saved on the target.
* The package manager doesn't know about the files and
  therefore fails to uninstall them, occupying space and causing
  uninstalled python scripts to remain executable.
* It's inconsistent, because python itself and autotools based
  projects already ship .pyo files.
* Probably .pyo files were deleted because .pyc files were
  available earlier, but this has changed and OE-Core's python
  now only generates optimized .pyo files. Deletion of .pyo was
  introduced in 2008, python/04-default-is-optimized.patch
  was introduced in 2009.

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
12 years agonetbase: merge init script updates from upstream
Andreas Oberritter [Mon, 21 May 2012 16:39:07 +0000 (18:39 +0200)]
netbase: merge init script updates from upstream

* Read /proc/mounts only once.
* Support many more network filesystem types.
* Remaining differences to netbase 4.47:
  - Uses the mountvirtfs keyword instead of mountkernfs
  - Doesn't use lsb functions
  - Doesn't print a warning if /etc/network/options exists
  - Doesn't use --exclude parameter for ifup, because
    busybox doesn't support it.

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
12 years agowpa-supplicant: fix packaging error since last commit
Andreas Oberritter [Tue, 22 May 2012 17:17:50 +0000 (19:17 +0200)]
wpa-supplicant: fix packaging error since last commit

* wpa-passphrase has its own package, but commit
  4a4c568e25a08e9f222d723f9819582c9f895c58 broke it.

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
12 years agolibnss-mdns: fix postinst scripts
Andreas Oberritter [Mon, 21 May 2012 14:53:53 +0000 (16:53 +0200)]
libnss-mdns: fix postinst scripts

* On upgrade, postinst ocassionally returned 1, so use a
  conditional instead of &&.
* Use sed patterns in order to make it work more generally.

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
12 years agoppp: avoid host contamination
Andreas Oberritter [Tue, 22 May 2012 22:31:58 +0000 (00:31 +0200)]
ppp: avoid host contamination

* Import patch from OE-Classic commit f7c547ede1bfc94c24f6315a23ff48a4c5b9ffaf.
* Fixes linking with libpcap (ppp filter).

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
12 years agolibtool: fixed parallel build related race
Enrico Scholz [Sun, 20 May 2012 17:54:37 +0000 (19:54 +0200)]
libtool: fixed parallel build related race

While building libtool, the libtool script itself will be regenerated
because OE modifies a dependency[1]. With -jX, this operation (-->
removal, creation of non-x file, 'chmod a+x') can happen at a time when
the script is going to be executed.  This can cause errors like:

| arm-linux-gnueabi-libtool: compile:  ccache arm-linux-gnueabi-gcc ...
| ...
| /bin/sh ./config.status libtool
| ...
| arm-linux-gnueabi-libtool: compile:  ccache arm-linux-gnueabi-gcc ...
| /bin/sh: ./arm-linux-gnueabi-libtool: Permission denied
| make[2]: *** [libltdl/libltdl_libltdl_la-lt__alloc.lo] Error 126

I am not sure whether the custom do_compile_prepend() is still needed.
For now only the issue above will be fixed by executing ./config.status
yet again.

[1] see 648290d5bf4d6ff50d3643bb7ad902dfc23aa702

Signed-off-by: Enrico Scholz <enrico.scholz@sigma-chemnitz.de>
12 years agoncurses: Avoid occasional builling failure when having parallel processable task
Xiaofeng Yan [Mon, 21 May 2012 06:59:22 +0000 (14:59 +0800)]
ncurses: Avoid occasional builling failure when having parallel processable task

ncurses failure non-gplv3 build (race issue) like the following \
error information:

| tic: error while loading shared libraries: /srv/home/pokybuild \
/yocto-autobuilder/yocto-slave/nightly-non-gpl3/build/build/tmp/\
work/x86_64-linux/ncurses-native-5.9-r8.1/ncurses-5.9/narrowc/lib\
/libtinfo.so.5: file too short
| ? tic could not build /srv/home/pokybuild/yocto-autobuilder/\
yocto-slave/nightly-non-gpl3/build/build/tmp/work/x86_64-linux/\
ncurses-native-5.9-r8.1/image/srv/home/pokybuild/yocto-autobuilder\
/yocto-slave/nightly-non-gpl3/build/build/tmp/sysroots/x86_64-linux\
/usr/share/terminfo
| make[1]: *** [install.data] Error 1

This is a race issue which is caused by
install.libs and install.data:

1) install.data needs run tic
2) tic needs libtinfo.so
3) install.libs would regenerate libtinfo.so
4) but install.data doesn't depend on install.libs, and they can run
   parallelly

So there would be errors in a very critical condition: tic is begining
to run at the same time when install.libs is generating libtinfo.so, and
this libtinfo.so is not integrity, then there would be the  above error.

Let task install.libs run before install.data for fixing this bug.

[YOCTO #2298]

Signed-off-by: Xiaofeng Yan <xiaofeng.yan@windriver.com>
12 years agoeglibc: package mtrace separately
Nitin A Kamble [Tue, 15 May 2012 16:31:32 +0000 (09:31 -0700)]
eglibc: package mtrace separately

add libc-mtrace as dependency for task-core-tools-debug

now eglibc-mtrace gets included in an sdk image and not in a non-sdk image.

This does not affect builds with uclibc.

This fixes bug: [YOCTO# 2374]

Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
12 years agorpm 5.4.0: respect to the arch when choose the alternatives
Robert Yang [Wed, 23 May 2012 05:55:39 +0000 (13:55 +0800)]
rpm 5.4.0: respect to the arch when choose the alternatives

There is a bug if we:
1) bitbake diffutils with MACHINE=crownbay
2) bitbake diffutils with MACHINE=qemux86
3) bitbake core-image-sato with MACHINE=crownbay

Then the diffutils.i586 would be installed to the crownbay's image, this
is because diffutils.i586 is newer than diffutils.core2, and rpm doesn't
respect to the arch priorities:

We have put the archs in order in _solve_dbpath:

crownbay/solvedb:core2/solvedb:i586/solvedb:all/solvedb

Fix rpm to respect to the order, for example, if it finds a pkg in both
core2/ and i586/, and the core2/ comes first, it should not use the one
in i586/ even if it's build time is newer.

Note: Don't worry about the _free(*ptr), it can check whether ptr is
NULL or not.

This is for the denzil branch, and the master branch also needs it.

[YOCTO #2360]

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
12 years agopackage_rpm.bbclass: Fix incremental rpm image generation
Robert Yang [Wed, 9 May 2012 01:15:46 +0000 (09:15 +0800)]
package_rpm.bbclass: Fix incremental rpm image generation

Fix the incremental rpm image generation, it didn't work since the code
has been changed.

The btmanifest should have a ".manifest" suffix, so that it can be moved
to ${T} by rootfs_rpm.bbclass:
mv ${IMAGE_ROOTFS}/install/*.manifest ${T}/

Note: The locale pkgs would always be re-installed.

[YOCTO #2440]

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
12 years agoclasses/base: add checksum file list at parse time
Paul Eggleton [Mon, 21 May 2012 15:56:15 +0000 (16:56 +0100)]
classes/base: add checksum file list at parse time

Connect the new fetcher file checksum code so that we get a list of the
files to be checksummed at parse time.

The file-checksums flag will not be read unless we are using a version
of BitBake that supports the function we call within it, so it is safe
to include this change even when the metadata will still be used with
older versions of BitBake.

Implements [YOCTO #2044].

Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
12 years agoexternal-sourcery-toolchain: ignore GNU_HASH issues with its packages
Christopher Larson [Mon, 21 May 2012 23:09:12 +0000 (16:09 -0700)]
external-sourcery-toolchain: ignore GNU_HASH issues with its packages

Signed-off-by: Christopher Larson <kergoth@gmail.com>
12 years agoRename 'external-csl' to 'external-sourcery'
Christopher Larson [Mon, 21 May 2012 23:09:11 +0000 (16:09 -0700)]
Rename 'external-csl' to 'external-sourcery'

This is a rename per the purchase of CodeSourcery by Mentor Graphics
Corporation, and associated naming change.

Signed-off-by: Christopher Larson <kergoth@gmail.com>
12 years agotcmode-external-csl: preferred external-csl-toolchain's gdbserver
Christopher Larson [Mon, 21 May 2012 23:09:10 +0000 (16:09 -0700)]
tcmode-external-csl: preferred external-csl-toolchain's gdbserver

Signed-off-by: Christopher Larson <kergoth@gmail.com>
12 years agogcc: Grok for callconvention-hard to enable hard float
Khem Raj [Tue, 1 May 2012 05:00:17 +0000 (22:00 -0700)]
gcc: Grok for callconvention-hard to enable hard float

If callconvention-hard is set then we build gcc defaulting
to hard-float ABI

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agoeglibc: Add ARM hf dynamic linker support
Khem Raj [Tue, 1 May 2012 04:59:39 +0000 (21:59 -0700)]
eglibc: Add ARM hf dynamic linker support

The work is done in glibc upstream we backport
the relevant patches

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agogcc-4.7: Add knowledge about arm hf dynamic loader
Khem Raj [Mon, 30 Apr 2012 22:55:34 +0000 (15:55 -0700)]
gcc-4.7: Add knowledge about arm hf dynamic loader

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agotune-mips64.inc: Add new tune file for mips64 big-endian
Khem Raj [Thu, 10 May 2012 21:00:57 +0000 (14:00 -0700)]
tune-mips64.inc: Add new tune file for mips64 big-endian

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agoxserver-xorg: Fix build for mips64
Khem Raj [Thu, 10 May 2012 20:58:43 +0000 (13:58 -0700)]
xserver-xorg: Fix build for mips64

mips64 N64 ABI has different size for int and pointer
which means the compiler will complain about conversions
which works fine for 32 bit since sizes are same.

amd64 is taken care of already so we take care of mips64
with this patch by understanding the difference in sizes
in compiler.h

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agosite/sh-common: Add missing caches variables to build glib-2.32
Khem Raj [Tue, 8 May 2012 18:53:08 +0000 (11:53 -0700)]
site/sh-common: Add missing caches variables to build glib-2.32

This is on the same tone as done for other architectures

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agonetbase: Add interface files for qemumips64 and qemush4
Khem Raj [Fri, 4 May 2012 05:55:49 +0000 (22:55 -0700)]
netbase: Add interface files for qemumips64 and qemush4

These files are alike other qemu machines they
help in creating a working network interface when
using qemumips64 or qemush4

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agorunqemu: Add qemush4 and qemumips64 knowledge
Khem Raj [Wed, 25 Apr 2012 22:47:51 +0000 (15:47 -0700)]
runqemu: Add qemush4 and qemumips64 knowledge

New machines need to be added and they also
have different kernel commandlines

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agolibc-package: Add sh4 and mips64 to arch options
Khem Raj [Wed, 25 Apr 2012 22:41:14 +0000 (15:41 -0700)]
libc-package: Add sh4 and mips64 to arch options

needed for new architecture support

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agoeglibc-2.15: Support mips64
Khem Raj [Wed, 25 Apr 2012 22:39:07 +0000 (15:39 -0700)]
eglibc-2.15: Support mips64

Add mips64 to binary locale arches
Enabling ports in addons causes wrong
machtype to be chosen for mips64 and
it always chose mips32 which is wrong
for mips64 so enable all the addons
we have and we were enabling them anyway

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agokernel-arch.bbclass: Map mips64{el} to mips KARCH
Khem Raj [Wed, 25 Apr 2012 22:37:52 +0000 (15:37 -0700)]
kernel-arch.bbclass: Map mips64{el} to mips KARCH

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agobinutils: Default to n64 when configured for mips64
Khem Raj [Wed, 25 Apr 2012 22:37:00 +0000 (15:37 -0700)]
binutils: Default to n64 when configured for mips64

Default ABI for mips64 is n64 so make it default in binutils

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agogcc-4.6, gcc-4.7: Add support for building mips64 cross compiler
Khem Raj [Wed, 25 Apr 2012 22:35:39 +0000 (15:35 -0700)]
gcc-4.6, gcc-4.7: Add support for building mips64 cross compiler

Defaults to n64 ABI

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agosite: Add mips64 eglibc and uclibc site files
Khem Raj [Wed, 25 Apr 2012 22:34:50 +0000 (15:34 -0700)]
site: Add mips64 eglibc and uclibc site files

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agoinsane.bbclass: Add mips64{el} to known machines
Khem Raj [Wed, 25 Apr 2012 22:32:12 +0000 (15:32 -0700)]
insane.bbclass: Add mips64{el} to known machines

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agosstate.bbclass: Make sure we don't have an empty fixmepath file
Mark Hatle [Fri, 18 May 2012 14:18:19 +0000 (17:18 +0300)]
sstate.bbclass: Make sure we don't have an empty fixmepath file

Jason Wessel noticed that a package without any fixmepath entries would
generate a sed warning about no input files.  This patch resolves that
by ensuring that an empty fixmepath file never gets written into the
sstate archive.  Also we avoid a second message by only doing xargs if
we got input.

Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoxserver-kdrive: Fix X server on PowerPC when built with GCC 4.7.x
Gary Thomas [Fri, 18 May 2012 14:18:18 +0000 (17:18 +0300)]
xserver-kdrive: Fix X server on PowerPC when built with GCC 4.7.x

Newer versions of GCC uncovered illegal code practice which can cause
runtime failures in the Xext/xace.c code.  This problem has been fixed upstream:
  http://cgit.freedesktop.org/xorg/xserver/patch/Xext/xace.c?id=6dae7f3792611aace1df0cca63bf50c50d93de43
  Subject: xace: Invalid reference to out-of-scope data.

Incorporate this patch into the recipe, with a PR bump.

Signed-off-by: Gary Thomas <gary@mlbassoc.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agolinux-yocto: policy cleanups
Bruce Ashfield [Fri, 18 May 2012 14:18:17 +0000 (17:18 +0300)]
linux-yocto: policy cleanups

Updating the meta SRCREVs to pickup configuration policy cleanups:

  49f931b meta/fishriver: remove redundant features and options
  51a6d3f meta/emenlow: remove redundant features and options
  101dd7f meta/crownbay: remove redundant features and options
  4110ecd meta/sugarbay: remove redundant features and options
  0f1304a meta/jasperforest: remove redundant features and options
  0a56a3b meta/common-pc-64: factor out SCSI CDROM option
  b71938a meta/common-pc-64: use usb-mass-storage feature
  0724f40 meta: add scsi cdrom feature
  438bca8 meta/common-pc: use usb-mass-storage feature
  c970881 meta: factor out SCSI options from the usb-mass-storage feature
  4c8135e meta: add scsi disk feature
  6872a81 meta: add scsi feature
  e706ec5 meta/sugarbay: factor out policy-related options
  8b7fbc2 meta/jasperforest: factor out policy-related options
  fea1b0e meta/fishriver: factor out policy-related options
  13bf9ab meta/emenlow: factor out policy-related options
  4748d50 meta/crownbay: factor out policy-related options
  44f592f meta/common-pc-64: factor out policy-related options
  5a3f5c7 meta/common-pc: factor out policy-related options
  1f5a10b meta/common-pc-64: use usb features
  4b87723 meta/common-pc: use usb features
  594ba05 meta: add ROOT_HUB_TT config option to the usb/ehci-hcd feature

Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agokern-tools: update LICENSE field to GPLv2
Bruce Ashfield [Fri, 18 May 2012 14:18:16 +0000 (17:18 +0300)]
kern-tools: update LICENSE field to GPLv2

The LICENSE field for kern-tools was generic and leads to QA warnings
from the license classs:

  "No generic license file exists for: GPL in any provider"

Updating to a specific GPL version that matches the source fixes the
warning.

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agolinux-yocto: intel BSP config changes
Bruce Ashfield [Fri, 18 May 2012 14:18:15 +0000 (17:18 +0300)]
linux-yocto: intel BSP config changes

Updating the meta SRCREV for the following fixes:

   1dfd60f meta/fishriver: move smp options from recipe-space
   012780a meta/emenlow: move smp options from recipe-space
   b59b1a5 meta/crownbay: move smp options from recipe-space
   74dc6ac meta/sugarbay: remove boot-live options
   a4bedcb meta/jasperforest: remove boot-live options
   4ae7b81 meta/sugarbay: use usb features
   30e7e8c meta/jasperforest: use usb features
   22d0c5d meta/fishriver: use usb features
   e262965 meta/emenlow: use usb features

Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agotest-reexec: Add script to address issues when task re-execution
Jiajun Xu [Fri, 18 May 2012 14:18:14 +0000 (17:18 +0300)]
test-reexec: Add script to address issues when task re-execution

The script is used to address build issues when tasks of different
recipes are re-executed. The script goes through all available recipes
and their tasks. The test results are saved in ./reexeclogs. Force build
logs are saved with prefix "force". Build failure logs are saved with
prefix "failed".

[YOCTO #2123]

Signed-off-by: Jiajun Xu <jiajun.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agomultilib.bbclass: Added multilib specific package QA.
Lianhao Lu [Fri, 18 May 2012 14:18:13 +0000 (17:18 +0300)]
multilib.bbclass: Added multilib specific package QA.

Added a new PACKAGEFUNCS function to check the multilib packages'
dependency.

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agomultilib.conf: Added multilib support for kmod.
Lianhao Lu [Fri, 18 May 2012 14:18:12 +0000 (17:18 +0300)]
multilib.conf: Added multilib support for kmod.

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years ago(libc-)package.bbclass: Added MLPREFIX to locale packages.
Lianhao Lu [Fri, 18 May 2012 14:18:11 +0000 (17:18 +0300)]
(libc-)package.bbclass: Added MLPREFIX to locale packages.

Added multilib prefix to the locale related package names/dependencies.

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoRevert "kmod: Use base_libdir for installing libkmod"
Koen Kooi [Tue, 15 May 2012 09:32:21 +0000 (11:32 +0200)]
Revert "kmod: Use base_libdir for installing libkmod"

The commit breaks pkgconfig and after discussing it with the kmod and udev maintainers the conclusion was reached that putting the libraries in /lib instead of /usr/lib is not supported.

This reverts commit 6b74f2461735272bd950a4f060dab6e778a36f92.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoxserver-xorg-lite: disable xmlto like full xserver-xorg does
Martin Jansa [Sat, 12 May 2012 12:18:33 +0000 (14:18 +0200)]
xserver-xorg-lite: disable xmlto like full xserver-xorg does

* xmlto-native is unusable (see YOCTO #2416
  xmlto/xsltproc stylesheets cannot be found even when they are installed in sysroot)

* fails like this:
| make[2]: Entering directory `/OE/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/xserver-xorg-lite/xserver-xorg-lite-2_1.11.2-r2/xorg-server-1.11.2/doc'
| /OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/bin/xmlto   xhtml-nochunks Xserver-spec.xml
| /OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/bin/xmlto   xhtml-nochunks Xinput.xml
| xmlto: /OE/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/xserver-xorg-lite/xserver-xorg-lite-2_1.11.2-r2/xorg-server-1.11.2/doc/Xinput.xml does not validate (status 3)
| xmlto: Fix document syntax or use --skip-validation option
| /OE/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/xserver-xorg-lite/xserver-xorg-lite-2_1.11.2-r2/xorg-server-1.11.2/doc/Xinput.xml:3: warning: failed to load external entity "http://www.oasis-open.org/docbook/xml/4.3/docbookx.dtd"
|                    "http://www.oasis-open.org/docbook/xml/4.3/docbookx.dtd">
|                                                                             ^
| /OE/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/xserver-xorg-lite/xserver-xorg-lite-2_1.11.2-r2/xorg-server-1.11.2/doc/Xinput.xml:40: parser error : Entity 'ldquo' not defined
| btaining a copy of this software and associated documentation files (the &ldquo;
|                                                                                ^
| /OE/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/xserver-xorg-lite/xserver-xorg-lite-2_1.11.2-r2/xorg-server-1.11.2/doc/Xinput.xml:40: parser error : Entity 'rdquo' not defined
|  of this software and associated documentation files (the &ldquo;Software&rdquo;
|                                                                                ^
| /OE/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/xserver-xorg-lite/xserver-xorg-lite-2_1.11.2-r2/xorg-server-1.11.2/doc/Xinput.xml:42: parser error : Entity 'ldquo' not defined
| btaining a copy of this software and associated documentation files (the &ldquo;
|                                                                                ^
| /OE/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/xserver-xorg-lite/xserver-xorg-lite-2_1.11.2-r2/xorg-server-1.11.2/doc/Xinput.xml:42: parser error : Entity 'rdquo' not defined
|  of this software and associated documentation files (the &ldquo;Software&rdquo;
|                                                                                ^
| warning: failed to load external entity "http://www.oasis-open.org/docbook/xml/4.3/docbookx.dtd"
| validity error : Could not load the external subset "http://www.oasis-open.org/docbook/xml/4.3/docbookx.dtd"
| Document /OE/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/xserver-xorg-lite/xserver-xorg-lite-2_1.11.2-r2/xorg-server-1.11.2/doc/Xinput.xml does not validate
| xmlto: /OE/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/xserver-xorg-lite/xserver-xorg-lite-2_1.11.2-r2/xorg-server-1.11.2/doc/Xserver-spec.xml does not validate (status 3)
| xmlto: Fix document syntax or use --skip-validation option
| make[2]: *** [Xinput.html] Error 13
| make[2]: *** Waiting for unfinished jobs....
| /OE/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/xserver-xorg-lite/xserver-xorg-lite-2_1.11.2-r2/xorg-server-1.11.2/doc/Xserver-spec.xml:4: warning: failed to load external entity "/xserver/doc/xml/xserver.ent"
|  <!ENTITY % defs SYSTEM "/xserver/doc/xml/xserver.ent"> %defs;
|                                                               ^
| Entity: line 1:
|  %defs;
|        ^

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
12 years agosanity.bbclass: check user can read and write to SSTATE_DIR
Joshua Lock [Sat, 12 May 2012 01:00:26 +0000 (18:00 -0700)]
sanity.bbclass: check user can read and write to SSTATE_DIR

The user needs read and write permissions to SSTATE_DIR, check
whether they have sufficient permissions and if not recommend
use of SSTATE_MIRRORS.

Signed-off-by: Joshua Lock <josh@linux.intel.com>
12 years agoxf86-video-vmware: update package to 12.0.2
Alexandru DAMIAN [Wed, 16 May 2012 14:27:01 +0000 (17:27 +0300)]
xf86-video-vmware: update package to 12.0.2

This patch updates the package version for the vmware xorg driver
to 12.0.2, following upstream. No other changes made beside updating
the checksums.

12 years agognupg: Update to upstream version 2.0.19
Laurentiu Palcu [Wed, 16 May 2012 13:47:49 +0000 (16:47 +0300)]
gnupg: Update to upstream version 2.0.19

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
12 years agognutls: update to 2.12.19
Bogdan Marinescu [Wed, 16 May 2012 12:07:37 +0000 (15:07 +0300)]
gnutls: update to 2.12.19

Tested with curl in qemux86/core-image-lsb.

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
12 years agoarchiver.bbclass: Add two default set for avoiding error when missing assignment
Xiaofeng Yan [Wed, 16 May 2012 08:59:45 +0000 (16:59 +0800)]
archiver.bbclass: Add two default set for avoiding error when missing assignment

Add two default set(SOURCE_ARCHIVE_LOG_WITH_SCRIPTS, \
SOURCE_ARCHIVE_PACKAGE_TYPE) to archiver.bbclass for avoiding \
building error when forgetting to assign to them.

Signed-off-by: Xiaofeng Yan <xiaofeng.yan@windriver.com>
12 years agometa/classes/sstate.bbclass: ensure sstate files have group R/W
Joshua Lock [Sat, 12 May 2012 01:22:20 +0000 (18:22 -0700)]
meta/classes/sstate.bbclass: ensure sstate files have group R/W

In order to make sharing sstate files easier chmod them with 0664 permissions so that they are readable and writable by the user and any other members of  their group.

Signed-off-by: Joshua Lock <josh@linux.intel.com>
12 years agokernel.bbclass: add deploy link to KERNEL_IMAGETYPE
Christopher Larson [Wed, 16 May 2012 01:25:23 +0000 (20:25 -0500)]
kernel.bbclass: add deploy link to KERNEL_IMAGETYPE

It's common to provide a non-machine-suffixed link in DEPLOY_DIR_IMAGE, so
let's be consistent and do so here as well.

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
12 years agooe.types: give the regex type more sane semantics
Christopher Larson [Wed, 16 May 2012 01:26:25 +0000 (20:26 -0500)]
oe.types: give the regex type more sane semantics

Currently, if a variable is unset or has an empty value, the regex type
will return a match object which always matches. Not all variable types
will necessarily have the same behavior for handling defaults. I believe
that returning a match object which matches nothing when a variable is
unset is superior to returning one which matches anything, and the user
can always explicitly request anything via '.*', if that's what they
want.

This constructs a null pattern object which will never match, and uses
it when encountering an unset or empty variable (currently, these two
things are one and the same, as maketype is handling the default. we may
well want to shift that logic into the individual types, giving them
more control over default behavior, but currently the behavior is at
least relatively consistent -- no difference between unset and empty
variables).

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
12 years agocsl-versions: fix bb.process.CmdError reference
Christopher Larson [Wed, 16 May 2012 01:28:24 +0000 (20:28 -0500)]
csl-versions: fix bb.process.CmdError reference

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
12 years agoxinetd: Update to 2.3.15
Saul Wold [Wed, 16 May 2012 19:39:31 +0000 (22:39 +0300)]
xinetd: Update to 2.3.15

Signed-off-by: Saul Wold <sgw@linux.intel.com>
12 years agofotowall: Add #include ui_wizard.h to ExportWizard.cpp
Andrei Gherzan [Thu, 10 May 2012 21:06:16 +0000 (00:06 +0300)]
fotowall: Add #include ui_wizard.h to ExportWizard.cpp

App/ExportWizard.cpp depends on wizard.h which depends on ui_wizard. The last one
should be already generated before compiling ExportWizard.cpp.

[YOCTO #2297]

Signed-off-by: Andrei Gherzan <andrei@gherzan.ro>
12 years agoexpat: update to upstream version 2.1.0
Marko Lindqvist [Thu, 10 May 2012 21:43:51 +0000 (00:43 +0300)]
expat: update to upstream version 2.1.0

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
12 years agoclasses/mirrors.bbclass: Point snapshot.debian.org mirror to working location
Khem Raj [Fri, 4 May 2012 21:56:33 +0000 (14:56 -0700)]
classes/mirrors.bbclass: Point snapshot.debian.org mirror to working location

If you point to snapshot.debian.net/archive/pool then it will fetch
you a html page which will end up in corrupt download. The locations
have changed for archives and here we point the mirror to right
location.

Signed-off-by: Khem Raj <raj.khem@gmail.com>
12 years agogdk-pixbuf: Fix lsb builds where dependency may be missing
Richard Purdie [Tue, 15 May 2012 18:43:37 +0000 (19:43 +0100)]
gdk-pixbuf: Fix lsb builds where dependency may be missing

We turn on the xlib depdendency for linuxstdbase builds but
were missing the dependency on libx11. This adds the missing
dependency.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agotcmode-external-csl.inc: specify the path you couldn't find
Peter Seebach [Tue, 15 May 2012 18:16:35 +0000 (13:16 -0500)]
tcmode-external-csl.inc: specify the path you couldn't find

While misconfiguring toolchains, I noticed that one of the
diagnostic messages gave me a little less information than I'd
like about what I'd done wrong.  Displaying the glob pattern
that couldn't be matched turns out to make it a lot easier to
figure out what you did wrong.  (Answer: Not enough coffee.)

Signed-off-by: Peter Seebach <peter.seebach@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agotcmode-external-csl.inc: Allow for overrides
Peter Seebach [Tue, 15 May 2012 18:16:34 +0000 (13:16 -0500)]
tcmode-external-csl.inc: Allow for overrides

Wind River uses binary toolchains provided by Code Sourcery, but which
have different values for a couple of the preset variables than the
generic toolchains the external-csl toolchain feature supports.  If
these values were ?= assignments, we could just assign values to them
and share the code, rather than keeping modified copies.

Signed-off-by: Peter Seebach <peter.seebach@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agosstate.bbclass: Optimize the generation and install path fixups
Mark Hatle [Fri, 11 May 2012 17:17:54 +0000 (12:17 -0500)]
sstate.bbclass: Optimize the generation and install path fixups

The fixmepath file that is generated contains a list of all of the files
that need their paths fixed.  In the previous version the fixmepath was
generated to include all of the files that sed may have changed.  In the
new version, we first grep the files to see if they contain a path that
needs to be changed, only then do we perform the sed operation on those files.

This results in a modest performance increate in the creation of the sstate
file.  The following numbers include the do_package and do_populate_sysroot
tasks on the perl recipe.

Before the change:
real    4m23.018s
user    1m57.067s
sys     1m33.327s

After the change:
real    4m13.083s
user    1m54.062s
sys     1m26.064s

However, a more significnt performance gain is felt during the
extraction/install of sstate cache files, as the fixmepaths file now has a
significantly smaller list of files to modify.

Before the change:
real    0m39.798s
user    0m11.158s
sys     0m12.642s

After the change:
real    0m25.511s
user    0m8.408s
sys     0m5.077s

(All numbers above were recorded with a cold filesystem cache on a machine
with 12 GB of ram.)

Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>