site.conf.sample: Fix broken SOCKS proxy setup and configuration
SOCKS proxy specification with git was using conflicting methods and
thus was failing when mixed SOCKS needs were in place (requiring no
proxy for some hosts and proxy for the rest)
- GIT_PROXY_COMMAND is an environment variable GIT uses to OVERRIDE
all proxy configuration in ~/.gitconfig or any other gitconfig. By
using it to configure, it was breaking havoc on site git
configuration or the one generated by bitbake in tmp/.
Renamed to OE_GIT_PROXY_COMMAND in meta/conf/site.conf.sample
(with a doc tidbit on the name chosen), meta/classes/base.bbclass.
- The gitconfig generated by bitbake was wrong. There was a typo error
(gitproxy vs gitProxy), thus all lines were being ignored. Fixed in
meta/classes/base.bbclass.
- The gitconfig generated was being placed in
${STAGING_DIR_NATIVE}/usr/etc/gitconfig; git was looking for it in
${STAGING_DIR_NATIVE}/etc/gitconfig. Fixed that in
meta/classes/base.bbclass, at the same time creating a
GIT_CONFIG_PATH variable, since it is also referenced in
generate_git_config() and have all instances refer to that.
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Thu, 1 Mar 2012 15:54:46 +0000 (15:54 +0000)]
toolchain-scripts: Exclude variables causing the recipe to become unecessarily machine specific
basehash changed from 30e97f9eeed1df8488b62b4fb47a3a0c to 72ea9a277e6599ae9052b169b9a94f1b
Variable TOOLCHAIN_CONFIGSITE_SYSROOTCACHE value changed from
/OE/oe-core/tmp-eglibc/sysroots/qemux86-64/usr/share/x86_64-oe-linux_config_site.d
to
/OE/oe-core/tmp-eglibc/sysroots/qemux86-64copy/usr/share/x86_64-oe-linux_config_site.d
Variable TIME value changed from 084543 to 085638
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Saul Wold [Thu, 1 Mar 2012 04:37:04 +0000 (20:37 -0800)]
netbase: fix QA Warnings
WARNING: For recipe netbase, the following files/directories were installed but not shipped in any package:
WARNING: /usr/sbin Signed-off-by: Saul Wold <sgw@linux.intel.com>
Saul Wold [Thu, 1 Mar 2012 04:36:08 +0000 (20:36 -0800)]
perl: remove empty directories to clean QA Warnings
WARNING: For recipe perl, the following files/directories were installed but not shipped in any package:
WARNING: /usr/lib/perl/site_perl
WARNING: /usr/lib/perl/site_perl/5.14.2 Signed-off-by: Saul Wold <sgw@linux.intel.com>
Saul Wold [Thu, 1 Mar 2012 07:46:21 +0000 (23:46 -0800)]
image-vmdk: Create image_vmdk class and setup image and image_types to use it
This creates a new image_vmdk class similar to live. The image_vmdk
class needs to have a hddimg created by the image-live class, so it
inherits it directly.
The changes to image_types is to ensure that both live and vmdk images
get the ext3 tools and dependencies.
Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Thu, 1 Mar 2012 02:27:07 +0000 (21:27 -0500)]
linux-libc-headers-yocto: bump to v3.2.8
Making two changes for BSPs that prefer linux-libc-headers-yocto:
- bumping to v3.2.8 headers
- stubbing out unecessary linux-yocto functions when headers
are being built
[YOCTO #2032]
Acked-by: Darren Hart <dvhart@linux.intel.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 29 Feb 2012 16:24:26 +0000 (16:24 +0000)]
image_types: Correctness fixes
* Add a newline to improve the output formatting
* Use set() to turn the list into a set of unique items to prevnt
the same image code running twice (for e.g. IMAGE_FSTYPES = "tar.gz tar.bz2")
* Support multiple compression extensions such as ".gz.u-boot"
* Fix basetype/type typo and fix multiple image generation
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
James Limbouris [Wed, 29 Feb 2012 06:56:30 +0000 (14:56 +0800)]
busybox: Enable 64 bit shell tests, and disable non-standard date format interpretation.
Without 64 bit shell tests, the timestamp comparison in initscripts' bootmisc.sh throws 'out of range' errors.
With CONFIG_FEATURE_DATE_COMPAT, date reads the YYYYMMddhhmm string as hhmmYYYYMMdd when setting the time from the timestamp in bootmisc.sh.
Signed-off-by: James Limbouris <james@digitalmatter.com.au> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Mon, 27 Feb 2012 12:43:49 +0000 (12:43 +0000)]
image_types: Refactor compression code into a generic compression solution
The current approach of adding each different compressed image type doesn't
scale. This patch changes the code so compressed images for each form are
automatically available using the form <type>.<compression type> in
IMAGE_FSTYPES. This doesn't change any existing externally visible behaviour
and the image generation process becomes more efficient as a result too.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Wenzong Fan [Tue, 14 Feb 2012 01:42:59 +0000 (09:42 +0800)]
gettext: Get gettext use its own libunistring
According to its documentation, gettext 0.18.1.1 needs either
libunistring or it will use its own internal version. Just add
a configure option '--with-included-libunistring' to force the
latter to avoid races.
[YOCTO#1934]
Signed-off-by: Wenzong Fan <wenzong.fan@windriver.com>
James Limbouris [Tue, 28 Feb 2012 03:49:06 +0000 (11:49 +0800)]
mtd-utils: Don't use host 'ranlib' or 'ar'.
If ranlib, ar, or cc are not supplied to make, it attempts to use ${CROSS}xxx.
Since ${CROSS} is not set, host ranlib may run, producing an unusable .a file on some platforms.
Signed-off-by: James Limbouris <james@digitalmatter.com.au>
multilib.bbclass: allow TARGET_VENDOR_virtclass-multilib to be overriden
If we set this bit, we can override the ugly "pokymllib32" to back to
"poky" (powerpc-pokymllib32-linux-gcc -> powerpc-poky-linux-gcc). I've
left this unset by default, but can be set by adding the following:
TARGET_VENDOR_virtclass-multilib-lib32 = "-poky"
Signed-off-by: Matthew McClintock <msm@freescale.com>
tune-ppce5500: consolidate ppce5500 and ppc64e5500 into one tune file
We don't need two files for this. Also this fixes some mutlilib build
issues where we were not able to select the multilib arch to be
ppce5500 or ppc64e5500.
Changes recently made to meta-fsl-ppc layer depend on this change as
well
Signed-off-by: Matthew McClintock <msm@freescale.com>
Bruce Ashfield [Mon, 27 Feb 2012 18:20:41 +0000 (13:20 -0500)]
linux-yocto: allow non-branched repositories to check out
Not all users of the checkout phase of linux-yocto have all
branches present. This is normal, and should be supported. By
checking for an empty KBRANCH we can avoid validating a branch
that isn't supposed to exist.
[YOCTO #2032]
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Paul Eggleton [Thu, 9 Feb 2012 17:38:21 +0000 (17:38 +0000)]
qt4-tools-nativesdk: improve and tidy up qmake build
* Use the same method to build qmake as in the qt4 target recipes
* We always have qmake from qt4-native so don't check if it exists -
just skip the check like we do in the qt4 target recipes. This
saves us a patch.
* Replace qt4-tools-native in DEPENDS with qt4-native
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Paul Eggleton [Thu, 9 Feb 2012 14:37:34 +0000 (14:37 +0000)]
qt4-tools-nativesdk: fix build of 4.7.4
Commit 4ccae37db1aa77a1d15098c3720ea6e2d383fbdc introduced a sed command
line to modify a file that only exists in 4.8.0, thus do_configure of
4.7.4 was failing; so make this conditional upon the file existing.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Paul Eggleton [Wed, 8 Feb 2012 17:00:25 +0000 (17:00 +0000)]
qt4: build qmake for the target
qmake was not being built at all (since we had the host version from
qt4-native) and the default is to build for the host machine within
configure. With a minor hack we can build qmake for the target as well,
which is useful if you want to build Qt 4 software on an actual device.
This is now installed as part of the tools package (together with uic,
moc etc.)
Additionally, add an environment setup script (installed in
/usr/share/qt4/environment-setup or /usr/share/qtopia/environment-setup
for the embedded version) as part of the mkspecs package since the
mkspecs will not work without the paths set up by this script. The code
to create this script was adapted from the qmake2 recipe in meta-oe.
Also move do_compile() from qt-${PV}.inc up to qt4.inc as both versions
have the exact same content.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Yi Zhao [Fri, 24 Feb 2012 06:51:28 +0000 (14:51 +0800)]
lsbtest: Add recipe for LSB tests and automate test
The recipe is used for LSB tests. The script LSB_Test.sh does the following things:
- setup LSB testing environment
- download LSB rpm packages with list file packages_list from remote
- install the packages
- execute LSB testing with profile file session
- collect the results
Install packages_list and session files into ${D}/opt/lsb-test.
Richard Purdie [Mon, 27 Feb 2012 12:16:31 +0000 (12:16 +0000)]
glib-2.0: Disable fam, we don't depend on it
Various people are reporting glib-2.0 failures from a dependency on
libfam creeping in uninvited. This patch explicitly disables it to
favour a deterministic build.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Eric Bénard [Mon, 20 Feb 2012 07:57:33 +0000 (08:57 +0100)]
pulseaudio: Really disable orc support
If orc is detected the compilation fails with missing headers which
should have been generated by orc (using meta-angstrom + meta-openembedded
and and armv7 target)
[RP: add RP bump, fix typos] Signed-off-by: Eric Bénard <eric@eukrea.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Saul Wold [Fri, 24 Feb 2012 18:59:20 +0000 (10:59 -0800)]
gnupg: Update checksum, fix configure and compliation issues
This recipe was added, but did not have the correct
checksum information for the LIC_FILES_CHKSUM or the
SRC_URI. Also disable Documetnation for now due to
older autotools issue.
[YOCTO #1966]
Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
automake: omit compilation of pyc files on install
* On install, automake calls py-compile, which previously
compiled python source code to pyc and pyo, which both
got packaged.
* The python interpreter in OE contains patches to enable
optimization (pyo) by default:
04-default-is-optimized.patch
99-ignore-optimization-flag.patch
* automake created pyc files by calling py_compile.compile()
and adding the c suffix manually, resulting in identical
byte code for both pyc and pyo files.
* py-compile-compile-only-optimized-byte-code.patch
applies to automake 1.11 and automake master, but older
versions require a slightly modified patch. However,
older versions are only pinned by chinook-compat and
nylon, so I left them untouched.
Signed-off-by: Andreas Oberritter <obi@opendreambox.org> Signed-off-by: Khem Raj <raj.khem@gmail.com>
The patch was imported from the OpenEmbedded git server
(git://git.openembedded.org/openembedded) as of commit id aa4585c5065e05c759f16e1e8623fc7f40640f1b.
Modified to apply to automake version 1.11.2 and to
include a patch header. Also renamed the patch.
Signed-off-by: Andreas Oberritter <obi@opendreambox.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Otavio Salvador [Fri, 24 Feb 2012 20:50:53 +0000 (20:50 +0000)]
qt4-native: set headerdir as done in regular builds
qmake uses a generated qt.conf file that expects headers to be found
at ${includedir}/qt4 but qt4-native were not following this layout
breaking the builds of native binaries that use qt4.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Eric BENARD [Sun, 26 Feb 2012 07:49:42 +0000 (08:49 +0100)]
gdb-cross-canadian: build gdb with python support
* python support is needed for providing finer scripting control
in gdb, and for remote controling gdb for example from qtcreator
* gdb/configure tries to autodetect python using python executable
with --includes --ldflags --exec-prefix to get the right flags
As it's difficult to achieve in OE's context, we generate a
script which will return the right values and make gdb's build
system happy. This idea was taken from the following article :
http://www.mentby.com/doug-evans-2/python-enabled-gdb-on-windows-and-relocation.html
* tested using angstrom & armv7 target & qtcreator 201005 & 2.4.0
* MJ: updated paths to python as disscussed here
http://lists.linuxtogo.org/pipermail/openembedded-core/2012-February/018222.html
now it's easier with cross-canadian staging dir and SDKPATH fixed :)
Signed-off-by: Eric Bénard <eric@eukrea.com> Cc: fcooper@ti.com Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Martin Jansa [Sun, 26 Feb 2012 07:49:41 +0000 (08:49 +0100)]
bitbake.conf: remove TARGET_ARCH from in SDKPATH
* SDKPATH shouldn't depend on TARGET_ARCH as discused here:
http://lists.linuxtogo.org/pipermail/openembedded-core/2012-February/018222.html
* introduce SDK_NAME_PREFIX so that distributions can overwrite only
this instead of whole SDK_NAME
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sat, 25 Feb 2012 15:53:08 +0000 (15:53 +0000)]
cross-canadian: Set STAGING_DIR_HOST correctly
As reported by Martin Jansa, the path to nativesdk sysroot was changing between
nativesdk and cross-canadian recipes. The problem was the incorrect deinfition of
STAGING_DIR_HOST in cross-canadian.bbclass.
Since nothing really uses the cross-canadian output in the sysroot, only the
packages, its not surprising this bug has gone un-noticed.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
license.bbclass: Symbolic links of generic license
This is to reduce the size of licenses added to images. With this
commit license.manifest, original license and generic license
adds about .5M to a core-image-minimal image, substantially less
than what is currently occuring when COPY_LIC_MANIFEST and
COPY_LIC_DIRS are set.
Signed-off-by: Elizabeth Flanagan <elizabeth.flanagan@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We should look for LICENSE at a package level first. If it's
not found, we should use the recipe level LICENSE. This adds a
bit more granularity to license manifests where needed.
Signed-off-by: Elizabeth Flanagan <elizabeth.flanagan@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Fri, 24 Feb 2012 01:58:47 +0000 (20:58 -0500)]
linux-yocto: improve checkout error handling and reporting
The typical workflow for linux-yocto simply uses a remote
upstream repository (Whether it is mirrored or not), and in this
case there are no issues with consistency in the format of the
resository that is unpacked into the WORKDIR.
When working with a local linux-yocto repository for kernel
development the remote vs local branches is not always consistent
between repositories.
The suggested/documented workflow has always been to use a
bare clone of linux-yocto, and use a second working tree repository
for development. Changes flow from the working tree to the bare
clone and then into the working directory for build. A common
mistake that happens with this workflow is that the non-bare,
working repository is used instead of the bare clone version.
If a non-bare repository is reference by the SRC_URI, then the
branches that are fetched into WORKDIR are not consitent. If the
MACHINE and META branches are not present, cryptic build errors
will result.
To solve this problem, the checkout code has been changed in
several ways:
- works with a newly proposed 'bareclone' option to bitbake
- detects if a bareclone is present in WORKDIR or not and
adjustst the checkout accordingly.
- if a non-bare clone is detected, machine and meta branches
are checked. If they are not present, or can't be created
a clear error message is produced
- instead of manipulating the refs directly in the git tree,
local tracking branches are (quietly) created for remote
branches. Enabling a better workflow in the WORKDIR kernel
repository.
This has been tested with linux-yocto remote upstreams, local
bare and non-bare respositories. All builds succeed or fail
with clear error messages.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Martin Jansa [Thu, 23 Feb 2012 14:54:14 +0000 (15:54 +0100)]
sstate.bbclass: improve performance of sstate package creation
* also fixes replacing paths for perl where cmd line was probably
too long for os.system(cmd) (it had 560410 characters because a lot of
files from sstate_scan_cmd).
* also print those 2 commands so we can find them in log.do_package
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
