Lei Liu [Thu, 21 Nov 2013 06:54:00 +0000 (00:54 -0600)]
base.bbclass: Fix incorrect setting of multilib PREFERRED_PROVIDER_virtual_pkg
PREFERRED_PROVIDER_virtual_pkg has been incorrectly set with more
than one multilib prefixes. For example, if we have two alternative
multilibs lib64 and lib32, PREFERRED_PROVIDER_virtual_pkg will be
set to lib32-lib64-pkg or lib64-lib32-pkg, depending on which
multilib shows up first in the list.
Signed-off-by: Lei Liu <lei.liu2@windriver.com> Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Fri, 22 Nov 2013 09:24:04 +0000 (09:24 +0000)]
ethtool: Fix ptest compile
buildtest-TESTS is a phony target and does nothing which results in a
do_install error since the tests aren't built. Since there isn't
a suitable make target but the number of tests are small, hardcode
the two to build to unbreak the build when ptest is enabled.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 20 Nov 2013 14:44:11 +0000 (14:44 +0000)]
buildhistory/distrodata: Adapt to updated fetcher API
The buildhistory code has fallback paths for older bitbakes for now. The
distrodata class is much less used and it can be assumed a recent bitbake
is used in that case rather than adding fallback code.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Thu, 21 Nov 2013 14:57:46 +0000 (14:57 +0000)]
oeqa/connman: disable unique test
The connman unique test starts another instance of connmand and then does a ps to
verify that there's only one of these running, on the assumption that the new
one has quit because there's already one running (started by init).
However, connmand is forking into the background straight away so there's a race
between running ps and the second connmand discovering the first and exiting.
This race can be seen because the test displays the output of ps, and by the
time that second ps has been executed the new connmand has exited.
This is a classic race condition and on a heavily loaded autobuilder inserting
an arbitrary sleep isn't wise. In the scheme of things this test isn't very
useful, so delete it.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Thu, 21 Nov 2013 15:25:36 +0000 (15:25 +0000)]
ia32-base: Remove cpio and ext3 defaults
On real IA hardware, neither the ext3 or cpio images are particularly useful
or used. cpio is legacy from initramfs and that specific image now overrides
FSTYPES accordingly. The size difference in filesystems makes ext3 as a file
format less useful, mainly being useful in the qemu case.
When needed users can still override the default FSTYPES so having
saner defaults makes sense. This improves build times and uses less
network bandwidth for builds and releases.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Qiang Chen [Mon, 18 Nov 2013 02:18:32 +0000 (10:18 +0800)]
util-linux: collect libs for related libxx-dev package
Commit 91c0f9e3 moves library files to base_libdir for safe while
symbol link libblkid.so, libmount.so and libuuid.so are placed in
libdir, not base_libdir.
Meanwhile, libblkid.la, libmount.la and libuuid.la are placed in
libdir too.
Thus they are missed by related develop package and
collected by util-linux's develop package at last. Fix it.
Chen Qi [Mon, 18 Nov 2013 07:20:44 +0000 (15:20 +0800)]
license.bbclass: fix missing of license files on ubuntu build host
The license_create_manifest function contains bashism, this will lead
to unexpected results on ubuntu build host, as sh is linked to dash on
ubuntu. Even if COPY_LIC_MANIFEST and COPY_LIC_DIRS are enabled, the
license files will still be missing on target.
Jack Mitchell [Mon, 18 Nov 2013 16:14:25 +0000 (16:14 +0000)]
libjson: update to 0.11 and rename to json-c
libjson is now known as json-c. Config.status is removed as it breaks
seperate build dir builds. Built without parallel make as it fails,
official word is not to bother trying.
Signed-off-by: Jack Mitchell <jmitchell@cbnl.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Ming Liu [Wed, 13 Nov 2013 05:40:59 +0000 (13:40 +0800)]
kernel.bbclass: move bundle_initramfs after kernel_link_vmlinux
${KERNEL_OUTPUT} is being renamed/restored in bundle_initramfs task, so we
must ensure bundle_initramfs run after kernel_link_vmlinux where the link
of vmlinux is created as the bootable image.
Signed-off-by: Ming Liu <ming.liu@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Phil Blundell [Thu, 14 Nov 2013 18:15:46 +0000 (18:15 +0000)]
bitbake.conf: Remove obsolete/unused MIRROR cruft
ADOBE_MIRROR, HANDHELDS_CVS and E_SVN were broken links and not used by
any recipe in oe-core.
FREEDESKTOP_CVS is no longer useful because all the source code that
matters is in git; no recipe in oe-core still uses the CVS repository.
E_MIRROR, FREEBSD_MIRROR, FREESMARTPHONE_GIT still point to valid-seeming
locations but there are no recipes in oe-core that use them. Any layers
which need these variables can define them for themselves.
GPE_SVN, GPE_EXTRA_SVN, GPEPHONE_MIRROR and GPEPHONE_SVN are not used by
any recipe in oe-core and the corresponding projects seem to be mostly
dead upstream. Again, any layers which still wish to use these variables
can define them locally.
All the above are just wasting space in bitbake's datastore and would be
better deleted.
Signed-off-by: Phil Blundell <philb@gnu.org> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Ming Liu [Wed, 13 Nov 2013 08:15:47 +0000 (16:15 +0800)]
qemu: explicitly disable xen support
We don't make use of xen and when building on Ubuntu 13.04 when
libxen-dev is installed on the build host you will get errors like the
following:
| /usr/include/x86_64-linux-gnu/bits/string3.h:81: warning: memset used with constant zero length parameter; this could be due to transposed parameters
| /usr/lib/gcc/x86_64-linux-gnu/4.7/../../../../lib/libxenguest.so: undefined reference to `lzma_alone_decoder@XZ_5.0'
| /usr/lib/gcc/x86_64-linux-gnu/4.7/../../../../lib/libxenguest.so: undefined reference to `lzma_code@XZ_5.0'
| /usr/lib/gcc/x86_64-linux-gnu/4.7/../../../../lib/libxenguest.so: undefined reference to `lzma_stream_decoder@XZ_5.0'
| /usr/lib/gcc/x86_64-linux-gnu/4.7/../../../../lib/libxenguest.so: undefined reference to `lzma_end@XZ_5.0'
This change disables xen for both -native and target packages but
since it is a PACKAGECONFIG a user could tune this to have xen support
in the target package.
Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com> Signed-off-by: Ming Liu <ming.liu@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Hans Beckerus [Fri, 15 Nov 2013 14:25:33 +0000 (15:25 +0100)]
initscripts: add missing dmesg.sh to run-level S
In commit faa8cc6c2a582a32c695f3f2b0d45b6892c769fd dmesg.sh was
added to the set of init.d scripts. But the script was never put
in any run-level. This patch will add dmesg.sh to run-level S.
Signed-off-by: Hans Beckerus <hans.beckerus AT gmail.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Phil Blundell [Fri, 15 Nov 2013 13:31:01 +0000 (13:31 +0000)]
kernel.bbclass: Use lzma not lzmash for decompressing
If kernel_do_compile() encounters a .lzma image for the initramfs, it
attempts to decompress it using the little-known "lzmash" utility. This
may not be present, which will cause the build to fail.
Fortunately, it appears that the more mainstream "lzma" tool is
compatible so we can use that one instead.
Signed-off-by: Phil Blundell <philb@gnu.org> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Tobias Henkel [Fri, 15 Nov 2013 07:36:41 +0000 (08:36 +0100)]
icecc: Use bb.utils.which instead of calling which
The icecc class often calls 'which' for determining paths. This leads
to many messages on stderr in case 'which' doesn't find the
executable. Using bb.utils.which is more appropriate here and doesn't
pollute stderr.
Irina Patru [Thu, 14 Nov 2013 08:54:19 +0000 (10:54 +0200)]
distrodata.bbclass: Add fetch2 handlers to svn case in checkpkg
The existing code in checkpkg from distrodata.bbclass had similar
functionality with fetch when searching for latest package version.
For packages that use svn protocol that part was rewrote in order
to use fetcher API.
It now calls latest_revision method from Svn class in fetch2 so
that it gets latest version.
[ YOCTO #1813 ]
Signed-off-by: Irina Patru <irina.patru@intel.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Phil Blundell [Thu, 14 Nov 2013 14:41:21 +0000 (14:41 +0000)]
boost: Add patch to avoid undefined references to boost::atomic::lockpool::get_lock_for()
Boost::thread uses functions from boost::atomic but doesn't actually
link with libboost_atomic. This works fine on platforms where
BOOST_ATOMIC_FLAG_LOCK_FREE is true but will lead to undefined
symbol references otherwise. Fix this by applying a patch from
the upstream bug tracker to add the missing library linkage.
Signed-off-by: Phil Blundell <pb@pbcl.net> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Lu Chong [Thu, 14 Nov 2013 02:47:58 +0000 (10:47 +0800)]
util-linux: Fix QA warnings about ALTERNATIVES
Previously, EXTRA_OECONF would be overwritten by EXTRA_OECONF_class-native
and EXTRA_OECONF_class-nativesdk.
As a result, some applications such as `last', `mesg' and `reset' didn't get
compiled.
The patch rewrote EXTRA_OECONF_class-native, EXTRA_OECONF_class-nativesdk
and EXTRA_OECONF variables to fix QA warnings.
usr/share/man/man1/last.1.util-linux does not exist
usr/share/man/man1/mesg.1.util-linux does not exist
usr/bin/last.util-linux does not exist
usr/bin/mesg.util-linux does not exist
bin/reset does not exist
Signed-off-by: Chong Lu <Chong.Lu@windriver.com>
[YOCTO #5338] Signed-off-by: Saul Wold <sgw@linux.intel.com>
Wenzong Fan [Thu, 14 Nov 2013 02:45:48 +0000 (21:45 -0500)]
udev: remove extra -dev/-dbg packages
We don't support multiple -dbg/-dev packages, the package can generate
them but the system does not correctly handle them. Just move all devel
stuffs into 'udev-dev' and all debug stuffs into 'udev-dbg'.
Signed-off-by: Wenzong Fan <wenzong.fan@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Hongxu Jia [Wed, 13 Nov 2013 08:46:07 +0000 (16:46 +0800)]
lib32-qt4-examples: fix QA warnings and INSANE_SKIP not work
Fix the following warnings:
...
lib32-qt4-examples: found library in wrong location: /usr/bin/qt4/examples/tools/echoplugin/plugin/libechoplugin.so
lib32-qt4-examples: found library in wrong location: /usr/bin/qt4/examples/tools/plugandpaint/plugins/libpnp_extrafilters.so
...
In oe-core f119566477243ce43b727492dc78b9cb3dd76de4, it added INSANE_SKIP
libdir for examples packages as it includes plugin shared libraries outside
of libdir.
But it hardcoded ${QT_BASE_NAME} as INSANE_SKIP's package name
and it didn't work while multilib enabled.
Add ${MLPREFIX} as package name's prefix in INSANE_SKIP
Laurentiu Palcu [Wed, 6 Nov 2013 13:07:58 +0000 (15:07 +0200)]
flex: fix m4 issue on target
Flex needs m4 to run (see below) and, since the create_wrapper
introduces a bash dependency on target, give the path to m4 binary in
the configure command line.
Snippet from the flex documentation:
"The macro processor m4 must be installed wherever flex is installed.
<...>
m4 is only required at the time you run flex."
Khem Raj [Tue, 12 Nov 2013 04:15:53 +0000 (20:15 -0800)]
libnl: Fix random segfaults due to memory corruption
This is a backport from upstream fixes a severe problem
w.r.t memory management, where it would result in random
segfaults in applications depending on libnl
Khem Raj [Tue, 12 Nov 2013 04:07:15 +0000 (20:07 -0800)]
tcl: Install header into 8.6 instead of PN-PV in user/include
This helps in compiling other programs like expect which
depend on private headers but 8.5, 8.6 and so on is enough
granularity and currently we had 8.6.x and so on which
means that expect recipe will need to be touched whenever there
is minor update of tcl.
Additionally the encode creating symlink to shared object in
patch and remove it from recipe
Refresh patches after making changes to Configure.in we
propertly generate configure and not patch is directly as
was the case.
Chunrong Guo [Tue, 12 Nov 2013 03:40:49 +0000 (11:40 +0800)]
mdadm: flag __SANE_USERSPACE_TYPES__ to include int-ll64.h for powerpc64
*PPC64 uses long long for u64 in the kernel, but powerpc's asm/types.h
prevents 64-bit userland from seeing this definition, instead defaulting
to u64 == long in userspace.
*fix the below error
|super-ddf.c:4542:5: error: format '%llu' expects argument of type 'long long unsigned int',
|but argument 5 has type '__u64' [-Werror=format=]
|dprintf("BVD %u has %08x at %llu\n", 0,
Hongxu Jia [Mon, 11 Nov 2013 11:56:54 +0000 (19:56 +0800)]
kernel-grub.bbclass: support /boot area within root partition
Previously, it supported the situation that /boot area with separate boot
partition:
...
menuentry "Update bzImage-3.10.10-WR6.0.0.0_standard-3.10" {
set root=(hd0,1)
linux /bzImage-3.10.10-WR6.0.0.0_standard root=/dev/sdb1 rw ip=dhcp
}
...
But didn't consider the situation that /boot within root partition:
...
menuentry "Update bzImage-3.10.10-WR6.0.0.0_standard-3.10" {
set root=(hd0,1)
linux /boot/bzImage-3.10.10-WR6.0.0.0_standard root=/dev/sdb1 rw ip=dhcp
}
...
Chen Qi [Mon, 11 Nov 2013 09:26:09 +0000 (17:26 +0800)]
sysvinit: fix problem in switching runlevels
Previously, if we switch to runlevel 1 and then switch back to runlevel
5, the network interface will be brought down and the NFS service will
not be restarted correctly.
The problem is that the networking and rpcbind services are brought down
in runlevel 1 but not brought up in runlevel 5.
This patch fixes the above problem. It's based on the assumption that
in sysvinit-based system, runlevel 1 does not have networking support.
This patch adjusts some init script parameters used by update-rc.d. It
makes sure that networking starts before rpcbind which in turn starts
before mountnfs.sh. When switching to runlevel 0, 1 and 6, the umountnfs.sh
is run first before stopping rpcbind service, and the network is brought
down afterwards.
Zhangle Yang [Mon, 18 Nov 2013 12:54:55 +0000 (20:54 +0800)]
python: Fix ptest not working problems
The run-ptest contained in python uses a Makefile. The path of the
Makefile is changed to ${libdir}/python/ptest in do_install_ptest.
However, the directory is wrong when the project is configured with
"--enable-multilib=lib32"
In addition, do_install_ptest is defined before "inherit ptest", so
it is overriden by the one in ptest.bbclass. do_install_ptest is
moved down.
Signed-off-by: Zhangle Yang <zhangle.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Saul Wold [Fri, 15 Nov 2013 01:52:13 +0000 (17:52 -0800)]
sysvinit: unmount the psplash lazyily
There is an race condition where psplash is not quite exited before the unmount occurs
causing a umount: /mnt/.psplash: target is busy message to appear, it's ok to lazyily
unmount and not get this message
[YOCTO #5244]
Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With this, one second ticks are only enabled if the width of a second is
five pixels or more. It is also possible to distinguish 1, 5 and 30
second ticks.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
pybootchartgui: Correct the X offset for the chart
This will make the first bar actually start within the graph. It will
also move the graph to the right so the names of the first tasks are
more likely to be visible.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This update the pybootchartgui code to the latest release from its new
location at "https://github.com/mmeeks/bootchart". This only imports
the relevant parts, and not all of bootchart2.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Sun, 17 Nov 2013 12:28:19 +0000 (12:28 +0000)]
buildhistory_analysis: fix error when comparing image contents
OE-Core commit b7de1eaac9eed559b2d68058f5de67de74a6cb58 added an extra
argument to the compare_dict_blobs() function but missed adding the
argument to one call to compare two versions of the image-info.txt file.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Sat, 16 Nov 2013 06:17:24 +0000 (01:17 -0500)]
kernel.bbclass: fix hardlink test
The commit [kernel.bbclass: test for hardlinks before installing] doesn't
work on all build machines without a space between the ! and the test for
hardlink equivalance. The build continues, while the test fails and no
copy is made into the sysroot. Eventually tasks like build_hdimg will fail
with the missing kernel with the root cause being less than obvious.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Thu, 14 Nov 2013 16:55:26 +0000 (11:55 -0500)]
kernel.bbclass: test for hardlinks before installing
commit df564c4 [kernel: Use hardlinks for do_populate_sysroot for speed]
does indeed speed up the build, but it also means that the KERNEL_OUPUT
and KERNEL_IMAGETYPE may be hardlinks.
If they are hardlinks, install complains that they are actually the same
file and stops the build. We can easily test and avoid the copy if the
file is already in place.
[YOCTO #5527]
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Wed, 13 Nov 2013 16:40:10 +0000 (16:40 +0000)]
linux-yocto/3.10: meta: ARM: OMAP3: Add USB PHY driver for Beagleboard
Updating the meta branch SRCREV to update the USB configuration:
The Beagleboard needs the USB PHY drivers in the kernel in order to enable
USB and Ethernet functionality. This fix ensures that they are built in
by tweaking the kernel config.
Tested on Beagleboard xM Rev. C2.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Thu, 14 Nov 2013 16:55:24 +0000 (11:55 -0500)]
linux-yocto/3.10: meta: ARM: OMAP3: Add USB PHY driver for Beagleboard
Updating the meta branch SRCREV to update the USB configuration:
The Beagleboard needs the USB PHY drivers in the kernel in order to enable
USB and Ethernet functionality. This fix ensures that they are built in
by tweaking the kernel config.
Tested on Beagleboard xM Rev. C2.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Wed, 13 Nov 2013 14:32:44 +0000 (14:32 +0000)]
classes/license: use "1" to set boolean variables
The expected usage of COPY_LIC_MANIFEST and COPY_LIC_DIRS appears to be
to set them to "1" to enable; however the test here is just testing
whether they have a value at all, so setting them to "0" would also
enable them which is somewhat disingenuous. Actually check if they are
set to "1" instead in order to fix this.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Wed, 13 Nov 2013 14:32:48 +0000 (14:32 +0000)]
classes/pkg_metainfo: remove
This is very old, and we now have pkgdata which is much more complete.
Nobody appears to be using this class, and even OE-Classic had no
current references to it.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Wed, 13 Nov 2013 14:32:47 +0000 (14:32 +0000)]
classes/pkg_distribute: remove
This appears to be a very old way of archiving recipes and associated
files. We have better ways of doing this now, and nobody appears to be
using this class - even OE-Classic had no current references to it.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Wed, 13 Nov 2013 14:32:39 +0000 (14:32 +0000)]
classes/insane: improve LIC_FILES_CHKSUM messages
* If the md5 parameter is specified with no value, report that the
checksum is not specified instead of reporting that it has changed
* If the md5 checksum has changed, point directly to the license file in
a way that is easy to copy and paste and give the line numbers in an
easy to read form, as well as asking the user to verify that the new
contents matches the current LICENSE value.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jeffrey C Honig [Wed, 13 Nov 2013 16:37:41 +0000 (11:37 -0500)]
perl: perl-ptest.inc polutes package dependencies when ptest not enabled
When ptest is not enabled, the populate_packages_prepend function runs
wheter ptest is enabled or not. This causes ptest packages to get in the
dependencies list when ptest is not enabled.
Signed-off-by: Jeffrey C Honig <jeffrey.honig@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Phil Blundell [Tue, 12 Nov 2013 13:43:21 +0000 (13:43 +0000)]
binutils: Add gnu-config-native to DEPENDS
do_configure() in binutils.inc includes an explicit call to
gnu-configize so we need to make sure that gnu-config-native is
present. Previously this was being dragged in with the rest of the
autotools stuff, but commit 54a3e2ee37003fc56af0339f857b0b6442790c26
disabled that for binutils-cross on the grounds that "we don't
autoreconf" the toolchain components. Fix this by adding
gnu-config-native itself explicitly to DEPENDS.
Signed-off-by: Phil Blundell <philb@gnu.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Phil Blundell [Tue, 12 Nov 2013 13:29:24 +0000 (13:29 +0000)]
insane, license: Trap MalformedUrl exceptions when parsing LIC_FILES_CHKSUM
bb.fetch.decodeurl() will throw if it doesn't like the look of the URL that
it's given. (Bitbake's idea of what constitutes a valid URL is somewhat
idiosyncratic so it is fairly easy to trip over this by mistake when writing
a recipe.)
If these exceptions are allowed to propagate all the way up to better_exec()
then we will get a large amount of python stack trace spew when they are
finally caught. Avoid that by catching them locally and throwing
bb.build.FuncFailed() with a suitable explanation instead.
Signed-off-by: Phil Blundell <philb@gnu.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
