Paul Eggleton [Thu, 5 Jan 2012 14:46:25 +0000 (14:46 +0000)]
buildhistory: add script to check for significant changes
Adds a buildhistory-diff script which can be used to analyse changes in
the buildhistory git repository (as produced by buildhistory.bbclass),
and report significant ones that may need manual checking to ensure they
aren't regressions (e.g. package size changed by more than a certain
percentage, files added/removed/changed in the image, etc.)
The implementation is actually split into a small script and a Python
module, in order to make the logic re-usable in a future web-based
interface.
Implements the first part of [YOCTO #1566].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Zhai Edwin [Thu, 5 Jan 2012 06:30:51 +0000 (22:30 -0800)]
matchbox-wm: Remove x-session-manager handling
Windows manager should be independent on session manager, especially when we
have multiple session manager like mini-x and matchbox session.
This commit remove session-manager stuff in matchbox-wm, as matchbox-session
already has duplicated code. Also adjust alternative priority for mini-X to
make it higher priority over matchbox-session.
Signed-off-by: Zhai Edwin <edwin.zhai@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Scott Garman [Thu, 5 Jan 2012 21:12:50 +0000 (13:12 -0800)]
insane.bbclass: use bb.process.Popen instead of subprocess.check_output
subprocess.check_output was only introduced in Python v2.7, so we
cannot use it. This refactors the QA test to use bb.process.Popen
instead.
This fixes the error:
AttributeError: 'module' object has no attribute 'check_output'
It no longer checks the return status of prelink-rtld, as that
case was simply adding noise. This QA test is intended to only
warn about specific paths that binaries could be linking to, not
handle the case where there is a missing library.
Signed-off-by: Scott Garman <scott.a.garman@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sat, 17 Dec 2011 10:08:53 +0000 (10:08 +0000)]
native.bbclass: Fix variable remapping coverage
When looking for RDEPENDS to process, bitbake iterates through PACKAGES
*and* PN. Since native.bbclass sets PACKAGES to be empty, its pointless
remapping the list of PACKAGES since this does nothing. There is a problem
since *_${PN} are used by bitbake but not remapped by the native.bbclass
class extension code.
This changes the code to remap _${PN} in both expanded and unexpanded
forms. As a result of this, various surprising dependencies are uncovered
and the patch rectifies those. These are real bugs since they're injecting
unneeded (unremapped) dependencies into the dependency chain.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Gary Thomas [Thu, 5 Jan 2012 06:30:45 +0000 (22:30 -0800)]
xserver-kdrive: work around issues with -Werror=address
A number of calls to REGION_INIT() use a static box which is flagged
as an error with -Werror=address. This patch works around the problem,
but should not be considered a final solution.
Upstream-Status: Inappropriate
Signed-off-by: Gary Thomas <gary@mlbassoc.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Phil Blundell [Thu, 5 Jan 2012 06:30:36 +0000 (22:30 -0800)]
python-native: correct prefix handling for oe-core sysroot layout
Don't try to guess a path based on ${HOST_SYS} since this is not where the
files in the sysroot are likely to be. Instead, just use the "plat_specific"
approach of taking STAGING_INCDIR and knocking off the trailing "/include" for
all cases. This is still evidently suboptimal but does at least seem to give
the correct results.
Signed-off-by: Phil Blundell <philb@gnu.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Scott Garman [Thu, 5 Jan 2012 06:30:35 +0000 (22:30 -0800)]
insane.bbclass: add QA tests for unsafe references to exec_prefix
Files under exec_prefix (commonly /usr) may not be available during
system recovery. exec_prefix may also be kept on a separate partition
that is mounted late in the boot process.
This QA test throws an warning if a binary in base_[bindir|sbindir|libdir]
is dynamically linked to a file under exec_prefix. The intention is to
turn this into an error in the near future.
It also checks executable non-binaries (e.g, shell scripts) in the above
base directories with a simple grep test to look for references to
exec_prefix. This test only produces a warning, since false positives
are likely.
This fixes [YOCTO #1008]
Signed-off-by: Scott Garman <scott.a.garman@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Scott Garman [Thu, 5 Jan 2012 06:30:32 +0000 (22:30 -0800)]
e2fsprogs: move libraries to base_libdir
The e2fsprogs utilities are installed into base_sbindir (/sbin)
and should not link to libraries under exec_prefix (/usr). So move
these libraries from libdir (/usr/lib) to base_libdir (/lib).
Signed-off-by: Scott Garman <scott.a.garman@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Scott Garman [Thu, 5 Jan 2012 06:30:31 +0000 (22:30 -0800)]
util-linux: move libraries to base_libdir
Various util-linux programs in base_sbindir (/sbin) linked to libraries
located under exec_prefix (/usr). Since this is not safe, move these
libraries from libdir (/usr/lib) to base_libdir (/lib).
Signed-off-by: Scott Garman <scott.a.garman@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Scott Garman [Thu, 5 Jan 2012 06:30:29 +0000 (22:30 -0800)]
ncurses: move libraries to base_libdir
Various utilities (including bash and the util-linux programs) located
in base_bindir (/bin) or base_sbindir (/sbin) dynamically link against
many of the ncurses libraries. So move these libraries from libdir
(/usr/lib) to base_libdir (/lib).
Signed-off-by: Scott Garman <scott.a.garman@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Saul Wold [Thu, 5 Jan 2012 00:46:25 +0000 (16:46 -0800)]
image_types: Fix rootfs size calcuation
The ROOTFS_SIZE calculation was not correctly taking into account
the IMAGE_ROOTFS_EXTRA_SPACE variable, it would only be applied if
the size as determined by the ((du * overhead) + extra space) was
greater than the IMAGE_ROOTFS_SIZE, so if the du * overhead was smaller
Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Phil Blundell [Wed, 4 Jan 2012 13:09:08 +0000 (13:09 +0000)]
bitbake.conf: remove some redundant/unnecessary variable exports
Specifically...
SLOT: apparently redundant, deleted.
STAGING_IDLDIR: likewise
QMAKE_MKSPEC_PATH: no longer exported, moved to qmake_base.bbclass
STAGING_SIPDIR: no longer exported, moved to sip.bbclass
Signed-off-by: Phil Blundell <philb@gnu.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Tue, 3 Jan 2012 13:41:33 +0000 (13:41 +0000)]
qt4: add version 4.8.0
Version 4.8.0 makes a few minor changes in the internal build system -
the following issues had to be worked around:
* The -qt-gif configure option has been removed. This is actually the
default and has been for some time, so remove it from qt4.inc.
* The mkspecs have been refactored requiring us to copy our g++.conf
file over the top of g++-unix.conf instead. Some modifications to this
file were also necessary to remove some settings that are now in other
conf files (and we don't modify those values in any case).
* The LD environment variable needs to be unset during configure, or
else the configure script overrides our value of QMAKE_LINK which
selects ${CXX} as our linker.
* QMAKE_CXX contains a reference to OE_QMAKE_CXX which the configure
script does not expect and cannot expand and this results in webkit
being disabled, so add a workaround for this.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Tue, 3 Jan 2012 13:41:32 +0000 (13:41 +0000)]
qt4: fix incorrect architecture for qt4 tools
Don't install the native staging versions of the qt4 tools in the bin
directory; we have been doing this for a very long time but it is no
longer necessary since we specify them via the QMAKE_UIC, QMAKE_MOC etc.
variables. Doing so was sometimes resulting in those executables being
the ones that end up being packaged (probably only sometimes due to
different date/time values on the staging files and the interaction
with make).
Fixes [YOCTO #1856].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Joshua Lock [Mon, 2 Jan 2012 21:50:32 +0000 (13:50 -0800)]
busybox: rename syslog.conf to syslog-startup.conf
Busybox 1.19 introduced basic support for an rsyslog inspired syslog.conf
whereas we've been shipping syslog.conf as a file to be sourced by the
syslog init script in order to configure which options busybox's syslog is
started with.
Busybox 1.19 in syslog mode chokes on our syslog.conf and doesn't start.
This patch renames the syslog.conf we ship to syslog-startup.conf in order
to prevent busybox trying to parse the file as an rsyslog style syslog.conf
Fixes [YOCTO #1848]
Signed-off-by: Joshua Lock <josh@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Nitin A Kamble [Tue, 3 Jan 2012 22:59:31 +0000 (14:59 -0800)]
vte: upgrade from 0.24.3 to 0.28.2
Added a patch to add introspection.m4 to avoid the following
configure error:
| src/Makefile.am:168: HAVE_INTROSPECTION does not appear in
AM_CONDITIONAL
| autoreconf: automake failed with exit status: 1
| ERROR: autoreconf execution failed.
NOTE: package vte-0.28.2-r0: task do_configure: Failed
This upgrade avoids configure issues with automake version 1.11.2
seen on the earlier version of vte:
| src/Makefile.am:155: `pkglibdir' is not a legitimate directory for
`PROGRAMS'
| src/Makefile.am:156: `pkglibdir' is not a legitimate directory for
`SCRIPTS'
| src/Makefile.am:178: variable `interpret_SOURCES' is defined but no
program or
| src/Makefile.am:178: library has `interpret' as canonical name
(possible typo)
| src/Makefile.am:229: variable `slowcat_SOURCES' is defined but no
program or
| src/Makefile.am:229: library has `slowcat' as canonical name (possible
typo)
| src/Makefile.am:203: variable `interpret_LDADD' is defined but no
program or
| src/Makefile.am:203: library has `interpret' as canonical name
(possible typo)
| autoreconf: automake failed with exit status: 1
| ERROR: autoreconf execution failed.
NOTE: package vte-0.24.3-r0: task do_configure: Failed
Did not upgrade to the latest version 0.31.0.
0.28.2 is the last version supporting gtk+ 2. Next versions have
dropped support for gtk+ 2, and they require gtk+ 3. It would be
more appropriate to move to next version after gtk+ 3 recipe is
available.
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Nitin A Kamble [Tue, 3 Jan 2012 22:59:30 +0000 (14:59 -0800)]
prelink: also look at nonstandard lib paths
Prelinking for x32 image showed that it was ignoring libraries
located at locations like /usr/libx32. Like that mips n32 has
library locations set as lib32
This commit modifies prelink.conf to look at libraries also
located at libx32 & lib32 locations.
Thanks to Mark Hatle for suggesting the fix.
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Dexuan Cui [Wed, 4 Jan 2012 09:33:18 +0000 (17:33 +0800)]
mc: fix configure with automake 1.11.2
As Nitin said, "automake version 1.11.2 has made use of dir variables
more strict, the pkglibexec var can not have SCRIPTS suffix. Using pkgdata
instead."
Fixes this error:
| contrib/Makefile.am:10: `pkglibexecdir' is not a legitimate directory for `SCRIPTS'
| autoreconf: automake failed with exit status: 1
NOTE: package mc-4.8.1-r0: task do_configure: Failed
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Joshua Lock [Wed, 4 Jan 2012 01:31:20 +0000 (17:31 -0800)]
cogl: ensure we can build on armv4
Without this patch cogl will not build for armv4 as it uses an
unsupported instruction. This changeset adds a patch from Wolfgang
Denk to add an extra guard around armv5 or above code.
Signed-off-by: Joshua Lock <josh@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Darren Hart [Tue, 3 Jan 2012 16:18:35 +0000 (08:18 -0800)]
distro: Add poky-tiny distro definition
Poky-tiny is intended for building very small OS images. The distro
definition sets the providers for the kernel and the runtime services.
It also reduces the eglibc component list and other DISTRO_FEATURE
elements.
Signed-off-by: Darren Hart <dvhart@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Nitin A Kamble [Tue, 27 Dec 2011 19:19:14 +0000 (11:19 -0800)]
quilt upgreade from 0.48 to 0.50
gnu_patch_test_fix.patch : this patch removed as it is already
upstream now.
Renamed quilt.inc to quilt-0.50.inc and moved the dupicate information
from native & target recipe files into the quilt-0.50.inc file.
Clean up the unpackaged files warning
These warnings are avoided:
WARNING: For recipe quilt, the following files/directories were
installed but not shipped in any package:
WARNING: /usr/bin/quiltrc
WARNING: /usr/share/emacs
WARNING: /usr/share/emacs/site-lisp
WARNING: /usr/share/emacs/site-lisp/quilt.el
The /usr/bin/quiltrc is purposely not packaged, as it is needed
for native recipe only.
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
This patch makes the link to the dependencies of libdw explicit, as recent ld
no longer implicitly links them. See
http://lists.fedoraproject.org/pipermail/devel/2010-March/133601.html as
a similar example of the error message you can encounter without this patch,
and https://fedoraproject.org/wiki/UnderstandingDSOLinkChange and
https://fedoraproject.org/wiki/Features/ChangeInImplicitDSOLinking for more
details.
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Tom Zanussi [Fri, 30 Dec 2011 01:29:06 +0000 (19:29 -0600)]
rpm2cpio.sh: make compression tests case-insensitive
In the rpm2cpio.sh script, the output of $COMPRESSION is tested for
certain lowercase strings such as 'xz' in order to determine the
decompression to use. The problem is that the output strings tested
are from the output of 'file', which uses different cases in different
versions e.g. file-5.09 prints:
tmp/sysroots/x86_64-linux/usr/bin$ ./file xxx.tar.xz: XZ compressed data
while file-5.03 prints:
tmp/sysroots/x86_64-linux/usr/bin$ ./file xxx.tar.xz: xz compressed data
In the former, the XZ string causes xz compressed payloads to
incorrectly fall through to the catch-all lzma case.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
Paul Eggleton [Fri, 30 Dec 2011 13:08:01 +0000 (13:08 +0000)]
buildhistory: improve git commit robustness
* Check if BUILDHISTORY_DIR exists before doing anything with it, in
case no tasks that would have created it have executed
* Ensure the git repo in BUILDHISTORY_DIR is initialised with "git init"
before attempting to do anything with it
* Check if any files have been added or changed before adding and
committing, to avoid an error from "git commit"
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Shane Wang [Fri, 30 Dec 2011 06:59:27 +0000 (14:59 +0800)]
distro_tracking_fields: update the info for this round of package upgrade
The update includes:
- update the info for linux-firware, quota, watchdog, hdparm, libomxil, oprofileui, puzzles, and matchbox-wm-2
- add one more recipe info for stat, which is depended by hdparm
- correct the wrong info for connman-gnome, which is updated by last manual check in error.
Shane Wang [Thu, 29 Dec 2011 09:05:01 +0000 (17:05 +0800)]
quota: upgrade to 4.00
The license change in rquota_server.c is the following two lines have been removed:
11a12,13
> * Version: $Id: rquota_server.c,v 1.22 2010/01/05 16:04:57 jkar8572 Exp $
> *
Nitin A Kamble [Wed, 28 Dec 2011 19:59:52 +0000 (11:59 -0800)]
avahi: fix configure with automake 1.11.2
automake version 1.11.2 has made use of dir variables more strict.
use of pkglibdir with DATA var results in automake errors.
This commits uses pkgdatadir var instead of pkglibdir to avoid
the strict check errors.
Fixes this error:
service-type-database/Makefile.am:21: `pkglibdir' is not a legitimate
directory for `DATA'
autoreconf: automake failed with exit status: 1
ERROR: autoreconf execution failed.
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Nitin A Kamble [Wed, 28 Dec 2011 17:34:00 +0000 (09:34 -0800)]
autotools.bbclass: fix cp error for empty dir
the .../usr/share/aclocal is empty for build from scratch. so avoid cp
error if the directory is empty.
Fixes this error, which is thrown before pseudo is built:
| cp: cannot stat
`/builddisk/build/build0/tmp/sysroots/x86_64-linux/usr/share/aclocal/*':
No such file or directory
NOTE: package libtool-native-2.4.2-r0.0: task do_configure: Failed
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Nitin A Kamble [Tue, 27 Dec 2011 20:47:59 +0000 (12:47 -0800)]
rpm: fix build with automake-1.11.2
automake version 1.11.2 has made use of dir variables more strict.
the use of pkglibdir with SCRIPTS & DATA vars results in automake
errors.
This commits uses pkgdatadir & pkgbindir vars instead of pkglibdir
to avoid the strict check errors.
This change also works with automake-1.11.1
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Saul Wold [Thu, 1 Dec 2011 22:45:39 +0000 (14:45 -0800)]
util-linux: Update to 2.20.1
Add patch remove setting CC, LD and LDFLAGS on commandline
since they are passed correctly via configure. Those setting
also caused the configure information about -std=gnu99 from
being passed correctly.
Removed patch that was fixed upstream
Added scanf configuration, since this is a cross-compile target check
libuuid removed an API that e2fsprogs uuidd used (uuid__generate_*),
thus we disable it in this patch and will now provide uuidd from
util-linux.
See http://sourceforge.net/projects/e2fsprogs/forums/forum/7053/topic/4639484
Andrei Gherzan [Thu, 29 Dec 2011 10:22:06 +0000 (12:22 +0200)]
wpa-supplicant - pulls GPLv3 license rpm in a non-GPLv3 build (libgnutls-extra)
Two different defconfigs are chosen, one for "with GPLv3 build" and one for "without".
In this way, libgnutls-extra is not pulled in a non-GPLv3 build.
Zhai Edwin [Tue, 27 Dec 2011 02:57:29 +0000 (10:57 +0800)]
gnome-keyring: fix gsettings_data_convert seg fault
gsttings_data_convert cause seg fault as gnome-keyring's schema is not
installed properly. As a fix, running glib-compile-schemas from glib-utils to
made "gschemas.compiled".
Darren Hart [Wed, 21 Dec 2011 08:24:46 +0000 (00:24 -0800)]
image_types: Ensure /init exists for cpio rootfs archives
In order for the kernel to accept an initramfs as a rootfs,
it must have /init - even if it is an empty file. Touch /init
to ensure it exists for cpio and cpio.gz image types.
When used with initrd and rootfs=/dev/ram0, this allows the
cpio and cpio.gz images to be used as the rootfs.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
