Paul Eggleton [Thu, 19 Jan 2012 10:32:12 +0000 (10:32 +0000)]
buildhistory_analysis: include related fields in output
Sometimes, when a value changes in the buildhistory it is useful to
know when a related (but not necessarily itself monitored) value
changes as it can help explain the change. For example, when the list
of installed packages for an image changes it could be caused by a
change to one of the image-related variables.
Related field changes are recorded as sub-items of each change.
Currently the only way to visualise these is via the buildhistory-diff
tool, so an example would be:
Changes to images/qemux86/eglibc/core-image-minimal (installed-package-names.txt):
locale-base-de-de was added
procps was added
* IMAGE_LINGUAS: added "de-de"
* IMAGE_INSTALL: added "procps"
Here we see that two additional packages have been added to the image,
and looking at the related changes to the two variables IMAGE_INSTALL
and IMAGE_LINGUAS we have the explanation as to why.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Thu, 19 Jan 2012 13:59:15 +0000 (13:59 +0000)]
classes/patch: avoid backtrace when patch does not apply
We don't need to see a Python stack backtrace when a patch does not
apply, just the error output from patch, so trap these kinds of errors
and ensure that we display the message and fail the task and nothing
else.
Fixes [YOCTO #1143]
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Martin Jansa [Thu, 19 Jan 2012 13:51:44 +0000 (14:51 +0100)]
task-base: rename VIRTUAL-RUNTIME-alsa-state to use underscore
* in http://lists.linuxtogo.org/pipermail/openembedded-core/2012-January/016226.html
we talked about
VIRTUAL-RUNTIME_alsa-state, but
VIRTUAL-RUNTIME-alsa-state was used instead, which is inconsistent with other VIRTUAL-RUNTIME vars
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Thu, 19 Jan 2012 10:32:11 +0000 (10:32 +0000)]
buildhistory_analysis: improve field handling robustness
Avoid errors when comparing changes for KEY = value files (package info
files and image-info.txt):
* Handle keys appearing and disappearing - this will help to handle PE
in package info files (which is only written when it is not blank) and
when we add additional fields in future.
* Handle when old value is 0 for numeric field (avoid division by zero)
* Report when numeric field was empty or missing rather than 0 (but
still treat it as 0 for comparison purposes)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Thu, 19 Jan 2012 10:32:10 +0000 (10:32 +0000)]
buildhistory_analysis: correctly handle whitespace when splitting lists
Don't specify any argument to the split() function when handling changes
to list type variables (e.g. PACKAGES) so that the values are split by
any whitespace and only split once for a block of multiple whitespace
characters.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Thu, 19 Jan 2012 10:32:09 +0000 (10:32 +0000)]
buildhistory: record additional image info
Record some additional information about images - the uncompressed size
of the final image as well as the values of various variables that may
have influenced its contents. This is recorded in a machine-readable
"image-info.txt" file similar in structure to the package history files.
Also add some code to analyse changes to these values. (Most of the
variable values aren't monitored directly but will be used as contextual
information when they change at the same time as the content of the
image changing.)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Nitin A Kamble [Tue, 17 Jan 2012 19:11:01 +0000 (11:11 -0800)]
grub-efi-native: fix errors with automake 1.11.2
This commit replaces earlier patch
(grub-1.99_fix_for_automake_1.11.2.patch) which did not work on all distros.
Fixes these errors observed with automake 1.11.2
The useof pkglibhas become more strict compared to the earlier release
of
automake resulting in these failures.
Fixed the files related to automake to avoid the issue.
| conf/Makefile.common:140: `pkglibdir' is not a legitimate directory
for `DATA'
| grub-core/Makefile.am:5: `conf/Makefile.common' included from here
| conf/Makefile.common:140: `pkglibdir' is not a legitimate directory
for `DATA'
| Makefile.am:6: `conf/Makefile.common' included from here
| autoreconf: automake failed with exit status: 1
| ERROR: autoreconf execution failed.
| conf/Makefile.common:150: `pkglibdir' is not a legitimate directory
for `SCRIPTS'
| grub-core/Makefile.am:5: `conf/Makefile.common' included from here
| conf/Makefile.common:140: `pkglibdir' is not a legitimate directory
for `DATA'
| grub-core/Makefile.am:5: `conf/Makefile.common' included from here
| conf/Makefile.common:150: `pkglibdir' is not a legitimate directory
for `SCRIPTS'
| Makefile.am:6: `conf/Makefile.common' included from here
| conf/Makefile.common:140: `pkglibdir' is not a legitimate directory
for `DATA'
| Makefile.am:6: `conf/Makefile.common' included from here
| autoreconf: automake failed with exit status: 1
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Khem Raj [Thu, 19 Jan 2012 05:39:16 +0000 (21:39 -0800)]
gnutls: Fix configure issue wrt to rm
Add -f option so if the files dont exist then we can
still continue. Fixes errors like
| + do_configure
| + for dir in . lib libextra
| + rm ./aclocal.m4 ./m4/libtool.m4 './m4/lt*.m4'
| rm: cannot remove `./m4/libtool.m4': No such file or directory
| rm: cannot remove `./m4/lt*.m4': No such file or directory
Khem Raj [Wed, 18 Jan 2012 03:01:05 +0000 (19:01 -0800)]
e2fsprogs_1.42.bb: Fix build fallout for uclibc
e2fsprogs 1.42 onwards uses fallocate() which uclibc
does not implement(yet). In most of places its use
is controlled and it only used when configure detects
fallocate being present but in this one case it missed
to check for fallocate being available so here we
add the check
Bruce Ashfield [Wed, 18 Jan 2012 17:07:55 +0000 (12:07 -0500)]
linux-yocto: restore WORKDIR as a patch source directory
In the switch to using src_patches and using patches in their
source directory, the scanning of WORKDIR migrated items like
config fragments was dropped. Adding WORKDIR back as a patch
directory restores the old functionality.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Saul Wold [Thu, 19 Jan 2012 00:00:57 +0000 (16:00 -0800)]
pkgconfig: Move pkg.m4 from -dev to package
If pkgconfig is installed you're doing development work and the -dev package
makes little sense in the pkgconfig case. Keeping the .m4 file associated with
the main package therefore makes more sense than having the user install both.
Andrei Gherzan [Mon, 16 Jan 2012 18:31:50 +0000 (20:31 +0200)]
libsoup: Update package from v2.2.100 to v2.2.105
I included a patch as well to solve a compile error. In this package, a dfprinf function
is declared. It will fail as eglibc provides it's function as well. So i renamed libsoup's
dprintf function into dprinfsoup.
Joshua Lock [Wed, 11 Jan 2012 19:19:27 +0000 (11:19 -0800)]
busybox: change syslog default to log to a file
This changes the default in syslog-startup.conf to log messages
to a file (/var/log/messages) because:
a) we already mount /var/log as a volatile mount by default
b) users are accustomed to looking at /var/log/messages as
most distributions don't ship with logread
Joshua Lock [Wed, 11 Jan 2012 19:02:46 +0000 (11:02 -0800)]
busybox: enable log level configuration in syslog-startup.conf and script
It's possible to configure busybox syslog to only log messages above a
certain level, expose this functionality through the syslog-startup.conf
and syslog init script.
Andrei Gherzan [Tue, 17 Jan 2012 11:28:31 +0000 (13:28 +0200)]
busybox/uglibc: remove obsolete check for "kernel24"
Busybox checks if MACHINE_FEATURES contains "kernel24". If so, CONFIG_FEATURE_2_4_MODULES will be on "y".
kernel24 is no longer present in any machine configuration. The same situation is in uglibc with a
different CONFIG mapping.
[YOCTO #1901]
Signed-off-by: Andrei Gherzan <andrei@gherzan.ro> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Martin Jansa [Tue, 17 Jan 2012 10:58:04 +0000 (11:58 +0100)]
gnutls.inc: set libz prefix
* without this patch host's zlib is used, similar to issue with
libreadline it's important if host's zlib is compatible with MACHINE
and in this case zlib was in DEPENDS so header was always available in
sysroot
lib/config.log:
configure:18288: checking whether to include zlib compression support
configure:18291: result: yes
configure:18786: checking for libz
configure:18808: x86_64-oe-linux-gcc -m64 --sysroot=/OE/shr-core/tmp-eglibc/sysroots/qemux86-64 -o conftest -O2 -pipe -g -feliminate-unused-debug-types -I/OE/shr-core/tmp-eglibc/sysroots/qemux86-64/usr/include -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed conftest.c /usr/lib/libz.so >&5
configure:18808: $? = 0
configure:18818: result: yes
configure:18825: checking how to link with libz
configure:18827: result: /usr/lib/libz.so
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Martin Jansa [Tue, 17 Jan 2012 10:58:03 +0000 (11:58 +0100)]
gnutls: add readline to DEPENDS and add --with-libreadline-prefix
* without this patch it looks for readline in host's /usr/lib
a) and fails if readline wasn't built before gnutls, because -I is set
and cannot find header file, this case is better
configure:33131: checking for readline
configure:33156: x86_64-oe-linux-gcc -m64 --sysroot=/OE/oe-core/tmp-eglibc/sysroots/qemux86-64 -std=gnu99 -o conftest -O2 -pipe -g -feliminate-unused-debug-types -I/OE/oe-core/tmp-eglibc/sysroots/qemux86-64/usr/include -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed conftest.c /usr/lib/libreadline.so >&5
conftest.c:240:31: fatal error: readline/readline.h: No such file or directory
b) and finds usable version there if readline was built and MACHINE is compatible with host
(like qemux86-64 on amd64 host) and then it's passing -L/usr/lib (without sysroot prefix)
to every build using libgnutls.la
configure:33131: checking for readline
configure:33156: x86_64-oe-linux-gcc -m64 --sysroot=/OE/shr-core/tmp-eglibc/sysroots/qemux86-64 -std=gnu99 -o conftest -O2 -pipe -g -feliminate-unused-debug-types -I/OE/shr-core/tmp-eglibc/sysroots/qemux86-64/usr/include -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed conftest.c /usr/lib/libreadline.so >&5
configure:33156: $? = 0
configure:33173: result: yes
configure:33185: checking how to link with libreadline
configure:33187: result: /usr/lib/libreadline.so
configure:33199: checking readline/readline.h usability
configure:33199: x86_64-oe-linux-gcc -m64 --sysroot=/OE/shr-core/tmp-eglibc/sysroots/qemux86-64 -std=gnu99 -c -O2 -pipe -g -feliminate-unused-debug-types -I/OE/shr-core/tmp-eglibc/sysroots/qemux86-64/usr/include conftest.c >&5
configure:33199: $? = 0
configure:33199: result: yes
configure:33199: checking readline/readline.h presence
configure:33199: x86_64-oe-linux-gcc -E --sysroot=/OE/shr-core/tmp-eglibc/sysroots/qemux86-64 -m64 -I/OE/shr-core/tmp-eglibc/sysroots/qemux86-64/usr/include conftest.c
configure:33199: $? = 0
configure:33199: result: yes
configure:33199: checking for readline/readline.h
configure:33199: result: yes
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Zumeng Chen [Mon, 16 Jan 2012 15:46:58 +0000 (23:46 +0800)]
lttng-modules: Update to 2.0
This patch intended to integrate the lttng-modules 2.0 package
containing the kernel tracer modules.
LTTng-modules 2.0 is currently in pre-release (-pre11), so we
have to update it when official released.
lttng-2.0 supports lttng-modules extra builds, which don't need
any patches on Linux kernel vs the previous version of lttng.
As described in README of lttng-modules-2.0, so far, it has been
tested in the latest kernel on x86 32/64-bit, and powerpc 32-bit,
, build tested on ARM. So does yocto kernel, the related tests
on qemux86 and qemuppc has been validated and a build test on
qemuarm passed too with this patcheset.
Zumeng Chen [Mon, 16 Jan 2012 15:46:35 +0000 (23:46 +0800)]
lttng-tools: Update to 2.0-pre16
This patch intended to integrate the Linux trace toolkit, which
is a suite of tools designed to extract program execution details
from the Linux operating system and interpret them.
lttng-tools 2.0 is currently in pre-release(-pre16), so we have to
update it when official released.
Saul Wold [Mon, 16 Jan 2012 16:43:57 +0000 (08:43 -0800)]
m4: refactor .bb and .inc files
The orignal had common code in the m4_1.4.16.bb file that was
in the m4.inc file. m4-native was then including the m4.bb file
and picking up the BBCLASSEXTEND, which causes the fetch to fail.
Paul Eggleton [Fri, 13 Jan 2012 17:22:36 +0000 (17:22 +0000)]
classes/packagehistory: remove now obsolete class
packagehistory.bbclass has been superseded by buildhistory.bbclass,
which gives more detailed output (including information on produced
images) as well as other enhanced functionality.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Paul Eggleton [Fri, 13 Jan 2012 17:35:46 +0000 (17:35 +0000)]
classes/package: remove out-of-date notice referring to the OE bugtracker
OpenEmbedded doesn't officially use a bug tracker anymore, and a message
asking the user to fix the metadata could be applied to almost any error
that occurs, so just remove the whole message.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Paul Eggleton [Fri, 13 Jan 2012 17:16:18 +0000 (17:16 +0000)]
conf/bitbake.conf: allow wget to output error messages
Use the -nv instead of -q command line option for wget so that if
fetching from a URL that the wget fetcher handles (e.g. http) fails, we
get a possibly useful error message (but it still remains relatively
quiet when the fetch succeeds.)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Mark Hatle [Fri, 13 Jan 2012 17:41:07 +0000 (11:41 -0600)]
prelink: Update to latest version
The latest version of prelink corrects a problem when attempting
to prelink MIPS architectures.
2012-01-04 Mark Hatle <mark.hatle@windriver.com>
* exec.c: Check that a section is larger then 0 bytes when
determining the section to segment mapping. This matches
the behavior of elfutils - readelf. Otherwise an empty
PROGBITS section at the end of a segment will cause a
failure.
[YOCTO #1463]
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Joshua Lock [Tue, 10 Jan 2012 16:32:08 +0000 (08:32 -0800)]
alsa-state: move state files to localstatedir
alsactl creates the state files in /var/lib/alsa by default so switch
alsa-state to use files in that location.
Further, update the alsa-state init script to have the location of the
state files sed'ed into the script at do_install time (so as to remove
hard coding of directory paths).
Enrico Scholz [Thu, 12 Jan 2012 16:30:24 +0000 (17:30 +0100)]
bitbake.conf: search /lib for -native builds
Changes to BUILD_LDFLAGS are required due to the recent movement of some
libraries from ${libdir} to ${base_libdir}. ${base_libdir} must be now
in the linker search path to avoid build problems of -native packages.
Nitin A Kamble [Thu, 12 Jan 2012 22:17:21 +0000 (14:17 -0800)]
grub-efi-native: fix errors with automake 1.11.2
Fixes these errors observed with automake 1.11.2
The useof pkglibhas become more strict compared to the earlier release
of
automake resulting in these failures.
Fixed the files related to automake to avoid the issue.
| conf/Makefile.common:140: `pkglibdir' is not a legitimate directory
for `DATA'
| grub-core/Makefile.am:5: `conf/Makefile.common' included from here
| conf/Makefile.common:140: `pkglibdir' is not a legitimate directory
for `DATA'
| Makefile.am:6: `conf/Makefile.common' included from here
| autoreconf: automake failed with exit status: 1
| ERROR: autoreconf execution failed.
| conf/Makefile.common:150: `pkglibdir' is not a legitimate directory
for `SCRIPTS'
| grub-core/Makefile.am:5: `conf/Makefile.common' included from here
| conf/Makefile.common:140: `pkglibdir' is not a legitimate directory
for `DATA'
| grub-core/Makefile.am:5: `conf/Makefile.common' included from here
| conf/Makefile.common:150: `pkglibdir' is not a legitimate directory
for `SCRIPTS'
| Makefile.am:6: `conf/Makefile.common' included from here
| conf/Makefile.common:140: `pkglibdir' is not a legitimate directory
for `DATA'
| Makefile.am:6: `conf/Makefile.common' included from here
| autoreconf: automake failed with exit status: 1
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Martin Jansa [Fri, 23 Dec 2011 10:18:24 +0000 (11:18 +0100)]
xorg-*: disable xmlto where it was failing
* meta-oe now provides xmlto-native
http://git.openembedded.org/meta-openembedded/commit/?id=ab8f876ca8fa223c183e3db2029cce88f3435a27
but not the stylesheets for buildhosts without them it's unusable and fails, because xorg macros
autodetecting xmlto available, see:
http://lists.linuxtogo.org/pipermail/openembedded-devel/2011-December/037066.html
after adding xmlto-native and stylesheets-native to DEPENDS we can enable it again
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
We need to depend on libsamplerate0 because it's detected automatically
and if we configure without it we get that issue at runtime trying
to run the alsaloop program:
No libsamplerate support.
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org> Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Darren Hart [Wed, 11 Jan 2012 23:04:24 +0000 (15:04 -0800)]
bootimg: Account for FAT filesystem overhead in image size
Fixes [YOCTO #1852]
The bootimg class wasn't accounting for non-trivial amount of space
required by the directory entries and FATs for the FAT filesystem.
This patch attempts to make an accurate prediction of FAT overhead and
adjusts the image size accordingly. It assumes no more than 16 directory
entries per directory (which fit in a single sector). It also assumes
8.3 filenames. With the ceiling functions rounding up to full sectors
and tracks, these assumptions seem reasonable.
In order to ensure the calculations are accurate, this patch forces the
FAT size to 32, rather than allowing mkdosfs to automatically select 12,
16, or 32 depending on the image being built.
Tested by setting BOOTIMG_EXTRA_SPACE=0 and building core-image-minimal
and core-image-sato for fri2-noemgd from meta-intel.
Signed-off-by: Darren Hart <dvhart@linux.intel.com> CC: Saul Wold <sgw@linux.intel.com>
Given a license string and whitelist and blacklist, determine if the
license string matches the whitelist and does not match the blacklist.
When encountering an OR, it prefers the side with the highest weight (more
included licenses). It then checks the inclusion of the flattened list of
licenses from there.
Returns a tuple holding the boolean state and a list of the applicable
licenses which were excluded (or None, if the state is True)
copyleft_compliance: add control of recipe types to include
In this context, recipe "type" refers to whether it is 'target', 'native',
'cross', etc. COPYLEFT_RECIPE_TYPES is a space separated list of types to
include. It defaults to 'target'.
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Shane Wang [Wed, 11 Jan 2012 11:44:33 +0000 (19:44 +0800)]
hdparm: upgrade to 9.37
The patch was imported from the OpenEmbedded server (http://git.openembedded.org/openembedded/tree/recipes/hdparm) as of commit id 4d2cb79dcecd056742f411a328f9f1f1113bf689.
And changes include:
- upgrade to the latest version 9.37 from 9.35.
- added license checksum.
- the license for wiper which is in hdparm is GPLv2.
Nitin A Kamble [Wed, 11 Jan 2012 01:33:31 +0000 (17:33 -0800)]
insane.bbclass: fix elf.arch not matching error for x32 kernel
For x32 the user space is 32bit and the kernel is 64bit.
So the elf.arch for vmlinuz is x86_64 and not x86. This commit
fixes this QA error thrown for x32 kernel.
ncurses-native: install to libdir, not base_libdir
For target, both base_libdir and libdir in sysroot can be used, as we pass
--sysroot to the toolchain. For native, we don't do this, and we also only add
-L<sysroot>/${libdir}, not -L<sysroot>/${base_libdir}, resulting in other
native recipes (like readline-native) failing to find the ncurses libraries.
readline-native only built successfully on hosts where it could fall back to
their ncurses/termcap rather than the one in the sysroot.
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
James Limbouris [Tue, 10 Jan 2012 08:37:48 +0000 (16:37 +0800)]
ghostscript: build cups filters
Cups filters are no longer built by default. Since ghostscript already depends
on cups, build the filters and package them as ghostscript-cups.
Ghostscript uses the cups-config script from the oe sysroots to determine some
of the target install paths, as well as to determine the linker path. The config
script gives out paths pointing to the sysroot, so some of the paths needed to
be adjusted in the Makefile.
Signed-off-by: James Limbouris <james@digitalmatter.com.au>
Andrei Gherzan [Fri, 6 Jan 2012 15:34:36 +0000 (17:34 +0200)]
wpa-supplicant: Compile without CONFIG_GNUTLS_EXTRA (PR BUMP)
CONFIG_GNUTLS_EXTRA is needed as support for TLS/IA which was designed to be used
in the EAP-TTLSv1. As we don't see any requirement for that protocol today we decided
to remove it from wpa-supplicant .config file.
This change includes PR bump.
