Paul Menzel [Sat, 3 Apr 2010 01:03:54 +0000 (01:03 +0000)]
enna: Move to own recipe folder and update to new Enna hg repository.
1. Create `enna.inc`.
2. Use `INC_PR`.
3. Move to new repository location.
4. Update license to LGPLv2.1+.
5. Build tested with GCC 4.4.3 and Binutils 2.20.1 for `MACHINE = "beagleboard".
6. Build tested with GCC 4.3.3 and Binuitls 2.18.50 for `MACHINE = "beagleboard".
7. Remove `dbus` from `DEPEND` since `edbus` already depends an it.
8. `libmusicbrainz`, `lightmediascanner` and `taglib` are not needed anymore as `DEPENDS`.
9. `do_configure_prepend()` fixes [1]. Thanks Koen.
10. libvalhalla 2.0.0 (`libvalhalla_hg.bb`) is needed for this version.
mesa-dri 7.8: disable egl for htcdream(this recipe is machine arch)
If we don't disable egl,we have some undefined functions at linking:
| xeglgears.o: In function `main':
| xeglgears.c:(.text+0x1afc): undefined reference to `eglCreateImageKHR'
| xeglgears.c:(.text+0x1c4c): undefined reference to `eglCreateImageKHR'
| xeglgears.c:(.text+0x1cf8): undefined reference to `glEGLImageTargetTexture2DOES'
| collect2: ld returned 1 exit status
Note that we don't have 3d(yet and maybe never) on htcdream
xf86-video-msm : bump git version(fixes gtk issue),rebased patch,also made it compile automatically
*We had an issue with gtk+ programs resulting in screen corruption,this is now fixed
with this version
*rebased the patches to make it work with the new version
*switched to machine arch(thanks XorA and hrw),that permit us to copy a kernel header in the sources,
unfortunately it also depends on another non-staged kernel header(kgsl_drm.h) that
was manually copied in the files dir
*The copy of the kenrel headers in the source directory permit us to compile it
(thanks mickeyl for how to do it corectly)
*removed neon for htcdream machine,if someone wants to support more machines or architectures,
he will have to find a better way and modify the recipe,maybe using a python @ function
that checks TARGET_CC_ARCH,and as CosmicPenguin said in #oe "Thats what open source is all about"
Roman I Khimov [Mon, 5 Apr 2010 14:12:47 +0000 (18:12 +0400)]
ragel-native: new recipe
Ragel compiles executable finite state machines from regular languages.
Ragel targets C, C++, Objective-C, D, Java and Ruby. Ragel state machines
can not only recognize byte sequences as regular expression machines do,
but can also execute code at arbitrary points in the recognition of a
regular language. Code embedding is done using inline operators that do
not disrupt the regular language syntax.
Mostly used to build other applications, thus -native version only.
Paul Eggleton [Sun, 4 Apr 2010 20:45:59 +0000 (21:45 +0100)]
opie-init: fix default font/icon size on hx4700
This patch adds a default qpe.conf for hx4700 so that icons/fonts are
a reasonable size (a direct copy of the one for spitz). Also adds a
1.2.4 version of the opie-init recipe.
Signed-off-by: Paul Eggleton <bluelightning@bluelightning.org>
Paul Eggleton [Sun, 4 Apr 2010 20:44:01 +0000 (21:44 +0100)]
opie-taskbar: force calibration on first start
Since we install default /etc/pointercal files that aren't perfect for
all devices, this patch forces Opie to do a screen calibration during
the first start wizard regardless of whether the pointercal file exists
or not.
Signed-off-by: Paul Eggleton <bluelightning@bluelightning.org>
We do not ship .desktop files in the image,so if we
search for them we end up with this:
++ find ./usr/share/applications -name '*.desktop'
find: `./usr/share/applications': No such file or directory
+ desktop=
in do_rootfs and it fails.
So commenting that part fixes it
Thanks mickeyl for telling me(#openmoko-cdevel in Freenode)
how to fix(so I didn't have to dig into it,and fixed quicker):
Apr 02 14:55:49 <mickeyl> just comment this line out
PR was not bumped because for mickeyl told me(also #openmoko-cdevel in Freenode)
that it was not necessary for images:
Apr 02 14:59:45 <GNUtoo> should I INC_PR the file?
Apr 02 14:59:53 <mickeyl> not necessary
Apr 02 14:59:59 <mickeyl> image files will be reread
Apr 02 15:00:06 <mickeyl> every time you build
gst-meta-base: fixed ivorbis dependency and bump PR
I had the following error:
| Collected errors:
| * satisfy_dependencies_for: Cannot satisfy the following dependencies for gst-meta-audio:
| * gst-plugin-ivorbis *
but... I also had that in my deploy dir: gst-plugin-ivorbisdec_0.10.28-r7.1.5_armv6.ipk
So I changed gst-plugin-ivorbis to gst-plugin-ivorbisdec
e2fsprogs.inc depends on util-linux-ng-native for native builds and bump PR
if We don't depend on util-linux-ng and that it has not been compiled yet
in e2fsck we get undefined references to:
e2fsck_free_context, check_super_block, e2fsck_fix_ext3_journal_hint,
e2fsck_journal_load,e2fsck_get_journal,e2fsck_check_ext3_journal,
e2fsck_journal_reset_super,setup_tdb
gst-plugins-bad 0.10.18 : also depend on orc and bump PR
If we depend only on orc-native and not also on orc
(and orc is not installed):
*it won't be able to find orc/orc.h
*orcc will be detected correctly,which will cause the following code:
"
do_configure_append() {
for i in $(find ${S} -name "Makefile") ; do
sed -i -e s:/usr/bin/orcc:${STAGING_BINDIR_NATIVE}/orcc:g $i
done
}
"
that is in gst-plugins.inc to sed the /usr/bin/orcc in
STAGING_BINDIR_NATIVE}/orcc with STAGING_BINDIR_NATIVE}/orcc
roughly a path like:
${TMPDIR}/staging/i686-linux/usr/bin/orcc
will result in
${TMPDIR}/staging/i686-linux/${TMPDIR}/staging/i686-linux/usr/bin/orcc:
Tom Rini [Thu, 1 Apr 2010 15:10:58 +0000 (15:10 +0000)]
insane.bbclass: Switch to endswith() and fix la/pc check
This changes insane.bbclass to use s.endswith("str") rather than s[-3:] ==
"str". The primary motivation for this is that we have tests that check for
'la' and 'pc' files when we really want '.la' and '.pc' to be what we check
for. Also, our -dev symlink test does if test1: if test2 and test3: action,
when it could just be if test1 and test2 and test3.
Signed-off-by: Tom Rini <tom_rini@mentor.com> Acked-by: Khem Raj <raj.khem@gmail.com> Acked-by: Marcin Juszkiewicz <marcin@juszkiewicz.com.pl>
projects / openembedded.git / log