**** BEGIN LOGGING AT Sun Apr 03 02:59:58 2011 Apr 03 04:38:25 hello all, I'm having a little bit of trouble building an initial image using OE Apr 03 04:39:36 eglibc-initial do_install is failing as it tried to remove files from the host system and gets a permission denied error Apr 03 04:40:56 I looked into recipies/eglibc/eglibc-initial.inc and inside do_configure() it has "--prefix=/usr" which seems wrong? Apr 03 16:41:52 03Paul Menzel  07master * r8adf7a5f7d 10openembedded.git/recipes/gdbm/gdbm_1.8.3.bb: Apr 03 16:41:52 gdbm: convert tabulator to spaces Apr 03 16:41:52 Signed-off-by: Paul Menzel Apr 03 20:35:09 which receipt can give me an idea how to modify /etc/inittab of busybox? Apr 03 20:44:55 dv__, of the top of my head I would go for something called base-files or similar **** BEGIN LOGGING AT Sun Apr 03 23:55:21 2011 **** BEGIN LOGGING AT Mon Apr 04 15:14:14 2011 Apr 04 15:31:24 03Michael 'Mickey' Lauer  07org.openembedded.dev * r37a35377de 10openembedded.git/recipes/vala/vala_0.12.0.bb: vala[-native]: add 0.12.0 Apr 04 17:30:33 03Martin Jansa  07master * rd872e0ca76 10openembedded.git/recipes/freesmartphone/opimd-utils_git.bb: Apr 04 17:30:33 opimd-utils: RDEPEND on ffphonelog instead of pyphonelog Apr 04 17:30:33 Signed-off-by: Martin Jansa Apr 04 17:30:44 03Martin Jansa  07master * r397f158896 10openembedded.git/conf/distro/shr.conf: Apr 04 17:30:44 shr: blacklist pstree in favor of psmisc Apr 04 17:30:44 Signed-off-by: Martin Jansa Apr 04 17:36:24 03Martin Jansa  07master * r73e10e3b08 10openembedded.git/conf/compatibility-providers.conf: Apr 04 17:36:24 compatibility-providers: use libav as ffmpeg provider Apr 04 17:36:24 * angstrom is using libav since cd4eacbe9eb3c396a5025fc7b7bac59f5bd36096 Apr 04 17:36:24 Signed-off-by: Martin Jansa Apr 04 17:36:34 03Martin Jansa  07master * rb7753cc968 10openembedded.git/recipes/e17/ (2 files in 2 dirs): Apr 04 17:36:34 elmdentica: fix build after azy changes from r58286 Apr 04 17:36:34 Signed-off-by: Martin Jansa Apr 04 17:36:36 03Martin Jansa  07master * r9a888d16d8 10openembedded.git/recipes/efl1/azy_svn.bb: Apr 04 17:36:36 azy: disable mysql tests Apr 04 17:36:36 Signed-off-by: Martin Jansa Apr 04 19:07:37 03Tom Rini  07master * rde42055b9a 10openembedded.git/recipes/ptpd/ (ptpd-1.0.0/add-limit-h.patch ptpd_1.0.0.bb): Apr 04 19:07:37 ptpd: Add a patch to bring in limits.h Apr 04 19:07:37 Taken from the old bugzilla: Apr 04 19:07:37 http://lists.linuxtogo.org/pipermail/openembedded-commits/2010-January/041030.html Apr 04 19:07:37 Signed-off-by: Tom Rini Apr 04 19:10:59 can somebody tell me if there's a file that I can "include" so that I can type "make" to build one of the components instead of bitbaking the component? Is there a big master include file or something? Apr 04 19:11:17 for instance, I JUST want to build the dsplink samples on OMAP3530 Apr 04 19:11:24 I've defined $DSPLINK Apr 04 19:11:25 but Apr 04 19:11:29 the toolchain isn't defined Apr 04 19:11:38 is there an easy solution to this or a workaround? Apr 04 19:14:04 I'll echo and check the log files ;) Apr 04 19:14:07 that'll get me there Apr 04 19:24:51 bitbake -c devshell recipe Apr 04 20:10:30 sorry for asking this question second time.. My gui was crashed and I can;t read answer... Can somebody point me to receipt that is adding custom string to /etc/inittab ? Apr 04 20:16:19 dv: cd openembedded/recipes; find . -name inittab Apr 04 20:19:17 dv: sysvinit Apr 04 20:55:13 03Philippe De Swert  07master * r09d88a9648 10openembedded.git/recipes/efl1/ (eio.inc eio_svn.bb): Apr 04 20:55:13 eio: new recipe and include file for eio Apr 04 20:55:13 Signed-off-by: Philippe De Swert Apr 04 20:55:13 Signed-off-by: Martin Jansa Apr 04 20:55:23 03Philippe De Swert  07master * r7352b69682 10openembedded.git/recipes/e17/ephoto_svn.bb: Apr 04 20:55:23 ephoto: new svn build recipe Apr 04 20:55:23 Signed-off-by: Philippe De Swert Apr 04 20:55:23 Signed-off-by: Martin Jansa Apr 04 20:58:59 03Martin Jansa  07master * r4b298625bd 10openembedded.git/recipes/e17/elmdentica_svn.bb: Apr 04 20:58:59 elmdentica: add maxrev, patch was applied upstream Apr 04 20:58:59 Signed-off-by: Martin Jansa Apr 04 21:24:08 03Chris Larson  07master * r156ea134e8 10bitbake.git/lib/bb/event.py: Apr 04 21:24:08 event: improve output for syntax errors in handlers Apr 04 21:24:08 Note: this includes IndentationError, which is a subclass of SyntaxError. Apr 04 21:24:08 Signed-off-by: Chris Larson Apr 04 21:24:10 03Chris Larson  07master * rfaf682dfc2 10bitbake.git/lib/bb/event.py: (log message trimmed) Apr 04 21:24:10 event: improve output when eventhandler exec fails Apr 04 21:24:10 - Name the event handler by its actual name, so the traceback shows it rather Apr 04 21:24:10 than 'tmpHandler'. Apr 04 21:24:11 - Rather than immediately aborting when encountering an event handler error, Apr 04 21:24:11 display an error message and try to continue. Apr 04 21:24:12 - Show a traceback for ordinary exceptions, skipping the first entry in the Apr 04 21:24:15 03Chris Larson  07master * rbe647dac9d 10bitbake.git/lib/bb/event.py: Apr 04 21:24:15 event: register event handler functions, not code objects Apr 04 21:24:15 Signed-off-by: Chris Larson Apr 04 21:24:42 (that was bugging me) Apr 04 21:26:24 still no fix for native:virtual: in path? :/ Apr 04 21:38:40 JaMa: done. sorry for the delay Apr 04 21:38:47 took all of 12 minutes to fix Apr 04 21:38:48 03Chris Larson  07master * r58bdaeb679 10bitbake.git/lib/bb/cooker.py: Apr 04 21:38:48 cooker: fix -b with BBCLASSEXTEND Apr 04 21:38:48 Signed-off-by: Chris Larson Apr 04 21:38:48 * kergoth rolls eyes Apr 04 21:39:38 yay, /me trying Apr 04 21:40:05 worked for me with coreutils-native at least Apr 04 21:40:16 also verified that specifying a nonexistant virtual gives a clean error Apr 04 21:40:22 e.g. virtual:nativeee:foo Apr 04 21:40:26 hmm Apr 04 21:44:21 kergoth: works for me too, thanks again! Apr 04 21:44:26 np Apr 04 21:44:50 kergoth: btw there are 2 simple patches for setup.py http://patches.openembedded.org/project/bitbake/list/, would you mind pusing them? would simplify my .ebuild a bit :) Apr 04 21:45:03 * kergoth looks Apr 04 21:45:52 btw, can we check for distro presence early at parsing stage? Apr 04 21:46:08 just to kick out people with angstrom-2008.1 and typos in DISTRO Apr 04 21:49:24 03Martin Jansa  07master * rf50a22ee1b 10bitbake.git/setup.py: Apr 04 21:49:24 setup.py: install fetch2 module Apr 04 21:49:24 Signed-off-by: Martin Jansa Apr 04 21:49:24 Signed-off-by: Chris Larson Apr 04 21:49:30 03Martin Jansa  07master * r2db367bc80 10bitbake.git/setup.py: Apr 04 21:49:30 setup.py: also install bitbake-{layers,diffsigs} scripts Apr 04 21:49:30 Signed-off-by: Martin Jansa Apr 04 21:49:30 Signed-off-by: Chris Larson Apr 04 21:50:35 Jay7: well, originally DISTRO was optional :) nowadays, that's not really the case, so we should add a check, perhaps at ConfigParsed time, but i'm not sure if that's early enough due to OVERRIDES Apr 04 21:50:37 hmm Apr 04 21:51:03 kergoth: well, even after parsing it should be better than strange errors :) Apr 04 21:51:21 you'll get no argument from me there Apr 04 21:51:28 just a matter of working out how to do it Apr 04 21:51:38 thank you :) Apr 04 22:36:02 the php recipe hurts Apr 04 22:41:38 the php recipe makes me hurt Apr 04 22:42:14 I'm pretty sure just saying "THREADS=pthreads" and LIBS=-lpthread doesn't trick it into doing pthreads Apr 04 22:42:54 I think we just luck out and don't enable any bits that _require_ pthreads Apr 04 22:56:21 http://pastebin.com/pXiPJfkd - Good idea or just too ugly of a combination? Apr 04 22:59:59 bah, damn tz(data|code) need updated again Apr 04 23:04:28 foerster: to e? Or noticing that the maintenance branch is out of date? Apr 04 23:05:54 master is on d, new is e Apr 04 23:06:01 fresh pull today broke on that Apr 04 23:06:34 what's broken? Apr 04 23:06:41 just tzdata Apr 04 23:06:47 ah Apr 04 23:06:50 as usual :) Apr 04 23:06:51 they released a new one again Apr 04 23:07:10 Of course, they did it when I'm in a hurry :) Apr 04 23:07:20 figures Apr 04 23:08:21 If nobody else sends a patch for it first, I'll try to send it later tonight. Apr 04 23:10:00 kergoth: http://pastebin.com/pXiPJfkd, good idea or too ugly? Apr 04 23:10:37 Tartarus: that's a rather large file, i'll need more context :) Apr 04 23:10:52 aside: screwed up my desktop's linux install, reinstalling ubuntu Apr 04 23:10:56 * kergoth rolls eyes Apr 04 23:11:09 kergoth: php5 rework and combine to use BBCLASSEXTEND Apr 04 23:11:26 Two files in there Apr 04 23:11:36 (and boy howdy is php a pain to cross compile) Apr 04 23:11:41 php has issues. Apr 04 23:12:08 the EXTRA_OECONF bits seem not ideal, but it's the best we can do lacking a target override, and the _pn- one is ugly too :) Apr 04 23:12:34 kergoth: yeah and I'm not convinced they're doing anything, but that's also just how it was before Apr 04 23:12:58 export LIBS_pn-php =" -lpthread " looks wrong Apr 04 23:13:05 flags won't survive overrides Apr 04 23:13:11 LIBS_pn-php = "-lpthread" Apr 04 23:13:12 export LIBS Apr 04 23:13:18 that's what you want, if LIBS *really* needs to be exported Apr 04 23:13:23 same for THREADS Apr 04 23:13:31 yeah Apr 04 23:13:40 looks reasonable otherwise Apr 04 23:13:40 But I'm not convinced php obeys them today as we do it Apr 04 23:13:56 php wouldn't allow modules b/c their m4 stuff is nuts. I had to hardcode -DHAVE_LIBDL here Apr 04 23:13:58 kergoth: what about the URI / PATCH stuff Apr 04 23:14:09 and getting the inbuilt mysqlnd driver working required another patch Apr 04 23:14:14 yeah, should determine if its needed -- this isn't the only case in OE where things are being exported when they shouldn't be, people don't seem to grasp that ${} is expanded by bitbake, not the shell Apr 04 23:14:45 Tartarus: re: URI/PATCH stuff, same comments as for EXTRA_OECONF Apr 04 23:14:55 you could put teh common in the main var, and append for pn-foo and virtclass-native Apr 04 23:15:09 rather than overriding for them Apr 04 23:15:51 SRC_URI / SRC_URI_append_pn-php ? Apr 04 23:17:41 yep Apr 04 23:17:45 k Apr 04 23:17:50 a bit less indirection that way Apr 04 23:17:54 I think I just had done some stupid stuff before coming to what I had there Apr 04 23:18:24 hehe. easy to try a bunch of stuff and end up with a pile of bits you built up to, have to trim back to the essentials often Apr 04 23:21:35 * Tartarus waits for this to finish before doing the final step, upgrade to a less full of (sometimes questionable) security holes Apr 04 23:41:37 03Tom Rini  07master * r9e160ddb72 10openembedded.git/recipes/php/ (8 files in 2 dirs): Apr 04 23:41:37 php: Update to 5.3.6 Apr 04 23:41:37 This is latest and closes various security issues. Apr 04 23:41:37 Signed-off-by: Tom Rini Apr 04 23:41:50 03Tom Rini  07master * r37cd472174 10openembedded.git/recipes/php/ (php-native.inc php-native_5.3.5.bb php.inc php_5.3.5.bb): Apr 04 23:41:50 php: Switch to using BBCLASSEXTEND for php-native Apr 04 23:41:50 Signed-off-by: Tom Rini Apr 04 23:41:56 03Tom Rini  07master * r4abe57b9bc 10openembedded.git/recipes/php/ (7 files in 2 dirs): Apr 04 23:41:56 php: Drop 5.2.13 Apr 04 23:41:56 This is unpinned and has a number of security issues. Apr 04 23:41:56 Signed-off-by: Tom Rini **** ENDING LOGGING AT Tue Apr 05 02:59:58 2011