**** BEGIN LOGGING AT Mon May 02 02:59:57 2011 May 02 06:55:33 03Andreas Mueller  07master * r54c90c53bd 10openembedded.git/recipes/maemo4/libgpsbt_0.1.bb: (log message trimmed) May 02 06:55:33 libgpsbt: add dbus-glib to DEPENDS May 02 06:55:33 Error message during configure was May 02 06:55:33 | checking for GLIB... yes May 02 06:55:33 | checking for DBUS... configure: error: Package requirements ( dbus-glib-1 >= 0.23 ) were not met: May 02 06:55:34 | May 02 06:55:34 | No package 'dbus-glib-1' found May 02 07:43:34 good morning May 02 07:56:11 good morning May 02 08:47:48 http://patches.openembedded.org/patch/2801/ May 02 08:47:53 http://patches.openembedded.org/patch/2799/ May 02 08:48:00 http://patches.openembedded.org/patch/2797/ May 02 09:25:39 hi May 02 09:31:48 hi, how can I make a package depend on a dlopened libvorbisidec.so.1 ? May 02 09:32:12 I tried adding tremor to RDEPENDS_${PN} and it doesn't get added May 02 09:46:26 GNUtoo|laptop: i don't understand. What do you mean by "dlopened" ? Do you want to prelink it for configury tests or something or just add build- or runtime depends? May 02 09:46:55 I mean that the app dlopen the library May 02 09:47:09 so it doesn't get picked automatically in RDEPENDS May 02 09:49:50 that would be RDEPENDS, yes May 02 09:50:58 the problem is that libvorbisdec1 is a dynamic package May 02 09:51:23 so I tried to add tremor May 02 09:51:38 like for libpng->libpng3 May 02 09:51:43 03Bernhard Reutner-Fischer  07master * rc3dd4fd5c9 10bitbake.git/doc/manual/usermanual.xml: May 02 09:51:43 manual: fix typo May 02 09:51:43 "is used signify" -> "is used to signify" May 02 09:51:43 Signed-off-by: Bernhard Reutner-Fischer May 02 09:51:50 but that didn't seem to fix it May 02 09:51:51 strange May 02 09:52:15 maybe tremor lacks somehitng May 02 09:52:54 i'm very sorry but i can't really look ATM .. May 02 09:53:02 ok May 02 09:55:49 let me look something May 02 09:56:20 Hi all May 02 09:57:01 Does anybody know why mirrors.openembedded.org is unreacheable? May 02 09:57:33 ahhh May 02 09:57:36 It worked May 02 09:57:38 but I did May 02 09:57:41 bitbake wesnoth May 02 09:57:47 instead of bitbake wesnot-wvga May 02 09:59:05 Anyone? May 02 09:59:56 Hello???????? May 02 10:00:59 no I've no idea May 02 10:01:05 ka6sox-away, ^^^ May 02 10:01:23 OK May 02 12:32:30 03Denis 'GNUtoo' Carikli  07org.openembedded.dev * r12665eb15a 10openembedded.git/recipes/wesnoth/wesnoth.inc: May 02 12:32:30 wesnoth,wesnoth-wvga: RDEPENDS on tremor(libvorbisdec1) May 02 12:32:30 Without that fix we have: May 02 12:32:30 error audio: Could not load music file '[...].ogg': Failed loading libvorbisidec.so.1: libvorbisidec.so.1: cannot open shared object file: No such file or directory May 02 12:32:30 Signed-off-by: Denis 'GNUtoo' Carikli May 02 12:34:12 kergoth_: are you there? May 02 12:35:24 kergoth_: i got a fix for bitbake regarding fetch2 May 02 12:35:37 RP: ^ you might be interested on it too May 02 12:36:18 it fails in case a recipe that uses git into SRC_URI and also uses SRCPV. Then bb.fetch2.get_srcrev fails. May 02 12:36:40 otavio: use bitbake from poky :) May 02 12:36:53 JaMa: I am using git master May 02 12:37:11 if the error says something about urldata doesn't have "branches" property or something like that May 02 12:37:24 JaMa: oh; you got it May 02 12:37:26 I was using git master, but cannot anymore :/ May 02 12:37:30 JaMa: yes, this is the case May 02 12:37:56 JaMa: where I get poky's bitbake? May 02 12:37:57 iirc it's triggered by AUTOREV.. May 02 12:38:05 in poky main repo May 02 12:38:12 /bitbake May 02 13:04:10 03Koen Kooi  07org.openembedded.dev * r7aec50b069 10openembedded.git/conf/distro/include/angstrom-2010-preferred-versions.inc: May 02 13:04:10 angstrom: switch to dbus 1.4.1 May 02 13:04:10 Signed-off-by: Koen Kooi May 02 13:17:32 hi JaMa May 02 13:17:47 do you know the angstrom equivalent of: May 02 13:18:37 /etc/X11/Xsession.d/90xXWindowManager May 02 13:25:19 GNUtoo|laptop: angstrom from meta-angstrom is now using /etc/X11/Xsession.d/90xXWindowManager too, don't know what they used before that May 02 13:25:33 ah ok May 02 13:25:39 I'm still on oe.dev May 02 13:25:51 because of the lack of games in meta-oe May 02 13:26:07 is it the time to switch? what is lacking? May 02 13:27:31 Hi all May 02 13:28:07 hi May 02 13:29:28 GNUtoo|laptop: for development yes.. ie you can add those games you're missing there May 02 13:29:43 ok May 02 13:29:50 since there is no games at all May 02 13:30:00 I'm using qt4-embedded recipe in my image and I've noticed that it installs a lot of things I don't need like qtopia etc etc... May 02 13:30:03 what's the procedure to add one? May 02 13:30:09 in what folder should they go May 02 13:30:29 How can I avoid installing qtopia or other things I don't need ? May 02 13:30:38 recipes-games look reasonable to me May 02 13:30:45 ok May 02 13:30:48 in meta-oe? May 02 13:30:52 y May 02 13:30:58 ok thanks a lot May 02 13:31:02 yw May 02 13:34:28 shr already switched to meta-oe for the images? May 02 13:34:34 *official images May 02 13:41:25 GNUtoo|laptop: there are both images now, actually 3 (shr-t, shr-u, shr-core) May 02 13:41:43 GNUtoo|laptop: and I'm building shr-core only for gta02 and nokia900 May 02 13:41:57 ok May 02 13:44:42 guys how can I exclude packages like qtopia from being installed when including qt-embedded in my image? May 02 13:44:54 I just need qt-embedded libraries May 02 13:45:15 I'm using meta-oe May 02 13:45:23 qt-embedded_4.7.2 May 02 13:45:36 mrAlmond, look in the image recipe you're bitbaking May 02 13:45:59 or in the tasks it includes May 02 13:46:07 or bitbake -g May 02 13:46:56 GNUtoo|laptop: btw: do you understand glib/gdbus-binding-tool/libshr-glib stack? May 02 13:47:22 no, why? May 02 13:47:49 I never looked into it May 02 13:47:56 after last glib upgrade I had to upgrade gdbus-binding-tool too and libshr-glib is failing to build May 02 13:48:05 1) because gdbus-binding-tool command line options changed May 02 13:48:18 2) after fixing it fails because there is no gdbusproxymanager anymore May 02 13:48:53 I think you should ask mickeyl May 02 13:49:00 he did the gdbus migration of fso May 02 13:49:20 and I wasn't able to fix it in libshr-glib and then run out of time and don't know where I'll find time to learn about gdbus and fix it myself May 02 13:49:41 but he is using vala-binding-tool and this problem is only in meta-oe/oe-core May 02 13:49:58 oe.dev had glib old enough :) May 02 13:51:00 mrmoku: ^^ I know you're busy with other stuff now, but do you have any idea about ^^ May 02 13:52:48 2) sounds like the gdbus migration of mickeyl May 02 13:55:22 mickey|office: could you please comment on last ~10 lines? thanks :) May 02 13:55:42 * JaMa done with highlighting May 02 13:55:46 and off May 02 14:13:10 anyeone having trouble with network manager ? May 02 14:15:42 JaMa: ok, will take a look May 02 14:18:27 JaMa: what version of glib do we have in meta-oe/oe-core? May 02 14:18:42 would 2.28.6 be new enough to be able to check on my laptop? May 02 14:19:04 2.28.6 May 02 14:19:54 * mickey|office reads the backlog May 02 14:20:11 and gdbus-binding-tool rev 286f4eaf48ea79821be5a4e6bf493566b156faa4 http://git.openembedded.org/cgit.cgi/meta-openembedded-contrib/commit/?h=shr&id=6028a6b7f81e317f229cae33bccd765ee1c4320a May 02 14:20:22 JaMa: good, will try to make it work then May 02 14:20:29 mickey|office: guess it's more kind of my problem :) May 02 14:20:36 ya, looks like it ,) May 02 14:24:27 note path is not quite right May 02 14:24:30 http://pastebin.com/Rnv1ZjGm May 02 14:28:32 Crofton|work: check libnl-1.0.pc May 02 14:28:51 Crofton|work: and last few patches for network-manager May 02 14:31:34 valhalla, hi May 02 14:32:00 GNUtoo|laptop: hi May 02 14:33:02 valhalla, can I PM you? May 02 14:36:19 JaMa: looks like gdbus-codegen has problems with methods having only one output arg May 02 14:41:59 GNUtoo|laptop : Thank you for the help...I was already looking at the qt-embedded recipes and .inc files but then I've noticed that the main problem for me is not qtopia but the whole qt documentation that is installed in the image : 250 MB of docs!!! May 02 14:42:16 mrAlmond, ok May 02 14:42:28 is this normal for a recipe made for an embedded system? May 02 14:42:55 installing also the docs is not needed May 02 14:42:57 IMHO May 02 14:44:13 mrmoku: generated .c code had extra ',' after last parameter? May 02 14:45:08 yup May 02 14:45:15 actually .h May 02 14:45:27 I've tried to fix it manually and then I just got another error May 02 14:45:52 ok May 02 14:45:58 :/ May 02 14:46:30 JaMa: will try to find out later tonight... May 02 14:46:31 maybe there is newer rev in gdbus-binding-tool and this fixed May 02 14:47:16 mrmoku: I've tried only what was latest on iirc Thursday.. May 02 14:47:25 mrmoku: thanks May 02 14:51:08 JaMa: trying current master May 02 14:53:03 JaMa: heh, does not work at all May 02 14:53:03 File "/usr/local/lib/gdbus-codegen/codegen.py", line 201, in declare_types May 02 14:53:04 self.h.write('#define %sTYPE_%s (%s_get_gtype ())\n'%(i.ns_upper, i.name_upper, i.name_lower)) May 02 14:53:07 AttributeError: Interface instance has no attribute 'ns_upper' May 02 14:53:25 ah :/ May 02 14:54:25 at least we can say that gdbus-codegen is changing quickly.. so hopefully it will be fixed soon :) May 02 14:55:50 hmm 286f4eaf48ea79821be5a4e6bf493566b156faa4 I see as latest in master and it's from 2011-04-08 May 02 14:56:04 are you using this repo? http://cgit.freedesktop.org/~david/gdbus-binding-tool/ May 02 14:56:44 JaMa, sounds like a preferred varesion issue May 02 14:57:01 Crofton|work: only 2 nm versions were fixed.. so maybe yes May 02 14:57:30 Crofton|work: but from your pastebin I have no idea which version is failing for you :) May 02 14:57:40 yeah May 02 14:59:25 mrmoku: here is something maybe newer http://git.gnome.org/browse/glib/log/?h=gdbus-codegen May 02 14:59:27 JaMa: indeed... might be python3 hitting me May 02 14:59:47 ahh... that's good May 02 15:00:51 hmm and it's at least partially merged to master May 02 15:01:18 so maybe glib in oe-core is too old instead :P May 02 15:02:00 even 2.29.2 is too old for this stuff :) May 02 15:02:05 :) May 02 15:02:14 so maybe in next stable 2.30 we will see such miracles May 02 15:02:26 * mrmoku cloning glib repo to try the one from gdbus-codegen branch May 02 15:04:09 mrmoku: great, if that works I can update gdbus-binding-tool recipe to use glib repo before we have new enough glib itself May 02 15:05:35 yup May 02 15:07:33 networkmanager-0.8.3.999-r0: task do_compile: Failed May 02 15:16:08 Crofton|work: extend 0001-respect-libnl-flags-also-in-backends-ip6-manager.patch to cover dns-manager Makefile.am too, May 02 15:20:27 At #yocto ... [10:37] < otavio> I am looking at porting some stuff I have done in OE to OE-core and I am trying to find where 'kernel-abiversion' went May 02 15:20:35 Someone has a glue about it? May 02 17:02:19 is there sth like virtual/init? May 02 17:51:44 JaMa|Off, ping May 02 18:27:28 What is the proper way of sending some patches to oe-core? Just mail them to oe-core ml? May 02 18:35:35 yes.. May 02 18:36:10 if they are very large, references to an external tree might be more appropriate.. but for most things simply using git to pull out the patches and setting up a summary is enough May 02 19:00:03 Crofton|work: pong May 02 19:00:24 looks like your suggerstion worked May 02 19:00:31 good May 02 19:00:38 do you want me to add an extra patch, or update the existing one? May 02 19:00:46 didn't you build libnl_git in between? May 02 19:00:51 no May 02 19:00:55 I mean some libnl-2.0 version May 02 19:00:59 I do not think so :) May 02 19:01:03 not on purpose May 02 19:01:09 good... please update existing one May 02 19:01:14 k May 02 19:01:14 as it's the same problem May 02 19:01:25 I'll add my sob to the patch May 02 19:01:31 will rebase mine into yours May 02 19:01:43 maybe also update 0.8.2 version or where it was also added May 02 19:02:02 iirc koen cherry-picked that version to maintenance branch May 02 19:02:10 the patch is in networkmanager, not a version speccific directory May 02 19:02:39 even better if it applies cleanly in both May 02 19:02:56 hmm, we'll need to check that ... May 02 19:09:49 I'm configuring VM's from the Wiki instructions. What happened to the openembedded-essential meta-package? http://www.openembedded.org/dl/packages/rpm/openembedded-essential-1.1-1.noarch.rpm May 02 19:40:35 How shall get Qt to be built with GLES support? I added 'PREFERRED_PROVIDER_qt4-x11-free = "qt4-x11-free-gles"' to my image recipe. But it still builds qt4-x11-free.bb instead of qt4-x11-free-gles.bb May 02 19:41:27 I put the PREFERRED_PROVIDER_qt4-x11-free in the DISTRO recipe not the IMAGE recipe. May 02 19:48:21 khem, the SSD is full...are we going to leave it that way? May 02 19:49:33 ka6sox *g* May 02 19:49:50 chn whats your machine? May 02 19:51:38 I'm building for a custom machine and have copied machine recipe from another board and modified it. Have also copied Angstrom distro files to set up a specific preferred version list. May 02 19:52:19 I can get a full system to be built with a working Qt on X but not with GLES configured. May 02 19:53:35 ka6sox-away, canyou talk with otavio about github sometime May 02 19:53:46 ChN and you have gles libs? May 02 19:54:06 Crofton|work, sure May 02 19:55:28 That might be something I have missed to specify in the build? What I do have is Xorg gles libraries as binaries from the board vendor (Freescale) copied to the target system for test. May 02 19:55:31 thaks May 02 19:56:08 I think he has done the homework and could set it up, but it makes sense for the infrastructure guys to be in the loopp May 02 19:56:20 he just happens to have a use case for github :) May 02 19:56:39 chn you need the dev libs too May 02 19:56:49 otherwise qt-gles cannt be compiled May 02 19:57:51 And then the qt4-x11-free-gles recipe will not be ran at all? May 02 19:58:03 you can run it by hand May 02 19:58:08 bitbake qt4-x11-free-gles May 02 20:00:02 Shouldt that recipe run if I have set 'PREFERRED_PROVIDER_qt4-x11-free = "qt4-x11-free-gles"' in my distro recipe? And then generate a build error if it misses libraries? May 02 20:01:18 ah right May 02 20:01:19 yes May 02 20:01:32 http://patches.openembedded.org/patch/2801/ May 02 20:01:37 http://patches.openembedded.org/patch/2799/ May 02 20:01:42 http://patches.openembedded.org/patch/2797/ May 02 20:01:48 http://patches.openembedded.org/patch/2795/ May 02 20:01:50 dv? May 02 20:02:11 fray: sent them May 02 20:02:19 woglinde_, me May 02 20:02:28 ya, I saw them come across.. May 02 20:02:54 normally I would have said you need more in the commit message.. but these are all fairly obvious so for typo like fixes it's not really needed.. May 02 20:03:27 woglinde_, bitbake qt4-x11-free-gles does not work since qt4-x11-free-gles.bb provides qt4-x11-free May 02 20:03:44 fray: yeah; the typo fixes I usually try to keep them small and trivial as possible May 02 20:03:50 chn yes May 02 20:03:51 fray, it was comment for me? May 02 20:04:03 PREFFRED_PROVIDER was correct to set May 02 20:04:22 nope.. otavio May 02 20:04:22 is someone working on syncing the bitbake of master with poky one? as JaMa say's the fetcher2 has some issues with git that seems fixes in poky's bb May 02 20:04:46 work is being done -- but a few key folks are still on vacation, they get back this week.. May 02 20:05:02 if you know of specific issues, make sure to send them to the oe-core list so they don't get ignored May 02 20:05:04 fray: ? nope for the bb question I guess May 02 20:05:29 woglinde_, any ideas about why it might not use the qt4-x11-free-gles.bb? May 02 20:05:50 I the annoying department NOTE: package qt4-x11-free-gles-4.7.2-r24.7: task do_compile: Failed May 02 20:05:55 yes people are working on it.. but it's as problem are found.. most of the on-going work is in the UI component sync -- unrelated to the fetcher May 02 20:07:06 ChN maybee some other setting is overwriting it May 02 20:07:12 grep the conf dir May 02 20:08:09 Have done that for a while now :-( will look futher with fresh eyes tomorrow into that specific. May 02 20:08:22 the other thing you can do is file a defect in the Yocto Project bugzilla... just be sure to mention it's an oe-core / poky sync issue.. May 02 20:09:06 Crofton|work, regarding qt4-x11-free-gles-4.7.2, are there known issues that might also exist for 4.7.1? May 02 20:09:19 not sure May 02 20:10:08 If one want to build the gles dev libraries, where is that enabled? May 02 20:11:07 chn normaly gles libs are vendor specific May 02 20:11:23 there are something from mesa May 02 20:11:35 but they are doing only softrendering May 02 20:59:36 jo ant May 02 21:00:04 hi woglinde_ May 02 21:05:15 otavio: ping May 02 21:05:54 otavio, ping May 02 21:07:58 ka6sox-away: I only ping him about his patch for compression management in bitbake. Patch still in limbo... May 02 21:08:46 woglinde_: you reported it too iirc. pls ack then otavio can push May 02 21:09:08 ant__, I only ping him about the repo clones he wants. May 02 21:09:27 yes, much more important stuff today :) May 02 21:10:35 ant__, I didn't see your ping...I was just realizing we needed to get this done ASAP. May 02 21:12:15 otavio just answered you May 02 21:12:37 well, 4 hours ago May 02 21:12:50 iirc he's in Brazil, afternoon now May 02 21:13:18 he...time for a Caipirinha May 02 21:15:46 ant__, ya dinner here. May 02 21:16:12 heh, im about to leave work and go home and drink a skol ;) May 02 21:19:30 ant__: hi :-) May 02 21:19:55 ant__: about? May 02 21:20:01 lzma May 02 21:20:07 hexor: are you brazilian too? May 02 21:20:16 it seems so May 02 21:20:22 hiya otavio May 02 21:20:31 ka6sox-work: :-) May 02 21:20:48 what are we trying to do? and how can I help? May 02 21:22:04 ka6sox-work: about github? May 02 21:24:25 ant__: i didn't get your reply May 02 21:24:45 ant__: btw, the commit is waiting for ack/pushing at https://projetos.ossystems.com.br/git/?p=users/otavio/org.openembedded.dev.git;a=summary May 02 21:24:59 otavio: about your patch for compression management in bitbake.conf. Patch still in limbo... May 02 21:25:20 ant__: most patches I sent latelly are on limbo ;-) May 02 21:25:27 needs a second ack May 02 21:25:36 then I'll push it if nobody May 02 21:25:55 otavio, yes, Github May 02 21:26:20 otavio, I just need a way to keep the mirror in sync. May 02 21:26:25 same for the S20modutils.sh -> S02modutils.sh change..pending in oe-dev and oe-core it seems May 02 21:26:49 ka6sox-work: they do that. May 02 21:26:52 then understand if S03udev is wrong or not May 02 21:26:55 ka6sox-work: we just need to create the orga May 02 21:27:03 thats already done May 02 21:27:13 I can talk to them about that... May 02 21:27:17 let me do that. May 02 21:27:18 ka6sox-work: good. May 02 21:27:28 what repos do you need? oe-core? meta-oe? May 02 21:27:48 ka6sox-work: user is one thing May 02 21:27:52 ka6sox-work: organization is another May 02 21:28:00 ka6sox-work: I'd say to add an organization May 02 21:28:02 no, I understand that. May 02 21:28:16 I wanted to start with a user and then I'll go to organization. May 02 21:28:20 ka6sox-work: I just found an user called openembedded May 02 21:28:29 I don't think this will become our "official" May 02 21:28:56 ka6sox-work: sure but having an organization doesn't hurt and avoids someone to make it May 02 21:29:10 ka6sox-work: I'd say to go with an orga to be at safest side May 02 21:29:37 this is mainly for the convenience of github May 02 21:29:39 ka6sox-work: you can keep clear on the repositories description they're a mirror and not official ones May 02 21:29:39 03Denis 'GNUtoo' Carikli  07org.openembedded.dev * r75599d20f9 10openembedded.git/recipes/prboom/prboom_2.5.0.bb: (log message trimmed) May 02 21:29:39 prboom: fix compilation May 02 21:29:39 Without that fix we have: May 02 21:29:39 arm-oe-linux-gnueabi-gcc [...] -o prboom [...] -L/usr/lib -lSDL -lpng -lm May 02 21:29:39 /usr/lib/libSDL.so: file not recognized: File format not recognized May 02 21:29:39 This commit was inspired by commit ba5581f2258fccaf87245f5672c20d3cc43d1849 : May 02 21:29:40 smpeg svn: fix compilation May 02 21:29:42 saves them storage May 02 21:29:56 Crofton|work: and time for users forking it May 02 21:30:15 yes May 02 21:30:27 it is a read only mirror, correct? May 02 21:31:21 Crofton|work: yep May 02 21:31:47 okay I get that now...converting user to organization. May 02 21:33:03 ka6sox-work: nice May 02 21:33:28 hey folks - trying to construct a recipe for a project set up a bit strange. The work is split in 3 packages, such that 'b' depends on 'a' and 'c' depends on both 'a' and 'b' - I've got the recipes set up, but if I simply try to build 'c' when none of the 3 are built yet, I am getting an error when building 'b' because 'a' isn't yet built (it is in progress, since bitbake is running multithreaded) May 02 21:33:42 am I missing something obvious to fix this? May 02 21:34:02 okay so we aren't planning on doing anything but mirroring this and then letting folks fork it and us pull patches? is that the idea << otavio? May 02 21:34:06 ka6sox-work: as repositories I'd say oe-core, meta-oe and bitbake May 02 21:34:17 ka6sox-work: yes May 02 21:34:44 ka6sox-work: plain oe I'd say to not mirror since we don't want people to keep working on it May 02 21:35:01 ka6sox-work: so making things harder seems the best alternative for it May 02 21:36:52 otavio, lets ask kergoth about bitbake....that one isn't in my perview. May 02 21:37:03 meta and core...np May 02 21:37:41 khem ping May 02 21:37:48 ka6sox-work: personally I see no problem in adding it but OK May 02 21:38:32 fine with me May 02 21:38:42 okay thats all I needed. May 02 21:38:43 i use github for my personal bitbake branches anyway :P May 02 21:38:51 ka6sox-work: https://github.com/kergoth/bitbake <= he uses it already May 02 21:39:00 ka6sox-work: so having it on the orga seems logical May 02 21:39:15 ka6sox-work: :-) May 02 21:39:18 kergoth: :-P May 02 21:39:20 otavio, having his personal isn't the same. May 02 21:39:28 this would be the mainline. May 02 21:39:36 okay let me work on getting that setup. May 02 21:39:51 converting first then creating the repos I want mirrored. May 02 21:41:24 hrms actual system on uclibc needs the future branch May 02 21:41:26 kergoth, the official is on berlios correct? May 02 21:41:31 no May 02 21:41:35 berlios svn is toast May 02 21:41:38 official is oe git May 02 21:41:47 oh, makes it simpler then...thanks May 02 21:41:51 release tarballs go on berlios, thats about it May 02 21:41:56 kk May 02 21:42:17 okay bbl...meeting May 02 21:42:20 kergoth: did you notice my problem with bb fetch2? May 02 21:42:28 kergoth: about git autoinc May 02 21:42:48 heh, i avoid anything remotely related to srcrev/srcpv/etc May 02 21:43:45 kergoth: :-( ok May 02 21:44:18 * kergoth is personally of the opinion that autorev is an abomination that should go away in favor of something like srctree, manage the git of a development tree outside of bitbake May 02 21:48:30 kergoth: I sort of agree however it is nice to be able to generate an image that uses the current source of packages May 02 21:48:41 kergoth: and this is a quite common thing where I work May 02 21:48:45 kergoth: we use it daily May 02 22:14:28 argh...'b' DEPENDS on 'a', yet if I build 'b' it errors out due to going too far in the build process before 'a' is finished (am running bitbake multithreaded if it makes a difference) ... what am I missing here? May 02 22:22:52 hbeck: when b depends on a, b's do_configure will run after a's do_populate_sysroot May 02 22:24:09 kergoth: That's what I would expect, but with DEPENDS_${PN} = "a" in recipe for b, I wasn't seeing that behavior. Trying to clean and re-run now to see if something was just goofed up May 02 22:24:25 DEPENDS is a recipe variable May 02 22:24:33 DEPENDS_${PN} will do absolutely nothing May 02 22:24:36 you're mixing up DEPENDS and RDEPENDS May 02 22:27:45 khem: do you see any problem with the busybox-mdev patch? May 02 22:30:06 kergoth: well, that's dumb on my part :P May 02 22:48:51 otavio: no, I wanted to make sure you tested it in that combination May 02 23:04:13 khem: ok May 02 23:18:38 khem, can I have the SSD so I can re-partition it? May 02 23:24:27 kergoth, I agree wrt to srctree. I have used srctree and it is much better than autorev May 02 23:52:55 hi guys, just a general oe question. I can cross compile ruby fine, but I want oe to deploy my cross compiled ruby in the target sysroot so that when I build ruby extensions it uses that ne instead of the native ruby which results in extensions being compiled for the wrong architecture. **** ENDING LOGGING AT Tue May 03 02:59:58 2011