**** BEGIN LOGGING AT Mon Dec 05 02:59:58 2011 Dec 05 06:22:29 gm Dec 05 06:51:07 morning. are there any planned releases for oe-core yet? Dec 05 07:14:30 good morning Dec 05 08:34:59 hi, all! If there are some new version of, for a example gstreamer, I should create new version of bb file by myself? Dec 05 08:37:34 yes and don't forget to send patches to ML Dec 05 08:38:28 JaMa: ok, thanks. Dec 05 08:40:44 tasslehoff: There was an OE-Core release in October, the 2011-1 branch Dec 05 08:42:43 RP__: can PRINC change go in in latest version? some people have issues with meta-smartphone using it already.. Dec 05 08:52:39 mornin Dec 05 08:54:34 Jin^eLD: hi Dec 05 09:03:17 is it possible to tell OE to generate a bootable ext3 image suitable to use with virtualbox or vmware? i.e. not just an ext3 image, but an image with mbr + grub + kernel and an ext3 partition inside? Dec 05 09:05:10 Jin^eLD: there is one with syslinux as bootloader Dec 05 09:05:58 zecke: what do I have to inherit, which class does it? Dec 05 09:07:03 I looked around in the classes dir yesterday but I guess I missed it Dec 05 09:07:09 or was grepping for wrong terms :> Dec 05 09:07:24 Jin^eLD: in directdisk is the term Dec 05 09:08:21 Jin^eLD: core-image-minimal-directdisk, in poky... you specify which 'image' to use and it will create a partition table with syslinux FAT (with the kernel) and EXTX Dec 05 09:10:06 boot-directdisk.bbclass: I think I found it, will have a look at the poky recipe as well Dec 05 09:10:08 thanks! Dec 05 09:12:51 hmm I wonder why livebb recipes have been removed from core? http://lists.linuxtogo.org/pipermail/openembedded-core/2011-July/006942.html Dec 05 09:22:34 hmm Dec 05 09:22:40 when I inherit the class I get NOTE: Error expanding variable do_bootdirectdisk Dec 05 09:29:45 JaMa: and what about patches? Dec 05 09:38:19 RP__: ok. I've asked here before, but I'll ask again. Would you guys say core is mature enough to plan something stable based on it march next year? Dec 05 09:44:12 tasslehoff: oe-core itself is ok, what you'll miss is probably already in meta-oe layer Dec 05 09:57:29 ant_work: yeah. I based my stuff on the angstrom setup scripts, so I have a bunch of layers to pick stuff from :) Dec 05 10:25:56 what's the best way to debug an "Error expanding variable" problem? Dec 05 10:26:53 I am attempting to use the boot-directdisk.bbclass but it does not really work out Dec 05 10:37:08 1. fixed problems in the new image that were discovered on saturday, RHoK coding for the penguin game Dec 05 10:37:15 oops wrong window ;> Dec 05 12:09:25 tasslehoff: when you say march next year, you mean release then or start development then? Dec 05 12:10:22 tasslehoff: I believe it is mature enough for use in products. There are some issues in 1.1 in the sstate and useradd areas, the 1.1.1 point release is being planned to address those Dec 05 12:17:14 morning all Dec 05 12:19:42 hi pb_ Dec 05 12:21:34 RP__: I mean release late march, so stable around mid february I guess Dec 05 12:23:21 hi, is there a quick way to find out which version of OE a OE tree is? Dec 05 12:23:28 is the version in a file or something? Dec 05 12:26:12 arm-angstrom-linux-gnueabi-gcc if I install my SDK to /usr/local, but not if I extract it to another place. how can I check if it looks for libs or something in an absolute location? Dec 05 12:35:00 disregard my question. I think I got it Dec 05 12:35:10 hmm. seems to me the it looks for libc.so.6 in an absolute path in /usr/local. is that the way it should be? Dec 05 12:35:23 disregard dm8tbr and answer me instead ;) Dec 05 13:44:43 tasslehoff: Next release for OE-Core will be around the April mark Dec 05 13:45:10 tasslehoff: so its a difficult call which release would make sense but I think OE-Core should be fine to use Dec 05 13:47:07 RP__: thanks for the info. Dec 05 13:57:33 is it intended that you should/must extract the sdk to /usr/local? Dec 05 13:58:29 ~seen khem Dec 05 13:58:34 khem is currently on #oe #uclibc, last said: 'problem is angstrom does PREMIRROR_append :('. Dec 05 14:51:00 btw I am doing recipes for the Qt 4.8.0 rc in case anyone is interested... I imagine I won't publish them until the final version comes out though Dec 05 15:10:08 hmm, interesting, I have a package that DEPENDS on boost-native, which aborts because boost headers can not be found, and the boost headers are indeed not there; when I bitbake boost-native it starts building gcc-cross-intermediate and eglibc and so on which confuses me Dec 05 15:10:37 why would it need that for a native package? and more importantly, why did this whole thing fail in the first place since DEPENDS was there? Dec 05 15:14:37 that does sound rather odd. what does bitbake -g say? Dec 05 15:15:55 I indeed did -g and it had a mypackage-native -> boost-native entry Dec 05 15:16:19 actually that was the first I checked, also did a -e to see that DEPENDS looks ok there Dec 05 15:23:40 now built boost-native manually, then build srecord-native - works Dec 05 15:23:49 can it be a bug in bitbake? Dec 05 16:19:49 03Joshua Lock  07master * rb8b2f4287c 10bitbake.git/lib/bb/fetch2/__init__.py: (log message trimmed) Dec 05 16:19:49 fetch2: fix Exception parameters when BB_STRICT_CHECKSUM enabled Dec 05 16:19:49 BB_STRICT_CHECKSUM caused a backtrace as the FetchError parameters are Dec 05 16:19:49 incorrectly specified such that FetchError is being passed 8 params Dec 05 16:19:49 when it's expecting 3. This fixes the parameters so we're passing a Dec 05 16:19:49 formatted string and the url. Dec 05 16:19:50 Signed-off-by: Joshua Lock Dec 05 16:20:00 03Robert Yang  07master * r24e9946080 10bitbake.git/lib/bb/command.py: (log message trimmed) Dec 05 16:20:00 Remove the async_cmds and sync_cmds from command.py Dec 05 16:20:00 In bitbake/lib/bb/command.py::Command::__init__, we have the following Dec 05 16:20:00 lines: Dec 05 16:20:00 for attr in CommandsSync.__dict__: Dec 05 16:20:00 command = attr[:].lower() Dec 05 16:20:00 method = getattr(CommandsSync, attr) Dec 05 16:21:28 03Matthew McClintock  07master * r71b53a3f07 10bitbake.git/lib/bb/siggen.py: Dec 05 16:21:28 siggen.py: If both sigs have a variable in it's whitelist then don't say it's changed Dec 05 16:21:28 Some BB_HASHBASE_WHITELIST variables are in the lists of variable Dec 05 16:21:28 dependencies for signatures. Ignore those differences in lists Dec 05 16:21:28 since this difference does not matter Dec 05 16:21:29 Signed-off-by: Matthew McClintock Dec 05 16:21:30 Signed-off-by: Richard Purdie Dec 05 18:03:56 # Supported values are auto, gnome, xfce, rxvt, xcreen, konsole (3.x only), none Dec 05 18:04:07 from default generated local.conf Dec 05 18:04:20 is 'xcreen' ok? Dec 05 18:05:01 * Jay7 is returning to the game slowly :) Dec 05 18:10:11 Jay7: looks like a typo in local.conf.sample... Dec 05 18:20:32 Unable to fetch URL http://kernel.org/pub/linux/utils/kernel/module-init-tools/module-init-tools-3.16.tar.bz2 from any source.' failed Dec 05 18:21:56 404 Dec 05 18:46:59 Jay7: you can add a KERNELMIRROR of your own Dec 05 18:47:12 Jay7: something like KERNELORG_MIRROR = "http://mirror.gbxs.net/pub" Dec 05 18:47:16 in local.conf Dec 05 18:47:32 khem: thanks Dec 05 18:47:32 I dont think all the packages we use are currently reinstated on ko org Dec 05 18:47:55 thats with oe-core btw Dec 05 18:48:13 yes, oe-core Dec 05 19:36:02 khem: ping. I saw hrw mention you when I asked and SDK question earlier today. was that because you are the/a man to pester about problems? Dec 05 19:36:26 * tasslehoff realizes that "no" is a very tempting thing to say :) Dec 05 19:38:19 In my oe-core built SDK, ldd shows that arm-angstrom-linux-gnueabi-gcc (was that right?) looks for libc.so.6 in a hardcoded path in /usr/local. Dec 05 19:39:32 attr[:].lower()? are you shitting me? how the hell did that get into bitbake, that's terrible Dec 05 19:39:39 * kergoth` rolls eyes Dec 05 19:47:47 tasslehoff: sort of Dec 05 19:47:50 :) Dec 05 19:48:27 tasslehoff: dont believe in ldd so much Dec 05 19:48:48 tasslehoff: gcc uses sysroot which should find the libraries relative to sysroot Dec 05 19:48:57 and its also built with a default Dec 05 19:49:08 sysroot if you dont specify one it will use that Dec 05 20:04:30 klibc-1.5.25.tar.bz2 is unfetchable from kernel.org too Dec 05 20:43:00 khem: ok. so my problem is maybe that I usually don't run the environment setup. Dec 05 20:44:20 khem: in oe-classic I could extract wherever I wanted to and set CC correctly. now I have to put it in /usr/local Dec 05 20:55:17 well, now my build env is up again Dec 05 21:13:05 humm I am trying to make a read-only-fs and now I am handling the runtime requirement of update-modules; has anyone sort it out? **** ENDING LOGGING AT Tue Dec 06 02:59:57 2011