**** BEGIN LOGGING AT Mon Feb 18 02:59:58 2013 Feb 18 08:22:50 good morning Feb 18 08:47:51 JaMa: in mountain view? Feb 18 08:48:07 ah, it was 7h ago ;D Feb 18 09:18:41 the openembedded-devel list doesnt allow non-subscribers to write to it? Feb 18 09:20:35 iirc yes Feb 18 09:21:59 meh... :/ Feb 18 09:24:14 josch: too many spammers I'm afraid :( Feb 18 09:24:24 morning all Feb 18 09:24:49 algright, subscribing and writing email again :) Feb 18 10:58:21 JaMa (and others): any plans of getting qt5 into OE soon? meta-qt5 doesn't seem to be "all done". Feb 18 11:00:45 By that I mean that building qtbase gave me ipk's that may work on target, but I miss sdk-stuff so I can test it :) Feb 18 11:18:45 how do add a different defconfig for busybox into my own layer ? Feb 18 11:31:07 rob_w: replacing the existing one Feb 18 11:50:15 is a .bbappend respected when building a recipe directly via -b ? Feb 18 11:52:50 rob_w: no, it won't be Feb 18 11:53:02 -b should be used sparingly in any case Feb 18 11:53:12 k thx Feb 18 11:54:54 can we build kernel using oe-core in which initramfs is embedded in it? Feb 18 11:55:00 ooh, pseudo performance tweaks. I'm definitely looking forward to that! Feb 18 11:55:20 Zagor: yes they do sound very good Feb 18 11:55:43 Noor: kexecboot does this; see meta-initramfs Feb 18 11:56:07 bluelightning: thanks Feb 18 11:58:53 bluelightning: do you know about someone who used OE to build cross compiler for ms windows? Feb 18 12:00:18 hrw: I don't, I'm afraid... though I'm sure I recall we had mingw recipes in oe-classic Feb 18 12:02:01 hrw uh why you need mingw? Feb 18 12:08:41 bluelightning: we are doing same kind of thing but having loops in rcipes Feb 18 12:09:05 this is due to kernel-image is added in rdepends in module.bbclass Feb 18 12:09:28 now kernel is dependent on image and image is dependent on linux recipe Feb 18 12:10:09 Noor: not sure what the correct way to resolve that one would be, but it can't be too hard Feb 18 12:16:14 woglinde: to build cross compiler for windows users? Feb 18 12:20:40 woglinde, ping Feb 18 12:22:40 Noor: EXTRA_IMAGEDEPENDS = "" Feb 18 12:26:08 ant_work: In my filesystem I have some kernel modules ... so that is why Feb 18 12:29:30 I see. I'd say, build your initramfs image additively because doing it in a subtractive way seems almost impossible Feb 18 12:36:42 Hi all Feb 18 12:43:24 Hi ascor Feb 18 12:52:26 blindvt whats up? Feb 18 12:52:54 hrw you have so many customers? Feb 18 12:53:02 on windows? Feb 18 12:53:29 woglinde, can you please relicense your uClibc execvpe patch as LGPLv2.1+ (i.e. add "or later")? Feb 18 12:55:45 woglinde: at Linaro we provide cross compilers for linux and windows. For now they are built with Crosstool-NG. I just wonder how much effort it would be to use OE for it Feb 18 12:56:16 woglinde: hi Feb 18 12:56:20 blindvt' hm did I not it is still in the bugtracker? Feb 18 12:56:52 blindvt' hm I just copied it from the other files by the way Feb 18 12:56:57 jo zecke Feb 18 12:59:00 woglinde, yes, it is here: bugs.uclibc.org/1465 If you agree on the "or later", i will read the patch and change it accordingly before applying it. Feb 18 13:00:33 wrong bugnumber Feb 18 13:00:59 woglinde, oops. 3829 Feb 18 13:01:46 Can you join wikipedia today ? Feb 18 13:03:11 ugh for two includes Feb 18 13:09:05 blindvt is Licensed under the LGPL v2.1 and later enough? Feb 18 13:09:21 woglinde, perfect Feb 18 13:09:43 woglinde, wait. I think it should be "or later", not and Feb 18 13:11:30 ah yes Feb 18 13:14:41 blindvt' have a look Feb 18 13:15:36 woglinde, excellent, thanks alot! Feb 18 13:16:31 woglinde, i guess that's the same patch as lives in oe, isn't it? Feb 18 13:16:36 yes Feb 18 13:17:12 should I relicense it too? Feb 18 13:17:26 or just wait until new upstream and it goes away anyway Feb 18 13:17:41 woglinde, you don't need to, i'll apply it to the stable branch too, so it will go away in oe. Feb 18 13:18:02 okay .) Feb 18 14:23:03 CalledProcessError: Command 'tar -cf - -C /tmp/del/gcc46-arm7/build/tmp-eglibc/work/armv7ahf-vfp-neon-oe-linux-gnueabi/gcc-cross-initial/linaro-4.6-r5/pkgdata -ps . | tar -xf - -C /tmp Feb 18 14:23:06 /del/gcc46-arm7/build/tmp-eglibc/pkgdata/armv7ahf-vfp-neon-oe-linux-gnueabi' returned non-zero exit status 2 with output tar: /tmp/del/gcc46-arm7/build/tmp-eglibc/work/armv7ahf-vfp-neo Feb 18 14:23:09 n-oe-linux-gnueabi/gcc-cross-initial/linaro-4.6-r5/pkgdata: Cannot chdir: No such file or directory Feb 18 14:23:15 ops Feb 18 14:23:23 anyway - I do wonder why it fails Feb 18 14:51:46 hrw: hmm, looks like pkgdata hasn't been populated for some reason Feb 18 14:52:13 bluelightning: nearly no one is using gcc 4.6 so it may be broken Feb 18 15:05:25 hi all Feb 18 15:12:45 hi pb_ Feb 18 15:13:07 pb_: do you have access to delete branches in the meta-openembedded repo? Feb 18 15:13:24 I accidentally pushed a branch there instead of -contrib and now I can't get rid of it Feb 18 15:52:37 bluelightning: I can try to delete the branch -- what is the name? Feb 18 15:52:49 cbrake: paule/python-bits Feb 18 16:05:23 bluelightning: ok, its gone. Git fetch does not seem to clear it out, but a clean clone shows its not there Feb 18 16:05:57 cbrake: thanks Feb 18 16:06:14 cbrake: I think you need to use -p - I've done that and it showed as gone from here Feb 18 16:07:29 bluelightning: yes, that did it, thanks! Feb 18 17:00:51 hrw: yes the one in Mountain View Feb 18 17:01:12 JaMa: nice place - was there once and would love to be again ;) Feb 18 17:02:47 it was closed 3y ago when I wanted to go there, so it was good to fulfil my 3y old plan :) Feb 18 17:04:46 ;) Feb 18 17:08:48 * ensc|w wishes, rp would take care about correct quoting... http://cgit.openembedded.org/bitbake/commit/?id=96088ebdec08e49ba9e8dbcac437bfcdc21f5983 changes behavior and gives bad surprises when files with '*' in their name are removed Feb 18 17:10:10 ensc|w: point that on ML Feb 18 17:10:11 ? Feb 18 17:11:08 bitbake ml is somehow broken for me :( gmane mirror (which I am using) has some postings, other are missing :( Feb 18 17:13:16 gmane was never updated to the new bitbake ml Feb 18 17:13:21 that repalced the old berlios one Feb 18 17:13:29 just subscribe to the new list, i'd say Feb 18 17:17:50 I do not like maillist traffic in my mailbox; nntp is a better solution Feb 18 17:19:41 good luck with taht Feb 18 17:40:05 hi, in the same recipe, I'd like to get a difference do_install for native and my target, is that possible ? Feb 18 17:40:14 different Feb 18 17:40:29 use overrides. grep for class-native and class-target Feb 18 17:44:14 morning Feb 18 17:49:23 moin Feb 18 18:55:29 do we have any usage info for a toolchain from -c populate_sdk? **** ENDING LOGGING AT Tue Feb 19 02:59:58 2013