**** BEGIN LOGGING AT Tue Dec 02 02:59:57 2008 Dec 02 03:23:30 03Khem Raj 07org.openembedded.dev * r146a0901eb 10openembedded.git/conf/distro/include/sane-srcrevs.inc: sane-srcrevs: Bump sane uclibc svn version which compiles Dec 02 05:48:11 03Jeremy Lainé 07org.openembedded.dev * r90d0da93bd 10openembedded.git/packages/linux/ (3 files in 3 dirs): linux-2.6.26: cleanup backport of Freescale NAND controller Dec 02 06:08:28 03Jeremy Lainé 07org.openembedded.dev * r08e1a194be 10openembedded.git/packages/linux/ (9 files in 2 dirs): linux-2.6.26: update patches for boc01 machine Dec 02 08:31:38 morning :) Dec 02 08:32:15 I have strange error while configuring gcc-cross-4.2.4-r3: http://tinderbox.openembedded.net/public/logs/2335534.txt Dec 02 08:32:33 is it about my host GMP/MPFR? Dec 02 09:17:39 03Koen Kooi 07org.openembedded.dev * r6570429cc9 10openembedded.git/packages/xorg-driver/xf86-video-omapfb_git.bb: xf86-video-omapfb: bump SRCREV for scaling tweak Dec 02 09:17:49 03Koen Kooi 07org.openembedded.dev * ra4d8beced4 10openembedded.git/packages/linux/linux-omap_git.bb: linux-omap git: bump SRCREV Dec 02 09:17:50 03Koen Kooi 07org.openembedded.dev * r69a5214dff 10openembedded.git/packages/linux/ (40 files in 8 dirs): linux-omap: add recipe to build 2.6.27 Dec 02 09:17:50 03Koen Kooi 07org.openembedded.dev * ra9f44be1cb 10openembedded.git/packages/linux/ (66 files in 6 dirs): linux-omap: add recipe to build 2.6.26 Dec 02 09:45:49 03Koen Kooi 07org.openembedded.dev * r478b479535 10openembedded.git/packages/gcc/ (gcc-4.2.4/gfortran.patch gcc-cross_4.2.4.bb): gcc: gcc 4.2.4 needs a different gfortran patch that 4.3.x Dec 02 09:55:34 03Koen Kooi 07org.openembedded.dev * r6c24f35983 10openembedded.git/conf/machine/omap3evm.conf: omap3evm: change XSERVER to Xorg Dec 02 09:56:34 03Koen Kooi 07org.openembedded.dev * rac0c17f9e9 10openembedded.git/packages/angstrom/angstrom-x11-base-depends.bb: angstrom-x11-base-depends: bump PR Dec 02 09:59:38 hm.. looks like Koen fix this now Dec 02 09:59:43 another try Dec 02 10:37:00 03Koen Kooi 07org.openembedded.dev * rd75e198291 10openembedded.git/packages/u-boot/u-boot_git.bb: u-boot git: update neuros git repo and bump SRCREV Dec 02 10:46:14 03Koen Kooi 07org.openembedded.dev * ra35d1d02f9 10openembedded.git/packages/linux/linux-neuros_git.bb: linux-neuros: update git repo and bump SRCREV Dec 02 10:48:42 morning Dec 02 11:09:04 mickey|ICxyz hi, I tried calling and sim access from cli-framework yesterday Dec 02 11:09:58 with success? Dec 02 11:10:09 no Dec 02 11:10:52 that's sad Dec 02 11:10:53 mickey|ICxyz on calling attempt I received something like Dec 02 11:10:55 what did it do? Dec 02 11:11:18 singleline doesn't support thisa nd that Dec 02 11:11:43 i need full logs Dec 02 11:12:58 and be sure to update Dec 02 11:12:59 i backported the new call handling to singleline over the weekend Dec 02 11:18:58 03Koen Kooi 07org.openembedded.dev * r54d6feaee5 10openembedded.git/ (conf/checksums.ini packages/policykit/policykit_0.9.bb): policykit: add 0.9 from poky Dec 02 11:40:44 morning all Dec 02 11:41:09 hey! Dec 02 11:41:14 how are you? Dec 02 11:41:35 zecke|assignment: Not too bad thanks :) Dec 02 11:41:56 zecke|assignment: How's things? Dec 02 11:42:22 hehe, I could complain... but I think I'm fine Dec 02 11:42:42 mickey|ICxyz I tried with the latest changes yesterday. wil provide you ith details tomorrow Dec 02 11:44:33 good Dec 02 11:44:41 hi rp Dec 02 11:45:02 Poky is finally publicaly on git FWIW :) Dec 02 11:45:20 cool. that should allow for better mergin Dec 02 11:59:04 03Paul Sokolovsky 07org.openembedded.dev * r3cabd6a460 10openembedded.git/packages/initrdscripts/ (files/87-kexecboot.sh initramfs-module-kexecboot_1.0.bb): initramfs-module-kexecboot: Fix kexec argument order so it works for any libc. Dec 02 12:07:04 * * OE Bug 4894 has been created by j.ongstreum(AT)gmail.com Dec 02 12:07:06 * * oops in pxa_ir_configure Dec 02 12:07:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4894 Dec 02 12:10:51 03Koen Kooi 07org.openembedded.dev * rc651ec19a2 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Dec 02 12:11:02 03Koen Kooi 07org.openembedded.dev * r8e5f8306ae 10openembedded.git/packages/packagekit/packagekit_git.bb: packagekit git: make it work again with the current opkg Dec 02 12:15:04 * * OE Bug 4878 has been RESOLVED (FIXED) by lukas.gorris(AT)gmx.de Dec 02 12:15:06 * * msm machine update/cleanup Dec 02 12:15:08 * * http://bugs.openembedded.net/show_bug.cgi?id=4878 Dec 02 12:22:48 03Paul Sokolovsky 07org.openembedded.dev * r7127c56047 10openembedded.git/packages/initrdscripts/ (3 files in 2 dirs): Dec 02 12:22:48 dropbear-backdoor: Set of files to provide passwordless root access for dropbear. Dec 02 12:22:48 * For DEBUGging images only. Dec 02 12:22:48 * Factored out from devimage_0.1.bb. Dec 02 12:22:51 03Paul Sokolovsky 07org.openembedded.dev * rdd79d20fda 10openembedded.git/packages/busybox/ (6 files in 4 dirs): Dec 02 12:22:53 busybox-static 1.11.3: Initial add, configs are copies of non-static version. Dec 02 12:22:55 * Configs will be patches in the next commit. Dec 02 12:22:57 03Paul Sokolovsky 07org.openembedded.dev * rbba10d357a 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Dec 02 12:23:00 03Paul Sokolovsky 07org.openembedded.dev * rb3eb70b5bb 10openembedded.git/packages/initrdscripts/ (files/30-bootmenu.sh initramfs-module-bootmenu_1.0.bb): initramfs-module-bootmenu: If dropbear available, start it if user drops to shell. Dec 02 12:23:06 03Paul Sokolovsky 07org.openembedded.dev * r2acd436f83 10openembedded.git/packages/busybox/ (4 files in 4 dirs): busybox-static 1.11.3: Actually enable static option. Dec 02 12:25:22 03Koen Kooi 07org.openembedded.dev * r31ecfb66d9 10openembedded.git/ (2 files in 2 dirs): policykit-gnome: add 0.9.2 Dec 02 12:25:40 psokolovsky: yay Dec 02 12:26:16 mickey|ICxyz: Hi! ;-) Dec 02 12:29:35 03Paul Sokolovsky 07org.openembedded.dev * r9b72d50c65 10openembedded.git/packages/klibc/klibc-utils-static_1.5.bb: Dec 02 12:29:35 klibc-utils-static 1.5: modprobe should live in /sbin Dec 02 12:29:35 * That's where various things expect to find it, in particular, Dec 02 12:29:35 kernel's kmod. Dec 02 12:29:45 03Paul Sokolovsky 07org.openembedded.dev * r3fd6991edd 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Dec 02 12:42:39 03Koen Kooi 07org.openembedded.dev * r62c026a92c 10openembedded.git/ (conf/checksums.ini packages/packagekit/packagekit_0.2.4.bb): packagekit: add 0.2.4 release Dec 02 13:04:12 grr Dec 02 13:04:32 gcc-cross_4.2.4 still fails to compile :\ Dec 02 13:05:09 *to configure Dec 02 13:05:15 http://tinderbox.openembedded.net/builds/62448/ Dec 02 13:06:56 03Paul Sokolovsky 07org.openembedded.dev * r5a453542d6 10openembedded.git/packages/images/initramfs-bootmenu-debug-image.bb: initramfs-bootmenu-debug-image: A version of initramfs-bootmenu-image with dropbear included. Dec 02 13:06:59 03Paul Sokolovsky 07org.openembedded.dev * r46f87cafbc 10openembedded.git/packages/tasks/task-devimage.bb: task-devimage: Pull in kernel symlink. Dec 02 13:07:03 03Paul Sokolovsky 07org.openembedded.dev * r479aacd6ae 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Dec 02 13:07:11 03Paul Sokolovsky 07org.openembedded.dev * r71f6e144f3 10openembedded.git/packages/images/initramfs-bootmenu-image.bb: initramfs-bootmenu-image: busybox' modprobe is broken, so pull klibc's. Dec 02 13:07:44 here is log: http://tinderbox.openembedded.net/public/logs/2339495.txt Dec 02 13:18:50 psokolovsky: if you see, the first try was to implement in klibc-static all the binaries you used for uniboot...last addition was 'wc...then we stopped against 'expr' Dec 02 13:22:58 ant_work: thanks, that's helpful! Dec 02 13:23:22 thesing: hi Dec 02 13:23:42 morning all Dec 02 13:23:43 thesing: Hi, please note: klibc-utils-static 1.5: modprobe should live in /sbin Dec 02 13:23:43

* That's where various things expect to find it, in particular, Dec 02 13:23:43

kernel's kmod. Dec 02 13:24:08 psokolovsky: you know, it seems it requires regexp, so no way to fit the small size Dec 02 13:24:37 ok, shouldn't be a problem for now ;-) Dec 02 13:24:39 03Koen Kooi 07org.openembedded.dev * r3ee7033007 10openembedded.git/packages/packagekit/gnome-packagekit_0.2.5.bb: gnome-packagekit: add 0.2.5 Dec 02 13:25:13 psokolovsky: at first we tried to reuse your init :-) Dec 02 13:25:35 taht's a bit of history for the records Dec 02 13:25:40 ant_work: I remember, we talked with thesing about trying to use klibc shell and utils Dec 02 13:25:43 ok Dec 02 13:28:36 03Koen Kooi 07org.openembedded.dev * r2a278ecd21 10openembedded.git/packages/pointercal/ (files/omap3evm/pointercal pointercal_0.0.bb): pointercal: add file for omap3evm Dec 02 13:28:59 ant_work: helo Dec 02 13:29:25 Jay7: have you installed fortran on your host? Dec 02 13:30:02 ant_work: it this error comes from host system? Dec 02 13:30:40 no, seems that I have no fortran Dec 02 13:31:24 long ago I had issues on Gentoo without fortran Dec 02 13:31:33 should have been fixed iirc Dec 02 13:31:40 but even... Dec 02 13:31:46 ant_work: hm.. should I install fortran? :) Dec 02 13:31:54 I'd try Dec 02 13:32:27 ok, lets try Dec 02 13:36:22 03Paul Sokolovsky 07org.openembedded.dev * r08ade1ed2e 10openembedded.git/packages/linux/linux-handhelds-2.6.inc: Dec 02 13:36:22 linux-handhelds-2.6: Add more verbose message about KERNEL_INITRAMFS_PATH. Dec 02 13:36:22 * Also, fix zImage file name in generated HaRET's startup.txt. Dec 02 13:36:24 03Paul Sokolovsky 07org.openembedded.dev * rbefb7d7e82 10openembedded.git/packages/linux/linux-handhelds-2.6-2.6.21/ (24 files in 24 dirs): Dec 02 13:36:27 linux-handhelds-2.6-2.6.21: Update from defconfigman r446 Dec 02 13:36:29 * Consistently enable serial8250 module. Dec 02 13:36:31 03Paul Sokolovsky 07org.openembedded.dev * rb2ef01929c 10openembedded.git/packages/linux/ (25 files in 25 dirs): Dec 02 13:36:34 linux-handhelds-2.6_2.6.21-hh20: Disable CONFIG_LOCALVERSION_AUTO. Dec 02 13:36:36 * This options is used to stick kernel's git rev to kernel version, Dec 02 13:36:38 but if the kernel is not from git, it can instead randomly pull up rev Dec 02 13:36:40 from some other repo. Dec 02 13:36:42 03Paul Sokolovsky 07org.openembedded.dev * rd898dd3fea 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Dec 02 13:38:20 Jay7: hmm...probably I built gcc-cross just before the last 2 koen's gfortran patches Dec 02 13:39:09 ant_work: I have installed fortran. now trying once more :) Dec 02 13:40:00 no, still fails Dec 02 13:40:12 psokolovsky: could you please expand the concept of 'defconfiigman'? Is it only for linux-hh ? Dec 02 13:40:54 btw, I can't search config.log.. Dec 02 13:41:13 ant_work: I proposed interested folks to look into it for adopting for other kernelsets. hrw even did once ;-) Dec 02 13:41:34 ant_work: http://defconfigman.sourceforge.net/ Dec 02 13:44:52 morgen Dec 02 13:45:05 hrw: lupus in fabula Dec 02 13:46:01 seems interesting.. Dec 02 13:46:21 gcc-cross-4.2.4 tries to configrure with fortran?.. Dec 02 13:52:44 Jay7: look at oebug 4130 Dec 02 13:53:28 !oebug 4130 Dec 02 13:53:29 * * Bug 4130, Status: NEW, Created: 2008-03-29 07:35 Dec 02 13:53:30 * * dmoulder(AT)qtome.com: gcc-cross 4.2.2 does not configure due to problems with gmp and mpfr Dec 02 13:53:31 * * http://bugs.openembedded.net/show_bug.cgi?id=4130 Dec 02 13:58:49 psokolovsky: I did more then one attempt to defconfigman Dec 02 13:59:32 ant_work: I have posted comment to this oebug Dec 02 14:01:09 I suppose after installing fortran you have to rebuild from scratch (mpfr-native,...) Dec 02 14:01:17 hm.. Dec 02 14:01:33 ok, np :) Dec 02 14:02:08 but config.log absence is strange.. Dec 02 14:03:33 btw Dec 02 14:03:50 how can I poing bitbake to my git repo Dec 02 14:04:07 now it can't get metadata info Dec 02 14:04:16 *point Dec 02 14:08:55 Jay7: if you are lazy (as me), don't change your scripts / local conf Dec 02 14:09:10 just checkouit in a new dir called org.openembedded.dev Dec 02 14:09:36 and then create a local branch Dec 02 14:13:30 git clone git://git.openembedded.net/openembedded org.openembedded.dev Dec 02 14:13:51 git checkout -b your-local-branch origin/org.openembedded.dev Dec 02 14:13:58 IIRC Dec 02 14:14:12 hmhm.. Dec 02 14:14:17 ~curse dd-wrt for using crappy shell ipkg Dec 02 14:14:19 May you be reincarnated as a Windows XP administrator, dd-wrt for using crappy shell ipkg ! Dec 02 14:16:18 Jay7: http://git.or.cz/gitwiki/GitCheatSheet -> Create/Branch Dec 02 14:22:40 ant_work: I have just checkout without -b option Dec 02 14:22:54 so I have branch named 'master' Dec 02 14:28:18 well, if you don't have commit-rights, no probs I think Dec 02 14:28:36 I feel sure committing in my branch :/ Dec 02 14:32:07 ah.. Dec 02 14:32:10 hm.. Dec 02 14:32:29 this is reason.. Dec 02 14:33:11 thesing: can you remember why there is sleep 3 in the init of initramfs-kexecboot? Dec 02 14:33:48 thesing: I commented it and still kexecboot seems to work... Dec 02 14:36:32 ant_work: I have talk about METADATA_BRANCH = "" and METADATA_REVISION = "" Dec 02 14:36:44 I've seen, in oestats Dec 02 14:36:49 bitbake tries to get git metadata from somewhere Dec 02 14:36:55 fatal: Not a git repository Dec 02 14:37:14 ant_work: no idea. Dec 02 14:59:55 ant_work: did you start bitbake in you local branch directory? Dec 02 15:08:09 Jay7: I use my old scripts / env-setup. I start building in /oe/build Dec 02 15:08:20 hm.. Dec 02 15:08:33 how did you clone the repo? Paths? Dec 02 15:09:42 ant_work: I've looking on your oestats at tinderbox Dec 02 15:10:01 03Steffen Sledz 07org.openembedded.dev * r2d263abd29 10openembedded.git/conf/checksums.ini: checksums.ini: add ethtool-6 (additional location) Dec 02 15:10:02 you provided metadata variables somehow :) Dec 02 15:10:24 iirc if you just clone, it creates an /oe/openembedded directory Dec 02 15:10:37 you can give destdir Dec 02 15:11:01 ant_work: no, I have git init && git add Dec 02 15:11:18 only for oe.dev Dec 02 15:11:43 ah.. Dec 02 15:11:50 I have understood Dec 02 15:12:01 I have the same tree in /oe like before with mtn (plus .git stuff in org.openembedded.dev Dec 02 16:33:25 I'm still pretty green to git, but I was wondering if there's a way to have a branch that's trimmed down to only contain the packages/recipes that I actually use, so that I don't have to wait for 6,271 recipes to be parsed every time I want to do a build, but that still gives me updates to those packages when I git-pull. is that possible? Dec 02 16:36:17 aside from git rm? Dec 02 16:36:26 you could rm the ones you dont want in your branch, but maintaining it would be a nightmare due to conflicts. you could use filter-branch to filter out the stuff you dont care about entirely, but you couldnt update it. i dont -think- you could do it with a subtree merge either, i think subtree merge only lets you do a subtree on one end of the merge, not both Dec 02 16:37:15 kergoth, conflicts? shouldn't git just say "dead file updated, don't care / ok." ? Dec 02 16:37:16 i dont know of a clean easy way to do what you want, offhand. of course, you could just make your BBFILES only contain the packages you care about, or you could mask out everything but what you want with BBMASK, or.. Dec 02 16:37:37 BBMASK will still parse them tho Dec 02 16:37:41 just not have them be usable Dec 02 16:37:47 I'd swear :) Dec 02 16:38:13 Tartarus: itll say upstream wants to make change foo to file A, but file A is removed here, conflict, you get to resolve it manually, most likely. but even if it did handle it the way he wants, which i dont think it does, he'd still have to keep removing newly added files in the package's FILESDIR/FILESPATH Dec 02 16:39:37 Tartarus: BBMASK prevents the parsing. Dec 02 16:39:50 Tartarus: it did when i wrote it, and it does today, see line 765 in lib/bb/cooker.py Dec 02 16:40:34 kergoth: can I BBMASK ? Dec 02 16:40:52 hmm, i'm not sure if python's regex is capable of that, off the top of my head, phobbes Dec 02 16:40:55 would have to look it up Dec 02 16:41:42 but you could do that with bbfiles. packages/(a|b|c|d|e|f|g)/*.bb Dec 02 16:42:28 phobbes: it would be nice to be able to do what you want with a branch though, i remember researching that myself for something unrelated to oe.. might have to look into it again Dec 02 16:42:47 kergoth: ah, BBFILES is cleaner. (rather then the double-negation of BBMASK) Dec 02 16:42:55 * kergoth nods Dec 02 16:42:58 kergoth: thanks Dec 02 16:43:15 well, (a|b|c) wont work i dont think, its a glob not a regex, but you could still likely do packages/{a,b,c,d,e}/*.bb like in shell Dec 02 16:43:21 which should be close enough Dec 02 16:43:32 heh Dec 02 16:43:40 * kergoth isn't quite awake yet Dec 02 16:43:41 git-filter-branch could just "remove" everything from upstream... Dec 02 16:43:48 okay. bbl Dec 02 16:44:00 zecke: yes, i already pointed that out, but he coucouldn't update it cleanly, he'd have to keep doing it ;P Dec 02 16:44:00 zecke: I'll look into it, thanks. Dec 02 16:44:34 since you cant merge from an unfiltered branch to a filtered branch Dec 02 16:44:41 due to all the rewritten commits Dec 02 16:45:20 kergoth: did you want me? Dec 02 16:45:39 phobbes: one possibility would be to remove the packages you dont want, and write a custom merge driver script for the merges to that branch specifically, which filters out changes to the removed files Dec 02 16:45:53 RP: if i did, i have no idea what i wanted now :) Dec 02 16:46:05 kergoth: :) Dec 02 16:46:13 how's it going at intel? Dec 02 16:46:29 ohand bits i mean Dec 02 16:46:49 kergoth: I don't mind waiting for git to download updates, mostly I just want to avoid parsing them. sounds like BBFILES will do what I want. Dec 02 16:46:55 * kergoth nods Dec 02 16:47:04 kergoth: Its an interesting ride with some bumps but I guess thats to be expected ;-) Dec 02 16:47:08 indeed Dec 02 16:50:03 hey RP :) how is life? Dec 02 16:51:50 man i'm rusty at debugging race conditions. i need to get less rusty in general Dec 02 17:04:38 Jin^eLD: Could be worse :) Dec 02 17:11:52 :) Dec 02 17:17:43 RP: can you give a hint about how to debug some mysterious bitbake logs? oebug 4784 Dec 02 17:21:26 RP: bitbake is like HAL 9000, talks only to Dr. Chandra ...err...Purdie Dec 02 17:31:41 ant_work: I have stuff about openzaurus-sa too, but kinda as warning, not error Dec 02 17:48:29 psokolovsky: I heavily suspect a module backported from 2.4 for collie or poodle. I don't remember exactly Dec 02 18:01:35 bbl Dec 02 18:02:27 configure: error: GMP 4.1 and MPFR 2.2.1 or newer versions required by fortran Dec 02 18:03:07 gcc-cross-4.2.4-r3 cannot be configured Dec 02 18:03:44 !oebug 4130 Dec 02 18:03:44 * * Bug 4130, Status: NEW, Created: 2008-03-29 07:35 Dec 02 18:03:45 * * dmoulder(AT)qtome.com: gcc-cross 4.2.2 does not configure due to problems with gmp and mpfr Dec 02 18:03:46 * * http://bugs.openembedded.net/show_bug.cgi?id=4130 Dec 02 18:04:12 is anybody here have such problems? Dec 02 18:04:54 nobody else has mentioned such a thing, and I don't think gcc-cross 4.2.4 has changed recently. Dec 02 18:05:29 do you, in fact, have gmp 4.1 and mpfr 2.2.1 or newer? Dec 02 18:11:21 pb_: on host system? Dec 02 18:11:37 in staging Dec 02 18:13:41 pb_: mpfr and gmp have installed before Dec 02 18:14:07 okay. then you need to work out why configure isn't finding them. Dec 02 18:14:07 libs installed into staging/i686-linux/usr/lib Dec 02 18:14:27 oh.. Dec 02 18:15:32 btw, I don't see config.log in work/armv5te-angstrom-linux-gnueabi/gcc-cross-4.2.4-r4/gcc-4.2.4 Dec 02 18:16:27 you wouldn't; it's in the build subdirectory Dec 02 18:17:25 ah.. yes.. Dec 02 18:17:55 /usr/bin/ld: cannot find -lc Dec 02 18:18:03 o_O Dec 02 18:22:20 is there anything else I'm supposed to be doing besides incrementing "FILE_PR" in my recipe in order to trigger a rebuild? Dec 02 18:23:38 or am I supposed to use "PR" instead? I'm not sure how FILE_PR snuck in there Dec 02 18:28:34 phobbes1: AFAIK FILE_PR should be gone now. Just use PR. Dec 02 18:33:17 thesing: thanks. Dec 02 18:53:22 hi Dec 02 18:53:25 ~seen khem Dec 02 18:53:28 khem is currently on #oe (23h 49m 27s) #uclibc (23h 49m 27s), last said: 'check unistd.h '. Dec 02 19:16:58 hi kalemas Dec 02 19:23:02 woglinde hi, it was inconvenient situation with zecke yestarday. Dec 02 19:38:19 if I install files to ${D}${datadir}/my_appname in a do_install_append() function, for an autotools-inheriting recipe, what's the proper way to tell bitbake to include those files when it does its packaging? They end up in ${WORKDIR}/images, but not in the rootfs Dec 02 19:40:11 phobbes search for FILES_${PN in the recipes Dec 02 19:41:59 woglinde: thanks, I'll check it out Dec 02 19:53:26 woglinde: that did it, thanks! Dec 02 19:55:35 phobbes1 no prob Dec 02 20:15:52 Jay7: fyi, I've just built gcc-cross-4.2.4-r4 w/o issues Dec 02 20:16:19 psokolovsky: which version of gcc have you on host system? Dec 02 20:16:37 Jay7: gcc version 4.1.2 (Gentoo 4.1.2 p1.0.2) Dec 02 20:16:50 hm.. same version Dec 02 20:16:58 i586-alt-linux-gcc (GCC) 4.1.2 20070626 (ALT Linux, build 4.1.2-alt2) Dec 02 20:17:42 psokolovsky: thanks for info, I'll go deeper :) Dec 02 20:28:12 hello, I already asked but no-one has given me the thing to put in the config to generate packages(ipk) for each depedencies... Dec 02 20:28:20 so what is it? Dec 02 20:28:35 GNUtoo RDEPENDS Dec 02 20:28:37 (there is even someone that told me that it didn't worked for him) Dec 02 20:28:39 ok thanks Dec 02 20:28:39 please read the manual Dec 02 20:28:51 shared library are discoverd automagic Dec 02 20:28:58 woglinde, no ipks Dec 02 20:29:13 *sigh* Dec 02 20:29:13 woglinde, what I means is create the ipk...not doing staging dir Dec 02 20:29:32 what is your question? Dec 02 20:29:42 have a dependency in a ipk Dec 02 20:29:50 or split a package into serval ipk Dec 02 20:30:02 more than oe does normaly Dec 02 20:30:13 woglinde, no...generate ipks for each deps...actually when i bitbake a package it generates an ipk only for that package but not for its deps Dec 02 20:30:46 *sigh* Dec 02 20:30:54 put all in task Dec 02 20:31:26 so there is no better ways? Dec 02 20:31:34 and thanks for the response Dec 02 20:31:54 nope Dec 02 20:32:11 ok thanks a lot Dec 02 20:41:40 valhalla, hello, I had no comment on bug 4834 for a long time...you said you would have try to (cross)compile boost with zlib for wesnoth...(but you were busy playing wesnoth(on your pc)...lol)did it work on your box? Dec 02 20:41:58 i didn't have time to work on it Dec 02 20:42:08 ah ok...np Dec 02 20:42:15 nor to play wesnoth on the pc :) Dec 02 20:42:35 i;m going to try it as soon as i can Dec 02 20:42:48 ok thanks Dec 02 20:42:56 hi elena Dec 02 20:43:18 hi woglinde Dec 02 20:43:44 GNUtoo why it shouldnt compile with zlib? Dec 02 20:43:51 hmhm.. Dec 02 20:43:55 it even works with bz2 Dec 02 20:44:00 hi jay7 Dec 02 20:44:14 * Jay7 found that here is no i686-linux/lib dir Dec 02 20:44:22 woglinde: hello :) Dec 02 20:45:02 woglinde, it compiles...if you remove '-sNO_ZLIB=1' \ otherwise you have that: http://bugs.openembedded.net/show_bug.cgi?id=4834#c9 Dec 02 20:46:06 woglinde, but If it was commited as it is...it must have been tested and it must have been working...so i suspect that the commiter used another libboost version than mine Dec 02 20:46:47 hm which version you are using? Dec 02 20:46:48 s/remove/don't remove Dec 02 20:47:08 PREFERRED_VERSION_boost ?= "1.33.1" Dec 02 20:47:15 urgs Dec 02 20:47:19 stone age Dec 02 20:47:23 use the newer one Dec 02 20:47:30 ok Dec 02 20:48:10 pplz, which libc should be taken by gcc-cross while configuring? Dec 02 20:48:11 so should the newer libboost be hardcoded in the bb file? such as specifying a minimum version... Dec 02 20:48:21 from i686-linux? Dec 02 20:48:22 no Dec 02 20:48:35 there is no support for this gnutoo Dec 02 20:48:38 s/don't remove/you have to/ Dec 02 20:48:38 *from staging/i686-linux Dec 02 20:48:46 woglinde, ok thanks a lot... Dec 02 20:49:10 om is to conservative at updating stuff Dec 02 20:49:31 ok...mabe I should change distro...and make my own images? Dec 02 20:49:37 and ?= means you can ovverride it Dec 02 20:49:50 yes...in the config? Dec 02 20:50:01 in your local.conf Dec 02 20:50:03 ok Dec 02 20:51:45 * Jay7 needs a bit of time of gcc guru :) Dec 02 20:54:16 good I am not guru Dec 02 20:55:06 and i bet there is no support for detecting which boost version is (cross)compiled...(so no way of specifying a different behavior for each boost version in the wesnoth bb file) Dec 02 20:55:16 so I should close the bug... Dec 02 20:57:44 http://paste.org.ru/?ctzsgr Dec 02 20:57:50 here is config.log Dec 02 20:59:32 libc is not located in any of -L paths Dec 02 21:00:08 should this be one from host system? Dec 02 21:10:55 hi Dec 02 21:11:04 who is graeme gregory? is he on here by any chance? Dec 02 21:12:39 i have a rather nasty input bug which seems to be caused by some commits of his upgrading the libx11 package. There is one about a keysym patch which looks a likely candidiate Dec 02 21:13:37 oliford hm then try it out Dec 02 21:14:14 hm.. Dec 02 21:14:44 hm? Dec 02 21:15:02 oliford xora is graeme I think but he isnt here right now Dec 02 21:15:07 I have i586-alt-linux-gnu, but gcc-cross tries to configure with --build=i686-linux --host=i686-linux Dec 02 21:15:09 ok Dec 02 21:15:19 Jay7 hm Dec 02 21:15:24 oliford: this is other 'hm' :) Dec 02 21:15:52 thanks, i'll pester him later, maybe file a bug report and just set the preferred_version in the mean time Dec 02 21:16:04 just my distro builds packages for i586 by default Dec 02 21:16:17 can this be a root of problem? Dec 02 21:16:39 Jay7 yes Dec 02 21:17:11 good :) Dec 02 21:17:31 but now here is other question.. what I should do?.. Dec 02 21:18:09 find the error? Dec 02 21:20:00 03Stanislav Brabec 07org.openembedded.dev * r4ff159c627 10openembedded.git/conf/checksums.ini: Added checksums for geda and emacs. Dec 02 21:20:02 03Stanislav Brabec 07org.openembedded.dev * r909edb6ddf 10openembedded.git/packages/guile/guile_1.8.5.bb: guile: Package Emacs lisp files as guile-el. Dec 02 21:20:05 03Stanislav Brabec 07org.openembedded.dev * r7d09f87695 10openembedded.git/packages/geda/ (15 files in 2 dirs): geda: Updated and packaged complete toolchain version 1.4.1. Dec 02 21:20:08 03Stanislav Brabec 07org.openembedded.dev * rb9b0f4343e 10openembedded.git/packages/guile/guile_1.8.5.bb: guile: Created cross-compilation capable guile-config. Dec 02 21:27:11 he ant Dec 02 21:27:29 ant___: hi Dec 02 21:27:52 woglinde hi Dec 02 21:27:52 hello Dec 02 21:28:58 woglinde: I'll check the console-image uclibcgnueabi, TZ settings and all Dec 02 21:29:19 if it is ok, the problems are in gpe (in fact opie-image seems immune) Dec 02 21:39:39