**** BEGIN LOGGING AT Fri Sep 25 02:59:58 2009 Sep 25 05:10:13 Running task 1947 of 2277 (ID: 1115, /work/build/foo/mainline/os/openembedded/recipes/e2fsprogs-libs/e2fsprogs-libs_1.41.5.bb, do_package_write_ipk) Sep 25 05:10:16 Exception: Message:[Errno 2] No such file or directory: '/work/build/foo/mainline/target/bar/tmp/work/armv5te-angstrom-linux-gnueabi/util-linux-ng-2.15-r3/install/e2fsprogs-libs-dbg' Sep 25 05:11:12 I think its creating an ipk for e2fsprogs-libs, but somehow is looking for its install -directory under util-linux-ng. Sep 25 05:11:25 any ideas what's going on and how to fix this? Sep 25 05:17:52 oh, following on myself. this seems to be bug #5149 Sep 25 05:53:28 hi, I just try to make an patch for checksum.ini, but after doing -> sorting I got an patch with - & +, so I just sorted the checksums.ini without my patch and I got the same + & -, so I think sombody don't sorted it Sep 25 05:53:56 please could some body check this and change it, so I can make a clean patch Sep 25 05:54:32 you still can if your entry is not beside a wrongly sorted one, just manually remove the extra + and - lines Sep 25 05:56:02 okay Sep 25 05:56:27 another question, "gitt diff" now gives me my changed files Sep 25 05:56:33 but not the added files Sep 25 05:56:43 commit your changes and use git format-patch Sep 25 05:56:54 thanks Sep 25 06:43:19 I did now (git add "file in recipe") and than (git commit) but if I do now (git diff) again I just get only my changed checksum.ini Sep 25 06:48:05 you have to "git add" all new and changed files, then "git commit" Sep 25 06:48:26 if you already made an invalid commit, you can start over with "git reset HEAD^" Sep 25 07:09:11 03Koen Kooi  07org.openembedded.dev * rb2144b1630 10openembedded.git/recipes/glib-2.0/ (6 files in 2 dirs): glib-2.0: add 2.22, latest stable Sep 25 07:09:12 03Koen Kooi  07org.openembedded.dev * r2fbcefe967 10openembedded.git/recipes/gtk+/ (5 files in 2 dirs): gtk+: add 2.18.0 Sep 25 07:09:12 03Koen Kooi  07org.openembedded.dev * ra7e6bcb33c 10openembedded.git/conf/machine/omap3-touchbook.conf: omap3-touchbook: also ship xf86-input-tslib Sep 25 07:09:20 03Koen Kooi  07org.openembedded.dev * rc03341eb96 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.org:openembedded into org.openembedded.dev Sep 25 07:09:21 03Koen Kooi  07org.openembedded.dev * r7497b15e9a 10openembedded.git/conf/ (2 files in 2 dirs): angstrom: switch to gtk 2.18 + glib 2.22 Sep 25 07:59:53 morning all Sep 25 08:22:46 03Martin Jansa  07shr/import * rd7f67ac565 10openembedded.git/recipes/ (2 files in 2 dirs): Sep 25 08:22:46 Themes: Fix typo and add shr-theme-neo to task-shr-feed Sep 25 08:22:46 Signed-off-by: Klaus Kurzmann Sep 25 08:22:48 03Martin Jansa  07shr/import * reac738aa2a 10openembedded.git/recipes/xserver-kdrive-common/ (3 files in 2 dirs): Sep 25 08:22:49 Xorg better usb mouse check, add ts_calibration also to Xsession, ie for running Xorg after install without reboot Sep 25 08:22:51 Signed-off-by: Klaus Kurzmann Sep 25 08:22:53 03Martin Jansa  07shr/import * rf19034468d 10openembedded.git/recipes/ (2 files in 2 dirs): Sep 25 08:22:56 Xorg: fix type Sep 25 08:22:58 Signed-off-by: Klaus Kurzmann Sep 25 08:23:00 03Martin Jansa  07shr/import * r89c7276e3d 10openembedded.git/ (6 files in 4 dirs): Sep 25 08:23:03 Latest xorg stuff 2009-09-23 Sep 25 08:23:05 Signed-off-by: Klaus Kurzmann Sep 25 08:23:09 03Martin Jansa  07shr/import * r8cb1e1121d 10openembedded.git/recipes/xorg-lib/ (5 files in 2 dirs): Sep 25 08:23:14 libpciaccesss: segfault fix from git master Sep 25 08:23:16 Signed-off-by: Klaus Kurzmann Sep 25 08:23:18 03Martin Jansa  07shr/import * r4d33871387 10openembedded.git/recipes/xserver-kdrive-common/xserver-kdrive-common/ (Xserver shr/Xserver): Sep 25 08:23:21 xorg: typo in XSERVER value, workaround for segfault when redrawing cursor Sep 25 08:23:23 Signed-off-by: Klaus Kurzmann Sep 25 08:23:27 03Martin Jansa  07shr/import * r7bd7533816 10openembedded.git/ (8 files in 5 dirs): Sep 25 08:23:30 Latest xorg stuff 2009-09-24 Sep 25 08:23:32 Signed-off-by: Klaus Kurzmann Sep 25 09:16:40 good morning Sep 25 09:19:54 03Koen Kooi  07org.openembedded.dev * rf750244ae8 10openembedded.git/recipes/linux/linux-omap_2.6.29.bb: Sep 25 09:19:54 linux-omap -2.6.29: use proper omap3-touchbook file Sep 25 09:19:54 * battery charging works Sep 25 09:19:54 * touchscreen works Sep 25 09:20:05 03Koen Kooi  07org.openembedded.dev * r02b33c980c 10openembedded.git/recipes/linux/ (3 files in 3 dirs): linux-omap-psp: add 2.6.31rc7 recipe Sep 25 09:20:05 03Koen Kooi  07org.openembedded.dev * rdb7815d260 10openembedded.git/conf/machine/include/omap3.inc: omap3: bump MACHINE_KERNEL_PR Sep 25 09:20:06 03Koen Kooi  07org.openembedded.dev * r6cbee2d68d 10openembedded.git/recipes/linux/ (32 files in 2 dirs): linux-omap-pm 2.6.29: add omap3-touchbook support Sep 25 09:24:33 My colleague has an HTC diamond, and looking in http://forum.xda-developers.com/ site he converted it to Android. During startup it use a 2.6.27 kernel compiled using, (I think) Angstrom oe recipe. Anybody knows if there is some link between Android and Openembedded ? Sep 25 09:26:59 recalcati: yes Sep 25 09:29:28 Looking at http://source.android.com/download it seems that it use git,repo, but openembedded dir are not present Sep 25 09:38:18 what is the trick to run some python code from within bitbake.conf ? Sep 25 09:40:06 I want to set a variable based on the stdout from a shell command, and figured I could use subprocess.Popen to achieve it, but I don't seem to be able to get it right :-/ Sep 25 10:13:02 03Koen Kooi  07org.openembedded.dev * r10d8a8386d 10openembedded.git/recipes/linux/linux-omap-pm-2.6.29/omap3-touchbook/defconfig: linux-omap-pm 2.6.29: tweak omap3-touchbook defconfig Sep 25 10:13:02 03Koen Kooi  07org.openembedded.dev * r7ab2e170a7 10openembedded.git/conf/machine/omap3-touchbook.conf: omap3-touchbook: switch to linux-omap-pm Sep 25 10:25:12 'lo woglinde Sep 25 10:25:26 jo rkirti Sep 25 10:26:29 woglinde: hello Sep 25 10:26:33 jo ant Sep 25 10:27:46 woglinde: you met Uli Lukas...I'd have some questions for him..about arm patches gone upstream for kexec-tools Sep 25 10:28:11 I didnt talk with Uli about kexec Sep 25 10:28:20 about Weizen, I know ;) Sep 25 10:28:28 no veltins Sep 25 10:28:38 :=) Sep 25 10:29:02 pilsner Sep 25 10:32:22 lunch Sep 25 10:32:35 passt gut dazu Sep 25 10:32:39 bbl Sep 25 10:32:44 03Koen Kooi  07org.openembedded.dev * r192b425629 10openembedded.git/recipes/pointercal/ (files/omap3-touchbook/pointercal pointercal_0.0.bb): pointercal: add omap3-touchbook support Sep 25 10:32:45 03Koen Kooi  07org.openembedded.dev * reba3242f63 10openembedded.git/recipes/xorg-driver/xf86-input-tslib_0.0.5.bb: xf86-input-tslib: rrecommend tslib-calibrate Sep 25 10:52:39 03Koen Kooi  07org.openembedded.dev * ra8f0ed36b9 10openembedded.git/recipes/ti/ (ti-dmai.inc ti-dmai/touchbook-support.patch ti-dmai_svn.bb): ti-dmai: add omap3-touchbook support Sep 25 11:41:50 re Sep 25 12:05:01 03Koen Kooi  07org.openembedded.dev * r926d4e42d0 10openembedded.git/recipes/ti/ (3 files): ti dsp recipes: replace some _beagleboard depends with _armv7a ones Sep 25 12:05:02 03Koen Kooi  07org.openembedded.dev * rd0fa63e22c 10openembedded.git/recipes/powervr-drivers/omap3-sgx-modules_1.3.13.1607.bb: omap3-sgx-modules: add proc patch for omap3517 and omap3 evms as well Sep 25 12:56:10 How is the initramfs-image supposed to be used? I extract it, edit the scripts and compress it? I I don't want to compile it into the kernel Sep 25 12:57:07 look at the kernel doku Sep 25 12:57:11 or in Sep 25 12:57:41 an oe doc? Sep 25 12:57:41 jo cbrake Sep 25 12:57:45 nope Sep 25 12:57:50 inside the linuxkernel Sep 25 12:58:00 woglinde: hello Sep 25 12:58:49 scripts/gen_initramfs_list.sh Sep 25 12:59:12 well, I know how initramfs works and how to create my own, but as there was already a recipe for one in oe I thought there was some kind of guide on how to use it and modify it for my needs Sep 25 12:59:28 hm oh Sep 25 12:59:36 I just want to add some functionality to it to switch_root into my main fs Sep 25 12:59:37 maybee ant can you help Sep 25 12:59:54 ant? Sep 25 13:00:02 !seen ant Sep 25 13:00:03 woglinde, I don't remember seeing ant. Sep 25 13:00:14 !seen ant_work Sep 25 13:00:15 ant_work (n=andrea@host214-85-static.34-85-b.business.telecomitalia.it) was last seen quitting from #kexecboot 46 minutes ago stating ({"ChatZilla} 0.9.85 {[Firefox} 3.5.3/20090824101458\]\"). Sep 25 13:18:01 hello Sep 25 13:23:38 morning Sep 25 13:26:02 has anyone experienced problems with the bayer2rgb gstreamer plugin ? Sep 25 13:38:11 lol Sep 25 13:38:18 never ever heard of this sink Sep 25 13:49:10 hi pb_ Sep 25 13:51:13 hi pb Sep 25 13:51:15 hi florian Sep 25 13:53:59 hi florian, woglinde Sep 25 13:54:03 I installed the initramfs image besides my kernel and added it to grub. I read the /init script in the initramfs image to figure out how its working. I wanted to know how to configure what fs it should switch_root to. The script uses the $ROOT_BOOT vriable. if it's not set it will print something like "Please add root=/dev/someting to the kernel command line and try again." when it says "kernel command line" does it really mean the "kernel line" in menu.lst (grub Sep 25 13:54:33 jovox yes Sep 25 13:54:40 the kernel line? Sep 25 14:05:25 ah, I read a bit futher in the init script and saw how it works Sep 25 14:07:53 I'm not sure the /proc/cmdline works correctly though.. I think the parsing fails when having the 'rw' flag passed to the kernel Sep 25 14:13:15 na, I was wrong.. something else is not working here.. Sep 25 14:16:56 Should I edit the script and set $BOOT_ROOT Sep 25 14:19:50 jovox just try dont ask so much Sep 25 14:22:26 re kergoth and crofton Sep 25 14:43:56 hi Sep 25 14:44:33 i'm trying to build python for powerpc Sep 25 14:44:59 but it always fails with this message: Sep 25 14:45:21 | /home/scflo/openembedded/tmp/staging/i686-linux/usr/bin/python: error while loading shared libraries: /home/scflo/openembedded/tmp/work/ppc603e-angstrom-linux/python-2.6.1-ml7.0/Python-2.6.1/libpython2.6.so.1.0: ELF file data encoding not little-endian Sep 25 14:45:22 | make: *** [sharedmods] Error 127 Sep 25 14:46:47 my local version is: "Python 2.6.2" Sep 25 14:47:13 does anyone know this problem? Sep 25 14:47:43 thats a new one Sep 25 14:47:46 to me anyway Sep 25 14:49:33 it looks as if it first wants to build version 2.6.1 for local use Sep 25 14:51:37 this is where pb__ normally jumps in with genious solution :-) Sep 25 14:55:29 hmm, that one rings a bell Sep 25 14:55:47 i also tried it on my opensuse machine with python 2.6.0-2.21.1 and there everything worked fine Sep 25 15:00:26 My rootfs is now booting ok through the initramfs. One thing I don't understand though. There is no /etc/fstab in the initramfs-image but something is mounting /dev/hda2 into /mnt.. There is really no problem with that but I would like to understand what's going on here.. the /init (in the initramfs) does not seem to mount the root fs at all Sep 25 15:27:56 i just uploaded the logs from both machines... Sep 25 15:27:57 this one is from my opensuse machine http://pastebin.org/24776 and this is the one that failed http://pastebin.org/24779 Sep 25 15:37:21 Florz2: it seems that the python build process is getting confused between host and target. Sep 25 15:37:41 looks like it's trying to run your powerpc binary on your i686 host, which is doomed to fail. the endianness mismatch will be the least of your problems there :-} Sep 25 15:38:06 I guess you must have some LD_LIBRARY_PATH or rpath problem. Sep 25 15:43:52 ok and how could i fix that? Sep 25 15:44:04 dunno, I think you have to debug this yourself. Sep 25 15:58:36 hi kergoth Sep 25 16:06:16 pb_: hello USA Sep 25 16:08:00 hi lrg Sep 25 16:08:03 ant__: hello europe Sep 25 16:10:35 pb_: I have a strange prob: in kexec-tools they check for the arch, using uname. My armv5te build produces a kernel whose uname is armv5tel.... Solution are 2: a) extend the array of arches[], now only "arm" b) use strncmp and the 3 first chars (arm) instead of strcmp. Sep 25 16:11:16 if I add "armv5tel" strcmp then matches...but how many archies to add? Sep 25 16:11:21 either of those is probably fine Sep 25 16:11:25 personally I would do (b) Sep 25 16:11:31 :) Sep 25 16:11:40 Uli Liukas patch was doing this... Sep 25 16:11:49 as you say, option (a) would lead to a lot of maintenance grief whenever a new arch appears Sep 25 16:11:53 mobving upstream they lost the 'n' in strncpy Sep 25 16:11:58 :/ Sep 25 16:12:19 it is just used to add a constant to the arm syscalls Sep 25 16:13:10 though, I wonder why kexec-tools can't just do this check at compile time (#ifdef __arm__) Sep 25 16:13:32 if the binary is built for arm then it isn't ever going to run on anything except an arm kernel, so checkin at run time seems a bit unnecessary Sep 25 16:14:25 +#ifdef __arm__ Sep 25 16:14:25 +#define __NR_kexec_load<------><------>__NR_SYSCALL_BASE + 347.. Sep 25 16:14:32 ^^ that seems the important bit Sep 25 16:14:57 right, that kind of thing Sep 25 16:15:03 is it valid for all arm*? Sep 25 16:15:13 I suppose yes Sep 25 16:15:15 yes, presumably Sep 25 16:15:19 ok Sep 25 16:15:22 thx Sep 25 16:15:48 presumably you've already tried to get the crazy kernel h4x0rz to define that syscall in the kernel's own headers Sep 25 16:16:09 re Sep 25 16:19:02 pb_: the patch-title would be 'fix insane arch detection'. Too rude? Sep 25 16:19:23 ant__: sounds about right to me Sep 25 16:19:31 ;) Sep 25 16:23:25 hey pb_ Sep 25 17:44:38 pb_: I just can't guess what will happen to other archs... Sep 25 17:47:10 I'll fix the arm case for the moment. can't test others right now Sep 25 17:53:13 morning all Sep 25 17:53:20 morning thesing Sep 25 17:55:28 re Sep 25 17:56:01 hi thesing Sep 25 17:56:04 jo khem Sep 25 17:59:37 03Martin Jansa  07shr/import * rb833462d99 10openembedded.git/recipes/xserver-kdrive-common/ (2 files in 2 dirs): Sep 25 17:59:37 Xorg: Xserver test is only for -mouse arg, USB_MOUSE is for both Sep 25 17:59:37 Signed-off-by: Klaus Kurzmann Sep 25 18:00:17 shr is merging to few back I have the feeling Sep 25 18:03:07 03Koen Kooi  07org.openembedded.dev * r832c767cdb 10openembedded.git/recipes/ffmpeg/ (ffmpeg_svn.bb omapfbplay_git.bb): ffmpeg: bump SRCREV Sep 25 18:03:08 03Koen Kooi  07org.openembedded.dev * rd7b07cab31 10openembedded.git/ (3 files in 3 dirs): mplayer: bump SRCREV to pull in a newer ffmpeg, update vo_omapfb Sep 25 18:08:38 03Koen Kooi  07org.openembedded.dev * r90b2410802 10openembedded.git/recipes/linux/linux-omap-psp/omap3517-evm/defconfig: linux-omap-psp 2.6.31: disable dss2 v4l2 since it breaks compilation Sep 25 18:31:38 otavio: ping Sep 25 18:32:11 methril_home: hello Sep 25 18:33:32 otavio: hello, i just sent you an offtopic e-mail Sep 25 18:33:55 otavio: if you didn't mind to answer some questions? I would be really happy :) Sep 25 18:34:50 methril_home: sure; let me take a look at it Sep 25 18:36:45 otavio: thank you very much Sep 25 18:44:55 hm, the preferred u-boot for beagleboard seems to be unbuildable. that's a bit of a shame. Sep 25 18:48:29 pb_: really? Stable or oe.dev? Sep 25 18:48:36 oe.dev Sep 25 18:49:16 u-boot-2009.05+r29+gitrd363f9cb0918a1b6b92e2e20d01543d0c4f53274-r29 is the offending version Sep 25 18:49:16 pb_: whats the issue? can't say I have had many issues but not built U-Boot for a few days for the Beagle. Sep 25 18:49:41 a bunch of errors in board.c: Sep 25 18:49:46 | board.c:126: error: inline function 'coloured_LED_init' cannot be declared weak Sep 25 18:49:50 and suchlike Sep 25 18:50:51 pb_: thats a problem in uboot upstream Sep 25 18:50:59 exposed with gcc 4.4. Sep 25 18:51:03 x Sep 25 18:51:31 khem: ahhh, good to know, still using 4.3 here Sep 25 18:51:34 pb_: I had a local fix somewhere but then I had to rebuild my laptop and it got los Sep 25 18:51:46 righto Sep 25 18:51:54 it looks like it should be trivial enough to fix, just annoying. Sep 25 18:51:59 yeah Sep 25 18:52:29 maybe a newer u-boot would be less broken. do you knwo if that has been fixed upstream in a more recent version? Sep 25 18:53:03 it was not fixed upstream when I last saw it couple of weeks back Sep 25 18:53:11 ok, never mind Sep 25 18:53:20 * pb_ back to fighting with stupid display controller now Sep 25 18:55:46 pb_: looks like it is one of those days ;-) Sep 25 18:56:56 heh, it seems Sep 25 19:16:34 re Sep 25 19:22:02 hi woglinde Sep 25 19:31:28 uh Sep 25 19:31:51 redhat now has released the virtio-disk driver for windows Sep 25 20:53:24 http://mechomaniac.com/node/32 Sep 25 20:53:35 powered by OE and Angstrom :) Sep 25 20:54:04 we are building stuff for people to do cool shit Sep 25 22:07:02 night all Sep 25 22:14:07 03Andrea Adami  07org.openembedded.dev * r1c1c4a1800 10openembedded.git/recipes/kexec/ (3 files in 2 dirs): kexec-tools_2.0.1: fix insane arch detection [arm], bump pref and PR Sep 25 22:26:02 hi Sep 25 22:26:31 does anyone have gconf problems with menuconfig in ubuntu? Sep 25 22:26:57 trying to do a custom angstrom kernel build, having issues with gconf Sep 25 22:28:57 znJon__: i think you'd better be more specific. Sep 25 22:29:07 menuconfig is a curses app, it doesn't care about gnome **** ENDING LOGGING AT Sat Sep 26 02:59:57 2009