**** BEGIN LOGGING AT Sat May 22 02:59:57 2010 May 22 03:51:56 good evening May 22 03:52:28 bug 5438 submitted May 22 03:52:57 it's for "nm" issue I experienced May 22 03:53:41 I set a component as toolchain, cause there was no other that is closer May 22 03:55:30 the image i am making for qemu are giving such error on booting in qemu May 22 03:55:38 root: mount: mounting rootfs on / failed : No such file or directory May 22 03:55:47 how do i resolve this?? May 22 03:56:58 while vfs says VFS : Mounted root ext3 file system ?? May 22 03:57:11 please help May 22 04:03:09 bug 5439 submitted on wxwidgets recipe May 22 04:11:39 bug 5440 submitted on wxwidgets recipe May 22 06:55:32 khem: bug5438 updated, thanx for looking May 22 07:46:54 khem|away: FYI tried gcc-4.5.0+eglibc-2.12+binutils-2.20.1+armv5t(spitz)+2.6.34 and the backtrace seems the same as last time :/ May 22 07:47:29 khem|away: used minimalist image is here http://jama.homelinux.org/org.openembedded.dev.eglibc.images/spitz/Angstrom-minimalist-image-mtdutils-eglibc-ipk-2010.4-test-20100521-spitz.rootfs.tar.gz May 22 13:11:52 While building omap3-console-image iam getting failed in do_rootfs May 22 13:12:00 satisfy_dependencies_for: Cannot satisfy the following dependencies for angstrom-zeroconf-audio May 22 13:12:06 libxtst6 * libxi6 * libxtst6 * libxi6 * libxtst6 * libxi6 * libxtst6 * May 22 13:12:16 opkg_install_cmd: Cannot install package angstrom-zeroconf-audio. May 22 13:12:26 Any one came across this problem May 22 13:15:21 ananth: I'll try here and see what happens May 22 13:15:39 ananth: oops, that must be the gumstix OE tree? May 22 13:15:56 ananth: that image does not seem to be in the main org.oe.dev tree May 22 13:16:16 I'm working on several problems. May 22 13:16:56 minimal-image no longer boots to a log-in prompt May 22 13:19:20 (on at91sam9g20 machine) May 22 13:25:17 yes that belong to gumstix OE tree .... So how to fix that problem any clues May 22 13:26:05 ? May 22 13:38:31 any pointers on how to debug init using dbg from non-interactive command line (via init script) ? May 22 16:01:19 khem: solved quasar build problem. May 22 16:01:34 was an issue with the cross compile patch, like you suspected. May 22 16:01:43 :) May 22 17:29:29 if /var/run is not tmpfs then sometimes ifup will fail on but May 22 17:29:33 just wanted to tell that May 22 17:29:46 ifup complains that ifstate states that interfaces are already configured May 22 17:30:25 re May 22 17:38:45 good morning florian and all May 22 17:40:52 hi IgorK May 22 17:55:13 JaMa|GoNe: ok. thanks for info May 22 18:20:23 hello all, even if you are away May 22 18:56:55 morning May 22 19:36:49 hi May 22 20:26:54 NetRipper, rather than setting the gcc version manually in local.conf we now use minimal distro for leo May 22 20:27:09 ok, i had built the console-image May 22 20:27:27 minimal-image !minimal distro May 22 20:27:33 oh May 22 20:27:53 you can also build console image with minimal distro May 22 20:28:28 ok.. haven't used openembedded before so i'll just let this happen :p May 22 20:28:34 learning as we go May 22 20:29:16 you should check the git diff output May 22 20:29:40 are you familiar with gdb ? May 22 20:29:52 the system is kind of slow lol May 22 20:30:06 i know gdb, but havent used it May 22 20:31:18 ok the problem is that now that we removed udev init is still dieing but don't see at which point May 22 20:31:39 hm ok May 22 20:32:16 the idea was to replace udev by adding devtmpfs and devtmpfs_mount kernel configs May 22 20:32:42 and it works. everything needed in /dev/ is present when root is mounted May 22 20:33:02 but after remounting rootfs init will die. May 22 20:33:36 !oebug 5435 May 22 20:33:37 * * Bug 5435, Status: UNCONFIRMED, Created: 2010-05-19 14:26 May 22 20:33:38 * * lukas(AT)htc-linux.org: machine htcleo: gcc version etc. May 22 20:33:39 * * http://bugs.openembedded.org/show_bug.cgi?id=5435 May 22 20:33:48 odd May 22 20:33:55 yes i read the bug May 22 20:33:55 NetRipper, check my last comment. I know you used the usbnet before on raphael May 22 20:34:08 maybe you can mange to set it up using some custom init script May 22 20:34:15 so we have an interactive debug shell. May 22 20:34:18 oh i didnt see the updated comments May 22 20:34:41 yes i fixed usbnet for raphael back then May 22 20:34:51 shouldnt be too hard for leo May 22 20:35:13 sounds good May 22 20:35:39 but you have a quite bare rootfs. except for /dev/ all the other 'dynamic' things are missing May 22 20:35:49 because lack of normal bootup May 22 20:36:18 will need ifconfig, telnetd (or sshd) May 22 20:36:46 ifconfig and dropbear are present May 22 20:36:51 by default in console-image May 22 20:36:58 ok May 22 20:37:30 I added extra stuff via conf/local.conf (pathnames always relative to ~/openembedded) May 22 20:38:08 so if you need anything extra you can look up the package name in recipes/ and add it in there May 22 20:38:14 ok May 22 20:42:57 NetRipper, what was the failing recipes again? May 22 20:43:07 in your previous build May 22 20:43:53 glib-2.0 May 22 20:44:13 ERROR: Task 1371 (virtual:native:/home/oedevel/openembedded/recipes/glib-2.0/glib-2.0_2.24.0.bb, do_compile) failed May 22 20:45:23 ok. let's just let this build run May 22 20:45:31 see if we get same error May 22 20:45:38 it failed on a libiconv error May 22 20:45:45 yep lets see :) May 22 20:48:05 NetRipper, what's the internet bandwidth ? May 22 20:48:12 100mbit up/down May 22 20:48:15 wow May 22 20:48:38 got a 2tb data limit May 22 20:49:11 if we don't remove sources directory everyday we won't hit that too fast :) May 22 20:49:40 hehe ;) May 22 20:50:15 by the way.. right now wiki points to your machine for 'most recent' haret builds May 22 20:50:32 ok May 22 20:50:37 i think you mentioned that before May 22 20:50:44 oh right, i still have a few todos for haret May 22 20:50:57 yes and you added that it's ok but pointed me to the fact that there's lots of unstable stuff May 22 20:50:59 etc May 22 20:51:06 yea exactly May 22 20:51:10 03Andrea Adami  07org.openembedded.dev * rd011dec899 10openembedded.git/recipes/linux/linux-rp_2.6.26.bb: May 22 20:51:10 linux-rp: fix SRC_URI of poodle_pm-r7.patch May 22 20:51:10 * patch is not yet in ${RPSRC} May 22 20:51:35 latest built can just be something like i hacked a device to be recognized as qsd or something May 22 20:51:49 and the point is that I think we should put some directory with 'clean' builds for haret.exe and haretconsole.tgz in the Kevin2 style May 22 20:52:22 well i think we should release a new version at some point :P May 22 20:52:25 it has quite some changes already May 22 20:53:10 when was the last release? May 22 20:54:23 almost 2 years ago May 22 20:54:32 september 28 2008 May 22 20:54:36 0.5.2 May 22 20:54:54 we should post it in haret-devel May 22 20:54:55 http://linuxtogo.org/cgi-bin/gitweb.cgi?p=groups/haret/haret.git;a=blob_plain;f=WHATSNEW;hb=50ea4ed24207baceb6fc7bb39612b023967454ef May 22 20:55:34 hm my development laptop is unbelievably hot May 22 20:56:23 oh we got an error May 22 20:57:37 i see May 22 20:57:51 it doesnt have an answer May 22 20:57:51 :P May 22 20:58:34 perl-native http://tinderbox.openembedded.net/public/logs/task/5870369.txt May 22 21:00:40 it uploads that log automatically? May 22 21:01:09 yes May 22 21:01:22 we should change the name tho May 22 21:01:38 dcovers_n_netripper_oeteam May 22 21:01:46 hehe May 22 21:04:58 dcordes, locally, are you building under root or a normal user? May 22 21:05:13 lol no normal user May 22 21:05:25 I have no clue why perl-native is giving us that permission error May 22 21:06:16 opendir(./../../../../../../../../..): Permission denied at ../../lib/File/Find.pm line 604 May 22 21:07:02 i'm wondering why it goes down so many levels May 22 21:07:10 hardly seems normal :p May 22 21:07:40 and why /home/oedevel/C ? i'd expect it to be within tmp directory May 22 21:07:52 yes looks strange May 22 21:15:09 check out the warning May 22 21:15:11 no write permission :p May 22 21:15:48 dcordes ^ May 22 21:17:16 no rw permissions on these files.. hence the permission denied i presume May 22 21:17:51 that's strange. I never cam across that. May 22 21:17:53 can i make modifications in the work directory and bitbake perl-native? May 22 21:17:57 to see if it works then? May 22 21:18:07 or will it overwrite May 22 21:18:26 yes let's wait what my approach bumping up preferred version does May 22 21:18:27 or will it restart completely (just asking to prevent waiting another horu :) May 22 21:18:35 ok May 22 21:20:13 check the modification to conf/local.conf May 22 21:21:18 ok May 22 21:21:33 i was working in the other window :) May 22 21:21:36 ctrl-a-a ;) May 22 21:23:45 NetRipper, that prevented me from detaching? May 22 21:24:16 you can always detach May 22 21:24:25 using ctrl-a-d May 22 21:24:32 but if you simply close the shell you'll end up in the other window yes May 22 21:25:00 well that's what I was doing. check query May 22 21:25:29 if you do ctrl-a-w you'll see the windows that are available May 22 23:03:05 kergoth_: any idea what this error means http://pastebin.com/bRX8sKSy May 22 23:03:29 it has been working well so far. Today I did a git pull of bitbake master May 22 23:03:41 which got me two of your latest commits May 22 23:30:09 * kergoth_ looks May 22 23:31:28 i just got bit by that, too. quite strange... lets see.. May 22 23:31:37 oddly, just a few minutes ago i was able to build fine May 22 23:32:06 khem: could you try a git revert of 05462fa79, see if that was it? May 22 23:33:57 the ">" is unbuildable thing looks like something didn't split the deps quite right May 22 23:33:59 for some reason May 22 23:38:32 khem: verified, that was it May 22 23:38:43 no idea why or how, or how i didn't see it with some local changes to OE, but.. May 22 23:38:47 quite odd May 22 23:44:05 khem: well, i'd push the revert if i wasn't having trouble talking to the git server May 22 23:44:06 heh May 23 00:02:03 kergoth_: 05462fa7908fc22988b3dc9d376798d0a46ccb5a is faulty one May 23 00:02:15 In expand, drop the unnecessary second regular expression match May 23 01:07:08 khem: hey man, are you there? May 23 01:35:35 khem: yes, that's what i said :) May 23 01:57:59 kergoth_: yea are you going to revert it ? May 23 02:00:08 kergoth_: would you mind applying http://permalink.gmane.org/gmane.comp.sysutils.bitbake.devel/1182 to bitbake ? May 23 02:03:13 ok, so May 23 02:03:25 I've followed each instruction in openembedded's site May 23 02:03:38 and ended up with /stuff directory and everything good May 23 02:03:45 then I do "bitbake nano" May 23 02:03:56 it went well until step 28, which is compiling coreutils May 23 02:04:29 then it gave me this: May 23 02:04:48 NOTE: Running task 21 of 578 (ID: 176, /stuff/openembedded/recipes/coreutils/coreutils-native_7.2.bb, do_populate_sysroot) May 23 02:04:48 ERROR: Error in executing: /stuff/openembedded/recipes/coreutils/coreutils-native_7.2.bb May 23 02:04:48 ERROR: Exception: Message:[Errno 2] No such file or directory May 23 02:05:08 it says no such file, but it won't give me a hint which file is missing May 23 02:05:16 anyone care to give me a hint on this? May 23 02:07:11 khem: as i said before, i couldn't push it due to issues talkingto the git server. will see if it works now May 23 02:07:13 khem: okay, pushed May 23 02:07:17 the revert, anyway May 23 02:07:38 is patchwork still out of commission? May 23 02:07:50 not applicable here, just wondering May 23 02:08:23 khem: heh, that patch has been in my inbox awaiting my attention since you sent it.. thanks for the reminder :) May 23 02:09:43 khem: k, that's on master too May 23 02:10:24 cool thx kergoth_ May 23 02:10:37 kergoth_: patchwork is working again May 23 02:10:37 np May 23 02:10:40 ah, good May 23 02:11:04 We should add link to bitbake-dev ml on OE wiki May 23 02:11:54 linf: thats not the real error you should check the logs of populate_sysroot for coreutils-native May 23 02:12:33 kergoth: I have locally update gettext to 0.18 and you wouldnt believe how much shit I was piled under May 23 02:14:23 kergoth_: now I have to update coreutils, libiconv, db, gnutls I could have fixed existing ones but I pulled in latest releases and fixed them instead May 23 02:14:42 I think I will put it up in a user branch for others to play around. May 23 02:22:06 linf: Pastebin your error May 23 02:23:56 the whole thing? May 23 02:24:18 let's try... May 23 02:24:20 look up in the log May 23 02:24:35 the logs are in the build tree /temp dir May 23 02:24:41 oh yeah, I've answered that in the hurry, let me try to find the log May 23 02:24:43 they are calle log.do_* May 23 02:24:47 right on May 23 02:26:17 a "find /stuff/ -iname log.do_*" should find it, right? sure about the file name? May 23 02:26:53 in /suff/build there's just the configuration file May 23 02:26:54 um, I'm using jlime, which I'm pretty sure uses open embedded... all I'm trying to do is get asian fonts working. I've got an asian ttf font, but I need to change the encoding to utf8 somehow. localedef doesn't come with jlime, and jlime's locale packages are all messed up. I'm pretty sure localedef is supposed to come with glibc, and I have an arm crosscompiler... but I don't really know where to start :P May 23 02:27:16 and all the /stuff/ directory is recursively owned by the use May 23 02:27:17 r May 23 02:27:48 oh... it build in my current dir May 23 02:28:38 auntie: localedef is packaged as a separate package May 23 02:28:44 auntie: you have to install that May 23 02:29:07 okay, but jlime doesn't even /have/ a localedef package May 23 02:29:12 oh May 23 02:29:19 it has a bunch of locale packages though May 23 02:29:29 auntie: did you search the feeds May 23 02:29:33 but the versions don't match with the feed file May 23 02:29:55 khem: yeah, there's no localedef package... just locale files May 23 02:30:08 http://repository.jlime.com/hp7xx/mongo/feed/locale/ May 23 02:30:15 auntie: hmm jlime should have put it up in feeds May 23 02:30:28 I don't even care about the locale files... I just want to change the encoding May 23 02:32:41 auntie: I think the locales they generate are done on build machine so they did not package localedef for the device May 23 02:32:59 auntie: best is to ask them May 23 02:33:25 or you can take oe stable and try to build it yourself May 23 02:33:48 that may be arduous dont know May 23 02:33:53 me neither May 23 02:34:11 I've used OE briefly before... hopefully I can just generate the locales somehow and copy them over :P May 23 02:34:16 ah well May 23 02:34:37 what I need to do is safe up for a device that's not so ancient :( May 23 02:35:34 auntie: and also depends how desperate you are :) May 23 02:35:53 yes you can easily May 23 02:36:01 that's curious: according to the latest do.log* file (found via 'find', sorted by atime), there isn't any mention about the error, logfile (log.do_install.28641) is like if it succeeded May 23 02:36:16 difficult is though to match the snapshot close to what jlime shipped on your device May 23 02:36:39 khem: should I post it anyway? May 23 02:36:44 linf: check log.do_populate_sysroot May 23 02:36:49 ok May 23 02:37:05 there's no such file May 23 02:37:15 or issue the bitbake nano command one more May 23 02:37:19 in fact, I guess step 28 is before concluding populate_sysroot May 23 02:37:26 and follow the errors as it spits out May 23 02:38:04 it end up saying that there's a file missing, but it doesn't say which May 23 02:38:52 File "/usr/lib/python2.6/subprocess.py", line 1092, in _execute_child May 23 02:38:53 raise child_exception May 23 02:38:53 OSError: [Errno 2] No such file or directory May 23 02:38:53 ERROR: Task 176 (/stuff/openembedded/recipes/coreutils/coreutils-native_7.2.bb, do_populate_sysroot) failed May 23 02:38:53 NOTE: Tasks Summary: Attempted 20 tasks of which 20 didn't need to be rerun and 1 failed. May 23 02:38:53 ERROR: '/stuff/openembedded/recipes/coreutils/coreutils-native_7.2.bb' failed May 23 02:40:02 khem: yikes, that sounds like a pain in the ass (re: gettext) May 23 02:40:14 * kergoth_ is working on removing unnecessary changes to OVERRIDES / calls to update_data from the bbclasses May 23 02:40:52 build failed,a nd i realized i hadn't attempted a complete one before i made my changes to make sure it was my code that broke it :) May 23 02:40:58 * kergoth_ waits on yet another build May 23 02:41:58 kergoth: yeah its a PITA to upgrate GNU tools May 23 02:42:21 linf: thats not enough info to help May 23 02:42:28 and my cystal ball is broken May 23 02:43:25 any other idea where could I find that filename? that would give me a hint of the issue, at least.. May 23 02:45:13 I've set distro to "angstrom-2008.1", but I've never download it, would that be the issue? May 23 02:46:27 no May 23 02:46:47 if thats all you see May 23 02:47:11 then the error is not clear to me May 23 02:47:23 alright, thanks for your time anyway May 23 02:47:48 what's the common value set for that variable? people just leave it as default? (DISTRO var) May 23 02:50:43 no May 23 02:50:52 you have to choose what distro you want May 23 02:50:57 have you read the manial May 23 02:51:00 manual ? May 23 02:51:46 ok, I'll start trying to understand that... I wanted to build something from scratch, not compile something to run in someone else's distro, so I had not took much time on that May 23 02:53:36 yes read getting started May 23 02:53:44 basically there is a lot of info May 23 02:53:48 that you can use May 23 02:53:57 you can try with DISTRO=minimal May 23 02:54:02 that works well too May 23 02:54:34 linf: btw. are you using .dev ? May 23 02:54:49 that's what I was looking for (and I've already read getting started, at least that getting started that instructs you how to install bitbake) May 23 02:55:46 I haven't had the option to decide on that, in fact, all I need for now is to compile a kernel for mini2440 that supports yaffs, since that online builder made almost everything I needed May 23 02:56:07 but then, I could use .dev or kernel's dev as I wish May 23 02:56:13 ok try with DISTRO=minimal May 23 02:56:15 there's no plug-and-play requirements May 23 02:56:56 well you can have a local.conf for any settings you want **** ENDING LOGGING AT Sun May 23 02:59:57 2010