**** BEGIN LOGGING AT Sun Jul 29 02:59:56 2007 Jul 29 03:42:41 03slapin 07org.oe.dev * r51a7fb0d... 10/ (8 files in 3 dirs): palms: Removed obsolete pxapwm-bl driver from machine configs Jul 29 03:42:45 03slapin 07org.oe.dev * r7809af1d... 10/ (7 files in 7 dirs): palms: Removed obsolete pxapwm-bl driver from defconfigs Jul 29 03:57:00 bluelightning: re: OPIE 0001081. I haven't had a zaurus with a working battery in long time, so I haven't been able to do any testing Jul 29 03:57:54 W8TVI: ok... seems like it was an upstream bug anyway Jul 29 03:58:00 from the comments Jul 29 03:58:19 yeah Jul 29 03:58:25 (I'm just closing off old previously resolved bugs atm) Jul 29 03:59:21 yeah Jul 29 03:59:50 I'm planning on getting a new battery for my two zauruses Jul 29 04:00:04 seems the price has come down Jul 29 04:01:15 ERROR: Importing sqlite3 failed, please install it. Jul 29 04:01:23 anyone tell me why my bitbake is barfing on this? Jul 29 04:01:31 especially since I have sqlite3 installed.... Jul 29 04:12:32 W8TVI: well, the good news at least is that Opie is still alive, assuming you still want to use it :) Jul 29 04:12:40 lol Jul 29 04:13:50 I kinda miss not having my portable kismet box :) Jul 29 05:20:19 so anyone want to help me out with this whacked error? ERROR: Importing sqlite3 failed, please install it. Jul 29 05:21:09 Zero_Chaos: You might want to give people a little more context to work with Jul 29 05:21:43 I type "bitbake" followed by just about anything and I get that error. My bitbake and monotone repo are up to the minute Jul 29 05:23:32 r848 does NOT have this issue while r849 and r850 do Jul 29 05:28:48 RP: "ERROR: Importing sqlite3 failed, please install it." I get this error with bitbake on r849 and r850, but r848 doesn't do it. I know you recently pushed updates... Jul 29 06:22:51 how do i run angstrom in neo qemu emulator. I have fic-gta01 images with me Jul 29 07:26:18 Zero_Chaos: apt-get install python-pysqlite2 Jul 29 07:28:41 koen http://lipax.paladine.org.uk/ <- whatcha think? :) Jul 29 07:52:02 good morning Jul 29 07:52:18 morning Jul 29 08:31:39 * Bernardo is away: Ausente por agora. Jul 29 09:36:59 * * OE Bug 2698 has been created by  Jul 29 09:37:00 * * Kernels with usbhost enabled lack drivers for usb devices Jul 29 09:37:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2698 Jul 29 10:29:18 hmmm Jul 29 10:29:25 bitbake 1.8.7 is quite slow with parsing Jul 29 10:29:32 * koen blaims slow disks Jul 29 10:56:05 hey zecke Jul 29 10:56:47 zecke: could you have a look at https://lists.berlios.de/pipermail/bitbake-dev/2007-July/000804.html ? Jul 29 10:57:15 mickey|NeoMeetin: you make me ditch studying math! Jul 29 11:06:21 hey lrg Jul 29 11:23:56 hey koen Jul 29 13:53:30 [anttix@klots build]$ bitbake --version Jul 29 13:53:30 ERROR: Importing sqlite3 failed, please install it. Jul 29 13:53:34 Who maintains bitbake? Jul 29 13:53:47 Where do I complain about idiotic sqlite3 import that breaks on Fedora Core? Jul 29 13:56:46 koen: is your ppc assembly problem still unsolved? Jul 29 13:57:11 tmbinc: I decided to ignore the problem for now Jul 29 13:57:16 anttix: apt-get install apt-get install python-pysqlite2 on debian Jul 29 13:57:27 (yes, pysqlite2 is for sqlite3) Jul 29 13:57:29 koen: good. Jul 29 13:57:53 anttix: a) use released versions Jul 29 13:58:05 anttix: b) if you don't do a) be clever enough for it Jul 29 13:58:20 anttix: c) don't judge if you have no clue, this will just sound offending Jul 29 13:59:29 I HAVE a clue :P Jul 29 13:59:43 Zecke I use stable 1.8 branch from SVN as recommended. Jul 29 13:59:57 And I have a clue. the module is named sqlite under Fedora and it's 3.0 branch. Jul 29 14:01:01 And I'm pretty pissed off about a breakage like this. I am preparing a proper patch that uses select sqlite_version(*) to determine the version of sqlite module. I'm just not sure where to submit it. Jul 29 14:08:03 anttix: you are [<-> ] that for to land on my ignore list Jul 29 14:08:10 s/for/far away/ Jul 29 14:10:29 anttix: you can use the bitbake-dev mailinglist at berlios or use the OpenEmbedded bug tracker Jul 29 14:11:51 zecke: Thanx. I'll send my patch to bitbake-dev list. Jul 29 14:23:17 what are likey causes of ERROR: QA issue: libfreetype.la failed sanity test (reference to workdir or installed)? Jul 29 14:34:34 hi, i try to build openzaurus 3.5.4 from org.oe.oz354x. Jul 29 14:34:40 i know it's old and unmaintained Jul 29 14:35:07 but now bb try to download sqlite-3.3.9.tar.gz Jul 29 14:35:15 and can't find it on sqlite.org Jul 29 14:35:26 can anybody help me? where can i download it Jul 29 14:35:44 why don't you build something that is actually maintained? Jul 29 14:35:51 or do a cvs co with a proper tag and tar.gz it by myself? Jul 29 14:36:22 i try to build oz 3.5.4 with opie 1.2.3 Jul 29 14:36:33 why don't you build something that is actually maintained? Jul 29 14:36:39 there is no newer "stable" version for collie Jul 29 14:36:48 ar: angstrom is very usable Jul 29 14:36:51 koen: what? for my collie Jul 29 14:37:01 it builds for collie? Jul 29 14:37:17 the only question would be how much flash do you have Jul 29 14:37:24 or if you want a root on sd or similar Jul 29 14:37:30 root on sd Jul 29 14:37:37 you shouldn't have a problem then Jul 29 14:37:51 hm. ok. i try it in org.oe.dev? Jul 29 14:38:53 http://www.angstrom-distribution.org/building-angstrom Jul 29 14:39:36 http://www.angstrom-distribution.org/unstable/images/collie/ Jul 29 14:40:25 wirelessdreamer: thanks so far. i'll try it out Jul 29 14:40:32 gl Jul 29 14:46:22 morning Jul 29 14:52:07 morning Jul 29 14:53:57 LOL http://ars.userfriendly.org/cartoons/?id=20070729 Jul 29 14:58:27 thats great :) Jul 29 15:00:26 wirelessdreamer: it's even funnier in my case, since I worked for them for 6 months and they all look like this sock ;) Jul 29 15:00:27 koen: it seems that having MACHINE set to something in the environment breaks the glibc-intermediate build (or something earlier that it depends on). How do you do your multimachine builds? Jul 29 15:00:44 it breaks binutils Jul 29 15:01:05 yep, that matches what I saw Jul 29 15:01:18 rwhitby: as Richard mentioned we unexport MACHINE in bitbake.conf Jul 29 15:01:35 for i in ep93xx h6300 fic-gta01 a780 a1200 e680 rokr-e2 at91sam9263ek qemuarm h2200 h4000 nslu2le ixp4xxle nokia770 omap5912osk poodle tosa h5000 magician htcuniversal htctornado hx2000 hx4700 c7x0 spitz akita netbook-pro progear guinness qemux86 mx31ads nokia800 efika nslu2be ixp4xxbe collie h3600 h3800 simpad htcwallaby ; do Jul 29 15:01:36 echo "MACHINE = \"$i\"" > conf/auto.conf Jul 29 15:01:49 rwhitby: so it should not be set when building binutils, could you verify it is not set when building binutils? Jul 29 15:01:50 zecke: guess what, it doesn't work and binutils still breaks Jul 29 15:02:06 I was doing: MACHINE=ixp4xxbe exec bitbake ... Jul 29 15:02:29 koen: that is a hypothesis, now back that with fact (e.g. a env in do_configure_prepend) Jul 29 15:02:50 zecke: it's not an hypothesis Jul 29 15:03:12 hey guys, let's not come to blows about this ... Jul 29 15:03:53 rwhitby: like yesterday, please show me the result of env when configuring binutils Jul 29 15:04:08 zecke: sorry - didn't see that yesterday. Jul 29 15:04:29 rwhitby: does it break on all distros? or is it only on centos? Jul 29 15:04:33 I will get onto that tomorrow (it's past midnight here). Thanks for the debugging tip. Jul 29 15:04:41 No, this is Debian Etch. Jul 29 15:05:00 and Debian Sarge Jul 29 15:09:48 koen: you still need to answer why you needed to move that directory before going to bed? did you know of any mediawiki exploit that would have been too dangerous to keep mw eight more hours around? Jul 29 15:10:48 zecke: the directory needed to be moved because it wasn't supposed to be there Jul 29 15:11:12 koen: why couldn't you wait eight more hours and talk before moving? Jul 29 15:11:25 zecke: and it was owned by 'root', so it surely must be an rootkit Jul 29 15:11:46 koen: see, if I'm tired and about to sleep I would send of an email... Jul 29 15:11:58 koen: ah ha, and moving it removes a 'rootkit'? Jul 29 15:12:16 no, it makes it less obvious Jul 29 15:12:34 koen: that is no reasoning at all, so what was the reason you needed to move the directory? Jul 29 15:12:49 zecke: are you really that thick? Jul 29 15:13:10 * rwhitby gets out the popcorn ... Jul 29 15:13:19 koen: and I think you should consider giving up some admin privileges to stop such "fastest sherif in the west"-behaviour Jul 29 15:13:45 koen: I know how easily I can get pissed off... Jul 29 15:14:03 zecke: I think some people shouldn't abuse the root acount to install random wikis into sites that don't want them Jul 29 15:14:18 hi all Jul 29 15:14:45 koen: see, I don't think your usage of 'abuse' is proper in this context. If you knew who created that dir with sudo, send a mail and wait eight more hours Jul 29 15:16:31 zecke: not sudo Jul 29 15:17:26 zecke: and I'm tired of repeatin myself, so go find someone else to pick on Jul 29 15:17:38 or fix bitbake bugs instead of wasting my time Jul 29 15:17:52 koen: well, I might just leave OE and BitBake due this episode Jul 29 15:18:04 and you help to making leaving eas Jul 29 15:18:05 y Jul 29 15:18:12 such a friendly development team is OE ... Jul 29 15:26:45 Yeah. Well, I support Koen. At least he does not jump on angry people like me ;) Jul 29 15:28:56 ERROR: Task 24 (/home/koen/OE/monotone/org.openembedded.dev/packages/gnome/gnome-power-manager_2.18.3.bb, do_compile) failed Jul 29 15:29:02 .. Jul 29 15:29:11 NOTE: Tasks Summary: Attempted 1322 tasks of which 1322 didn't need to be rerun and 0 failed. Jul 29 15:50:59 * * OE Bug 2700 has been created by bugzilla(AT)stalag99.net Jul 29 15:51:01 * * ipkg-utils missing in Angstrom feeds Jul 29 15:51:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2700 Jul 29 16:00:17 hi all Jul 29 16:12:59 * * OE Bug 2700 has been RESOLVED (WONTFIX) by Jul 29 16:13:01 * *  ipkg-utils missing in Angstrom feeds Jul 29 16:13:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2700 Jul 29 16:41:54 anttix: Did you have a patch to make sqlite work under Fedora? Jul 29 16:42:19 koen: The locking issues worry me a bit. How easily are you hitting that? Jul 29 16:44:50 RP: Yes, I'm posting it to bitbake-dev Jul 29 16:45:04 RP: It's simple actually Jul 29 16:45:54 RP: in lib/bb/persist_data.py find a line import sqlite3 and replace it with from pysqlite2 import dbapi2 as sqlite3 Jul 29 16:46:18 RP I mean "import sqlite3" becomes "from pysqlite2 import dbapi2 as sqlite3" Jul 29 16:46:59 Please try it and let me know. I'm a bit worried, because bbimage doesn't work for me because I only have older version of OE tree to test against. Have to wait for a newer version to build. Jul 29 16:47:21 anttix: That fails on Debian :-( Jul 29 16:47:40 RP: make sure python-sqlite2 is installed Jul 29 16:47:49 RP: For debian it's python-pysqlite2 I believe Jul 29 16:48:01 RP: see http://initd.org/tracker/pysqlite/wiki/PysqlitePackages Jul 29 16:51:52 anttix: The strange thing is import sqlite3 just works, pysqlite2 seems to need more packages :/ Jul 29 16:54:58 It's still a mystery why it works. There is no such file in python libs under debian. Or is it somewhere and I'm missing it? Jul 29 16:57:42 anttix: Its a directory Jul 29 16:58:00 RP: it's quite easy to trigger the lock, it happens when 2 tasks start at the same time Jul 29 16:58:21 koen: I just did that with lots of tasks and it seemed to survive :/ Jul 29 16:58:29 hmmm Jul 29 16:58:57 RP: parsing is also very slow, but I can live with that for the moment Jul 29 16:59:08 koen: Slower than it used to be? Jul 29 16:59:16 RP: Did You try my change on Fedora? Jul 29 16:59:22 hi all Jul 29 16:59:28 anttix: I don't have Fedora Jul 29 16:59:59 hi steliosk Jul 29 17:00:14 koen: Parsing is slower :-( Jul 29 17:00:34 RP Hi ! Jul 29 17:01:02 has anyone noticed any difference between binaries generated with OE on i686 and amd64 ? Jul 29 17:01:14 RP: Ok, I'll do some more hacking on Debian and see where it gets me. Jul 29 17:01:39 anttix: ok, thanks Jul 29 17:01:47 steliosk: People have mentioned it now and again Jul 29 17:01:54 steliosk: gcc3 used to have a bug like that for crosscompiling Jul 29 17:02:12 well, gcc3 still has that bug, but I moved on to gcc4 Jul 29 17:02:19 it seems it happens with gcc4 Jul 29 17:02:56 i noticed it when trying to build uboot for sequoia Jul 29 17:03:11 it builds fine on i686 even with icecc.bbclass Jul 29 17:03:17 but fails on amd64 Jul 29 17:04:00 complaining about segments been too big Jul 29 17:04:04 koen: I can't get that lock problem to show up at all :-( Jul 29 17:04:59 http://rafb.net/p/ccguaU40.html Jul 29 17:06:18 RP: 3 bitbake threads on a fast machine? Jul 29 17:07:13 RP: I dont get it every time, so it might have been a few anomalies Jul 29 17:07:30 as in, it doesn't happen on my other buildmachine Jul 29 17:08:34 koen: I'm running fetchall with 4 threads on my desktop... Jul 29 17:08:49 koen: The default timeout is 5 seconds, perhaps we should just try incerasing it slightly Jul 29 17:09:08 A stupid question: how do You get bitbake to run in two or more threads? Jul 29 17:09:25 Does it compile things in parallel then? Jul 29 17:10:58 anttix: It does. You set BB_NUMBER_THREADS=x Jul 29 17:11:23 koen: Try adding ,timeout=10 to the connect tring in persist_data.py Jul 29 17:11:59 timeout=1 doesn't cause me any problems here :/ Jul 29 17:14:47 koen : Any plans to get mtd-utils "fixed" to a certain date ? Jul 29 17:16:53 RP: thanx, I'll try Jul 29 17:17:52 steliosk: RP and I talked about it, no concrete plan yet Jul 29 17:18:40 koen : Any particular reason not to do it ? Jul 29 17:19:03 steliosk: Dates make no sense with git is the main one Jul 29 17:19:51 koen RP: we am trying to get a OE livecd going where you will not need to download anything Jul 29 17:20:08 koen RP: i.e work without an inet connection Jul 29 17:20:17 steliosk: So the git version should have a preference of -1 and we should use a released version Jul 29 17:20:38 its an option Jul 29 17:21:11 RP: next year 8GB usb keys with an OE (or poky) build env :) Jul 29 17:21:28 koen: Poky images were nearly on there this year ;-) Jul 29 17:21:48 i think the same should be true for anything requiring updates (kernel etc etc) Jul 29 17:21:49 (or with proper funding, a c3200 that emulates usb-storage) Jul 29 17:22:03 well rm-work can do miracles Jul 29 17:22:05 :) Jul 29 17:22:35 with enough ram you can get OE to build on ram only ;) Jul 29 17:23:16 I've been building with -c rmall for a while now Jul 29 17:23:51 steliosk: with enough ram people will run C# and Java on it ... not a good option ;) Jul 29 17:23:56 that and tinder autoclean saves up a lot of space Jul 29 17:26:02 has anyone seen any usb dongles with ram instead of flash ? Jul 29 17:26:35 few years ago i had seen sram based with battery backup, but can't find anything now Jul 29 17:28:27 03rpurdie * r851 10/ (2 files in 2 dirs): persist_data: Remove bogus testing code Jul 29 17:28:29 koen: Are you running multiple bitbakes on your machine? Jul 29 17:28:50 that too Jul 29 17:28:59 koen: If so, the above might fix the locking issues Jul 29 17:29:17 * RP had left a rather nasty hardcoded path in :/ Jul 29 17:34:03 hi all Jul 29 17:34:09 hi flo_lap Jul 29 17:34:10 hey flo_lap Jul 29 17:51:59 koen: Did increasing the timeout or the svn changes help? Jul 29 17:52:17 haven't tested yet, sorry Jul 29 17:52:26 ok, np Jul 29 17:52:27 * koen is watching glibc eat through locales Jul 29 17:53:14 hi, all! how could I get previous LR in data abort handler on ARM? Jul 29 17:54:58 RP: it seems to have worked Jul 29 17:55:32 koen: I know how to fix the parsing slowdown too :} Jul 29 17:56:26 by buying everyone a 5THz pentium7? Jul 29 17:56:50 Fix base.bbclass to avoid calling something Jul 29 17:57:42 dang Jul 29 17:57:55 RP: http://rafb.net/p/9ecM2p53.html :( Jul 29 17:58:29 (still with timeout=1) Jul 29 17:58:41 koen: Was that with the svn change? Jul 29 17:58:46 yes Jul 29 17:58:53 Try the default timeout of 5 Jul 29 17:58:56 r851 Jul 29 18:00:47 where is the timeout defined? Jul 29 18:12:16 03rpurdie 07org.oe.dev * r0ecb1852... 10/ (1 classes/base.bbclass): base.bbclass: Only take the slow path if SRC_URI_OVERRIDES_PACKAGE_ARCH is actually set to something Jul 29 18:18:09 RP: parsing it much faster again, thanks! Jul 29 18:48:26 does anyone see an error in this line? SRC_URI = "${SOURCEFORGE_MIRROR}/sourceforge/synergy2/synergy-${PV}.tar.gz" Jul 29 18:49:10 nm i named the bb wrong Jul 29 18:49:44 wirelessdreamer: remove the /sourceforge as well Jul 29 18:50:27 wirelessdreamer: have you looked at packages/synergy already? Jul 29 18:51:15 someone in #angstrom asked about it, i didn't bother to check if it was built, thanks for the heads up Jul 29 18:58:48 RP: http://rafb.net/p/pjV5wh69.html Jul 29 19:49:49 psokolovsky_: could you have a look at #2698 ? Jul 29 19:50:47 03koen 07org.oe.dev * r30d4ffe6... 10/ (1 packages/synergy/synergy_1.3.1.bb): synergy: update to 1.3.1 Jul 29 19:55:30 koen: Hi! Well, I took care of h2200 flash, but didn't yet have chance to look into that. I won't be able to do that for few more weeks due to vacation, so feel free to enable needed options in hx4700 defconfig - I'll merge them to defconfigman later Jul 29 19:58:42 psokolovsky_: ok, will look at squeezing it in the defconfig Jul 29 19:58:50 psokolovsky_: enjoy your holidays Jul 29 20:02:03 thanks Jul 29 20:06:11 * koen heads to bed Jul 29 20:29:48 koen: Still there? Jul 29 20:51:07 koen: The common pattern for those .bb files is their change MACHINE Jul 29 20:51:24 Which means it can't find the cache Jul 29 21:06:27 koen: Those BB_URLDATA issues should be resolved in OE.dev now Jul 29 21:20:57 03rpurdie * r852 10/ (6 files in 4 dirs): event.py: Add ConfigParsed Event after configuration parsing is complete Jul 29 21:27:10 what are likely causes of CROSS COMPILE Badness: /usr/include in INCLUDEPATH: Jul 29 21:29:49 wirelessdreamer: Something broken in the package which is letting that path creep in Jul 29 21:32:30 03rpurdie 07org.oe.dev * r871de5bd... 10/ (1 conf/bitbake.conf): bitbake.conf: Forcefully expand the cache path - we don't want this to change beyond this point Jul 29 21:32:35 03rpurdie 07org.oe.dev * r4f833d2e... 10/ (1 classes/sanity.bbclass): sanity.bbclass: Run sanity checker after configuration parsing for bitbake > 1.8.6 Jul 29 21:32:40 03rpurdie 07org.oe.dev * r2690bc18... 10/ (1 classes/sanity.bbclass): santiy.bbclass: Undo accidental change Jul 29 21:36:00 03rpurdie 07bitbake-1.8 * r853 10/lib/bb/persist_data.py: persist_data.py: Try harder to import an sqlite package. Show how to change the connection lock timeout Jul 29 21:36:39 anttix: Hopefully that will solve your problem too Jul 29 21:36:40 03rpurdie * r854 10bitbake/lib/bb/persist_data.py: persist_data.py: Try harder to import an sqlite package. Show how to change the connection lock timeout Jul 29 21:37:25 RP: I'll try that in a moment Jul 29 21:44:02 Yay, these changes even help bitbakes memory footprint nicely :) Jul 29 21:45:00 RP: bitbake itself now runs on FC6, waiting for the build process to reach bbimage ;) Jul 29 21:45:26 anttix: Is that python 2.4 btw? Jul 29 21:45:48 I think sqlite3 is python 2.5 only... Jul 29 21:45:48 RP: python-2.4.4-1.fc6 Jul 29 21:46:00 Right, that explains the problem Jul 29 21:46:29 You running unstable debian with python 2.5? Jul 29 21:46:42 anttix: ubuntu edgy Jul 29 21:47:41 RP: Yeah, You are probably right. My debian system is stable with python 2.4 and I've been trying to figure out where sqlite3 dir is :P Jul 29 21:48:13 RP: Unfortunately I had to deal with some other issues in the mean time so I didn't have much time to hack that particular problem :( Jul 29 21:48:22 I think its solved for python 2.4 and 2.5 now :) Jul 29 21:48:30 I sure hope so. Jul 29 21:48:33 Thanx ;) Jul 29 21:49:13 Thanks for testing. It is appreciated (and useful) :) Jul 29 21:49:31 Just try and make the bug reports polite ;-) Jul 29 21:50:23 Yeah, well, I was pissed off. Sorry for that. Happens :( Jul 29 21:52:07 is sqlite3 an integral part of python 2.5? Jul 29 21:52:22 anttix: I'm not 100% sure, I think it could be Jul 29 21:55:17 Well, the strange thing is, that bbimage still fails for me. It might be that I have too old OE metadata. Jul 29 21:55:29 anttix: What is the failure? Jul 29 21:56:08 | + bbimage -n dbe61-586-20070729214027 -t jffs2 -e /home/anttix/OpenEmbedded/rbox/packages/images/dbe61-image.bb Jul 29 21:56:08 | ERROR: Importing sqlite3 and pysqlite2 failed, please install one of them. A 'python-pysqlite2' like package is likely to be what you need. Jul 29 21:56:08 NOTE: Task failed: /home/anttix/OpenEmbedded/build/tmp/work/586-linux/dbe61-image-1.0-r0/temp/log.do_rootfs.10587 Jul 29 21:56:22 ouch :/ Jul 29 21:56:38 bbimage uses python-native from OE itself. Jul 29 21:56:40 * RP tries an image Jul 29 21:56:55 Why on earth does it do that? Jul 29 21:57:10 PATH I guess :/ Jul 29 21:57:39 Yeah! It's executed from a script that has all set for cross-compiling Jul 29 21:59:05 We should tweak image.bbclass to use the system's python then Jul 29 21:59:42 I haven't python-native build in my testing environment which is why I haven't seen this Jul 29 22:00:39 Yeah, at least libxml2-native depends on it. Jul 29 22:01:43 I use --without-python here... Jul 29 22:02:48 There is also a bitdoc in bitbake/bin dir that's probably affected as well Jul 29 22:03:06 Does OE call bitdoc though? Jul 29 22:03:30 Don't know, I just looked into the dir to see what else is there which might suffer ;) Jul 29 22:04:44 Its just bbimage Jul 29 22:05:10 seems so ;) Looks like Your machine has faster grep ;) Jul 29 22:07:43 No, I just know what bitdoc is Jul 29 22:08:13 Can you ask a running python where the python binary its using is? Jul 29 22:08:23 I was just thinking about the same thing ;) Jul 29 22:08:34 And I don't know :( Jul 29 22:09:00 But I looked into bbimage and it uses #!/usr/bin/env python to execute itself. Jul 29 22:09:34 It could be replaced to /usr/bin/python, but it would probably break on systems where python is somewhere else Jul 29 22:14:13 sys.executable Jul 29 22:14:25 I just found that too :) Jul 29 22:15:15 Well, it could be great if bitbake set a variable that image.bbclass can use. Or bbimage could use it to re-execute itself. Jul 29 22:15:38 I'm looking at that now Jul 29 22:21:44 anttix: mtn pull, I have difficulty testing this one Jul 29 22:22:36 just a moment Jul 29 22:23:43 Pulling. Maybe You can bitbake python-native, that should trigger the bug. Jul 29 22:24:10 anttix: That isn't something I can easily do with my setup Jul 29 22:24:30 Well, I can on one box but the build won't get anywhere near an image before tomorrow ;-) Jul 29 22:24:39 ok, I'll try to test it ;) Jul 29 22:25:08 The problem with my setup is, that it uses an old OE tree, but I can probably copy the needed files over manually. Jul 29 22:25:31 We haven't had the guts to move to glibc 2.5 in the middle of our firmware freeze process :P Jul 29 22:25:51 anttix: Fair enough. Its a simple patch you should be able to apply Jul 29 22:26:11 one line to bitbake.conf, one change to image.bbclass Jul 29 22:26:37 does anyone here know who "maintains" the jack packages in bitbake? Jul 29 22:26:57 wirelessdreamer: s/bitbake/OE/ ;-) Jul 29 22:27:03 wirelessdreamer: but no Jul 29 22:27:13 (I don't know) Jul 29 22:27:13 true :) Jul 29 22:27:44 wirelessdreamer: Its a sure way to upset bitbake developers, some more than others mind ;-) Jul 29 22:28:13 any way i could try and find out? i'm working on anding some packages to oe and would like someone with input that has a similar interest in it :) Jul 29 22:28:40 wirelessdreamer: Ask on the mailing list. I presume you've checked the Maintainers file? Jul 29 22:28:41 duely noted, i'm trying to break myself of bad habits still Jul 29 22:29:02 i'm still a bit green, didn't know we had a maintainers file :) Jul 29 22:29:36 but no, its not listed there now that you mention it :) Jul 29 22:31:12 ok, the mailing list is the best place other than here to ask then Jul 29 22:31:48 RP: Success. Jul 29 22:32:19 RP: However there seems to be another bbimage line in image.bbclass that uses fakeroot to launch it. Maybe it should be modified too. Jul 29 22:33:58 cannot find -lqt-mt, anyone know what lib this is listed as in oe? Jul 29 22:36:43 wirelessdreamer: Doesn't qt have it? Jul 29 22:37:15 i assumed so, i haven't knowingly though tried to build anything that would actually need to link against it yet Jul 29 22:37:17 qt-x11 or qt4-x11? Jul 29 22:37:48 or was it even qt-x11-free I can't remember from the top of my head :P Jul 29 22:38:05 i'm trying to build a qt3 app, so i'm trying to build qt-x11-free-3.3.7 now Jul 29 22:40:16 anttix: Hmm, probably yes Jul 29 22:40:44 03rpurdie 07org.oe.dev * r0a0e903f... 10/ (1 classes/image.bbclass conf/bitbake.conf): image.bbclass: Use system python explicitly, don't use python-native version Jul 29 22:42:34 anttix: I've pushed that change Jul 29 22:46:18 RP: Cool. Ok, that should be it for now. I think I'll call it a day. It's 2am around here. Thanx for help. Jul 29 22:46:30 anttix: np Jul 29 22:47:23 Good night Jul 29 22:47:33 anttix: 'night Jul 29 22:50:47 RP: you seem to be the best person to check with this about :) is there a variable i can set in bitbake.conf (or elsewhere) so i can run a devshell over ssh in the same term it was started in? Jul 29 22:52:05 wirelessdreamer: Not at the moment, no due to the multithreading :-( Jul 29 22:55:50 not even i way i could exec the process then just kill the shell when i'm done? Jul 29 22:56:07 s/even i way/even a way/ Jul 29 22:56:07 wirelessdreamer: You can cheat and use the -b option Jul 29 22:56:39 wirelessdreamer: or you can run the script generated in the work/*/temp/run*do_devshell* Jul 29 22:57:05 wirelessdreamer: We need to find a better way really. See the disucssion on the bitbake list Jul 29 22:57:51 i think i'm only subscribed to angstrom lists right now, it would probably make seince for me to at least join in on oe too Jul 29 22:58:57 I should get some sleep. 'night all Jul 29 23:00:57 thanks for the help :) Jul 29 23:04:59 * * OE Bug 2701 has been created by antti.andreimann(AT)mail.ee Jul 29 23:05:00 * * ed libxslt to 1.1.21 Jul 29 23:05:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2701 Jul 29 23:06:59 * * OE Bug 2702 has been created by antti.andreimann(AT)mail.ee Jul 29 23:07:01 * * ed libxml2 to 2.6.29 Jul 29 23:07:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2702 Jul 30 01:37:23 koen: thanks for the info, and just fyi, not everyone uses ubuntu ;-) Jul 30 01:48:11 i have this issue compiling nano Jul 30 01:48:15 | block-qcow.c:26:18: error: zlib.h: No such file or directory Jul 30 01:48:23 but zlib is in satging Jul 30 01:49:00 * T0mW thinks libiconv is evil Jul 30 01:56:11 this the pastebin Jul 30 01:56:13 http://pastebin.com/m7a21bef1 Jul 30 02:12:34 * psycoman libiconf is the evil >] Jul 30 02:42:30 Zero_Chaos: ubuntu, that is a breakfast food, right? Jul 30 02:43:13 T0mW: it's actually linux with a windows front end, used by newbs Jul 30 02:43:47 heh, Mandriva here Jul 30 02:43:58 ouch Jul 30 02:44:32 * T0mW makes warding signs against distro angst Jul 30 02:45:09 not to start a holy war, but I haven't noticed any elements from Windows in 'buntu ever Jul 30 02:45:46 maybe you're thinking of Kubuntu Jul 30 02:46:37 BHSPitMonkey: "not I" said /me Jul 30 02:46:54 k Jul 30 02:47:24 I know nothing about it, I've too much work to do to mess with a new distro. Jul 30 02:47:49 I came to a breaking point with mandriva, and I had heard so much about ubuntu Jul 30 02:48:20 like it a bunch, most mature "desktop" distro I've used Jul 30 02:53:43 BHSPitMonkey: the only thing that concerns me about ubuntu (gentoo as well) is package version stability. Sometimes I don't want the latest + greatest, last years version is fine as long as it works today. Jul 30 02:54:10 BHSPitMonkey: I'm a bit afraid that if I tried those distros out that I'd have versions changing all the time? Jul 30 02:55:18 software versions? unless you add software repos in ubuntu for backports and such, you stay locked in with the software versions that were out when that distro version was released Jul 30 02:55:25 save for security updates and such Jul 30 02:56:10 e.g., gaim went 2.0 with pidgin right after the most recent ubuntu release, and you're stuck with gaim unless you go out and get a pidgin package. Jul 30 02:57:23 I actually -do- like to have the latest and greatest (stable releases), so this gets on my nerves sometimes, but eh, they have reasons. and, it's not that hard to go find a package or a repo Jul 30 02:57:50 funny how you had ubuntu and gentoo in the same breath :P Jul 30 02:57:51 BHSPitMonkey: well, I have some Point of Sale stuff that I need to keep tight control over... Jul 30 02:57:57 true Jul 30 02:58:25 Dapper is still the current LTS release, for the people who feel the same way as you Jul 30 02:58:41 it's from june of last year, but it's still stable Jul 30 02:58:56 if you guys all want to talk about how great ubuntu is, please use #ubuntu Jul 30 02:59:05 * Zero_Chaos didn't mean to start a distro conversation Jul 30 02:59:07 Zero_Chaos, are you kidding? #ubuntu is horrible Jul 30 02:59:17 :) **** ENDING LOGGING AT Mon Jul 30 02:59:57 2007