**** BEGIN LOGGING AT Sat Jul 23 23:59:56 2005 Jul 24 01:57:40 moin Jul 24 01:57:45 Hello...can anyone help me with an issue? Jul 24 01:57:54 I'm working with Nylon, OE based, and when I try to call bitbake it returns the following error: Jul 24 01:58:01 ERROR: Unable to open conf/bitbake.conf Jul 24 01:59:12 vbcn: most likely your BBPATH environment variable is wrong/missing Jul 24 02:00:35 my BBPATH is set on the env.sh script like that: export BBPATH=${OEBASE}:${OEBASE}/openembedded Jul 24 02:01:02 and the file bitbake.conf is set on /openembedded/conf/bitbake.conf Jul 24 02:01:43 (I set 777 permission on it, but it doesn't seem to solve the problem) Jul 24 02:01:57 vbcn: bitbake -DDD and see what files it tries to read Jul 24 02:03:24 I get the same ("Unable to open...") error when I call bitbake, no matter with flags or parameters I use :( Jul 24 02:03:57 vbcn: No, -DDD traces itself... but if you know better... Jul 24 02:05:54 Excuse me, I didn't whant to mean that Jul 24 02:06:11 i've tried "bitbake -DDD", but I get the same error :( Jul 24 02:07:38 vbcn: yes but you should have at least ten to twenty lines of debugging Jul 24 02:07:43 output Jul 24 02:09:16 I get only one output line with the "ERROR: Unable to open conf/bitbake.conf" Jul 24 02:10:17 I suppouse that if the command that you say doesn't work, the problem is worse than I though Jul 24 02:11:06 it's strange because I built a few days ago all the Environment succesfully, and the process ran bitbake several times without any problems :-S Jul 24 02:11:11 I strongly doubt that it doesn't work Jul 24 02:11:52 vbcn: not finding conf/local.conf indicates a wrong BBPATH Jul 24 02:12:07 vbcn: echo $BBPATH please Jul 24 02:13:00 oops....echo $BBPATH returns a blank line Jul 24 02:13:17 it seems to be missing to store the path? Jul 24 02:19:05 vbcn: dunno I'm no shell expert Jul 24 02:20:21 thanks anyway, I appreciate your help. No I know there's something wrong with the BBPATH variable Jul 24 02:20:23 thank you Jul 24 02:23:17 good morning all Jul 24 02:23:56 koen: moin Jul 24 02:24:12 koen: my tinderclient.bbclass contains an error Jul 24 02:24:25 it does? Jul 24 02:24:33 * koen is busy waking up Jul 24 02:27:16 moin Jul 24 02:27:40 koen: i have trying more than week build opie from oe sources. it fail while build or builded image doesn't work. can you share "local.conf" files for me which you use for hh.org snaphosts ? Jul 24 02:29:34 aquadran: cp local.conf.sample local.conf, DISTRO=familiar-0.8.3 Jul 24 02:29:44 aquadran: and I don't know anything about opie Jul 24 02:32:52 hey CoreDump|home Jul 24 02:33:07 zecke_: what's the error? Jul 24 02:35:00 koen: not sending failure or success Jul 24 02:35:15 ah, that explains the yellow Jul 24 02:43:11 aquadran: rootfs fails because of a missing ipkg.conf, right? Jul 24 02:43:18 not sure why... Jul 24 02:51:02 JustinP: hmm, i don't know. i didn't noticed that. Jul 24 02:51:57 that's what's failing for me right now Jul 24 02:52:12 after I fixed the pcre /usr/lib/libc.a bug Jul 24 02:52:26 zecke_: I'm setting up the OpenSlug tinderbox client now. Jul 24 02:52:35 Do you have a sample config which I can copy from? Jul 24 02:52:48 what is this tinderbox thing? Jul 24 02:52:49 rwhitby-away: conf/tinder.conf Jul 24 02:53:15 JustinP: http://ewi546.ewi.utwente.nl/OE_qa/OpenEmbeddedBuild/status.html Jul 24 02:54:32 and http://article.gmane.org/gmane.comp.handhelds.openembedded/5264 Jul 24 02:55:43 koen: what should BUILDNAME and SUBJECT be set to for OpenSlug ? Jul 24 02:57:05 'Tinder Client OpenSlug' and 'Tinder Client OpenSlug report' or something like that Jul 24 02:57:06 rwhitby-away: SUBJECT must be "OpenEmbedded Tinder Log" Jul 24 02:57:11 ok Jul 24 02:57:25 or at least start with OpenEmbedded... Jul 24 02:57:30 let me change that in tinder.conf Jul 24 02:57:53 TINDER_BUILD = "openslug-sluggy-sluggy-image" Jul 24 02:58:07 koen: and TINDER_BUILDNAME does not exist :( Jul 24 02:58:44 so it is: #TINDER_BUILD = "Tinder Client Test-NameOfClient" ? Jul 24 02:58:57 zecke_: is simply "OpenSlug" and "Unslung" suitable for TINDER_BUILD? Jul 24 02:58:59 without spaces and / Jul 24 02:59:03 well I gather that it's some sort of automated build and testing system, but none of those links said anything like that.... Jul 24 02:59:26 rwhitby-away: hmm do you build everything slug related? Jul 24 02:59:47 JustinP: http://zecke.blogspot.com/ Jul 24 02:59:57 JustinP: but it might be off as I was really really tired Jul 24 02:59:59 zecke_: we will have a cron job on the official build machine which will build OpenSlug and Unslung Jul 24 03:00:28 they will be the official builds. Others may do personal builds. Jul 24 03:01:17 rwhitby-away: if you specify TINDER_VERBOSE_REPORT you will send one mail per Task Jul 24 03:01:17 bbiab 30min Jul 24 03:01:24 rwhitby-away: if not only one big compile log Jul 24 03:05:32 03koen 07org.oe.dev * rb56b830a... 10/conf/tinder.conf: Fix up tinder.conf some more Jul 24 03:05:41 zecke_: did you push the fix for the bbclass yet? Jul 24 03:07:02 zecke_: cool :-) Jul 24 03:07:03 koen: I need to get mails first ;) Jul 24 03:09:41 all right, I should be going to sleep now....I hope someone either applies my pcre patch or finds an alternative Jul 24 03:13:48 zecke_: fetchmail: No mail for tinderbox-oe@gmx.net at pop.gmx.net Jul 24 03:15:25 zecke_: did you turn the gmx spamfilter off? Jul 24 03:17:40 koen: no but who should send mails? Jul 24 03:18:06 zecke_: Venture Capitalists wanting to give us meagabucks Jul 24 03:22:17 that is useful spam Jul 24 03:31:07 zecke_: http://pastebin.com/319558 Jul 24 03:31:16 are those settings correct? Jul 24 03:32:15 ibot: test Jul 24 03:32:15 Oh, no! There's a test and I haven't studied! Jul 24 03:32:21 could some of those (e.g. _ERROR, _LOG, _MAILTO, _TREE, _REPORT, _START) be defaulted somehow ? Jul 24 03:32:24 * koen hopes the one big log will pass gmx.net Jul 24 03:32:58 rwhitby: looks good Jul 24 03:33:22 koen: you'd prefer different settings? Jul 24 03:34:15 rwhitby: I'm not sure gmx.net will like a 5MB mail Jul 24 03:34:22 can some defaults be put in tinderclient.bbclass? Jul 24 03:34:32 koen: what should I change? Jul 24 03:36:20 or can tinder.conf be defaults, and we include that in local.conf and override it? Jul 24 03:37:41 koen: how do I stop it being a 5MB mail? Jul 24 03:38:24 rwhitby: build less ;) or do verbose reports Jul 24 03:38:30 rwhitby: but let us try it first Jul 24 03:39:06 hail pb_ Jul 24 03:39:26 hail zecke Jul 24 03:39:35 * zecke_ awards pb_ for giving the homeless a $HOME... Jul 24 03:42:50 zecke: mm? Jul 24 03:44:40 zecke_; you should have tinderbox mail (a small one) Jul 24 03:45:55 zecke_: did that work? Jul 24 03:45:58 hi all Jul 24 03:46:04 florian: good morning! Jul 24 03:46:25 rwhitby: did you mail tinderbox-oe? Jul 24 03:46:54 yep Jul 24 03:47:09 2005-07-24 03:39:12 1Dwdt2-0004po-6J <= tinderclient@nslu2-linux.org H=localhost ([66.159.209.54]) [127.0.0.1] P=esmtp S=33152 Jul 24 03:47:09 2005-07-24 03:39:15 1Dwdt2-0004po-6J => tinderbox-oe@gmx.net R=dnslookup T=remote_smtp H=mx0.gmx.de [213.165.64.100] Jul 24 03:47:09 2005-07-24 03:39:15 1Dwdt2-0004po-6J Completed Jul 24 03:48:02 http://ewi546.ewi.utwente.nl/mail/from Jul 24 03:50:10 so it didn't make it? Jul 24 03:51:20 sounds like you are blacklisted at gmx Jul 24 03:51:51 it is not even in the spam filter Jul 24 03:52:28 zecke: how can I see the revision history of a file in the monotone repository? Jul 24 03:52:41 either using the web interface or the shell would be fine Jul 24 03:53:31 annotate, log Jul 24 03:53:46 I tried that, but neither seemed to work Jul 24 03:53:58 viewmtn is using monotone log ;) Jul 24 03:54:02 pb@lurch:~/oe/oe/packages/gcc$ monotone log gcc-cross-sdk_3.3.4.bb Jul 24 03:54:03 pb@lurch:~/oe/oe/packages/gcc$ monotone annotate gcc-cross-sdk_3.3.4.bb Jul 24 03:54:03 monotone: misuse: No such file 'gcc-cross-sdk_3.3.4.bb' in revision 7d0136e54bdf205173cc235d2b4a5d747c2bd8bc Jul 24 03:54:03 pb@lurch:~/oe/oe/packages/gcc$ Jul 24 03:57:59 pb_: it wants a path Jul 24 03:58:10 meaning what? Jul 24 03:58:21 pb_: cd openembedded ; monotone annatate packages/gcc/gcc-cross-sdk_3.3.4.bb Jul 24 03:58:28 without the typos Jul 24 03:58:43 ah Jul 24 03:58:53 thanks, yeah, that works Jul 24 03:58:58 annotates is a database op (confusing & annoying) Jul 24 03:59:01 the path has to match the entry in the manifest Jul 24 03:59:19 now, can I get it to show the revisions in any more readable format? Jul 24 03:59:32 "80f3b9d4e175b65ba8e7040fdcdfe98c15fc09fd" doesn't mean much to me Jul 24 03:59:46 monotone cat revision 80f3b9d4e175b65ba8e7040fdcdfe98c15fc09fd Jul 24 04:00:45 or monotone log Jul 24 04:00:48 what I really want is for "annotate" to show me the date and author of each revision, kind of like cvs does. Jul 24 04:00:53 me too Jul 24 04:01:19 I had some shell magic to do that Jul 24 04:01:25 * koen searches for that Jul 24 04:03:48 monotone annotate packages/sylpheed/sylpheed_2.0.0beta6.bb | awk '{print $1}' | tail -n1 | sed s/:// | xargs monotone list certs Jul 24 04:03:58 that gives the log for the last line Jul 24 04:04:10 thanks Jul 24 04:05:53 monotone needs a bit of user interface added around it which doesn't expose the hashes so much ... Jul 24 04:05:53 03freyther 07org.oe.dev * r017e300c... 10/classes/tinderclient.bbclass: Jul 24 04:05:53 classes/tinderclient.bbclass: Jul 24 04:05:53 -Fix indention of the final report mail. Now success or failure Jul 24 04:05:53 should be mail to the tinderbox. Jul 24 04:05:57 03koen 07org.oe.dev * r761267fc... 10/conf/tinder.conf: more info on VERBOSE_REPORT Jul 24 04:08:51 rwhitby: OpenSlug has landed Jul 24 04:09:09 we might have an timezone issue... Jul 24 04:09:41 :) Jul 24 04:10:24 zecke_: well, I'm in one timezone, the build is in another, and you're in another, so I hope Tinderbox handles it properly ... Jul 24 04:11:09 zecke_: make tinderstart use UTC Jul 24 04:12:52 koen: could we delete the data_dir and content of html_output once more? Jul 24 04:13:21 rwhitby: update your tree please to get a fixed tinderclient.bbclass Jul 24 04:19:00 mickey|zzZZzz: even the regular gcc-cross seems to suffer from the FP problem in an openzaurus build. I guess this bug is not specific to the SDK. Jul 24 04:19:01 will do Jul 24 04:19:17 mickey|zzZZzz: maybe gcc 3.3 just lacks the right bits to do openzaurus style floating point Jul 24 04:22:50 koen: and could you give me write access to the files in local_conf and bin/ Jul 24 04:23:32 zecke_: try now Jul 24 04:26:45 koen: could we clean data_dir and html_output? ;) Jul 24 04:27:29 done Jul 24 04:28:13 rwhitby: you could start a build now Jul 24 04:32:25 do I need to change any of the variables in my local.conf for the timezone stuff (i.e. TINDER_START) ? Jul 24 04:32:57 rwhitby: yes... but I do not know to what yet? Jul 24 04:33:08 using gmtime instead of localtime should give utc Jul 24 04:33:20 but I need to make the tinderbox to process mails as utc as well Jul 24 04:34:24 So <<>> ? Jul 24 04:34:54 yes but the issue is with process_mails Jul 24 04:36:10 03:36:14 Jul 24 04:36:32 http://ewi546.ewi.utwente.nl/OE_qa/cgi-bin//tinder.cgi?tree=OpenEmbeddedBuild&start-time=1122183042&display-hours=24 Jul 24 04:36:42 you show up... Jul 24 04:37:04 http://ewi546.ewi.utwente.nl/OE_qa/cgi-bin//gunzip.cgi?tree=OpenEmbeddedBuild&full-log=1122204615.7653 Jul 24 04:39:54 zecke_: I think I killed bitbakeqa Jul 24 04:40:23 koen: how? Jul 24 04:40:43 rm /data/tinderbox*/data_dir/* -rf Jul 24 04:41:42 yes Jul 24 04:42:00 and same for html_output Jul 24 04:42:19 koen: the data_dir can be removed completely as it gets created... Jul 24 04:44:58 zecke_: you should see Unslung appear now too ... Jul 24 04:54:14 koen: did you delete the data? Jul 24 04:55:58 want me to do it again? Jul 24 04:56:47 koen: no I wonder from where he gets the h3900... build Jul 24 04:57:44 koen: I think I confused tinderbox with starting a new build when a old one hasn't been finished yet... Jul 24 04:58:44 heh Jul 24 04:58:51 ok, what do I need to rm? Jul 24 04:59:53 koen: I'm starting a new build and see what happens... Jul 24 05:00:04 ok Jul 24 05:02:18 rwhitby: 'tinderbox: starttime:', is greater than the tinderbox: timenow. Jul 24 05:02:46 I'm using gmtime as pasted above ... Jul 24 05:02:49 I need to study for my drivers license... I will fix the timezone issue later Jul 24 05:03:21 rwhitby: and I said it will only work when the tinderbox expects UTC as welll Jul 24 05:03:43 so should I change it back to localtime until then ? Jul 24 05:04:09 no leave it like that for now Jul 24 05:08:53 * koen install OE in his workstation Jul 24 05:09:09 lets see I can get the tinderboox to work Jul 24 05:10:11 does anybody here have the eleet user-mode-linux ski11z? Jul 24 05:10:24 nope Jul 24 05:10:34 drat Jul 24 05:10:35 isn't mickey|zzZZzz the uml master? Jul 24 05:10:43 ah, is he? Jul 24 05:10:46 mickey|zzZZzz: wake up! Jul 24 05:10:50 tch, lazy germans Jul 24 05:10:53 iirc he is using OE to make uml images Jul 24 05:11:28 ./packages/meta/uml-image.bb Jul 24 05:12:21 ah yes Jul 24 05:12:26 cya later Jul 24 05:12:28 I could drive to Frankfurt and wake him uf if it is really necessary ;-) Jul 24 05:12:40 florian: hehe. no, it is not that urgent :-) Jul 24 05:12:58 * zecke_ thinks tinderbox isn't as robust as one expects... Jul 24 05:13:09 zecke_: which tinderbox version are you using? Jul 24 05:13:26 pb_: tinderbox2 from mozilla/webtools Jul 24 05:13:35 hm, maybe that is your error Jul 24 05:13:41 I think mozilla is still using tinderbox1 Jul 24 05:13:45 iirc, tinderbox2 was abandoned Jul 24 05:13:53 really? Jul 24 05:14:04 pb_: but OSDL did work on tinderbox2 as well? Jul 24 05:14:14 zecke: I thought OSDL was working on tinderbox3 Jul 24 05:14:17 pb_: and the author works on tb3 outside of mozilla Jul 24 05:14:31 which, aiui, is based on tinderbox1 rather than tinderbox2 Jul 24 05:15:22 pb_: which tb did you use? Jul 24 05:15:28 1, I think Jul 24 05:15:48 http://www.mozilla.org/projects/tinderbox/ Jul 24 05:16:24 that does not sound like "tb2 is not done, go away"?! Jul 24 05:17:47 true Jul 24 05:17:57 can you tell by looking at the web page which version is in use? Jul 24 05:18:14 for example http://tinderbox.mozilla.org/showbuilds.cgi?tree=SeaMonkey-Ports Jul 24 05:18:20 one probably can... Jul 24 05:18:36 that looks like tinderbox1 output to me, but maybe tinderbox2 looks similar Jul 24 05:19:19 output looks similiar but the Administration looks different... Jul 24 05:27:29 moin, can anybody tell me how I can check out bitbake with svn? Jul 24 05:28:30 DeGT: no ask berlios Jul 24 05:29:32 rwhitby-away: congrats you've broken tinderbox ;) Jul 24 05:30:39 heh Jul 24 05:30:58 pb_: mozilla use tb1 (showbuilds.cgi is the indicator) Jul 24 05:40:05 zecke: ah, right, as I suspected Jul 24 05:40:22 I fear that probably means they will be immune to your DoS attacks Jul 24 05:41:22 pb_: probably Jul 24 05:41:43 I just need to fix process_mails Jul 24 05:44:09 the latest change to tb1 was in 2004 Jul 24 05:44:24 sam eapplies for tb2 :( Jul 24 05:44:32 doh Jul 24 05:44:46 maybe you should try to chase down the osdl tb3 stuff Jul 24 05:44:58 I never managed to find anyone who knew anything about that. Jul 24 05:45:08 pb_: I'm currently chasing Jul 24 05:45:21 http://www.johnkeiser.com/mozilla/tbox3.html ? Jul 24 05:45:22 http://www.osdl.org/cgi-bin/osdl_development_wiki.pl?Linux_Kernel_Tinderbox Jul 24 05:45:31 zecke_: ah, sounds like you are one step ahead of me Jul 24 05:45:41 koen: that is another TB3 ;) Jul 24 05:45:49 aaargh ;) Jul 24 05:45:51 but using postgress... Jul 24 05:45:57 crumbs Jul 24 05:46:34 I tempted to ask jg what he runs on fd.o but... Jul 24 05:46:44 zecke_: we can probably check FROM: headers in procmail if some nitwits want to sabotage us Jul 24 05:46:54 I don't think jg actually runs it. you could ask in #freedesktop, though. Jul 24 05:47:01 maybe someone like daniels or ajax would know Jul 24 05:47:19 http://www.freedesktop.org/Software/TinderboxWiki Jul 24 05:47:45 koen: the OSDL TB3 is based on the TB3 you posted Jul 24 05:47:47 ah, anholt Jul 24 05:48:26 good news for zecke's bsd troubles Jul 24 05:48:52 oh what did I miss?! Jul 24 05:49:18 Eric is a freebsd guru Jul 24 05:49:32 http://www.livejournal.com/users/anholt/ Jul 24 05:50:10 zecke_: see the june 22nd entry Jul 24 05:58:43 Did anyone here notice problems building for x86 based systems? Jul 24 05:59:22 not yet Jul 24 05:59:40 but I havent' really built anything in a week or so Jul 24 05:59:50 glibc doesn't seem to build anymore... Jul 24 05:59:59 for arm it is ok. Jul 24 06:00:25 there's a note on that in the mailinglist or bugzilla Jul 24 06:00:42 pb_: was it NTPL? Jul 24 06:00:46 ah.. let me see Jul 24 06:00:47 florian: yah, some dude complained of that on the mailing list the other day Jul 24 06:01:11 apparently the epia configuration, which is the only x86 that I test regularly, is still using some older glibc. Jul 24 06:01:37 * florian waits for Thunderbird to come up Jul 24 06:01:50 I'm not sure what's gone wrong with the new one. It sounds a bit weird. Jul 24 06:03:09 pb_: i tried latest and an older one which seem to suffer from the same problem Jul 24 06:03:25 florian: which older one did you try? Jul 24 06:04:12 pb_: 2.2.3+cvs 20050420 Jul 24 06:04:32 eh. s/2.2/2.3 Jul 24 06:04:44 try 2.3.2+cvs20040726 Jul 24 06:05:36 okay Jul 24 06:06:07 I'll look at the problem with the new one later on. Jul 24 06:06:27 meantime, can I talk you into updating gcc-cross-sdk to use 3.4.4? Jul 24 06:07:11 hmm I think I will stick to tb2 for now Jul 24 06:07:24 righto Jul 24 06:07:46 I will fix processmails and bugs and look for something better Jul 24 06:08:02 pb_: yes.. i guess that should be easy, i'll do this as soon as i have enough room for my workstation keyboard again :-) Jul 24 06:08:11 florian: excellent :-) Jul 24 06:10:57 bbiab, gotta go over to the office for a little while Jul 24 06:11:21 it is sunday?! addict Jul 24 06:11:59 zecke_: yes it is sunday ;-) Jul 24 06:14:20 hrm... Thunderbird is a pain on a 200MHz machine :-/ Jul 24 06:40:34 rwhitby: could you please stop using gmtime for now? Jul 24 06:43:37 sure - want me to change it back to localtime? Jul 24 06:45:11 rwhitby: for now yes Jul 24 06:45:17 ok, just kicked off a localtime build. let me know if that causes any problems ... Jul 24 06:46:03 * koen sees an extra UTC collumn at http://ewi546.ewi.utwente.nl/OE_qa/OpenEmbeddedBuild/status.html Jul 24 06:47:46 zecke_: 13 + 2 = 11? Jul 24 06:48:29 this is really strange Jul 24 06:48:54 it goes wrong for 8 entries, but corrects itself Jul 24 06:56:15 it showed logs correctly Jul 24 06:56:29 but then Slung send some data and tinderbox was wrong again... Jul 24 06:56:41 aha Jul 24 06:57:41 zecke_: have you seen http://venge.net:9000/ ? Jul 24 06:58:12 koen: yes but they do not publish what they use? Jul 24 06:58:22 buildbot Jul 24 06:58:33 yeah just seen that in the caption as well Jul 24 06:58:38 pb_: This version of glibc builds without trouble. Jul 24 06:58:38 the gstreamer people use it too Jul 24 06:59:33 15:57 <@njs`> it's a slightly weird fit to monotone, since these tools assume there is a linear history moving ever forward, which assumption monotone, err, breaks Jul 24 06:59:50 zecke_: that was about buildbot, but I think it applies to tb too Jul 24 07:00:02 right Jul 24 07:04:28 hmm - maybe the monotone people should add a "--use-latest-common-ancestor-instead-of-earliest" command line option ... Jul 24 07:04:51 yeah Jul 24 07:05:01 but you can force it with explicit_merge Jul 24 07:07:55 koen: buildbot is interesting but... Jul 24 07:08:12 it isn't perl? Jul 24 07:08:34 python Jul 24 07:09:00 eeks! Jul 24 07:09:15 * koen gets some snake repellent Jul 24 07:11:49 koen: could you run tinder.cgi as www-data manually once? Jul 24 07:12:25 tinderbox1? Jul 24 07:13:23 done Jul 24 07:15:03 t2 ;) Jul 24 07:15:09 any weird output? Jul 24 07:15:13 judgment day Jul 24 07:15:31 www-data@bitbake:~$ /data/tinderbox2/cgi-bin/tinder.cgi Jul 24 07:15:31 www-data@bitbake:~$ /data/tinderbox2/cgi-bin/tinder.cgi --daemon-mode Jul 24 07:15:31 www-data@bitbake:~$ Jul 24 07:16:03 hmm Jul 24 07:18:12 http://ewi546.ewi.utwente.nl/OE_qa/OpenEmbeddedBuild/status.html Jul 24 07:18:19 do you see h3900 build? Jul 24 07:18:52 right at the top Jul 24 07:19:05 and also a yellow bar? Jul 24 07:19:12 http://ewi546.ewi.utwente.nl/OE_qa/cgi-bin//gunzip.cgi?tree=OpenEmbeddedBuild&full-log=1122214447.24795#err2 Jul 24 07:19:52 yellow with "errs: 1" Jul 24 07:25:30 I hope it won't go away... Jul 24 07:29:50 how does bugzilla integration work? Jul 24 07:30:08 does it talk to bugzilla with xmlrpc or something? Jul 24 07:32:09 koen: bugzilla needs to send emails Jul 24 07:32:46 can tb report new bugs? Jul 24 07:33:19 yes we need to make bugzilla mail tinderbox-oe Jul 24 07:33:39 and bugs will show up in a column, with one click you enter the bugzilla Jul 24 07:34:03 when we've a VC_mtn we could also utilize the monotone QA options Jul 24 07:35:32 zecke_: the h3900 errors moved Jul 24 07:39:24 looks like the slung reports really bug tinderbox2 Jul 24 07:41:30 re Jul 24 07:41:32 florian: very good Jul 24 07:45:49 zecke_: can't you just drop the db and start a new build? Jul 24 07:46:42 koen: things look good now?! Jul 24 07:48:33 http://ewi546.ewi.utwente.nl/OE_qa/cgi-bin//tinder.cgi?tree=OpenEmbeddedBuild&start-time=1122194702&display-hours=48 Jul 24 07:48:43 apart from them being yellow, they look good Jul 24 07:48:54 but I still think starting with a fresh db is a good thing Jul 24 07:53:40 we've cleaned up the data_dir twice already Jul 24 07:54:15 why are the build yellow? Jul 24 07:54:42 in progress Jul 24 07:55:10 hmmmm Jul 24 07:55:15 and errors vanished as it looks only at the latest logfile Jul 24 07:55:25 why are there gaps in the timeframe? Jul 24 07:55:54 the one is compiling flite ther other gcc Jul 24 07:56:02 ah, I see Jul 24 07:56:14 * koen sets up a tinderbuild on his workstation Jul 24 07:56:28 ~stab ubuntu for not having monotone 0.21 Jul 24 07:56:29 * ibot runs at ubuntu for not having monotone 0.21 with an origami Swiss Army knife, and inflicts a nasty paper cut. Jul 24 07:57:33 zecke_: so if the next mail comes in, the bar will have no gaps? Jul 24 07:57:35 I think I will send tinderbox: timezone: Zone Jul 24 07:57:54 koen: like it just happened Jul 24 07:58:05 does tinder.bbclas parse tinder.conf? Jul 24 07:58:10 koen: the log is changing because I only send the current progress... Jul 24 07:58:11 no Jul 24 07:58:37 koen: do you've a bandwidth limitiation on ewi? Jul 24 07:58:50 not that I know off Jul 24 07:59:10 koen: do you get into trouble when running the source checker? Jul 24 07:59:12 and if the student standards apply it's 15GB/week to internet Jul 24 07:59:26 diskspace is the only concern Jul 24 08:00:05 they will phone my boss' boss when we do 200GB/day or so :) Jul 24 08:00:46 hehe Jul 24 08:11:54 zecke_: what does tinderbox use from sending emails? Jul 24 08:12:19 s/from/for/ Jul 24 08:13:18 koen: the tinderclient? Jul 24 08:13:25 yes Jul 24 08:13:29 koen: the python mail implementation Jul 24 08:13:50 does that need an local MTA? Jul 24 08:14:29 currently yes Jul 24 08:15:17 line 103 of tinderclient.bbclass Jul 24 08:16:31 afternoon Jul 24 08:17:04 hey Jul 24 08:17:30 hi zecke_ Jul 24 08:20:28 hey reenoo_ Jul 24 08:23:09 hi koen Jul 24 08:29:31 zecke_: I've added a tinderbox, could you see if it gets added correctly? Jul 24 08:31:59 koen: you do not deliver emails Jul 24 08:32:10 koen: what's your 3figures.patch for libtool supposed to do? Jul 24 08:32:14 Jul 24 17:30:27 localhost postfix/smtp[21836]: E05491160698: to=, relay=smtp.utwente.nl[130.89.1.83], delay=1, status=sent (250 2.0.0 j6OFUTb13585 Message accepted for delivery) Jul 24 08:32:25 reenoo_: make libtool support 1000 again Jul 24 08:32:34 reenoo_: needed for gnome, among others Jul 24 08:32:41 koen: what is your from? Jul 24 08:32:42 it is a regression in libtool Jul 24 08:32:53 Jul 24 17:30:27 localhost postfix/qmgr[10506]: EDC951165AC5: from=, size=468464, nrcpt=1 (queue active) Jul 24 08:33:46 koen: it did not show up in the from... Jul 24 08:33:55 koen: I bet GMX is just delaying mail delivery Jul 24 08:34:11 or not... Jul 24 08:34:20 koen: yeah, but where's it from and why is it again ltmain.sh and not ltmain.in? Jul 24 08:34:30 s/again/against/ Jul 24 08:37:19 koen: that update delay now is not explainable... Jul 24 08:37:56 [Sun Jul 24 16:44:35 2005] Variable: 'tinderbox: timenow:', is not of the form MM/DD/YY HH:MM:SS, or unix date, or your clock is set incorrectly, or the mail was delayed for a long time. Jul 24 08:38:11 and since then updates seem disabled... Jul 24 08:39:46 reenoo_: it's from the url in the cset, and ltmain.sh seemed the right place (I didn't know we used ltmain.in) Jul 24 08:41:32 zecke_: could you check gmx.net spamfilter? Jul 24 08:41:37 checked Jul 24 08:41:39 * koen goes to the kitchen to do the dishes Jul 24 08:43:08 koen: all other patches are against ltmain.in. for some people ltmain.sh is not re-generated from ltmain.in as a result of that which effectively means that all of our essential libtool fixes get lost. Jul 24 08:43:22 koen: can you remove that Build.Update..Tinder-ClientTest-EWI.1122218657 Jul 24 08:43:28 koen: and see what happens? Jul 24 08:52:14 god I love gmx Jul 24 08:52:14 delivered a mail with just 17hours of delay Jul 24 08:53:18 <_husam_> Hi, I'm port linux to HP h6300 and I'm at the stage where I'm trying to get gpe up and running on teh device Jul 24 08:53:31 <_husam_> so I'm winder if someone can help me Jul 24 08:53:34 zecke_: Sounds familiar... Siegen University suffers from these problems too. Jul 24 08:54:21 _husam_: Hi, sounds good! Which parts of the hardware are supported? Jul 24 08:55:04 <_husam_> I manage to get the touchscreen to work and anther guy writting the aduio driver Jul 24 08:55:16 <_husam_> for some reason my gpe build using oe crashing Jul 24 08:56:01 <_husam_> the device support OMAP 1510 Jul 24 08:56:14 <_husam_> it's anther IPAQ device Jul 24 08:56:25 _husam_: okay... when does it crash? Jul 24 08:56:36 florian: the mail was delivered to gmx directly... Jul 24 08:56:44 <_husam_> after I login to gpe Jul 24 08:57:00 <_husam_> I can get login screen for GPE with no problem Jul 24 08:57:09 _husam_, florian: -> #gpe Jul 24 08:57:35 <_husam_> thanks reenoo Jul 24 08:58:02 zecke_: well... that's a shame. But someone here counted 40 hops to send a mail form one office to the next ;-) Jul 24 08:59:41 ouch Jul 24 09:01:39 zecke_: do I have to change PARALELL_MAKE when I'm using icecc? Jul 24 09:01:53 micropal_: the bigger the better Jul 24 09:02:38 micropal_: you queue your compile jobs in the icecream scheduler Jul 24 09:06:07 zecke_: where should that build.update file be? Jul 24 09:09:15 koen: oh you did not delete it? Jul 24 09:09:24 I can't find it Jul 24 09:09:38 it vanished Jul 24 09:10:46 * koen wonders where his tinderbox mails goto Jul 24 09:10:57 koen: /data/tinderbox2/data_dir/OpenEmbeddedBuild/db Jul 24 09:11:08 Build.Update..Tinder-ClientTest-EWI.1122220286 Jul 24 09:12:27 removed it Jul 24 09:12:54 it looks like a stale file... Jul 24 09:13:02 is the build uid... Jul 24 09:28:18 koen: I don't think you're even using a version of ltmain.sh with the patch applied Jul 24 09:28:36 koen: grep -A 3 'each of the things' tmp/work/libtool-native-1.5.10-r1/libtool-1.5.10/ltmain.sh Jul 24 09:29:16 reenoo_: so you're saying the patch is {harm,use}less ? Jul 24 09:29:30 koen: run the grep command I pasted Jul 24 09:29:56 I didn't find icecc deb's so I made some if anyone wants it: http://folk.uio.no/alexawo/debian/ice-i386/ Jul 24 09:30:06 grep: tmp/work/libtool-native-1.5.10-r1/libtool-1.5.10/ltmain.sh: No such file or directory Jul 24 09:30:36 heh Jul 24 09:31:01 koen@bitbake:~/OE/build/tmp/familiar/work/i686-linux$ grep -A 3 'each of the things' libtool-native-1.5.10-r1/libtool-1.5.10/ltmain.sh Jul 24 09:31:03 # Check that each of the things are valid numbers. Jul 24 09:31:07 case $current in Jul 24 09:31:09 0 | [1-9] | [1-9][0-9] | [1-9][0-9][0-9]) ;; Jul 24 09:31:12 *) Jul 24 09:31:44 right Jul 24 09:31:47 no effect Jul 24 09:32:01 that explains some things Jul 24 09:32:11 * koen adapts patch to work on ltmain.in Jul 24 09:33:07 koen: a success log should soon be sent Jul 24 09:33:28 zecke_: great Jul 24 09:34:00 koen: I wonder how this delay happens. I will do a local setup during this week Jul 24 09:35:56 ugh. monotone screwed up the log message Jul 24 09:36:05 ~lart monotone Jul 24 09:36:05 * ibot DoSes monotone Jul 24 09:36:30 afternoon all Jul 24 09:36:37 hey RP Jul 24 09:36:38 hey RP Jul 24 09:37:02 RP: did you get my mail on the pdaX w100 claims? Jul 24 09:37:03 Looks like 2.6.12-rc3 works (ish) Jul 24 09:37:17 03rw 07org.oe.dev * re8b84bc3... 10/packages/libtool/ (libtool-native_1.5.10.bb libtool_1.5.10.bb): libtool: disable 3figures.patch for now. see bug #33 for details.MT: ---------------------------------------------------------------------- Jul 24 09:37:20 koen: I did. Looks interesting. I'll be interested to see what they've done Jul 24 09:37:40 2.6.12-rc3 does report two false soft lockups upon bootup... Jul 24 09:37:53 RP: I was hoping you could contact them asking if they will use EXA or some funky russian hack Jul 24 09:37:54 Makes me think we should just turn off the softlockup detector ;-) Jul 24 09:38:19 koen: I'll contact them and ask if I can see what they've done :) Jul 24 09:38:25 reenoo_: monotone didn't screw up the message, you did. Don't remove the newline next time Jul 24 09:38:44 fwiw, you can do the same with cvs Jul 24 09:38:48 koen: I supplied the message via -m on the command line Jul 24 09:39:37 koen: there's no newline to remove there :) Jul 24 09:39:59 * koen wonders how that happened Jul 24 09:41:24 RP: from what I hear the X people are really liking EXA over XAA Jul 24 09:43:33 I don't follow it too closely but that would seem to be the case Jul 24 09:43:55 I'm just worrying about the low level stuff and assuming something like mallum can sort that bit out easily ;- Jul 24 09:43:57 ;-) Jul 24 09:44:19 The xserver causes me far too many headaches (spent the last week tracking a bug in it) Jul 24 09:44:45 I'm worried about the pda people having accell in some weird hacking way, which is no use to others Jul 24 09:44:58 koen: can you spare another link to EXA and W100? Jul 24 09:45:02 /pda/pdaX/ Jul 24 09:45:05 Nothing would surprise me :-/ Jul 24 09:45:48 zecke_: http://www.oesf.org/forums/index.php?showtopic=13874&pid=89164&st=0&#entry89164 and http://dot.kde.org/1119948104/ Jul 24 09:46:16 zecke_: and you are forced to like EXA ;) Jul 24 09:46:32 koen: I think we need to use http post instead of email... Jul 24 09:46:50 zecke_: http would be excellent Jul 24 09:47:03 sure zack is doing it... I hope he will do the 100% Jul 24 09:47:15 instead of stopping at 90% again Jul 24 09:47:50 maybe he should eat more meat Jul 24 09:47:57 * koen hides Jul 24 09:49:08 too bade I can not type a pipe symbol here Jul 24 09:50:22 * RP ponders - Should I upgrade the Zaurus kernel despite the softlockup "bug" report... Jul 24 09:51:15 koen: could you check if I sent a bigger mail recently? Jul 24 09:51:39 http://ewi546.ewi.utwente.nl/mail/from Jul 24 09:52:03 koen: sent != received Jul 24 09:52:16 ah Jul 24 09:53:06 2005-07-24 18:50:33 1DwjgP-0007iR-2U <= tinderbox@ewi546.ewi.utwente.nl H=localhost (bitbake.utwente.nl) [127.0.0.1] U=zecke P=esmtp S=39161 Jul 24 09:53:09 2005-07-24 18:50:33 1DwjgP-0007iT-4o <= tinderbox@ewi546.ewi.utwente.nl H=localhost (bitbake.utwente.nl) [127.0.0.1] U=zecke P=esmtp S=835 Jul 24 09:53:13 it was not yet received, and is not in the gmx mailbox Jul 24 09:53:21 oh 3.9MB Jul 24 09:53:39 and a 1.4mb mail before that Jul 24 09:55:34 zecke_: and Xorg 6.9/7.0 is shipping with EXA Jul 24 09:55:43 but with XAA as the default renderer Jul 24 09:56:44 man gmx sucks ass... Jul 24 09:58:24 any perl guru around? Jul 24 10:01:02 reenoo_dinner: I fixed the patch and merged the heads, libtool now has all patches applied to ltmain.in Jul 24 10:01:21 * koen hugs meld Jul 24 10:02:24 03koen 07org.oe.dev * r6311b4a3... 10/packages/libtool/ (libtool-1.5.10/3figures.patch libtool_1.5.10.bb): apply 3figures.patch to ltmain.in instead of ltmain.sh Jul 24 10:06:19 hi mickeyl Jul 24 10:06:35 hey mickeyl Jul 24 10:06:43 hi folks Jul 24 10:06:58 needing a lot of sleep on your old day? Jul 24 10:07:32 heh Jul 24 10:07:32 mickeyl: Did you read my comments about 2.6.13-rc3-mm1? I'm probably in favour of bumping the kernel version in this big jump, despite the false softlockup thing... Jul 24 10:07:32 03mickeyl 07org.oe.dev * rf9eeaf8a... 10/packages/initscripts/initscripts-1.0/tosa/keymap-2.4.map: tosa keymap: populate some more keys to make USB keyboards behave Jul 24 10:07:36 03mickeyl 07org.oe.dev * r709d2624... 10/packages/meta/meta-sdk.bb: meta-sdk: add opie libraries Jul 24 10:07:40 03mickeyl 07org.oe.dev * r078ff68c... 10/packages/initscripts/initscripts-1.0/tosa/keymap-2.4.map: tosa keymap: add num lock and caps lock Jul 24 10:07:43 03mickeyl 07org.oe.dev * r295b25bd... 10/packages/initscripts/initscripts-1.0/tosa/keymap-2.4.map: Jul 24 10:07:43 tosa keymap: map holding the left dotted button to AltGr and Jul 24 10:07:43 map AltGr+Left|Right to Incr_Console and Decr_Console Jul 24 10:08:19 Before we get carried away with the keymaps, can ber please start some documentation btw? :) Jul 24 10:08:33 RP: I agree. Bump the devel kernel Jul 24 10:08:41 can *we* please Jul 24 10:08:53 mickeyl: is the SDK shaping up? Jul 24 10:09:00 koen: no idea, i can't build it :/ Jul 24 10:09:15 i've added the opie stuff, but i can't build it since it chokes on soft float Jul 24 10:09:23 i'll try to launch a hard float build later Jul 24 10:10:23 koen: what's the preferred revision of libtool? (merge here) with or without the 3figures patch? Jul 24 10:11:15 mickeyl: monotone sync Jul 24 10:11:50 http://monotone.vanille.de/viewmtn/headofbranch.psp?branch=org.openembedded.dev doesn't show more than one head :) Jul 24 10:11:55 hmm Jul 24 10:12:07 it's local here Jul 24 10:12:28 so... without or with the patch? Jul 24 10:12:41 with, but please don't merge Jul 24 10:12:45 hmm Jul 24 10:12:55 i don't think i can revert that Jul 24 10:13:02 http://monotone.vanille.de/viewmtn/revision.psp?id=91f9e83267ccfd470d7b09f7e18aa2f8cd9383e0 already merges it Jul 24 10:13:08 yeah Jul 24 10:13:15 in the second _after_ i completed the merge here Jul 24 10:13:30 hmm maybe disapprove will do Jul 24 10:14:10 2 changesets cannot invert Jul 24 10:15:35 *shrug* Jul 24 10:15:40 * mickeyl copies oe.db to mickey.db Jul 24 10:16:25 handy if you have the master db on the same fs as the personal one :) Jul 24 10:16:29 can dump one all the time Jul 24 10:17:04 hehe Jul 24 10:18:02 koen@bitbake:~/OE/monotone/openembedded$ monotone merge Jul 24 10:18:02 monotone: misuse: branch 'org.openembedded.dev' is merged Jul 24 10:18:04 * koen happy Jul 24 10:18:06 03koen 07org.oe.dev * r7207f6f2... 10/packages/gpe-mini-browser/gpe-mini-browser_0.14.bb: add gpe-mini-browser, a *web* browser for the people who didn't know what it meant. Jul 24 10:18:15 koen: did you actually test the patch? (you didn't bump PR on libtool-native) Jul 24 10:18:34 reenoo_: I tested it, and I merged in your PR bump Jul 24 10:19:03 libtool != libtool-native Jul 24 10:19:18 libtool-native sets PR independently Jul 24 10:19:27 aargh Jul 24 10:20:56 pr bumped Jul 24 10:21:44 grep -A 3 'each of the things' libtool-native-1.5.10-r1/libtool-1.5.10/ltmain.sh Jul 24 10:21:45 # Check that each of the things are valid numbers. Jul 24 10:21:45 case $current in Jul 24 10:21:45 0 | [1-9] | [1-9][0-9]*) ;; Jul 24 10:21:45 *) Jul 24 10:22:57 very good Jul 24 10:36:18 zecke_: on the ewibox you can send mail to the local www-data user Jul 24 10:36:28 * chouimat is away: laundry Jul 24 10:41:24 A quick monotone question - what's the correct procedure for renaming a file/directory? Jul 24 10:41:32 see here: Jul 24 10:42:06 function mtmv() { mt rename $1 $2 && mv $1 $2; } Jul 24 10:42:12 for mt=monotone Jul 24 10:42:23 + command line arguments Jul 24 10:42:31 note that this only works for files Jul 24 10:42:43 monotone doesn't care about dirs, so you would need to write another function for that Jul 24 10:42:49 So you need to run mv and mt rename ?! Jul 24 10:42:58 exactly Jul 24 10:43:11 although I'd use the order I shows Jul 24 10:43:16 s/shows/shown/ Jul 24 10:43:19 first monotone rename then mv though Jul 24 10:43:23 right Jul 24 10:44:02 Ok, thanks :) Jul 24 10:44:15 mickeyl: won't mt rename ; mt update do the same? Jul 24 10:44:26 never tried Jul 24 10:44:29 it would take much longer though Jul 24 10:44:29 since rename would be a db operation Jul 24 10:44:35 Thankfully I used mt rename first anyway, then got confused as much hadn't happened :) Jul 24 10:44:52 koen: no. you need to manually perform the mv. Jul 24 10:44:57 koen: I tried that and didn't see it take effect. I was doing both a file rename and a directory rename though Jul 24 10:45:35 and rename, commit, update? Jul 24 10:46:19 Ah, I didn't commit it :) Jul 24 10:46:21 koen: the mv has to be done manually. Jul 24 10:46:25 the db || local copy seperation is a bit fuzzy Jul 24 10:46:40 koen: that's by intention/design Jul 24 10:46:42 I'm still nervous about commiting :) Jul 24 10:46:52 just pull and update before doing it Jul 24 10:46:55 the rest will follow :) Jul 24 10:48:22 and see http://article.gmane.org/gmane.comp.handhelds.openembedded/5271 Jul 24 10:48:33 mickeyl: good morning Jul 24 10:48:54 heya pb Jul 24 10:49:09 hi pb_ Jul 24 10:49:13 hi rp Jul 24 10:50:01 RP: what kind of bug where you tracking down in X? Jul 24 10:50:18 koen: an Xomap one Jul 24 10:50:25 ah Jul 24 10:50:27 nokia voodoo Jul 24 10:50:34 indeed Jul 24 10:51:06 didn't they hack xcalibrate? Jul 24 10:52:31 koen: You've seen the sources. The hacked quite a few things ;-) Jul 24 10:53:14 yeah Jul 24 10:53:22 THey should push upstream really soon now Jul 24 10:54:59 koen: Some stuff like Xomap won't be accepted Jul 24 10:55:18 I was affraid of that Jul 24 10:55:27 any particular reason? Jul 24 10:56:13 * chouimat is back. Jul 24 10:57:26 koen: Copying the damage layer into the omap code would be a primary reason Jul 24 10:59:11 03rpurdie 07org.oe.dev * r356722c1... 10/packages/linux/ (6 files in 2 dirs): linux-oz-2.6: Upgrade to 2.6.13-rc3-mm1, update several ipaq and tosa patches Jul 24 10:59:29 I think I got that right.. ;-) Jul 24 10:59:29 RP: did you read lkml lately? Jul 24 10:59:37 mickeyl: bits of it. why? Jul 24 11:00:22 RP: there is someone posting collie touschreen patches. unfortunately he neither knows that we exist or doesn't want to cooperate with us. I would have thought he'd contacted John Lenz as the official collie maintainer in 2.6, but it doesn't look like it Jul 24 11:00:37 mickeyl: Its ok, he does know about us Jul 24 11:00:43 good to hear Jul 24 11:00:53 why is he feeding ts patches through lkml and not alkml ? Jul 24 11:00:54 You'll note both John and I are in the cc list ;-) Jul 24 11:01:10 * france is back (gone 12:07:59) Jul 24 11:01:20 before I run out the door Jul 24 11:01:25 Well, its input system so they should go through the input maintainer Jul 24 11:01:35 mickeyl: it may not be arm specific Jul 24 11:01:44 besides, at first glance the patch shouldn't be called collie-ts.c but rather ucb1x00-ts.c because it looks like it's not collie specific. SIMpad has the same ts iirc Jul 24 11:01:50 The mcp stuff is interesting and has put rmk's nose out of joint though :) Jul 24 11:02:13 mickeyl: getting patchs through rmk is always difficult, he dislikes hh.org,has stated so many times. Jul 24 11:02:19 mickeyl: That may be. If he can get the code in, we can then submit patches to tweak it ;-) Jul 24 11:02:19 well, given that Zecke ported the mcp stuff to 2.6 mid 2004 but rmk didn't want it to go in Jul 24 11:02:23 k Jul 24 11:02:25 good plan Jul 24 11:02:54 france: yeah, but we zaurus people are not hh.org :D Jul 24 11:02:54 mickeyl: any patch that does not have to goto rmk, shouldn't. Jul 24 11:03:02 Given I don't have much to do with Collie, its John's call. He's more interested in poodle so any collie work is probably good by both of us Jul 24 11:03:17 sure Jul 24 11:03:23 to be honest... collie 2.6 is a dead end anyway Jul 24 11:03:36 i don't think many people would use it without SD Jul 24 11:03:39 I've told him as much offlist Jul 24 11:03:51 but he's still keen so... Jul 24 11:04:00 getting the MCP and UCB stuff into 2.6 is worthwhile anyway Jul 24 11:04:10 We may yet talk kergoth into a driver :) Jul 24 11:04:15 * france is away: Away Jul 24 11:04:16 heh Jul 24 11:04:18 i wouldn't count on that .) Jul 24 11:04:33 :) Jul 24 11:05:07 I'm off for food. Its nice to be back uptodate with the kernel though :) Jul 24 11:05:27 yeah. we will have to sort out pcmcia and udev/hotplug asap Jul 24 11:05:35 then we can ship 3.5.4 Jul 24 11:06:32 last time i tried udev it freaked out with alignment errors when reading its rules Jul 24 11:07:13 is anyone still working on libtool? mickeyl? koen? Jul 24 11:07:20 not me Jul 24 11:09:43 koen: ping Jul 24 11:10:07 hmm Jul 24 11:10:08 ERROR: libtool_1.5.10.bb:36: unparsed line: '}' while parsing /local/pkg/oe/org.openembedded.dev/packages/libtool/libtool-native_1.5.10.bb Jul 24 11:10:18 mickeyl: that's why I ask Jul 24 11:10:21 ah Jul 24 11:10:38 mickeyl: just want to avoid the merging ping pong Jul 24 11:11:30 reenoo_: could you pull and update to see if it's still there? Jul 24 11:13:18 koen, RP: looks like one of you created multiple heads again Jul 24 11:14:16 koen: and yes, libtool.bb still has a bogus extra "}" Jul 24 11:14:27 03koen 07org.oe.dev * rdd4ebed2... 10/packages/libtool/libtool-native_1.5.10.bb: bump PR Jul 24 11:14:46 koen: AFAICT from viewmtn anyway Jul 24 11:15:37 File "/usr/lib/meld/prefs.py", line 52, in ? Jul 24 11:15:37 import gconf Jul 24 11:15:37 ImportError: No module named gconf Jul 24 11:15:43 ~lart gentoo Jul 24 11:15:44 * ibot readies the nuke launcher and fires some rounds at gentoo Jul 24 11:16:47 and the heads are gone Jul 24 11:18:57 koen: would you mind fixing the extra '}' in the libtool .bb while you're at it? Jul 24 11:19:56 koen: one of your merges went wrong somehow Jul 24 11:26:07 * koen looks at it Jul 24 11:37:53 reenoo_afk: fixed, it will show up at vanille after it allows me to connect again Jul 24 11:40:07 03koen 07org.oe.dev * rd73cf33a... 10/packages/libtool/libtool_1.5.10.bb: remove stray } Jul 24 11:41:14 * CoreDump|home plays with UAE on his Z Jul 24 11:52:28 bleh, the mouse is b0rked Jul 24 11:55:22 * den1 bows to you Jul 24 11:55:42 anyone here working on apache module bb's? Jul 24 11:55:47 (apxs) Jul 24 11:59:01 Argh! I constanly get an error about "unexpected token `'INHIBIT_PACKAGE_STRIP','" if i try to build a kernel package :-/ Jul 24 11:59:20 Anyone seen this before? Jul 24 11:59:49 florian: check your DISTRO Jul 24 12:00:22 koen: in linux-hotplug package " $(INSTALL) -d $(etcdir)/hotplug/{usb,pci} " fail for me(and also other lines in Makefile after that), it create "/etc/hotplug/{usb,pci}" dir. but should be "pci" and "usb". do you have idea Jul 24 12:00:24 03law 07org.oe.dev * r566032cf... 10/packages/sylpheed/ (files/sylpheed-gnutls_2.0.0rc.patch sylpheed_2.0.0rc.bb): added sylpheed2.0.0rc bb and patch file Jul 24 12:00:43 pb_: familiar-0.8.2 Jul 24 12:00:45 koen: what it could be ? Jul 24 12:05:04 koen: that is a bit annoying for, every time i want rebuild tree of packages, i need fix it by hand Jul 24 12:06:35 koen: mybe i have wrong version of "install" Jul 24 12:06:58 koen: main is "install (coreutils) 5.2.1" Jul 24 12:23:23 * chouimat is away: tv Jul 24 12:23:45 florian: hm, that should be okay Jul 24 12:23:57 what does the offending piece of script look like? Jul 24 12:34:09 <_husam_> I have built Xomap but when I tried to run I get "/lib/libm.so.6: version `' not found (required by /usr/bin/Xomap" Jul 24 12:34:31 <_husam_> but libm.so.6 is installed Jul 24 12:34:45 _husam_: did you run ldconfig? Jul 24 12:34:57 <_husam_> No .. I haven't Jul 24 12:35:05 _husam_: give it a try Jul 24 12:35:24 <_husam_> I just did ... getting the same error Jul 24 12:35:41 :( Jul 24 12:35:51 aquadran: it isnt install that does that substitution, its the shell. oe is spawning its shell scripts with /bin/sh, but assuming that that shell knows how to expand {,}, which is flawed. we shouldnt be relying on bashisms like that. Jul 24 12:36:24 <_husam_> den1: Yah I know :) ... I think there spomething wrong with my build Jul 24 12:38:29 hi kergoh Jul 24 12:38:31 kergoth, too Jul 24 12:47:06 hey Jul 24 12:52:52 pb_: indeed... i can't imagine having seen this effect before. Jul 24 12:52:53 zecke_: Jul 24 12:52:54 >>>RCPT To: Jul 24 12:52:54 <<< 450 4.3.2 {mx063} Too many mails (mail bomb), try again in 2 hour(s) 12 minute(s) Jul 24 12:52:57 ... Deferred: 450 4.3.2 {mx063} Too many mails (mail bomb), try again in 2 hour(s) 12 minute(s) Jul 24 12:53:02 zecke_: all hail to gmx Jul 24 12:53:24 hey zecke, you about? Jul 24 12:53:43 zecke: i'm in Berlin Zoo station atm :P Jul 24 12:54:11 and using a very lagged wireless access point Jul 24 12:54:36 koen: was it you who generated multiple heads or me? Jul 24 12:55:03 RP: no idea Jul 24 12:55:48 I did but I thought I'd merged them. Looking at the graph confuses the issue :) Jul 24 12:56:30 your merge created another head Jul 24 12:57:06 or rather, my commit created divergence Jul 24 12:57:18 That's all there was on the server at the time though. I guess we were just unlucky with the timing... Jul 24 12:57:42 we're not using the same server :) Jul 24 12:58:18 that probably doesn't help :) Jul 24 13:01:01 kergoth: thanks, it my system(debian) sh was linked into dash, so i changed it to bash, and it works. i hope it's ok for system :) Jul 24 13:01:10 bleh Jul 24 13:01:28 * koen gets a *lot* of 'couldn't send for the past 4 hours' mails Jul 24 13:01:31 ~lart gmx Jul 24 13:01:31 * ibot gives gmx a "free" copy of Windows and then charges double for "Upgrades" Jul 24 13:04:20 * mithro collects his 300 emails Jul 24 13:07:33 yay 50% done Jul 24 13:09:06 aquadran: that should be fine. /bin/sh as ash or dash is perfectly reasonable. its oe making that flawed assumption. should fix that eventually Jul 24 13:11:08 * koen tries another tinderbox build Jul 24 13:11:34 koen: if you need a free email account, I could provide it Jul 24 13:12:01 CoreDump|bbl: we decided to go the http post route Jul 24 13:12:12 since tinderbox doesn't like delayed mails Jul 24 13:12:19 ah i c Jul 24 13:27:23 * chouimat is back. Jul 24 13:47:05 03nail 07org.oe.nslu2-linux * rf3173306... 10/packages/ncftp/ncftp_3.1.9.bb: Upstream upgrade Jul 24 13:47:09 03nail 07org.oe.nslu2-linux * r5b948a2a... 10/packages/man/ (man/man.conf man_1.5p.bb): Upstream upgrade Jul 24 13:47:13 03nail 07org.oe.nslu2-linux * r9203ea8d... 10/packages/streamripper/streamripper_1.61.10.bb: Added streamripper Jul 24 13:47:16 03nail 07org.oe.nslu2-linux * rf3826b85... 10/packages/groff/ (groff/groff.patch groff_1.19.1.bb): Added. Needed for man Jul 24 13:47:20 03nail 07org.oe.nslu2-linux * rf9ab42d4... 10/packages/procps/ (5 files in 2 dirs): updated version, patched to support OE cross install-sh Jul 24 13:47:49 * NAiL hums innocently Jul 24 13:57:38 03nail 07org.oe.nslu2-linux * rcacdd0f9... 10/packages/screen/ (screen-4.0.2/configure.patch screen_4.0.2.bb): Added screen Jul 24 13:58:50 koen|tv: that would explain a bit of the merging chaos we saw earlier today. would you mind switching to the same primary server the rest of us use? Jul 24 13:59:18 koen|tv: (i.e. monotone.vanille.de) Jul 24 13:59:53 nite guys Jul 24 14:00:34 'night zecke_ Jul 24 14:00:48 koen|tv: I'm switching code tomorrow to use HTTP post instead of mail... Jul 24 14:05:10 'night zecke Jul 24 14:08:45 reenoo_: well since I can't access vanille most of the time I commit to my db and sync that, just like all of you do Jul 24 14:08:51 I just sync once an hour Jul 24 14:11:15 well.. mickeyl asked us to use "monotone pull && monotone merge && monotone sync" to push for a reason Jul 24 14:13:25 also, I believe the extra "monotone pull" reduces the risk of breaking the dbs on the servers Jul 24 14:14:04 pull is a one way operation limiting any potential damage to the local db Jul 24 14:16:51 but in the monotone world it doesn't matter when you sync Jul 24 14:17:04 since your commits have an explicit ancestor Jul 24 14:17:22 only the moment of committing counts Jul 24 14:23:31 note the 'monotone merge' in that series of commands. I don't see how you could create multiple heads that way (the timeframe and thus the probability is rather low anyway) Jul 24 14:24:17 it isn't Jul 24 14:24:28 that is exactly what happend tonight Jul 24 14:24:53 aha Jul 24 14:25:40 NAiL: you got screen compiling? Any idea if it works with OZ? Jul 24 14:25:55 it works on familiar Jul 24 14:26:25 JustinP: not me. I just committed it. But it works for me. Try it and see ;-) Jul 24 14:26:26 NAiL: did you add or update screen? Jul 24 14:26:39 screen has been in OE for a while Jul 24 14:26:53 but suffered glib breakage last time I checked Jul 24 14:27:00 really? It wasn't in my tree before I added it Jul 24 14:27:05 it's in nonworking Jul 24 14:27:07 atleast not in packages Jul 24 14:27:15 aha Jul 24 14:27:32 we really should go over the nonworking packages once in a while Jul 24 14:27:39 they suffer from extreme bitrot Jul 24 14:28:25 hmmm.....it's in the nslu2 branch...I'd rather not propagate from that one...I have enough merge issues as it is Jul 24 14:28:41 03mickeyl 07org.oe.dev * r905af256... 10/packages/linux/ (3 files in 2 dirs): Jul 24 14:28:41 openzaurus-pxa27x: autoload usb host module Jul 24 14:28:41 openzaurus-pxa: remove outdated bluetooth patch, use local WE patches Jul 24 14:28:45 03mickeyl 07org.oe.dev * rc38b841b... 10/conf/machine/spitz.conf: spitz.conf: ship all of e2fsprogs in default image Jul 24 14:28:49 03mickeyl 07org.oe.dev * r8824e706... 10/conf/machine/zaurus-clamshell-2.4.conf: zaurus-clamshell-2.4.conf: ship usb host modules in default image Jul 24 14:28:53 03mickeyl 07org.oe.dev * r6e10609c... 10/packages/misc-binary-only/ (prism3-firmware_1.8.4.bb prism3-firmware_1.8.3.bb): Jul 24 14:28:53 prism3-firmware: Jul 24 14:28:53 - add v1.8.4 Jul 24 14:28:53 - fix SRC_URI in v1.8.3 Jul 24 14:28:53 - set PACKAGE_ARCH to noarch Jul 24 14:29:26 hmm Jul 24 14:29:30 was it really noarch? Jul 24 14:29:33 or rather 'all' ? Jul 24 14:30:32 we have both Jul 24 14:30:34 iirc Jul 24 14:30:41 noarch is a redhat thing Jul 24 14:30:51 *sigh* Jul 24 14:31:09 wouldn't hurt if we decide on one Jul 24 14:31:47 hmm Jul 24 14:32:07 all .bb's seem to use all Jul 24 14:32:11 * mickeyl fixes the two prism packages Jul 24 14:32:24 koen: hmm. ok. Jul 24 14:34:18 03mickeyl 07org.oe.dev * r20bf7ae2... 10/packages/misc-binary-only/ (prism3-firmware_1.8.3.bb prism3-firmware_1.8.4.bb): prism3-firmware: s/noarch/all/ Jul 24 14:34:23 i think if merging would automatically update the working copy we would have less merges, as well Jul 24 14:34:41 because if you forget to update, then one the next pull you will have to merge again Jul 24 14:34:53 even if no one did changes except you Jul 24 14:35:15 yeah Jul 24 14:35:39 nothing that can't be fixed with a couple of aliases and functions though Jul 24 14:35:46 I'm just trying to figure out how to avoid as many problems as possible Jul 24 14:35:48 right Jul 24 14:35:59 hmm. bummer, i have the libtool } again Jul 24 14:36:13 perhaps my cache Jul 24 14:36:15 * mickeyl cleans Jul 24 14:37:01 http://monotone.vanille.de/viewmtn/revision.psp?id=fca1070eaa90db3b30a008267883aa0c7b5425dd Jul 24 14:37:14 it shows what i've explained. without updating you fork yourself Jul 24 14:37:17 ~lart mickeyl Jul 24 14:37:17 * ibot declares mickeyl a moron Jul 24 14:37:40 I guess we could write a mandatory wrapper script or something Jul 24 14:37:45 * JustinP likes monotone but it seems to have some warts Jul 24 14:37:59 *nod* a couple of functions folks need to source in their bashrc and we would be set Jul 24 14:38:13 could you start with something like that on the MonotonePhraseBook ? Jul 24 14:38:14 doesn't openslug use some special makefiles or something? Jul 24 14:38:23 we can later add them to 'contrib' or so Jul 24 14:38:42 ah, righto, a wrapper script Jul 24 14:38:43 even better Jul 24 14:41:11 jesus...I pull, it finshes, I pull again and there are 6 more revs Jul 24 14:43:52 mickeyl: hmm. I can try to. I should be finishing a paper and learning for a couple of exams though :/ Jul 24 14:44:04 do that instead Jul 24 14:44:04 speaking of which Jul 24 14:44:10 we can do the wrapper later Jul 24 14:55:56 03mickeyl 07org.oe.dev * rf32d65a1... 10/packages/libtool/libtool-native_1.5.10.bb: libtool: fix parsing error. functions need to have at least one line Jul 24 14:56:00 03mickeyl 07org.oe.dev * rc547435e... 10/packages/libtool/ (3 files): libtool: functions need to have at least one line Jul 24 14:56:37 g'night Jul 24 14:56:45 mickeyl: good night Jul 24 15:01:00 <[g2]> nite mickey|zzZZzz Jul 24 15:02:06 * chouimat is away: stuff Jul 24 15:03:46 hey people Jul 24 15:03:47 in Berlin now :P Jul 24 15:03:54 guten abend Jul 24 15:03:58 hi mithro Jul 24 15:04:08 have you seen master hacker Holger yet? Jul 24 15:08:16 so what is up? Jul 24 15:08:20 apart from the sky? Jul 24 15:08:42 bugs... lots of bugs! Jul 24 15:10:12 a big floating yellow orb that burns my eyes when I try to see wtf it is Jul 24 15:11:08 hey ljp Jul 24 15:11:11 how goes Australia? Jul 24 15:11:27 chilly Jul 24 15:11:37 seeing as I'm no longer in it :P Jul 24 15:12:18 cloudy and rainy last week Jul 24 15:12:40 we finally got the first qtopia phone in the office here Jul 24 15:13:41 that sounds cool Jul 24 15:13:49 http://linuxdevices.com/articles/AT6230413091.html Jul 24 15:14:39 hi ljp Jul 24 15:15:27 hi Jul 24 15:18:49 hey pb_ Jul 24 15:19:39 ljp: it any good? Jul 24 15:22:18 its really light Jul 24 15:22:50 and they customized the interface, it doesnt look like qtopia phone at all Jul 24 15:26:42 hi mithro Jul 24 15:31:45 bah ha ha! Jul 24 15:31:47 Most obvious distinction is that BSD (and I guess FreeBSD) is dead. Jul 24 15:31:47 Linux has yet to add that feature... or perhaps Linus is simply rejecting the patch for now. My guess is that he didn't want to junk up the kernel with rotting and decaying matter. He's probably looking for a good user space implementation of death. We'll see. Jul 24 15:53:01 good night Jul 24 16:00:33 JustinP: does pcre build for you with the current libtool-native btw? Jul 24 16:00:52 JustinP: (and without your own patch) Jul 24 16:14:05 Is there a problem it seems monotone is hanging at 23.6 mb Jul 24 16:14:36 And I got a warning monotone: warning: No branches found. Jul 24 16:17:21 it's not exactly hanging. the initial pull takes ages Jul 24 16:17:50 renoo_busy thanx than I will make some coffee in between Jul 24 16:17:53 up to a couple of hours depending on your machine Jul 24 16:18:24 amd64 1gb ram Jul 24 16:19:09 I guess a bit over an hour in that case Jul 24 16:19:54 hehehe Jul 24 16:27:09 reenoo_bosy: checking now Jul 24 16:31:16 renoo_busy: I cleaned any libtool and pcre I had and pcre is failing again Jul 24 16:34:23 JustinP: hmm. could you check if the resulting ltmain.sh has all of our libtool patches applied? Jul 24 16:34:44 one sec....I'm trying again Jul 24 16:35:07 looks like 3figures waas applied to libtool-native Jul 24 16:35:55 libtool-native is compiling again, I'll try pcre again in a sec Jul 24 16:36:52 nope..doesn't seem to work. It rebuilt libtool and pcre and its still failing Jul 24 16:37:32 I really don't know how I can check the ltmain.sh... Jul 24 16:38:30 grep PB tmp/work/pcre-4.4-r1/pcre-4.4/ltmain.sh Jul 24 16:38:49 http://dioxide.randomvoids.com# This breaks install into our staging area. -PB Jul 24 16:38:55 oops, forget the URL Jul 24 16:39:01 # This breaks install into our staging area. -PB Jul 24 16:39:24 odd Jul 24 16:39:37 that's our patched version Jul 24 16:42:47 it's still failing with the same /usr/lib/libc.a failure Jul 24 16:46:28 just a wild guess.. has libtool-cross been rebuilt? Jul 24 16:46:49 03nail 07org.oe.nslu2-linux * r784c0507... 10/packages/meta/openslug-packages.bb: Added man, man-pages, ncftp, psmisc, screen and streamripper Jul 24 16:53:56 reenoo_busy: no, only libtool-native. I cleaned out libtool* in tmp/work and tmp/stamps and libtool-native is all that was rebuilt before pcre Jul 24 16:55:38 JustinP: could you try rebuilding libtool-cross? Jul 24 16:56:00 (bitbake libtool-cross should do that) Jul 24 16:57:21 reenoo_busy: doesn't he need to clear an existing build? Jul 24 16:58:08 Luke-Jr: I've bumped PR. so, no. Jul 24 17:00:24 Luke-Jr: the only libtool in work or stamps is libtool-native Jul 24 17:00:24 baking libtool-cross Jul 24 17:00:42 (of course it *may* still be picking up -cross form staging or something....) Jul 24 17:05:45 renoo_busy: nope, rebuild of libtool-cross didn't help Jul 24 17:07:33 what does this give you? "grep PB tmp/work/pcre-4.4-r1/pcre-4.4/arm-linux-libtool" Jul 24 17:15:56 sorry, will need a few minutes, build machine is being very slow ATM Jul 24 17:16:12 np Jul 24 17:25:43 renoo_busy: that gives me the same thing: # This breaks install into our staging area. -PB Jul 24 17:26:06 ok. so it's using the patched libtool Jul 24 17:26:40 best reply to the bug with a description of your findings Jul 24 17:27:12 maybe one of our resident libtool gurus (pb_ and kergoth) will have a look at it Jul 24 17:27:45 I've run out of ideas why it might work here but not for you ;) Jul 24 17:31:15 thanks for trying Jul 24 17:31:20 I'm composing a reply right now Jul 24 17:31:36 my fix still does work...at least for now Jul 24 17:31:45 could you test and see if it breaks things on your system? Jul 24 17:32:41 the patch looked rather safe Jul 24 17:32:43 I realize it's brute force and a horrible hack, but it works... Jul 24 17:33:06 but it's not the right thing to do as this might be a systematic error Jul 24 17:33:12 I agree Jul 24 17:33:30 however it would be good to have the hack in there for other people until the real problem is sniffed out Jul 24 17:34:48 hmm. I'll think about it Jul 24 17:34:54 off to bed for now Jul 24 17:34:57 'night all Jul 24 17:52:16 03daka 07org.oe.nslu2-linux * r5bbc6ec4... 10/packages/meta/openslug-native.bb: Add gcc-symlinks, fix the files libpthread.so and libc.so from libc6-dev (that really should be fixed in there) Jul 24 20:12:42 03daka 07org.oe.nslu2-linux * r2eeab934... 10/packages/meta/openslug-native.bb: Add binutils-symlinks, remove the libc.so and libpthread.so fix since things work without it Jul 24 20:12:46 03daka 07org.oe.nslu2-linux * r27ad08fd... 10/packages/busybox/busybox_1.00.bb: Jul 24 20:12:46 Check for existance of 'test' before assuming it isn't there and alias test to 'busybox test'. Jul 24 20:12:46 When building a rootfs the previos implementation cased a few serious (non-fatal) errors showing up with tinderbox. Jul 24 23:12:12 hi all Jul 24 23:22:08 morning Jul 24 23:44:01 morning Jul 24 23:44:53 anyone got wellenrieter to work on a 3100+dlink card? Jul 24 23:47:31 well, it works w/ OZ 3.5.3 on a SL-C1000 (which is close enough to a 3100) Jul 24 23:47:46 not w/ a dlink card though ;) Jul 24 23:48:08 hmmm, maybe my network settings.. **** ENDING LOGGING AT Sun Jul 24 23:59:56 2005