**** BEGIN LOGGING AT Wed Mar 01 10:59:56 2006 Mar 01 11:00:39 njs: did the rosterify work out? Mar 01 11:01:18 koen: I tried it with both 0.24 and 0.26pre-whatever -- with 0.24, it looks like you got lucky, and not only did it pick a really stupid base, but running with --lca it hit an old bug in the lca algorithm! and still picked a bad base. Mar 01 11:02:35 and with 0.26pre? Mar 01 11:02:36 koen: however, when I pulled up monotone-viz to see what the right LCA was, and told monotone to use that directly with explicit_merge, then the actual merge did become trivial -- just 1 conflict, some kernel config file that had a timestamp in a comment at the top, and the timestamp had changed. Mar 01 11:02:40 koen: with 0.26, it gave me that same conflict directly, without any fiddling around or anything. it also ran about 5 times faster, just to mock its elder and less able ancestor, or something :-) Mar 01 11:02:46 koen: err... it just went. I didn't time it or anything. Mar 01 11:02:48 but no problems. Mar 01 11:03:22 that's pretty good news Mar 01 11:06:10 is your last question there about whether I was using mainline monotone or not? if so, the answer is "yes" Mar 01 11:07:07 do you have a static version of 0.26pre which I can use to do some tests? Mar 01 11:07:41 njs: That sounds quite positive. Thanks for checking that out :) Mar 01 11:08:18 hmm, how static do you need? is dependency on libstdc++6 okay? Mar 01 11:08:40 that would probably okay Mar 01 11:08:46 debain/stable over here Mar 01 11:10:41 -rw-r--r-- 1 root root 874164 2005-10-02 07:29 /usr/lib/libstdc++.so.6.0.6 Mar 01 11:11:17 RP: np. it was a bit of a trip down memory lane actually, I'd managed to supress the memories of most such workaround over the last few months... Mar 01 11:11:17 RP: was helpful to understand what the problem really was, too; I'd have suggested the explicit_merge trick, but I didn't realize it was applicable. Mar 01 11:11:42 njs: and if possible a copy of the rosterified db :) Mar 01 11:12:10 koen: gimme a sec for the linker to run :-) Mar 01 11:12:14 :) Mar 01 11:12:49 ibot: botmail for zecke: dominion will serve a 0.26 db soon(TM) Mar 01 11:13:43 RP: you wouldn't have any updates gtkmm/glibmm packages lying around? Mar 01 11:13:51 koen: hmm, it has The Revisions That Must Not Be Named in it... Mar 01 11:14:12 koen: I guess you can kill them off easily enough, though. Mar 01 11:14:43 koen: no, sorry Mar 01 11:15:16 hmm, I guess people have been, like, improving monotone or something, apparently make wants to rebuild lots of files. will be a minute... Mar 01 11:15:37 njs: I have all day :) Mar 01 11:18:59 oh, and small warning, I doubt this applies to you guys, but we _just_ (like, a hour ago) tightened up monotone's checking of filename charsets. so it's possible if you have weird non-ascii, non-utf8 filenames checked in (which should be impossible, but we haven't had time to make sure it hasn't happened yet :-)), then it might fail. Mar 01 11:19:14 koen: http://frances.vorpus.org/~njs/OE-bad-rosterified.db.bz2 Mar 01 11:19:38 wgetting Mar 01 11:20:07 koen: http://frances.vorpus.org/~njs/monotone-0.26pre-latest.bz2 Mar 01 11:20:29 thanks Mar 01 11:20:50 OE-bad-rosterified is RP's OE db, from before he removed the revs that were killed Mar 01 11:20:50 then run through rosterify Mar 01 11:21:19 initially I want to compare first pull timings Mar 01 11:21:34 after that it'd be testing to see if we can get it to break Mar 01 11:22:06 hrw|work: ping Mar 01 11:23:50 CoreDump|work: pong Mar 01 11:24:29 koen: nod Mar 01 11:24:38 koen: please break it! I am being scared by the lack of bugs found so far :-( Mar 01 11:24:38 koen: and you guys are really good at breaking things, it seems ;-) Mar 01 11:25:17 haha, phear my push of death =) Mar 01 11:25:29 CoreDump|work: altboot 1.0? Mar 01 11:25:29 koen: and the first pull timings are still being worked on in the next phase of work, as I guess you know from #monotone. Mar 01 11:25:49 njs: yes, and I read the mailinglist as well Mar 01 11:26:16 njs: I used 'delta storage format' a lot last weekend in converstations ;) Mar 01 11:26:20 raduga: its coming ;) I just read your update on OZ.org regarding cxxxx machines Mar 01 11:26:31 raduga: s/raduga/hrw Mar 01 11:26:54 what's the problem with kernel 2.4? Mar 01 11:27:24 koen: oh my. Mar 01 11:27:24 koen: I'm sorry. Mar 01 11:27:28 koen: I have to use it a lot in conversations, I don't particularly wish that fate on other people :-) Mar 01 11:27:29 CoreDump|work: apm problems, on/off button does not work correctly iirc etc Mar 01 11:28:09 right, the button is fu*ked. But suspend / resume works fine with test5 AFAICT Mar 01 11:29:03 CoreDump|work: anyway we skip them Mar 01 11:29:03 hrw|work: anyways, was just wondering, gotta go back to work ;) later Mar 01 11:53:53 koen: ping Mar 01 11:54:10 hrw|work: pong Mar 01 11:55:26 koen: angstrom-2006.9.conf contain orinoco-modules_h3x00 = 0.13e - should not it be placed in machine configs rather? Mar 01 11:56:46 hrw|work: probably, but I hate all the whining I get when touching ipaq machine files Mar 01 12:02:27 heh. Mar 01 12:17:45 starting pull with monotone 0.26 Mar 01 12:27:35 20% done :) Mar 01 12:28:08 huh, I guess that's not _so_ bad... Mar 01 12:28:08 after 10 minutes? Mar 01 12:28:11 yes Mar 01 12:28:19 fscking awesome speed Mar 01 12:28:39 800/4599 revs Mar 01 12:28:47 so slightly less than 20% Mar 01 12:29:17 ibot: calc 80 / 6 Mar 01 12:29:18 koen: where from I can pull? Mar 01 12:29:23 it may slow down as you go, that's been seen sometimes... Mar 01 12:29:37 oh well, if you're happy, cool :-) Mar 01 12:29:46 hrw|work: only localhost ewi Mar 01 12:30:23 and we'll keep working on making it _actually_ fast, not just fast if your expectations are already lowered by old versions :-) Mar 01 12:30:27 koen: ah.. cant you serve it? iirc 0.26pre use other port.. Mar 01 12:31:22 hrw|work: note you have to use mainline monotone from this week to pull, or it will get pissed off at the root suture stuff Mar 01 12:31:23 hrw|work: on port 80, 22 and old-monotone are open Mar 01 12:32:04 ah.. ok Mar 01 12:32:09 http://ewi546.ewi.utwente.nl/OE/monotone-0.26pre-latest.bz2 Mar 01 12:32:11 that one :) Mar 01 12:32:14 njs: just fetched your latest one Mar 01 12:32:25 cool Mar 01 12:32:50 hrw|work: if you use ssh port forwarding you could pull Mar 01 12:32:53 port 4691 Mar 01 12:33:49 hmm.. how it goes... Mar 01 12:34:03 njs: you can make it ~60% faster by making it multithreaded Mar 01 12:34:19 njs: one cpu is idling during pull over here Mar 01 12:38:57 koen: that's the cpu you use to play movies while you're waiting for the download to finish Mar 01 12:39:08 :) Mar 01 12:39:12 njs: on headless server? Mar 01 12:39:27 hrw|work: aalib in screen ;) Mar 01 12:39:35 koen: libcaca Mar 01 12:39:47 that remind me that I had to build mplayer with caca Mar 01 12:41:58 monotone: 23.8 M | 634 | 4,816/17,910 | 1,337/4,599 Mar 01 12:42:59 so on one opteron full pull is 1.5-2h Mar 01 12:43:18 looks like it Mar 01 12:43:44 ~lart mplayer cvs for being unbuildable Mar 01 12:43:44 * ibot eats mplayer cvs and falls over dead for being unbuildable Mar 01 12:43:46 roughly 1 rev/second Mar 01 13:14:49 hi all Mar 01 13:19:12 hi Rui Mar 01 13:19:34 I have a little question about monotone. I followed the description in the getting started part of the oe.handhelds.org wiki and got into some trouble. After I set up this line monotone --db=/stuff/oe.db pull ewi546.ewi.utwente.nl "org.openembedded.{dev,dreambox}" . The terminaloutput says this monotone: 41,1 M | 1,9 M | 18.792 | 4.778 | 2.475 . It is now about 24 hours ago that I set up that line. Is this normal? Have I've got jus Mar 01 13:19:53 ~ewi Mar 01 13:19:54 ewi is probably ewi546.ewi.utwente.nl - main backup server for OE monotone. Hosts database snapshot: http://ewi546.ewi.utwente.nl/OE/OE.db.bz2 Mar 01 13:20:03 get snapshot and then pull Mar 01 13:20:35 <_law_> has anyone a working kismet.conf for me (i use hostap modules) Mar 01 13:20:47 <_law_> or only source string Mar 01 13:22:18 lrg: The asoc fix works, thanks. Will push into OE shortly Mar 01 13:22:26 hrw|work: Mar 01 13:22:27 14:24 < CIA-1> * commands.cc (CMD(merge)): don't say "misuse" if it's Mar 01 13:22:27 14:24 < CIA-1> just already merged. Mar 01 13:22:53 koen: noticed Mar 01 13:23:32 grrrr Mar 01 13:23:36 01 14:26 -!- Irssi: #monotone #oe : koen (dominion.kabel.utwente.nl) Mar 01 13:23:48 channels shared Mar 01 13:23:51 some bozo removed the extra warnings from GettinStarted Mar 01 13:24:43 mr_nice: the gettingstarted wiki tells you it will take a long time, so follow the advice given in that warning Mar 01 13:27:42 koen: ok. I thougt 24 h are a long time - how many days takes it normaly? Mar 01 13:27:43 hrw|work: dominion.kabel.utwente.nl is now server 0.26 and 0.26pre Mar 01 13:27:48 ok Mar 01 13:27:56 mr_nice: 8 hours on my build machine Mar 01 13:27:58 RP: cool, I'm now fixing tosa with dapm, etc Mar 01 13:29:47 koen: started pulling Mar 01 13:30:21 koen: could you please give me a guess about the revs written? have it to be the same nuber as the revs in to be compleate? Mar 01 13:31:18 mr_nice: it has to be the same number Mar 01 13:32:00 RP: how much work we need to get 2.6 for oz 3.5.4.1? Mar 01 13:32:55 hrw|work: It shouldn't be that much work - just a lot of changes from .dev Mar 01 13:35:00 koen: thx - thats great so I have a gues how long it will take :) Mar 01 13:35:04 cya all Mar 01 13:37:37 koen: 8 minutes and still 0/4599 Mar 01 13:37:57 koen: but probably first need to fetch more then 11M Mar 01 13:38:25 yes, about 20M Mar 01 13:38:37 your hitting my bandwidth limit :) Mar 01 13:39:56 ;) Mar 01 13:42:40 19.6M Mar 01 13:43:26 upto 23.8 has to be Mar 01 13:43:34 started revs Mar 01 13:43:42 20.6M is all Mar 01 13:48:42 and its about 1rev/s here too (axp1700+) Mar 01 13:57:19 monotone: 36.6 M | 634 | 17,714/17,910 | 4,552/4,599 Mar 01 13:57:21 460revs after 14 minues Mar 01 13:58:39 real 97m29.603s Mar 01 13:58:39 user 97m37.583s Mar 01 13:58:39 sys 1m23.731s Mar 01 13:59:22 the wall clock says it took 1 hour and 37 minutes Mar 01 13:59:41 much faster Mar 01 14:01:17 11 hours vs 1.6 hours Mar 01 14:01:33 but is it faster than other SCMs Mar 01 14:01:47 hi Mar 01 14:01:56 hi Bernardo Mar 01 14:02:04 if so monotone stays , better the devil you know and all Mar 01 14:03:18 it's now actually faster as svk in my tests Mar 01 14:03:35 which is an unfair test, since the svk server wasn't on localhost Mar 01 14:22:36 1055revs Mar 01 14:29:33 hi, anybody can help me take a look at http://pastebin.com/578400 Mar 01 14:29:50 i faild to 'fetch bootstrap-image' in bitabke -i mode Mar 01 14:33:46 ~lart handhelds-sa maintainer Mar 01 14:33:46 * ibot slaps handhelds-sa maintainer upside and over the head with one freakishly huge killer whale named hugh Mar 01 14:33:59 minipanda: looks like handhelds-sa is fscked a bit Mar 01 14:35:45 hrw|work: looks more like a fscked cache/too old bitbake Mar 01 14:36:34 NOTE: bb.build.FuncFailed:("can't read from file '%s' (%s)", ('/home/hzhang/netspectrum/wrt/20060228/build/tmp/staging/arm-linux/kernel/kernel-abiversion', )) while evaluating: Mar 01 14:36:48 tmp/staging/arm-linux/kernel/kernel-abiversion is only AFTER building of kernel Mar 01 14:37:08 so recipe cannot use ${KERNEL_VERSION} Mar 01 14:39:37 em .... Mar 01 14:40:24 minipanda: first update bitbake to current version Mar 01 14:40:51 hrw|work: ok, svn up now Mar 01 14:42:42 03xora 07org.oe.dev * rd3623b9b... 10/packages/tinymail/ (files/gtk-doc.m4 files/gtk-doc.make tinymail_svn.bb): tinymail_svn.bb : add gtk doc stuff for _law_ Mar 01 14:42:46 03rpurdie 07org.oe.dev * rc231cb3f... 10/packages/linux/linux-openzaurus_2.6.15.bb: linux-oz-2.6: Update to ASoC 0.10rc3 and tweak the headphone switch patch Mar 01 14:49:25 03koen 07org.oe.dev * r4d485370... 10/packages/dia/dia_0.94+0.95pre3.bb: dia: add 0.95-pre3 Mar 01 14:49:52 XorA: what's the trick you use to boot from SD? Mar 01 14:50:39 koen: just change root=/dev/mmcblk0p1 Mar 01 14:51:59 1600revs Mar 01 14:52:03 XorA: thanks Mar 01 15:00:33 morning Mar 01 15:06:57 * minipanda 's bitbake update to revisiton 383 Mar 01 15:14:25 Does anyone here have editing access to the OZ site? I can't remember my password and the lost password feature appears to be broken/slow... Mar 01 15:23:51 03koen 07org.oe.dev * rdfe9588c... 10/packages/libgpewidget/libgpewidget_0.110.bb: libgpewidget: update to 0.110 Mar 01 15:29:43 Cwiiis: broken - sf.net does not allow to send mails from project sites Mar 01 15:30:05 hi ho Mar 01 15:30:08 Cwiiis: mail me from your address used at oz website and I will set pass for you Mar 01 15:30:13 mickeyl|office: ho hi Mar 01 15:30:25 hrw|work: Thanks - what's your address? Mar 01 15:30:51 use irc@hrw.one.pl Mar 01 15:32:13 hrw|work: Sent Mar 01 15:32:14 hrw|work: do you think you can make it for next years FOSDEM? It would be great to have you there as well. We plan to do two OE presentations and a combined OE/OZ/Familiar/Opie stand Mar 01 15:32:26 GPE as well, of course Mar 01 15:32:54 mickeyl|office: I have fosdem 2007 in plans Mar 01 15:33:00 hrw|work: excellent! Mar 01 15:33:12 1h50m with monotone 0.26-pre = 2500 revs Mar 01 15:33:20 hey mickeyl|office Mar 01 15:33:43 hi CoreDump|home Mar 01 15:34:11 hello hrw|work hi all Mar 01 15:34:25 hey CoreDump|home Mar 01 15:35:41 hi koen, hi CoreDump|home Mar 01 15:35:54 CoreDump|home: how about you and FOSDEM 2007 ? Mar 01 15:36:12 mickeyl|office: if I can get the day off, why not =) Mar 01 15:36:26 I hope you had a good time there? Mar 01 15:37:25 hrw|work: alright, since svn up, bitbake -i, fetch bootstrap-image done :-) Mar 01 15:37:35 hrw|work: I'll so the last of the altboot testing on collie today. Should be ready by the end of the day Mar 01 15:37:42 CoreDump|home: great Mar 01 15:38:37 CoreDump|home: Is it easy to use your collie SD package stuff btw? Mar 01 15:38:51 Cwiiis: dead simple Mar 01 15:39:01 I'll have to check it out this week then :) Mar 01 15:39:05 ;) Mar 01 15:39:21 Cwiiis: and there is no other root-on-sd scripts/hacks anymore Mar 01 15:39:46 Is ipkg-link still being used? Mar 01 15:40:26 <[lala]> RP: starting devmand: /etc/zaurusd/scripts/zaurus-hinge: 37: xrandr not found Mar 01 15:40:39 <[lala]> RP: what's that about? Mar 01 15:41:23 mehhh 61 revs to pull Mar 01 15:41:31 good news: we will get poodle ;) Mar 01 15:41:42 CoreDump|home: about 1 minute with the new monotone :) Mar 01 15:41:51 CoreDump|home: that'd be cool. FOSDEM is always a great happening. It's not just about the presentations, but also about meeting the guys in person and having contacts to the people who actually use your stuff :) Mar 01 15:41:52 doh Mar 01 15:42:00 mickeyl|office: indeed Mar 01 15:42:01 [lala]: devmand is written for X :) Mar 01 15:42:13 hrw|work: excellent! Mar 01 15:42:17 http://www.oesf.org/forums/index.php?showtopic=17632&view=findpost&p=116531 Mar 01 15:42:18 s/people/suckers hehe Mar 01 15:42:23 mickeyl|office: and seeing mickeyl drink leffe Mar 01 15:42:24 CoreDump|home: heh Mar 01 15:42:30 koen: :D Mar 01 15:42:41 indeed. I have a slight leffe brun consume problem Mar 01 15:42:54 <[lala]> koen: that's what i had in mind as well ... but zaurusd seems to require it Mar 01 15:42:56 especially the days when I'm outside belgium Mar 01 15:43:17 good news: we will get poodle ;) Mar 01 15:43:19 yay Mar 01 15:43:45 I hope that I will not be the one who will get it Mar 01 15:44:23 * CoreDump|home nods Mar 01 15:44:55 it needs to go to someone who has time to hack on it Mar 01 15:45:11 koen: we have someone with time? Mar 01 15:45:20 ehm..... Mar 01 15:45:44 we can do it like with the 6000. everyone can look at a different aspect. RP could check some kernel issues, I can look about wifi cards behaving odd on that model, coredump could check altboot etc. etc. Mar 01 15:46:13 sounds like a plan Mar 01 15:46:24 mickeyl|office: for me tosa is 2nd to be marked as non-maintained machine Mar 01 15:46:56 hrw|work: that came as a little surprise for me. I always thought that tosa was a nice machine Mar 01 15:47:30 CoreDump|home: look at oz-devel and recent tosa fixes.. Mar 01 15:47:37 it has FSCKED keymap since 3.5.3 Mar 01 15:47:46 noone fixed it Mar 01 15:47:48 hrw|work: that will change. there are a lot of tosas out there. when we have 2.6 on the device, our distribution will get more attractive than it's now. Mar 01 15:48:00 it cannot suspend/resume with working usb wifi Mar 01 15:48:04 * lrg is fixing tosa audio and touch atm Mar 01 15:48:05 etc etc etc Mar 01 15:48:26 lrg: yay Mar 01 15:48:27 we maintain c7x0 and collie. Mar 01 15:48:44 cxx00/2.6 are maintained Mar 01 15:48:54 i can do more work on tosa, but not until summer has passed Mar 01 15:49:08 hrw|work: got a question about 3.5.4.1 Mar 01 15:49:12 mickeyl|office: write PhD first Mar 01 15:49:18 CoreDump|home: dont ask to ask but ask Mar 01 15:49:18 hrw|work: *nod* absolutely Mar 01 15:49:21 is it 3.5.4 + 2.6 kernel or a .dev snapshot Mar 01 15:49:26 [lala]: If you're using opie, you can remove the zaurus-hinge script for now - it should be checking if that binary exists before trying to run it. It handles rotating the screen under GPE Mar 01 15:49:29 CoreDump|home: 3.5.4 + 2.6 kernel Mar 01 15:49:41 excellent =) Mar 01 15:49:43 CoreDump|home: and all updates which are needed to get it working good Mar 01 15:49:56 CoreDump|home: .oz354fam083 is for 3.5.4+ releases Mar 01 15:50:02 works damn fine as it is. The kernl has very few bugs Mar 01 15:50:16 CoreDump|home: if soemthing is not in .oz354fam083 then it does not exist in 3.5.4+ Mar 01 15:50:23 hrw|work: ok Mar 01 15:50:56 well the biggest problem with 2.6 is that it changes your jffs2 partitions so that a 2.4 kernel oopses during mount. Mar 01 15:51:09 no kexec to 2.4 =( Mar 01 15:51:24 * CosmicPenguin is going to be *VERY* happy with multimachine, I can just feel it Mar 01 15:51:50 I've been saying that for months Mar 01 15:51:56 ;) Mar 01 15:52:23 koen: dumb question - what is multimachine? Mar 01 15:52:39 * RP also loves multimachine Mar 01 15:52:49 HopsNBarley: it allows you do build for multiple machines and reuse toolchains and stuff Mar 01 15:53:07 mickeyl|office: did you see http://www.kdedevelopers.org/node/1829 ? Mar 01 15:53:11 re: poodle, If I get it for a few weeks, I suspect i can sort out 2.6 on it in sort order. There can't be that much wrong with it... Mar 01 15:53:22 s/sort/short/ Mar 01 15:53:24 RP: yay! Mar 01 15:53:33 ah, guess you can't do that now - though I do parallel openslug and ucslugc builds... Mar 01 15:53:39 sounds great, and i'll love it too! Mar 01 15:54:32 koen: yeah. finally some kind of event interoperability between qt and gtk Mar 01 15:54:38 koen: and the lower layers of course Mar 01 15:55:02 mickeyl|office: and the glib mainloop integration Mar 01 15:55:27 mickeyl|office: it means you can start looking at using dbus for opie/fluffie Mar 01 15:55:53 koen: yeah. Fluffie/Qt4 will happily use dbus Mar 01 15:58:02 for(rok = rok_start, a = 0; rok <= rok_end; rok++, a++) Mar 01 15:58:15 javascript suxx but atleast it work Mar 01 15:59:09 monotone: 30,9 M | 634 | 12033/17910 | 3150/4599 Mar 01 15:59:40 hail pb_ Mar 01 16:02:37 mickeyl|office: http://www.handhelds.org/hypermail/oe/68/6816.html Mar 01 16:02:39 RP: headphone has "right mixer" off, thats a "little" bug :) Mar 01 16:03:06 greentux_alt2: Are you not in headset mode? It worked when I tested it this morning Mar 01 16:04:40 koen: impressive! Mar 01 16:04:47 6 times is better than what I hoped for Mar 01 16:06:44 bbl Mar 01 16:06:49 ~bon appetit Mar 01 16:06:50 extra, extra, read all about it, bon appetit is smacznego. Guten Appetit. Eet Smakelijk. God Appetitt. Buon Appetito. Buen apetito Bom Apetite. buen apetito Mar 01 16:10:40 RP: you are right, it was headphone(!)... only for later finalising Mar 01 16:11:51 Is it possible to compile the collie kernel without the gcc 2.9x toolchain at the moment? Mar 01 16:12:19 Cwiiis: no :-( Mar 01 16:12:27 ah, scratch that then :) Mar 01 16:12:50 ~zauruskernels Mar 01 16:12:51 extra, extra, read all about it, zauruskernels is By default Zaurus use 2.4.18 or 2.4.20 and you need GCC 2.95.3 to compile them - all is described in OE wiki: http://openembedded.org/cgi-bin/moin.cgi/ZaurusKernels There is also work on 2.6: http://openembedded.org/cgi-bin/moin.cgi/ZaurusKernels26 Mar 01 16:14:34 ibot: no, zauruskernels is By default Zaurus used 2.4.18 or 2.4.20 which need GCC 2.95.3 to compile - all is described in OE wiki: http://openembedded.org/cgi-bin/moin.cgi/ZaurusKernels Now we work on moving all models to 2.6 kernels: http://openembedded.org/cgi-bin/moin.cgi/ZaurusKernels26 Mar 01 16:14:35 okay, hrw|work Mar 01 16:20:01 hi Mar 01 16:21:22 * JustinP really hates people sometimes Mar 01 16:23:57 lrg + RP: perhaps you have an idea: after "arecord -c 1 -r 48000 -f S16_LE | aplay" all other apps work very fine! no problem. but without the (possible) init from arecord th emic was down. Mar 01 16:24:31 lrg: what does arecord, what other apps (i think they are using oss emu) not do? Mar 01 16:26:47 greentux_alt2: do you mean the mic does not work without arecord ? Mar 01 16:27:14 lrg: yes, after loading the init states from RP the Pi/Phone app had no mic. Mar 01 16:27:28 lrg: i assume te pi/phone uses oss (/dev/dsp r/w) Mar 01 16:27:41 lrg: perhaps there are a little problem in the oss emulation? Mar 01 16:27:56 greentux_alt2: can you cat /sys/device/platform/soc-audio/dpm Mar 01 16:28:05 hail mickeyl|office Mar 01 16:28:26 RP: suspend worked for you? Mar 01 16:28:55 greentux_alt2: Yes, even with music playing Mar 01 16:29:10 greentux_alt2: same here, np on mainstone Mar 01 16:29:25 RP: in th emoment nothing happens. will look deeper :) Mar 01 16:29:36 03xora 07org.oe.dev * rbcba7ea1... 10/packages/vnc/ (tightvnc/Makefile tightvnc_1.3dev7.bb): tightvnc_1.3dev7.bb : add tightvnc recipe Mar 01 16:30:11 lrg: root@borzoi:~# cat /sys/devices/platform/soc-audio/dpm Mar 01 16:30:12 Mono Out 1: Off Mar 01 16:30:12 Out 3: Off Mar 01 16:30:12 Right PGA Mux: Off Mar 01 16:30:12 Left PGA Mux: On Mar 01 16:30:14 Left ADC: On Mar 01 16:30:16 Right ADC: Off Mar 01 16:30:18 Mic Bias: On Mar 01 16:30:20 Left DAC: Off Mar 01 16:30:22 Right DAC: Off Mar 01 16:30:24 Left Out 1: Off Mar 01 16:30:26 Right Out 1: On Mar 01 16:30:28 Left Out 2: Off Mar 01 16:30:30 Right Out 2: Off Mar 01 16:30:32 Mono Mixer: Off Mar 01 16:30:34 greentux_alt2: In future, use pastebin Mar 01 16:30:34 DAC Mute: On Mar 01 16:30:36 ... Mar 01 16:30:36 WM8750 Clock: Off Mar 01 16:30:38 PM State: D3hot Mar 01 16:30:40 ~pastebin Mar 01 16:30:40 ~pastebin Mar 01 16:30:46 ~flood Mar 01 16:30:51 it has been said that pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.com/, or http://pastebin.ca, or http://channels.debian.net/paste Mar 01 16:30:53 methinks pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.com/, or http://pastebin.ca, or http://channels.debian.net/paste Mar 01 16:30:53 RP: sorry, yes! Mar 01 16:30:54 somebody said flood was spewing loads of output into a channel; *very* rude in most channels and often grounds for banning. If you want to show a lot of output to someone, ask them to join you in #flood and paste the output there. Mar 01 16:32:42 greentux_alt2: dapm looks fine, the ADC is on and the mic is supplied with bias (power) Mar 01 16:33:08 lrg: in headset mic is the "right" channel? Mar 01 16:33:21 * lrg checks Mar 01 16:33:50 greentux_alt2: This was after running arecord so though so everything was working? Is it the same if you run it after booting and switching to headset mode but before you run arecord? Mar 01 16:33:51 lrg: sorry, above do u see the working output... Mar 01 16:34:03 RP: will do that Mar 01 16:34:13 wow.. someone use qtopia under pdaX - inside of qvfb... Mar 01 16:34:50 sounds sick Mar 01 16:34:56 is sick Mar 01 16:35:03 more sick then x/qt Mar 01 16:35:13 =) Mar 01 16:35:45 greentux_alt2: headset mic is left chn Mar 01 16:36:57 after rebooting: http://pastebin.ca/44111 Mar 01 16:37:04 all off, no zaurusd Mar 01 16:37:13 switch_vd runs Mar 01 16:38:15 mode is hodephone. Mar 01 16:39:00 now i switch to headset with alsamixer right? Mar 01 16:39:18 greentux_alt2: looks fine for no active streams, you need to either run aplay or arecord to see the bits powered up. Mar 01 16:39:49 greentux_alt2: Not with alsamixer. Use the script as I detailed in the email Mar 01 16:39:58 RP: ok Mar 01 16:40:26 /etc/zaurusd/scripts/zaurus-mixer-callback Headset Off Mar 01 16:40:41 greentux_alt2: Was just typing that :) Mar 01 16:41:17 http://pastebin.ca/44112 Mar 01 16:41:22 something changed. right? Mar 01 16:42:54 what should i do now? test the voip app or the arecord? Mar 01 16:43:02 greentux_alt2: you need to run your app now Mar 01 16:43:25 lrg: ok start kphone/pi, no dialing... Mar 01 16:43:56 lrg: nothing changed Mar 01 16:44:01 lrg: will try a call Mar 01 16:44:30 lrg: damn! it works! Mar 01 16:45:35 lrg: will try again. (i love that soundchip). Mar 01 16:49:15 lrg: suspend/resume and works... Mar 01 16:56:13 I'm planning to do the xlibs rename/move this evening (x11/x11, xpm/xpm -> xlibs/libx11, xlibs/xpm, etc.). Any objections? Mar 01 16:56:42 nope Mar 01 16:56:53 I do not touch .dev even Mar 01 16:56:58 monotone: 35,9 M | 634 | 16954/17910 | 4365/4599 Mar 01 16:57:11 3h15m Mar 01 16:58:10 s/xpm,/libxpm,/ Mar 01 17:02:25 pH5: fine with me :) Mar 01 17:03:51 hey Mar 01 17:04:03 rwhitby: I have one Unslung question Mar 01 17:05:05 hmm.. we need to make something like http://people.kde.nl/ (The People Behind KDE website) - "The People Behind OE" Mar 01 17:05:22 what do you think? Mar 01 17:05:50 agreed, who can do it? Mar 01 17:06:17 hi zecke Mar 01 17:06:26 zecke: why you always ask trouble questions? Mar 01 17:06:27 :) Mar 01 17:07:26 and then maybe a website for the people behind the website for the people behind oe ?? Mar 01 17:08:29 =) Mar 01 17:09:08 real 219m10.116s Mar 01 17:09:08 user 140m5.715s Mar 01 17:09:08 sys 1m53.158s Mar 01 17:09:18 moreal 219m10.116s Mar 01 17:09:18 user 140m5.715s Mar 01 17:09:18 sys 1m53.158s Mar 01 17:09:21 ops Mar 01 17:09:30 it was: time ~/monotone-0.26pre-latest pull dominion.kabel.utwente.nl org.openembedded.dev --db baza-0.26-pre.db Mar 01 17:10:38 ok. time to go home - cu Mar 01 17:13:18 cu hrw|gone Mar 01 17:16:35 hail zecke Mar 01 17:16:39 zecke: you can test monotone 0.26pre-thismorning now Mar 01 17:17:45 org.oe.dev has two heads Mar 01 17:18:50 not on vanille.de Mar 01 17:20:06 oh, sorry. I guess 601faaded8a6427e29dc8dc43433fe9be662c96c is one of those mysterious to-be-deleted-locally revisions? Mar 01 17:20:14 yes Mar 01 17:22:46 hi, is this the latest one? monotone: already up to date at f076172bf11be4df22001e51898b11d9f9e2050d Mar 01 17:24:15 * CoreDump|home wished the revisions had a better name, like YYYYMMDD-$VERSION_OF_THE_DAY Mar 01 17:25:03 monotone: already up to date at 3a42af814b71596f330d557943e49bdcd8b602d6 Mar 01 17:25:10 CoreDump|home: Submit a patch? Mar 01 17:25:17 nahh Mar 01 17:25:17 CoreDump|home: anything with dates breaks across timezones Mar 01 17:26:02 it was just an idea, anything is better than hashes (if you need to compare them at least) Mar 01 17:26:26 isn't that the purpose of hashes? Mar 01 17:26:30 easier comparision Mar 01 17:26:45 um not if you do the comparsion by hand heh Mar 01 17:27:33 this is taking ages Mar 01 17:27:43 untarring a rootfs from cf to sd on a c700 Mar 01 17:28:04 koen: use altboot for that heh Mar 01 17:28:13 it mounts them async for that ;) Mar 01 17:28:32 I doubt that helps on a c700 :( Mar 01 17:29:14 http://oz.hentges.net/openembedded/altboot/latest-snapshot/c7x0/ <- untested ;) Mar 01 17:31:28 my kernel boots straight from sd :) Mar 01 17:32:45 that's cheating =D Mar 01 17:35:56 the do_package phase of the bake just walks the image/ directory to make a package, correct? Mar 01 17:38:25 hmm that would be install/ or not? Mar 01 17:39:14 do_install: image/ -> install, do_package install/ -> .ipk IIRC Mar 01 17:40:15 hmmmm looks like SD auto-mounting has problems on collie at least Mar 01 17:41:10 i'm trying to figure out how bb knows which files to put in the package. the 2800 examples use various schemes. in some of them there is no FILES defined at all. In others, they are explicitly listed. just looking for some light on the issue. Mar 01 17:41:34 HopsNBarley: There is a default set of FILES_ defined Mar 01 17:42:42 any volunteers for adding libata to OE? Mar 01 17:43:07 Is dbus-native-0.61 known to work? Mar 01 17:43:46 RP: can't find FILES anywhere in bitbake hierarchy... where is it set? Mar 01 17:43:51 03rpurdie 07org.oe.dev * r0abc6467... 10/packages/libtool/libtool-native_1.5.10.bb: libtool-native: Remove unneeded DEPENDS (the original problem this works around is fixed). Mar 01 17:45:20 hrw|gone: sorry for that Mar 01 17:45:26 HopsNBarley: conf/bitbake.conf Mar 01 17:45:35 koen: I have read your email Mar 01 17:46:03 koen: I think there is no easy way to incrementally convert? Mar 01 17:46:20 i see FILESDIR and FILESPATH... i'll read some code and hopefully come back with better questions! thanks Mar 01 17:46:42 HopsNBarley: Note this file is in openembedded, not bitbake Mar 01 17:47:16 got it. Mar 01 17:47:27 zecke: maybe via monotone-dumb Mar 01 17:49:02 zecke: njs also had a solution for vendors with private branches Mar 01 17:50:01 koen: tar.gz install w/ async on collie: <1m for gpe-image Mar 01 17:50:23 collie doesn't have broken cache and ide bugs Mar 01 17:50:41 RP: ahhh.... Mar 01 17:58:37 03pH5 07org.oe.dev * r71d2eba1... 10/packages/ (192 files in 75 dirs): xlibs: move X libraries to a common folder and unify naming to tarball names Mar 01 18:02:44 pH5: rock Mar 01 18:09:51 Crofton: hey Mar 01 18:09:58 hey Mar 01 18:10:14 now I see you I should merge svn again :) Mar 01 18:10:21 well the feedback is quite low, I wonder why Mar 01 18:10:43 I guess everyone is perfectly happy with monotone afterall Mar 01 18:11:08 heh Mar 01 18:11:53 people are lazy Mar 01 18:12:03 I'm sure they'll have lots to say about svn if they were forced to use dit Mar 01 18:12:50 03coredump 07org.oe.dev * r6e871c66... 10/packages/altboot/ (4 files in 4 dirs): altboot: Last changes before the stable relase Mar 01 18:13:23 hey mreimer Mar 01 18:13:31 zecke, monotone IS getting better ... Mar 01 18:13:42 i wish monotone had a '-n' flag, ala make. would be useful for propagate Mar 01 18:13:48 How is the new version of monotone? I saw some speed improvement remarks Mar 01 18:14:11 Is there anything explaining the multiple head problem in terms I can understand? Mar 01 18:14:20 Crofton: initial pull in 1.5 hours Mar 01 18:14:44 as opposed to almost forever :) Mar 01 18:15:13 Crofton, take two cars and drive them headlong into eachother ... theoretically they "could" combine into one car ... usually you end up with a mess Mar 01 18:15:28 hmmm Mar 01 18:15:41 I think I need something slightly more technical :) Mar 01 18:16:18 Crofton: we hit a bug in monotone 0.25 Mar 01 18:16:37 monotone 0.26 handled the multiple head situation without problems Mar 01 18:16:40 so this problem will not be as bad in .26? Mar 01 18:16:41 ok Mar 01 18:16:50 hmm should also work now (since latest kernel update) Mar 01 18:17:11 alsa always doesnt find any soundcard Mar 01 18:18:05 law_: what Z do you have again ? Mar 01 18:18:17 husky c760 Mar 01 18:18:46 RP: is husky a corgi.c ? Mar 01 18:19:01 lrg: yes Mar 01 18:19:18 03pH5 07org.oe.dev * rf16c11d5... 10/packages/ (167 files in 77 dirs): all over the place: adapt DEPENDS to new xlib names Mar 01 18:19:50 * RP still needs to test the c7x0 devices Mar 01 18:20:08 RP: ah, thanks. It looks like an init problem. Mar 01 18:20:26 It's not matching the codec to the machine... Mar 01 18:20:48 should i post a lsmod ? Mar 01 18:21:44 law_: no, RP knows the machine very well. We've made some recent changes to asoc and corgi.c in asoc, so it's possible somethings been missed Mar 01 18:21:58 ah ok Mar 01 18:22:05 law_: I'm building a kernel now and will look into it Mar 01 18:22:15 RP: nice Mar 01 18:22:21 I'll also see if I can get the mic working with it... Mar 01 18:23:08 ~hail RP Mar 01 18:23:10 * ibot bows down to RP and chants, "I'M NOT WORTHY!!" Mar 01 18:51:06 03pH5 07org.oe.dev * r45ff7b0c... 10/packages/ (5 files in 2 dirs): adapt various DEPENDS, part 2: libxdmcp Mar 01 20:17:26 hi to all, i have an issue (dependecy issue i think) Mar 01 20:17:52 hi gremlin[it] Mar 01 20:18:20 i'm have done some changed to oe configs to build a h3600 image with kernel 2.6 Mar 01 20:18:39 recently (in last 10 days about) i got this error Mar 01 20:18:46 | ERROR: Cannot satisfy the following dependencies for task-bootstrap: Mar 01 20:18:46 | kernel-module-mcp-core-2.6 Mar 01 20:18:55 hi all - to take a long story short, which is the correct statement in local.conf to set the TARGET_OS variable to Linux? Here is the long version :Well, I have a little trouble configuring my local.conf file. I've red the file and changed everything to my needs. I've also read, that the default TARGET_OS is Linux so i don't uncommented it. Now I've got a error ERROR: The following variable(s) were not set: TARGET_OS. After this I uncommented the Mar 01 20:19:34 but i have deleted all 'mcp-core' modules from my config files (ad not exist in oe original) ... :( Mar 01 20:19:38 any clue ? Mar 01 20:19:42 hi mhentges :) Mar 01 20:20:03 gremlin[it]: rebuild task-bootstrap Mar 01 20:20:35 i try but i got the error at the end ... Mar 01 20:20:48 mr_nice: I guess you're not using a DISTRO so TARGET_OS isn't being set to anything. It would only "default" if you use a distro Mar 01 20:20:53 koen: looks like our videos are online now - yikes Mar 01 20:22:06 eeks! Mar 01 20:22:17 lrg: fosdem.org or free-electrons? Mar 01 20:22:56 koen: free-electrons, you should have an email to your handhelds address Mar 01 20:23:11 nice Mar 01 20:23:18 * koen suspects the uni mailserver has indigestion Mar 01 20:23:34 koen: I only got mine about 10 mins ago Mar 01 20:24:35 hmmm Mar 01 20:24:51 apparently the only device with the right codecs is my ipaq Mar 01 20:26:22 koen: hi, i've updated to latest OE.db and bitbake ... Mar 01 20:26:25 RP: if I set the DISTRO = "familiar-unstable" then i get this error ERROR: Unable to parse conf/bitbake.conf (conf/local.conf:79: unparsed line: ' DISTRO = "familiar-unstable"') The MACHINE variable is set to MACHINE = "simpad" is the familiar-unstable setting illegal if the MACHINE is set to simpad? Mar 01 20:26:46 koen: and now fetch bootstrap-image is ok, but i still failed to build ... Mar 01 20:27:02 koen: please take a look at http://pastebin.com/578966 Mar 01 20:27:28 koen (or someone else), this is probably a FAQ : - /, but how do I get some random package (e.g. 'atd') into my opie-image build and gpe-image? Maybe by defining EXTRA_IMAGEDEPENDS (used in rootfs_ipk.bbclass)? Mar 01 20:28:02 atd is already in gpe image :) Mar 01 20:28:20 ok, use bash as the example, then :) Mar 01 20:28:30 tnb: the quickest way is to poke at BOOTSTRAP_EXTRA_(R)DEPENDS in ipaq-pxa270.conf Mar 01 20:28:59 tnb: No, you need to add it to the image dependencies. We don't have then but could probably use some EXTRA_OPIE_IMAGE_RDEPENDS and EXTRA_GPE_IMAGE_RDEPENDS people could use... Mar 01 20:29:12 These would work like EXTRA_BOOSTRAP_RDEPENDS Mar 01 20:29:21 minipanda: rm /home/hzhang/netspectrum/wrt/20060228/build/tmp/cache -rf Mar 01 20:29:31 mr_nice: That should work - I'd work out why you get the parse error Mar 01 20:29:53 what are the _append versions of these for? I put the packages there initially Mar 01 20:30:21 tnb: did you doo foo_append = "bar" or foo_append = " bar"? Mar 01 20:30:30 _append needs an extra space Mar 01 20:30:30 tnb: append versions of which variable? Mar 01 20:30:49 and, would it go in RDEPENDS or DEPENDS or both? (their use is still unclear) Mar 01 20:30:56 BOOTSTRAP_EXTRA_RDEPENDS_append = " udev module-init-tools bash" Mar 01 20:30:59 for example Mar 01 20:31:03 tnb: You're using .dev? Mar 01 20:31:27 no, fam083 Mar 01 20:31:56 (oz354fam083) Mar 01 20:32:06 Which file uses an _append on BOOTSTRAP_EXTRA_DEPENDS - that sounds like bad practise... Mar 01 20:32:19 For the branch it will need to be added to both variables Mar 01 20:32:32 This all changed in .dev which just uses RDEPENDS now Mar 01 20:32:52 RP: conf/machine/ipaq-pxa270.conf Mar 01 20:33:24 RP: ok (both), I'll give it a try Mar 01 20:34:10 tnb: Note that RDEPENDS takes a package name and DEPENDS takes a bitbake target name which can be different Mar 01 20:34:19 koen: well, it failed again, even after i removed whole /build/tmp Mar 01 20:34:34 RP: ah, that could be my problem Mar 01 20:34:39 koen: and it's still the same error like my last post Mar 01 20:35:11 rwhitby: can more than two USB disks be hooked to the NSLU when using unslung? Mar 01 20:35:50 koen: That append line is confusing in ipaq-pxa270.conf. If there's a reason, I can't see it :) Mar 01 20:36:43 zecke: I have some bitbake updates. I've rewritten some of the git fetcher and enhanced the mirrors handling code to accept multiple src mirrors. Just testing the changes now... Mar 01 20:37:26 RP: Well. Get back to real work! Mar 01 20:37:46 RP: if you do not listen you could move the 'command runner' to the fetcher module Mar 01 20:37:51 btw, why do so many machine conf files append module-init-tools to BOOTSTRAP_EXTRA_RDEPENDS when it is pulled by task-bootstrap.bb already? Mar 01 20:38:06 RP: and make the premirror and tarballstash use any other SCM system :) Mar 01 20:38:13 RP: use cvs to get the git tarball :) Mar 01 20:38:54 pH5: historical reasons, I suppose Mar 01 20:39:05 task-bootstrap didn't use to do that Mar 01 20:39:32 pH5: That sounds like something we can cleanup Mar 01 20:40:01 zecke: I suggested moving the "command runner" to the fetcher module a while back - I thought you had concerns about it though? Mar 01 20:40:32 RP: ah, just general concern when using os.system Mar 01 20:40:52 RP: I agree though that bitbake should set exported variables when running commands Mar 01 20:41:43 zecke: so we agree having it wrapped in one function in the fetcher module would be safer than the mutliple os.system calls in each fetcher module as at the moment? Mar 01 20:42:16 I'll move it, we can then worry about making it more secure... Mar 01 20:43:52 koen .. do u have far distance ESP debugging ability ?!?!?! i rebuild again and again, clening and not ... but nothing change ... after u said "rebuild" start to work without errors !!! :) :) :) Mar 01 20:44:19 my crystal ball came back from repairs ;) Mar 01 20:44:33 hihihi :) Mar 01 20:45:14 koe wgat is the irc chan for cirrus 93xx ??? Mar 01 20:45:24 s/koe/koen/ Mar 01 20:45:49 gremlin[it]: #ep93xx Mar 01 20:45:57 thanks Mar 01 20:46:01 RP: right Mar 01 20:56:44 RP: after I add a package to DEPENDS/RDEPENDS how much do I need to rebuild? Mar 01 20:57:03 zecke: I like the idea of the stash using all available SCMs. I suspect we can get into nasty loops doing that though :) Mar 01 20:57:16 tnb: Probably task-bootstrap Mar 01 20:57:24 a 'build opie-image' build the dependency, but did not include it in the base tarball Mar 01 20:57:33 ok, thanks Mar 01 20:57:52 RP: I would be happy to use scp instead of http Mar 01 20:58:06 * Philippe is back (gone 22:35:12) Mar 01 20:59:13 hi Philippe Mar 01 20:59:20 zecke: Indeed. I just tried multiple stash urls including a file:// and an http://. The file:// was unsupported :-/ Mar 01 20:59:33 zecke: Multiple stash sources is supported though Mar 01 21:00:06 I'm busy implementing a couple of Fourier Transformations and I think I'm too dump Mar 01 21:00:09 dumb even Mar 01 21:00:10 if I want to update libxv_2.2.2 to the one from X11R7.0 (libxv_1.0.1) - until bitbake gets epoch support, is it okay to set the version to libxv_X11R7.0-1.0.1 ? Mar 01 21:00:51 RP: bingo! that did it, thanks! Mar 01 21:01:54 pH5: That will just break when we do get epoch support :-/ Mar 01 21:03:07 pH5: Set the version to the correct string including the epoch and people with problems will just have to set a PREFERRED_VERSION... Mar 01 21:03:15 pH5: That way ipkg can still do its job Mar 01 21:03:52 Is it correct that packages with an epoch set have colons in their file names? Mar 01 21:04:46 pH5: I guess so. You could omit the colon part and just set PV in the file to include that bit I guess... Mar 01 21:05:03 RP: how will it break? once there is epoch support, the next version will be 1:1.0.1, which should sort after X11R7.0-1.0.1. Mar 01 21:07:43 pH5: Read the version policies page on the wiki and the debian link. It will compare X with "" and conclude X sorts higher Mar 01 21:08:23 pH5: I think anyway - try ipkg-compare-versions to be sure :) Mar 01 21:08:38 RP: thanks, I will :) Mar 01 21:08:59 zecke: with Unslung 6.7 (in alpha testing now), you can have as many disks behind a hub as you like. Mar 01 21:09:00 pH5: hang on, you're right, For some reason I was thinking there was a colon in both expressions :-/ Mar 01 21:09:15 hey pH5 Mar 01 21:09:20 * RP evidently can't multitask :-/ Mar 01 21:09:24 zecke: but I expect you would want to run OpenSlug or DebianSlug, being a linux firmware developer yourself. Mar 01 21:11:14 rwhitby: right, so it is time to get a NSLU2 from my paypal account Mar 01 21:11:23 any other ideas on how to spend money useful? Mar 01 21:14:47 well I'm sure I could send it to pb_ Mar 01 21:15:32 I guess you could, but that wouldn't be terribly useful. Mar 01 21:15:47 better to spend it on buying leffe for mickey. Mar 01 21:16:51 or scotch for kergoth, or something. Mar 01 21:17:27 * kergoth chuckles Mar 01 21:20:23 zecke: it's always a good time to get an NSLU2 :-) Unless you can find an NAS100d or DSMG600 instead ;-) Mar 01 21:24:13 rwhitby: no, a ixp2800 :) Mar 01 21:24:38 heh Mar 01 21:25:33 imagine a slug with 10x1.4GHz microengines Mar 01 21:29:27 rp: I just want to inform you, that it now works :). I had to set both the DISTRO and the TARGET_OS variable. cya - and many thanks for such an great community :). This irc channel is very frendly to new users! Mar 01 21:33:03 koen|tv: i am not a germ, no, i am italy :) Mar 01 21:36:13 koen|tv: videos have now been mentioned on lwn.... Mar 01 21:37:09 lrg: end of day i want to say tnx to you and rp for the support @afternoon. all works perfect. good work! Mar 01 21:37:45 greentux: np Mar 01 21:38:48 lrg: watch from time to time: http://www.linux-kongress.org/2006/ Mar 01 21:40:12 gcc 4.1 is out Mar 01 21:40:22 which should have eabo support Mar 01 21:40:39 whats eabo? Mar 01 21:42:10 ooops Mar 01 21:42:12 eabi that is Mar 01 21:42:30 yay Mar 01 21:42:59 no matching glibc though Mar 01 21:47:56 koen|tv: what is missing? Mar 01 21:53:31 night all Mar 01 21:57:22 http://cgi.ebay.de/miniLAMP-Linux-Distribution-fuer-Linksys-NSLU2_W0QQitemZ5871702247QQcategoryZ26808QQrdZ1QQcmdZViewItem Mar 01 21:57:25 lol Mar 01 21:57:38 rwhitby: just for some bucks you can buy Unslung on a USB stick Mar 01 21:58:21 zecke as usual Mar 01 22:00:20 pH5: a glibc release with matching eabi bits Mar 01 22:02:51 koen: glibc 2.3.6 doesn't include those bits? Mar 01 22:03:50 pH5: afaik it doesn't Mar 01 22:06:51 pH5: the gcc 4.1 from http://ewi546.ewi.utwente.nl/tmp/eabi/ needs an updated ldflags.patch Mar 01 22:06:54 hi there. Mar 01 22:07:05 hey ArthurMaciel Mar 01 22:07:14 are you, Ken Koen? Mar 01 22:07:34 no, I'm not Ken Mar 01 22:08:01 ArthurMaciel: he is Klark Kent Mar 01 22:08:15 Oh, that's my nick name between my friends. Mar 01 22:09:47 zecke: and you are Bruce Wayne? Mar 01 22:09:57 guys, I'm trying to build an h2200 image, but it fails when trying to compile gettext. Mar 01 22:10:09 Is there any site where I can found corrections for these errors? Mar 01 22:12:21 koen: how did you find out Mar 01 22:13:29 who is Alfred here? Mar 01 22:15:07 dunno haven't seen him in a while Mar 01 22:16:12 03koen 07org.oe.dev * r2968a65d... 10/conf/distro/angstrom.conf: angstrom: prefer softfloat till we get proper eabi Mar 01 22:16:53 zecke: I used my spidersense Mar 01 22:17:08 lol Mar 01 22:17:26 guys, any idea on gettext errors? Mar 01 22:17:32 koen: prefer softfloat to what? Mar 01 22:17:58 mreimer: hardfloat Mar 01 22:18:24 it now uses that hidious softfpa Mar 01 22:18:46 koen: so when we get eabi, each binary will get tagged with what kind of fp it uses? Mar 01 22:18:58 ArthurMaciel: gettext errors in what sense? How to use gettext? compile it? Mar 01 22:19:04 mreimer: yes, and you can mix and match them at will Mar 01 22:19:18 ah ok Mar 01 22:19:31 and stuff like iwmmx works Mar 01 22:19:42 yes, while compiling it. Mar 01 22:19:48 koen: in userland, or the kernel, or both? Mar 01 22:19:54 mreimer: both Mar 01 22:19:57 nice Mar 01 22:21:08 koen aehm that means you can use soft when you think its better than the hardfloat support? Mar 01 22:21:27 woglinde: yes, without apps suddenly crashing Mar 01 22:21:57 koen but then cpu hasnt fpu than there is nothing new Mar 01 22:22:20 then = when Mar 01 22:32:43 please, could anyone take a look at http://rafb.net/paste/results/D17yuU86.html ? Mar 01 22:37:41 is there a place to browse the metadata online? Mar 01 22:38:56 * Philippe is away: visual contact - melancholic dreams Mar 01 22:41:16 * pH5 goes to dream of an eabi support Mar 01 22:42:09 mickeyl|office: ping Mar 01 22:42:16 mickeyl|office: bug 658 Mar 01 22:46:41 Zero_Chaos: yes Mar 01 22:46:48 Zero_Chaos: viewmtn on ewi Mar 01 22:47:19 zecke: url? Mar 01 22:48:30 monotone.vanille.de? Mar 01 22:49:01 ahh, thanks Mar 01 22:49:12 it has a link to viewmtn Mar 01 22:49:49 An error has occurred Mar 01 23:17:25 zecke: Do you think its reasonable to have the try_mirror function check for the file in DL_DIR first (as per http://www.rpsys.net/openzaurus/temp/bitbake_stash-r0.patch) ? Mar 01 23:31:15 let me take a look Mar 01 23:32:42 RP: Isn't it checked somewhere else already? Mar 01 23:33:05 RP: also what about the md5 file? Mar 01 23:53:01 zecke: The way the git fetcher uses try_mirror, the check is needed and isn't covered elsewhere. The md5 file is a good question Mar 01 23:54:03 Well, its needed in that it makes things more efficient rather than having wget try and download the file which would fail if you were say offline Mar 02 00:12:52 RP: that check there can't hurt. But I have the feeling that I already check in tryMirror Mar 02 00:15:05 can anyone help me with http://rafb.net/paste/results/hjW22N49.html when trying to run bitbake nano for a h2200 machine with familiar-0.8.3 distro and target linux Mar 02 00:16:15 zecke: The only checks I can find are within the wget fetcher... Mar 02 00:17:34 ArthurMaciel: try my local.conf: http://sermons.desiringGod.org/~mreimer/local.conf Mar 02 00:19:57 are you religious? Mar 02 00:23:01 ArthurMaciel: everybody has a "religion"; depends on what you mean :-) I'm a Christian Mar 02 00:23:36 I see :-). I found that url very nice. Mar 02 00:23:51 In a good way... not in a funny one. Mar 02 00:23:57 good :-) Mar 02 00:24:38 I really should be Christian... with all those virtues to support the mental torture I may try to place upon you ;-) Mar 02 00:24:45 need to go to bed now :) Mar 02 00:24:50 see you. Mar 02 00:24:51 night zecke Mar 02 00:25:19 ArthurMaciel: there are better reasons than that :-) Mar 02 00:25:33 I really know. Just kidding now... Mar 02 00:25:46 I know :-) Mar 02 00:28:57 matt, do you work with development? Mar 02 00:29:09 ArthurMaciel: what kind? kernel? application? Mar 02 00:29:10 I mean, do you profit with it? Mar 02 00:29:42 I'm a programmer/sys/netadmin for a hosting company, and I've done a bit of linux handhelds kernel programming for pay Mar 02 00:29:45 I don't know... just asking your profession, if you don't mind. Mar 02 00:29:55 oh, interesting... Mar 02 00:30:36 But do you have a preview of how many time you'll take to develop a kernel for a embedded device? This h2200 project took about 2 years to get at a stable point... Mar 02 00:31:33 yes, this one took a lot of time because the only people who were working on it were people who knew little about the whole porting process, like me. Another port would take a lot less time Mar 02 00:32:16 but you've said you've developed kernel for handhelds... Mar 02 00:32:54 who are the guys that know here (h2200)? You, Openadcker, Koen, Joshua, who else? Mar 02 00:32:56 that's just been in the last few months, after I learned a lot porting h2200 Mar 02 00:33:04 oh, I see.. Mar 02 00:33:52 others have helped with building kernels and GPE, others with testing, but the above list plus Szabolcs Gyurko pretty much covers it Mar 02 00:34:54 I remembered him, but didn't know how to write :-) Mar 02 00:35:28 I'm a medicine student and I really would like to "mix" medicine and computing... Mar 02 00:36:03 don't know... maybe improving machines... maybe with research and databases... maybe with bioinfo algorithms Mar 02 00:36:16 That's why I think this list so interesting... Mar 02 00:36:56 do it! it is very interesting. Mar 02 00:36:59 You can make a whole specific totally-configurable kernel for a tiny device... Imagine if what engineers can do for the medicine... Mar 02 00:37:15 no if * Mar 02 00:37:22 could* Mar 02 00:37:38 English gets confused when I start dreaming about it :-) Mar 02 00:37:47 :-) Mar 02 00:37:48 about the future of it... Mar 02 00:40:43 man, bitbake takes time to handle its files, huh? There's a long time since it started from 0% Mar 02 00:42:43 ArthurMaciel: tip: use "bitbake -i", then 'parse', then 'build gpe-image' Mar 02 00:42:55 you can issue multiple build commands, cancel, etc. Mar 02 00:43:54 oh, okay... I'll use it. Could you tell me if it is possible to set the ipaq to mount the CF every time it is inserted? Mar 02 00:48:42 I would like also to click on an html file on the file browser and be redirected to the gpe-mini-browser... Mar 02 00:50:01 isn't there any mime-type definitions here? Mar 02 00:52:48 ArthurMaciel: auto-mounting of CF should already be happening, if udev is configured correctly. I haven't tried that Mar 02 00:53:00 ArthurMaciel: I don't know. that would be a good question for #gpe Mar 02 00:53:09 well, I had to mount /dev/hda1 on /mnt/cf Mar 02 00:53:15 to see its contents Mar 02 00:53:23 wrong channel... sorry. Mar 02 01:13:36 matt, same error with your local.conf file... http://rafb.net/paste/results/4J8QSk44.html Mar 02 01:15:18 ArthurMaciel: are you using org.openembedded.dev or .oz354fam083? my local.conf is for .oz354fam083 Mar 02 01:16:29 oz354fam083 Mar 02 01:16:48 this error I had when I used to play with oe many months agor.. Mar 02 01:16:50 ago* Mar 02 01:17:26 I've deleted oe completely and reinstalled it two days ago, since linux booted here and intend to make a gtkmm package in order to develop for ipaq.. Mar 02 02:31:46 * jnc arrives home from work Mar 02 02:31:57 hi, folks. I cann't download the bluez package from sourceforge.net. and the oesources.org can't providded bluez-2.25 package. Mar 02 02:32:15 what shall I do ? Mar 02 02:32:40 panic. Mar 02 02:40:28 good night to all. Mar 02 02:41:11 ArthurMaciel, it is morning here .. :) Mar 02 02:41:25 what a interesting world, huh? Mar 02 02:42:33 can anyone give me an answer to a python question? Mar 02 02:43:05 maybe #python? Mar 02 02:43:35 maybe rtfm :) Mar 02 02:43:42 lol Mar 02 02:44:04 I don't know python, yet (I hope - if it is not difficult) Mar 02 02:45:28 I think I have the solution Mar 02 02:45:42 I needed to figure out how pyc files are created Mar 02 02:46:29 oh, would you like to tell us? Mar 02 02:59:33 just import them Mar 02 02:59:53 the are made automatically, if you have write perms in the directory Mar 02 04:32:18 03mwester 07org.oe.dev * r863394fb... 10/packages/nslu2-binary-only/ (3 files in 2 dirs): unslung-rootfs: rc.sysinit fix - clean .pid files in /var/run Mar 02 08:46:32 morning Mar 02 08:46:46 molrning Mar 02 08:46:56 morning all Mar 02 08:47:04 <_law_> morning all Mar 02 08:47:54 good morning all Mar 02 08:48:10 hey koen Mar 02 08:55:28 morning Mar 02 09:01:35 lrg: does your fosdem presentation is available? Mar 02 09:02:11 hrw|work: http://free-electrons.com/community/videos/conferences/en Mar 02 09:02:33 thx Mar 02 09:03:39 Philippe just sent a mail asking us to post our talks, so they should be on the fosdem website soon Mar 02 09:03:47 * lrg and koen have their 15 minutes of fame Mar 02 09:04:52 yes! Mar 02 09:05:10 lrg: actually you have more fame, you're mentioned in the Jeff Waugh keynote Mar 02 09:05:47 * lrg hides Mar 02 09:06:50 someone was drunk on bug tracker yesterday then :-) Mar 02 09:09:21 speaking about presentations... need to find time, read already made OE presentations, create own one in Polish and go to Polish linux confs.. Mar 02 09:09:25 but this need time ;( Mar 02 09:09:31 lrg: hi, do you have a doc about the the mixer? what do which mixer and so on? Mar 02 09:10:35 greentux: not yet, but for borzoi out1 is hp vol, out 2 is spk vol Mar 02 09:11:09 greentux: I'm finishing stuff for Dirk on Tosa now, and I'll be starting alsa-lib changes after that Mar 02 09:11:27 lrg: ok, i want to change to things, perhaps you know the magic knob: while headset i hear me slef (thats good, but its to much) Mar 02 09:12:09 lrg: the second is, that if i speak and i make a break the agc increase the noise ... Mar 02 09:14:08 greentux: that's called sidetone. you can lower it with the bypass volumes Mar 02 09:14:42 morning all Mar 02 09:14:45 lrg: ok tried something with "bypass"... sidetone for the forst or the second problem? Mar 02 09:14:47 morning RP Mar 02 09:15:28 greentux: yiu may also want to tweak your alc settings for your particular headset. It may be worth downloading the wm8750 datasheet so you can find the best values. These values could then be used in the config file Mar 02 09:15:34 hey Richard Mar 02 09:15:48 greentux: bypass is first problem Mar 02 09:15:50 lrg: alc = autom line control? Mar 02 09:16:12 greentux: automatic level control - datasheet has good description Mar 02 09:16:18 lrg: the datasheet is available at your website? Mar 02 09:16:21 yes Mar 02 09:18:49 lrg: 62 pages :) Mar 02 09:20:12 lrg: and a complex power management mode... fine Mar 02 09:25:55 greentux: i think the alc only has 2 pages Mar 02 09:27:23 lrg: yes but the whole soundchip is very complex. i didnt expected :) Mar 02 09:49:57 <_law_> RP, could you find some time for alsa-testing on husky ? Mar 02 09:54:32 _law_: I have the kernel - just need a minute to flash, test and debug. Last night didn't really work out time wise :-/ Mar 02 09:54:55 someone really should invent a way to debug in your sleep Mar 02 09:55:00 btw: if latest asoc use machine name instead of codec name then what it use on c7x0 machines? Mar 02 09:55:59 I'm trying to learn my way around oe, but I cannot get the glibc to compile with MACHINE="x86" and DISTRO="familiar-unstable". Should it work? I first tried it with the revision "jan2006prebreakage" and it failed I also tried with the database I pulled yesterday. Mar 02 09:56:14 Missed one dot there :) Mar 02 09:57:16 hrw|work: iirc it will use corgi Mar 02 09:57:40 hrw|work: All the c7x0 will be known as "Corgi" Mar 02 09:57:48 (for sound purposes) Mar 02 09:57:59 ok Mar 02 09:59:43 <_law_> so perhaps i have do load snd_soc_corgi ? Mar 02 09:59:51 <_law_> and remove snd_soc_wm8731 Mar 02 09:59:59 _law_: That should be loaded. You also need wm8731 Mar 02 10:04:14 btw2: I'm moving with hosting today so in next few days I can have problems with mails Mar 02 10:06:26 * koen cleans up after teh xlibs changes Mar 02 10:12:43 Helpdesk seems to be busy. Mar 02 10:14:00 Cockroach-: I think only epia x86 is actively supported Mar 02 10:18:21 hmm I see. Mar 02 10:19:01 Cockroach-: you should be able to copy the setting you need from epia.conf to local.conf Mar 02 10:19:43 You see I'm trying to figure out the best process to develope software with x86 PC, of cource compiling against the same version libraries that exist on the actual device (h3600) Mar 02 10:20:02 XorA: I will look into that Mar 02 10:20:04 Cockroach-: machine=h3600 then Mar 02 10:20:07 Cockroach-: bitbake meta-sdk Mar 02 10:20:39 hrw|work: That creates arm-binaries right? I cannot link x86 software (which I'm doing) against those Mar 02 10:21:11 as said, I'm just trying to get things to work, 'bitbake nano' didn't go throuch with MACHINE="x86" since glibc doesn't compile Mar 02 10:21:54 Failed to open /dev/rtc: Device or resource busy (it should be readable by the user.) Mar 02 10:22:12 zaurus 2.6.14-rc1 does not have /dev/rtc at all? (we have it in /dev) Mar 02 10:22:40 Cockroach-: I think you need the GLIBC_* stuff from epia.conf, but its not a build I have ever done Mar 02 10:23:16 Cockroach-: MACHINE="epia" is the only x86 build I've been able to make work. And I can only get it to work with gcc 3.4.4 (not with 4.x) and I needed to disable jgc. Mar 02 10:23:35 How do you guys develope new software from scratch then? With ARM-proto boards and/or directly on IPAQ:s? Mar 02 10:24:14 I don't know if my approach is the right one. Mar 02 10:26:56 * koen mentions meta-sdk once more Mar 02 10:27:11 koen: doing it as we speak =) Mar 02 10:27:41 ~lart e17 Mar 02 10:27:41 * ibot takes out a seltzer bottle and sprays e17 in the face. You know, one of those old-school seltzer bottles clowns have? Yeah those. Anyway, consider yourself spritzed Mar 02 10:28:55 03koen 07org.oe.dev * rd95b1f7d... 10/packages/gnome/libwnck_2.12.1.bb: libwnck: add dependency on gtk+ Mar 02 10:30:10 03koen 07org.oe.dev * r3b1583d9... 10/conf/distro/angstrom-2006.9.conf: angstrom-2006.9: move some preferred providers around so angstrom can override stuff in includes like e17 defining a gcc version (WTF?) Mar 02 10:36:26 koen: e17 stuff use cpp on device Mar 02 10:37:40 03frederic 07org.oe.dev * rd36f6928... 10/packages/linux/ (4 files in 3 dirs): Mar 02 10:37:40 opensimpad: Update to WE-18 Mar 02 10:37:40 see http://bugs.openembedded.org/show_bug.cgi?id=595 Mar 02 10:37:40 see http://monotone.vanille.de/viewmtn/revision.psp?id=7237d2d3dbeea22100e5f34a64509cb73978750c Mar 02 10:37:44 03frederic 07org.oe.dev * raafd39c8... 10/packages/linux/ (4 files in 3 dirs): disapproval of revision 'd36f69289f03c75db72375cba8d47b713a10221e' Mar 02 10:45:15 aaaaaargh Mar 02 10:46:39 koen: thats one of those bad revs? Mar 02 10:47:16 hrw|work: yes Mar 02 10:47:36 ~curse frederic Mar 02 10:47:38 May the fleas of a thousand camels infest your most sensitive regions, frederic ! Mar 02 10:51:35 * koen wants monotone 0.26 Mar 02 10:54:07 koen: should I mention again that there's a straightforward workaround to make that merge work, even without 0.26? :-) Mar 02 10:54:44 njs: could you mail that to koen@handhelds.org? Mar 02 10:55:25 njs: I'll let the guilty person suffer a few hours more in the mean time **** ENDING LOGGING AT Thu Mar 02 10:59:56 2006