**** BEGIN LOGGING AT Sun Aug 24 02:59:56 2008 Aug 24 03:50:51 Yes, but you're gone... Aug 24 04:28:47 beagle newbie looking to see if anyone is using fc8 as host devel machine... libtool and a couple of others are eating my lunch Aug 24 07:59:00 hi, how can I add new tasks - like "make install" - to a bitbake provess? Aug 24 07:59:34 process Aug 24 10:49:13 gm mickeyl Aug 24 10:49:35 hey Crofton|work Aug 24 11:11:30 morning Aug 24 11:11:35 hows life OE stylee Aug 24 11:14:46 g Aug 24 11:14:47 gm Aug 24 11:14:49 that is Aug 24 11:17:18 frakking xilinx fpga tools have taken me a day to install Aug 24 11:17:19 not much of a party around here today, everyone musta been clubbing last night Aug 24 11:17:27 heh Aug 24 11:17:40 I think everyone is on vacation or something Aug 24 11:17:47 probably it is summer] Aug 24 11:17:54 yeah Aug 24 11:17:56 I try and avoid taking hols in summer Aug 24 11:18:13 my shoulder has been crap, so no point Aug 24 13:07:29 does anyone here really understand boost? Aug 24 13:22:22 Crofton, no, but what's your problem, anyway? Aug 24 13:24:24 I'm building boost 1.3.36 and it added -mt (for thread safe to the lib names) Aug 24 13:25:02 Ah understanding its build system is even worse :) But shouldn't it add -mt for thread safety? Aug 24 13:25:30 well the exisiting files package the non -mt names Aug 24 13:25:44 I'm thinking of adding * to package eitehr Aug 24 13:25:49 Doesn't it mean its not threadsafe? Aug 24 13:25:50 and see what happens :) Aug 24 13:26:07 apparently it can be built both ways Aug 24 13:26:13 Well, if you build both thread and non-thread you're likely to package both, indeed. Aug 24 13:26:48 well, it only seems to build one at a time Aug 24 13:27:05 yeah, probably. Luckily I don't remember that much of boost :) I find it way too cumbersome to be used, unless there's no choice. Aug 24 13:27:17 no choice in my case Aug 24 13:27:18 Then, perhaps, you need to build it twice? Aug 24 13:27:31 Yeah, I know, you don't always have choice... Aug 24 13:28:21 ok, F8 installs both Aug 24 13:28:51 same for F( Aug 24 13:28:54 F9 Aug 24 13:28:58 I don't even know what's F8 ;-) Aug 24 13:29:01 And F9 :) Aug 24 13:29:04 I'll look harder at the build options Aug 24 13:29:07 Fedora Aug 24 13:29:10 ah Aug 24 13:29:22 I thought you're referring to function keys on my keyboard :) Aug 24 13:29:27 heh Aug 24 13:29:37 Well, you can always lookup their recipes? Aug 24 13:29:41 yeah Aug 24 13:40:51 they don't build build ldconfig softlinks them Aug 24 14:14:36 maybe not ldconfig .... Aug 24 14:48:03 khem ping Aug 24 15:15:12 ~seen rschuster Aug 24 15:15:13 rschuster was last seen on IRC in channel #oe, 14d 21h 16m 3s ago, saying: 'woglinde: this is in openssl.inc for example. can't believe that no one else has seen this'. Aug 24 15:15:21 ~seen thebohemian Aug 24 15:15:21 thebohemian was last seen on IRC in channel #oe, 3d 2h 14m 49s ago, saying: 'hi henning'. Aug 24 15:15:29 ~seen zecke Aug 24 15:15:30 zecke was last seen on IRC in channel #oe, 4d 1h 10m 20s ago, saying: 'cool'. Aug 24 15:22:35 laibsch robert is at froscon Aug 24 15:22:49 zecke was driving to netherlands Aug 24 15:22:51 ah, OK Aug 24 15:22:56 thanks Aug 24 15:23:05 you are a close enough substitute ;-) Aug 24 15:23:14 Mind taking a look at classpath-native? Aug 24 15:23:25 I think you have an interest in Java, too, right? Aug 24 15:23:29 laibsch whats the problem? Aug 24 15:23:38 fails to compile Aug 24 15:23:49 do you have gettext-0.17 enabled? Aug 24 15:23:53 http://tinderbox.openembedded.net/public/logs/828374.txt Aug 24 15:24:00 let me see Aug 24 15:24:21 aj okay Aug 24 15:24:26 apparently not Aug 24 15:24:30 0.14 Aug 24 15:24:30 it is not this error Aug 24 15:24:40 make -j 3 Aug 24 15:24:50 I guess that could be a problem Aug 24 15:25:47 Let's see Aug 24 15:30:38 woglinde: nope, PARALLEL_MAKE="" does not fix it Aug 24 15:31:20 hm Aug 24 15:31:27 the error is strange Aug 24 15:33:09 hm ah Aug 24 15:33:11 seems Aug 24 15:33:24 I never checked in my ecj patches Aug 24 15:35:56 go right ahead Aug 24 15:36:09 don't keep 'em all to yourself Aug 24 15:36:46 hehe yes Aug 24 15:36:51 will do Aug 24 15:43:32 hm Aug 24 15:44:16 have to test the lambda closure in lambdaclosure Aug 24 15:51:45 mooo Aug 24 15:52:30 hi zecke Aug 24 15:52:38 gm zecke Aug 24 15:52:42 moo Aug 24 15:52:43 how is .nl? Aug 24 15:52:48 http://www.jwz.org/gruntle/phones.html <- cool Aug 24 15:53:08 Crofton|work: I have fast internet, all I need to be happy, oh and casis drink, and fast food :) Aug 24 15:55:45 one day I should go to the dnalounge Aug 24 15:57:03 he zecke Aug 24 16:00:03 zecke: hehe Aug 24 16:00:45 Crofton|work: is that supposed to be good? Aug 24 16:01:40 dnalounge? Aug 24 16:01:46 Is that some genetics stuff? Aug 24 16:01:51 Laibsch: club in SF Aug 24 16:02:20 San Francisco? Aug 24 16:02:26 Bist Du da in nächster Zeit? Aug 24 16:03:01 Are you trying to gain over koen as most-travelling-OE-person alive? Aug 24 16:03:05 ;-) Aug 24 16:05:39 dunno if it is any good, but I'm curious what jwz did with his money :) Aug 24 16:13:50 woglinde: pong Aug 24 16:17:46 khem how do I enable uclibc_nptl? Aug 24 16:22:49 he mickeyl Aug 24 16:27:35 woglinde: raise the recipe priority thats all Aug 24 16:27:58 khem hm Aug 24 16:28:13 khem will it work with armv4? Aug 24 16:28:36 servus Aug 24 16:31:51 grüezi Aug 24 16:34:29 *g* Aug 24 16:34:35 "gruess gott" Aug 24 16:36:06 RP: ping? Aug 24 16:36:44 woglinde: hi Aug 24 16:37:43 hi mrnice Aug 24 16:41:53 does anybody knows which script was used to convert from mtn to git? Aug 24 16:46:20 felipec, it's in contrib directory in OE, I believe. Aug 24 16:47:35 felipec, and I didn't like your preaching on ohloh, although technically I take your side ;-) Aug 24 16:51:27 polyonymous: heh, I had to defend my opinion =/ Aug 24 16:51:45 felipec, that's exactly the point - you didn't have to ;-) Aug 24 16:54:41 polyonymous: yeah... I need to ignore those guys Aug 24 16:55:45 felipec, I also think that this discussion unnecessarily offends mtn developers, who at least have tried. Aug 24 17:04:15 polyonymous: perhaps... I didn't try to attack mtn, just point out deficiencies in the way pidgin devs use it... in any case, software isn't perfect, sometimes major redesigns are needed, should that offend the people that did the first design? Aug 24 17:05:05 felipec, I'm not going to defend your opponent, he did show quite some ignorance and attitude, that is out of question. Aug 24 17:05:49 polyonymous: you can argue that git is a drastically major redesign of mtn :) Aug 24 17:05:54 oops, I think I misunderstood you. Well, not it shouldn't, but depending on how you put it, it may. Especially if they're still willing to develop it further. Aug 24 17:12:30 polyonymous: well, they shouldn't be offended, maybe they made mistakes, and the new design is better, or maybe the new design has flaws, that's when dispassionate discussion becomes important. If you get emotionally attached to your code so you take criticism on it personally, then things become ugly Aug 24 17:17:51 * NAiL knows his code is utter shit, so he doesn't care to much about criticism (unless it's constructive) ;) Aug 24 17:18:00 what is the common shell languague in .bb files? is it save to use bash stuff there? Aug 24 17:20:00 NAiL: heh, m Aug 24 17:20:17 NAiL: that's the spirit (I do the same) Aug 24 17:21:01 I have incredibly enough managed to get some of my code in the kernel though... Aug 24 17:25:27 NAiL: well, the code that you replaced must have been shittier :) Aug 24 17:27:52 that's the point of collaborative development; we all fix everybody else's crap Aug 24 17:30:59 felipec: It wasn't replacing anything, it was adding support for a device ;) Aug 24 17:33:31 NAiL: hehe, well, if it was approved it must mean something Aug 24 17:33:39 anyway, gotta go Aug 24 17:36:29 woglinde: I think it will work with armv4t but not with armv4 Aug 24 17:36:50 woglinde: I have tested with armv5t Aug 24 17:37:30 woglinde: its EABI that puts restriction not nptl Aug 24 18:43:49 khem hm why doesnt with armv4? Aug 24 18:44:00 nptl is only eabi? Aug 24 18:44:37 iirc yes Aug 24 18:45:03 why this? Aug 24 18:45:07 Or at least, you need to do some hacking for OABI NPTL Aug 24 18:45:13 if it's at all possible Aug 24 18:45:21 has to do with the thread register stuff Aug 24 18:45:32 * Tartarus hasn't had to think about this for a long time Aug 24 18:46:01 why shoud nptl threads use other registers? Aug 24 18:46:07 or depend on it Aug 24 18:47:20 hmm Aug 24 18:47:28 Maybe i'm mis-remembering Aug 24 18:48:13 khem will awnser Aug 24 18:48:16 *g* Aug 24 18:48:18 :) Aug 24 20:27:44 hi Aug 24 20:27:48 hi hrw Aug 24 20:27:53 from twente? Aug 24 20:28:20 as usual Aug 24 21:33:22 I tried bitbake nano for testing the installation Aug 24 21:33:38 How long will this test take? Aug 24 21:34:12 I thought a couple of minutes, but he goes on for 2 hours now Aug 24 21:35:21 If I had know it would take that long I also could have started bitbake qt4-x11-free Aug 24 21:35:28 Proton23: Well, first OE has to build the toolchain, then with the resulting toolchain it needs to build all the dependencies for nano, then finally nano -- so how long it takes depends on the image one selected as the context within which to build nano. Many hours is not unreasonable. Aug 24 21:36:26 If you bitbake qt4-x11-free, bitbake will detect that many of the dependencies are already built (to build nano) and it won't rebuild those. So little time lost, in reality. Aug 24 21:36:40 ahh Aug 24 21:36:43 ok :) Aug 24 21:37:00 qt is broken Aug 24 21:37:01 it seems Aug 24 21:37:11 laibsch reported something Aug 24 21:37:17 :( Aug 24 21:37:27 yeah :( Aug 24 21:38:34 since when? because I already tried it yesterday and though I did something wrong Aug 24 21:38:52 and tried it from scratch again Aug 24 21:40:29 hrw|gone, good to hear from yuo Aug 24 21:58:53 woglinde: do you have some time? Aug 24 21:59:06 hm a little bit Aug 24 22:00:56 woglinde: can you have a look at http://pastebin.ca/1183746 ? it is my aufs .bb file . it is now generic for every 2.6 versions Aug 24 22:01:16 woglinde: just if I forgot some oe style stuff Aug 24 22:01:33 hm Aug 24 22:01:34 woglinde: I was not shure about CVS dates Aug 24 22:01:42 better write a bug report Aug 24 22:01:48 so more people see it Aug 24 22:01:57 sorry I never yet have done with cvssrc Aug 24 22:02:05 but there is semothing in the doku Aug 24 22:02:12 oemanual about Aug 24 22:03:32 woglinde: I will look at it. else the .bb is ok? Aug 24 22:04:09 woglinde: the pkg_postinst() stuff is from madwifi Aug 24 22:06:44 hm seems okay Aug 24 22:06:47 if it works fine Aug 24 22:09:16 woglinde: thanks. do you think a mail to oe-devel is ok? bugtracker is the wrong place to get suggestions I think Aug 24 22:09:41 made both Aug 24 22:09:47 woglinde: ok Aug 24 22:09:49 bugtracker and oe Aug 24 22:10:07 woglinde: I will do so Aug 24 22:10:27 woglinde: aufs is quite usefull :) Aug 24 22:14:09 woglinde: I think thread cancellation would be an issue with nptl/oabi Aug 24 22:14:34 re khem Aug 24 22:14:44 woglinde: oabi uses sjlj and I am not sure if it will work well with nptl Aug 24 22:14:51 hm Aug 24 22:14:54 I will test Aug 24 22:15:13 woglinde: Good you will be one of a kind :) Aug 24 22:16:18 woglinde: I never tested it with oabi and I am not motivated to fix it either if it breaks on oabi, nptl as long as it works with EABI we are good for future Aug 24 22:16:29 hm okay Aug 24 22:17:10 btw it there any mips machine conf we have in oe which uses 2.6 kernel Aug 24 22:17:25 wrt used 2.4.x Aug 24 22:17:48 khem: dream's stuff? Aug 24 22:18:43 zecke: hmm Aug 24 22:19:33 I could use wgt634u Aug 24 22:20:47 Do we have any resident rpm experts? Aug 24 22:21:12 RP: I have used it and I know how to write spec files Aug 24 22:21:22 but thats all :) Aug 24 22:21:46 * RP is wondering if its possible to use it without being foor for installing packages Aug 24 22:21:59 er, without being root :) Aug 24 22:22:45 Even fakeroot fakechroot didn't help since the package I wanted to install was outside the chroot :/ Aug 24 22:23:03 hmmm I was about to suggest fakeroot Aug 24 22:25:54 RP: I have built packages as non-root user but installs were always as super user Aug 24 22:26:23 My only remaining idea is to share the packages over http and then it would be able to access them inside the chroot Aug 24 22:26:55 Keeping in mind I'd want to do this from rootfs_rpm.bbclass, I don't like the idea though... Aug 24 22:27:34 RP: isnt it a security breach Aug 24 22:27:51 khem: amongst other problems Aug 24 22:28:11 This whole experience hasn't made me an rpm/yum fan :/ Aug 24 22:28:35 why cant you use root Aug 24 22:28:56 khem: in rootfs_rpm.bbclass? Aug 24 22:29:01 RP: maybee --root will help? http://www.rpm.org/max-rpm/s1-rpm-install-additional-options.html#S2-RPM-INSTALL-ROOT Aug 24 22:29:15 mr_nice: Its implremented using chroot Aug 24 22:29:49 Once inside it you can't access packages from the filesystem outside the chroot Aug 24 22:31:29 RP: hey. do you know how to restart the ciabot stuff on amethyst? i did not touch ciabot in ages Aug 24 22:31:37 (+email commit) Aug 24 22:31:48 RP: but you can copy the file into the tmp of you chroot before chrooting Aug 24 22:31:49 mickey|dinner: monotone or git? Aug 24 22:32:08 khem: I guess so. That is ugly as anything though :( Aug 24 22:32:10 RP: both are not running, but anyone would do Aug 24 22:32:33 mickey|dinner: There is no bot running for git at least, it just makes xmlrpc calls to the cia server Aug 24 22:32:41 RP: well or you can make your chroot mount other partitions/directories Aug 24 22:32:53 hmm Aug 24 22:32:54 khem: without root? Aug 24 22:33:05 ok, i wonder why our commit mails have been stalling for days now Aug 24 22:33:07 inside chroot you are root Aug 24 22:33:28 mickey|dinner: someone at GMIT turned off my machine once again... Aug 24 22:33:32 khem: with fakeroot/fakechroot, you won't be able to mount Aug 24 22:33:32 mickey|dinner: may be its waiting for mtn shutdown Aug 24 22:33:35 hmm Aug 24 22:33:39 your machine? Aug 24 22:33:40 RP: yes but without chroot you could combine --root with --dbpath Aug 24 22:33:42 mickey|dinner: so no mtn2git.py is running Aug 24 22:33:55 why am i not aware that serious infrastructure movements have been done? Aug 24 22:33:58 mickey|dinner: means no one is pushing to git.oe.net, means no xmlrpc... Aug 24 22:34:07 last i knew was that ciabots were running on amethyst Aug 24 22:34:08 mr_nice: The chroot is part of rpm's internal implementation of --root Aug 24 22:34:08 mickey|dinner: haeh? Aug 24 22:34:27 and i have not seen any announcements on the list for that moving. perhaps i missed that Aug 24 22:34:35 mickey|dinner: If the monotone to git merges weren't happening that would explain the lack of git commit messages Aug 24 22:34:38 RP: ok Aug 24 22:34:47 mickey|dinner: I have no idea how the monotone ones work Aug 24 22:34:52 ah Aug 24 22:34:58 well, previously there was a bot running Aug 24 22:35:02 that waited for monotone triggers Aug 24 22:35:13 then sending out commit mails and launching XMLRPC to ciabot Aug 24 22:35:25 mickey|dinner: I suspect its just the monotone bot thats failed Aug 24 22:35:30 RP: I am still pushing into mtn should I start using git ? Aug 24 22:35:37 khem: not yet Aug 24 22:35:53 * zecke hides now :) Aug 24 22:35:58 zecke: Any progress? ;-) Aug 24 22:36:12 zecke: too slow ;-) Aug 24 22:36:13 ok Aug 24 22:36:24 fair enough, so there's nothing for me to reinstantiate Aug 24 22:36:38 i have been looking why the ciabot stopped running Aug 24 22:36:43 RP: sure. I mostly rewrote the Qt font backend of QtWebKit and should see 100/100 on acid3 seen Aug 24 22:36:50 but if you say it hasn't been running for ages, i'm fine with that Aug 24 22:37:15 mickey|dinner: I'm not saying that. I'm saying the git side of things is fine and I have no idea whats up with monotone Aug 24 22:37:26 zecke: great, nice work :) Aug 24 22:37:27 heh Aug 24 22:37:35 ok Aug 24 22:37:38 too much distributed sysadmin work Aug 24 22:37:45 *shrug* Aug 24 22:38:24 nite Aug 24 22:38:31 good idea Aug 24 22:38:32 g'night Aug 24 22:39:14 'night mickey|zzZZzz Aug 24 22:39:25 RP: was that progress you were asking for? :} Aug 24 22:39:39 let me try again Aug 24 22:39:47 was it that kind of progress you were asking for? Aug 24 22:40:18 zecke: I was trying to be polite and not ask directly ;-) Aug 24 22:40:31 zecke: Did you have a chance to look at the mtn->git conversion issues? Aug 24 22:41:31 RP: I'm still in front of this author -> name, mail address map Aug 24 22:41:50 zecke: Did you see the wiki page that was started? Aug 24 22:42:54 RP: yes, but that is not complete for the easy part :) Aug 24 22:43:00 Laibsch: I think we have something you could help with? :) Aug 24 22:43:16 oh, we already talked :) Aug 24 22:43:25 zecke: ok, great :) Aug 24 22:43:30 the tough bit is whenever someone used --author Aug 24 22:43:55 We just preseve what was there? Aug 24 22:44:09 http://openembedded.org/~zecke/all_names Aug 24 22:45:44 zecke: I could certainly help clean some of that up Aug 24 22:45:53 zecke: What format do we need the result in? Aug 24 22:46:25 RP: python parsable, so basicly a dict so I can build forward and backward map for mtn2git.py Aug 24 22:46:45 and that is where I confuse Laibsch Aug 24 22:47:28 zecke: Which is what you started at the top of that file? Aug 24 22:47:33 yes Aug 24 22:47:51 if someone starts from the bottom we can easily merge Aug 24 22:48:06 ok. I can give that a go now if it would help? Aug 24 22:48:25 sure, makes it twice as fast... but again it is really boring Aug 24 22:48:40 I used to data entry as a summer job ;-) Aug 24 22:52:05 damn it... rewrote everything and it is still broken and more broken than before :( Aug 24 22:52:59 zecke: the conversion script? Aug 24 22:53:08 WebKit Aug 24 22:53:14 ah :/ Aug 24 22:54:41 anyway time to watch Code Rush now Aug 24 23:35:39 zecke: http://www.rp.rpsys.net/all_names2 Aug 24 23:39:31 RP: Are you still working on that? Aug 24 23:39:46 If not, I'd like to take a stab at the remaining blanks Aug 24 23:39:52 Laibsch: No, I'm calling it a night at that Aug 24 23:39:57 Laibsch: feel free Aug 24 23:39:57 Cool Aug 24 23:40:32 Laibsch: You realise the "" outside the [] is for the final email address we'll use in the conversion? Aug 24 23:40:46 (on the right hand side) Aug 24 23:40:55 "" on the left is the name Aug 24 23:40:59 thanks for pointing that out Aug 24 23:41:10 I was mostly going to fill in a few real names Aug 24 23:41:25 Maybe I should put this in a wiki Aug 24 23:41:31 So we can track changes Aug 24 23:41:38 In case I introduce some errors Aug 24 23:41:39 ok. I think there are some email addresses from our original table that could go in too Aug 24 23:41:45 On a temporary talk page Aug 24 23:41:54 Laibsch: The format is kind of important since its python readable Aug 24 23:42:00 I know Aug 24 23:42:10 This is not meant to really become a wiki page Aug 24 23:42:17 I'll show you in a minute Aug 24 23:42:25 no, I just wanted to make sure that was understood Aug 24 23:43:19 http://wiki.openembedded.net/index.php?title=Talk:Monotone_to_Git_ID_mapping&action=edit Aug 24 23:43:25 Copy the text from there Aug 24 23:43:30 That is your text now Aug 24 23:43:38 The page itself will be rather unusable Aug 24 23:43:51 But it allows us to easily track changes Aug 24 23:44:01 so zecke can have a look over our work Aug 24 23:50:17 Laibsch: ok Aug 24 23:51:01 'night all Aug 24 23:51:04 * RP -> Zzzz Aug 25 00:10:01 rwhitby: rw@openembedded.org., is that you? Aug 25 00:18:21 zecke: http://wiki.openembedded.net/index.php?title=Talk:Monotone_to_Git_ID_mapping&action=edit Aug 25 00:18:29 Copy the test in the input box Aug 25 00:18:36 or edit it there **** ENDING LOGGING AT Mon Aug 25 02:59:57 2008