**** BEGIN LOGGING AT Thu Dec 14 02:59:57 2006 Dec 14 07:55:41 good morning everyone Dec 14 07:57:58 bonjour vlo Dec 14 07:58:06 hello Dec 14 08:00:26 good morning everyone Dec 14 08:00:32 morning Dec 14 08:00:47 * hrw lost network yesterday evening Dec 14 08:01:56 good morning koen, hrw Dec 14 08:03:24 hi koen, hi hrw, hi vlo, Genesis and hi all Dec 14 08:04:28 devkitidp-pxa255-kernel_2.6.11.bb - who maintain it? Dec 14 08:04:42 koen, : so have you try AGG ? Dec 14 08:05:11 hrw, : i think we shouldn't removed the maintener section in bb files :( Dec 14 08:07:27 -rw-r--r-- 1 hrw hrw 17927528 Dec 14 00:48 gpe-image-collie-20061213223736.rootfs.jffs2 Dec 14 08:07:55 collie in .dev suxx Dec 14 08:29:52 hrw: gnutils 1.6.0 fails to build for powerpc Dec 14 08:32:18 koen: then fix it - works for arm and x86 Dec 14 08:33:06 koen: Debian has it built for powerpc without patching Dec 14 08:33:46 I suspect it's related to libstc++.la b0rkage Dec 14 08:34:11 so you have broken toolchain? Dec 14 08:34:36 powerpc makes the breakage more obvious Dec 14 08:34:42 see XorA|goner's mail to the list Dec 14 08:35:55 I saw that Dec 14 08:37:08 opieII thread is going to be more and more interesting Dec 14 08:37:31 I hope lorn unbreaks the 'release' and build system Dec 14 08:38:38 I do not Dec 14 08:39:31 bye - time to go to work Dec 14 08:40:54 morning Dec 14 08:41:16 hey XorA Dec 14 08:43:12 | powerpc-angstrom-linux-gcc -mcpu=603e -DHAVE_CONFIG_H -I. -isystem/data/build/koen/OE/build/tmp/angstrom/staging/powerpc-angstrom-linux/include -I/data/build/koen/OE/build/tmp/angstrom/staging/i686-angstrom-linux/include -I/data/build/koen/OE/build/tmp/angstrom/staging/i686-angstrom-linux/include -MT keyserver.lo -MD -MP -MF .deps/keyserver.Tpo -c keyserver.c -fPIC -DPIC -o .libs/keyserver.o Dec 14 08:43:15 spot the error Dec 14 08:43:58 XorA: libmudflap and libsupc++ la files need sedding as well Dec 14 08:44:13 koen: I suspect all *.la in cross do Dec 14 08:44:32 WTF is libmudflap?? Dec 14 08:44:36 those a the only only with i686 according to grep Dec 14 08:44:48 libmudflap is some gcc thingo Dec 14 08:45:47 koen: wierd reports on c860 flashing Dec 14 09:06:22 hrw|gone: NOTE: package gnutls-1.6.0-r2: task do_build: completed Dec 14 09:06:36 hrw|gone: that's after sedding out -I/staging/i686 Dec 14 09:08:02 re Dec 14 09:08:59 hrw|work: what happens if you build gnutls 1.6.0 for arm after you built a progear image? Dec 14 09:09:14 koen: I do not have such builddir Dec 14 09:09:33 http://ewi546.ewi.utwente.nl/tinderbox/showbuilds.pl?tree=OpenEmbeddedBuilds looks ok Dec 14 09:12:57 !oebug 1682 Dec 14 09:12:58 * * Bug 1682, Status: NEW, Created: 2006-12-14 01:40 Dec 14 09:12:59 * * koen(AT)dominion.kabel.utwente.nl: gnutls suffers from cross-build contamination Dec 14 09:13:00 * * http://bugs.openembedded.org/show_bug.cgi?id=1682 Dec 14 09:32:21 all hail QA meister zecke Dec 14 09:32:24 ~hail zecke Dec 14 09:32:26 * ibot bows down to zecke and chants, "I'M NOT WORTHY!!" Dec 14 09:33:00 yeah, I need that *shivers* and waits for his exam(ination) at two o'clock Dec 14 09:34:39 hals und beinbruch Dec 14 09:35:00 hehe nice :) Dec 14 09:35:39 zecke: bluetooth yah brain and we will google for you Dec 14 09:36:23 koen: I think I will try to have a lightning talk as well... using kergoth's notes on buildsystem Dec 14 09:36:38 "Fix your fucking buildsystem - god damn it" will be the title :) Dec 14 09:37:25 I will pass the abstract through the censorship department though... Dec 14 09:38:45 or do we have a better talk for OE? Dec 14 09:56:17 zecke : hi ! Dec 14 09:56:35 hey Dec 14 09:57:02 zecke : check #1495 i have posted a patch for icecc Dec 14 09:57:41 zecke : cross compilling works. now trying to get native and cross-kernel supported Dec 14 09:57:57 awesome :) Dec 14 09:58:40 zecke : i am thinking of moving the "blacklists" in variables which can be set from local.conf. What do you think ? Dec 14 09:59:52 Ifaistos: sure, default to a default though :) Dec 14 10:00:13 Ifaistos: icecc.bbclass needs to check for ICECC_PATH instead of assuming it's set Dec 14 10:00:13 getVar('TARGET_FPU', d, True) as we want to have TARGET_FPU expanded in all cases Dec 14 10:00:57 koen : ok. will do that Dec 14 10:02:39 Ifaistos: --silent is new for icecc-create-env? Dec 14 10:02:42 Ifaistos: anything else? Dec 14 10:03:12 zecke : You can set the as and define the output path/filename for the tar Dec 14 10:03:12 added_gcc is new as well? Dec 14 10:03:40 Ifaistos: talk to coolo and ask if we is interested in a proper patch Dec 14 10:05:20 zecke : basicaly is "compatible" with the standard icecc-create-env which expects only gcc/g++ as arguments plus it adds a couple more functions Dec 14 10:05:29 AND R0, R0, R0 looks like a good old NOP on ARM right ? Dec 14 10:06:07 Ifaistos: okay well, I will take a look at the diff and push it to the icecream svn Dec 14 10:06:27 Ifaistos: and you rock :) Dec 14 10:07:33 vlo: mov r0,r0 Dec 14 10:07:44 that's the canonical NOP for arm Dec 14 10:07:50 but yours looks like one as well :) Dec 14 10:07:58 and isnt a NOP Dec 14 10:07:58 it will update the flags register Dec 14 10:08:34 cause that's what my mx31 is doing all the time after (my guess) init dies ... Dec 14 10:09:29 or quits ... Dec 14 10:09:46 which shouldn't happen Dec 14 10:13:04 zecke : Thanks ! Well this was my first bbclass work ;) Dec 14 10:15:06 hrw|work: I will look into the coreutils-native failure on edgy Dec 14 10:17:22 re Dec 14 10:20:23 ok Dec 14 10:23:34 zecke: "bmz c7x0 E-17" will go after repull Dec 14 10:24:35 is there something which easily allows you to build a program with -g option for debug ? Dec 14 10:24:47 sure Dec 14 10:25:03 set the optimisation flag variable Dec 14 10:26:10 03hrw 07org.oe.oz354x * r1274d84a... 10/ (3 files in 3 dirs): Dec 14 10:26:10 epsilon: fix includes, closes #1623 (from .dev) Dec 14 10:26:10 applied changes from 2ff0b12bedc3fd85cd0cbbce60e7586f5ae352b5 (.dev) Dec 14 10:26:10 through fb24b0f3b51e2ab57fb4f0291af7e4c6e178b0da Dec 14 10:27:45 thx zecke Dec 14 10:27:52 fscking konq/emb Dec 14 10:28:06 hrw|work: no not konqe Dec 14 10:28:17 good morning! Dec 14 10:28:19 hrw|work: fscking kde1/2/3 buildsystem Dec 14 10:28:22 hey Laibsch Dec 14 10:29:07 OE is just building grep-native for me. Couldn't we assume this as provided? I guess all modern distros have grep as default. Dec 14 10:29:37 gnu grep? Dec 14 10:29:52 Laibsch: ;) Dec 14 10:30:26 Laibsch: all modern __GNU linux__ distributions might have GNU grep installed Dec 14 10:31:30 Laibsch: same applies for GNU make, GNU sed, GNU cp, GNU find :) Dec 14 10:34:55 OK, so it seems like this does not make sense as default. But it will apply to quite many users. Anyone setting this in local.conf and willing to share what they assume as provided? Dec 14 10:35:23 I guess it should be worth the effort. Dec 14 10:35:35 https://bugs.freedesktop.org/show_bug.cgi?id=9142 â–’ Dec 14 10:35:41 dbus security Dec 14 10:39:10 okay later Dec 14 10:39:14 zecke: can you try coreutils 5.3.0 from .dev? Dec 14 10:39:31 will look into it Dec 14 10:41:19 hi Cliff Dec 14 10:45:00 03koen 07org.oe.dev * r8ed5d163... 10/ (3 files in 3 dirs): dbus 1.0.1: apply patch to fix CVE-2006-6107 Dec 14 10:53:04 03koen 07org.oe.dev * r5a6cf1f2... 10/ (1 packages/dbus/dbus.inc): dbus.inc: glib-2.0 isn't needed any more Dec 14 10:59:33 03koen 07org.oe.dev * r5ab38f8d... 10/ (10 files in 4 dirs): dbus: add 1.0.2 Dec 14 10:59:45 03koen 07org.oe.dev * r01463f0f... 10/ (1 conf/distro/angstrom-2007.1.conf): angstrom: prefer dbus 1.0.2 Dec 14 11:01:28 hrw|work: hello Marcin Dec 14 11:01:39 hrw|work: over 60 bugs fixed over the course of the last week and only 30 new bugs openened. Dec 14 11:01:50 Laibsch: which is good Dec 14 11:02:03 yes, good work! Dec 14 11:02:06 ERROR: Build of /home/hrw/devel/oe/org.openembedded.dev/packages/poboxserver/poboxserver_1.2.5.bb do_fetch failed Dec 14 11:02:16 oh? Dec 14 11:02:24 retrying on other machine Dec 14 11:02:24 Lemme take a look. Dec 14 11:03:27 yes, seems like the server has changed. Dec 14 11:03:35 I'll investigate and fix. Dec 14 11:04:57 It is still linked from his own homepage. I'll just host the source on gakusei.sf.net Dec 14 11:05:51 12:30:07 ERROR 404: Not Found. Dec 14 11:07:14 Laibsch: ok Dec 14 11:19:41 morning Dec 14 11:19:52 hi mickeyl Dec 14 11:22:44 http://www.engadget.com/2006/12/14/puppybot-has-a-nokia-770-for-a-head/ Dec 14 11:23:01 cool stuff Dec 14 11:23:16 next year i'm buying one of those robot construction sets Dec 14 11:23:19 just dunno which one yet Dec 14 11:23:28 will put a Neo1973 as head :D Dec 14 11:23:35 although it lacks a camera Dec 14 11:25:53 03hrw 07org.oe.dev * r3da2caa6... 10/ (4 files in 3 dirs): u2nl: update to 1.3 Dec 14 11:30:56 http://jobs.apple.com/index.ajs?BID=1&method=mExternal.showJob&RID=27&CurrentPage=1 <- wow apple offers nice jobs Dec 14 11:33:33 zecke : what's the best way to evaluate if a parameter passed from local.conf is empty ? Dec 14 11:34:10 s = getVar('FOO') or "default" Dec 14 11:35:11 zecke : can default be an expression ? os.popen blah blah ? Dec 14 11:35:33 sure Dec 14 11:35:45 if getVar('FOO') returns none the right site is evaluated Dec 14 11:36:39 zecke : thanks Dec 14 11:37:59 zecke : i have a 7 machine cluster compiling now (6 i686 and one amd64).... now it seems configure takes for ever to finish :) Dec 14 11:38:35 hehe Dec 14 11:38:44 I think most of the time is spent on autoreconf Dec 14 11:39:34 zecke : i wonder if the threaded version of bitbake will make any difference... Dec 14 11:42:14 NOTE: package wireshark-0.99.4-r0: task do_compile: started Dec 14 11:49:20 failed Dec 14 11:50:04 hrw|work: NOTE: package poboxserver-1.2.5-r2: task do_fetch: completed Dec 14 11:50:45 Ifaistos: That's great news. So are your latest changes the breakthrough to get things compiling with icecc? Dec 14 11:51:32 mickeyl: I wonder if compiling with icecc (one machine ubuntu eft and one ubuntu dapper) will corrupt tmp/ again? Dec 14 11:51:42 never used icecc Dec 14 11:51:45 * mickeyl points to zecke Dec 14 11:51:51 Laibsch : it seems so. What i have posted only cross-compiles. working to get native and cross-kernel working Dec 14 11:52:31 mmm, link all OE dev machines with icecc into a worldwide mega compiler farm Dec 14 11:52:51 03Laibsch 07org.oe.dev * rbde1fc71... 10/ (1 packages/poboxserver/poboxserver_1.2.5.bb): poboxserver: upstream sources gone (temporarily?). Host at gakusei.sf.net now. Dec 14 11:55:14 XorA : Then take over the world :) Dec 14 12:01:55 Laibsch: contacted upstream about archive? Dec 14 12:02:06 no Dec 14 12:02:17 but sounds like a good idea ;-) Dec 14 12:04:08 zecke: in past we had "bittest source_checker" which checked availability of sources. do we have it or was it dropped at all? Dec 14 12:04:27 hrw|work: it's called bitbake_qa now Dec 14 12:05:13 http://www.openembedded.org/wiki/BitTest Dec 14 12:06:29 thx Dec 14 12:13:50 03hrw 07org.oe.dev * rfd56d089... 10/ (5 files in 2 dirs): dbus: package libdbus into separate package - opie images will get smaller now Dec 14 12:23:32 03hrw 07org.oe.dev * rbe6037f1... 10/ (1 packages/dbus/dbus.inc): dbus: make lib recommend dbus Dec 14 12:26:42 03koen 07org.oe.dev * rdfd3feed... 10/ (1 packages/dbus/dbus.inc): dbus.inc: revert granularity change till someone comes up with something that doesn't break every dbus using app Dec 14 12:27:40 !bootstrap images are too big anyway to fit on collie Dec 14 12:28:09 we can work on fixing that Dec 14 12:28:39 * koen is looking forward to OE's unit tests for binary sizes Dec 14 12:28:46 opie is 16M, gpe is 17M Dec 14 12:28:52 hrw|work: tried compiling with -Os? Dec 14 12:29:02 i thought Os is breaking something? Dec 14 12:29:06 recently read that on alkml Dec 14 12:29:11 that saved 1MB last time Dec 14 12:29:29 mickeyl: perl, but perl now has -O1 in angstrom Dec 14 12:29:36 ah that was ust perl Dec 14 12:29:38 just, even Dec 14 12:29:43 who needs perl anyway`? Dec 14 12:29:44 * mickeyl hides Dec 14 12:29:49 mickeyl: git Dec 14 12:30:34 task-gpe NEED task-basing Dec 14 12:31:09 OT: anyone know if there's a list with glib data structures and their operational costs in O notation? Dec 14 12:31:17 and if so, where? :D Dec 14 12:31:22 * mickeyl brews coffee Dec 14 12:31:33 someone want to fight with wireshark? Dec 14 12:31:44 hrw|work: I tried last month and gave up Dec 14 12:32:24 (wireshark) Dec 14 12:33:10 did they break something? i added ethereal and it worked Dec 14 12:33:20 months ago, that is Dec 14 12:33:46 wireshark->ethereal should have been s:ethereal:wireshark:g Dec 14 12:33:46 mickeyl: ethereal is not fetchable Dec 14 12:33:55 but somehow it doesn't build Dec 14 12:33:57 koen: cross compile badness Dec 14 12:34:11 hrw|work: if you fix that, it still won't compile Dec 14 12:35:26 mickeyl: you know about '-b foo.bb -c devshell'? Dec 14 12:35:59 not really. educate me, what can i do with that? Dec 14 12:36:18 it dumps you into as shell in ${S} with all the env vars set up Dec 14 12:36:23 s/as/a/ Dec 14 12:36:40 it's very usefull Dec 14 12:36:51 which env vars in particular? Dec 14 12:36:58 (after inheriting devhell.bbclass, whoch angstrom does) Dec 14 12:36:58 i.e. what can i do with that now i have a shell prompt Dec 14 12:36:59 I liked old devshell anyway Dec 14 12:37:25 it's the like old devshell, but more obvious and userfriendly Dec 14 12:37:36 mickeyl: 'make' just works(TM) Dec 14 12:37:41 oh Dec 14 12:37:43 that's cool Dec 14 12:38:01 hrw-source-mirror-fetcher build is progressing in nice red color Dec 14 12:38:04 food time Dec 14 12:38:11 bbl Dec 14 12:38:12 ~bon appetit Dec 14 12:38:16 extra, extra, read all about it, bon appetit is smacznego. Guten Appetit. Eet Smakelijk. God Appetitt. Buon Appetito. Buen apetito Bom Apetite. buen apetito Dec 14 12:38:39 I've just started with OE. Having trouble with bitbake not being able to find conf/bitbake.conf Dec 14 12:38:43 ~lart g_slist_length Dec 14 12:38:43 * ibot beats g_slist_length over the head with a microkernel Dec 14 12:39:01 they're really crawling through the complete list to gather the lenght Dec 14 12:39:02 *sigh* Dec 14 12:39:32 that makes it O(n) Dec 14 12:39:34 :/ Dec 14 12:39:47 so, double bookkeeping for me Dec 14 12:39:51 Pakmanorama: Have you read GettingStarted on the wiki? Dec 14 12:40:03 Pakmanorama: Have you configured local.conf? Dec 14 12:40:13 Laibsch: yeah, read it a couple of times Dec 14 12:40:25 Pakmanorama: Have you set up the environment variables? Dec 14 12:40:33 Laibsch: I think I've done local.conf OK Dec 14 12:41:05 hrw|work: Sent a mail to upstream. I wonder if he will be able to decipher what I was trying to tell him ;-) Dec 14 12:41:14 Laibsch: yes path ok... BBPATH=/oe/build:/oe/org.openembedded.dev Dec 14 12:41:50 Laibsch: am I supposed to copy bitbake.conf into /oe/build/conf ? Dec 14 12:41:54 mickeyl: have you tried glib >2.6 with openmoko? Dec 14 12:42:11 hi all Dec 14 12:42:12 no chance yet Dec 14 12:42:37 mickeyl: if you, let me know, the irex dudes where wondering about that as well Dec 14 12:42:55 Pakmanorama: no Dec 14 12:43:05 k Dec 14 12:43:14 k Dec 14 12:45:35 Pakmanorama: Can you paste a bit of info? Dec 14 12:45:38 !paste Dec 14 12:45:57 -> http://rafb.net Dec 14 12:46:28 What vars have you set, what directory are you in when calling bitbake, paste your local.conf, etc. Dec 14 12:46:52 Laibsch: I just uncommented the REMOVE_THIS_LINE stuff in /oe/build/conf/local.conf and it didn't give the expected error Dec 14 12:47:01 Laibsch: ok, will paste Dec 14 12:47:01 Are you in /oe/build when calling bitbake? Dec 14 12:47:09 Laibsch: yes Dec 14 12:47:26 Pakmanorama: Then of course it will not work. It say quite clearly you need to READ the conf file! Dec 14 12:47:31 Laibsch: running inside "screen" if that makes a difference Dec 14 12:47:34 no Dec 14 12:47:40 I do that all the time. Dec 14 12:47:51 You need to fix your local.conf Dec 14 12:48:27 Laibsch: I uncommented the line to try and see if local.conf was being read Dec 14 12:48:42 But you did not read the conf file. Dec 14 12:49:06 READ it, understand it and configure it and I think you will be happy. Dec 14 13:03:31 init fails after my linux boot: how to easily add some "printf" in the init package and rebuild it so that I can see exactly where it fails ? Dec 14 13:03:41 http://pastebin.ca/278842 Dec 14 13:04:50 Pakmanorama: Making progress? Dec 14 13:05:34 because of course, when I rebuild init, bitbake decompresses the source again, overwriting all the changes I could make in the init.c file for instance Dec 14 13:08:26 Laibsch: http://rafb.net/paste/results/G5nwmE46.html Dec 14 13:09:01 Laibsch: read, read again, re-read and once more to be sure :) Still can't see what I'm doing Dec 14 13:09:08 Laibsch: wrong Dec 14 13:09:14 Pakmanorama: http://www.hrw.one.pl/ - read 'my openembedded environment II' post Dec 14 13:09:42 hrw|work: thanks, will look Dec 14 13:21:03 Let me summarize: Your config is http://rafb.net/paste/results/G5nwmE46.html, "set|grep BB" show you BBPATH=/oe/build:/oe/org.openembedded.dev, you checked out the OE tree into /oe/org.openembedded.dev (/oe/org.openmbedded.dev/conf/bitbake.conf is there and non-empty), but calling "bitbake nano" in /oe/build directory errors out for you with conf/bitbake.conf not found? Dec 14 13:22:43 plus your local.conf is at /oe/build/conf/local.conf ? Dec 14 13:23:40 Ive just been hit by !oebug 1680 Dec 14 13:23:45 !oebug 1680 Dec 14 13:24:56 * * Bug 1680, Status: NEW, Created: 2006-12-13 03:14 Dec 14 13:24:56 * * andrea.adami(AT)gmail.com: JFFS2 angstrom gpe-image won't boot Dec 14 13:24:56 * * http://bugs.openembedded.org/show_bug.cgi?id=1680 Dec 14 13:25:42 btw - someone tried 2.6.19 on Z? Dec 14 13:25:49 hrw|work: just now Dec 14 13:26:07 XorA: and 2.6.19 == 1680? Dec 14 13:26:08 hrw|work: but it wont boot as / is fscked Dec 14 13:26:35 hrw|work: I dont know if its 2.6.19 that fscked it, or an oebug Dec 14 13:27:10 XorA: does it work with 2.6.18? Dec 14 13:27:24 hrw|work: actuall I think I might just have written a tar.gz to jffs2 Dec 14 13:27:45 fsck I have Dec 14 13:27:51 Pakmanorama: Was my above summary correct? Dec 14 13:28:08 Laibsch: exactly, I just finished checking it three times :) Dec 14 13:28:48 Laibsch: I noticed that the sample local.conf has a colon before the = sign in the BBFILES line Dec 14 13:29:09 Laibsch: tried it with and without with no difference Dec 14 13:30:39 Then I think it should work. Maybe you can paste the output of grep -v '^$' ../build/conf/local.conf|grep -v '#' as that is more readable. But I think your config is OK. Maybe one of the experts can take a look. Dec 14 13:30:52 s/paste/pastebin Dec 14 13:31:08 Pakmanorama: BBFILES is not really the problem. Dec 14 13:31:51 Laibsch: is there something I can do to ensure that the conf/local.conf is being read by bitbake? Dec 14 13:31:59 hrw|work: Of course. I think Pakmanorama was not going to dump his stuff straight here. For me the difference is paste and post. But I guess pastebin might be clearer. Dec 14 13:32:27 Pakmanorama: Be in /oe/build when you call bitbake. Dec 14 13:32:54 Pakmanorama: Just had an idea. Paste the output of "bitbake -D -D -D nano" Dec 14 13:33:04 koen|away: NOTE: package wireshark-0.99.4-r0: task do_compile: completed Dec 14 13:33:14 applaud hrw|work Dec 14 13:33:36 Laibsch: ERROR: Unable to open conf/bitbake.conf Dec 14 13:33:51 That is all? Dec 14 13:34:06 Laibsch: Yeah, I tried that before I came on here ;) Dec 14 13:34:27 Well, but the -D should have given you more verbose output. Dec 14 13:34:42 Have you really tried with -D -D -D? Dec 14 13:34:54 Laibsch: single line output as pasted :( Dec 14 13:35:00 Strange. Dec 14 13:35:28 Using installed or SVN version of bitbake? Dec 14 13:35:50 Maybe you want to reinstall bitbake, but I am really poking into the dark here. Dec 14 13:36:03 Laibsch: svn info Dec 14 13:36:03 Path: . Dec 14 13:36:03 URL: http://svn.berlios.de/svnroot/repos/bitbake/branches/bitbake-1.6 Dec 14 13:36:03 Repository UUID: fd90740e-a5ea-0310-a7a2-b3005cc77bd6 Dec 14 13:36:03 Revision: 719 Dec 14 13:36:04 Node Kind: directory Dec 14 13:36:06 Schedule: normal Dec 14 13:36:08 Last Changed Author: zecke123 Dec 14 13:36:10 Last Changed Rev: 626 Dec 14 13:36:12 Last Changed Date: 2006-11-01 21:11:36 +1000 (Wed, 01 Nov 2006) Dec 14 13:36:13 Pakmanorama: Don't do that! Dec 14 13:36:14 Properties Last Updated: 2006-12-14 20:22:26 +1000 (Thu, 14 Dec 2006) Dec 14 13:36:16 oops, sorry Dec 14 13:36:29 Laibsch: oops Dec 14 13:36:35 That is what pastebin is for. http://rafb.net Dec 14 13:37:33 Laibsch: perhaps I'm having an isue with python or some stuff Dec 14 13:37:51 Laibsch: or psyco JIT Dec 14 13:38:11 paste the output of "head /oe/build/conf/local.conf" Dec 14 13:38:32 Paste means http://rafb.net (just for starters ;-)) Dec 14 13:38:41 Laibsch: lol...ok Dec 14 13:38:44 ~pastebin Dec 14 13:38:46 hmm... pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.ca, or http://channels.debian.net/paste Dec 14 13:39:18 or maybe I didnt, as koens Angstrom images are buggered as well Dec 14 13:40:17 Laibsch: http://pastebin.ca/278872 Dec 14 13:40:34 ibot: thanks Dec 14 13:40:34 Pakmanorama: no problem Dec 14 13:40:47 hrw|work: thanks Dec 14 13:48:29 hum, at least I now can have a shell with the boot option init=/bin/sh ... so it's definitely init which has a problem Dec 14 13:50:08 03koen 07org.oe.dev * r3485994a... 10/ (1 packages/dbus/dbus_0.95.bb): dbus: drop 0.95 Dec 14 13:52:30 Pakmanorama: Where is the output of "head /oe/build/conf/local.conf"? Dec 14 13:52:45 Pakmanorama: Disregard Dec 14 13:54:05 Pakmanorama: OK, now I am really lost as to why your setup is not working. It really should. What distro are you using? Anything deb based? Then google for openembedded-essential and install this meta-package. Dec 14 13:54:24 Laibsch: Ubuntu Dec 14 13:54:36 03hrw 07org.oe.dev * rba11475e... 10/ (6 files in 5 dirs): wireshark: updated to 0.99.4 (it was ethereal before) Dec 14 13:54:52 Pakmanorama: Which version? I am also on ubuntu. Dec 14 13:54:55 03hrw 07org.oe.dev * raf2ecc64... 10/ (1 packages/scw/scw_0.4.2.bb packages/scw/scw_0.4.3.bb): scw: update to 0.4.3, point to Debian mirrors Dec 14 13:55:16 hmm nokia wants their machine back... looking for a new idle machine... Dec 14 13:55:36 zecke: which machine you lose? Dec 14 13:56:12 03hrw 07org.oe.dev * rd2e53229... 10/ (3 files in 2 dirs): cdtool: update to 2.1.8 Dec 14 13:56:37 Laibsch: Breezy Dec 14 13:56:43 Pakmanorama: Include "deb http://www.openembedded.org/dl/ packages/" in /etc/apt/sources.list or some separate file in /etc/apt/sources.listd/. Then "sudo apt-get update;sudo apt-get install openembedded-essential" Dec 14 13:56:48 hrw|work: autobuild, looking for a new powerful machine in the rack Dec 14 13:56:57 Pakmanorama: Breezy? That stuff is ancient! Dec 14 13:57:14 Laibsch: Yeah I know. I works though :) Dec 14 13:57:17 Pakmanorama: I suggest upgrading at least to dapper. Dec 14 13:57:39 Pakmanorama: Sorry, but I think this might be because you are running very old software. Dec 14 13:59:16 mickeyl: ping Dec 14 13:59:50 Laibsch: hmmm..apt seems to need build-essentials, quilt and a few others Dec 14 14:00:10 zecke: pöng Dec 14 14:00:16 Pakmanorama: Of course, you always had those installed anyways, didn't you? Dec 14 14:00:24 zecke: test passed? Dec 14 14:00:36 mickeyl: mind if I apply for a lighning talk? Dec 14 14:00:42 Pakmanorama: That is the only function of that meta-package, to make sure necessary software is installed. Dec 14 14:00:42 hmm sloooow wlan Dec 14 14:00:50 zecke: @ FOSDEM? Dec 14 14:01:06 mickeyl: yes AT FOSDEM Dec 14 14:01:10 zecke: not at all, go ahead. Dec 14 14:01:32 mickeyl: OE can only have talk, so we need to judge carefully Dec 14 14:01:53 zecke: we should only coordinate that we don't talk about the same stuff Dec 14 14:02:22 i would rant :) Dec 14 14:02:49 excellent Dec 14 14:02:50 :D Dec 14 14:03:04 Laibsch: Ok, looks like I need to upgrade :( Dec 14 14:03:20 That is what I thought. Dec 14 14:03:26 It is not so bad, usually. Dec 14 14:03:34 Don't forget the backup, though. Dec 14 14:03:48 And don't upgrade straight in eft, that is not supported. Dec 14 14:03:54 Go for dapper first. Dec 14 14:04:20 I compile with OE on dapper (and eft). So no need to go all the way if you don't want to. Dec 14 14:05:03 who overloaded monotone.oe.org? Dec 14 14:05:27 Laibsch: thanks for all your help, it is much appreciated Dec 14 14:05:55 Pakmanorama: de nada. Dec 14 14:06:17 tried stracing bitbake? Dec 14 14:06:25 Be sure to come back. OE is awesome. BTW, may I ask what you intend to do? Dec 14 14:06:36 Pakmanorama: ? Dec 14 14:07:33 Laibsch: We have a gaming terminal that we'd like to run it on Dec 14 14:08:12 03hrw 07org.oe.dev * raccbdc42... 10/ (5 files in 3 dirs): masqmail: update to 0.2.21, switched to Debian mirror Dec 14 14:08:32 Pakmanorama: I see. koen was suggesting for you to strace bitbake. Do you feel like it? Dec 14 14:09:10 Laibsch: yeah, I don't know if that would help as it runs through python but I guess it can't hurt Dec 14 14:09:16 Personally, I would rather upgrade. And you will not get around having certain software which I am not sure is available in breezy. Dec 14 14:10:11 your choice. Dec 14 14:11:35 Laibsch: stat64("conf/bitbake.conf", 0xbfc3a758) = -1 ENOENT (No such file or directory) Dec 14 14:11:44 Laibsch: hmm Dec 14 14:12:22 03hrw 07org.oe.dev * rb66f8aa9... 10/ (1 packages/live555/live555_20051005.bb): live555: point SRC_URI to one of places where it is available Dec 14 14:13:22 I assume your python version is lacking a function which is used for determining the part before conf/local.conf. As I said, I would rather upgrade. But trying to install python from dapper might fix this for you. Dec 14 14:13:59 How long is breezy going to stay supported? Dec 14 14:14:07 Not much longer, I think. Dec 14 14:15:49 Laibsch: working Dec 14 14:15:59 Cool Dec 14 14:16:35 Laibsch: bitbake config file was installed to /usr/bitbake/conf and bitbake was looking in /usr/share/bitbake/conf Dec 14 14:16:40 but don't forget to install quilt and all the other software. You will need it and waste yours and other ppl's time when you run into trouble. Dec 14 14:16:59 /usr? Dec 14 14:17:16 Laibsch: yup, that's where is was looking Dec 14 14:17:25 You were always talking about /oe! And showed me that indeed the file was there. Dec 14 14:17:39 Something is wrong! Dec 14 14:18:08 OE usually does not and should not touch /usr! Dec 14 14:18:46 Laibsch: hmm...that's where bitbake was looking Dec 14 14:18:49 My /usr is mounted ro to make sure it only gets changed when I want it to. Dec 14 14:19:33 Pakmanorama: It certainly won't help you trying to copy it there, now. It will only open a chance for you to completely fsck up your system. Dec 14 14:20:03 I REALLY advice you to upgrade Dec 14 14:20:28 Laibsch: I will put a system together at work. Too risky to upgrade my laptop :) Dec 14 14:20:45 Laibsch: too much crap on it and not enough time to sort it out...lol Dec 14 14:23:51 You could boot into a LIVE CD and compile for OE there. Dec 14 14:24:30 or use [g2]'s chroot thing Dec 14 14:25:07 koen: he did it? Dec 14 14:25:42 03hrw 07org.oe.dev * r40da8c32... 10/ (4 files in 2 dirs): c7x0/akita/spitz: fix order or zaurus includes Dec 14 14:25:55 hrw|work: he has had a ubuntu dapper chroot for a while now Dec 14 14:26:18 ah.. Dec 14 14:27:32 Laibsch: I have just remembered that we have a server configured with dapper on it Dec 14 14:28:21 Laibsch: will openembedded-essentials pick up all packages I need? Dec 14 14:28:48 Pakmanorama: yes Dec 14 14:30:31 excellent. Thank you all for your help. Dec 14 14:33:32 crazy fetch is progressing Dec 14 14:35:15 morning Dec 14 14:38:12 zecke: s/OpenSnyc/OpenSync Dec 14 14:38:58 okay Dec 14 14:39:03 besides that Dec 14 14:41:33 zecke: ok for me Dec 14 14:44:41 zecke: looks ok Dec 14 14:45:48 hopes keithp will be present when I bitch about fontconfig Dec 14 14:45:54 I know how to make me popular Dec 14 14:47:20 ;D Dec 14 14:53:00 ~lart devshell badly Dec 14 14:53:01 * ibot shoots devshell badly in his sleep Dec 14 14:53:27 or rather 'bitbake -c' without -b Dec 14 14:53:39 right Dec 14 14:53:45 it seems to hang without -b Dec 14 14:53:51 forgot to file a bug about that Dec 14 14:56:58 bitbake trunk is broken for devshell Dec 14 14:57:14 * RP sighs Dec 14 14:57:46 yo RP Dec 14 14:57:53 hi mickeyl Dec 14 14:57:58 "morning" all in fact :) Dec 14 14:58:09 heh Dec 14 14:58:15 fscked up your sleep/wake rhythm? Dec 14 14:58:39 mickeyl: no, just complying with ugt ;-) Dec 14 14:58:47 heh, righto Dec 14 15:00:58 list of non-fetchable sources decreased Dec 14 15:04:41 RP : Hi ! Dec 14 15:12:16 brb, reboot Dec 14 15:13:06 MACHINE = "ipaq-pxa270" Dec 14 15:13:14 we still have this machine config? Dec 14 15:13:21 I was thinking that this was dropped.. Dec 14 15:17:22 mickeyl: hsvc in OE is your toy? Dec 14 15:19:01 well Dec 14 15:19:21 more or less, yeah Dec 14 15:19:29 mickeyl: it is not fetchable... Dec 14 15:19:51 ok, i'll have a look Dec 14 15:32:25 ixp4xxbe lack virtual/kernel setting Dec 14 15:32:45 ixp4xxle probably too Dec 14 15:35:27 hrw|work: koen@bitbake:~/OE/monotone/org.openembedded.dev/conf/machine$ grep kernel include/ixp4xx.conf Dec 14 15:35:28 PREFERRED_PROVIDER_virtual/kernel ?= "ixp4xx-kernel" Dec 14 15:35:52 RP: no time to hack on xmlrpcserver yet. I'm sending it to you over the weekend Dec 14 15:39:56 koen: not here Dec 14 15:40:55 03koen 07org.oe.dev * r7d5c4e9e... 10/ (1 conf/machine/ipaq-pxa270.conf): ipaq-pxa270: drop as indicated before. This machine has been split into hx4700 and hx2000 Dec 14 15:40:59 ixp4xxle.conf:PREFERRED_PROVIDER_virtual/kernel ?= "ixp4xx-kernel" Dec 14 15:40:59 hrw|work: heh, right Dec 14 15:41:10 hmm.. why then bitbake complain.. Dec 14 15:48:50 MACHINE = "ixp4xx" Dec 14 15:48:54 NOTE: multiple providers are available for virtual/kernel (nslu2-kernel, linux-ezx, native-kernel, handhelds-sa, devkitidp-pxa255-kernel, compulab-pxa270, linux-jlime-jornada6xx, linux-efika, linux-rp, e680-kernel, linux-amsdelta-2.6, linux, linux-smdk2440, linux-h6300-omap1, linux-dht-walnut, linux-titan-sh4, linux-ks8695, linux-navman, nslu2-linksys-kernel, mx31ads-kernel, a780-kernel, linux-omap1, triton-kernel, mx21ads-kernel, netbook-pro-kernel, linu Dec 14 15:49:08 hm Dec 14 15:50:10 now I'm lost. Dec 14 15:50:13 hrw|work : some of them sure are not ixp Dec 14 15:50:32 ixp4xx has PREFERRED_PROVIDER_virtual/kernel ?= "ixp4xx-kernel" and bitbake has problem with it.. Dec 14 15:50:39 hrw|work: X <----- you're here Dec 14 15:51:05 you're not lost if you don't care where you are Dec 14 15:51:24 www.perdu.com Dec 14 15:51:53 koen : Does efika compile for angstrom ? Dec 14 15:52:44 koen : or is it the other way around :) Dec 14 15:52:52 Ifaistos: apart from libusb, yes Dec 14 15:53:15 http://www.angstrom-distribution.org/unstable/images/20061213/efika/ Dec 14 15:53:24 although I forgot to upload the kernel Dec 14 15:53:41 but the kernel from efika.info works well enough to boot it Dec 14 15:54:05 koen : i want to give something to the compile cluster to work on, as i have to go for some appointment Dec 14 15:54:57 koen : mine is still at the customs... Dec 14 15:55:00 ~lart fedex Dec 14 15:55:01 * ibot forces fedex to use Outlook Express Dec 14 15:57:44 they allready use it ... Dec 14 15:59:28 :) Dec 14 16:01:30 mickeyl: No problem, I've been too busy to think about it atm anyway :-( Dec 14 16:01:37 ok, cool Dec 14 16:02:26 mickeyl: Next week I'm away but I have holiday over Xmas so I'm hoping to get some work done on bitbake then :) Dec 14 16:02:58 reminds of that dilbert where the PHB tries to make Alice take holidays Dec 14 16:04:07 good Dec 14 16:04:13 XorA: Sadly I have other priorities than bitbake when at work :-( Dec 14 16:04:45 RP: its funny as we all get jobs to do with embedded linux, we get less time to actually work on OE Dec 14 16:05:08 * mickeyl just submitted first patch to Maemo Dec 14 16:05:47 XorA: Its sad, isn't it :-/ Dec 14 16:06:12 someone needs to incorporate oe.com and find some spare VC Dec 14 16:06:29 VC? Dec 14 16:06:34 venture capital Dec 14 16:06:36 Venture Capital Dec 14 16:07:09 OH would love to have me working on poky which basically equates to OE long term. We need me working on paid projects though Dec 14 16:07:09 Volture Capital Dec 14 16:07:37 Ifaistos: :-D too true Dec 14 16:07:37 its the only time the bean counters are allowed to gample, so they get ruthless Dec 14 16:07:37 03hrw 07org.oe.dev * rd53336ad... 10/ (3 files in 2 dirs): (log message trimmed) Dec 14 16:07:37 maradns: update to 1.0.39 (last 1.0.x version) Dec 14 16:07:37 http://www.maradns.org/faq.html#version1 say: Dec 14 16:07:37 MaraDNS 1.0 will continue to be fully supported by upstream until December 21, 2007 Dec 14 16:07:37 This means that MaraDNS 1.0 questions will still be answered and bug fixes will Dec 14 16:07:37 still be applied. Dec 14 16:07:37 After 2007/12/21, MaraDNS 1.0 will no longer be fully supported by upstream; Dec 14 16:07:48 is VC == getting it from being in the long term? Dec 14 16:08:08 err behind Dec 14 16:08:12 can't type today Dec 14 16:08:37 You give us your idea, we get 51% and if things go well we through you out of the company Dec 14 16:08:37 chouimat: bend over and take it like a man Dec 14 16:08:37 XorA: hehe Dec 14 16:09:41 JustinP: microemacs is your toy iirc... Dec 14 16:14:03 I'm trying to make a bb for gst-python, but the install is failing, I think because the path to the libraries isn't getting passed correctly to the gcc, but this mess: -L/opt/oe/build-hx/tmp/staging/arm-linux/lib/.libs -L/opt/oe/build-hx/tmp/work/armv5te-linux/python-gstreamer-0.10.6-r0/image/opt/oe/build-hx/tmp/staging/arm-linux/lib Dec 14 16:14:09 any ideas what would cause this? Dec 14 16:14:47 notice the '.libs' and the funky concatenating of paths Dec 14 16:15:03 looks like libtool breakage Dec 14 16:15:14 hey koen Dec 14 16:16:38 koen|away: I don't know much about libtool. Is it likely a problem with some libtool thing that ships with gst-python? I do see ltihooks.py and ltmain.sh in the tarball Dec 14 16:26:04 2.6.20-rc1 released Dec 14 16:27:49 hrw|work: for now it looks like 2.6.20 will be an easy upgrade Dec 14 16:28:24 yes Dec 14 16:28:31 Linux v2.6.20-rc1 Dec 14 16:28:31 Dec 14 16:28:32 .. and so the stabilization phase starts. Dec 14 16:28:43 ops. Dec 14 16:44:13 hi psokolovsky_ Dec 14 16:44:19 Hi! Dec 14 16:45:15 hrw|work: for now only irqchip/desc --> irq_chip/desc hit me and system.h uses local_irq_save/restore without including interrupt.h or irqflags.h Dec 14 16:46:56 hey lrg Dec 14 16:47:00 hey all Dec 14 16:47:24 pH5: testing new changes on mainstone tomorrow Dec 14 16:47:42 pH5: if all is ok then I'll email patch Dec 14 16:47:51 lrg: thanks! I'm looking forward to it Dec 14 16:48:04 pH5: should make things easier :) Dec 14 16:49:20 hello lrg Dec 14 16:49:54 [ARM] Add sys_*at syscalls Dec 14 16:49:54 Dec 14 16:49:56 Later glibc requires the *at syscalls. Add them. Dec 14 16:49:58 RMK Dec 14 16:50:14 hey Marcin, sent email before seeing your reply Dec 14 16:50:18 so linux-libc-headers 2.6.20 in OE will be without patc Dec 14 16:50:20 lrg: ;) Dec 14 16:50:44 hey vlo, booting init now ? Dec 14 16:51:36 now it runs init, but init hangs ... but I can have shell with init=/bin/sh Dec 14 16:51:49 this is very strange Dec 14 16:53:10 cu Dec 14 16:54:11 vlo: I'm doing work soon, maybe next week. I'll build a clean image Dec 14 16:56:47 lrg: great. I hope my image will work one day Dec 14 16:59:34 I'm missing a good debug technique with the embedded JTAG probe, I can work the proc step by step and see the assembly being executed, but I would need to be able to link it with gdb efficiently Dec 14 17:04:04 but now I've learnt to use quilt, so I've been able to add "printf"'s (I know it's ugly, but it works quite well) in the init c code, and I can see where the problem is Dec 14 17:04:08 and it works ! Dec 14 17:04:17 I see my messages Dec 14 17:04:20 yepee Dec 14 17:05:40 endian-little? someone finally split up the site files? Dec 14 17:05:44 nifty Dec 14 17:08:26 ok, got to go, see you all later Dec 14 17:09:24 kergoth: yep Dec 14 17:09:38 kergoth: Jamie lenehan did the ground work and RP refined it Dec 14 17:10:03 kergoth: are you volunteering to factor out the the 'common' site file? Dec 14 17:11:27 * kergoth hides Dec 14 17:11:58 i'd still like to see a repository of the m4 macros associated with the test results, so we could populate the site file in one fell swoop on a new platform Dec 14 17:12:43 that openmoko thing looks promising Dec 14 17:12:51 * kergoth 's catching up on news/updates Dec 14 17:13:18 kergoth: http://www.openembedded.org/repo/org.openembedded.dev/contrib/site-conf/README Dec 14 17:13:27 kergoth: it's like christmas ;) Dec 14 17:13:50 hah, excellent Dec 14 17:16:47 drat Dec 14 17:16:56 it requires autoconf >=2.60 Dec 14 17:17:27 * mrz80 is away: Metaphysically I'm still here; physically, I'm elsewhere Dec 14 17:17:29 kergoth: hehe, where is your paypal account so I can send you money :) Dec 14 17:18:00 kergoth: only builtins and binutils are complete Dec 14 17:18:10 koen: it does? Dec 14 17:18:44 zecke: yes Dec 14 17:18:53 koen: pastebin the error please Dec 14 17:19:08 zecke: heard of jlibtool? Dec 14 17:19:12 no Dec 14 17:19:22 C libtool implementation thats in the apr source tree Dec 14 17:19:32 sounds scary Dec 14 17:19:33 cuts the apr build time from 1 minute 10 seconds with gnu libtool to 27 seconds with jlibtool Dec 14 17:19:37 same syntax Dec 14 17:20:33 hmm.... sounds good :-) Dec 14 17:21:12 zecke: http://rafb.net/paste/results/AgNLP332.html Dec 14 17:21:14 different .la format, seemingly, so cant mix and match them, and i'm not sure what will need modifying for crosscompilation, but its worth trying in oe Dec 14 17:24:09 zecke : I am splitting the "blacklists" in icecc into 2 parts Dec 14 17:24:31 zecke: should I open a bug and attach the config.cache files? Dec 14 17:24:44 kergoth: so at the end of the year we optimize m4? Dec 14 17:25:03 zecke : one called "section" which allows you to blacklist a whole "section" i.e native,cross,cross-kernel,target Dec 14 17:25:14 koen: oh fix it right away Dec 14 17:25:29 zecke: I'm affraid I have too little time for that atm Dec 14 17:25:48 replace [2.60] with [2.59] in configure.ac Dec 14 17:26:00 zecke: I compiled and installed 2.61 Dec 14 17:26:14 zecke : and a number of package specific ones, as many as the sections. What do you think ? Is it ok or too complicated ? Dec 14 17:27:04 Ifaistos: native,cross,cross-kernel are not sections but okay :) Dec 14 17:27:29 koen: autoconf-native-2.61? Dec 14 17:27:44 * hrw|gone for a moment only Dec 14 17:27:54 hrw|gone: native compile Dec 14 17:27:59 ah Dec 14 17:28:29 yes we should update autoconf to 2.61, run the test suite, add kergoth's sizeof macro and finally make it pass the test suite Dec 14 17:28:40 zecke : You are right. Any other word that could define them ? also where do the kernel fits ? Dec 14 17:30:17 zecke: http://www.openembedded.org/~koen/site-files/armv5te-linux-gnueabi.site Dec 14 17:35:17 zecke: it looks a bit short Dec 14 17:35:30 koen: we run 6 tests Dec 14 17:35:30 * koen <- food Dec 14 17:35:51 the other m4 files are pretty much empty and require the copy and paste Dec 14 17:42:51 Is there any parameter that holds the cross-kernel we are using ? Dec 14 17:43:49 KERNEL_CC_SUFFIX or such Dec 14 17:43:54 Ifaistos: but there is one, I don't know the name Dec 14 17:44:17 kergoth: do we have a proper jlibtool.m4 as replacement for libtool.m4? Dec 14 17:45:43 koen|away: I will send an abstract for another talk at FOSDEM Dec 14 17:46:10 koen|away: Lightning Talk is pretty special as it can attract many different people who are too bored to leave the room Dec 14 17:50:49 Hello Dec 14 17:53:19 hi sirfred Dec 14 17:59:14 zecke : is 'class' better than 'section' ? Dec 14 18:02:05 Ifaistos: anything not section is fine :) Dec 14 18:02:12 Ifaistos: section conflicts with deb/ipk Dec 14 18:03:14 zecke : i'll use 'class' and we see how it goes. Dec 14 18:04:08 zecke : it will be something like ICECC_CLASS_NATIVE_BL Dec 14 18:38:56 re Dec 14 18:39:03 Ifaistos: STAGING_KERNEL_DIR Dec 14 18:42:17 I have a stupid question Dec 14 18:42:25 mccarthy: go Dec 14 18:42:35 I update GPE svn about 2 hours ago and am trying to use a bb file that points to the svn Dec 14 18:42:53 but for some reason when I build it uses the revision *right before* my checking (I think) Dec 14 18:43:01 is says 8878 while the website shows 8879 Dec 14 18:43:04 mccarthy: SRCDATE is set to a date Dec 14 18:43:15 mccarthy: either specify a revision BitBake should download Dec 14 18:43:28 03freyther 07org.oe.dev * rff1c5f34... 10/ (1 contrib/site-conf/configure.ac): Dec 14 18:43:28 contrib/site-conf: No need to require autoconf 2.60 Dec 14 18:43:28 Specially as OE uses autoconf 2.59 itself Dec 14 18:43:31 mccarthy: or set SRCDATE tempoarirly to christmas Dec 14 18:43:45 zecke, cool. how do I set a revision? Dec 14 18:43:50 mccarthy: it takes the repo at last midnight Dec 14 18:43:54 do I need to modify the bb file? Dec 14 18:44:03 koen, good to know Dec 14 18:44:36 set SRCDATE_pn-="now" in local.conf Dec 14 18:44:51 that should trigger the special case in the fetcher Dec 14 18:45:02 mccarthy: or go to the workdir and do 'svn up' Dec 14 18:46:06 mccarthy: http://bitbake.berlios.de/manual/ch03s06.html Dec 14 18:48:41 thanks guys Dec 14 18:49:10 koen, what do you work on now that you are out of familiar territory (pretty punny, huh?0 Dec 14 18:49:14 feedback to the manual is welcome Dec 14 18:49:27 mccarthy: angstrom-linux.org Dec 14 18:49:35 mccarthy: http://www.angstrom-distribution.org/ Dec 14 18:50:02 mccarthy: I am playing with this as well: http://dominion.kabel.utwente.nl/koen/cms/another-new-toy-genisi-efika Dec 14 18:51:44 hrw: koen|away RP: the oe builds moved from a nokia owned machine to a vmware server so build times will increase Dec 14 18:51:55 ok Dec 14 18:52:15 hrw: koen|away RP : but it will be more easy to move these things (even to Xen...) Dec 14 18:52:17 zecke: can you share autoscripts? then I can run some of them on amd64 3500+ (32bit mode) Dec 14 18:52:33 it is still running on a dual xeon though Dec 14 18:52:46 zecke: I have machine at celf for own usage Dec 14 18:52:48 hrw: I posted them once, can post them again Dec 14 18:53:05 zecke: oeml? Dec 14 18:53:11 no Dec 14 18:53:49 zecke: can you resend them to me? thx Dec 14 18:54:00 they are pretty lame though, I will go to a pub now and will mail them tomorrow Dec 14 18:54:28 they need /bin/sh == /bin/bash though Dec 14 18:54:36 no problem Dec 14 18:54:41 I have # there Dec 14 18:55:38 hrw: http://rafb.net/paste/results/sWuFy494.html Dec 14 18:55:52 you need to have a build/MACHINE/conf/local.conf set up Dec 14 18:56:00 ok Dec 14 18:56:05 thx Dec 14 18:56:08 and you need to adjust the path to bitbake Dec 14 18:56:16 and to_build Dec 14 18:56:24 if you want other configurations Dec 14 18:56:27 zecke: Did the underlying OS change? Dec 14 18:56:49 no, it is still edgy eft Dec 14 18:57:09 it might miss a package or two Dec 14 18:57:09 zecke: Wasn't it fedora based? Dec 14 18:57:15 hell no Dec 14 18:57:34 zecke: does this script take care of restarting builds? Dec 14 18:57:41 zecke: I'm just confused them. Its strange poky should have has as many issues on edgy :-/ Dec 14 18:58:12 hrw: it is a crap script Dec 14 18:58:31 hrw: build, build, build -> clean all -> build, build, build -> clean all Dec 14 18:58:38 ok Dec 14 19:08:18 zecke : If you like i can give you access to use the icecc cluster to run them Dec 14 19:08:39 As of next Tuesday, C and C++ will be flushed in favor of COBOL. Please update your programs. Dec 14 19:09:13 chouimat: But I prefer Fortran 77 on punch cards.... Dec 14 19:09:21 hvontres|poodle: hehe Dec 14 19:09:47 * hvontres|poodle never actually USED a punch card, but he has seen them... Dec 14 19:10:02 koen, does angstrom run on the ipaq h5550? Dec 14 19:10:04 * hvontres|poodle did use Fortran 77 in school... Dec 14 19:10:36 hvontres|poodle: my father still has his punch cards for his old programs in FORTRAN IV (iirc) Dec 14 19:10:46 chouimat: Can we use ADD ONE TO COBOL instead ? Dec 14 19:11:32 hvontres|poodle: :) Dec 14 19:18:41 did someone thought about updating gnu-config? Dec 14 19:21:05 mccarthy: it can, but it wouldn't be very usefull, since it requires a 2.6 kernel Dec 14 19:21:42 what is wrong with a 2.6 kernel? Dec 14 19:22:28 not everything works on h5xxx with 2.6 last I heard Dec 14 19:25:28 NOTE: package autoconf-native-2.61-r0: task do_build: completed Dec 14 19:27:06 NOTE: package automake-native-1.9.6-r5: task do_build: completed Dec 14 19:28:01 hrw: emacs is mine, not microemacs ;-) Dec 14 19:28:13 ah. ok Dec 14 19:28:46 i am trying to figure out if the package i am processing is the kernel. To see if for example is a xxxxx-cross i use bb.data.inherits_class("cross"). For the kernel what value should i use ? Dec 14 19:29:08 I tried both "linux" and the machine name but both evaluate to false Dec 14 19:36:27 Ifaistos: if PN == PREF_PROV_virtual/kernel Dec 14 19:36:57 Ifaistos: or 'if recipe inherited kernel.bbclass' Dec 14 19:37:18 well it seems kernel.bbclass is the correct one ;) Dec 14 19:37:47 ;D Dec 14 19:39:17 beware next changesets Dec 14 19:39:29 here be dragons Dec 14 19:40:35 03hrw 07org.oe.dev * rf2601be6... 10/ (14 files in 3 dirs): autoconf: added 2.61 (with DEF_PREF=-1 as not yet tested) Dec 14 19:40:39 03hrw 07org.oe.dev * r4a6c98f5... 10/ (3 files in 2 dirs): automake: added 1.9.6 (DEF_PREF=-1 as not yet tested) Dec 14 19:40:56 both builds Dec 14 19:58:25 meh Dec 14 19:58:59 ~lart exohelpdesk Dec 14 19:59:00 * ibot stabs exohelpdesk Dec 14 19:59:07 ibot: botsnack Dec 14 19:59:08 koen: thanks Dec 14 19:59:58 re Dec 14 20:00:32 Jin^eLD: wb Dec 14 20:04:33 has anyone played with the Philips RC9800i remote control? Dec 14 20:04:53 turned out it runs linux and is easy to hack/put custom firmware onto Dec 14 20:06:28 whats the current oe source mirror/ Dec 14 20:06:32 packetfactory.net is down Dec 14 20:23:50 kergoth: atm we have none Dec 14 20:24:13 kergoth: hrw will get one at CELF and OE should get its own 1U box to host somewhere as well Dec 14 20:26:32 iirc we have 40-70 recipes which are unfetchable Dec 14 20:27:56 some of them block other recipes anyywa Dec 14 20:32:49 ah Dec 14 20:33:10 good to hear celf is being supportive Dec 14 20:33:39 kergoth: what you are doing for work now? Dec 14 20:34:34 after leaving ohand i sat on my ass for about two months looking around for another full time telecommute position, then gave up and sent a resume to the company i used to work for (the wearables for soldiers), and within an hour and half i was rehired Dec 14 20:34:39 so back where i was Dec 14 20:35:22 thx for info Dec 14 20:35:31 ~kergoth Dec 14 20:35:33 from memory, kergoth is Chris Larson, an open source developer currently working for CSC, Inc. in Phoenix, AZ. Core developer on OpenZaurus, OpenEmbedded/BitBake, and TSLib. Often the resident BOFH, and often an ass. Newbies and the clueless beware. Email at clarson@kergoth.com. Dec 14 20:36:23 kergoth: your nickname sounds Klingon, a coincidence or is it really related? Dec 14 20:37:36 coincidence Dec 14 20:37:49 its justa word i pulled out of my ass one day in an attempt to come up with something not already taken Dec 14 20:38:33 ah :) there was a time where I was.. well.. learning Klingon :) Dec 14 20:38:43 hehe. cant say i've ever done that Dec 14 20:39:52 hmm, has anyone encountered an ud2a illegal instruction thing on x86? Dec 14 20:40:18 actually, kergoth speaks fluent Klingon Dec 14 20:40:26 ljp ))) Dec 14 20:40:54 i was waiting on someone translating Opie to Klingon Dec 14 20:40:56 there's enough oddball languages that actually exist without spending time on one thats fictional :P Dec 14 20:41:21 yeah.. in the end I dropped Klingon and started with Suomi Dec 14 20:41:31 60 recipes failed to fetch. some of them blocked other Dec 14 21:03:42 Hello Dec 14 21:04:14 hi Dec 14 21:05:57 Running ping tells me "ping: unknown host www.google.com", and yet I'm in here Dec 14 21:06:00 speaking Dec 14 21:06:07 I cannot connect to any webpage Dec 14 21:06:10 :| Dec 14 21:06:14 :) Dec 14 21:06:40 time to kill machine - bitbake giac Dec 14 21:09:02 Is the internet down or something ;) Dec 14 21:12:28 cu Dec 14 21:17:33 off topic, anyone have any recommendations on books on cryptography? know nothing about it, other than the basics, and am thinking about picking up applied cryptography Dec 14 21:33:56 kergoth: Schneiers books seem pretty good, but I'm not expert. Dec 14 21:35:27 my memories of what i learned about higher math is pretty weak, so i expect i wont understand the implementation details anyway, just need to understand all the concepts :) Dec 14 21:46:15 kergoth, for historical color, check ou t "the codebreakers" tome and also "the ultra secret". Dec 14 21:48:22 hrw|gone: Looks like the pobox guy had to close his website because Sony registered POBox as a trademark. Oh boy! Dec 14 21:48:43 That is absolutely ridiculous. Dec 14 21:51:06 ? Dec 14 22:08:18 urgs Dec 14 22:08:18 http://zaitcev.livejournal.com/108617.html Dec 14 22:35:00 chouimat: qpobox is a Japanese input method in OE that became unfetchable. hrw had asked about it. I contacted the upstream author and got the response above. Dec 14 22:40:23 Laibsch: Was there any alternative location? Dec 14 22:40:42 RP: yes, my HD ;-) and now gakusei.sf.net. Dec 14 22:41:03 Laibsch: cool :) Dec 14 22:41:15 upstream is planning to release it again later anyways. And continue development. Dec 14 22:43:56 Laibsch: That's good to know Dec 14 22:44:13 sure Dec 14 23:34:38 icecc is now handling cross-kernel :) Dec 15 01:16:47 03mreimer 07org.oe.dev * rc50c1fd9... 10/ (5 files in 2 dirs): gstreamer: add the latest 0.10 versions of gstreamer and gst-plugins-*. Dec 15 02:40:11 Laibsc1: ok **** ENDING LOGGING AT Fri Dec 15 02:59:57 2006