**** BEGIN LOGGING AT Mon Feb 19 03:00:00 2007 Feb 19 03:04:04 I added the following to my local.conf file:OPENSLUG_EXTRA_DEPENDS = "python-native" Feb 19 03:04:04 OPENSLUG_EXTRA_RDEPENDS = "python python-serial" Feb 19 03:04:49 I'm not clear on what a perferred provider is yet. Feb 19 03:07:05 OPENSLUG_EXTRA_RDEPENDS should be right according to the openslug distro file Feb 19 03:08:07 one would assume the depends version of that would be correct Feb 19 03:08:12 but I can't see any mention Feb 19 03:08:14 how does it fail? Feb 19 03:12:04 It seems the differences between the new slugos.conf and the openslug.conf are the root cause for the troubles. I can see the problem with the way the EXTRA_RDEPENDs are initialized in the slugos.os vrs the openslug.conf file. The question in my mind is "is there a recomended way of customizing an image such as the slugos-image?" Feb 19 03:12:49 The openslug.conf uses a ?= to define the openslug_extra_rdepends, where the slugos.conf uses an =. Feb 19 03:16:53 a *recommended* way.. that is something you'll need to wait around here for :) Feb 19 03:17:45 Right now I'm trying to add my extra packages to packages/tasks/task-slugos.bb Feb 19 04:02:18 Alrighty! I've managed to waste half a day on my new recipies for oe Feb 19 04:04:06 I have a huge problem however.. I cannot call "install" nor oenote, oedebug nor any other function from within my do_*_append or do_*_prepend function Feb 19 04:04:11 If I do it fails with a syntax error Feb 19 04:17:06 recommended way was just emailed to nslu2 linux list Feb 19 04:17:14 DISTRO_EXTRA_* variables are now used. Feb 19 04:20:40 anyone have any idea why doing *anything* inside an append or prepend function in a recipe would fail with a syntax error? Feb 19 04:29:33 Bah.. I give up. It's broken so I'm just using addtask Feb 19 05:14:02 Ok next issue. What is the recommended method for using pkg-config if any? Feb 19 06:02:30 And here's another nasty one. Is anyone using autotools packages with make files created by kdevelop :) Feb 19 06:18:58 ouch. fixed that Feb 19 06:33:15 morning all Feb 19 06:46:39 03Laibsch 07org.oe.dev * r7c25cded... 10/ (1 packages/grub/grub_0.93.bb packages/grub/grub_0.97.bb): grub: fix SECTION, sanitize, separate out -doc package. Closes 1838. Feb 19 06:46:44 03Laibsch 07org.oe.dev * rc34a7087... 10/ (1 removal.txt): removal.txt: added libmpeg2 Feb 19 06:46:48 03Laibsch 07org.oe.dev * r72dbfb4a... 10/ (10 files in 2 dirs): samba: Make fetching more robust (closes 1885). Unify some more. Sanitize. Feb 19 06:46:53 03Laibsch 07org.oe.dev * r971d18af... 10/ (1 conf/machine/include/collie-2.4.conf): prefer openzaurus-sa kernel and 64-0 layout for collie. Feb 19 06:47:00 03rwhitby 07org.oe.dev * r16171ae5... 10/ (1 conf/machine/nslu2be.conf conf/machine/nslu2le.conf): nslu2{b,l}e.conf: Fixed location of OVERRIDES so that it doesn't clobber the work of tune-thumb. Feb 19 06:47:06 03rwhitby 07org.oe.dev * r6a082250... 10/ (5 files in 5 dirs): slugos: Added support for DISTR_EXTRA_{DEPENDS,RDEPENDS,RRECOMMENDS} Feb 19 06:51:37 ~ping rwhitby Feb 19 06:51:41 pong rwhitby Feb 19 06:51:54 Ifaistos: gu Feb 19 06:52:02 hi ! Feb 19 06:52:05 hi Feb 19 06:52:34 got a minute ? i want to ask you a few things about uclibc in slugos Feb 19 06:53:17 sure Feb 19 06:53:38 (note that it hasn't been supported for a long time, so I may not know the answers) Feb 19 06:55:21 i am building uclibc-0.9.28-svn for ppc and noticed that it ships uclibc headers and headers from other apps in the rootfs Feb 19 06:55:34 have you noticed anything similar ? Feb 19 06:56:47 Ifaistos: I don't build uclibc myself, so I don't know. Feb 19 06:57:44 do you have a uclibc rootfs to check ? Feb 19 06:58:19 i am trying to figure out if this version specific (to 0.9.28) all effects all versions of uclibc Feb 19 06:59:07 here is what i get -> http://www.pastebin.ca/361956 Feb 19 07:11:29 morning koen Feb 19 07:21:44 Ifaistos: talk to saladino in #openslug Feb 19 07:22:30 rwhitby : thanks.will do Feb 19 07:22:46 rwhitby : thanks.will do Feb 19 07:24:35 hi Feb 19 07:25:46 i am trying to build a system with uclibc and for some reason gcc-cross-initial-3.4.4 does not seem to compile with thumb and thumb-interwork support Feb 19 07:26:34 it fails at linking the uclibc after warning that libgcc of gcc3.4.4 does not support interworking Feb 19 07:35:55 morning Feb 19 07:36:02 Laibsch: ping Feb 19 07:39:49 pong Feb 19 07:39:54 Good morning hrw Feb 19 07:40:03 got my mail? Feb 19 07:40:17 Laibsch: which one? Feb 19 07:40:45 I believe I sent you two. Feb 19 07:40:51 * Laibsch checks Feb 19 07:40:52 tosa&kernel is one Feb 19 07:41:09 second one hit spambox Feb 19 07:41:19 bug 605 the other one. Feb 19 07:41:35 Spam? Never! ;-) Feb 19 07:41:54 I'd be interested to know what triggered the spam warning for you? Feb 19 07:42:01 Any .jp IP filter? ;-) Feb 19 07:42:29 X-Spam-Status: No, score=2.2 required=5.0 tests=BAYES_50,BIZ_TLD, DNS_FROM_RFC_ABUSE autolearn=disabled version=3.1.0 Feb 19 07:43:01 Laibsch: so which Z you want to get? tosa or spitz? Feb 19 07:45:07 So, basically .biz and rfc-abuse.org. But that line says, it did NOT hit the spam filter. Feb 19 07:45:30 hrw: Which one is more needed now or likely to be more needed soon? Feb 19 07:46:11 Laibsch: my spamtrap is lower then required 5.0 - too much spam passed in past Feb 19 07:46:54 saladino : rwhitby told me to talk to you regarding uclibc in slugos Feb 19 07:47:22 Laibsch: I prefer to give you spitz, but i you want/can do tosa debugging then tosa Feb 19 07:47:24 saladino : i am building uclibc-0.9.28-svn for ppc and noticed that it ships uclibc headers and headers from other apps in the rootfs Feb 19 07:47:48 saladino : have you noticed something similar in your builds ? Feb 19 07:47:53 Ifaistos i am not able to compile uclibc Feb 19 07:47:54 hrw: spitz sounds nice. I was a bit hesitant since AFAIK there are two tosa but only one spitz. Feb 19 07:48:04 it fails Feb 19 07:48:36 hmm, i must go to work Ifaistos, i'll log from there un about 40 mins Feb 19 07:48:37 hrw: As I said, I cannot see I have the skill set necessary to debug the screen issues that YamaKen reported. Feb 19 07:48:53 saladino : ok. just ping me Feb 19 07:49:10 ok Feb 19 07:49:18 hrw: I think I got him interested in his toy again. He is skilled. Maybe he will fix it. Feb 19 07:51:07 Laibsch: spitz is quite popular so we have devs with it Feb 19 07:51:50 I see. Feb 19 07:52:09 Then it certainly makes more sense to hold on to the tosa Feb 19 07:53:40 * Laibsch is looking forward to 640x480 Feb 19 07:56:50 Laibsch: but if you will write to me in japanese then I will del it from spambox without even checking ;D Feb 19 07:59:43 ;-) Feb 19 08:00:32 Laibsch: ★貴方の様々なニーズにお応えするパートナーをご提供★ Feb 19 08:00:43 Laibsch: one of cjkv spams which I got Feb 19 08:00:51 X-Spam-Status: Yes, score=31.7 required=5.0 tests=BAYES_99,DNS_FROM_RFC_POST, NO_REAL_NAME,RAZOR2_CF_RANGE_51_100,RAZOR2_CF_RANGE_E4_51_100, RAZOR2_CF_RANGE_E8_51_100,RAZOR2_CHECK,SPF_SOFTFAIL,URIBL_AB_SURBL, URIBL_JP_SURBL,URIBL_OB_SURBL,URIBL_SBL,URIBL_SC_SURBL,URIBL_WS_SURBL autolearn=disabled version=3.1.7 Feb 19 08:00:56 hardspam even ;) Feb 19 08:03:17 talks about them offering you a partner to fulfill your needs. Feb 19 08:03:24 me wonders what kind of partner and what kind of needs. Feb 19 08:03:31 ;-) Feb 19 08:03:41 want whole spam to read? Feb 19 08:03:51 No thank you. Feb 19 08:03:53 ;D Feb 19 08:04:00 CIA-20: wake up Feb 19 08:04:01 This one is also coming from the Chinese, really. Feb 19 08:04:05 hi kOEn Feb 19 08:04:21 There Japanese is mostly just as lousy as their German/English/Polish Feb 19 08:04:26 their Feb 19 08:05:47 good morning Feb 19 08:11:28 morning all Feb 19 08:13:36 hi Dirk Feb 19 08:14:18 hey Marcin Feb 19 08:15:11 hrw: How's your new job? Feb 19 08:19:13 good morning all Feb 19 08:19:14 do13_: absorbing Feb 19 08:19:22 hey koen Feb 19 08:19:55 do13_: do you know why TS calibrating on tosa dies? Feb 19 08:21:28 This city lack shops with used computers/parts ;( Feb 19 08:21:44 I need 64/128MB DDR to fill 4th slot onboard Feb 19 08:25:06 hrw: which kernel? Feb 19 08:25:18 do13_: 2.6.17 Feb 19 08:26:03 hrw: ts_calibrate hangs? Feb 19 08:27:05 do13_: did not checked with ts_calibrate but gpe/opie calibrate gets stuck after final touch. reboot req Feb 19 08:27:25 I'll reflash tosa to check Feb 19 08:28:20 hrw: I have seen this with opie. Plain ts_calibrate works Feb 19 08:29:57 hrw: no idea. I haven't recognized this with gpe. Feb 19 08:31:24 ah. ok so it looks like 'yet-another-opie-bug' ;( Feb 19 08:31:26 03hrw 07org.oe.dev * r3a184ece... 10/ (1 classes/insane.bbclass): Feb 19 08:31:26 insane.bbclass: added dict entries for mipsel, i386, x86_64 Feb 19 08:31:26 x86_64 can not be detected properly with this code - needed change: Feb 19 08:31:26 - ELFFile.my_assert(self.data[ELFFile.EI_CLASS], chr(ELFFile.ELFCLASS32)) # only 32 bits Feb 19 08:31:26 + ELFFile.my_assert(self.data[ELFFile.EI_CLASS], chr(ELFFile.ELFCLASS64)) Feb 19 08:31:32 03hrw 07org.oe.dev * rb7774331... 10/ (1 classes/insane.bbclass): insane.bbclass: added ia64, alpha, hppa, m68k, mips, s390, sparc based on Debian packages Feb 19 08:32:07 do13_: anyway for next release I plan to ship /etc/pointercal Feb 19 08:32:43 hrw: workaround :) Feb 19 08:38:38 but good one ;D Feb 19 08:40:28 do13: Sorry for jumping in. are you maintaining mtd-sharp-flash-hack-r0.patch? Feb 19 08:41:16 It does not apply cleanly when compiling angstrom linux-rp-2.6.20-r2 for collie. Feb 19 08:41:28 I can let you know more details in a minute. Feb 19 08:43:33 TypeError: sequence item 1: expected string, tuple found Feb 19 08:44:01 morning hrw. have you ever had a chance to test #1806? Feb 19 08:44:47 hvontres|home: kill me Feb 19 08:44:52 morning Feb 19 08:44:57 hi XorA Feb 19 08:45:24 hi Feb 19 08:45:32 hrw: nah... that would be too harsh. :) Feb 19 08:45:45 ~pester hrw to test #1806 Feb 19 08:45:47 hrw to test #1806: Are we there yet? .. Are we there yet? .. Are we there yet? see http://beverlys.net/LJ/BuggingYou.swf Feb 19 08:46:31 Ifaistos Feb 19 08:46:34 are you there? Feb 19 08:46:38 yep Feb 19 08:48:13 Ifaistos as i told you before, im trying to compile ucslugc Feb 19 08:48:42 and the uclibc linking stage fails due to the lack of thumb and thumb-interwork support in gcc3.3.4 Feb 19 08:48:46 i am trying to solve that Feb 19 08:50:00 i am using 4.1.1 and have compile uclibc 0.9.28-svn for powerpc, x86 (i486), i586 and i686 Feb 19 08:50:19 no arm for the moment though Feb 19 08:50:30 humm Feb 19 08:50:31 ok Feb 19 08:51:38 what i have noticed is that its sipping varius headers files to the rootfs Feb 19 08:51:46 shipping Feb 19 08:52:22 not sure if the problem is with uclibc or one of the apps Feb 19 08:53:20 gettext for example looks like it depends on gettext-dev (strange) which depends on uclibc-dev Feb 19 08:53:32 yep, it is strange Feb 19 08:53:45 when i manage to compile and run it i'll look at that Feb 19 08:56:19 i think zecke is the man for gcc issues Feb 19 08:56:38 but why don't you try a newer version ? Feb 19 08:59:23 hrw: any idea why http://rafb.net/p/9Jzpgo66.html errors out with 'TypeError: sequence item 1: expected string, tuple found' ? Feb 19 08:59:25 Ifaistos i don't know, im a bit newbie with this, but happens that i've got another outdated compilation root dir that compiles correctly with that version Feb 19 09:00:07 so i think that latest changes broke the compilation, something about compile flags i supose Feb 19 09:00:10 koen: are the /snapshots for 0.32 mtn actually produced by a 0.32 mtn or some beta/test version? Feb 19 09:00:23 XorA: 0.32 Feb 19 09:00:35 koen: because my mtn 0.32 doesnt work with them Feb 19 09:01:02 koen: nope Feb 19 09:01:24 XorA: it are snapshots taking from the db the server uses Feb 19 09:01:39 and that's using the static 0.32 build Feb 19 09:02:49 koen: wierd, I was using a static build from venget.net as well, Ill have to retry tonight Feb 19 09:03:08 hrw: I know why Feb 19 09:03:21 hrw: zecke is trying to throw tupples in to a string Feb 19 09:05:48 koen: so when we arrive in Brussels how do we find you in the pub? Feb 19 09:07:04 XorA: by finding the OE crowd? Feb 19 09:07:20 koen: Ive never met them :-) Feb 19 09:08:00 XorA: which time you arrive in Brussels? and which airport Feb 19 09:08:08 hrw: Charleoi at 14:35 Feb 19 09:08:21 XorA: Astrid hotel? Feb 19 09:08:21 XorA: you know lrg, right? Feb 19 09:08:21 hrw: so I figure Brussels at 16:30 Feb 19 09:08:42 hrw: yeah astrid Feb 19 09:10:11 XorA: mickeyl, zecke, dirk, I will be at BRU on 18:30, then travel to astrid and we can take you. Feb 19 09:11:03 I suspect we'll need some food before beer Feb 19 09:11:10 hrw: if Astrid is 15 mins from train station then I shall already have made it there Feb 19 09:11:21 does Astrid have a bar? Feb 19 09:12:36 no Feb 19 09:12:56 bah, I cant prop that up then Feb 19 09:14:58 hmm Feb 19 09:15:13 I'll be at the station somewhere between 17.00 and 18.16 Feb 19 09:18:20 someone could just ring me when they arrive? Feb 19 09:19:22 koen : I think i found out what causes the problem with the failed rootfs when using multimachine Feb 19 09:19:51 koen : it needs the deploy dir for all targets to be the same Feb 19 09:20:31 koen : as the first time it runs it generates the ipkg-collateral and update-rc.d for the the first target Feb 19 09:20:40 XorA: your cell number is? Feb 19 09:21:15 +44 77 250 91 558 Feb 19 09:21:17 koen : but then when you try to run it for a second time, it assumes that the packages allready exist Feb 19 09:22:18 koen : so if you have a deploy dir something like this DEPLOY_DIR = "${TMPDIR}/deploy/${MACHINE}" Feb 19 09:22:32 right Feb 19 09:22:52 multimachince touch DEPLOY_DIR for that reason Feb 19 09:24:58 koen : i see 3 options -> 1) make sure that all machines use same DEPLOY_DIR Feb 19 09:25:28 koen : force the recration of the packages for all targets Feb 19 09:25:48 koen : or case them somewhere and copy them to any new targets Feb 19 09:26:19 4) split packages dir to separated ones: ipk/arm/ ipk/all/ ipk/armv5te/ etc Feb 19 09:26:28 koen : first one is the easiest but too messy IMHO Feb 19 09:27:24 hrw : right Feb 19 09:31:05 good morning Feb 19 09:31:45 hi florian Feb 19 09:40:45 any idea what might be causing this -> http://rafb.net/p/lDTWWD85.html Feb 19 09:42:23 Efika Feb 19 09:42:37 Ifaistos: it should be efika (small chars) Feb 19 09:42:59 Ifaistos: *** Error: Package name contains illegal characters, (other than [a-z0-9.+-]) Feb 19 09:48:05 its exactly as the previous efika kernel .bb Feb 19 09:48:29 anyone knows how to deal with this error "BINFMT_FLAT: reloc outside program 0xff07af7c (0 - 0x163800/0x697e0)" Feb 19 09:48:49 just changes source uri, workdir and added patches Feb 19 09:52:04 hrw: I figured out the errot Feb 19 09:52:33 bb.error("foo %d", bar) vs bb.error("foo %d" % bar) Feb 19 09:53:39 03koen 07org.oe.dev * rbc83c512... 10/ (1 classes/insane.bbclass): Feb 19 09:53:39 insane.bbclass: fixes Feb 19 09:53:39 * change "," to "%" so python doesn't die if an error occurs Feb 19 09:53:39 * spell architecture correctly Feb 19 09:53:39 * add arm/eabi tupple that works for stripped binaries, but fails for -dbg binaries (abiversion 0 vs 97) Feb 19 09:56:23 koen, hrw : any chance you will look into the multimachine issue ? Feb 19 09:56:40 Ifaistos: not before 12 Feb 19 09:57:16 hrw : well according to local time here... this was half our ago :) Feb 19 09:57:38 haha Feb 19 09:57:52 hrw: there's no issue if you don't mess with DEPLOY_DIR Feb 19 09:57:55 ehm Feb 19 09:57:59 s/hrw/ifaistos/ Feb 19 09:58:12 ~lart git for being just some random commands that dont actually seem to have anything to do with source control Feb 19 09:58:12 * ibot chops git in half with a free AOL CD for being just some random commands that dont actually seem to have anything to do with source control Feb 19 09:58:26 XorA: wrong! Feb 19 09:58:36 ~lart Linus for using already taken name Feb 19 09:58:36 * ibot takes out a cattle prod and gives Linus a good jolt for using already taken name Feb 19 09:59:00 hrw: nah, I got the right git, the fact its a pile of polished turd is the main problem Feb 19 09:59:05 Ifaistos: if you use DISTRO="angstrom-2007.1" it will sort them seperate image dirs and leave deploy/ipk intact Feb 19 10:01:19 koen : yes i know, but i think putting all packages of all arches in the same 'pool' will create issues when updating feeds Feb 19 10:02:31 Hi! Feb 19 10:02:49 Ifaistos, did you have the same problem even after rebuild from scratch? Feb 19 10:03:22 FYI, I do mult-mac builds and don't have any issues due to that. I have regular issues due to ipkg & ipkg-utils brokenness Feb 19 10:03:43 but those easy to workaround (I suggested you yesterday one trick) Feb 19 10:03:43 psokolovsky_ : Yes. i posted a few line up what i have found about it Feb 19 10:04:01 Ifaistos, sorry, just turned on my box Feb 19 10:04:21 psokolovsky_ : it seems it needs a comon ipkg dir for all arches build Feb 19 10:04:23 3444 messages since the first one, 2.95 years ago, for an average of 7.5 hours between messages Feb 19 10:04:32 almost 3 years of cia logging for me Feb 19 10:05:04 as the first time it runs it generates the ipkg-collateral and update-rc.d for the the first target Feb 19 10:05:09 Ifaistos, it *has* it (angstrom at least) Feb 19 10:05:13 2401 messages since the first one, 2.77 years ago, for an average of 10.12 hours between messages Feb 19 10:05:27 hrw: http://thomas.apestaart.org/log/?p=442 Feb 19 10:06:12 roxx Feb 19 10:06:12 psokolovsky_ : i making a new distro and thought to have seperate package dir for each arch Feb 19 10:07:06 Ifaistos, again, just FYI, Angstrom's way works mostly well. any issues I saw were in ipkg* stuff. Feb 19 10:09:10 psokolovsky_ : will have to follow its way (angstrom) as i have no time to mess with this in a more detailed manner at the moment Feb 19 10:09:50 so one dir to rule them all :) Feb 19 10:10:12 Ifaistos, I unfortunately also have no time to fix ipkg-utils bugs (and slowness), even though I tried to start on that... Feb 19 10:22:35 2.6.20 defines efika as "Efika" with a capital E Feb 19 10:22:51 2728 messages since the first one, 2.44 years ago, for an average of 7.85 hours between messages Feb 19 10:23:53 Ifaistos: then patch it to not do it Feb 19 10:24:35 hrw : working on it Feb 19 10:30:33 hrw: http://www.openembedded.org/~koen/rootfs-ipk-with-sorted-deploy.patch Feb 19 10:34:32 sort of Feb 19 10:38:57 hrw: should I push it? Feb 19 10:39:59 koen: I do not think that this is finished thing Feb 19 10:40:10 what about splitting packages into subdirs etc? Feb 19 10:46:52 hrw: I sent the patch as an rfc to the list Feb 19 10:55:47 hrw: could you give your thought on that patch? Feb 19 10:55:59 yes and will Feb 19 10:57:45 koen: will you take two ID and fingerprint of your gnupg key to fosdem? Feb 19 10:58:12 hi * Feb 19 10:59:25 i try to build w/ following instructions http://www.openembedded.info/wiki/Getting_started_with_OE Feb 19 11:00:29 but i got error when build libtool-native Feb 19 11:00:47 "source directory already configured; run "make distclean" there first" Feb 19 11:00:49 that one? Feb 19 11:01:02 hrw, yeah, Feb 19 11:01:04 that error Feb 19 11:01:08 19 11:31 < koen> Lazerdye: you didn't read the docs Feb 19 11:01:08 19 11:31 < koen> Lazerdye: especially the section about not having symlinks in the path Feb 19 11:01:16 same problem Feb 19 11:01:48 hrw, oh, thx i will try again Feb 19 11:06:49 koen: your patch is wrong a bit Feb 19 11:08:56 hi Liam Feb 19 11:12:38 hrw: do you have some time for me? Feb 19 11:17:03 no Feb 19 11:25:47 koen: test my version Feb 19 11:30:29 subversion is braindead Feb 19 11:30:52 svn up Feb 19 11:30:56 svn: Failed to add directory 'rt2x00': object of the same name is already scheduled for addition Feb 19 11:36:43 morning Feb 19 11:38:59 hi CoreDump|home Feb 19 11:44:00 koen|away: thanks, they should work with .debug and stripped files Feb 19 11:44:23 koen|away: if the header ist stripped it is not an elf file anymore Feb 19 11:48:30 zecke: which version of gtk is needed for your web wrapper? can you please give me the link to the source? Feb 19 11:48:51 zecke: I hope that recent addons to insane will help Feb 19 11:49:56 hey Marcin, morning all Feb 19 11:50:16 hey mr Girdwood Feb 19 11:51:02 hrw: what kind of IDs? Feb 19 11:54:03 re Feb 19 11:56:23 koen: Feb 19 11:56:36 hrw: Feb 19 11:57:40 koen: I want to check does 0xAE111A91 gnupg key is really yours Feb 19 11:57:56 koen: no one signed that key so how can I be sure? Feb 19 12:14:02 hrw: I get that part Feb 19 12:14:21 I;m just wondering what kind of ID you want, passport,, drivers license, etc Feb 19 12:14:32 koen: any goverment one with photo Feb 19 12:14:41 koen: I'll write to oeml Feb 19 12:15:04 I'll bring my passport Feb 19 12:15:55 thx Feb 19 12:16:36 koen|away: okay, debugging files, that is interesting Feb 19 12:17:19 mr_nice: which wrapper? http://www.openembedded.org/~zecke/git/olpc/sugar/.git/? Feb 19 12:20:56 or not write Feb 19 12:21:55 koen: I will be on gpg keysigning party at sunday. Laibsch wanted to be but have to leave at saturday so we discussed about making small gpgparty in OE team. Feb 19 12:27:32 zecke: yes - thx. I am on searching for a stable browser for xynth. xynth has gtk support so I remembered me that you was writing something about a gtk wrapper for a browser core Feb 19 12:28:20 anyone knows how to deal with this error "BINFMT_FLAT: reloc outside program 0xff07af7c (0 - 0x163800/0x697e0)" Feb 19 12:30:20 03koen 07org.oe.dev * r61b58fb2... 10/ (1 classes/insane.bbclass): insane.bbclass: fix some more ',' vs '%' issues Feb 19 12:30:38 * koen takes the ',' key away from zecke Feb 19 12:34:25 03koen 07org.oe.dev * r4ae4da5e... 10/ (1 packages/esound/esound_0.2.36.bb): esound: sed out insanity in configure.ac Feb 19 12:45:12 hello all Feb 19 12:45:14 likewise: good morning! Feb 19 12:45:15 hey likewise Feb 19 12:45:27 hey koen, Jin^eLD Feb 19 12:46:05 well, I have been up for a few hours already (don't think of me joining carnaval or such) Feb 19 12:46:08 :-) Feb 19 12:46:19 koen : 2.6.20 for efika is working ok:) Feb 19 12:46:28 Ifaistos: cool Feb 19 12:46:59 Hi Ifaistos Feb 19 12:47:03 will push it later in the afternoon Feb 19 12:48:48 likewise : hi Feb 19 12:55:54 Laibsch: http://makotonen.livejournal.com/14747.html Feb 19 13:05:38 die subversion, die Feb 19 13:05:56 hi cbrake Feb 19 13:06:00 cbrake: nice article Feb 19 13:06:10 hrw: hello. Thanks! Feb 19 13:08:37 "load at 30.02. Try again later" Feb 19 13:08:37 gmane must be busy Feb 19 13:21:36 im getting a "BINFMT_FLAT: reloc outside program" error at run-time for several apps cross compiled for ARM....cant get this problem solved.... Feb 19 13:23:57 Sender: which toolchain? Feb 19 13:24:28 well...based on uclinux....but its a proprietary toolchain... Feb 19 13:25:22 Sender: so ask the vendor of this toolchain? Feb 19 13:26:04 that would be a problem...but can you tell me what causes this error..in general...? Feb 19 13:26:31 Sender: proprietary tooclahin's in general cause all kind of weird things ;) Feb 19 13:27:17 Sender: it is because you forgot to add --dio2379spe9=ad28n option to each second gcc invocation Feb 19 13:27:18 yeah...but it changes very littel from the uclinux chain... Feb 19 13:27:22 mickeyl: guten morgen! Feb 19 13:27:28 hi mickeyl Feb 19 13:27:35 hrw: huh? Feb 19 13:27:36 morgen Feb 19 13:27:37 cheers Feb 19 13:28:30 Sender: you asked for help with propertiary toolchain which is a thing we do not support here. Feb 19 13:29:19 its basically a uclinux toolchain with some addons....what would your answer be if i said uclinux...? Feb 19 13:29:48 Sender: I would ask for machine/distro setup and try to forward you to someone who know more Feb 19 13:30:30 machine is arm7 (sigma8622) ...you want local distro? thats fedora core 5 Feb 19 13:30:52 forward me away :) Feb 19 13:31:05 Sender: OE has own definition of machine/distro ;d Feb 19 13:31:32 im not using OE.... Feb 19 13:31:43 Sender: #elinux maybe then? Feb 19 13:31:57 but i figured this is the place to find people with some embedded knowledge... Feb 19 13:32:04 asking there as well.. Feb 19 13:32:36 Sender: tried with free toolchains? Feb 19 13:32:58 is http://openembedded.org/ or is it just my connection? Feb 19 13:32:58 Sender: it can be toolchain related, it can be target machine problems etc Feb 19 13:33:03 aem down I mean Feb 19 13:33:18 hrw....can't....this is work related and the entire development wing uses this toolchain i have to comply with the others... Feb 19 13:34:54 Sender: Might be worth checking to see if your employer has a support contract with the toolchain vendor.. I know at my $EMPLOYER they've got all sorts of arrangements with the embedded compiler devs they deal with Feb 19 13:35:54 NAbyss: got their manual....dont ever read it! Feb 19 13:36:14 Sender: good time to start? :D Feb 19 13:36:19 cu Feb 19 13:36:24 lets attack this differently... Feb 19 13:36:37 how would you approach trying to solve the problem? Feb 19 13:36:51 what would you check...? Feb 19 13:36:53 disable binreloc? Feb 19 13:37:07 koen...cant....mmu less environment.... Feb 19 13:37:14 Sender: What differs between binaries that do and don't cause that error? Feb 19 13:38:09 Compilation options etc Feb 19 13:38:16 good question....apparently nothing....building a project consisting of several binaries to be made....all with same configuration ..few work...most dont Feb 19 13:38:47 rwhitby: I see you added the EXTRA_ stuff? I'll give it a try... Feb 19 13:39:18 Sender: Does the same configuration for the same binary always work? Feb 19 13:41:23 for the ones that work...yes they always work....those that dont...always dont Feb 19 13:42:28 Well, that's the first step, reproducibility.. what differentiates the ones that fail from those that work.. Size? Linked libraries? Feb 19 13:44:07 sized are about the same.... Feb 19 13:44:35 libraries....is there a quick way to check which libraries are used by a binary? Feb 19 13:44:39 ldd Feb 19 13:44:42 rwhitby: I'm ready to try angstrom on my slug: is the Angstrom-bootstrap-image-test-20070218-ixp4xxle.rootfs.jffs2 image compatible with reflash? Feb 19 13:45:56 "not a dynamic executable" obviously... Feb 19 13:46:42 Hmm. Feb 19 13:46:46 ldd wont process an arm-elf flat executable Feb 19 13:46:52 Sender: Any other obvious differences in the apps? Feb 19 13:47:26 not really....both pretty much reference the same libs....but i cant be sure...need to see symbol lists... Feb 19 13:47:46 and relocation table....but i need a tool to do that... Feb 19 13:48:13 since ldd wont do it dor me.. Feb 19 13:48:56 There may be something in binutils that'll do it Feb 19 13:49:18 nm Feb 19 13:49:19 ? Feb 19 13:49:25 No idea Feb 19 13:51:03 anybody knows where ARCH_FLAGS_FOR_TARGET is seted ? Feb 19 13:51:51 * koen hands saladino a copy of grep Feb 19 13:51:59 koen Feb 19 13:52:02 i've greped Feb 19 13:52:10 and i have seen in some files Feb 19 13:52:32 the thing is that i have to root dirs to compile Feb 19 13:52:36 one old Feb 19 13:52:42 and one newer Feb 19 13:52:45 to compile ucslugc Feb 19 13:52:55 the old one compiles gcc with thumb and interwork support Feb 19 13:53:02 and the new one not and uclibc fails Feb 19 13:53:18 koen: http://lists.freedesktop.org/archives/compiz/2007-February/001413.html read that as well? Feb 19 13:53:48 so i've been diffing run.do_compile files and i've seen that in the new one that variable is not seted while in the old ones it is seted to mthumb and mthumb-interwork Feb 19 13:54:12 * zecke stops to work early today Feb 19 13:58:00 koen: what is reported on debug files? Feb 19 13:59:33 zecke: http://rafb.net/p/djGixa26.html Feb 19 14:03:54 koen: okay will build it now Feb 19 14:04:09 zecke: could you pastebin that compiz mail? Feb 19 14:04:17 lists.fd.o is unreachable from here :( Feb 19 14:04:39 from here as well Feb 19 14:04:52 koen: http://rafb.net/p/Hp4z8T86.html just retry Feb 19 14:04:54 CoreDump|home: http://rafb.net/p/Hp4z8T86.html Feb 19 14:04:57 all: http://rafb.net/p/Hp4z8T86.html Feb 19 14:05:05 * zecke feels like Axel Springer Feb 19 14:05:18 monotone.openembedded.org is pingeable but pulls just freeze :\ Feb 19 14:05:25 zecke: thanks Feb 19 14:05:35 CoreDump|home: so use a mirror Feb 19 14:06:14 I take it that is considered normal then? Feb 19 14:06:43 koen: http://svn.o-hand.com/view/poky/trunk/meta/classes/insane.bbclass?rev=1195&r1=1187&r2=1195 is that sane? Feb 19 14:07:29 zecke: that would improve it a lot Feb 19 14:13:45 koen: did you push the changes for multimachine? Feb 19 14:14:01 Ifaistos: what changes? Feb 19 14:15:02 koen: i was reading the email you send regarding the rfc Feb 19 14:17:03 no, I'm waiting for some more feedback Feb 19 14:21:05 ~ping me Feb 19 14:21:08 pong ifaistos_ppc Feb 19 14:25:12 greetings Feb 19 14:25:43 is in bitbake any option to run -c clean after every package build? Feb 19 14:29:53 03koen 07org.oe.dev * r7eec4d13... 10/ (1 packages/matchbox2/matchbox-panel-2_svn.bb): mb-panel-2: rreplace mb-panel-1 Feb 19 14:34:19 re Feb 19 14:34:27 hrw: wb Feb 19 14:41:31 sck... gnome-vfs has badness Feb 19 14:41:35 c-c badness Feb 19 14:43:42 hrw: yes, the configure.ac add /usr/include as convience to find openssl/samba Feb 19 14:43:55 damn, I should learn to spell again... Feb 19 14:44:20 convenience Feb 19 14:48:42 is there any planned meeting of the OE team at fosdem ? Feb 19 14:49:02 at the beer event? Feb 19 14:49:34 on friday? Feb 19 14:49:40 ifaistos_ppc: I think same procesure as every/last year. Hang around together Feb 19 14:51:01 i am thinking of join you, but maybe for a day Feb 19 14:51:05 I'll be touroning on Friday, should be there for the beer event Feb 19 14:51:41 I will be at ~15h at the airport and will stay/hack there until mickeyl's and dirk's flight arrives Feb 19 14:52:04 I would like you all to know, the temperature here has gone above 0 (C) before noon first the first time in a month Feb 19 14:52:21 hehe Feb 19 14:53:05 We have just had a month of "global cooling" Feb 19 14:53:08 here is North Pole? :) Feb 19 14:53:16 January was "global warming" Feb 19 14:53:31 Last night a friends glass of wine froze by the hot tub Feb 19 14:54:56 someone is running enviromental strenght tests :) Feb 19 14:56:21 i trying to figure out what would be the best day to come Feb 19 15:09:14 Is there a way to install an ipk with arguments that are passed to the postinst script (and obviously) a way to process such arguments within pkg_postinst() ? Feb 19 15:19:56 hillct: not that I know of Feb 19 15:20:14 hillct: and it sounds stupid :) Feb 19 15:20:29 is in bitbake any option to run -c clean after every package build? Feb 19 15:20:54 Sleep_Walker: not in bitbake, there is a hack that works to some degree Feb 19 15:21:03 zecke: for example ipkg install mypackage --with-extracoolness=true Feb 19 15:21:05 Sleep_Walker: INHERIT += "rm_work" into your local.conf Feb 19 15:21:22 zecke: thx, for this hint Feb 19 15:21:37 vim /etc/default/mypakage extracoolness=true Feb 19 15:22:41 the idea being, the arg would permit firing off some additional postinst step for advanced users, or suitable for installation at distro setup time but not after initial installation Feb 19 15:23:34 zecke: so you're thinking that postinst should just source /etc/default/mypackage and get the setting from there Feb 19 15:23:53 perhaps that is a better approach Feb 19 15:24:11 hillct: no, I do not believe this is part of postinst Feb 19 15:24:34 well, I could easily enough make it part of postinst Feb 19 15:24:46 and perhaps that would do the trick Feb 19 15:24:52 hillct: I can kill kittens, normally I don't Feb 19 15:25:06 why not? It's fun Feb 19 15:25:17 I see your point though Feb 19 15:25:25 hillct: The issue with uebercool users is, they change their opinion very often Feb 19 15:25:41 hillct: so to change this setting they would have to remove and reinstall the package? that does not sound cool at all Feb 19 15:26:01 hillct: man kind had an answer: Separation: Let us separate install and configuration Feb 19 15:26:13 hillct: install once, configure everytime you want to Feb 19 15:26:36 well no. This is not for routine setting changes. this would be for firing off a config script as part of the installation Feb 19 15:26:37 hillct: and god said, use human readable configuration files Feb 19 15:26:45 the script would be there after install too Feb 19 15:27:12 ok. I see your point Feb 19 15:27:25 it doesn't apply in this case, but i see your point ;) Feb 19 15:27:40 hillct: don't do that, you break automatic installations Feb 19 15:28:01 ok Feb 19 15:30:38 03tmbinc 07org.oe.dreambox * r1d2f5240... 10/ (1 conf/machine/dm600pvr.conf): conf/machine/dm600pvr.conf: don't rely on recent kernel, generate NFI1 Feb 19 15:30:42 03tmbinc 07org.oe.dreambox * recd47f97... 10/ (1 packages/base-files/base-files/filesystems): base-files: give ext3 priority over ext2 via /etc/filesystems Feb 19 15:30:45 03tmbinc 07org.oe.dreambox * rc2438035... 10/ (1 packages/dreambox/dreambox-dvb-modules.bb): dreambox-dvb-modules/dm600pvr: update drivers to 20070219 Feb 19 15:30:49 03tmbinc 07org.oe.dreambox * r3d4da889... 10/ (1 packages/dreambox/dreambox-secondstage.bb): dreambox-secondstage/dm600pvr: update to v49 Feb 19 15:30:53 03tmbinc 07org.oe.dreambox * rb421884c... 10/ (1 packages/enigma/enigma/opendreambox/userbouquet.33fc5.tv): enigma: fixup default bouquet Feb 19 15:30:56 03tmbinc 07org.oe.dreambox * r41583d0c... 10/ (1 packages/enigma/enigma_cvs.bb): enigma: use proper defaults for dm600pvr Feb 19 15:33:52 pb_: RP: koen : Tough question should staged .la files of -native have installed=yes? Feb 19 15:35:08 zecke: I arrive 5 minutes later then mickeyl and dirk Feb 19 15:36:06 hrw: this would make it acceptable to wait for you as well Feb 19 15:37:14 zecke: how do you travel from airport -> hotel? Feb 19 15:37:30 we're going to use the good'ole'train Feb 19 15:37:42 koen: hehe, I assume we will use the train, or organize a taxi ;) Feb 19 15:37:48 Dirk and me found it last time Feb 19 15:37:53 i think we can do it this time again Feb 19 15:37:53 :) Feb 19 15:37:54 koen: could you do me a favor and run "dpkg -S Automake/Struct.pm" on your system? Feb 19 15:38:21 koen: answer the question above (libgettext is broken badly, libtool itself too) Feb 19 15:38:36 * Crofton|home has been memorizing the map of Brussels Feb 19 15:39:02 koen@bitbake:/data/build/koen/OE/build/tmp/angstrom/work/armv5te-angstrom-linux-gnueabi/eds-dbus-1.4.0+svn20070215-r5/install$ dpkg -S Automake/Struct.pm Feb 19 15:39:02 automake1.9: /usr/share/automake-1.9/Automake/Struct.pm Feb 19 15:39:28 koen: thanks! Feb 19 15:39:42 automake1.9: /usr/share/automake-1.9/Automake/Struct.pm Feb 19 15:39:42 automake: /usr/share/automake-1.10/Automake/Struct.pm Feb 19 15:39:46 here Feb 19 15:40:19 sigh Feb 19 15:40:26 still fails wtf Feb 19 15:40:54 CoreDump|home: how? Feb 19 15:41:13 1.10? Feb 19 15:41:26 http://rafb.net/p/mg7HpJ25.html Feb 19 15:41:30 zecke: 1.9 Feb 19 15:41:43 it's a slugos build fwiw Feb 19 15:42:56 * CoreDump|home bitbakes automake-native Feb 19 15:48:37 koen: I wonder if I should call hostname in insanity and if I see mvista, montavista, timesys refuse to work? Feb 19 15:49:44 zecke: would be a nice touch =) Feb 19 15:57:25 heh Feb 19 15:58:34 hmm, we don't "like" timesys? Feb 19 16:12:35 has anybody (else) run a "mtn db check" on the latest snapshot? I get a "file integrity error", but despite that it seems to work okay. Is this a problem? I'm double checking with a clean download (and md5) now. Feb 19 16:14:19 that should be "file structure is corrupted" Feb 19 16:15:22 HopsNBarley: I'm getting the same w/ an up-to-date DB Feb 19 16:16:25 lemme ask about ramifications over in #monotone... Feb 19 16:16:39 thanks for the sanity check. Feb 19 16:18:34 HopsNBarley: this sounds like an sqlite error (pragma integrity) Feb 19 16:21:01 hrw: where are angstroms hardware tasks* hiding? Aren't there any? Feb 19 16:21:17 packages/angstrom/ Feb 19 16:21:49 hmm Feb 19 16:23:27 insane.bbclass is not enabled by default? Feb 19 16:23:34 any1 here using rs232 port on omap 5912 ? Feb 19 16:23:47 putumutukas: ask Crofton ;) Feb 19 16:24:07 heh Feb 19 16:24:10 he already has :) Feb 19 16:24:15 :D Feb 19 16:24:23 I do not know the answer Feb 19 16:24:30 hi Feb 19 16:24:33 I hope it is a generic question :) Feb 19 16:25:12 i was just wondering how no1 has faced that problem before Feb 19 16:25:58 I'm not sure how many people actually read/write the serial ports from OE-derived distros Feb 19 16:26:16 I'll bbiab and take a quick look at my osk Feb 19 16:26:26 although I have a bunch to do before leave for FOSDEM Feb 19 16:26:38 Latest snapshot (2/19) image has this error as well. Feb 19 16:28:30 thnx- i cannot await that you to spend the precious time on my problem but i would be very grateful if you'd do spend some :P Feb 19 16:30:05 putumutukas: what is a problem at all? Feb 19 16:30:34 i get empty chars when i read from ttyS0 Feb 19 16:30:53 it sends out via ttyS0 fine but doesn't receive anything :S Feb 19 16:31:26 putumutukas: checked cable and signal voltage level? Feb 19 16:31:44 yes Feb 19 16:31:50 works fine on pc-s Feb 19 16:32:10 you connect simple wiresonly cable omap5912osk<>pc?? Feb 19 16:32:26 yes Feb 19 16:32:29 null modem cable Feb 19 16:32:52 putumutukas: you are sure that omap5912osk board has 12V rs232 and not 5V one? Feb 19 16:33:01 is there any tool for making bb files? Feb 19 16:33:08 03florian 07org.oe.dev * re9ca3ba6... 10/ (1 packages/xserver-common/xserver-common_1.14.bb): xserver-common: Make 0.14 default again after uploading the source package. Feb 19 16:33:13 03florian 07org.oe.dev * r8d32cbe0... 10/ (3 files in 2 dirs): gpe-conf: remove 0.2.2, add 0.2.4 Feb 19 16:33:35 jrn_: $EDITOR Feb 19 16:33:50 but if i omap would have 12V and would not work on 5V then it would be impossible to use the console ;) Feb 19 16:34:14 ^^ Feb 19 16:34:16 hmm Feb 19 16:34:43 putumutukas: so other question.. does serial console in u-boot works? Feb 19 16:34:48 yes Feb 19 16:34:56 my problem is that i actually do not have the slightest idea, what this all means ^^ Feb 19 16:35:12 (...that i have to write in a bbfile) Feb 19 16:35:35 putumutukas: both directions? Feb 19 16:35:37 03freyther 07org.oe.dev * r49cb44f6... 10/ (1 classes/insane.bbclass): classes/insane.bbclass: Remove whitespace Feb 19 16:35:41 03freyther 07org.oe.dev * rdba464b9... 10/ (1 classes/insane.bbclass): Feb 19 16:35:41 classes/insane.bbclass: Make the .so-check check for symlinks (merge poky r1258,r1195) Feb 19 16:35:41 The .so check now looks for symlinks ending with .so. These are a typical Feb 19 16:35:41 trait of -dev packages and kills a lot of false positives. Feb 19 16:35:41 This was authored by ross and was taken from the poky svn repository. Feb 19 16:35:49 can't i just tell bitbake, where the source code is that has to be compiled and so it does? Feb 19 16:35:50 03freyther 07org.oe.dev * r44766714... 10/ (1 classes/insane.bbclass): (log message trimmed) Feb 19 16:35:50 classes/insane.bbclass: check stages .pc and .la for sanity Feb 19 16:35:50 native ones: Feb 19 16:35:50 .la files should have installed=yes (at least I believe Feb 19 16:35:51 koen: it would be nice if you could send me two EABI files one binary, one debg one Feb 19 16:35:52 this). And somehow some believe this as well, some don't though... Feb 19 16:35:53 some like fakeroot love to install two .la files... Feb 19 16:35:57 non-native: Feb 19 16:36:35 it should work both ways, otherways i wouldn't be able to configure the console Feb 19 16:37:14 putumutukas: 2.6.19-omap1 kernel? Feb 19 16:37:45 yes Feb 19 16:37:50 from http://ossie.mprg.org/download/osk/ Feb 19 16:38:05 putumutukas: ah.. Croftons kernels Feb 19 16:38:16 yap :) Feb 19 16:38:27 yeah Feb 19 16:38:43 image sort of follows boostrap-image Feb 19 16:38:56 since by knowledge on that area is very fuzzy i try to understand what others have done :) Feb 19 16:39:02 console works fine, but I am concerned something stays attached to serial port Feb 19 16:39:54 something grab-s the data before it reaches ttyS0 Feb 19 16:40:29 03tmbinc 07org.oe.dreambox * rcc95dd06... 10/ (1 packages/enigma/enigma_cvs.bb): enigma: update to 20070220 Feb 19 16:43:51 later, going home now Feb 19 16:46:25 cu Feb 19 16:51:59 03freyther 07org.oe.dev * r65003f86... 10/ (1 classes/insane.bbclass): classes/insane.bbclass: missing import os was added in the .so check Feb 19 16:57:53 putumutukas, did you comment out the line that starts getty on ttyS0 and reboot? Feb 19 16:58:47 Zero_Chaos: Feb 19 16:58:58 grr - not being able to build a cross compiler is really cramping my style Feb 19 16:59:05 ~botmail for zecke: http://ewi546.ewi.utwente.nl/tmp/zecke/ Feb 19 17:12:19 hi Feb 19 17:14:47 uhm, if I want to build angstrom - why do I need qemu? Feb 19 17:15:29 * Jin^eLD reads the distro page Feb 19 17:15:43 hi * Feb 19 17:15:57 Jin^eLD: glibc locale generation Feb 19 17:16:08 how do i told bitbake re-run a specify task (ex : task 100) Feb 19 17:17:00 koen: I am looking for an EABI system, gcc 4.1.1, glibc 2.5 (that would match the vendors toolchain, for arm), would you suggst I take angstrom as my "base" distro to crosscompile some packages? Feb 19 17:17:58 yes Feb 19 17:18:01 btw quemu configure fails with ERROR: "gcc" looks like gcc 4.x - which is true indeed - I do have gcc 4.x on my system Feb 19 17:19:33 Jin^eLD: apt-get install gcc-3.4 Feb 19 17:20:28 03koen 07org.oe.dev * reeb59395... 10/ (1 packages/tasks/task-openmoko.bb): task-openmoko: fall back to matchbox-panel-1 since the default session script still uses version 1 instead of version 2 Feb 19 17:20:28 ok, that means I'm screwed :) let's see if fedora will allow me to install the old gcc additionally Feb 19 17:21:13 Jin^eLD: have your tried the fedora quemu package? Feb 19 17:21:42 hvontres|zzZZzz: doesn't work Feb 19 17:21:45 actually no, I haven't.. I just setup for angstrom build and started building Feb 19 17:21:49 oh Feb 19 17:21:54 hvontres|zzZZzz: qemu needs patches to be actually usefull Feb 19 17:23:48 koen: ahh, ok. I didn't realize that...:( I just have been running with binary locale genereation turned off for now Feb 19 17:24:02 hmm ok, as I thought... can't intsall an additonal rpm Feb 19 17:24:13 conflicts with new stuff Feb 19 17:24:32 Jin^eLD: you can turn locale generation off like hvontres|home said, if you have a sucky RPM distro Feb 19 17:24:55 indeed.. Feb 19 17:24:59 Jin^eLD: which version of fedora are you on? Feb 19 17:25:11 6 Feb 19 17:25:39 koen: so basically we need qemu with all of the patches in qemu-native.bb? Feb 19 17:26:49 Jin^eLD: have you tried the gcc-compat rpm's? Feb 19 17:27:11 no, I guess I missed them Feb 19 17:27:14 let me check Feb 19 17:27:27 florian, psst, we have started populating #tinymail on gimpnet :) Feb 19 17:28:03 hvontres|home: those are not in the official repos, right? Feb 19 17:28:41 pvanhoof: ok, i should register a nick there and configure my client a little bit. Feb 19 17:28:51 Jin^eLD: I thought they were on the install CD, I am still waiting on yumex to load :( Feb 19 17:29:04 florian, owk Feb 19 17:29:06 yum search gcc-compat did not give any results Feb 19 17:29:10 Jin^eLD: slow build machine Feb 19 17:29:24 florian, feel free to invite-over the KC hackers that like tnyml :) Feb 19 17:29:27 hmm Feb 19 17:30:17 i'm pretty sure nomis uses gimpnet frequently :) Feb 19 17:30:45 hvontres|home: well I do not see anything useful on rpmfind and yum did not tell me much either Feb 19 17:31:13 aah Feb 19 17:31:18 I think I have something now Feb 19 17:31:22 compat-gcc-34 Feb 19 17:31:29 I was looking for gcc-compat Feb 19 17:31:56 yum, thats it Feb 19 17:32:07 aem, yes I mean, not yum :) Feb 19 17:32:30 Jin^eLD: yup. Not sure if that's gonna help or not... but it could be worth a shot Feb 19 17:33:00 well, since my caching with disabled locale generation just finished, I guess I will proceed without qemu Feb 19 17:33:18 I anyway only need to crosscompile a couple of packages for EABI Feb 19 18:18:16 hi mickeyl Feb 19 18:18:52 Hello Feb 19 18:21:56 hey sirfred Feb 19 18:23:31 koen: Hi Feb 19 18:25:33 hi woglinde Feb 19 18:29:38 woglinde: hi Feb 19 18:32:24 hi pb hi mr_nice Feb 19 18:32:55 mr_nich I will flash my simpad and see if it runs Feb 19 18:33:23 woglinde: :) Feb 19 18:40:19 03koen 07org.oe.dev * r4a494696... 10/ (1 conf/distro/include/angstrom.inc): angstrom: inherit the new and improved insane.bbclass by default so everyone shares and enjoys our QA efforts Feb 19 18:40:44 woglinde: do you remember your mtn password? Feb 19 18:41:04 mr_nice no Feb 19 18:43:11 woglinde: :( Feb 19 19:12:28 hi, can anybody tell me the difference between flite-, flitedyn- and flitedyn16- output in opie-reader? Feb 19 19:13:48 Has linux-rp 2.6.20 support for MMC card on zaurus (husky)? Feb 19 19:14:27 sirfred: yes Feb 19 19:15:16 koen: ok. I'm just trying to restart with a 2.6.20 using kexec, but my SD didn't mount. Sure I've messed something Feb 19 19:16:51 sirfred: SD detection is now a kernel thread, so you might need to pass 'rootdelay=5' to the kernel cmdline Feb 19 19:17:12 koen: I think I'm having some trouble with udev and /dev entries. Feb 19 19:17:59 koen: I've created an ext2 rootfs on a CF card copying what I have in my jffs2 root partition Feb 19 19:18:11 ah Feb 19 19:18:19 did you create dev nodes as well? Feb 19 19:18:26 you need a few Feb 19 19:18:27 I copied them using tar Feb 19 19:19:06 It should be able to replicate the dev entries in the target filesystem, but I'm getting some errors about /dev/.udev already existing or something so. Feb 19 19:19:07 sirfred: you need these: http://www.openembedded.org/repo/org.openembedded.dev/files/device_table-minimal.txt Feb 19 19:19:16 koen: Thanks. Let's see Feb 19 19:19:35 sirfred: remove the /dev.udev, it should be created on the fly Feb 19 19:19:48 koen: And when should it be deleted? Feb 19 19:19:53 koen: During shutdown? Feb 19 19:20:10 something like that Feb 19 19:20:23 koen: Perhaps I need to delete it manually before running kexec, so. Feb 19 19:20:40 udev does some scary stuff with bindmounts and ramdisks Feb 19 19:21:09 Hmmm, I'm not reaching oe.org Feb 19 19:21:45 It seems to be badly loaded Feb 19 19:22:54 koen: can I ask you a off topic question? Feb 19 19:23:07 sure Feb 19 19:26:58 koen: I am trying to build a toolchain out of the oe logs. but I am a bit puzzeled on the glibc task. I get a "compiler support for __thread is required". I have builde the binutils and the stage1 gcc with the same options as in the logs. after that I unpacked the glibc-ports to glibc and tryed to use the options used in glibc-intermediate Feb 19 19:27:29 logs? Feb 19 19:28:54 koen: I suppose that kernel thread is kmmcd, isn't it? Feb 19 19:29:16 koen: e.g. tmp/work/arm-oabi-angstrom-linux/gcc-cross-4.1.1-r10/temp/log.do_configure.2885 Feb 19 19:30:35 hi, all! Feb 19 19:30:58 koen, how to make insane.bbclass to emit warnings instead of erors? Feb 19 19:34:24 Strange, I have that [kmmcd] running, there's a /sys/bus/mmc directory, but no entries in /sys/bus/mmc/devices Feb 19 19:34:46 I wonder if starting with kexec could be related with this Feb 19 19:36:04 slapin_nb: you mean "how can I fix those errors?", right? Feb 19 19:36:32 koen, I'm not maintainer of these packages, so I can't fix them. Feb 19 19:36:40 that's not true Feb 19 19:36:58 I don't maintain a shitload of packages and I still fix them Feb 19 19:37:32 To add a new .bb to OE by non-developers like me, the procedure is to open a bug, isn't it? Feb 19 19:38:10 IIRC any change requires that Feb 19 19:39:49 hey people! Feb 19 19:39:55 koen|away: thanks Feb 19 19:40:21 koen|away: no idea why objdump changes the osabi from 0 (SYS V) to 97 (ARM) Feb 19 19:42:16 koen|away: http://rafb.net/p/8nVrGM73.html, probably it makes not too much sense to check the osabi at all Feb 19 19:46:29 zecke, folks are screaming after recent insane.bbclass changes ;-) Feb 19 19:48:02 psokolovsky_: hehe, thanks to my sith power I knew this in advance and wrote a nice log message Feb 19 19:50:43 zecke: could you send a mail to the oe list as well? Feb 19 19:51:24 koen: is insane.bbclass enabled by default (it isn't hopefully) Feb 19 19:51:36 zecke: in angstrom it is Feb 19 19:52:02 that is insane Feb 19 19:52:24 what should be the content? What insane gained, or why stuff is broken ATM? Feb 19 19:52:27 zecke: http://www.openembedded.org/bonsai/view/rev/24813/ Feb 19 19:52:49 zecke: what insane gained and people should help to fix the problems Feb 19 19:53:10 I still have the .la file question for -native packages Feb 19 19:56:25 03ifaistos 07org.oe.dev * r86785386... 10/ (36 files in 3 dirs): /packages/linux: Add kernel 2.6.20 for efika Feb 19 19:57:09 off topic : some info for meet on friday ? Feb 19 19:58:54 gremlin[it]: what do you need to know about friday (guessing you mean the FOSDEM beer event) Feb 19 20:00:43 philippe exactly :) ... just how to meet .. we exchange cell numbers ? Feb 19 20:02:19 gremlin[it]: Well I do not know exactly what my schedule will be. But usually it is go to the pub, move upstairs where all the noise is. Feb 19 20:02:36 Grab beer and find Koen :-) Feb 19 20:03:31 ok ... easy and straight :) Feb 19 20:05:38 gremlin[it]: and otherwise you just randomly make friends :-) Feb 19 20:06:30 mhhh true ... how many woman developer there are ?!?!?! *LOL* Feb 19 20:08:00 gremlin[it]: there should be a few. But the *OMG* woman hacker puts them off... Feb 19 20:09:03 OMG ? Feb 19 20:09:35 gremlin[it]: Oh My God??? Feb 19 20:10:06 HIHIIH :P Feb 19 20:10:24 gremlin[it]: just get to the place and search for wookey and someone looking worse than wookey (me) :} Feb 19 20:10:26 'morning' all Feb 19 20:10:42 evening already ;) Feb 19 20:10:48 RP: morning:) Feb 19 20:10:56 gremlin[it]: just make sure you don't mistake Alan Cox for zecke then :-) Feb 19 20:12:16 hehe Feb 19 20:13:03 search for some one worse than Wookey and Alan Cox ;) Feb 19 20:13:13 any1 here who's familiar with omap5912 rs232 port ? Feb 19 20:13:19 slimmer than rms though Feb 19 20:13:41 * philippe feels he competes with zecke in the looks section now :-) Feb 19 20:13:47 hehe Feb 19 20:13:58 theu look the same ??? both with a red fedora on top ??? :P Feb 19 20:14:24 gremlin[it]: no fedora in my case... Feb 19 20:14:25 gremlin[it]: I lost my hat in a plane last year, didn't manage to get a new one yet (well too lazy to search one) Feb 19 20:14:36 gremlin[it]: mine wasn't red though Feb 19 20:14:59 hahah but will be A.Cox at fosdem ? Feb 19 20:15:15 03ifaistos 07org.oe.dev * re87859c3... 10/ (3 files in 3 dirs): Feb 19 20:15:15 /packages/uclibc/uclibc-0.9.28: Add config for i486 Feb 19 20:15:15 Since the default now for x86 target is arch=i486 this is need for uclibc to build Feb 19 20:15:19 03ifaistos 07org.oe.dev * r66d74689... 10/ (1 packages/uclibc/uclibc_0.9.28.bb): uclibc_0.9.28.bb: Bump PR Feb 19 20:15:23 03ifaistos 07org.oe.dev * rbc13ceb9... 10/ (4 files in 4 dirs): Feb 19 20:15:23 /packages/uclibc/uclibc-svn: Add config for i486 Feb 19 20:15:23 Since the default now for x86 target is arch=i486 this is need for uclibc to build Feb 19 20:15:23 Bump PR Feb 19 20:15:32 I used one when i was younger ... a black one ... Feb 19 20:15:46 RP: Hello? Feb 19 20:15:57 gremlin[it]: I think Alan was there last year. Feb 19 20:16:08 ~seen RP Feb 19 20:16:28 rp is currently on #gpe #oe #openzaurus #handhelds.org. Has said a total of 1 messages. Is idling for 6m 3s, last said: ''morning' all'. Feb 19 20:17:38 hi sirfred Feb 19 20:18:05 btw .. i'm not young, small, round and wear black :) Feb 19 20:18:09 RP: Hi. I was thinking about trying myself to change that bits in w100fb Feb 19 20:18:16 I've seen Alan once in London but never managed to meet him... Feb 19 20:18:24 sirfred: go for it :) Feb 19 20:18:34 RP: OK. Feb 19 20:19:15 RP: http://www.openembedded.org/~zecke/Web-Site/Old%20Events_files/alan.jpg Feb 19 20:19:28 very, very, very old pic Feb 19 20:19:35 RP: What do you think about changing the way to store the init_mode in w100fb_mach_info Feb 19 20:19:48 RP: I think that perhaps we could use directly the FB_ROTATE_* defines. Feb 19 20:20:03 RP: Instead of INIT_MODE_ROTATED + INIT_MODE_FLIPPED Feb 19 20:20:31 sirfred: Yes, probably if we're using the rotate field properly Feb 19 20:20:50 zecke: You do look a bit younger in that :) Feb 19 20:20:57 hehe Feb 19 20:21:09 RP: That's one of the goals. ;-) Feb 19 20:21:19 which one u are zecke ? Feb 19 20:21:29 gremlin[it]: 2nd from the left Feb 19 20:21:39 gremlin[it]: or the one with glasses Feb 19 20:21:58 RP: I think this is normal, this was before I was dead :) Feb 19 20:22:05 Why only Alan has beer? Feb 19 20:22:15 Or are you hiding yours? Feb 19 20:22:16 i think we need a "hall of fame" with pictures from OE developers :) Feb 19 20:22:22 sirfred: hehe, don't remember Feb 19 20:22:28 zecke: How long ago was it? Feb 19 20:22:30 zecke: :) Feb 19 20:22:37 RP: http://www.openembedded.org/~zecke/Web-Site/Old%20Events_files/hall.jpg Feb 19 20:23:03 RP: I think this was 2001 and I was in high school at this time, so it was before useles uni assignments stole my nights... Feb 19 20:23:05 koen, the package-index script, does it create a PACKAGE file based on the files inside the current directory? Feb 19 20:23:35 BTW: this was old Team Opie Germany Feb 19 20:23:51 koen, Im asking cause I want to make multiple feeds, divided by interest. So if it simply reads what is inside that folder I can put all wanted packages there and run package-index Feb 19 20:24:10 zecke looks a lot like a well-known kernel hacker.... Feb 19 20:24:32 zecke: You're making me feel old ;-) Feb 19 20:26:30 RP: I better don't mention I feel old Feb 19 20:28:31 zecke: We're both too young to feel like this ;-) Feb 19 20:29:05 yeah, actually I feel quite well (most of the time) Feb 19 20:31:58 03kristoffer 07org.oe.dev * r4c745bba... 10/ (1 packages/initscripts/initscripts-1.0/jornada6xx/devices): Feb 19 20:31:58 initscripts/initscripts-1.0/jornada6xx/devices : Cleanups & Additions Feb 19 20:31:58 * Addition of /dev/ttySC0 - 2 + /dev/irda0 (with correct maj/min) Feb 19 20:31:58 * Removal of ttySAx Feb 19 20:37:38 first "victims" of the new insane.bbclass -> http://rafb.net/p/9tMoQv14.html Feb 19 20:38:50 Ifaistos: this is ironic. libtool's .la file is not sane Feb 19 20:39:50 zecke : ironic or not the build fails... Feb 19 20:42:50 g'night all Feb 19 20:43:57 Kristoffer: it uses the dir you pass it Feb 19 20:44:08 Kristoffer: although I suspect you pass it "." as dir Feb 19 20:46:25 Ifaistos: installed=yes and no references to $TMPDIR/work Feb 19 20:47:23 Ifaistos: either help fixing these bugs, fix them by hand, or disable fatal errors for 6 Feb 19 20:49:19 zecke : that's why i am testing insane :) Feb 19 20:52:59 koen, oki, thx Feb 19 20:53:45 zecke : its a good thing that QA will become part of the developing cycle. will take sometime for people to adjust Feb 19 20:54:38 Ifaistos: fail fast (CROSS COMPILE Badness) was the first step, this is the 2nd, and the final one will be writing notes into a notebook and I will review them weekly Feb 19 20:57:00 Hopefully, I just cleaned some hard coded paths out of some of my configure files ... Feb 19 20:57:34 Ifaistos: fixing .la files is tricky, I don't know what influences them. 20.000 lines of shell code scares be somehow Feb 19 20:59:41 zecke : could testing on them be a bit more "relaxed". just report not fail Feb 19 21:00:00 Ifaistos, of course, then no one is motivated to fix ... Feb 19 21:00:00 Ifaistos: change the make_fatal method Feb 19 21:00:08 Ifaistos: if class == 0 or class == 6 Feb 19 21:00:13 zecke: http://rafb.net/p/YCSNiR49.html Feb 19 21:01:12 koen: libfakeroot.la is evil, there are two staged Feb 19 21:03:59 well my time is up.... wife factor :) Feb 19 21:04:20 night everyone Feb 19 21:05:20 koen: I think we could disable the OSABI check, I'm pretty confident that we cross compile for linux Feb 19 21:05:45 zecke: okay Feb 19 21:06:09 feel free to disable them and making .la non fatal Feb 19 21:06:09 woglinde: can you please give me your changes you made on the config files? Feb 19 21:06:47 zecke: http://www.openembedded.org/more-quality-assurance-tests Feb 19 21:07:19 koen: the gettext bits are scary Feb 19 21:07:26 koen: they refer to .la files inside the WORKDIR Feb 19 21:14:28 03koen 07org.oe.dev * ra4676fba... 10/ (1 classes/insane.bbclass): insane.bbclass: report, but don't die on broken .la files Feb 19 21:14:30 hmm no more work on hokidachi this night... Feb 19 21:20:32 nite Feb 19 21:21:01 koen: now only checking file and dir permissions is left to do Feb 19 21:36:17 bd2: what is connected there? Feb 19 21:36:21 err, sorry:) Feb 19 22:08:37 hi Feb 19 22:10:10 Good night Feb 19 22:31:15 rwhitby Feb 19 22:31:30 did you saw my mail? i compiled correctly! Feb 19 22:31:36 :) Feb 19 23:17:52 Jin^eLD: trying to build qemu-native with compat-gcc-34 installed. I should know more in a few hours :) Feb 19 23:23:26 :) Feb 19 23:23:49 well I could start it as well Feb 19 23:23:54 and tell you :) Feb 19 23:24:15 actually, let me try Feb 19 23:31:50 hvontres|home: at least it passed configure now, compiling... Feb 19 23:35:07 hvontres|home: NOTE: package qemu-native-0.8.2+cvs20060723: completed Feb 19 23:35:15 it works Feb 19 23:35:37 so I guess you can have faith in your running compilation process Feb 20 01:04:43 koen|away: sir? Feb 20 01:05:30 koen|away: I will assume you did not mean to hilight me Feb 20 01:19:19 How would I go about creating and submitting a package for org.openembedded.dev? I've got bitbake recipes and source for ftplib-3.1.1 that does not require anything from opie Feb 20 01:24:17 hi * Feb 20 01:24:59 after build dropbear package, i try to nano package, but bitbake get error Feb 20 01:25:01 NOTE: Not using a cache. Set CACHE = to enable. Feb 20 01:25:06 ERROR: no files to build. Feb 20 01:51:57 Jin|away: heh... well now to try glibc.... Feb 20 01:56:09 guys. I have a question about SVN SRC_URI paths Feb 20 01:56:27 the format that is used there doesn't seem to allow for a directory name for the checkout to go into Feb 20 01:57:55 ie I can issue "svn co svn://svnserve1.selserve/${project_name}/trunk checkout_dir" and get the contents of trunk checked out to the dir "checkout_dir" Feb 20 01:58:24 is there a way to specify that directory in the SRC_URI field? Feb 20 02:00:23 jselect: localdir I think Feb 20 02:01:07 ah thanks rwhitby I'll try that now Feb 20 02:01:39 See /bitbake/lib/bb/fetch/svn.py Feb 20 02:01:55 localpath is it Feb 20 02:02:12 (line 51 of that file) Feb 20 02:02:14 That was my next step to look at the fetch code Feb 20 02:02:15 thanks Feb 20 02:02:44 damn version control maintainer here at work decided to name two directories "default" under different paths :) Feb 20 02:05:38 hmm doesn't work how I'd hoped Feb 20 02:06:04 The unpack function seems to read llocalpath and assumes that is the replaced name for the source Feb 20 02:11:30 ah forget it. localpath defines the name of the tar.gz created from the checkout Feb 20 02:11:39 new do_fetch time Feb 20 02:30:39 newest rev has broken quilt Feb 20 02:30:44 or recent rev at least Feb 20 02:30:53 doesn't build Feb 20 02:30:58 fails on make install **** ENDING LOGGING AT Tue Feb 20 02:59:57 2007