**** BEGIN LOGGING AT Wed Aug 08 02:59:57 2007 Aug 08 06:15:00 polyonymous, hi, are you around by any chance? Aug 08 06:17:19 morning Aug 08 06:20:37 hi hrw Aug 08 06:30:18 polyonymous, hi Aug 08 06:30:56 ibot: spell orthopaedist Aug 08 06:30:58 possible spellings for orthopaedist: orthopedist orthopedists orthopedist's Aug 08 06:34:04 03koen 07org.oe.dev * r2c0fad76... 10/ (1 packages/maemo3/hildon-thumbnail_0.11.bb): hildon-thumbnail: add 0.11 Aug 08 06:34:09 03raj.khem 07org.oe.dev * rc441a97d... 10/ (4 files in 3 dirs): eglibc: fix staging, SRC_URI and add a backport fix from gcc Aug 08 06:46:12 koen: hello Aug 08 06:46:59 koen: I am able to build the whole stuff with eglibc but it is not booting :( Aug 08 06:47:21 Freeing init memory: 120K Aug 08 06:47:21 Kernel panic - not syncing: No init found. Try passing init= option to kernel. Aug 08 06:47:27 hi koen Aug 08 06:47:47 khem_: boot with 'init=/bin/busybox sh' maybe? Aug 08 06:47:59 lemme try that Aug 08 06:48:09 yeah thats next in line Aug 08 06:48:39 grrr Aug 08 06:48:56 the uni network locks me out of freedesktop.org and ibiblio Aug 08 06:49:22 so I can't complete a build because I can't download the new gstreamer... Aug 08 06:49:24 koen: wtf.. Aug 08 06:49:42 Freeing init memory: 120K Aug 08 06:49:42 Failed to execute /bin/busybox. Attempting defaults... Aug 08 06:49:42 Kernel panic - not syncing: No init found. Try passing init= option to kernel. Aug 08 06:51:59 * * OE Bug has been RESOLVED by sybren(AT)stuvel.eu Aug 08 06:52:00 * *  Zaurus C3000 drains battery on stand-by Aug 08 06:52:02 * * http://bugs.openembedded.org/show_bug.cgi?id= Aug 08 06:53:47 khem_: boot any good system and chroot into the one which you built Aug 08 06:53:57 khem_: looks like you lack libs in it Aug 08 06:53:59 koen: Did u have a chance to look at the patch I posted Aug 08 06:54:09 hrw: yeah looks like that Aug 08 06:54:19 hrw: I dont have a glibc based build Aug 08 06:54:22 running one now Aug 08 06:54:32 khem_: then fetch angstrom one? Aug 08 06:54:39 * hrw -> doctor Aug 08 06:54:48 my right wrist is killing me ;( Aug 08 07:01:01 hrw|gone: you are right my eglibc libs are missing from rootfs Aug 08 07:07:07 When and how is rootfs created Aug 08 07:09:58 * * OE Bug 2721 has been created by  Aug 08 07:10:00 * * various opie/palm related fixes Aug 08 07:10:02 * * http://bugs.openembedded.org/show_bug.cgi?id=2721 Aug 08 07:29:58 * * OE Bug 2721 has been RESOLVED (INVALID) by Aug 08 07:30:01 * *  various opie/palm related fixes Aug 08 07:30:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2721 Aug 08 07:31:12 NOTE: preferred version 2.5 of glibc-intermediate not available (for item glibc-gconv-iso8859-1) Aug 08 07:31:18 any ideas anyone? Aug 08 07:32:02 khem: did you try chrooting into the eglibc image from a known-good image? Aug 08 07:42:58 * * OE Bug 2507 has been REOPENED by Aug 08 07:43:00 * *  building for angstrom-2007.1 tries to build glibc 2.3.2 Aug 08 07:43:02 * * http://bugs.openembedded.org/show_bug.cgi?id=2507 Aug 08 07:44:33 koen: no, however I see that rootfs/lib does not have any thing from eglibc Aug 08 07:44:39 koen: where does it get copied Aug 08 07:44:47 in the build proces Aug 08 07:45:12 during do_rootfs, the packages should depend on eglibc via our shlib detection Aug 08 07:50:52 koen: it seems thats not happening Aug 08 07:51:20 koen: the libs are there in tmp/work/armv5te-angstrom-linux-gnueabi/eglibc-2.6 Aug 08 07:51:20 +svn20070808-r0/image/lib Aug 08 07:51:33 but they dont appear in tmp/rootfs Aug 08 07:54:26 do the created packages look ok? Aug 08 07:54:28 hey zecke_ Aug 08 07:54:41 e.g libc6.....ipk Aug 08 07:58:02 re Aug 08 08:03:53 koen: I see only libc6-dev_2.6+svn20070808-r0_armv5te.ipk Aug 08 08:04:08 hey XorA Aug 08 08:04:17 but in glibc case there is libc6_2.5-r6_armv5te.ipk also Aug 08 08:04:51 khem: so it looks like it didn't create libc6_2.6+svn20070808-r0_armv5te.ipk Aug 08 08:06:22 koen: hmmm ok how come the build completed without it Aug 08 08:06:22 koen: where should I look for fixing this ? Aug 08 08:07:11 bitbake -b eglibc_svn.bb -c install -f Aug 08 08:07:22 and see which files it installs Aug 08 08:07:43 (tmp/work/armv5te....../eglibc-..../image/) Aug 08 08:09:43 koen: install looks good Aug 08 08:09:55 koen: all libs are there in image/lib Aug 08 08:10:05 bitbake -b eglibc_svn.bb -c package -f Aug 08 08:10:14 any weird warnings there? Aug 08 08:10:23 lemme try that Aug 08 08:10:43 that will collect the files and put them in tmp/work/armv5te....../eglibc-..../install/ Aug 08 08:12:21 mornig Aug 08 08:15:02 koen: the install directory has mixture of eglibc and glibc folders Aug 08 08:16:26 03marek.vasut 07org.oe.dev * r76fec634... 10/ (1 packages/libopie/libopie2_cvs.bb): libopie2: use proper var for CVS version Aug 08 08:16:33 03marek.vasut 07org.oe.dev * r88c64741... 10/ (1 packages/libqpe/libqpe-opie_cvs.bb): libqpe: use proper var for CVS version Aug 08 08:16:38 03hrw 07org.oe.dev * rf048a9ba... 10/ (4 files in 2 dirs): palm machine configs: fix names in comments Aug 08 08:18:28 khem: it is collected from ../image, so it should be only from eglibc Aug 08 08:18:42 bitbake -b eglibc_svn.bb -c package_write -f Aug 08 08:18:52 that will output the filenames it creates Aug 08 08:30:30 koen: I am going to try this patch http://pastebin.ca/649623 Aug 08 08:30:54 koen: I will fire a clean build and sleep now and will continue tomorrow Aug 08 08:31:49 khem: those lines only work in the virtual OE namespace, they don't have anything to do with packaging errors Aug 08 08:32:02 koen: ok Aug 08 08:32:24 could you pastebin the output of package_write? Aug 08 08:32:32 (run -c install -f first) Aug 08 08:33:43 koen: I deleted whole tmp Aug 08 08:34:07 started fresh build I made note of it and will check tomorrow Aug 08 08:34:51 koen: are these bitbake -c stuff documented somewhere Aug 08 08:35:15 I believe should be in bb manual somewhere Aug 08 08:35:38 koen, if you are not interested in that patch I sent, I dont mind Aug 08 08:40:59 koen: can you remind me proper way of disapprove? Aug 08 08:41:15 mtn disapprove revnumber - and then? Aug 08 08:41:25 mtn sync Aug 08 08:41:37 or if the commit is in the past, mtn merge Aug 08 08:51:15 03hrw 07org.oe.dev * r936655d7... 10/ (1 packages/libqpe/libqpe-opie_cvs.bb): disapproval of revision '88c64741ce48efad01b051c81bcb07df08267748' Aug 08 08:51:41 hrw: and a mail to the list why it was disapproved Aug 08 08:54:13 sent Aug 08 09:28:15 good morning Aug 08 10:21:54 hi. I have a machine with the following /etc/ipkg/arch.conf: arch all 1, arch any 6, arch noarch 11, arch arm 16, arch ereader 21 (comma is line break). how should my PACKAGE_EXTRA_ARCHS look like to build compatible packages (without changing arch.conf)? Aug 08 10:30:42 mickeyl: good morning! Aug 08 10:30:48 Guten Morgen koen Aug 08 10:31:14 koen: i see eglibc is progressing, very cool. how far are we from an angstrom-eglibc booting? Aug 08 10:31:43 mickeyl: not far I think Aug 08 10:31:46 great Aug 08 10:32:10 mickeyl: the big trouble is that eglibc and glibc can't be seperated using multimachine like glibc and uclibc Aug 08 10:32:23 koen: no? why's that? Aug 08 10:32:35 it's the same thing Aug 08 10:32:40 oops Aug 08 10:32:48 eglibc is basically a patched glibc Aug 08 10:32:51 do_stage() { Aug 08 10:32:52 # FIXME: this removes files from staging Aug 08 10:32:52 # make sure there isn't a conflicting libc in staging Aug 08 10:32:53 # this should be solved differently Aug 08 10:32:53 rm -f ${STAGING_LIBDIR}/libc.so.6 Aug 08 10:33:01 oh Aug 08 10:33:04 i see Aug 08 10:33:29 so this is not another build-from-the-groundup thing, but rather a patched version that's not (yet?) accepted by upstream? Aug 08 10:33:38 yes Aug 08 10:33:50 they import glibc svn into their eglibc svn tree AIUI Aug 08 10:34:10 aah, i see. interesting. one needs to see the benefits of this approach Aug 08 10:34:16 less compatibility hell compared to uclibc Aug 08 10:34:16 propably because mr drepper refuse to apply patches to make x-compiling easier Aug 08 10:34:47 btw, glibc 2.6 + gcc 4.2 works Aug 08 10:35:10 good to know. should we upgrade A soon? Aug 08 10:35:21 I'm not sure Aug 08 10:35:23 oh btw., i think we can remove the default preference from gtk+ 2.10.14 Aug 08 10:35:44 florian: what do you think re. gtk+? Aug 08 10:36:24 hi mithro Aug 08 10:36:29 how are you ? Aug 08 10:37:21 mickeyl: pretty good Aug 08 10:37:32 good. how are your studies going on? Aug 08 10:37:35 did I tell you that I graduated last week on Tuesday? Aug 08 10:37:41 woo Aug 08 10:37:43 CONGRATS! Aug 08 10:37:52 great! what are you up to next? Aug 08 10:37:52 mithro: congrats Aug 08 10:38:20 mickeyl: I think we should... there are some open issues, but we need some motivation to fix these. Aug 08 10:38:30 mickeyl: looking for jobs Aug 08 10:38:45 being really slack about putting together my resume Aug 08 10:39:06 get a tin cup and hang out at trade shows Aug 08 10:39:16 was waiting to see if I got a scholarship at Uni to do some SDR stuff, but it looks like that has fallen through Aug 08 10:39:55 florian: ok. i'm going to do it then. Aug 08 10:40:22 mithro: which area do you want to work on/in? Aug 08 10:40:40 mickeyl: do you know which patches from the previous releases are missing? Aug 08 10:40:47 mickeyl: ideally I would like to work on something FOSS related Aug 08 10:41:10 florian: no, let me check Aug 08 10:41:40 how can I enforce that packages built in OE against glibc 2.3.5 can be installed on a system with a glibc 2.3.2 without the need to use -force-depends? Aug 08 10:42:47 mickeyl: but there is pretty slim pickings in that area here in adelaide Aug 08 10:43:30 mickeyl: seen the new imac? Aug 08 10:43:33 mickeyl: _law_ Did a pretty good job making GTK 2.11 work properly... including SMS-like input and keypad navigation. Aug 08 10:43:35 florian: i can't see anything substantial missing except the keynav patches. and i think a couple of those are upstream yet aren't they? Aug 08 10:43:49 florian: oh really? didn't know that. it's not in oe Aug 08 10:43:49 mickeyl: yes Aug 08 10:44:13 koen: not yet. i bought lots of things yesterday, i'm broke now Aug 08 10:44:31 mickeyl: they are brushed alu now Aug 08 10:44:42 mickeyl: he intended to submit these... Aug 08 10:44:42 including alu keyboard Aug 08 10:44:48 hi, all! I have .ext2 image generated by genext2fs, now I need to turn it into bootable hard disk image, is it possible? Geometry is known. Aug 08 10:44:50 _law_: ping Aug 08 10:45:04 koen: but with a glossy screen... what a pain :-/ Aug 08 10:45:22 the case and keyboard looks really great Aug 08 10:45:52 mickeyl: http://helmsdeep.homelinux.net/~pat/oe/tornado/ Aug 08 10:46:35 <_law_> mickeyl, pong Aug 08 10:47:02 florian: you can buy the keynoard seperately Aug 08 10:47:25 florian: how about gtk 2.12? Aug 08 10:47:51 _law_: hi. i heard you had a set of gtk+ 2.11 ready? we'd appreciate it seeing in .dev :) Aug 08 10:48:38 florian: ah. i'd say go for it w/ DEFAULT_PREFERENCE=-1 Aug 08 10:49:07 well, i wouldn't use their keyboards anyways Aug 08 10:49:17 it's ms natural for me -- nothing else comes close Aug 08 10:49:22 koen: 2.12 is not ready... or do you know more than me? Aug 08 10:49:42 mickeyl: sure Aug 08 10:49:47 koen, I suspect my libtool-1.5.24 is clsely related to the need to add the -lstdc++ flag to builds for teh Efika Aug 08 10:50:09 for example libusb adds that flag globally, but is not needed at least for the OSK Aug 08 10:51:07 should we make the -lstdc++ apply only for efika/ppc in libusn Aug 08 10:51:27 do you really need full stdc++ ? Aug 08 10:51:32 supc doesn't work? Aug 08 10:51:34 <_law_> mickeyl, http://bugzilla.gnome.org/show_bug.cgi?id=460858 i have also posted them to bugzilla Aug 08 10:51:53 mickeyl, not familiar with supc Aug 08 10:52:21 the problem remains, there is something fishy in the toolchain used on Efika wrt to -lstdc++ handling Aug 08 10:52:50 <_law_> have to go now. bbl Aug 08 10:53:21 ibot: botmail to _law_: ok, thanks. could you commit them w/ DEFAULT_PREFERENCE=-1 when you have a chance? Aug 08 10:56:23 florian: I think I got confused with glib 2.12 Aug 08 10:56:26 ehm Aug 08 10:56:26 2.14 Aug 08 10:57:41 koen: :-) Aug 08 10:59:34 03mickeyl 07org.oe.dev * rbacc3c96... 10/ (64 files in 5 dirs): Aug 08 10:59:34 gtk+: drop 2.10.9 to 2.10.12, promote 2.10.14 as the new default version Aug 08 10:59:34 2.11 will be added soon w/ DEFAULT_PREFERENCE=-1. If you still need 2.6.10, Aug 08 10:59:34 please set your PREFERRED_PROVIDER accordingly. Aug 08 10:59:39 03koen 07org.oe.dev * r99ddf3e8... 10/ (4 files in 3 dirs): roadmap: add gtk2 version Aug 08 10:59:44 03koen 07org.oe.dev * r4cdf2bbe... 10/ (1 packages/roadmap/files/cross.patch): roadmap: add missing patch Aug 08 11:04:10 hi woglinde Aug 08 11:05:04 hi florian Aug 08 11:05:17 hi all Aug 08 11:06:04 koen : thanks for the tip on the FIC, will let it connected on the usb hub all the time Aug 08 11:06:56 or remove the battery when you're not using it/transporting it Aug 08 11:09:01 glibc_2.3.3 should not be available for arm. it does not contain any patches (especially no arm-ioperm.patch which makes the build fail). file a bug? Aug 08 11:09:07 Zero_Chaos: http://bugs.openembedded.org/show_bug.cgi?id=2722 Aug 08 11:09:46 * koen tries to setup and OE autobuilder Aug 08 11:09:51 ehm Aug 08 11:09:59 * * OE Bug 2722 has been created by  Aug 08 11:10:01 * * Seppuku.bbclass triggers recursion error Aug 08 11:10:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2722 Aug 08 11:10:09 ~botmail for zecke: http://bugs.openembedded.org/show_bug.cgi?id=2722 Aug 08 11:10:14 koen, when you get teh AB running, try libtool-1.5.24 Aug 08 11:10:27 koen do you understand german? Aug 08 11:10:29 I am guessing it will work for all but Efika ... Aug 08 11:10:41 woglinde: jawohl! Aug 08 11:10:46 koen -> http://www.heise.de/newsticker/meldung/93966 Aug 08 11:11:01 we are violating patents Aug 08 11:11:45 ". In dem 2004 anerkannten Patent 6,784,873 geht es um eine virtuelle Tastatur, die auf einem berührungsempfindlichen Bildschirm angezeigt wird." Aug 08 11:11:57 vkb on ts Aug 08 11:12:04 koen yes Aug 08 11:12:13 "berührungsempfindlichen Bildschirm" Aug 08 11:12:21 touchscreen Aug 08 11:12:22 that's a neat word (or 2) Aug 08 11:12:25 yes Aug 08 11:13:12 “patent Trollerlei” Aug 08 11:13:21 :-) Aug 08 11:14:33 woglinde, koen: http://fl0rian.wordpress.com/2006/02/21/prior-art/ Aug 08 11:14:41 * koen eats some tom-ym soup Aug 08 11:14:52 how to apply the sound statefile in the attatchment of Aug 08 11:14:53 we had a similar one before :-) Aug 08 11:14:58 ~hail http://live.gnome.org/GnomeCookbook Aug 08 11:14:58 * ibot bows down to http://live.gnome.org/GnomeCookbook and chants, "I'M NOT WORTHY!!" Aug 08 11:14:59 !oebug 2667 Aug 08 11:15:01 * * Bug 2667, Status: NEW, Created: 2007-07-19 08:07 Aug 08 11:15:02 * * uncle.grog(AT)gmail.com: Sound problems with recent kernels a.k.a. asound statefiles need updating Aug 08 11:15:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2667 Aug 08 11:15:21 the patent is from 2004. there is a c64 emulator having a virtual keyboard much longer IMHO Aug 08 11:16:09 hm they filed it in at August 4, 2000 Aug 08 11:16:32 florian yes I know prior-art Aug 08 11:17:16 woglinde: sure... but that more or less about the same thing but with apple on the other side. Aug 08 11:18:13 lol Aug 08 11:18:22 apple file in the same patent Aug 08 11:18:27 they are so dump Aug 08 11:18:37 aeh dull Aug 08 11:18:39 * koen still can't connect to fd.o Aug 08 11:18:44 ~lart uni network morons Aug 08 11:18:44 * ibot hits uni network morons with an anvil and laughs with a contralto voice ... Haha Ha HA Ha Aug 08 11:20:42 :D Aug 08 11:21:22 he mickeyl Aug 08 11:21:29 lol the iphone simcard cloning Aug 08 11:21:38 could somebody help? Aug 08 11:23:23 hi woglinde Aug 08 11:24:03 Is Munich Uni of Technology any good? Aug 08 11:24:14 well Aug 08 11:24:23 it's not exactly known for its IT department Aug 08 11:24:29 mickeyl, koen: That bug is one for zaurus users :-) Aug 08 11:24:30 but it shouldn't be bad Aug 08 11:24:33 :) Aug 08 11:24:55 I got an email from a guy there about SDR on a Davinci board Aug 08 11:24:56 florian: i even forgot where i put my Zaurus battery Aug 08 11:25:10 florian: and i mean, the whole series of Zauruses, not one battery ;) Aug 08 11:25:21 Crofton|home: its quite likely to be a good choice... the Munich area is nice as well - but living there is expensive. Aug 08 11:25:33 mickeyl: oh well Aug 08 11:25:37 So I should visit then :) Aug 08 11:25:47 no chance to unpack them since the move Aug 08 11:26:11 My wife's parents met south of Munich after the war Aug 08 11:26:32 mickeyl: heh Aug 08 11:27:18 Institue for Integrated Systems Aug 08 11:27:26 mickeyl: Is one of the OE clamshells idle currently? Aug 08 11:28:04 crofton it is a good uni Aug 08 11:29:05 I need to get the guy interested in OE also Aug 08 11:32:36 03florian 07org.oe.dev * r841bb9cf... 10/ (1 packages/linux/linux-hackndev-2.6/palmtx/defconfig): linux-hackndev-2.6: Update palmtx defconfig to make it build again. Aug 08 11:32:40 03florian 07org.oe.dev * r7dac8ac9... 10/ (1 packages/linux/linux-hackndev-2.6_svn.bb): linux-hackndev-2.6: bump PR Aug 08 11:32:45 03koen 07org.oe.dev * r87682066... 10/ (1 classes/seppuku.bbclass): Aug 08 11:32:45 seppuku.bbclass: fix indentation Aug 08 11:32:45 * grrrrr, stupid python Aug 08 11:32:51 crfoton what is he doing? Aug 08 11:32:56 the guy? Aug 08 11:33:03 florian: for sure. i don't think hrw and zecke are using theirs a lot Aug 08 11:33:26 *g* zecke has no time Aug 08 11:33:30 yea Aug 08 11:33:40 woglinde, wireless communication Aug 08 11:33:48 crofton ah hm Aug 08 11:36:05 mickeyl: I do not have time too, but I would like to play with GPE on then and try to improve support for them a little bit. Aug 08 11:39:53 florian: like the annoying 'backlight switches off' bug? Aug 08 11:41:16 koen: I don't know, do I have a Zaurus? ;) Aug 08 11:41:36 koen: but let me guess: ipaq-sleep? Aug 08 11:42:17 florian: most likely Aug 08 11:42:46 it seems 'gpe' assumes you have a powerbutton that can be held to enable backlight and that your LCD is transflective Aug 08 11:43:06 but on a zaurus backlight of means screen unreadable Aug 08 11:45:38 koen: indeed... the defaults might be bad for most modern devices Aug 08 11:46:31 * koen thinks every LCD should be at least transflective Aug 08 11:56:27 koen: yeah Aug 08 11:56:38 I don't think transflective works nicely with a high-rez LCD Aug 08 11:58:36 eekee: it works with 110dpi what I have on my laptop which lacks this feature Aug 08 12:01:05 oh o.o; Aug 08 12:03:09 I was thinking of the segment shadows you see in watches, & I dunno how polarizing reflectors work Aug 08 12:04:32 bye Aug 08 12:04:37 anyway, I'm a bit worried abotu my spitz. Had to reboot many times because I couldnt' make the screen come back & a couple of times had to pull the battery out because it wouldn't come back after reset. Now it's not coming back after I pull the battery out :/ Aug 08 12:04:42 bai woglinde Aug 08 12:05:01 or after I put it back in even :o) Aug 08 12:05:50 ah, it's on now. maybe have to leave it for a bit after I put the battery in or something. It's all rather worrisome Aug 08 12:07:29 03lenehan 07org.oe.dev * r05cc5d66... 10/ (3 files in 3 dirs): Aug 08 12:07:29 p3scan 2.9.05d: Fix up configure.in to work with the recently updated Aug 08 12:07:29 m4/autotools. It wasn't happy about the lack of a new line at the end of Aug 08 12:07:29 configure.in. Aug 08 12:10:00 I could probably cook up a script to keep the backlight on when needed... Aug 08 12:10:28 well, not script as it would have to know when the last key was pressed, but.. Aug 08 12:11:53 koen : should i use fic1 or fic2 as machines ? Aug 08 12:12:21 fic1=fic-gta1 Aug 08 12:13:04 gta01 Aug 08 12:13:38 btw I tried setting one of the buttons to run "bl 120" but that resulted in the buttons no longer working at all. Aug 08 12:15:10 steliosk: I'll mail it tomorrow Aug 08 12:15:18 steliosk: I included an openmoko sticker :) Aug 08 12:16:06 heh thanks :) Aug 08 12:23:58 * * OE Bug 2721 has been REOPENED by Aug 08 12:24:00 * *  various opie/palm related fixes Aug 08 12:24:02 * * http://bugs.openembedded.org/show_bug.cgi?id=2721 Aug 08 12:25:30 i have some troubles bbing task-e-x11 Aug 08 12:25:44 still have to be more precise Aug 08 12:25:57 makedepend native dont compile Aug 08 12:26:11 i ahve deleted all my tmp dir and rebuild all but dont work Aug 08 12:26:12 http://pastebin.com/d5937d1b8 Aug 08 12:26:18 here is the error Aug 08 12:28:51 re Aug 08 12:28:58 mickeyl: I do not have OE clamshell Aug 08 12:29:06 mickeyl: I have OE tosa and alix Aug 08 12:29:43 ok. who did you give the C3K ? Aug 08 12:31:09 laibsch on fosdem Aug 08 12:31:26 mickeyl: it is listed on projectdevices wiki page Aug 08 12:39:58 * * OE Bug 2723 has been created by waldecir(AT)gmail.com Aug 08 12:40:00 * * makedepend do_compile problem: No rule to make target `makedepend.@APP_MAN_SUFFIX@', needed by `all-am'. Aug 08 12:40:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2723 Aug 08 12:41:09 qemu-native uses host gcc now or still requires gcc 3.4.4 ? Aug 08 12:43:01 03mickeyl 07org.oe.dev * re51f6d1c... 10/ (1 packages/matchbox2/matchbox-panel-2_svn.bb): matchbox-panel2 ship plugin *.la in debug package (QA fix) Aug 08 12:43:10 03mickeyl 07org.oe.dev * r54fa1e3c... 10/ (1 packages/openmoko2/openmoko-terminal2_1.0.0.bb): openmoko-terminal2: supply sane font defaults to mrxvt Aug 08 12:44:11 steliosk: host Aug 08 12:44:22 steliosk: I found second problem with Alix board Aug 08 12:44:32 steliosk: 2.5" cables are impossible to get Aug 08 12:44:42 hmmm ok, then we can build it using icecc also ;) Aug 08 12:56:14 koen, is there an easy way to try build with gcc-4.2.x? Aug 08 12:56:23 hrw : any idea why its using gcc 3.4 instead of gcc 4.1.2 ? Aug 08 12:56:47 this must add -lstdc++ for Efika is going to drive me insane Aug 08 12:57:18 steliosk: qemu? Aug 08 12:57:26 qemu-native Aug 08 12:57:46 steliosk: qemu can be built only with gcc3 - upstream refuse to use gcc4 Aug 08 12:59:11 hrw : hmmm so that did not change then.... I was under the impression it had Aug 08 12:59:36 you need to have gcc3 installed in host Aug 08 12:59:50 and icecc to build a package for it Aug 08 13:00:30 we have a patch in OE that makes it build with gcc4 Aug 08 13:01:15 Crofton|home: angstrom-2007.1.conf, line 176 Aug 08 13:01:24 koen : in bugtracker ? Aug 08 13:01:32 no, in .dev Aug 08 13:01:50 but i am building dev Aug 08 13:02:09 just synced with the repo Aug 08 13:02:09 if it finds 3.4 it will use 3.4 Aug 08 13:02:26 since that's what the qemu dudes recommend Aug 08 13:02:36 * steliosk removes gcc3 to see Aug 08 13:03:03 steliosk: why? gcc 3.4 is the recommended option Aug 08 13:03:27 i want to try and see if icecc/gcc4 will work Aug 08 13:03:45 with qemu Aug 08 13:04:05 its one of the packages that's excluded by default from using icecc Aug 08 13:06:51 its distributing build jobs :) lets see if it builds Aug 08 13:07:01 03koen 07org.oe.dev * r583da5d8... 10/ (1 packages/matchbox2/matchbox-panel-2_svn.bb): matchbox-panel-2: .a and .la files go into -dev packages, not into -dbg packages Aug 08 13:07:29 koen, I compare the lines from libtool for OSK and Efika, and cannot figure out why one links c++ ok, and the other fails Aug 08 13:07:40 koen, So maybe it is a problem in gcc? Aug 08 13:08:32 Crofotn|home : Are you using gcc 4.1.2 ? Aug 08 13:08:36 yes Aug 08 13:08:45 whatever default for angstrom is Aug 08 13:08:52 which is 4.1.2 Aug 08 13:09:17 and comparing compile logs for libusb for OSK (omap5912osk) and Efika Aug 08 13:09:23 so gcc 4.1.2 + newer libtool = fails to builds c++ ? Aug 08 13:09:30 well Aug 08 13:09:51 older libtool needs the CXXFLAGS_powerpc = -lstdc++ hack Aug 08 13:10:23 libusb used -lsdtc++ for all machines, but it is not needed for OSK Aug 08 13:11:05 SO I think I will try builds with 4.2.1 and see if the behavior imporves Aug 08 13:11:22 and I can work on other stuff while the build runs :) Aug 08 13:11:24 koen : Do you get the shared libriaries build in OSK ? Aug 08 13:11:39 steliosk, they are build Aug 08 13:11:43 libstdc++.so? Aug 08 13:11:49 yes Aug 08 13:12:05 I'll post some logs in a bit Aug 08 13:12:09 hang on Aug 08 13:24:35 http://pastebin.ca/649856 Aug 08 13:24:47 libtool and gcc lines for OSK and Efika builds Aug 08 13:25:37 Note this appears in both: # Aug 08 13:25:37 /home/balister/oe/tmp/cross/powerpc-angstrom-linux/ Aug 08 13:25:37 # Aug 08 13:25:37 lib/libstdc++.so Aug 08 13:26:57 Crofton: gcc3-build-cross.inc line 47 Aug 08 13:27:13 it works as intended on everythign except ppc Aug 08 13:28:06 gcc4? Aug 08 13:30:54 ok, that file included in gcc4 cross Aug 08 13:34:22 Crofotn|home : I have not looked into the problem but i believe that's a libtool issue Aug 08 13:36:50 well Aug 08 13:37:04 libtool does not know about the underlying machine Aug 08 13:37:25 it feeds nearly identical lines to gcc for linking Aug 08 13:37:41 and the OSK version finds stdc++, and ppc does not Aug 08 13:38:00 and we have to have bb files to link c++ for ppc Aug 08 13:39:55 s/hack/have/ Aug 08 13:47:17 Crofotn|home : did you try to build any other gcc+ app ? Aug 08 13:47:37 s/gcc+/c++ Aug 08 13:48:15 got one in mind? Aug 08 13:48:35 bitbake omniorb Aug 08 13:49:42 interesting, no -lstdc++ line there Aug 08 13:49:53 hmm Aug 08 13:50:00 maybe it does not use libtool Aug 08 13:51:11 What we would really nice is to know what platforms do not build with libtool-1.5.24 Aug 08 13:51:26 It is 50% for me :) OSK, good, Efika bad Aug 08 13:51:49 03pH5 07org.oe.dev * r1e604e78... 10/ (6 files in 3 dirs): Aug 08 13:51:49 oprofile: add 0.9.3, update cvs, drop 0.9.1 left-overs Aug 08 13:51:49 - the no_arm_mapping_syms and opcontrol_bashisms patches are applied upstream Aug 08 13:51:53 03koen 07org.oe.dev * r0bb67ca8... 10/ (1 classes/update-rc.d.bbclass): update-rc.d: merge in ${D} cleanups from poky Aug 08 13:52:57 mickey|DIY, RP: http://lists.linuxtogo.org/pipermail/openembedded-devel/2007-August/002754.html Aug 08 13:56:17 koen, the inc fiel you refer to only operates on the .a library Aug 08 14:03:14 morning Aug 08 14:05:16 ~lart CSS Aug 08 14:05:16 * ibot installs PocketPC on CSS's PDA Aug 08 14:05:43 css is good Aug 08 14:05:46 browsers stuxx Aug 08 14:09:23 hrw: simple quition if I have
  • that css magic do I use to style that Aug 08 14:10:10 I cant change the format of :-( Aug 08 14:10:32 Crofton|home: sed -i "s|dependency_libs\s*=\s*.*|dependency_libs='-L${CROSS_DIR}/${TARGET_SYS}/lib -lgcc_s -lc -lm '|" ${CROSS_DIR}/${TARGET_SYS}/lib/libstdc++.la || true Aug 08 14:11:33 my sed is poor Aug 08 14:12:06 you are adding a -L to the cross libdir? Aug 08 14:13:03 we are replacing the contents of dependency_libs Aug 08 14:13:13 but I think that is a red herring Aug 08 14:13:25 what is dependency libs? Aug 08 14:13:55 I think the question we must answer is "Why do some bb files need the -lstdc++ for PPC/Efika?" Aug 08 14:13:56 the stuff it needs to link to itself Aug 08 14:14:28 XorA, li.blah a { ... } ? Aug 08 14:15:36 damn, I have a copy of "sed and awk" right here Aug 08 14:16:14 XorA: li.blah a.blah { styles } Aug 08 14:16:49 or li.blah a {} if you want any in
  • Aug 08 14:16:57 ~seen kergoth Aug 08 14:16:59 kergoth is currently on #wowi-lounge (8d 18h 31m 24s) #oe (8d 18h 31m 24s) #wowhead (8d 18h 31m 24s) #norganna (8d 18h 31m 24s) #wowace (8d 18h 31m 24s) #elinux (8d 18h 31m 24s). Has said a total of 361 messages. Is idling for 15h 47m 35s, last said: 'hehe'. Aug 08 14:20:10 hrw: ah cheers, I musta bread deaded it first time around Aug 08 14:20:19 s/bread/brain/ Aug 08 14:20:48 XorA: the common problem with css for people is that elements inherit from parent which inherit from parent which .... Aug 08 14:21:11 hrw: yeah I was finding that :-) Aug 08 14:21:26 hrw: that firebug is awesome btw Aug 08 14:21:31 yes Aug 08 14:21:59 XorA: firebug talk was on my list during fosdem. but it was cancelled ;( Aug 08 14:22:36 I particulary like the fact it tells you what style you overrode to get where you are Aug 08 14:23:56 * XorA drupal theme almost works now Aug 08 14:29:45 03pH5 07org.oe.dev * r227d9d16... 10/ (1 packages/angstrom/angstrom-x11-image.bb): angstrom-x11-image: replace openmoko-dialer with openmoko-dialer2 Aug 08 14:29:49 03koen 07org.oe.dev * rbf37569e... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom: add 2008.1 file, see comments within Aug 08 14:30:19 what is 2008.1? Aug 08 14:30:33 Crofton|home: see comments within :) Aug 08 14:30:49 * Crofton|home kicks koan Aug 08 14:30:52 er koen Aug 08 14:31:11 # This is a testbed for unstable and/or untested things, while angstrom-2007.1 Aug 08 14:31:12 # is aiming for stability and a release. Aug 08 14:31:13 # Use this at your own risk, we welcome bugreports filed at Aug 08 14:31:13 # http://bugs.openembedded.org Aug 08 14:31:14 # In doubt, use DSITRO="angstrom-2007.1" Aug 08 14:31:27 gah Aug 08 14:32:00 * koen can't spell Aug 08 14:32:16 wooooh Aug 08 14:32:21 I can access fd.o again Aug 08 14:35:08 * Crofton|home rediscovers bash.org Aug 08 14:36:47 * koen hands Crofton|home thesis.org Aug 08 14:37:27 awesome Aug 08 14:39:17 http://qdb.us/ Aug 08 14:39:33 http://qdb.us/bottom <---- interesting Aug 08 14:40:33 * Crofton|home is thinking about asking for a quote Aug 08 14:45:27 "iphone shuffle, no screen and it phones random people" Aug 08 14:47:16 hehe Aug 08 14:48:22 http://bash.org/?785888 Aug 08 14:48:42 koen: that's a major refactoring. I'm afraid of subtle changes. did you test some images? Aug 08 14:48:55 koen ROFL Aug 08 14:49:14 mickey|DIY: with package.bbclass? Aug 08 14:49:17 ya Aug 08 14:49:26 not yet, you'd have to ask RP Aug 08 14:49:56 you will get messages about stripping debiug files Aug 08 14:51:03 debus I mean Aug 08 14:53:18 noglitch: did you mention a at91 patch for 2.6.22? Aug 08 14:56:37 BFD: /home/hrw/devel/OH/poky/build/tmp-poky/work/armv5te-poky-linux-gnueabi/xmodmap-1_1.0.2-r0/image/usr/bin/.debug/xmodmap: warning: sh_link not set for section `.ARM.exidx' Aug 08 14:56:46 that one Aug 08 14:56:55 right Aug 08 14:57:11 you get those when you run gdb with the current bbclass as wel Aug 08 14:57:30 koen: They're currently in the RUNSTRIP logfiles Aug 08 14:58:01 koen: I've asked on the arm-linux list, the binutils list and have now mailed Paul Brook directly about them since he added that section support to binutils ;-) Aug 08 14:58:12 :) Aug 08 14:58:41 mickey|DIY: Those changes seem stable in poky FWIW Aug 08 14:59:31 RP: i don't doubt that. i'm just afraid of different behaviour thanks to possible other things where poky and .dev differ Aug 08 14:59:34 koen: not yet... -rc5 Aug 08 14:59:40 s/possible// Aug 08 15:00:10 mickey|DIY: I know, it needs a RFC/T... Aug 08 15:00:24 My gut feeling says it should be ok though Aug 08 15:01:11 that's why I sent it to the mailinglist :) Aug 08 15:02:33 It would be possible to split that patch into about four separate logical patch units Aug 08 15:02:47 (if that helps) Aug 08 15:04:24 splitting always helps Aug 08 15:04:45 svn d -crev >rev.diff helps ;) Aug 08 15:05:58 03koen 07org.oe.dev * r6fcbcfd5... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom-2008.1: fix typo in comments Aug 08 15:07:39 bye Aug 08 15:08:52 cute, a mmc host test kernel module Aug 08 15:09:08 sounds usefull for collie Aug 08 15:09:18 (on LAK) Aug 08 15:15:35 03koen 07org.oe.dev * r6fcbcfd5... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom-2008.1: fix typo in comments Aug 08 15:20:50 mickey|DIY: http://svnweb.openmoko.org/trunk/src/target/OM-2007.2/artwork/themes/openmoko-standard-2/gtk-2.0/openmoko-dialer?rev=2666&view=diff&r1=2666&r2=2665&p1=trunk/src/target/OM-2007.2/artwork/themes/openmoko-standard-2/gtk-2.0/openmoko-dialer&p2=/trunk/src/target/OM-2007.2/artwork/themes/openmoko-standard-2/gtk-2.0/openmoko-dialer Aug 08 15:20:59 mickey|DIY: "IOError: [Errno 2] No usable temporary directory found in ['/tmp', '/var/tmp', '/usr/tmp', '/usr/lib/cgi-bin']" Aug 08 15:21:58 nasty Aug 08 15:24:20 no idea what's up there Aug 08 15:24:26 i'll tell the admin Aug 08 15:40:29 koen: hmm the web server suffers from a full partition Aug 08 15:40:34 that's the reason Aug 08 15:40:40 ouch Aug 08 15:41:21 ibot: botmail to LaF0rge: sita is full, please increase the simfs size Aug 08 15:42:06 * koen hopes that isn;t the scm server Aug 08 15:42:18 03koen 07org.oe.dev * r6fcbcfd5... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom-2008.1: fix typo in comments Aug 08 15:42:21 if it is I can hear people curse centralized scms Aug 08 15:43:54 koen: morning Aug 08 15:44:00 hey khem Aug 08 15:44:08 koen: I have the log for package_write Aug 08 15:44:28 could you put it online somewhere? Aug 08 15:44:35 koen: strangely its generating the locale packages starting with glibc- Aug 08 15:47:33 has anyone setup OE on OS X? Aug 08 15:49:15 koen: I am trying but some times upload suck on my connection I dont know why Aug 08 15:51:29 koen: http://paste.debian.net/34225 Aug 08 15:56:34 koen: it seems we need a separate eglibc-package.bbclass Aug 08 15:57:50 koen: may be create it as a separate package like uclibc Aug 08 16:20:46 gnah Aug 08 16:20:56 I know I shouldn't have tried to compile ffmpeg Aug 08 16:22:57 hello jg Aug 08 16:23:24 koen: any idea why .debug crap and some unstripped libraries end up in the filesytems? Aug 08 16:23:46 most likely bad packaging Aug 08 16:24:13 the unstripped stuff can be fixed by reviewing the patch on the mailinglist :) Aug 08 16:25:56 koen: which one? Aug 08 16:26:11 "[oe] merging package.bbclass with poky, please review diff" Aug 08 16:28:55 good evening Aug 08 16:28:56 florian: can you check in which packages those files are? my latest bitbake angstrom-x11-image installed some -dbg packages into the rootfs and I have no clue why. Aug 08 16:30:31 pH5, florian: check the Packages file in deploy and see if a non -dbg package depends on a -dbg package Aug 08 16:30:36 can anybody give me a hint on how to use the intel ixp4xx npe stuff on oe ? Aug 08 16:31:00 koen: oh... advanced stripping for the packages.. Aug 08 16:31:09 pH5: i'll check Aug 08 16:31:49 03koen 07org.oe.dev * r6fcbcfd5... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom-2008.1: fix typo in comments Aug 08 16:32:52 déjà vu Aug 08 16:35:07 hmmm Aug 08 16:35:21 why did I just eat all eggs before reading http://www.barisione.org/blog.html/p=96 ? Aug 08 16:37:05 koen: uhhhh Aug 08 16:44:46 koen I built a mamona i586 but my system believe that it is i386 instead of i586 so somethings are breaking... (dpkg for instance) do you have any idea? Aug 08 16:44:57 hmmm Aug 08 16:45:06 you'd have to ask RP about dpkg stuff Aug 08 16:45:39 03koen 07org.oe.dev * r6fcbcfd5... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom-2008.1: fix typo in comments Aug 08 16:49:29 koen: did u find something in the logs I posted Aug 08 16:51:12 KhemWork: nothing obvious Aug 08 16:51:31 KhemWork: I added a eglibc-package.bbclass Aug 08 16:51:54 koen: cool once I get this packaging sorted out it will help in locale Aug 08 16:52:48 koen: once it works may be it should be a separate package unless you are thinking of dropping glibc in favor of eglibc ") Aug 08 16:53:04 03koen 07org.oe.dev * r6fcbcfd5... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom-2008.1: fix typo in comments Aug 08 16:53:33 KhemWork: what do you mean with 'a seperate package'? Aug 08 16:53:44 koen: Like uclibc glibc Aug 08 16:54:05 a third library Aug 08 16:54:08 could you explain a bit more? Aug 08 16:54:44 OE splits on the host triplet, e.g. amr-linux-gnueabi vs arm-linux-uclibc-gnueabi Aug 08 16:54:49 I think it is just structural I think we added all eglibc stuff inside package/glibc Aug 08 16:54:54 but since eglibc is jsut like glibc we can't do that Aug 08 16:55:03 I thought it would be more symbolic to have packages/eglibc Aug 08 16:55:21 yeah thats true thats why I think its ok Aug 08 16:57:32 koen: I see that install/glibc folder is empty inside eglibc install Aug 08 16:57:45 but this is populated in glibc install Aug 08 16:58:09 how is a an application package going to be "differentiated" in the feeds ? now libc uclibc are 2 different things Aug 08 16:58:54 what is libc? Is it == glibc? Is it a meta-package? Aug 08 16:58:57 steliosk: if it is an eglibc question than I already trashed my suggestion Aug 08 16:59:44 steliosk: DEPLOY_DIR = "${TMPDIR}/deploy/${ANGSTROM_MODE}" Aug 08 16:59:58 steliosk: after that I trust the sanity of the uploader :) Aug 08 17:01:01 KhemWork : Yes it was. I was thinking about it, when i first saw eglibc (not your patches) Aug 08 17:02:11 koen : I have a feeling the eglibc should be on a different branch for the time Aug 08 17:02:14 steliosk: actually eglibc is glibc made embedded so ideally there should be only eglibc for embedded systems no glibc but that what I think may be not everyone Aug 08 17:02:48 steliosk: on a branch, why? Aug 08 17:02:57 steliosk: uclibc isn't on a branch Aug 08 17:03:30 and the work is very package specific its not touching common parts so much so prolly branch would be overkill Aug 08 17:03:38 KhemWork : True. We need to handle eglibc as glibc and uclibc and let the user/developer decide on what to use Aug 08 17:05:12 KhemWork koen : I think we could at least, as the beggining do so, so we can test it and then integrated it into .dev without too many things breaking Aug 08 17:05:24 steliosk: what would break? Aug 08 17:05:50 KhemWork koen : it should be "standard procedure" when doing things like that Aug 08 17:05:56 steliosk: so far koen has been doing great in getting the stuff in nothing broke except eglibc itself :) Aug 08 17:05:59 no it shouldn't Aug 08 17:06:04 it's called .dev for a reason Aug 08 17:06:37 koen: Is it safe to use angstrom-2008.1 Aug 08 17:06:48 and people have to manually select it, so if things break it's their fault Aug 08 17:07:00 as in "doctor, it hurts when I poke in my eye" Aug 08 17:07:05 KhemWork: it should be Aug 08 17:07:18 koen: ok Aug 08 17:07:27 KhemWork: it works on my at91 board Aug 08 17:07:41 I can try it on osk Aug 08 17:07:52 koen : well i have a different view on that, but thats a discussion that we would probably have to do in OEDEM Aug 08 17:11:29 03koen 07org.oe.dev * r6fcbcfd5... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom-2008.1: fix typo in comments Aug 08 17:17:52 angstrom-2008 is with 2.6.22 wow thats nice Aug 08 17:18:09 koen: are there any plans for GCC 4.2 Aug 08 17:18:57 not before angstrom-2007.1 sees a release :) Aug 08 17:19:05 k Aug 08 17:19:32 I wish 2008 would ship with eglibc Aug 08 17:20:01 re Aug 08 17:21:45 KhemWork: if eglibc works it will ship with glibcm, uclibc and eglibc flavours Aug 08 17:22:35 wow cool Aug 08 17:25:18 koen: where can I find the docs on bitbake -c options ? Aug 08 17:27:08 bitbake -c listtasks Aug 08 17:27:17 that will list all tasks per recipe Aug 08 17:32:48 koen: in OE Build Configuration: it would be nice to report LIBC being used Aug 08 17:35:28 line 428: oe_runconf: command not found Aug 08 17:41:25 03florian 07org.oe.dev * rd8f46a90... 10/ (1 packages/tasks/task-gpe-base.bb): gpe-task-base: Back to Vera fonts... Dejavu is too big. Aug 08 17:41:41 03florian 07org.oe.dev * rd8f46a90... 10/ (1 packages/tasks/task-gpe-base.bb): gpe-task-base: Back to Vera fonts... Dejavu is too big. Aug 08 17:41:54 03florian 07org.oe.dev * rec5c1ed6... 10/ (5 files in 3 dirs): gpe-bootsplash: Update logos in order to reflect latest GPE release. Aug 08 17:42:06 03florian 07org.oe.dev * r445066af... 10/ (1 packages/gpe-bootsplash/files/splash-l.svg): gpe-bootsplash: Fix text Aug 08 17:42:19 03koen 07org.oe.dev * r6fcbcfd5... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom-2008.1: fix typo in comments Aug 08 17:42:31 03koen 07org.oe.dev * r4f6697ba... 10/ (1 packages/xorg-lib/pixman_0.9.4.bb): Aug 08 17:42:43 pixman: add 0.9.4 Aug 08 17:42:51 * the xserver and cairo will depends on this in upcoming releases Aug 08 17:43:00 03koen 07org.oe.dev * rbbb9c3e2... 10/ (1 packages/cairo/cairo_git.bb): cairo: add pixman to DEPENDS for git version Aug 08 17:43:11 03koen 07org.oe.dev * re9119f5e... 10/ (1 packages/glibc/eglibc-package.bbclass): eglibc: add eglibc-package.bbclass based on glibc-package.bbclass Aug 08 17:43:20 03koen 07org.oe.dev * r76d3b20f... 10/ (1 packages/glibc/eglibc_svn.bb): eglibc: use eglibc-package.bbclass Aug 08 17:43:31 03koen 07org.oe.dev * r9337a5ff... 10/ (1 packages/ffmpeg/ffmpeg_svn.bb): fffmpeg svn: fix build Aug 08 17:47:48 mtn: 2 heads on branch 'org.openembedded.dev' Aug 08 17:47:48 mtn: [left] 70aa18fe06d3e3a4ad005666e02d3c53f91a5fa1 Aug 08 17:47:48 mtn: [right] 9337a5ff3440772705d464c123d99b9bcbe72cd0 Aug 08 17:47:48 mtn: misuse: you have no private key to make signatures with Aug 08 17:47:48 mtn: misuse: perhaps you need to 'genkey ' Aug 08 17:49:02 khem yes Aug 08 17:49:08 take one for now Aug 08 17:49:27 mtn -r 9337a5ff3440772705d464c123d99b9bcbe72cd0 up Aug 08 17:49:30 woglinde: heh left or right Aug 08 17:49:36 depends Aug 08 17:49:51 either you will have koens or florians changes Aug 08 17:49:53 thats the one I have Aug 08 17:50:01 koens is what I want Aug 08 17:50:08 or wait 20 minutes or so for the automerger to kick in Aug 08 17:50:16 ah i see Aug 08 17:50:17 ah Aug 08 17:50:26 didnt know about autokicker yet Aug 08 17:50:40 aeh automerger Aug 08 17:50:49 so how does automerger work Aug 08 17:51:17 someone does a pull update merege push :) Aug 08 17:51:28 Crofton|home: no, it's automated :) Aug 08 17:51:57 Crofton|home: that's where the oe@openembedded.org commit come from Aug 08 17:52:29 its a push-pull automerger :) Aug 08 17:52:38 I can not find genkey on my host Aug 08 17:52:46 mtn genkey? Aug 08 17:53:08 * KhemWork this shell should be closed and new one opened Aug 08 17:53:10 mtn enkey you mean? Aug 08 17:53:14 ups genkey Aug 08 17:53:23 no I will go to basketball training Aug 08 17:53:24 bye Aug 08 17:58:55 koen : bootstrap-image for fic-gta01 done :) Aug 08 18:03:06 steliosk, I tried another C++/libtool lib and had failure on Efika Aug 08 18:03:21 steliosk: people still use bootstrap-image? Aug 08 18:03:24 but can build package on OSK (without -lsdtc++ hack) Aug 08 18:03:38 adding angstrom to the front is to many cahracters :) Aug 08 18:04:41 its a benchmark :) Aug 08 18:06:11 it feels like there is a to stop isse, one need ppc toolchain, two need libtool with C++ target Aug 08 18:06:12 Crofton|home : you mean same toolchain different lib to build ? Aug 08 18:06:19 no Aug 08 18:06:27 I mean it works on arm toolchain Aug 08 18:06:33 but not on ppc toolchain Aug 08 18:06:37 same version libtool Aug 08 18:07:55 all-angstrom-linux-gnueabi armv5te-angstrom-linux-gnueabi omap5912osk-angstrom-linux-gnueabi Aug 08 18:08:03 are there any tiny images that are current and work? Basically I just need a shell, mtdutils, etc. Aug 08 18:08:04 all of them are target packages right Aug 08 18:08:48 cbrake : angstrom-minimal-image is the smallest Aug 08 18:08:57 steliosk: thanks Aug 08 18:13:25 Crofton|home : No idea why :( Probably will need to make some simple "hello world" programs to see the output and determine where the problem is Aug 08 18:15:22 yeah Aug 08 18:15:28 that is a good idea Aug 08 18:16:15 I need to the worlds simplest c++ libtool lib Aug 08 18:16:18 tomorrow Aug 08 18:59:22 re Aug 08 19:01:23 03ifaistos 07org.oe.dev * r0312f6be... 10/ (7 files in 3 dirs): Aug 08 19:01:27 packages/asterisk/asterisk_1.2.24.bb: New package asterisk_1.2.24 Aug 08 19:01:34 Digium release note :Version 1.2.24 is the final 1.2 release that contains normal bug fixes. Aug 08 19:01:39 The 1.2 branch will only be maintained with security fix releases from now until it is completely deprecated Aug 08 19:01:48 mental note : get 1.4 working from the bugtracker Aug 08 19:01:56 03cbrake 07org.oe.dev * r1d12a1b6... 10/ (13 files in 3 dirs): compulab-pxa270-2.6.22: add 2.6.22 kernel Aug 08 19:17:59 * * OE Bug 2540 has been RESOLVED (FIXED) by skoroneos(AT)digital-opsis.com Aug 08 19:18:01 * *  Asterisk 1.4.x recipe Aug 08 19:18:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2540 Aug 08 19:19:41 that's all for today. Aug 08 19:19:47 bye Aug 08 19:32:17 03ifaistos 07org.oe.dev * r2e543815... 10/ (5 files in 3 dirs): Aug 08 19:32:27 packages/asterisk/asterisk_1.4.7.bb : New package asterisk 1.4.7 Aug 08 19:32:36 Package was created by Ovidiu Sas, posted on bugtracker as #2540 Aug 08 19:37:13 03koen 07org.oe.dev * r7b2cd756... 10/ (1 packages/asterisk/asterisk_1.4.7.bb): asterisk 1.4.7: try to make conform a bit more to the styleguide Aug 08 19:45:31 moniing Aug 08 19:49:47 hey hrw Aug 08 19:50:59 koen: ready for tomorrow breakage? Aug 08 19:51:54 I don't expect too much breakage Aug 08 19:52:40 koen: asterisk 1.4.10 was released. Aug 08 19:53:37 osas: then provide recipe for it? Aug 08 19:54:27 is there any way to speed up udev on boot, or do most people just use static dev entries if you want fast boot? Aug 08 19:54:42 just renaming the files will do it Aug 08 19:54:59 cbrake: disable nearly all virtual consoles in the kernel Aug 08 19:55:48 cbrake: use busybox/mdev? Aug 08 19:56:44 thanks for the suggestions -- I'll give them a try ... Aug 08 20:02:13 re Aug 08 20:35:12 hrw, should we reduce the number of consoles as a general rule? Aug 08 20:35:26 re crofton Aug 08 20:35:36 hi Aug 08 20:35:53 Crofton|home: iirc it is only possible by extra patch for kernel Aug 08 20:36:05 Crofton|home: and it is good stuff Aug 08 20:36:18 hm, guess I need to look closely next time I am config'ing a kernel Aug 08 20:36:19 he hrw Aug 08 20:45:21 re khem Aug 08 20:45:26 argh... Aug 08 20:45:41 why 2.5" -> 2.5" ata cables must be so hard to find.. Aug 08 20:45:43 woglinde: whats up Aug 08 20:46:14 khem I am still working at midpath Aug 08 20:46:20 hrw: because usually you hook up the disk directly to the board Aug 08 20:46:30 s/board/connector/ Aug 08 20:47:11 koen: which on alix means 90° hdd <>board Aug 08 20:47:20 nice Aug 08 20:47:29 call it 'art' and sell it for $$$$$$$ Aug 08 20:47:32 woglinde: nice Aug 08 20:48:09 hrw: I remember something about 90° risers Aug 08 20:49:05 koen: I need cable Aug 08 20:49:19 for now I know where to buy one in UK Aug 08 20:50:31 * koen hugs efika board for HD mount Aug 08 20:51:24 koen: alix has other nice thing - only 12V needed Aug 08 20:51:29 dc-dc onboard Aug 08 20:51:44 nice Aug 08 20:52:03 a picopsu is nice as well Aug 08 20:52:11 indeed Aug 08 20:56:09 koen: adding eglibc.bbclass helped I think now I see all libraries are installed Aug 08 20:56:22 yay! Aug 08 20:56:27 koen: it still does not boot though :( Aug 08 20:56:43 * koen revokes his rejoicing license Aug 08 20:56:46 khem no init? Aug 08 20:57:30 woglinde: it is there Aug 08 20:57:40 but it does not find it Aug 08 20:57:48 khem sorry I meant init dont start Aug 08 20:57:55 hm maybee softfloat problem Aug 08 20:58:04 the kernel finds init but cant execute it Aug 08 20:58:13 Kernel panic - not syncing Attempted to kill init Aug 08 20:58:19 yes Aug 08 20:58:21 oh well the error message is different Aug 08 20:58:31 from last time so there is some progress Aug 08 20:58:31 hm Aug 08 20:58:33 sure? Aug 08 20:58:42 last time I mean yesterday Aug 08 20:58:44 then they have changed Aug 08 20:59:50 I think I will use chroot Aug 08 20:59:53 and debug this Aug 08 21:03:37 koen: http://blog.haerwu.biz/2007/08/08/why-alix-board-is-nice-and-why-not-quite/ Aug 08 21:06:54 * koen reads Aug 08 21:08:09 hrw: is a c7500 detectable at runtime? Aug 08 21:09:25 "A display of the 250-horsepower Intel Chopper, built by Intel and Orange County Choppers to showcase Intel embedded technology. Features "a digital dashboard, GPS navigation, wireless connectivity, fingerprint recognition for security, and digital cameras that replace rear-view mirrors," MontaVista says" Aug 08 21:09:44 ui orange county choppers Aug 08 21:09:50 I know this tv-series Aug 08 21:10:20 MACHINE=intel-chopper :) Aug 08 21:11:07 koen: rahter not Aug 08 21:11:23 and a 100 000 USD price tag Aug 08 21:11:57 chouimat: that's like 5 euros nowadays Aug 08 21:12:08 ouch Aug 08 21:12:10 koen yup ... Aug 08 21:14:56 hrw, my mother grew up sort of near Bromley Aug 08 21:15:08 Crofton|home: I never was there Aug 08 21:16:18 but I see from your blog you may go there Aug 08 21:16:34 heureka Aug 08 21:16:38 midpath is in oe Aug 08 21:16:46 I need to go by sometime and take a picture of the neighborhood Aug 08 21:16:50 only some work to do for cacao-clcd Aug 08 21:16:53 only some work to do for cacao-cldc Aug 08 21:17:30 nice lookin bike, shame about the branding Aug 08 21:17:54 koen: and it is not c7500 but sl-7500 Aug 08 21:18:07 Crofton|home: tried to get uni sponsoring for OEDEM? Aug 08 21:18:17 ah right Aug 08 21:18:19 oedem Aug 08 21:18:38 need to ask rschuster if tarent will do some sponsoring Aug 08 21:19:05 oedem.. need to talk with mallum about it Aug 08 21:19:49 koen, that would be tough Aug 08 21:20:18 I need some more contracting work so I can list it as a business expense :) Aug 08 21:20:36 http://en.wikipedia.org/wiki/Crofton_Park Aug 08 21:20:41 source of irc nick :) Aug 08 21:22:39 Crofton|home: we have offers of sponsorships Aug 08 21:22:51 iirc TT this time Aug 08 21:23:05 we are hitting the big time! Aug 08 21:23:56 http://www-int.etec.uni-karlsruhe.de/seiten/conferences/wsr08/ Aug 08 21:24:07 I really should submit a paper to this workshop Aug 08 21:24:15 that wouild be easy to get money for Aug 08 21:24:18 from uni Aug 08 21:24:38 you should look into it :) Aug 08 21:26:19 crofton karlsruhe is nice Aug 08 21:26:40 I have a few months to come up with a topic Aug 08 21:27:24 I need to see if I can line up funding from another org also, need to bring in as much travel money as possible Aug 08 21:28:30 * KhemWork noticed that he had init=/bin/busybox sh and that does not work Aug 08 21:28:48 koen: EGLIBC BOOTS !!!! Aug 08 21:29:15 khem hui Aug 08 21:29:22 khem what was the error? Aug 08 21:29:25 congrats!! Aug 08 21:29:37 woglinde: I had wrong init in my bootargs Aug 08 21:29:56 sometimes its nice to visit uboot and kick it Aug 08 21:30:07 khem lol Aug 08 21:31:15 next thing I will do is to setup locale generation in cross env which is one of inherent advantages we have in eglibc Aug 08 21:31:44 then try OE on OMAP2430 Aug 08 21:31:49 grrr Aug 08 21:31:53 I want one of those Aug 08 21:32:09 2430s? Aug 08 21:32:29 I have one for a limited time so I want to use it in good time Aug 08 21:33:53 Crofton|home: so how does source pulling works in OE does it pull the sources from several localtions or are all srcs tar files stored on some central location Aug 08 21:34:35 both Aug 08 21:34:46 ok Aug 08 21:34:50 It can pull from pretty much anywhere net accesible Aug 08 21:34:57 but angstrom keeps a source cache Aug 08 21:35:11 yeah its on my host too Aug 08 21:35:21 or do u mean a cache elsewhere Aug 08 21:35:23 You can pull from a local source dir Aug 08 21:35:39 that would be cool Aug 08 21:35:45 watch the screen Aug 08 21:35:45 how do you do that Aug 08 21:35:56 file:// Aug 08 21:36:10 I use to export from svn and use that Aug 08 21:36:18 this was a long time ago Aug 08 21:36:23 k Aug 08 21:36:28 I now grab straight from svn Aug 08 21:36:35 this is for my SDR work Aug 08 21:36:45 thats better with good DSL connections these days Aug 08 21:36:57 I just upgraded my connection to 6Mbps Aug 08 21:38:10 I certainly get my moenys worth from my DSL line Aug 08 21:38:18 me too Aug 08 21:38:22 although the uni conncection is much faster :) Aug 08 21:38:32 besided my family watches streaming videos too :) Aug 08 21:38:45 heh Aug 08 21:38:56 * hrw -> bed Aug 08 21:38:57 bye Aug 08 21:39:00 gn Aug 08 21:39:03 nite hrw Aug 08 21:39:11 good night all Aug 08 21:39:44 I find that there are few left on US East and west cost Aug 08 21:39:52 s/cost/coast/ Aug 08 21:39:58 heh Aug 08 21:40:44 and after working hours on east coast Aug 08 21:42:24 So OE supports eglibc now? Aug 08 21:42:38 yep Aug 08 21:43:10 it builds and boots atleast Aug 08 21:43:45 Crofton|home: Though the password is empty for root when I ssh into the box it still asks for it and I just press enter Aug 08 21:43:47 yay! Aug 08 21:43:53 good, I will put that OE supports uclibc, eglibc, and glibc in my thesis :) Aug 08 21:44:04 yes sure :) Aug 08 21:44:14 crofton hehe Aug 08 21:44:15 hopefully no one actually reads it Aug 08 21:44:27 khem next week I will test eglibc too Aug 08 21:44:31 and OE will be the first one to have it in working sense Aug 08 21:44:43 we need an autobuild farm Aug 08 21:44:48 woglinde: thanks I would like someone with PPC do stuff on it Aug 08 21:44:55 hm Aug 08 21:45:01 I only have a simpad for now Aug 08 21:45:04 we can try building for efika Aug 08 21:45:09 there is some work we did for ppc85xx Aug 08 21:45:09 qemu-ppc? Aug 08 21:45:14 qemu-mips? Aug 08 21:45:20 heh qemu-all Aug 08 21:45:38 hm right Aug 08 21:45:49 I am taking a day of tomorrow as its my birthday :) Aug 08 21:45:51 I have german settop box dbox2 Aug 08 21:45:56 which ppc mpc823 Aug 08 21:46:00 +is Aug 08 21:46:14 oh Aug 08 21:47:27 mpc8540 and mpc8548 I think Aug 08 21:48:44 woglinde: I have angstom-console-image running now suppose I want to add some more packages to the build how would I do it Aug 08 21:49:22 like say ntpdate is one application I need I am sure there is a way to add it to my OE build without much hassle Aug 08 21:50:08 bitbake whatever you like scp it over and do ipkg install Aug 08 21:50:36 ah cool Aug 08 21:50:40 ANGSTROM_EXTRA_INSTALL += "apps I want" in local.conf Aug 08 21:50:51 so is ipkg something like dpkg Aug 08 21:51:08 it's dpkg + apt in one Aug 08 21:51:14 with a zillion bugs added Aug 08 21:51:18 oh wow Aug 08 21:51:54 dpkg bugz * apt bugz = ipkg bugz ? Aug 08 21:52:03 and then some Aug 08 21:52:15 koen: I just had a look at the state of sync between poky and OE.dev and its not bad. I had a go at merging our base.bbclass and bitbake.conf changes Aug 08 21:52:23 (both ways) Aug 08 21:52:26 RP: nice Aug 08 21:52:30 * RP hugs meld Aug 08 21:52:34 hi rp Aug 08 21:52:40 hi woglinde Aug 08 21:52:50 RP: I like apples FileMerge.app Aug 08 21:52:56 rp merge the git.bb back to poky *g* Aug 08 21:53:04 * koen is too dumb to figure out kdiff3 Aug 08 21:53:19 koen hm select always c Aug 08 21:53:23 woglinde: What did you do to it? Aug 08 21:53:24 *hide and duck* Aug 08 21:53:32 rp making it working? Aug 08 21:53:46 without sneaking to many hostapps Aug 08 21:53:51 +in Aug 08 21:53:54 woglinde: I've only used git-native ;-) Aug 08 21:53:58 RP: I'd like to have the bbclasses in sync between poky and OE at all times Aug 08 21:54:10 yeah even git-native fails with zlib.h Aug 08 21:54:11 we can work on syncing packages after that :) Aug 08 21:54:14 without Aug 08 21:55:25 koen: I try my best with that, its not that bad atm... Aug 08 21:55:42 koen: package.bbclass being the main exception :/ Aug 08 21:55:46 yeah Aug 08 21:56:07 I usually wait a few days before looking at poky changes to let your autobuilder catch up Aug 08 21:56:22 and insane.bbclass. I need to ask ross whether he wants the changes in the before I overwrite them with zecke's version ;-) Aug 08 21:56:55 koen: Probably a good move :) Aug 08 21:57:10 woglinde: so I see. Your changes make sense ;-) Aug 08 21:57:14 * koen wants to get seppuku working Aug 08 21:58:05 woglinde: However you inheirt native then inherit git, you lose ;-) Aug 08 21:58:24 RP: is http://bugs.openembedded.org/show_bug.cgi?id=2722 a sideeffect of recent bitbake changes? Aug 08 21:58:33 rp hm why? Aug 08 21:58:54 woglinde: Check the contents of say do_install after you've done that Aug 08 21:59:00 RP: the debian policy translate x86, i386, i486, i586, i686 and pentium name all to i386 arch name in deb package, so I changed the control file generation of package_deb.bbclass to translate this too... what do you think about this approach? do you see any problem? I believe that another good thing to do is translate the final names of packages too... Aug 08 21:59:02 woglinde: native must come after autotools Aug 08 21:59:21 rp hm no Aug 08 21:59:30 I had problems with that Aug 08 21:59:50 because native overwrites the do_staging Aug 08 22:00:09 and that catched me on other package Aug 08 22:00:15 vivijim: Do the tools enforce that? I've prefer to leave that open if possible Aug 08 22:00:43 woglinde: native before autotools will break. Trust me on this Aug 08 22:00:52 hm it works Aug 08 22:01:01 otherwise I wouldnt check it in Aug 08 22:01:07 woglinde: What does its do_install look like? Aug 08 22:01:17 * koen notices it's midnight Aug 08 22:01:17 native? Aug 08 22:01:19 woglinde: You just think it works ;-) Aug 08 22:01:21 koen yes Aug 08 22:01:28 yes, dpkg and apt translate everything to i386 so it is really impossible to install i586 package using dpkg and apt Aug 08 22:01:33 RP: ^ Aug 08 22:02:00 vivijim: Are you suggesting to merge the feeds? Aug 08 22:02:33 AIUI apt refuses to install foo_i586.deb Aug 08 22:02:56 RP: only for deb packages... like dpkg and apt do Aug 08 22:02:59 rrp -> http://pastebin.ca/650403 Aug 08 22:03:39 03koen 07org.oe.dev * r17a80253... 10/ (4 files in 2 dirs): asterisk: add missing files Aug 08 22:03:44 03rpurdie 07org.oe.dev * r3ac53a3a... 10/ (1 packages/lsof/lsof_4.78.bb): lsof: Build fixes from poky to avoid /usr/include/ Aug 08 22:03:49 03woglinde2 07org.oe.dev * r846702a8... 10/ (9 files in 3 dirs): (log message trimmed) Aug 08 22:03:49 midpath: inital support for midpath Aug 08 22:03:49 * puts java-cldc1.1 support in oe Aug 08 22:03:49 * split in servals packages, because more than one native UI is Aug 08 22:03:49 support Aug 08 22:03:51 * its support for now qte, qt3x11, x11 and awt Aug 08 22:03:53 * there is more work to do for integrate more modules Aug 08 22:03:54 vivijim: We can't merge the feeds, that would break the way OE works Aug 08 22:03:55 03rpurdie 07org.oe.dev * r5866a539... 10/ (1 conf/bitbake.conf): bitbake.conf: Merge in section and dependency setting for -dbg and -dev packages (from poky). Start to allow multiple entries in PKG_CONFIG_PATH (from poky) Aug 08 22:03:59 03rpurdie 07org.oe.dev * rcfffbf72... 10/ (1 classes/gtk-icon-cache.bbclass): gtk-icon-cache: Add comment from poky's version Aug 08 22:04:09 vivijim: We could refer to them all as i386 though Aug 08 22:04:12 03rpurdie 07org.oe.dev * r7c8c03be... 10/ (1 classes/base.bbclass): base.bbclass: Tweaks from Poky (extra comment, lib handling bugfix to not clobber a variable name and unset DISTRO (to unbreak setting DISTRO as an environment variable) Aug 08 22:04:18 03koen 07org.oe.dev * r122762c7... 10/ (3 files in 2 dirs): autoconf: make 2.61 the default, as discussed on the mailinglist Aug 08 22:04:58 koen: gcc still needs autoconf 2.13 on its top dir Aug 08 22:05:25 AFAIK we don't regenerate it's autofoo Aug 08 22:05:30 i hope OE just uses whatever comes with gcc and do not regenerate Aug 08 22:05:38 I suppose I should fire of a build or two soon Aug 08 22:05:44 cool as long as thats the case its does not matter Aug 08 22:05:59 RP: I'm not sure if merge is the correct world for this... I believe that all feeds can coexist but when generating the debian package just translate the file name package and the Architecture field in debian/control file Aug 08 22:06:14 RP s/world/word/g Aug 08 22:06:22 vivijim: Yes, I can understand why we might need to do that Aug 08 22:07:13 * Crofton|home needs to make a hello,world shared library using libtool and compile for Efika Aug 08 22:14:15 woglinde: can midpath-qt be built against qt 4.3 with qt3 compat enabled? Aug 08 22:14:29 RP can or can't? if can't take a look at cputable inside dpkg source code but If can, do you believe that would be a good solution add this kind of patch to package_deb.bbclass: Aug 08 22:14:29 if bb.data.getVar('TARGET_ARCH', d, 1) in ['x86', 'i386','i486','i586','i686', 'pentium']: data = 'i386' and change the file name too? Aug 08 22:14:39 koen dont know Aug 08 22:14:49 I didnt try Aug 08 22:15:09 but It should not be so hard to convert it Aug 08 22:15:26 I can ask guillaum if he can do this Aug 08 22:16:12 vivijim: I agree we need to do something like that. I'd prefer something more extendable than the if though ;-) Aug 08 22:16:39 RP: uhm I got your point... Aug 08 22:18:19 vivijim: The neat thing here is that TARGET_ARCH is an override so things like DPKG_ARCH = "${TARGET_ARCH}" DPKG_ARCH_x86 = "i386" will work Aug 08 22:18:54 at least I think it is an override... Aug 08 22:21:05 I believe so Aug 08 22:21:07 it is Aug 08 22:24:20 RP: so if I understood correctly all that we need to change is the fields.append(["Architecture: %s\n", ['TARGET_ARCH']]) to fields.append(["Architecture: %s\n", ['DPKG_ARCH']]) isn't it? Aug 08 22:25:24 RP: I was crazy about the file name... dpkg-deb that do it using the Architecure field of control... Aug 08 22:26:06 * vivijim needs more coffe Aug 08 22:28:20 * Crofton|home needs more beer Aug 08 22:31:17 hehehehehe good idea! Aug 08 22:39:24 * koen wonders why bsd utils sucks so hard Aug 08 22:39:31 'sleep 10h' doesn't work Aug 08 22:40:09 hehe Aug 08 22:40:38 also doesn't work Aug 08 22:43:07 args Aug 08 22:43:32 how can I commit with monotone and leave one file out Aug 08 22:44:31 --exclude --exclude Aug 08 22:44:36 okay Aug 08 22:44:39 thanks Aug 08 22:50:09 * Crofton|home updates exmap-console from Poky Aug 08 22:54:42 yay exmap-console builds Aug 08 22:54:45 thanks open hand! Aug 08 22:57:37 vivijim: Yes, you just need to change that Aug 08 22:57:43 Crofton: :) Aug 08 22:57:53 Crofton: That got some bugfixes today Aug 08 22:58:26 I need it working so I can write a paper based on some numbers for it soon :) Aug 08 22:58:32 RP good idea, I'll try this later Aug 08 22:59:22 Crofton|home: Its nice to see people using it :) Aug 08 22:59:59 If the paper is not too bad, I'll let you guys know where it is :) Aug 08 23:00:05 cacao-cldc is now in oe Aug 08 23:00:08 We started to use SRCREV in Poky and the crachs appeared :/ Aug 08 23:00:11 cracks Aug 08 23:00:22 Crofton|home: Please do! :) Aug 08 23:00:44 woglinde: cool! Aug 08 23:01:02 RP: I use SRCREV locally with some care Aug 08 23:01:08 flo_lap but there is still work to do Aug 08 23:01:24 (makes committing changes annoying, since I have to change SRCREV -> SRCDATE everytime) Aug 08 23:01:39 koen: You've not tried PREFERRED_VERSION_foo = "...${SRCREV}" then? Aug 08 23:01:44 nope Aug 08 23:01:59 koen: It breaks but I have a fix for bitbake Aug 08 23:02:02 no angstrom release yet :) Aug 08 23:02:08 woglinde: yeah... the whole thing is quite complex... Aug 08 23:02:25 flo_lap I saw the hello world midlet on qte2 Aug 08 23:02:49 * koen heads to bad Aug 08 23:02:50 ehm Aug 08 23:02:51 bed Aug 08 23:02:54 'night all Aug 08 23:02:56 gn Aug 08 23:02:57 I hope I made the most thinks right Aug 08 23:02:58 mee too Aug 08 23:03:00 nite koen Aug 08 23:03:02 'night koen Aug 08 23:03:06 nite all others Aug 08 23:03:29 'night woglinde Aug 08 23:03:29 flo_lap I am glad if you would test it under gpe Aug 08 23:03:51 around this time, this channel sounds like the end of a Walton's episode Aug 08 23:03:52 but I am away at friday for the whole weekend Aug 08 23:03:57 no internet Aug 08 23:04:28 RP, update from Poky enough for you? Aug 08 23:05:06 Hi. I noticed there is not a package like libcgi-perl for OE. I'm trying to compile a new one. Aug 08 23:05:15 Creating a .bb. Aug 08 23:05:23 woglinde: i'll try if i manage to get a free minute Aug 08 23:05:28 hrm Aug 08 23:06:47 mmm. It seems it's part of the Perl core... never mind. Reading.... Aug 08 23:08:36 holy leaping lizard batman, three heads! Aug 08 23:09:11 Crofton|home: fine Aug 08 23:09:21 thanks Aug 08 23:10:09 Have you guys heard of the iRack? Aug 08 23:20:29 hi. I have a small problem with e-wm (e17) via bitbake. it always fails because it can't find -lecore_dbus. edbus is installed. any ideas? Aug 08 23:21:20 Mmm. No CGI module in the default installation. Aug 08 23:37:22 'night all Aug 08 23:37:35 gn Aug 08 23:41:31 kergoth, ping Aug 08 23:43:19 pong Aug 08 23:43:45 hey Aug 08 23:44:02 mindfunk IM'ed by accident today Aug 08 23:44:08 he says hi Aug 08 23:44:20 and Stev Johnson? quit Aug 08 23:44:29 I think I have the name right, that was hours ago Aug 08 23:50:08 hi, any idea about this? Aug 08 23:50:09 http://209.85.135.104/search?q=cache:KG5DCgsTqGsJ:rafb.net/p/6XUSMo91.html+makedepend.%40APP_MAN_SUFFIX%40&hl=cs&ct=clnk&cd=1 Aug 08 23:50:24 I have the same problem ... is it solved already or is it error on my side? Aug 08 23:50:57 I have seen some chater, have you updated? Aug 08 23:52:17 Crofton, this morning, yes Aug 08 23:53:19 ah ... its a bug I think Aug 08 23:53:36 yeah, it may have been fixed today, not sure Aug 08 23:54:29 Crofton|home, ok Aug 08 23:54:36 thanks for your help :) Aug 08 23:54:43 good luck Aug 08 23:54:54 thanks Aug 08 23:57:20 it's not fixed ;) Aug 08 23:59:58 * * OE Bug 2724 has been created by  Aug 09 00:00:01 * * makeinfo-native depends on util-macros-native Aug 09 00:00:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2724 Aug 09 00:39:21 !oebug 2723 Aug 09 00:39:23 * * Bug 2723, Status: NEW, Created: 2007-08-08 05:39 Aug 09 00:39:24 * * waldecir(AT)gmail.com: makedepend do_compile problem: No rule to make target `makedepend.(AT)APP_MAN_SUFFIX(AT)', needed by `all-am'. Aug 09 00:39:25 * * http://bugs.openembedded.org/show_bug.cgi?id=2723 Aug 09 00:39:57 * * OE Bug 2725 has been created by ranaldo(AT)unina.it Aug 09 00:39:59 * * Gpe login respawn continuosly Aug 09 00:40:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2725 Aug 09 01:11:26 parsing issues in some opie stuff Aug 09 01:56:19 mwester: ping Aug 09 01:58:52 whois thomasg Aug 09 02:10:00 03crofton 07org.oe.dev * r036e2e71... 10/ (5 files in 2 dirs): exmap-console : Update from Poky. Builds now. Aug 09 02:10:05 03woglinde2 07org.oe.dev * r36843ed1... 10/ (12 files in 4 dirs): (log message trimmed) Aug 09 02:10:05 cacao: first support of cacao in oe Aug 09 02:10:05 * for now we only support cacao-cldc Aug 09 02:10:05 * with cacao-cldc and midapth wie have midp support in oe Aug 09 02:10:05 * opera-mini and serval other midlets should work now Aug 09 02:10:07 * installing the packages is not enough Aug 09 02:10:09 * you have to write shellscripts which looks like Aug 09 02:10:11 03rpurdie 07org.oe.dev * rdb976a98... 10/ (1 conf/bitbake.conf): bitbake.conf: Weakly assign DL_DIR so it can be set from the environment (from poky) Aug 09 02:33:24 are there several parsing errors in .dev right now or is there something wrong with me? Aug 09 02:45:31 seriously, that is pretty annoying that bitbake won't let me continue but I know I didn't change the repo so I'm guessing these parse errors are in monotone not just local Aug 09 02:53:55 Zero_Chaos: yeah, I'm seeing them too Aug 09 02:54:27 okay, I deleted them all because it's just a bunch of opie stuff, but really annoying that people leave the repo broken **** ENDING LOGGING AT Thu Aug 09 02:59:57 2007