**** BEGIN LOGGING AT Sun Jul 22 02:59:58 2007 Jul 22 04:34:18 koen: actually no. when I flash the n800 kernel. it does not boot. Jul 22 04:34:54 when I boot the rootfs from sd. it does boot. Jul 22 04:35:09 whichi suppose is pivot_root Jul 22 06:01:08 Hi folks, trying to build OE for the first time. Getting this error: "package gnu-config-native-0.1+cvs20050701-r5: task do_patch: failed" Jul 22 06:01:16 Any ideas? Jul 22 06:02:01 (doing bitbake helloworld) Jul 22 08:14:19 koen : hi Jul 22 08:37:12 hi all Jul 22 09:17:21 03koen 07org.oe.dev * rc9345b9b... 10/ (1 packages/geoclue/geoclue_git.bb): geoclue: cath up with recent developments Jul 22 11:42:33 sweet, my neo bv4 can make calls now Jul 22 11:42:54 the trick: use angstrom, not openmoko...... Jul 22 11:44:00 lol Jul 22 11:46:29 at guadec we found that that the openmoko defconfig breaks suspend/resume Jul 22 11:46:35 the one in OE works.... Jul 22 11:47:24 * koen starts saving for OEDEM 2007 Jul 22 11:47:53 linking is sooo slow Jul 22 11:48:19 mickey|zzZZzz: ping Jul 22 11:50:49 zecke: what's the status of webkit integration into the rssreader? Jul 22 11:51:21 koen: mjs has vacations and I'm trying to find some one to review the API patch Jul 22 11:51:47 koen: the API would be enough for the RSS-reader, for a full fledged browser a lot of is missing Jul 22 11:52:33 koen: but, I follow KISS at that point and only add what it is needed. There are ~100 delegates inside the WebView of the Mac and we currently have 6 of them Jul 22 11:55:46 zecke: did you see http://blogs.gnome.org/nigeltao/2007/07/19/wheres-my-rocket-powered-themes/ ? Jul 22 11:56:08 ah, yes you did :) Jul 22 11:56:51 koen: should I get insulting? Jul 22 11:58:06 koen: at least xan of epiphany turned up in #webkit Jul 22 11:59:38 "at Trolltech, we'll be shipping Qt/WebKit with Qt4.4" Jul 22 11:59:47 so the qt tarbal will get 700MB bigger? Jul 22 12:00:09 yes, but the price will be the same :) Jul 22 12:00:32 I wonder if they will use the fast KJS or the ueber fast QtScript as JavaScript implementation Jul 22 12:02:38 zecke: http://software.twotoasts.de/?page=midori , then click 'news' Jul 22 12:04:10 *YAWN* Jul 22 12:22:59 * * OE Bug 2676 has been created by jstoefer(AT)schuett-elektronik.de Jul 22 12:23:01 * * angstrom-x11-image does not boot on h3850 Jul 22 12:23:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2676 Jul 22 12:23:53 morning ! Jul 22 12:30:04 03koen 07org.oe.dev * re9e7550c... 10/ (3 files in 2 dirs): Webkit: update local copy of Makefiles Jul 22 12:46:01 steliosk, where you looking for me? Jul 22 12:49:24 hm virtual/bla shouled be used when more than package can deliver it? Jul 22 12:49:45 without virtual when it is the same package with other version numbers? Jul 22 12:51:01 hm no Jul 22 12:51:19 update-alternatives-cworth_0.99.154.bb:RPROVIDES_${PN} = "update-alternatives" Jul 22 12:51:25 update-alternatives-dpkg_1.13.22.bb:RPROVIDES_${PN} = "update-alternatives" Jul 22 12:52:54 Crofotn|home: Yes. Did you get that Xilinx II board that was around 200-300$ for academia ? Jul 22 12:55:27 Crofton|home: Yes. Did you get that Xilinx II board that was around 200-300$ for academia ? Jul 22 13:03:45 I have the ml403 Jul 22 13:04:52 http://www.digilentinc.com/Products/Detail.cfm?Nav1=Products&Nav2=Programmable&Prod=XUPV2P Jul 22 13:04:54 this one? Jul 22 13:22:23 Crofton : yes this is the one Jul 22 13:23:51 I was buying on my "business account" Jul 22 13:24:05 I may be able to buy one through a uni account in a few weeks though Jul 22 13:30:38 Crofton : did you find any time to play with it ? Any first impressions ? Jul 22 14:08:08 steliosk, not yet Jul 22 14:21:01 morning Jul 22 14:21:12 hi chouimat Jul 22 14:40:48 03mickeyl 07org.oe.dev * r65577cf9... 10/ (4 files in 2 dirs): Jul 22 14:40:48 * add edje_editor, an editor for edje files Jul 22 14:40:48 * add enna, a media center application based on EFL Jul 22 14:40:48 * add exhibit, a picture viewer application based on EFL Jul 22 14:40:56 03mickeyl 07org.oe.dev * r7eb8746b... 10/ (3 files in 2 dirs): taglib[c]: unify formatting Jul 22 14:40:59 03mickeyl 07org.oe.dev * rf1a35ea5... 10/ (4 files in 2 dirs): libmusicbrainz: update from 2.1.2 to 2.1.4, add 3.0.1 (DEFAULT_PREFERENCE=-1) Jul 22 14:41:04 03mickeyl 07org.oe.dev * rc2ae3a88... 10/ (4 files in 3 dirs): enna: fix svn build (broken locale problem) Jul 22 14:41:11 03mickeyl 07org.oe.dev * r8644022f... 10/ (1 packages/efl1/esmart_cvs.bb): esmart: point to proper edje_cc Jul 22 14:41:15 03mickeyl 07org.oe.dev * r99efc70e... 10/ (1 classes/efl1.bbclass classes/efl_library.bbclass): classes: remove outdated efl1.bbclass, package *.a/*.la in common EFL plugin directories to PN-dev Jul 22 14:41:20 03mickeyl 07org.oe.dev * rd8c169d4... 10/ (1 packages/efl1/etk-native packages/efl1/etk-native_cvs.bb): add etk-native (non-working yet) Jul 22 14:41:26 03mickeyl 07org.oe.dev * refa64132... 10/ (9 files in 5 dirs): efl++: move inside efl1 since it's now a subproject of E17 Jul 22 14:45:52 mickey|zzZZzz: du bist da! Jul 22 14:50:44 morning all Jul 22 14:51:05 hi RP Jul 22 14:52:30 hey RP Jul 22 14:53:19 mickey|zzZZzz: where can I find locale-is-broken.patch ? Jul 22 15:01:12 * RP tries to get Linus to pull the LED and backlight trees... Jul 22 15:01:29 RP: didn't he pull yet? Jul 22 15:02:26 zecke: no Jul 22 15:02:52 your request is +1 week old? Jul 22 15:03:09 zecke: 6 days Jul 22 15:03:09 * mickeyl waves to RP Jul 22 15:03:14 hi mickeyl Jul 22 15:03:14 the day I browsed lkml.org I saw it :) Jul 22 15:03:20 RP: got a minute for a /query ? Jul 22 15:03:25 mickeyl: sure Jul 22 15:13:35 03rpurdie * r839 10bitbake/ (3 files in 3 dirs): Improve certain error messages Jul 22 15:14:21 yeah Jul 22 15:18:29 mickeyl: mtn add packages/e17/enna -R ; mtn commit packages/e17/enna -m "enna: add missing patch" Jul 22 15:19:09 * mickeyl nods Jul 22 15:24:17 I've built an image for my handheld but it hangs after "Freeinig init memory: 92k" Jul 22 15:24:35 is this because the UIDs are all wrong, and if so do I need to rebuild everything as root? Jul 22 15:25:39 (asus 730, used the opie-kdepim-image target) Jul 22 15:26:00 lmm: no idea Jul 22 15:27:05 hangs on init may have EABI userland/non-EABI kernel as reason Jul 22 15:27:09 zecke : just read your reply on the e.V. hre you need 30 founding members to get started Jul 22 15:27:30 zecke : 5 or 7 won't be any problem Jul 22 15:28:03 mickeyl: or wrong console, or wrong floating point emulation, or wrong rootpartition... Jul 22 15:28:24 steliosk: yes, I think we get enough people together but I don't feel like doing the paper work :} Jul 22 15:28:44 zecke: florian said KC could help there Jul 22 15:28:57 BTW anyone knows of any other OE presentations (other = not those mentioned in the wiki) Jul 22 15:29:23 steliosk: mickeyl's Jul 22 15:29:30 steliosk: download.vanille.de -> FOSDEM200*.pdf Jul 22 15:29:37 dunno whether they are linked from the wiki Jul 22 15:29:38 steliosk: but they are in a proprietary format :) Jul 22 15:29:50 heh, thanks to poppler, pdf is not that proprietary Jul 22 15:30:02 bbiab, switching computers Jul 22 15:32:52 hmm, can anyone of your compile libsigc++ on DISTRO=angstrom*? Jul 22 15:33:01 getting heaps of c++ errors here Jul 22 15:33:19 mickeyl: could you hold of merging openmoko kernel recipes Jul 22 15:33:44 mickeyl: at guadec we found out that suspend/resume works with the .dev kernels, but not with the openmoko kernels Jul 22 15:34:02 (running custom userspace) Jul 22 15:35:35 interesting. where's the difference? Jul 22 15:35:54 afaik only the defconfig Jul 22 15:39:36 i hope the problem vanishes with me setting up a .dev environment on buildhost and deprecating the overlay Jul 22 15:39:56 and keeping kernelcoders away from OE ;) Jul 22 15:40:36 * koen votes for using the ezx scripts to add patches to OE from svn instead of pulling straight from svn Jul 22 15:40:57 coming back to my libsigc++ question... anyone? Jul 22 15:41:42 mickeyl : wait a sec Jul 22 15:43:44 mickeyl: 1.2 or 2.0? Jul 22 15:44:36 2.0 Jul 22 15:44:42 both 2.0.17 and 2.0.6 error out here Jul 22 15:44:48 I'm running bb now Jul 22 15:44:50 starting with namespace errors Jul 22 15:44:56 NOTE: package libsigc++-2.0-2.0.17-r1: task do_configure: started Jul 22 15:45:17 no providers? Jul 22 15:45:25 er name of bb file? Jul 22 15:45:31 bitbake libsigc++-2.0 Jul 22 15:45:49 hmmm Jul 22 15:45:50 ok Jul 22 15:45:55 namespace erros Jul 22 15:46:05 03koen 07org.oe.dev * rc293cc4e... 10/ (1 packages/gtk-webcore/midori_0.0.3.bb): Jul 22 15:46:05 midori: update to 0.0.3 Jul 22 15:46:05 * is still using gtk-webcore instead of webkitgdk Jul 22 15:46:05 darn Jul 22 15:46:09 03mickeyl 07org.oe.dev * r706cfac4... 10/ (1 packages/e17/enna/locale-is-broken.patch): enna: add missing patch. thanks koen for spotting Jul 22 15:46:31 guess no one tested that since upgrading toolchain Jul 22 15:46:42 they have no newer release either Jul 22 15:46:44 gcc 4.x c++ cahnges? Jul 22 15:47:31 i'm afraid so Jul 22 15:47:34 probably more strict compliance Jul 22 15:47:55 yeah Jul 22 15:48:04 hi guys Jul 22 15:48:21 mickeyl: I built gobby a while ago with gcc 4.x.x Jul 22 15:48:23 in wonder whether I should just ditch it and going boost instead Jul 22 15:48:32 heh, boost Jul 22 15:48:45 mickeyl: read the mtn mailinglist, nothing but trouble with boost Jul 22 15:48:49 what is libsigc++? Jul 22 15:48:49 which offers more or less the same functionality Jul 22 15:48:51 mtn is dumping boost Jul 22 15:48:53 ewww boost :) Jul 22 15:48:57 koen: really? Jul 22 15:49:05 Crofton: type-safe signal/slot library Jul 22 15:49:22 which i need for EFLpp Jul 22 15:49:26 * Crofton|home needs to figure out signal/slot stuff Jul 22 15:49:49 mickeyl: try poking murrayc Jul 22 15:49:54 this will to communicationi between processes? Jul 22 15:49:56 * chouimat likes signal and slot Jul 22 15:50:17 Crofton no only between objects ... and it also depend on the toolkit you use Jul 22 15:51:00 * chouimat is listening to Komm Gib Mir Deine Hand by Onkel Tom Angelripper on Ein Troepfchen Voller Glueck [Amarok] Jul 22 15:52:06 http://womble.decadentplace.org.uk/c++/syntax-errors.html Jul 22 15:52:39 Crofton|home: here an interesting blog about signal and slot http://labs.trolltech.com/blogs/2007/06/15/boost-signals-slots-with-qt/ Jul 22 15:59:00 the default kernel config for my machine is wrong (doesn't include the touchscreen); how do I get to menuconfig myself when building? Jul 22 15:59:19 bitbake -c configure Jul 22 15:59:22 cd into the directory Jul 22 15:59:32 call make menuconfig ARCH=whatever CROSS_COMPILE=whatever Jul 22 15:59:41 then copy the new .config into your metadata Jul 22 15:59:41 then build again Jul 22 16:00:29 where should be what, linux-2.6.21-hh12 ? Jul 22 16:02:07 ood afternoon, all ! Jul 22 16:03:42 Iam trying to build an angstrom image for netbook right now, and I want it to build against OABI. Jul 22 16:04:27 mickeyl : it fails on angstrom/efika also Jul 22 16:04:56 I have added 'netbook' to the list of OABI - devices in angstroem.conf, however, bitbake tell sme it is going to build against 'gnueabi' - how comes ? Jul 22 16:05:47 steliosk: k, thanks for testing Jul 22 16:05:58 jan_js maybee you made a mistake? Jul 22 16:06:14 mickeyl : Do you need to build log ? Jul 22 16:06:40 steliosk: no, thanks, it's probably the same failure as here. Jul 22 16:06:48 * mickeyl tries an svn checkout Jul 22 16:07:41 woglinde: absolutely possible :-) I tried to copy the way it is done for collie. Jul 22 16:08:13 jan_js please /msg me the line Jul 22 16:13:51 oops Jul 22 16:14:31 sigh Jul 22 16:14:37 still errors out Jul 22 16:14:39 however differently Jul 22 16:15:43 woglinde: and at 2 more places further down the angstrom.conf Jul 22 16:16:23 jan_js did you register? Jul 22 16:16:29 I didnt get msh Jul 22 16:16:32 msg Jul 22 16:16:37 no, not yet. Jul 22 16:17:14 woglinde: can I send it to you as a brief file ? Jul 22 16:17:47 than put it pastebin Jul 22 16:19:39 hi florian Jul 22 16:19:48 re Jul 22 16:22:22 hi florian Jul 22 16:26:58 woglinde: sorry, took a while: http://pastebin.ca/629709 Jul 22 16:28:27 jan_js okay that looks fine Jul 22 16:29:26 woglinde: But bitbake insist on TARGET_OS = "linux-gnueabi" Jul 22 16:29:42 hm is MACHINE=netbook set? Jul 22 16:29:49 please run bitbake with -e | less Jul 22 16:29:58 und look what MACHINE is ste too Jul 22 16:30:30 woglinde: ok Jul 22 16:30:47 ar right Jul 22 16:31:07 ah error found Jul 22 16:31:10 in != not in Jul 22 16:33:03 args sorrry Jul 22 16:33:58 woglinde: ? Jul 22 16:34:21 I looked false at TARGET_OS_UC Jul 22 16:34:33 # ARM_ABI=${@['','oabi'][bb.data.getVar('MACHINE',d) in ['collie','h3600', 'h3800', 'simpad', 'htcwallaby']]} Jul 22 16:34:36 ARM_ABI="oabi" Jul 22 16:35:19 output from bibtake bla -e Jul 22 16:42:55 mmh nasty Jul 22 16:43:01 now it breaks in the toolchain Jul 22 16:43:12 | /home/pkg/oe/fic-gta01/tmp/cross/arm-angstrom-linux-gnueabi/include/c++/bits/functexcept.h:43: error: expected constructor, destructor, or type conversion before 'namespace' Jul 22 16:44:48 sometimes i hate compiled languages Jul 22 16:44:49 hm yes which compiler? Jul 22 16:45:16 mickeyl oe-branch or openmoko? Jul 22 16:45:36 org.openembedded.dev Jul 22 16:45:38 gcc version 4.1.2 Jul 22 16:45:44 hm Jul 22 16:45:45 or rather, g++ Jul 22 16:45:48 use uclibc Jul 22 16:45:51 *g* Jul 22 16:45:56 no option Jul 22 16:46:01 at least not in the short term Jul 22 16:47:07 hm Jul 22 16:47:55 sigh guess it means going boost then Jul 22 16:47:57 mickeyl, did you see the link I pasted to common C++ problems? Jul 22 16:48:10 it is help ful figuring out problems when you move to gcc-4.x Jul 22 16:48:26 Crofton|home: yes, but that's in the toolchain. i doubt that they have a common C++ error in a toolchain header Jul 22 16:48:49 ah Jul 22 16:49:07 mickeyl but why no other has thus error Jul 22 16:49:11 I compile a lot of c++ stuff for angstrom Jul 22 16:49:15 or you using gcc 4.2.1? Jul 22 16:49:23 no, 4.1.2 Jul 22 16:49:34 sigc++ is probably the thing that makes most heavy usage of advanced C++ features Jul 22 16:49:45 so I'm not surprised it doesn't crashes nowhere else Jul 22 16:50:12 oh hm at sigc++ Jul 22 16:50:16 I wondered Jul 22 16:50:31 I thought it was in g++ compilation Jul 22 16:50:40 no, sorry. toolchain builds ok Jul 22 16:50:45 just that damn signal slot library Jul 22 16:50:59 hehe I know Jul 22 16:51:14 used for long in dreambox and dbox2 Jul 22 16:51:32 mickeyl: at least it doesn't use a preprocessor Jul 22 16:51:47 zecke: i have nothing against preprocessor usage once it works Jul 22 16:51:51 grr Jul 22 16:51:58 mickeyl 2.0 or 1.2? Jul 22 16:52:04 woglinde: 2.0 Jul 22 16:52:27 of course it compiles perfectly on x86 Jul 22 16:52:41 which makes me think our toolchain build is to blame Jul 22 16:52:42 mickeyl please look at debian project Jul 22 16:52:50 maybee they have somefixes Jul 22 16:52:56 good idea. let me check Jul 22 16:53:04 Version: 2.0.17-2 Jul 22 16:53:20 no arch exclusion Jul 22 16:54:32 hm we hm no patches Jul 22 16:54:38 nope, no patches Jul 22 16:54:43 just build changes Jul 22 16:54:49 so g++ patches Jul 22 17:00:46 hm my network connections is to slow Jul 22 17:00:57 koen: eek, I have double scrollbars :} Jul 22 17:02:05 zecke: better than no scrollbars :) Jul 22 17:02:17 *g* Jul 22 17:02:20 oooooh Jul 22 17:02:24 SDIO mail on LAKML Jul 22 17:14:58 koen : i saw on oesf that you have a 4gb sd card working in angstrom? Jul 22 17:15:47 my transcent 4gb sdhc card is having problems, i partitioned a new ext2 partition on it then when i try to copy data on to it, it crashes Jul 22 17:16:11 tried to set a different block size also, using mke2fs -b 1024 Jul 22 17:38:02 koen: I've just noitced the SDIO mail too, interesting Jul 22 17:50:56 Hi folks, is there a coLinux setup page somewhere? Telling what base tools needs to be installed? Jul 22 17:52:00 hm last time I tried colinux on winxp I crashed Jul 22 18:05:12 * RP reboots woglinde Jul 22 18:28:42 03Laibsch 07org.oe.dev * r86395f96... 10/ (5 files in 2 dirs): Jul 22 18:28:42 apt: upgrade to 0.7.3 Jul 22 18:28:42 * 0.7.2 has become unfetchable on the Debian mirrors Jul 22 18:28:42 * 0.7.2 was broken (failed do_patch), the same applies for 0.7.3. Jul 22 18:28:42 IOW, it was broken before, it is still broken now, but Jul 22 18:28:42 hopefully less so. Jul 22 18:28:45 * more information in bug 2017 comment 6 Jul 22 18:42:12 RP: the idea is to implement mmc_block on top of the sdio framework Jul 22 18:42:21 RP: so prepare for more sd breakages Jul 22 18:42:34 :-( Jul 22 19:06:35 ping hrw|gone Jul 22 19:21:00 pong Laibsch Jul 22 19:21:11 hi guys Jul 22 19:22:59 * * OE Bug 2677 has been created by  Jul 22 19:23:01 * * nokia 800 kernel not booting Jul 22 19:23:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2677 Jul 22 19:24:44 ljp, what time is it there? Jul 22 19:25:29 05:25 Jul 22 19:26:02 ug up early :) Jul 22 19:27:17 ya. i dont sleep much Jul 22 19:27:25 sleep is overrated Jul 22 19:27:40 heh Jul 22 19:28:16 are you trying angstrom on n770 also? Jul 22 19:28:43 not at the moment.. on a n800 Jul 22 19:28:55 need maemo on the n770 Jul 22 19:28:56 I have a 770, no 800 :( Jul 22 19:29:27 just got the n800 on wednesday Jul 22 19:29:37 cool Jul 22 19:29:44 much nicer display Jul 22 19:29:49 t Jul 22 19:29:54 larger? Jul 22 19:30:00 clearer Jul 22 19:30:02 I find the text small on the 770 Jul 22 19:30:07 and I am getting farsighted Jul 22 19:36:05 so Jul 22 19:36:11 any progress on the palm's recentely Jul 22 20:03:09 hello all Jul 22 20:06:32 hello koen Jul 22 20:07:05 hey likewise Jul 22 20:09:59 koen, you went to guadec? Jul 22 20:10:04 gm Jul 22 20:10:13 gm Crofton|home Jul 22 20:11:05 likewise: yes Jul 22 20:11:30 was it fruitful (in whatever way)? Jul 22 20:11:46 Crofton|home: Sound on the EFIKA userland test (works for me (tm)) http://www.pastebin.ca/629951 Jul 22 20:12:22 likewise: yeah, talked a lot with matthijs from irex Jul 22 20:12:46 interestng, what image do you use? Jul 22 20:13:22 Crofton|home: some angstrom image (I think console, with some EXTRA in there). I'm trying to get it to go again. Jul 22 20:13:26 it=the build Jul 22 20:13:39 I was sorking from bootstrap-image Jul 22 20:14:09 I guess I'll try console imaage Jul 22 20:14:31 I can test in a bit I guess. Jul 22 20:14:47 * koen gets back to harry potter Jul 22 20:15:14 * Crofton|home had read the first chapter .. Jul 22 20:15:57 girlfriend did a reading marathon over the weekend and finished this afternoon Jul 22 20:16:04 I cannot bother with harry Jul 22 20:16:06 :-) Jul 22 20:16:12 likewise: same here, now I have the book :) Jul 22 20:17:28 so who died? Jul 22 20:17:58 ssshhh Jul 22 20:17:59 arg Jul 22 20:17:59 shut up! Jul 22 20:17:59 :) Jul 22 20:17:59 no spoiling for a week or so Jul 22 20:18:02 did you buy the book? Jul 22 20:18:24 zecke: atleast half of year will pass Jul 22 20:21:13 hmm angstrom-minimal-image Jul 22 20:21:17 any tothers? Jul 22 20:22:26 angstrom-console-image? Jul 22 20:22:33 where is that? Jul 22 20:23:36 maybe it's legacy. It used to exist. I am not up-to-date, I have been not been able to follow OE lately... Jul 22 20:23:55 packages/angstrom Jul 22 20:24:39 so ANGSTROM_EXTRA_INSTALL will add packages to the image? Jul 22 20:25:28 Crofton|home: yes, it used to. Please grep to see if that's currently still the case. Jul 22 20:25:34 yeah Jul 22 20:25:43 I was reading the bb file :) Jul 22 20:26:22 ANGSTROM_EXTRA_INSTALL += "libao-alsa-plugin \ Jul 22 20:26:23 mpg321 alsa-utils-aplay alsa-utils-amixer alsa-utils-speakertest madplay" Jul 22 20:26:31 ah Jul 22 20:26:35 ok Jul 22 20:26:40 local.conf? Jul 22 20:26:43 yes Jul 22 20:26:50 took me a while myself... Jul 22 20:27:27 after I got everything working, I switched to uclibc, which broke some things again... that's the current state. Jul 22 20:29:47 well, lets see what I get from this piece of info Jul 22 20:43:10 bye Jul 22 20:52:01 interesting: http://www.burtonini.com/blog/computers/katachi-2007-07-22-20-50 Jul 22 21:10:08 koen, I find it wonderful - bzr, git and svn in one project ;) wonder why no cvs and visual sourcesafe :) Jul 22 21:10:30 and yes, I understand it's not his fault :) Jul 22 21:21:05 I think koen's interest is mainly flipping through photographs, not the number of version control systems involved :-) Jul 22 21:22:50 Wouldn't that be boring? :) Jul 22 21:25:34 yes, it would be too user friendly I guess Jul 22 21:32:32 especially for koen :) Jul 22 21:35:26 koen: I guess at least the breakage is for a good cause :/ Jul 22 21:35:32 Hi folks, I'm trying to build OE for the first time. I'm getting this error: ERROR: No buildable providers available for required build target nano Jul 22 21:35:38 koen: The suspend/resume problem is controlled by a Kconfig option btw Jul 22 21:35:46 Any idea what could cause it? Jul 22 21:36:33 drmikecrowe: did you set BBFILES in local.conf? did you export BBPATH? Jul 22 21:37:05 zecke: Here's BBPATH=/opt/stuff/build:/opt/stuff/org.openembedded.dev Jul 22 21:37:23 drmikecrowe: did you export it? Jul 22 21:37:35 drmikecrowe: you understood your are not forced to use /opt/stuff? Jul 22 21:37:46 zecke: yes, that was from echo $BBPATH Jul 22 21:37:57 Here's what I have in local.conf: BBFILES := "/opt/stuff/org.openembedded.dev/packages/*/*.bb" Jul 22 21:38:10 drmikecrowe: do the test BLA="abc" Jul 22 21:38:15 drmikecrowe: echo $BLA Jul 22 21:38:16 yes, I understand I can put it anywhere Jul 22 21:38:39 drmikecrowe: so did bitbake say it parsed +4000 files? Jul 22 21:39:08 zecke: Yes. I've fixed some sanity errors, not this buildable target message Jul 22 21:39:43 drmikecrowe: I wonder if richards last commit improved such error messages as well Jul 22 21:39:50 drmikecrowe: any special MACHINE, DISTRO? Jul 22 21:40:17 zecke: I checked out last night, not today Jul 22 21:40:19 drmikecrowe: I would propose to use -v or -vv or -vvv to check which provider might not be buildable. This can only be some sort of toolchain Jul 22 21:40:28 Here's the build: http://pastebin.com/d4820f45f Jul 22 21:40:36 morning Jul 22 21:41:24 drmikecrowe: use -vvv to see what is missing Jul 22 21:41:28 zecke: Check this out: NOTE: No providers of build target virtual/gcc Jul 22 21:42:03 are you sure 'MACHINE = "palmtt3"' is correct? Jul 22 21:42:46 zecke: Yes Jul 22 21:43:14 I was trying to build on cygwin earlier, and was getting further. Now, I'm trying to use coLinux. Just setting up Jul 22 21:43:33 I have no palmtt3 config Jul 22 21:43:38 drmikecrowe: this machine, if it even exists in conf/machine, does not set TARGET_ARCH Jul 22 21:44:37 zecke: TARGET_ARCH is set in include/palm.conf Jul 22 21:44:48 zecke: doh, I think I see the issue: BUILD_SYS="i686-pc-cygwin" Jul 22 21:45:21 evening .. Jul 22 21:45:29 looks like coLinux is i486-pc-linux-gnu Jul 22 21:45:31 BobOfDoom: well, I have no OE tree here but if woglinde says there is no "palmtt3.conf"... Jul 22 21:45:43 drmikecrowe: well, this would fail later... Jul 22 21:46:25 zecke: Here: org.openembedded.dev/conf/machine/palmtt3.conf Jul 22 21:46:33 http://www.openembedded.org/repo/org.openembedded.dev/conf/machine/palmtt3.conf Jul 22 21:49:39 drmikecrowe: did you copy the OE tree from cygwin or check it out again? maybe try checking out with monotone again.. it might have missed something? Jul 22 21:50:08 RP: the SD suspend problem? Jul 22 21:50:24 BobOfDoom: I'm wondering if that might be it. I did just copy it (it's a big download) Jul 22 21:50:44 I still have the big OE.mtn file Jul 22 21:51:35 drmikecrowe: yeah you won't have to download anything again if you still have that file. just do mtn --db=OE.mtn checkout or whatever it is again. keep your sources (tarballs) directory though Jul 22 21:51:47 * XorA|gone hunts for his lost P0 phone Jul 22 21:52:23 zecke: It was that BUILD_SYS="i686-pc-cygwin" I was trying to use. It's building now. Thanks BobOfDoom Jul 22 21:52:42 Koen ; making progress Jul 22 21:52:54 drmikecrowe: cool. :) Jul 22 21:54:33 koen : The Linux boot on NBP must be reversible. In other words, it MUST be possible to go back to WinCE. I have altered the partition command and now the CE image will load. Problem is, ha, ha, it will not execute. Gosh, isn't this fun! Jul 22 21:55:32 BobOfDoom: You don't have a T3, do you? Jul 22 21:55:38 koen can you do me a little favor and configure libsigc++ and than show me whats in line 80 of sigc++config.h Jul 22 21:55:41 If only I could talk to Wookey Jul 22 21:55:42 Crofton|home: hmm, 2.6.20.11 in OE misses the sound (ALSA and PPC device 5200), Furthermore, the module seems not to be copied to the rootfs. Jul 22 21:55:55 night Jul 22 21:56:04 nite zecke Jul 22 21:56:09 hey woglinde Jul 22 21:56:16 Crofton|home: missing, as in, you need to run ARCH=powerpc make menuconfig Jul 22 21:56:19 he jonb Jul 22 21:56:27 Crofton|home: Enable sound, and the PPC 5200 device. Jul 22 21:56:30 I mean "wotcha mate" Jul 22 21:56:37 good nite Jul 22 21:57:05 Also, Crofton, your baby asleep? Jul 22 21:57:11 or can do it someone else with a glibc setup Jul 22 21:57:51 hm Jul 22 21:58:30 Does anyone know where I can get an OpenZaurus E17 root image from (that runs off a 2.6.9 kernel)? Jul 22 21:58:50 use angstrom Jul 22 21:59:11 Angsrtrom needs a 2.6.17 kernel, according to koen Jul 22 21:59:37 yes Jul 22 21:59:49 2.6.9 is horrible old with security patches Jul 22 22:00:24 So at the moment, I'm buggered, 'cause I can't get a 2.6.17 kernel to boot on my device. Guess what it is woglinde Jul 22 22:02:07 Gosh everyone's leaving Jul 22 22:02:22 it's me isn't it Jul 22 22:02:30 paranoid, me Jul 22 22:03:04 of course it's you, came and scared everyone off. Jul 22 22:03:04 jonb hm I think you are lost with 2.6.9 Jul 22 22:03:14 *g* Jul 22 22:03:32 but what is the device, I'm curious already :) Jul 22 22:03:35 polyonymous can you please try to build libgsigc++? Jul 22 22:03:49 woglinde, hang on, lemme pull first Jul 22 22:03:59 I'm somwhat outdated Jul 22 22:04:27 44 revs :) Jul 22 22:04:27 but I need it with glibc setup Jul 22 22:04:32 oh that's grand. how the [insert rude word here] am i supposed to get this stuff working, when no one who knows anything about the machine will give me some answers? Jul 22 22:04:34 mostly e17 Jul 22 22:04:35 I'm building with glibc Jul 22 22:04:43 fine Jul 22 22:04:50 poly : it's a Netbook Pro. Jul 22 22:05:02 jonb, I don't even know what's that :) Jul 22 22:05:26 and I'm not sure how you're supposed to get this stuff working... Jul 22 22:05:55 Better than a Zaurus, better than a poxy iPaq, better than any other mobile computing device. Because of the battery and he form factor. Jul 22 22:06:11 Sorry "the form factor" Jul 22 22:06:24 Believe me, it's the business. Jul 22 22:06:39 yeah but 2.6.9 helps you nothing Jul 22 22:06:50 You're right there woglinde. Jul 22 22:07:10 you cand put it in safe watch it every day Jul 22 22:07:16 one time Jul 22 22:07:18 *g* Jul 22 22:07:26 Wot's the matter with 2.6.9 Jul 22 22:08:01 no security support Jul 22 22:08:15 My problem is, it doesn't have support for the Orinoco card (well, I am talking about the only working OS image I have) Jul 22 22:08:52 And, Angstrom rootfs will not work with it. koen says it needs 2.6.17 Jul 22 22:09:23 baking now, had to do some manual merging. Jul 22 22:10:00 polyonymous when it fails please show me line 80 of sigc++config.h Jul 22 22:10:11 will do :) Jul 22 22:11:07 real problem is the bootloader it's a bit wierd and no one seems to know anything about it. Frustrating. Jul 22 22:12:16 BUT : I have OE & bb up and running and have built a zImage of 2.6.17 Jul 22 22:12:28 So at least I am trying!!! Jul 22 22:13:49 woglinde : There are people out there with the knowledge, but no one is sharing. How frustrating is that? Jul 22 22:14:10 jonb sorry openzaurus is dead nobody cares Jul 22 22:14:51 there is an e17 rootfs for it, koen's got screen dumps of it (no files though!) Jul 22 22:15:13 just want to get he machine useable Jul 22 22:15:23 then will get angstrom up on it Jul 22 22:15:29 arrgh Jul 22 22:16:33 woglinde, I came back to see that after parsing it bailed out with no provider, what's the target? Jul 22 22:17:19 sorry Jul 22 22:17:21 libsigc++-2.0 Jul 22 22:17:29 ok :) Jul 22 22:18:40 Oh well, I am continuing various avenues of investigation. As I told koen, if I find out anything truly useful, I will update the Ang / OE wiki. No need to be sorry. I *will* crack this thing. I've beaten much more complex stuff than this, be sure of that. Jul 22 22:19:20 Good night. Jul 22 22:19:23 woglinde, ;-))) /* #undef SIGC_USING_STD */(Symbol) Jul 22 22:19:43 polyonymous right this is the erorr Jul 22 22:19:55 yeah, doesn't take a rocket scientist to see :) Jul 22 22:20:05 it must #define #SIGC_USING_STD(symbol) Jul 22 22:20:07 args Jul 22 22:20:13 #define SIGC_USING_STD(symbol) Jul 22 22:20:18 yeah Jul 22 22:20:26 I think it's autoconf version issue Jul 22 22:20:31 autoheader to be precise Jul 22 22:20:37 hm Jul 22 22:20:45 I think autoheader is right Jul 22 22:21:02 the definition seems wrong Jul 22 22:21:21 Well, it seems to define macro with parameter Jul 22 22:21:30 and my guess is that our autoheader doesn't support it. Jul 22 22:21:56 of course there's an option of no autoheader supporting it... Jul 22 22:22:07 but I'm trying to make sense of it :) Jul 22 22:22:11 yes Jul 22 22:22:19 mickeyl will kiss you Jul 22 22:22:21 *g* Jul 22 22:22:26 hmm... Jul 22 22:22:39 Have you looked at sigc++-config.h.in ? Jul 22 22:22:47 yes Jul 22 22:22:59 Looks like this stuff should never be run through autoheader... Jul 22 22:23:29 I've seen that before when doing ebuild for postgresql 7.* ;-) Jul 22 22:23:29 hm no Jul 22 22:23:41 than the definition is wrong Jul 22 22:23:45 it's just that some stuff shouldn't be fully autoreconfed Jul 22 22:23:55 autoheader seems to do the right think Jul 22 22:24:04 no, Jul 22 22:24:10 the right thing is not to touch the fiel Jul 22 22:24:28 look at like 62 in .h.in Jul 22 22:24:32 line Jul 22 22:26:10 EXTRA_AUTORECONF to exclude autoheader ? Jul 22 22:26:46 hm cat >> config.h after autotools Jul 22 22:27:35 cat what? Jul 22 22:27:52 SIGC_USING_STD(Symbol) Jul 22 22:28:00 or echo Jul 22 22:28:52 it doesnt test anything, so its safe on evry compiler I guess Jul 22 22:29:21 I'd prefer proper fix. Jul 22 22:30:13 * polyonymous is -c rebuilding Jul 22 22:30:42 why do you call namespace std { using ::Symbol; } nothing, anyway? Jul 22 22:31:04 /me? Jul 22 22:31:19 I only spotted the error Jul 22 22:31:25 after mickeyl has problems Jul 22 22:31:26 nevermind Jul 22 22:31:35 http://rafb.net/p/GRjGFL85.html Jul 22 22:31:46 I am not really good at c++ Jul 22 22:32:10 woglinde, anyway, do you want to patch your mtn or do you want me to push it? :) Jul 22 22:32:21 please push it Jul 22 22:32:27 hang on then Jul 22 22:32:30 I am working on fastjar and midpath Jul 22 22:32:36 and its mainly for mickeyl Jul 22 22:34:54 pushed Jul 22 22:38:10 tanks Jul 22 22:38:11 thanks Jul 22 22:38:46 you're welcome :) Jul 22 22:41:08 I hereby declare my resignation from OE and Angstrom Jul 22 22:41:12 Good bye Jul 22 22:41:14 Have fun Jul 22 22:41:30 laibsch why? Jul 22 22:41:32 Laibsch, I think it's not the best you can do. Jul 22 22:41:39 * polyonymous thinks of devel list Jul 22 22:42:00 http://www.angstrom-distribution.org used to be there just a minute ago Jul 22 22:42:04 It is gone now Jul 22 22:42:09 This is still a fiefdom Jul 22 22:42:14 I am sick of dealing with it Jul 22 22:42:23 Sorry, wrong link Jul 22 22:42:32 http://www.angstrom-distribution.org/mw used to be there just a minute ago Jul 22 22:43:05 oh well... Jul 22 22:43:17 People just can't get along well... Jul 22 22:44:20 maybee I am on the wrong devel-list Jul 22 22:44:38 woglinde, the wiki thing. Jul 22 22:45:46 to be true, I find the very idea of wiki so unappealing that arguing about which one is to be used isn't of much importance to me :) Jul 22 22:46:08 03polyonymous 07org.oe.dev * rb9a8109f... 10/ (1 packages/libsigc++-2.0/libsigc++-2.0_2.0.17.bb): Jul 22 22:46:08 libsigc++-2.0: disable autoheader in autoreconf Jul 22 22:46:08 since it uses non-autoheader .h.in which gets broken Jul 22 22:46:19 thank you, CIA, just about time :) Jul 22 22:59:49 polyonymous: I think you should appreciate that I was not forcing my decision on others Jul 22 23:00:02 I prefer a tool and I wanted to be able to use it. Jul 22 23:00:32 Should others like to use it, too, they should have been free to do share it with me. Jul 22 23:00:57 That is the whole idea Jul 22 23:01:04 s/is/was/ Jul 22 23:01:20 Laibsch, I do not object your actions... Jul 22 23:01:39 I actually can't commend regarding the matter, since it's irrelevant to me. Jul 22 23:01:48 It was nonetheless important for me to make sure you understand Jul 22 23:02:03 it is irrelevant to you if somebody deletes your work? Jul 22 23:02:15 OK, not your work, but my work. Jul 22 23:02:24 no, I mean the whole wiki thing. Jul 22 23:02:42 And I tried very hard for it not to escalate Jul 22 23:02:49 I think this is laughable, too. Jul 22 23:02:55 But that is just koen-style Jul 22 23:02:59 as for removal... well... I don't think it's nice thing to do, but then again, I have no good idea on what was going on. Jul 22 23:03:06 Ask hrw tomorrow how I tried Jul 22 23:03:19 ohwell... Jul 22 23:03:33 There are just people you cannot get along with Jul 22 23:03:43 koen is trying very hard to prove he is Jul 22 23:03:44 But anyway I think calling oe a fiefdom may be offending for others. Jul 22 23:03:54 OK, you are right Jul 22 23:04:05 oe is a bit separate Jul 22 23:04:27 oe is not a fiefdom Jul 22 23:04:40 but essentially the only distro left is angstrom Jul 22 23:04:51 and angstrom *is* a fiefdom Jul 22 23:05:02 thus I have no choice but to leave both Jul 22 23:05:41 well... I don't know.. Jul 22 23:05:51 it's a sad story Jul 22 23:06:08 polyonymous can you help me? Jul 22 23:06:16 woglinde, how do I know ahead of time? Jul 22 23:06:28 when I should use virtual/bla and when I can leave the virtual out? Jul 22 23:07:06 woglinde, that's not the part of oe I'm expert in ;-) I think you need to use virtual/ if the feature in question may be provided by more than one package Jul 22 23:07:21 like kernel, iconv, etc. Jul 22 23:07:44 hm I see packages doing that without virtual Jul 22 23:08:04 doing what ? Jul 22 23:08:32 I'd think that the package that relies on particular implementation should do it without virtual Jul 22 23:25:39 godd nite Jul 23 00:05:13 Anybody using coLinux here to compile OE? **** ENDING LOGGING AT Mon Jul 23 02:59:56 2007