**** BEGIN LOGGING AT Thu Apr 08 02:59:56 2010 Apr 08 06:25:15 good morning Apr 08 06:42:36 slightly offtopic: anyone know if there's a way (or addon) to reply to an individual message withing a digest? Apr 08 06:42:48 for thunderbird Apr 08 07:16:22 03Koen Kooi  07org.openembedded.dev * r8775a28f83 10openembedded.git/recipes/nautilus/ (3 files in 2 dirs): nautilus-cd-burner: add 2.25.3 + patches to remove runtime dep on 'eject' and 'gnome-mount' Apr 08 07:16:33 03Koen Kooi  07org.openembedded.dev * readf48fd11 10openembedded.git/recipes/tasks/angstrom-task-gnome.bb: angstrom-task-gnome: add gnome-mount Apr 08 07:16:34 03Koen Kooi  07org.openembedded.dev * ra8ce2ebbef 10openembedded.git/recipes/abiword/abiword_2.8.3.bb: abiword: add 2.8.3 Apr 08 08:12:20 broonie: ping Apr 08 08:23:51 hey guys Apr 08 08:24:44 When a bitbake crashes with "ERROR: This autoconf log indicates errors, it looked at host includes." it's typically due to a -L/usr/lib in the config.log, right? Apr 08 08:57:49 03Koen Kooi  07org.openembedded.dev * r437d812748 10openembedded.git/recipes/networkmanager/network-manager-applet_0.8.bb: nm-applet: enable BT PAN Apr 08 08:57:50 03Koen Kooi  07org.openembedded.dev * r692bd00724 10openembedded.git/recipes/tasks/angstrom-task-gnome.bb: angstrom-task-gnome: add modemmanager Apr 08 08:57:51 03Koen Kooi  07org.openembedded.dev * r186db950ff 10openembedded.git/recipes/gnome/metacity_2.30.0.bb: metacity: remove def_pref from 2.30 Apr 08 08:57:51 03Koen Kooi  07org.openembedded.dev * r6f2b45c1f2 10openembedded.git/recipes/gnome/gnome-terminal_2.30.0.bb: gnome-terminal: add 2.30.0 Apr 08 08:57:52 03Koen Kooi  07org.openembedded.dev * rd11fc0c845 10openembedded.git/recipes/gnome/gnome-bluetooth_2.30.0.bb: gnome-bluetooth: add 2.30.0 Apr 08 08:57:53 03Koen Kooi  07org.openembedded.dev * rc44db2b9e8 10openembedded.git/recipes/networkmanager/modemmanager_0.3.bb: modemmanager: add 0.3 Apr 08 08:57:54 03Koen Kooi  07org.openembedded.dev * r48dd41af75 10openembedded.git/recipes/gnome/gnome-utils_2.30.0.bb: gnome-utils: add 2.30.0 Apr 08 08:57:55 03Koen Kooi  07org.openembedded.dev * r247a0a81b5 10openembedded.git/recipes/gphoto2/ (5 files in 2 dirs): Apr 08 08:57:55 gphoto2: add 2.4.8 Apr 08 08:57:55 libgphoto2: add 2.4.8 Apr 08 09:00:19 03Koen Kooi  07org.openembedded.dev * r826a4c44f9 10openembedded.git/recipes/gnome/gnome-bluetooth_git.bb: gnome-bluetooth git: lower default pref for angstrom Apr 08 09:00:41 porcoesphino: probably -I/usr/include in that case, but yes Apr 08 09:01:10 I've found a reference to it Apr 08 09:01:29 I mean, I grep-ed config.log for 'configure:' Apr 08 09:01:45 and manually went through it so I could of missed something Apr 08 09:02:29 but the only thing I could find, was that it looks for a working copy of readline to use, and searches a bunch of places on my host machine, but in the end cant find a working copy Apr 08 09:02:33 could it be that? Apr 08 09:02:41 ha Apr 08 09:02:46 apparently it is Apr 08 09:03:53 well, since it can't find a working copy of readline I assume it's throwing away those search paths. So I deleted those lines from config.log, and it seems to be running away nicely now Apr 08 09:04:17 am I wrong in my assumption, or is there anything else that might go wrong due to those actions? Apr 08 09:06:37 any python wiz here can tell me how to tell OE not to build (actually install) an egg, but install the package directly unter site-packages/ ? Apr 08 09:06:47 How can i remove the "inherit update-rc.d" for angsrom and shr from bluez? Apr 08 09:12:47 You can't, really. Why do you want to do that? Apr 08 09:13:19 pb__: update-rc.d: /etc/init.d/bluetooth: file does not exist Apr 08 09:14:07 so that file exists for other DISTROs, but not angstrom or shr? Apr 08 09:14:12 angstrom and shr are using udev to start bluez and because of that there is no init script Apr 08 09:14:27 i think so Apr 08 09:14:44 i haven't tried to compile it without the --enable-udevrules Apr 08 09:15:01 ah, I see Apr 08 09:15:22 in that case, I think your easiest option is probably to remove the inherit altogether and implement the same functionality by hand for other distros. Apr 08 09:15:44 alternatively, you might find that something along the lines of: Apr 08 09:15:48 inherit = "update-rc.d" Apr 08 09:15:51 inherit_shr = "" Apr 08 09:15:55 inherit ${inherit} Apr 08 09:15:58 does what you need Apr 08 09:16:13 obviously you'd need to pick a variable name that isn't a reserved word, but you get the idea. Apr 08 09:16:47 pb__: ok, will try that, thx Apr 08 09:24:54 ant_work: pong (please just ask questions!) Apr 08 09:25:49 hi Apr 08 09:26:17 I was creating some kernel configs for 2.6.34-rc3 Apr 08 09:26:29 for machine tosa I could not find CONFIG_TOUCHSCREEN_WM97XX Apr 08 09:27:05 isn't the driver upstream? Apr 08 09:27:44 Yes, it is. Apr 08 09:28:12 What do you mean when you say you could not find that config option? Apr 08 09:28:36 eh..it wasn't just listed in 'touchscreen' Apr 08 09:29:01 I left CONFIG_TOUCHSCREEN_ADS7846 knowing it's wrong Apr 08 09:29:15 So if you take a look at the Kconfig entry for the driver... Apr 08 09:29:29 just wondering if I need some patches Apr 08 09:29:40 as for 2.6.29 Apr 08 09:30:19 You shouldn't have needed any patches for 2.6.29 either. Apr 08 09:30:34 Have you tried looking at the Kconfig entry for the driver? Apr 08 09:30:47 It shows a dependency on AC97_BUS. Apr 08 09:30:59 sorry, can't do it right now Apr 08 09:31:07 ah.I disabled sound... Apr 08 09:31:13 This means that the driver can't be enabled unless you've selected AC97. Apr 08 09:31:28 I see..I'm compiling a minimal kernel Apr 08 09:31:44 That'll be it, you need your AC97 subsystem up and running for AC97 devices. Apr 08 09:31:52 ok, I'll have to enable AC97 for tosa Apr 08 09:31:54 thx Apr 08 09:32:49 If a driver isn't appearing it's almost always due to a dependency not having been enabled. Apr 08 09:39:08 ping mickey|zzZZzz Apr 08 09:42:45 morning Apr 08 09:43:19 hi hrw Apr 08 09:52:43 Hi, bitbaking a xorg-image, I have this http://pastebin.org/141143 error at the end of the file xorg-image-1.0-r0/temp/log.do_rootfs.29204 . Ideas? Apr 08 10:20:51 03Michael Lippautz  07org.openembedded.dev * r163489ebbf 10openembedded.git/recipes/ (3 files in 3 dirs): Apr 08 10:20:51 python-lxml: Fix recipe. Apr 08 10:20:51 * Use latest version Apr 08 10:20:51 * Remove nonworking copy and unused patch Apr 08 10:20:51 * Move recipe to python/ dir Apr 08 10:20:52 Signed-off-by: Michael Lippautz Apr 08 10:31:35 hello Apr 08 10:32:33 when i build my distro, i get: ERROR: '/home/hanibal/oe/openembedded/recipes/madwifi/madwifi-ng_r2826-20071105.bb' failed Apr 08 10:32:33 because it's not on the website anymore Apr 08 10:33:06 how can i prevent a recipe being build, without changing oe? Apr 08 10:35:14 03Koen Kooi  07org.openembedded.dev * re1ead8542b 10openembedded.git/recipes/gnome/gconf-editor_2.30.0.bb: gconf-editor: add 2.30.0 Apr 08 10:35:15 03Koen Kooi  07org.openembedded.dev * r03fdb2997c 10openembedded.git/recipes/gnome/gnumeric_1.10.1.bb: gnumeric: add 1.10.1 Apr 08 10:35:16 03Koen Kooi  07org.openembedded.dev * rcfa7ad3874 10openembedded.git/recipes/policykit/policykit_0.96.bb: policykit: fix pam auth Apr 08 10:35:16 03Koen Kooi  07org.openembedded.dev * r0551e46608 10openembedded.git/recipes/pidgin/pidgin.inc: pidgin: enable audio and vv when available Apr 08 10:52:31 03Koen Kooi  07org.openembedded.dev * r55693a05e9 10openembedded.git/recipes/gnome/gnome-power-manager_2.30.0.bb: gnome-power-manager: add 2.30.0 Apr 08 10:54:05 stefan_schmidt: pong Apr 08 10:54:41 alo mickeyl Apr 08 10:55:14 mickeyl: good afternoon Apr 08 10:56:03 mickeyl: Python 2.6 compile problem in OE. Ever seen this: http://pastebin.com/HruVmngR ? Apr 08 10:57:28 hi hrw, morning pb__ Apr 08 10:57:46 stefan_schmidt: d'oh, never seen that one Apr 08 10:58:00 sounds like a header/lib mismatch Apr 08 10:58:14 can you check that the python-native version doesn't pick up any local headers? Apr 08 11:03:23 mickeyl: hmm, will try Apr 08 11:15:08 stefan_schmidt, Running ubuntu? I got the same problem yesterday, and I'm almost positive it was in a tree that used to compile fine. Only thing new was some Ubuntu upgrades I think. Apr 08 11:15:37 tasslehoff: no, debian unstable with updates from today Apr 08 11:15:41 I threw out my TMPDIR (after some other messing around) and I'm recompiling now. Apr 08 11:15:50 tasslehoff: hmm, ok Apr 08 11:16:15 tasslehoff: I'm trying to get down to the problem. Let me know if the rebuild works out. Apr 08 11:16:36 stefan_schmidt, will do Apr 08 11:17:04 mickeyl: is there a reason we have python-native 2.6.4 and python 2.6.2 for the target? Apr 08 11:17:42 not in particular Apr 08 11:17:47 for me 2.6.2 native was enough Apr 08 11:17:51 but someone chose to update Apr 08 11:17:57 ok Apr 08 11:18:15 perhaps the problem is solved if we update the non-native version to .4 as well Apr 08 11:18:26 * stefan_schmidt gives the log a deeper look Apr 08 11:18:39 mickeyl: That would be the next try on my list Apr 08 11:19:44 cool Apr 08 11:20:07 I've asked on the mailing list, but I'll repeat myself here. I'm creating an SDK for Beagleboard-development, and I need some way of getting headers like into it. linux-libc-headers only gives me the vanilla kernel. Any suggestions? Apr 08 11:20:48 is the beagleboard kernel installing any headers? Apr 08 11:20:58 if so, then depend on it Apr 08 11:24:13 mickeyl: why do some python recipe indicate their PR with ml (instead of r) ? Apr 08 11:24:34 mickeyl, it checks out the linux-omap kernel and does a buch of patches. I'll try just depending on linux-omap. thanks. Apr 08 11:27:43 mlip: ml is for Michael Lauer (which would be me) and the rootfs of that are historic. Apr 08 11:27:48 err Apr 08 11:27:50 not rootfs Apr 08 11:27:51 roots Apr 08 11:27:51 heh Apr 08 11:28:20 these days i don't mind if we substitute all that with r0 step by step during updates Apr 08 11:28:56 mickeyl: enough ego with mickeyterm, mdbus, mioctl and such? ;) Apr 08 11:29:26 ok, machting python version work Apr 08 11:29:35 stefan_schmidt: correct :D Apr 08 11:30:00 mcorrect, that is Apr 08 11:30:00 Will send a mail to the list which would bump it. Apr 08 11:30:04 mickeyl: heh Apr 08 11:30:09 myes Apr 08 11:30:11 :D Apr 08 11:31:03 mickeyl: iirc you play music, isn't? Apr 08 11:31:46 mickeyl: about licensing, how would you proceed with http://people.dsv.su.se/~fk/beatrix_home.html Apr 08 11:31:49 ant_work: yeah, i torture string instruments and electronic instruments quite frequently Apr 08 11:31:59 is very restrictive and the author didn't bother to answer me Apr 08 11:32:25 do you have an URI to the license text? Apr 08 11:32:39 mickeyl: and in past you tortured computers to generate music Apr 08 11:32:58 hrw: oh yeah! :) Apr 08 11:33:35 http://people.dsv.su.se/~fk/beatrix_download.html Apr 08 11:34:19 mickeyl: I think about interfacing with http://www.bgmi.it/?a=showproduct&b=1 Apr 08 11:34:37 ..and have an alternative to http://www.bgmi.it/?a=showproduct&b=2 Apr 08 11:35:17 I throw astone in beagleboard lake Apr 08 11:38:28 hmm Apr 08 11:38:41 cool stuff Apr 08 11:39:19 http://marcin.juszkiewicz.com.pl/2010/04/08/what-makes-a-good-developer-board/ Apr 08 11:40:36 hrw: btw, what's your masterplan to convert canonical to OE? ;) Apr 08 11:45:28 mickeyl: ah thought of something like that ;) Apr 08 11:45:49 Hello! Apr 08 11:45:51 My OE compile spends a lot of time on "generating locale". Does it have to? How can I tell it only to bother with "en-us"? I tried setting IMAGE_LINGUAS="en-us" in my meta-toolchain.bb, but it didn't do much Apr 08 11:46:35 Anyone can tell me how to fix this warning everytime I use bitbake: "NOTE: preferred version git of u-boot not available (for item u-boot)" ?? Apr 08 11:48:52 XorA: http://www.youtube.com/watch?v=jHWDSnWk2jU Apr 08 11:55:32 hrw: wow Apr 08 11:55:57 XorA: I knew that you will like it Apr 08 11:57:39 * XorA has a lego radio controlled car, but not that cool Apr 08 12:01:50 XorA: http://www.sakoman.com/OMAP3/a-script-for-partitioningformatting-a-bootable-sdmicrosd-card.html does this still work for you ? Apr 08 12:02:26 mlip: there is a later version in OE in contrib/angstrom Apr 08 12:03:20 XorA: thx, somehow grep (ofc it's not my fault) missed it Apr 08 12:03:23 ;) Apr 08 12:03:37 03Koen Kooi  07org.openembedded.dev * raf23f72bc8 10openembedded.git/recipes/cups/ (cups14.inc cups_1.4.3.bb): cups 1.4.x: cleanups Apr 08 12:03:39 03Tarjei Knapstad  07org.openembedded.dev * r6ba355a440 10openembedded.git/recipes/cups/ (5 files in 2 dirs): cups: add 1.4.3 Apr 08 12:04:47 mlip: there is even a .bb file to build a target package :-) Apr 08 12:05:07 XorA: mhm ? Apr 08 12:05:51 angstrom/omap3-mkcard.bb Apr 08 12:07:38 ah. found http://wiki.openembedded.net/index.php/How_to_disable_generation_of_locales. Is there any danger in providing qemu-arm natively and disabling locale generation? Apr 08 12:07:43 nice! Apr 08 12:09:02 tasslehoff: you seem to be mixing something up there! Apr 08 12:09:18 tasslehoff: if you disable locale generation you dont need to build qemu-arm Apr 08 12:10:34 XorA: and this script works for you ? Apr 08 12:10:48 mlip: yes Apr 08 12:10:53 XorA, If I disable locale generation, I'm told to provide qemu-arm natively. Installing qemu-kvm-extras on Ubuntu gets the build going again. Apr 08 12:11:25 partitioning works, but somehow i cant find the p1/p2 block devices :/ Apr 08 12:11:54 tasslehoff: you misread Apr 08 12:12:10 mlip: hdparm -z /dev/sdX or replug card Apr 08 12:12:11 tasslehoff: there is an or in the middle of that paragraph that you seem to have read as an and Apr 08 12:13:10 XorA, hmm. yes. Apr 08 12:15:11 XorA, actually it's only GLIBC_GENERATE_LOCALES I need to set. thanks for the heads up. :) Apr 08 12:15:17 hrw: well, thought i could do everything in one step Apr 08 12:15:25 tasslehoff: coolio Apr 08 12:15:45 the fs nodes are actually present in /dev, but the script seems to miss them (maybe is too fast) Apr 08 12:16:02 it works on fedora, no idea about lesser distros :-) Apr 08 12:16:24 it worked for me under debian Apr 08 12:16:29 have gentoo Apr 08 12:16:36 but I had some improvements even one day Apr 08 12:18:35 script is too fast Apr 08 12:19:01 sleep 1 after sfdisk command solves this problem Apr 08 12:19:23 mlip: for MMC we added one sec sleep too in kexecboot Apr 08 12:19:43 ..nut with 2.6.34 it seems PCMCIA/CF is needing too Apr 08 12:20:14 actually this should not be necessary because sfdisk writes and re-reads it before quitting ? Apr 08 12:20:38 I'm talking in general Apr 08 12:20:40 mlip: will be delay in udev most likely Apr 08 12:20:48 script should really wait for nodes to appear Apr 08 12:20:53 no idea how to do that Apr 08 12:21:06 mhm Apr 08 12:21:09 XorA: he..polling is the long and hard way.. Apr 08 12:21:28 we thought about it and Jay7 finally opted for sleep 1 Apr 08 12:21:30 :D Apr 08 12:21:56 I can't imagine how you could do it from a shell script, though ;) Apr 08 12:24:29 Anyone can tell me how to fix this warning everytime I use bitbake: "NOTE: preferred version git of u-boot not available (for item u-boot)" ?? Apr 08 12:25:33 dcopeto: ignore it Apr 08 12:26:11 XorA: should we just add a sleep to this script ? Apr 08 12:26:23 or should I fork it ? Apr 08 12:26:25 mlip: Id accept a change that did that yes Apr 08 12:26:59 sending it to the ML then (in a few ...) Apr 08 12:27:25 XorA: ok, thanks. I thought this could be causing my problem in compiling u-boot... I have an error in file sha1.c and can't find any reference about it! Apr 08 12:34:06 mickeyl, I tried http://pastebin.ca/1858838 , but omapfb.h isn't in the SDK. any ideas? Apr 08 12:36:31 tasslehoff: if your software needs special headers that are not in the kernel, you might either add them to the kernel build or bring them with the software that needs it in a local dir Apr 08 12:56:38 mickeyl, the kernel is checked out from git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6.git (linux-omap_2.6.29.bb), and the header I need is in there. I'm just not able to bring it into the sdk. Apr 08 12:58:45 tasslehoff: take a look at one of the linux classes Apr 08 12:58:59 some of them stage additional headers as per device overrides Apr 08 12:59:01 e.g. Apr 08 12:59:10 do_stage_append_nhk15 () { Apr 08 12:59:10 install -D -m 666 ${S}/drivers/video/nomadik/sga_interface.h ${STAGING_KERNEL_DIR}/drivers/video/nomadik/sga_interface.h Apr 08 12:59:12 ... Apr 08 12:59:21 could do the same for omap Apr 08 13:00:28 hello Apr 08 13:01:02 i am sitting in front of an mpc8323EVRADDC and trying to get a working kernel Apr 08 13:01:23 but it stops right after Uncompressing Kernel Image ... OK Apr 08 13:01:50 does this msg comes from the kernel? or is the kernel not doing anything at all? Apr 08 13:02:54 i used make mpc83xx_defconfig for configuration Apr 08 13:02:55 mickeyl, the header can already be found in <..>/staging/beagleboard-angstrom-linux-gnueabi/kernel/include/linux/omapfb.h. Apr 08 13:03:13 * tasslehoff feels like he's nagging :) Apr 08 13:05:10 tasslehoff: hmm, i'm not too familiar with the sdk packages; if the kernel-dev package is installed fine, but lacking the header, then perhaps the header needs to be added to the kernel-dev package Apr 08 13:05:19 obviously it installs fine in staging already Apr 08 13:10:57 mickeyl, what do you mean by "the kernel-dev package"? should my depend on linux-omap drag in kernel-dev_2.6.29-r51.5_beagleboard.ipk? Apr 08 13:19:12 tasslehoff: I should point out that no userspace program should ever include that header, to do so is a bug Apr 08 13:19:13 03Koen Kooi  07org.openembedded.dev * r9e1803c61b 10openembedded.git/recipes/gtk-engines/murrine_0.90.3.bb: murrine: add 0.90.3 Apr 08 13:19:15 03Koen Kooi  07org.openembedded.dev * r74817a51a9 10openembedded.git/recipes/gtk-engines/gtk-engine-aurora_1.5.1.bb: gtk-engine-aurora: add 1.5.1 Apr 08 13:19:15 03Koen Kooi  07org.openembedded.dev * r85ce1e8f61 10openembedded.git/recipes/gtk-engines/elementary-gtk-theme_1.0.bb: elementary-gtk-theme: add 1.0 Apr 08 13:19:16 03Koen Kooi  07org.openembedded.dev * r03490510a2 10openembedded.git/recipes/gtk-engines/elementary-gtk-theme_1.0.bb: elementary-gtk-theme: also depend on aurora Apr 08 13:19:17 03Koen Kooi  07org.openembedded.dev * rdf9cefd295 10openembedded.git/recipes/gtk-engines/ (elementary-icon-theme-enable.bb elementary-icon-theme_2.2.bb): elementary-icon-theme: add 2.2 from openpandora OE overlay Apr 08 13:19:18 03Koen Kooi  07org.openembedded.dev * rca135d82b4 10openembedded.git/recipes/gtk-engines/murrine_git.bb: murrine: add git version Apr 08 13:19:21 03Koen Kooi  07org.openembedded.dev * rd60ef74778 10openembedded.git/recipes/gtk-engines/gtk-engines_2.20.0.bb: gtk-engines: add 2.20.0 Apr 08 13:35:09 XorA, Our code is heavily inspired by omapfbplay, which also uses that header directly. I haven't explored other ways to do it (just as efficiently), but maybe I must Apr 08 13:36:20 03Michael Lippautz  07org.openembedded.dev * r6360c49be6 10openembedded.git/contrib/angstrom/omap3-mkcard.sh: Apr 08 13:36:20 omap3-mkcard.sh: Adding sleep and umounts to fix race cond. Apr 08 13:36:20 * Adds sleep to wait for blocknodes in /dev Apr 08 13:36:20 * Adds unmounts to fix possible automounts Apr 08 13:36:20 Signed-off-by: Michael Lippautz Apr 08 13:36:45 tasslehoff: the reason for linux-libc-headers is to sanitise kernel headers and move them to usespace, kernel ones are not supposed to be used directlu Apr 08 13:36:59 tasslehoff: better to carry a local copy of that header in your app Apr 08 13:42:48 XorA, I'll take that advice. The fact that linux-libc-headers uses the vanilla linux kernel while my beagleboard build uses linux-omap with patches is something I shouldn't need to worry about? Apr 08 13:42:59 nope Apr 08 13:43:10 linux-libc-headers does not depend on target Apr 08 13:43:24 and 2.6 ones can be used on machine which runs 2.4 kernel Apr 08 13:46:38 is contrib/oe-stylize.py safe/popular enough to call it on every recipe and push result? Apr 08 13:47:27 hrw, great Apr 08 13:47:28 * JaMa is thinking about script appending checksums to every recipe and then stylize them with this script Apr 08 13:47:33 JaMa: rather not Apr 08 13:47:47 hrw, XorA. mickeyl: thanks for the help (and patience) Apr 08 13:47:50 * mwester_ groans at the mad "make it pretty" mass-commits that seem to have infected OE lately. Apr 08 13:48:13 mwester_: does that change work for you? Apr 08 13:48:41 JaMa, I suggest you set up an environment to build SlugOS/BE and SlugOS/LE Apr 08 13:49:03 mwester_: IMHO better to "make it pretty" before branching new stable, than diverging after branch more than needed Apr 08 13:49:17 JaMa, re: changes -- they work for me; the correct kernel is now built. It doesn't solve the problem for other ixp4xx distros, but I guess they'll have to solve it themselves. Apr 08 13:50:24 mwester_: do you know about any other distro using PREFERRED_VERSION for linux-ixp4xx? Apr 08 13:50:29 JaMa, I disagree. Then you get a broken stable, because you are assuming that everyone will have tiime to find all the problems caused in their distros and fix them. I still, for example, do not have the package feeds building for SlugOS since a set of commits broke them two weeks ago. Apr 08 13:50:59 JaMa: Nagios, IIRC and Angstrom. The former might be dead, but I think Angstrom is still a viable OE distro. Apr 08 13:51:13 mwester_: my git grep didn't show any (but you're right someone can be off tree) Apr 08 13:51:19 No, they're in there. Apr 08 13:52:06 conf/distro/include/angstrom-2007-for-openmoko.inc:PREFERRED_VERSION_linux-ixp4xx ?= "2.6.21.6+svnr${SRCREV}" Apr 08 13:54:02 included only in dead openmoko.. Apr 08 13:54:22 but also machine/include/ixp4xx.inc :/ Apr 08 13:55:45 03Stefan Schmidt  07org.openembedded.dev * r742f0a5c1d 10openembedded.git/recipes/lowpan-tools/lowpan-tools_git.bb: Apr 08 13:55:45 lowpan-tools_git.bb: Bump SRCREV Apr 08 13:55:45 Gain some fixes and monitor device support. Apr 08 13:55:47 03Stefan Schmidt  07org.openembedded.dev * rbe0e5268d2 10openembedded.git/conf/machine/imote2.conf: Apr 08 13:55:48 imote2.conf: Change flash size to 30 Mbyte Apr 08 13:55:48 We have a32 Mbyte flash chip but better leave some space for jffs2. Apr 08 13:55:49 03Stefan Schmidt  07org.openembedded.dev * r1ba1263ed0 10openembedded.git/conf/distro/include/angstrom-2008-preferred-versions.inc: Apr 08 13:55:49 angstrom-2008-preferred-versions.inc: Bump python version to 2.6.4 Apr 08 13:55:49 This is done to match the python-native version. Without this we seem to get Apr 08 13:55:49 trouble with some header mismatch or similar: Apr 08 13:55:49 ./libpython2.6.so: undefined reference to `_PyParser_Grammar' Apr 08 13:55:49 Signed-off-by: Stefan Schmidt Apr 08 13:55:50 Acked-by: Koen Kooi Apr 08 14:00:31 03Stefan Schmidt  07org.openembedded.dev * r25bd61497a 10openembedded.git/recipes/linux/ (linux/imote2/defconfig linux_2.6.33.bb): Apr 08 14:00:31 linux_2.6.33.bb: Add iMote2 as supported board. Apr 08 14:00:31 Also add a defconfig for this board. Apr 08 14:18:45 03Mike Westerhof  07org.openembedded.dev * r198179be17 10openembedded.git/conf/distro/include/preferred-slugos-versions.inc: (log message trimmed) Apr 08 14:18:46 SlugOS: preferred-slugos-versions.inc - fix kernel SRCREV selection Apr 08 14:18:46 This makes the PREFERRED_VERSION selection for the SlugOS kernel work Apr 08 14:18:46 again after a recent "pretty-ification" change for SRCREV handling. Apr 08 14:18:46 Note that this is a sub-optimal solution, but it is required to correct Apr 08 14:18:46 build failures for SlugOS. A long-term solution needs to either put Apr 08 14:18:47 the SRCREV back to where it was (a semi-global config file), or it needs Apr 08 14:47:41 Hello! Apr 08 14:48:19 How to change the PREFERRED_VERSION_ for u-boot?? Apr 08 14:48:41 I'm using PREFERRED_VERSION_u-boot = "2009.08" and it always builds me the 2010-rc1 version! Apr 08 14:52:34 * hrw -> out Apr 08 15:18:43 03Martin Jansa  07org.openembedded.dev * ra84caeecc4 10openembedded.git/recipes/linux/ (2 files in 2 dirs): Apr 08 15:18:43 linux-openmoko-2.6.32: enable MMC_S3C in defconfig Apr 08 15:18:43 Signed-off-by: Martin Jansa Apr 08 16:09:38 morning Apr 08 16:15:57 hi kergoth Apr 08 16:36:21 * JaMa running fetchall for every oe.dev recipe... Apr 08 16:42:52 JaMa: glutton for punishment? :) Apr 08 16:43:09 script for moving checksums :) Apr 08 16:43:30 ah :) Apr 08 16:43:30 i.e. "yes" :-} Apr 08 16:43:34 hehe Apr 08 16:43:42 is there something like "world" but with all available versions? find + bitbake -b fails for BBCLASSEXTEND = "native" :/ Apr 08 16:44:23 maybe I should reuse oe-source-checker.py Apr 08 16:45:05 nope, no such thing. isn't there a patch for -b w/ bbclassextend floating around? Apr 08 16:45:12 i seem to remember seeing such a thing on the list Apr 08 16:45:46 yes there was but IIRC only for 1.8 Apr 08 16:46:10 don't think that area of the code is that different, shouldn't be hard to get to apply Apr 08 16:46:16 and forward port wasn't so easy (so I gave up.. expecting author to send it) Apr 08 16:48:54 * kergoth looks at the patch Apr 08 16:49:18 kergoth: thanks :) Apr 08 17:58:24 I guess this is bad? http://pastebin.com/UfiYM3g1 Apr 08 17:58:56 Trying to run Xorg on h5550, org.openembedded.dev, x11-gpe-image Apr 08 18:40:08 03Roman I Khimov  07org.openembedded.dev * r56083e9b32 10openembedded.git/recipes/gd/gd_2.0.35+2.0.36rc1.bb: Apr 08 18:40:08 gd: build with freetype support Apr 08 18:40:08 Support for freetype in gd is required for some packages and it looks Apr 08 18:40:08 like it was disabled previosly just because of build failures. Version Apr 08 18:40:08 2.0.35+2.0.36rc1 builds fine with it, so enable freetype for it. Apr 08 18:40:09 Signed-off-by: Roman I Khimov Apr 08 18:40:13 03Roman I Khimov  07org.openembedded.dev * r9614d5509e 10openembedded.git/recipes/sarg/ (files/sarg.conf sarg_2.2.7.1.bb): Apr 08 18:40:13 sarg: new recipe Apr 08 18:40:13 Squid Analysis Report Generator is a tool that allow you to Apr 08 18:40:13 view "where" your users are going to on the Internet. Apr 08 18:40:14 Sarg provides many informations about Squid users Apr 08 18:40:14 activities: times, bytes, sites, etc... Apr 08 18:40:15 Signed-off-by: Roman I Khimov Apr 08 19:03:19 03Vladimir Sorokin  07org.openembedded.dev * rd65aa5a8f6 10openembedded.git/recipes/binutils/ (binutils_2.20.1.bb binutils_2.20.bb): Apr 08 19:03:19 binutils: Add x86_64_i386_biarch patch for 2.20 & 2.20.1 Apr 08 19:03:19 Signed-off-by: Vladimir Sorokin Apr 08 19:03:19 Signed-off-by: Roman I Khimov Apr 08 19:03:29 03Vladimir Sorokin  07org.openembedded.dev * r97bf345f59 10openembedded.git/classes/insane.bbclass: Apr 08 19:03:30 insane.bbclass: add x86_64 architecture for linux-uclibc Apr 08 19:03:30 Signed-off-by: Vladimir Sorokin Apr 08 19:03:30 Signed-off-by: Roman I Khimov Apr 08 19:04:02 Hello I am having a problem when doing a "bitbake nano" Apr 08 19:04:07 tar: /home/kyle/sources/config_cvs.sv.gnu.org__20050701.tar.gz: Cannot open: No such file or directory Apr 08 19:05:47 can anyone tell me where dbus-dev recipe is, as listed in say line 58 of task-fso-toolchain-target.bb? Apr 08 19:07:36 I was wondering if "-dev" suffixes are somehow automatically ignored when need be. Apr 08 19:08:09 dickelbeck: dbus Apr 08 19:09:27 what does the presence of the "-dev" trailer mean? Apr 08 19:13:58 well the question is probably not as stupid as it sounds. I know in a normal package name sense it is for building with a library. Apr 08 19:14:06 03Antonio Ospite  07org.openembedded.dev * ra014576736 10openembedded.git/recipes/tzcode/tzcode-native_2010f.bb: Apr 08 19:14:06 tzcode-native_2010f: Update tzdata SRC_URI and checksum Apr 08 19:14:06 tzdata2010g is no longer available, use tzdata2010h Apr 08 19:14:06 Signed-off-by: Antonio Ospite Apr 08 19:14:06 Acked-by: Roman I Khimov Apr 08 19:14:12 dickelbeck: exactly that Apr 08 19:14:15 03Antonio Ospite  07org.openembedded.dev * rdfe237398a 10openembedded.git/recipes/tzdata/ (tzdata_2010g.bb tzdata_2010h.bb): Apr 08 19:14:15 tzdata: move tzdata_2010g.bb to tzdata_2010h.bb Apr 08 19:14:15 tzdata_2010g is no longer available; change the checksums in the new Apr 08 19:14:15 recipe Apr 08 19:14:16 Signed-off-by: Antonio Ospite Apr 08 19:14:16 Acked-by: Roman I Khimov Apr 08 19:15:04 XorA: what is mysterious is that this package name is not exactly implemented anywhere that I can see. There must be some magic name manipulation WRT "-dev" in bitbake? Apr 08 19:15:36 Yes Apr 08 19:15:44 dickelbeck: all -dev type stuff when building a recipe gets put in -dev packages when building Apr 08 19:15:56 For each recipe there's a PACKAGES variable that by default has -dev and -dbg and -doc Apr 08 19:16:00 dickelbeck: dont mix the concept of recipe with concept of package Apr 08 19:16:03 And some filemasks to put stuff in some places Apr 08 19:16:57 OK, it is beginning to come back. Been about 18 months since I read the manual.... Apr 08 19:17:56 My memory non-volatility does not extend that long I see. Apr 08 19:18:02 Bye. Apr 08 19:49:18 03Koen Kooi  07org.openembedded.dev * rcd7cf4bdd8 10openembedded.git/recipes/gnuplot/gnuplot_4.4.0.bb: gnuplot: add 4.4.0 Apr 08 19:49:19 03Koen Kooi  07org.openembedded.dev * re42f9ca7aa 10openembedded.git/recipes/abiword/abiword_2.8.3.bb: abiword 2.8.3: fix regression that caused the collab plugin being dropped from the list Apr 08 19:49:20 03Koen Kooi  07org.openembedded.dev * r243f74f834 10openembedded.git/recipes/gtk-engines/ (elementary-icon-theme_2.2.bb elementary-icon-theme_2.3.bb): elementary-icon-theme: update to 2.3 Apr 08 19:53:09 download world will take longer than expected, does someone have some estimate how much space to I need in downloads dir (to keep it running whole night?) Apr 08 19:54:07 now I'm somewhere near 1300 recipes from 8300 Apr 08 19:55:35 re Apr 08 20:09:20 hmm. Apr 08 20:09:50 * kergoth_ thinks about possibly leveraging the logging module in master Apr 08 20:30:29 does someone know sizeof(intx)? Apr 08 20:30:47 JaMa: my DL_DIR was ~30GB iirc Apr 08 20:30:59 hrw: with SCM checkouts? Apr 08 20:31:03 yes Apr 08 20:31:10 hrw: I'm already at ~16G good Apr 08 20:32:20 thanks Apr 08 20:47:03 it makes me sad that a pylint on the bb package fails with "RuntimeError: maximum recursion depth exceeded while calling a Python object".. we must do some funky stuff :) Apr 08 20:55:32 03Roman I Khimov  07org.openembedded.dev * r9bacc628c3 10openembedded.git/recipes/beep/beep_1.2.2.bb: Apr 08 20:55:32 beep: fix building and packaging Apr 08 20:55:32 GNU_HASH was missing and beep and beep-doc packages were listed Apr 08 20:55:32 in PACKAGES twice. Apr 08 20:55:32 Signed-off-by: Roman I Khimov Apr 08 20:55:44 03Roman I Khimov  07org.openembedded.dev * rc4aa10367b 10openembedded.git/recipes/linux-ha/ (2 files in 2 dirs): Apr 08 20:55:44 cluster-resource-agents: fix headers define extraction Apr 08 20:55:44 Header define extraction routine tries to execute test binary which Apr 08 20:55:44 may or may not work depending on target architecture. Redo it using Apr 08 20:55:44 preprocessor only. Apr 08 20:55:44 Fixes wrong OCF_ROOT_DIR. Apr 08 20:55:45 Signed-off-by: Roman I Khimov Apr 08 20:55:45 03Roman I Khimov  07org.openembedded.dev * r8a9cae9876 10openembedded.git/recipes/pacemaker/ (2 files in 2 dirs): Apr 08 20:55:46 pacemaker: fix headers define extraction and uclibc build Apr 08 20:55:46 Header define extraction routine tries to execute test binary which Apr 08 20:55:47 may or may not work depending on target architecture. Redo it using Apr 08 20:55:47 preprocessor only. Apr 08 20:55:48 Fix uclibc build also, doesn't link properly without fstack-protector. Apr 08 20:55:48 Signed-off-by: Roman I Khimov Apr 08 21:56:44 http://people.redhat.com/kzak/git-scripts/git-groupedlog is cute Apr 08 22:34:27 okay, update_data() needs to be renamed to something sane Apr 08 22:37:18 seems a sane name to me, although somewhat generic... Apr 08 22:42:28 mwester: I'm thinking a "finalize" method of the datastore. similar in concept to the finalise() function in the parse.. final steps to be performed against whatever Apr 08 23:17:24 * kergoth_ thinks about trying to wedge in some unit testing into bitbake Apr 08 23:20:18 hi kergoth. how could we track down the notes about 'installed but not shipped in any package'? making them QA so these are logged? Apr 08 23:20:28 track down? Apr 08 23:20:51 yea, these are scrolling on screen and are not in the logs irc Apr 08 23:20:58 so log it Apr 08 23:21:11 I've seen right now some glibc issues about /etc/rpc Apr 08 23:21:18 no notes or warnings or errors from bitbake are logged, unless they're from tasks and you run master, and even then its only if the task fails Apr 08 23:21:24 other than specific cases Apr 08 23:21:27 like legacy-staging.log Apr 08 23:21:47 I was thinking about a new QA Apr 08 23:22:20 those errors don't come from insane/sanity. Apr 08 23:22:45 I see Apr 08 23:22:55 pretty sure its package.bbclass emitting them Apr 08 23:23:08 yes, do_package Apr 08 23:23:11 the QA messages are handled by insane, which isn't guaranteed to always be inherited if package is Apr 08 23:23:17 its up to the distro to inherit insane Apr 08 23:25:19 hm..I have to ask woglinde about strongarm/eabi. collie is still building own OABI toolchain :/ Apr 09 00:06:30 03Chris Larson  07org.openembedded.dev * r444d86f372 10openembedded.git/ (10 files in 2 dirs): Apr 09 00:06:30 Using NotHandled/Handled in event handlers is deprecated Apr 09 00:06:30 Signed-off-by: Chris Larson Apr 09 00:10:43 hi everybody! Apr 09 00:11:05 I am new to OE, and I have some questions,.. Apr 09 00:12:05 I didn't understand exactly how angstrom distribution is related to OE Apr 09 00:12:46 what is the angstrom distribution exactly? I see that OE can build a whole linux system from scratch Apr 09 00:13:17 if you don't know what a linux distribution is, I don't think we can help you. Apr 09 00:13:25 hehehe Apr 09 00:13:27 I know Apr 09 00:13:29 sorry Apr 09 00:13:37 I mean, for example Apr 09 00:13:54 when I put in local.conf Apr 09 00:14:06 the var DISTRIBUTION='angstrom-2008' Apr 09 00:14:21 OE is a repository of shared metadata. Multiple linux distributions can work together in the repository without duplicating effort. Apr 09 00:14:26 angstrom is one of the main distros OE can build Apr 09 00:14:38 uhmmm Apr 09 00:14:49 and the angstrom distribution Apr 09 00:15:17 put their files on your server, I mean, its just to do a git pull, and everything is updated? Apr 09 00:15:47 ? Apr 09 00:16:02 no, git pull updates the metadata, which you can then use to BUILD the distro Apr 09 00:16:04 I am building the angstrom 2008 Apr 09 00:16:05 OE is not a distro Apr 09 00:16:08 OE is a buildsystem Apr 09 00:17:19 If you want to build the last version of angstrom, what would you do? Apr 09 00:17:34 I'd ask in #angstrom Apr 09 00:17:44 hehehe, they don't answer Apr 09 00:17:53 but thanks Apr 09 00:17:57 thanks a lot =) Apr 09 00:26:25 03Chris Larson  07org.openembedded.dev * r71d0e0501b 10openembedded.git/recipes/tasks/task-base.bb: Apr 09 00:26:25 task-base: change the 'set' logic to work a bit better Apr 09 00:26:25 Signed-off-by: Chris Larson Apr 09 00:39:44 03Andrea Adami  07org.openembedded.dev * r000ac398f9 10openembedded.git/recipes/kexecboot/linux-kexecboot-2.6.33+2.6.34-rc3/ (6 files in 6 dirs): linux-kexecboot: try to standardize Zaurus defconfigs for 2.6.34. WIP. Apr 09 00:44:55 kergoth_, you could also refer them to the angstrom website Apr 09 00:45:17 03Andrea Adami  07org.openembedded.dev * rae50bef3b5 10openembedded.git/recipes/kexecboot/linux-kexecboot-2.6.33+2.6.34-rc3/ (collie/defconfig poodle/defconfig): linux-kexecboot: add forgotten loglevel=3 to collie and poodle cmdline. Apr 09 00:49:15 Crofton|work: ah, good point :) Apr 09 00:58:20 although that won't get him .dev **** ENDING LOGGING AT Fri Apr 09 02:59:56 2010