**** BEGIN LOGGING AT Wed Jan 21 02:59:57 2009 Jan 21 03:16:38 03Michael 'Mickey' Lauer  07org.openembedded.dev * r591bc386af 10openembedded.git/packages/bluez/ (bluez4_4.27.bb bluez4_4.7.bb): bluez4: 4.7 -> 4.27; improve package granularity and fix DEPENDS Jan 21 04:15:07 mickeyl: ping Jan 21 04:18:21 pong Jan 21 04:18:44 you're lucky -- usually i should be deep asleep Jan 21 04:19:03 mickeyl: who owns the CIA scripts for OE git checkin reporting? Jan 21 04:19:12 it's linking to the wrong URL for cgit. Jan 21 04:19:49 anyone who has access to the server is allowed to touch 'em Jan 21 04:20:23 do you have a patch? Jan 21 04:20:42 nope, but I'm working out what the URL should look like now. Jan 21 04:21:00 can you tell me what the current setting is? should be near the top of the cia script Jan 21 04:21:46 it current tells CIA: http://cgit.openembedded.net/?p=openembedded.git/?a=commit;h=591bc386af2d50c85da59825132704b35ecb9ab3 Jan 21 04:21:54 it should tell CIA: http://cgit.openembedded.net/cgit.cgi?url=openembedded/commit/&id=591bc386af2d50c85da59825132704b35ecb9ab3 Jan 21 04:22:50 hmm Jan 21 04:22:59 can't find it Jan 21 04:23:35 my $gitweb_url = "http://source.winehq.org/git"; Jan 21 04:23:38 hmm Jan 21 04:24:09 is this in git-notify or perhaps on the CIAbot? Jan 21 04:24:11 is there a git-notify post-receive hook or something? Jan 21 04:24:41 it's usually in the hook that sends the XML RPC message to CIA - it includes a link for the commit diff Jan 21 04:24:57 the bit that creates http://cia.vc/stats/project/openembedded/.message/e0e0ca Jan 21 04:25:16 the URL field of that is what's wrong Jan 21 04:25:35 "Generated by git-notify script for CIA" Jan 21 04:26:01 http://cia.vc/stats/project/openembedded/.message/e0e0ca/xml Jan 21 04:26:06 says it's on amethyst. Jan 21 04:26:40 fun Jan 21 04:26:46 i can't see where that url comes from Jan 21 04:26:52 definitely not from $gitweb_url Jan 21 04:27:00 must be looking in the wrong git-notify **** BEGIN LOGGING AT Wed Jan 21 04:28:46 2009 Jan 21 04:28:47 monotone yeah Jan 21 04:28:54 survey might come in handy at a later point Jan 21 04:29:02 we can add it then Jan 21 04:29:10 rwhitby: we call Jan 21 04:29:10 I suspect it points a a dead machine :) Jan 21 04:29:11 /home/git/git-notify -c OpenEmbedded -m openembedded-commits@openembedded.org -r openembedded.git -u http://gitweb.openembedded.n Jan 21 04:29:11 et/ $refname $oldrev $newrev Jan 21 04:29:30 so the uri is http://gitweb.openembedded.net Jan 21 04:30:16 i wonder how you see cgit in there Jan 21 04:30:35 mickeyl: looks like a redirect Jan 21 04:30:56 gitweb -> cgit Jan 21 04:31:01 ah Jan 21 04:31:02 fun Jan 21 04:31:08 ok, so should i try Jan 21 04:31:16 -u http://cgit.openembedded.net/cgit.cgi?url=openembedded/commit/&id=$newrev Jan 21 04:31:17 ? Jan 21 04:31:26 or change the redirect? Jan 21 04:31:35 (when i find it) Jan 21 04:31:50 I doubt the redirect is doing a rewrite of the URL Jan 21 04:32:38 right, it's a dumb redirect Jan 21 04:32:42 mickeyl: does that git-notify have cia stuff in it? Jan 21 04:32:43 nothing we could not change though Jan 21 04:32:51 yes, it has Jan 21 04:33:01 we're setting the uri via -u Jan 21 04:33:19 i guess rewriting would make more sense Jan 21 04:33:27 to fix all the broken older commits Jan 21 04:35:11 looking at http://people.openezx.org/wyrm/git-notify (assuming your git-notify is similar) it seems it's written for gitweb URLs Jan 21 04:35:46 stupid cgit people going and creating a new URL format for no good reason ... Jan 21 04:36:01 bummer Jan 21 04:36:38 If someone can provide me the proper changes to the apache config I'd add a ReWrite directive Jan 21 04:37:02 gn Jan 21 04:37:10 night Crofton Jan 21 04:37:30 mickeyl: does it use rewrite module already? Jan 21 04:37:34 no Jan 21 04:37:50 other alternative is to hack the git-notify script Jan 21 04:37:59 that won't fix the older ones Jan 21 04:38:08 so I'd rather rewrite Jan 21 04:38:11 ok Jan 21 04:39:27 mickeyl: if you email me the current apache config I can take a look at it, since I'm the one asking. Jan 21 04:40:05 http://rafb.net/p/aZLwaq50.html Jan 21 04:40:10 ah, email... oh well ;) Jan 21 04:40:29 i guess we're looking for something like Jan 21 04:40:44 RewriteEngine on Jan 21 04:40:53 bbiab (meeting) Jan 21 04:48:42 mickeyl: so we need to change '/?a=commit;h=...' to '/commit/&id=...' Jan 21 04:49:02 RewriteRule ^\?p=openembedded.git/\?a=commit;h=([a-z0-9]+)$ http://cgit.openembedded.net/cgit.cgi\?url=openembedded/commit/&id=$1 Jan 21 04:49:05 something like that? Jan 21 04:49:22 it doesn't work yet Jan 21 04:49:25 but should be close Jan 21 04:57:38 hmm, for some reason the rewriteengine doesn't work at all atm. Jan 21 04:57:38 heh Jan 21 05:06:17 how do you define non-greedy in RE again? Jan 21 05:18:35 85.180.157.182 - - [21/Jan/2009:06:18:24 +0100] [gitweb.openembedded.net/sid#7c5b80][rid#1bda0c8/initial] (2) init rewrite engine with requested uri / Jan 21 05:18:35 85.180.157.182 - - [21/Jan/2009:06:18:24 +0100] [gitweb.openembedded.net/sid#7c5b80][rid#1bda0c8/initial] (3) applying pattern '^\?p\=openembedded\.git(.*)' to uri ' Jan 21 05:18:35 85.180.157.182 - - [21/Jan/2009:06:18:24 +0100] [gitweb.openembedded.net/sid#7c5b80][rid#1bda0c8/initial] (1) pass through / Jan 21 05:19:02 any idea why it requests uri / when I'm reqeusting the whole thing? (?p=...) Jan 21 05:19:16 is '?' not working with the RewriteEngine? Jan 21 05:53:01 mickeyl: ah ? is a QUERY thing Jan 21 06:01:39 mickeyl: maybe http://stackoverflow.com/questions/404684/modrewrite-check-for-custom-query-string-in-url ? Jan 21 06:02:15 hmm - maybe it can't match in the query string .... Jan 21 06:04:01 sabine wakes up Jan 21 06:04:07 i need to get breakfast Jan 21 06:04:10 bbl, good speed Jan 21 06:06:56 mickey|breakfast: see section 3 in http://www.the-art-of-web.com/system/rewrite/1/ Jan 21 06:07:13 need to use RewriteCond to match in the query string Jan 21 06:07:28 then %N to use that in RewriteRule Jan 21 10:03:07 good morning Jan 21 10:08:45 03Koen Kooi  07org.openembedded.dev * r919c436148 10openembedded.git/packages/mythtv/mythtv_0.21.bb: myththv: bump PR Jan 21 10:08:55 03Koen Kooi  07org.openembedded.dev * r6f5958334a 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Jan 21 10:08:56 03Koen Kooi  07org.openembedded.dev * r5d387e64a1 10openembedded.git/contrib/angstrom/build-feeds.sh: angstrom feed builder: add pimlico Jan 21 10:10:12 mickeyl: good morning Jan 21 10:12:33 good morning pb! Jan 21 10:12:34 hi florian Jan 21 10:13:22 hi mickeyl Jan 21 10:37:10 good morning @ all Jan 21 10:38:06 03Michael 'Mickey' Lauer  07org.openembedded.dev * r5e32ee3d4d 10openembedded.git/ (4 files in 2 dirs): Jan 21 10:38:06 paroli: new recipe; combined application for Openmoko 2009.x Jan 21 10:38:06 this supersedes paroli-dialer and paroli-gui which are now removed Jan 21 10:38:43 morning Jan 21 10:39:32 are this: "ERROR: QA staging was broken by the package built above" quiet normal for building the org.openembedded.dev branch? Jan 21 10:41:21 no. Jan 21 10:41:33 yes ;) Jan 21 10:42:04 the reason are some bad entries in ncurses *.la libs... Jan 21 10:43:29 is there any good advices for solving this (others then bitbake -k ....) Jan 21 10:44:01 morning pplz :) Jan 21 10:46:39 hi Юрий Jan 21 10:46:42 <_boto> hello Jan 21 10:47:09 <_boto> i try to mount the rootfs from nfs Jan 21 10:47:15 <_boto> it partially works Jan 21 10:47:23 hrw: :) Jan 21 10:48:55 <_boto> but the os does not boot up completely waiting for nfs server reply Jan 21 10:49:01 Jay7: I know, 'privet' would be better but how to enter cyrylic with US keyboard...;D Jan 21 10:49:24 _boto: show me kernel cmdline Jan 21 10:49:31 ping http://bugs.openembedded.net/show_bug.cgi?id=4978 Jan 21 10:49:44 <_boto> i get the famous output: "nfs: server not responding, still trying" Jan 21 10:49:45 !oebug 4978 Jan 21 10:49:46 * * Bug 4978, Status: UNCONFIRMED, Created: 2009-01-12 21:26 Jan 21 10:49:48 * * rep.dot.nop(AT)gmail.com: busybox hwclock.sh runlevel script fails without GETOPT_LONG Jan 21 10:49:49 * * http://bugs.openembedded.net/show_bug.cgi?id=4978 Jan 21 10:50:24 _boto: I use this: "Kernel command line: mem=64M console=ttyS0,115200 nfsroot=192.168.1.112:/srv/nfs/at91sam9263ek ip=192.168.1.250:::255.255.255.0" and at91sam9263ek boots fine Jan 21 10:50:27 <_boto> bootargs=console=ttyS2,115200n8 noinitrd root=/dev/nfs rootwait rootfstype=nfs rw nfsroot=192.168.178.20:/home/abotorabi/oe-nfs,nolock,tcp,tcp dyntick=e ip=dhcp nohz=off Jan 21 10:50:42 <_boto> aha Jan 21 10:50:44 hrw, patch attached to 4978; same for bug 4979 Jan 21 10:50:51 <_boto> you use a fix ip? Jan 21 10:51:03 <_boto> i use dhcp Jan 21 10:51:03 hrw: one time I have using old laptop without cyrillic letters on keyboard :) from that time I have not looking at keyboard anymore while typing :) Jan 21 10:51:37 <_boto> something is curious ... Jan 21 10:52:02 _boto: I use static - otherwise interface is changed during boot and rootfs is gone Jan 21 10:52:03 <_boto> mounting the rootfs works in principle Jan 21 10:52:12 <_boto> ah, i understand Jan 21 10:52:19 <_boto> that my be the problem Jan 21 10:52:24 <_boto> gonna try a static ip Jan 21 10:52:30 Jay7: I never had to type cyrylic Jan 21 10:52:52 Jay7: my russian is very rusted. its nearly 20 years since I learnt it Jan 21 10:53:51 hrw: my german should be alike :) Jan 21 10:54:55 <_boto> hehe Jan 21 10:56:04 !oebug 4979 Jan 21 10:56:06 * * Bug 4979, Status: UNCONFIRMED, Created: 2009-01-12 21:36 Jan 21 10:56:07 * * rep.dot.nop(AT)gmail.com: opendreambox setups lack nbd client support Jan 21 10:56:08 * * http://bugs.openembedded.net/show_bug.cgi?id=4979 Jan 21 10:56:31 <_boto> i see that i configured the system to get its ip via dhcp Jan 21 10:57:13 <_boto> that may be the cause of problem; the nfs is retrieved with a fixed ip, shortly after the system changes its ip due to dhcp and the nfs is lost Jan 21 10:57:25 <_boto> hrw: may that be the cause? Jan 21 10:57:55 <_boto> so i try to fix also the ip of eth0 to the same ip as in bootargs Jan 21 10:58:24 _boto: I had such problem when setup this board. but it was my first time with nfsroot Jan 21 10:58:35 <_boto> aha Jan 21 10:58:40 blindvt: 4979 is opendreambox distro so I can not touch it Jan 21 10:59:16 hrw, oh. but the nbd stuff is generic Jan 21 10:59:17 blindvt: 4978 can be applied and then you should request adding it into stable on proper ML Jan 21 10:59:30 hrw, which list would that be? Jan 21 11:00:22 openembedded-stable Jan 21 11:01:41 hrw, ok, thanks alot. Let me send some of that stuff to the -devel list for a start Jan 21 11:02:59 <_boto> yupiieeee Jan 21 11:03:09 <_boto> hrw: it works now, thanks for the hints :-) Jan 21 11:04:03 morning channel Jan 21 11:04:47 03Bernhard Reutner-Fischer  07org.openembedded.dev * r34aabfd099 10openembedded.git/packages/busybox/ (8 files in 2 dirs): Jan 21 11:04:47 busybox: use short options for hwclock - closes #4978 Jan 21 11:04:47 Both busybox as well as big version support the same short options Jan 21 11:04:47 but the busybox version may be build with getopt_long turned off. Jan 21 11:05:15 ant_work: morning :) Jan 21 11:05:42 Jay7: hey, rebuilt? Jan 21 11:06:05 03Mark Brown  07org.openembedded.dev * r1bd70f27e9 10openembedded.git/conf/machine/include/tune-arm1176jzf-s.inc: Jan 21 11:06:05 Add tuning for ARM1176JZF-S core Jan 21 11:06:05 This core is featured in processors such as the S3C6410. Jan 21 11:06:18 ant_work: no, I have misspelled 'console-image' when started build :)) Jan 21 11:06:49 rebuilding now.. Jan 21 11:07:14 03Mark Brown  07org.openembedded.dev * rf9c43b03f9 10openembedded.git/ (4 files in 2 dirs): alsa: update to 1.0.18 Jan 21 11:08:27 any advice to solve failed sanity tests (QA) on ncurses libs... Jan 21 11:09:19 hi ant_work Jan 21 11:09:37 florian: hello Jan 21 11:10:05 Jay7: possibly broonie could help for the mmc subsystem issue Jan 21 11:10:41 ant_work: good :) Jan 21 11:11:23 but this looks like more generic issue.. Jan 21 11:11:42 yes..this looks like an asynch issue Jan 21 11:12:11 for sure some wait state would solve Jan 21 11:12:22 but we don't like to wait ;-) Jan 21 11:12:35 (for unnecessary periods) Jan 21 11:14:25 I'm trying to find some guru :) Jan 21 11:15:00 iirc rootwait is/was an infinite loop Jan 21 11:15:02 hrw, thanks Jan 21 11:17:40 ant_work: seems that udev working alike :) Jan 21 11:17:51 Jay7: https://kerneltrap.org/mailarchive/linux-kernel/2007/5/10/89285/thread Jan 21 11:18:31 but this is in kernel Jan 21 11:21:07 * * OE Bug 4984 has been RESOLVED (FIXED) by openembedded(AT)haerwu.biz Jan 21 11:21:09 * * Add tuning for ARM1176JZF-S core Jan 21 11:21:11 * * http://bugs.openembedded.net/show_bug.cgi?id=4984 Jan 21 11:22:07 * * OE Bug 4978 has been RESOLVED (FIXED) by openembedded(AT)haerwu.biz Jan 21 11:22:09 * * busybox hwclock.sh runlevel script fails without GETOPT_LONG Jan 21 11:22:11 * * http://bugs.openembedded.net/show_bug.cgi?id=4978 Jan 21 11:22:19 * * OE Bug 4969 has been RESOLVED (FIXED) by openembedded(AT)haerwu.biz Jan 21 11:22:21 * * Bump ALSA to 1.0.18 Jan 21 11:22:22 * * http://bugs.openembedded.net/show_bug.cgi?id=4969 Jan 21 11:24:42 hrw: ALSA 1.0.19 was released since I did that :/ Jan 21 11:26:54 broonie: happens :-) Jan 21 11:27:34 hrw: but thanks Jan 21 11:28:38 broonie: I think that you should just request r/w Jan 21 11:29:35 hrw: If people are happy with that - I thought the usual procedure was that people got fed up applying your patches and suggested it for you :) Jan 21 11:29:44 seeing as broonie had r/w under mtn, he really should request Jan 21 11:30:30 XorA: Well, I had root on a server but that's not *quite* the same thing :) Jan 21 11:30:56 broonie: wolfson@oe.org was always a valid commit key :-) Jan 21 11:33:19 it had exactly one commit Jan 21 12:02:16 woglinde, I think I'm missing some deps for gettext: http://rafb.net/p/qTt5d174.html Jan 21 12:03:04 woglinde, ah no...it's part of glibc Jan 21 12:03:16 so uclibc doesn't have it Jan 21 12:11:06 * * OE Bug 4505 has been RESOLVED (FIXED) by lrg(AT)kernel.org Jan 21 12:11:08 * * gtk+-2.12.11-r3 fails to link Jan 21 12:11:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4505 Jan 21 12:12:09 Gnutoo, uClibc does have it (UCLIBC_HAS_GNU_ERROR) but any portable program should definitely not use it Jan 21 12:13:47 Gnutoo, we have lots and lots of support for legacy cruft and glibc compat stuff, really Jan 21 12:14:09 Gnutoo, if something specific is missing that makes sense then please let us know, as always Jan 21 12:18:35 Gnutoo: you are uclibc dev? Jan 21 12:19:19 Gnutoo, if your gcc has support for it, you should turn off -ftree-dominator-opts, tree loop optimizers, strength reducing and set proper alignment and stack boundary for your target IFF you care about size. I wouldn't use -fexpensive-optimizations, but YMMV Jan 21 12:20:41 hrw, i don't think he is, at least i don't remember him offhand (I'm the maintainer) Jan 21 12:21:29 blindvt: I had to ask you not him - wrong typed Jan 21 12:21:38 03Koen Kooi  07org.openembedded.dev * r50dbefb891 10openembedded.git/packages/blueprobe/blueprobe_0.18.bb: blueprobe: bump PR Jan 21 12:22:14 hrw, ah, yes. So: yes :) Jan 21 12:23:01 blindvt: nice to meet you Jan 21 12:23:18 hrw, my pleasure Jan 21 12:24:33 ok thanks a lot Jan 21 12:29:54 Gnutoo, i've 2 smallish patches accumulated for gettext to make it behave, fwiw: http://repo.or.cz/w/buildroot.git?a=tree;f=package/gettext;hb=HEAD HTH Jan 21 12:30:14 blindvt, thanks a lot Jan 21 12:31:35 fortunately i don't need all those insane and bloated locale stuff myself :P Jan 21 12:33:51 03Fraxinas  07org.openembedded.dreambox * r5efc44bac9 10openembedded.git/packages/enigma2/enigma2.bb: add gst-plugin-cdio for audio-cd playback Jan 21 12:33:52 03Fraxinas  07org.openembedded.dreambox * rc5d844d755 10openembedded.git/packages/enigma2/enigma2.bb: add gst-plugin-vcdsrc for playback of "original" VCDs Jan 21 12:33:53 03Felix Domke  07org.openembedded.dreambox * r86127d5970 10openembedded.git/: Jan 21 12:33:53 Merge branch 'org.openembedded.dreambox' of git://git.opendreambox.org/git/fraxinas/opendreambox-1.5 into org.openembedded.dreambox Jan 21 12:33:55 Conflicts: Jan 21 12:33:57 packages/enigma2/enigma2.bb Jan 21 12:37:03 03Michael 'Mickey' Lauer  07org.openembedded.dev * r0a192f3f71 10openembedded.git/conf/distro/openmoko.conf: openmoko.conf: move to bluez4 Jan 21 12:37:04 03Michael 'Mickey' Lauer  07org.openembedded.dev * r104ab0d94d 10openembedded.git/packages/bluez/ (4 files in 2 dirs): bluez4: packaging tweaks; remove bluez4-utils in favour of that Jan 21 12:39:41 hello all Jan 21 12:44:33 03Michael 'Mickey' Lauer  07org.openembedded.dev * r5291f1c8f6 10openembedded.git/conf/distro/include/sane-srcrevs.inc: sane-srcrevs: bump FSO bundle Jan 21 12:45:09 yo likewise Jan 21 12:47:34 elo mickeyl Jan 21 12:48:00 mickeyl: I will push some python stuff related to rpm/yum Jan 21 13:07:11 hi hrw Jan 21 13:07:12 sounds good Jan 21 13:08:20 03Michael 'Mickey' Lauer  07org.openembedded.dev * rb5ed661065 10openembedded.git/packages/bluez/bluez4_4.27.bb: bluez4: s/alsa-libs/alsa-lib/ Jan 21 13:09:08 mickeyl: hey Jan 21 13:09:09 all: hey Jan 21 13:09:11 another attempt to get rpm support in OE Jan 21 13:30:24 oh.. Jan 21 13:30:45 * Jay7 tries to ask on #linux about some interface to wait for devices Jan 21 13:32:44 Jay7: like netlink? Jan 21 13:33:35 robtaylor: I needed wait in init-process to all storage devices with partitions be detected by kernel and filled in /proc/partitions Jan 21 13:33:54 Jay7: i guess you're not using hal? Jan 21 13:34:03 oh, in init! Jan 21 13:34:09 ook Jan 21 13:34:19 robtaylor: yes, I have single init process with /proc :) Jan 21 13:34:24 and may be with /sys Jan 21 13:34:33 Jay7: yeah, netlink is your answer here, i think Jan 21 13:34:45 robtaylor: where I can read about this? Jan 21 13:35:13 Jay7: see udev source (lib/libudev-monitor.c) Jan 21 13:35:27 robtaylor: thanks :) Jan 21 13:35:58 (or at least, i think thats what you want ;)) Jan 21 13:36:04 Jay7: np, good luck Jan 21 13:36:28 robtaylor: I'm going to check this :) Jan 21 13:36:52 Jay7: you can probably just use that library directly pretty easily Jan 21 13:37:25 well, excpet that is GPL :/ Jan 21 13:40:18 robtaylor: hm.. I have no lib/ dir in udev-1.24 Jan 21 13:40:41 Jay7: udev/lib Jan 21 13:41:29 robtaylor: which udev version you checking? :) Jan 21 13:41:44 I have udevmonitor.c file Jan 21 13:41:48 is it? Jan 21 13:42:20 Jay7: i'm looking at git head Jan 21 13:42:37 udev/lib/libudev-monitor.c Jan 21 13:42:49 can you give me url? Jan 21 13:43:09 git://git.kernel.org/pub/scm/linux/hotplug/udev Jan 21 13:43:58 Jay7: dunno what version of udev you have there. udev doesnt have x.y versionsing, its get versions like '137' Jan 21 13:44:07 (which the current version) Jan 21 13:44:11 124 is my :) Jan 21 13:44:25 seems too old now :) Jan 21 13:45:02 ah, yeah, the lib happend after that Jan 21 13:45:14 i'm sure the code is somewhere in there though ;) Jan 21 13:45:23 Jay7: use rootdelay maybe? Jan 21 13:45:26 make sure you use the right udev code for your kernel version Jan 21 13:45:41 hrw: I don't have root device :) Jan 21 13:46:01 kexecboot started from initramfs Jan 21 13:46:25 ah right Jan 21 13:46:59 Jay7: i am of course presuming 2.6 kernel here ;) Jan 21 13:47:27 robtaylor: yep, sure :) Jan 21 13:54:23 hm.. netlink and kernel uevents is right thing.. Jan 21 13:55:12 but seems that we want impossible thing Jan 21 13:57:29 ant_work: hey :) Jan 21 13:58:31 ant_work: we can't wait for all user's storage devices to appear.. Jan 21 13:58:51 because we don't know what exactly devices user have inserted.. Jan 21 13:59:28 we can only wait some time to allow kernel to fill /proc/partitions Jan 21 14:05:05 hi! im am using openembedded to build the Angstrom Distribution for an at91sam9260-ek. The kernel starts but doesnt find the JFFS2. :-( What mdtblock should i use? Jan 21 14:09:48 Jay7, so, if you cannot 'udevadm settle', then it sounds a bit like you'd want something like this, adjusted to your desired filesystems: http://paste.debian.net/26538/ Jan 21 14:11:30 blindvt: there is known partitions (from fstab) :) we have unknown devices and partitions.. so better way is just to wait sometime.. Jan 21 14:51:57 03Michael 'Mickey' Lauer  07org.openembedded.dev * r4f89a0ec59 10openembedded.git/ (2 files in 2 dirs): gsm0710muxd: bump to 0.9.3 release Jan 21 15:06:38 admins: did something change on the ML server? : Relay access denied (state 14). Jan 21 15:07:14 that was from my GMail account, which has been subscribed for months, no changes on my side. Maybe Google changes? Jan 21 15:07:17 thorsten1: does it find nand? Jan 21 15:10:43 hrw , doesnt find the jffs2 filesystem in nand! :-( i see in the kernel conf from angstrom that jffs2 ist not installed in kernel module, this night i want to compile it again :-) Jan 21 15:10:59 hopefully it runs after that :-) Jan 21 15:11:14 thorsten1: I use at91sam9263ek board here but with nfsroot Jan 21 15:11:51 thorsten1: check does it have jffs2 and nand in kernel config at all Jan 21 15:17:53 nand ist found :-) in kernel config in oe/package/linux/linux-2.6.20/at91sam9260/defconf the setting is CONFIG_JFFS2_FS = n, that might be the error :-) Jan 21 15:21:29 blindvt, i didn't remove the "optimisations" but I used a patch...and it still doesn't compile...i'll see that later... Jan 21 15:23:14 likewise, I just moved dns Jan 21 15:23:30 do other oe.org names resolve Jan 21 15:24:36 likewise, ping Jan 21 15:25:24 Host git.openembedded.org not found: 3(NXDOMAIN) Jan 21 15:25:29 www.openembedded.org has address 80.81.242.146 Jan 21 15:25:33 openembedded.org has address 80.81.242.146 Jan 21 15:25:33 never was there :) Jan 21 15:25:41 I think it is a CNAME issue Jan 21 15:25:48 bugs.openembedded.org has address 80.81.242.146 Jan 21 15:25:52 lists.openembedded.org has address 85.214.40.226 Jan 21 15:26:12 I think lists should be a CNAME to openembedded.org for some reason Jan 21 15:26:37 Crofton: nameresolve.com is new dns hoster? Jan 21 15:26:58 yes Jan 21 15:27:11 ok, I changed lists to match the old config Jan 21 15:27:19 I wasn't sure why it was a CNAME Jan 21 15:27:27 what should git be? Jan 21 15:27:34 same as www? Jan 21 15:27:48 /bin/sh: git.openembedded.net: command not found Jan 21 15:27:54 git.openembedded.net has address 80.81.242.146 Jan 21 15:28:00 so same Jan 21 15:28:20 www, bugs, git are on amethyst. lists are on ltg Jan 21 15:28:25 right Jan 21 15:28:57 I'm still getting results from easydns Jan 21 15:29:06 * * OE Bug 4994 has been created by malcolm.crossley2(AT)gefanuc.com Jan 21 15:29:07 git.oe.org is in Jan 21 15:29:08 * * Invalid libstdc++.la and libsupc++.la files Jan 21 15:29:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4994 Jan 21 15:29:18 I am getting from nameresolve.com Jan 21 15:29:39 what do you get for lists now? Jan 21 15:30:08 added gitweb also Jan 21 15:30:24 lists.openembedded.org has address 85.214.40.226 Jan 21 15:30:34 not a cname Jan 21 15:30:36 hmmm Jan 21 15:30:41 need to check something Jan 21 15:30:46 ;; AUTHORITY SECTION: Jan 21 15:30:46 openembedded.org. 83064 IN NS ns4.nameresolve.com. Jan 21 15:31:27 openembedded.org needs to point at ltg Jan 21 15:31:36 according to old easydns config Jan 21 15:33:13 Jay7: re Jan 21 15:33:22 ant_work: :) Jan 21 15:33:40 http://www.sxlist.com/techref/ti/msp430/rwsdmmcs.htm Jan 21 15:33:43 ^^ Jan 21 15:33:59 likewise, when you see lists be a CNAME, can you try again, I'll bbiab Jan 21 15:34:59 Jay7: if even they retry 50 times...it means there is no precise delay..just a guess Jan 21 15:36:41 33 commits added... Jan 21 15:37:23 ant_work: I'll move scan_devices more closer to key reading loop Jan 21 15:37:36 it can take some delay Jan 21 15:37:48 but sleep should be added anyway.. Jan 21 15:38:02 what puzzles me is that CF interface should be even slower Jan 21 15:38:17 but if I insert CF all is seen (CF+SD) Jan 21 15:38:45 may be CF should be initialized somehow Jan 21 15:38:49 I even found sleep(3) in most initscripts for CF init Jan 21 15:38:55 in this time SD can be processed Jan 21 15:39:05 really don't know... Jan 21 15:39:11 I'm too.. Jan 21 15:39:19 so sleep(3) looks reasonable :) Jan 21 15:39:27 SD is simpler then CF Jan 21 15:40:11 hrw: what is scanned at first? is it async? Jan 21 15:40:16 no idea Jan 21 15:40:28 kernel guru wanted :) Jan 21 15:47:41 Jay7: tell problem Jan 21 15:47:56 Jay7: i'll read back log... Jan 21 15:48:09 likewise: that is :) Jan 21 15:48:43 likewise: we should wait for block devices (storages) will be detected and all partitions will be filled in /proc/partitions Jan 21 15:49:17 but we don't know which card will be inserted (if one) Jan 21 15:49:19 Jay7: are you mounting a rootfs? Jan 21 15:49:30 likewise: no :) Jan 21 15:49:38 any storage Jan 21 15:49:45 Jay7: you want to detect insertion in userspace, any storage? Jan 21 15:49:54 SD/CF/microdrive and may be usb Jan 21 15:50:06 Jay7: so, all sd devices? Jan 21 15:50:19 Jay7: so it's more of a udev question? Jan 21 15:50:32 likewise: just already inserted devices at least Jan 21 15:51:22 Jay7: i think udev does run for already inserted devices (detected by the kernel) also Jan 21 15:51:48 likewise: we don't have udev (neither the space for) Jan 21 15:52:07 likewise: other question - is devices detected in async? Jan 21 15:52:36 Jay7: hmm, in recent kernels, yes more and more async detection is added Jan 21 15:52:39 *are Jan 21 15:54:04 my vote for stopping this discussion then and use sleep(3) :) Jan 21 15:54:53 even when we will read UEVENT's and process them we should wait some time Jan 21 15:55:14 easier is just wait :) Jan 21 16:00:07 Jay7: can you change your tactic and just deal dynamically with devices getting inserted? Jan 21 16:01:28 robtaylor: I can but this will make code more complicated.. Jan 21 16:01:41 Jay7: may be a lot more reliable in the end Jan 21 16:02:03 * robtaylor usually finds these things are a trade off between coding time and customer service time ;) Jan 21 16:02:20 easiest way is to use inotify on /proc/partitions :) Jan 21 16:02:43 if inotify works on procfs files Jan 21 16:02:58 Jay7: for WRITE on slow cards there are patches flowing Jan 21 16:03:05 https://kerneltrap.org/mailarchive/linux-kernel/2008/11/12/4121444 Jan 21 16:04:58 robtaylor: yes, this is trading :) Jan 21 16:05:32 03Jeremy Lainé  07org.openembedded.dev * r378f7f0fbf 10openembedded.git/packages/tzdata/tzdata_2007k.bb: Jan 21 16:05:32 tzdata: make it possible to set default timezone Jan 21 16:05:32 * make it possible to choose the default timezone by Jan 21 16:05:32 setting the DEFAULT_TIMEZONE variable Jan 21 16:08:42 * Jay7 is going to try inotify on /proc/partitions Jan 21 16:09:13 then we can make scan_devices() event-driven :) Jan 21 16:13:00 * Jay7 is going to eat something and do some tests Jan 21 16:19:37 re Jan 21 16:19:40 hi blindvt Jan 21 16:21:41 hm Jan 21 16:21:42 strange Jan 21 16:22:00 03Michael 'Mickey' Lauer  07org.openembedded.dev * r585ed3fed2 10openembedded.git/packages/connman/ (connman.inc connman_0.8.bb): connman: add RDEPENDS to dynamic packages, enable more plugins Jan 21 16:22:15 gnutoo is the second person who hits the erorr_ürint_progname again Jan 21 16:23:43 hi woglinde Jan 21 16:23:47 he pb Jan 21 16:25:13 damn flew Jan 21 16:25:26 er? Jan 21 16:25:26 whole family got it Jan 21 16:25:30 oh, flu Jan 21 16:25:34 args Jan 21 16:25:36 sorry Jan 21 16:25:36 heh Jan 21 16:25:37 heh Jan 21 16:26:00 cant remeber I was this much ill Jan 21 16:26:11 yeah, it's no fun Jan 21 16:38:01 Jay7: sys/block should superceed proc/partitions Jan 21 16:38:07 oneday Jan 21 16:38:09 http://oss.oracle.com/pipermail/ocfs2-tools-devel/2007-November/000475.html Jan 21 16:38:42 ant_work: hm.. Jan 21 16:38:47 pb_: poor 'wogrippe' Jan 21 16:40:22 Jay7: let use sleep(3) and /proc/partitions for this release Jan 21 16:41:03 (it's ready now and you have other works in queue ;-) Jan 21 16:43:00 ant_work: ok, seems that is better plan :) Jan 21 16:43:12 how is your build? Jan 21 16:43:23 building :) Jan 21 16:43:34 Running task 2529 of 3283 Jan 21 16:43:48 ah..console-image Jan 21 16:44:44 if you find time pls check and sanitize devicesan.c Jan 21 16:44:57 git complains about whitespaces Jan 21 16:45:17 when we patch that file Jan 21 16:49:47 ant_work: what patch complains? Jan 21 16:50:07 I've remember only typo fix against devicescan.c :) Jan 21 16:50:26 ah.. debug Jan 21 16:51:53 hm.. Jan 21 16:52:06 actually the last patch was for Rescan Jan 21 16:52:13 and was kexecboot.c Jan 21 16:52:33 I can send you new patches against 4.0 :) Jan 21 16:53:06 but check for mixed tab and spaces! Jan 21 16:53:28 old patches was made with -b Jan 21 16:53:41 now I don't use -b :) Jan 21 16:54:57 ok, pls send me the last patches when you're ready Jan 21 16:57:52 ant_work: heh, right Jan 21 16:57:55 zecke: hail zecke Jan 21 16:58:02 ~seen thebohemian Jan 21 16:58:03 ~seen thebohemian_ Jan 21 16:58:05 thebohemian was last seen on IRC in channel #oe, 7d 1h 24m 40s ago, saying: 'GNUtoo-desktop: I've written a proposal. see ML'. Jan 21 16:58:09 i haven't seen 'thebohemian_', zecke Jan 21 16:58:31 woglinde: hey, do you know if robert is at the c-base tonight? Jan 21 16:58:42 Ahe zecke Jan 21 16:58:48 zecke sorry dont know Jan 21 16:59:11 maybee he is somewhere else, because he is not at work this week Jan 21 17:07:40 zecke: will you visit the space station? Jan 21 17:08:07 maybe Jan 21 17:31:58 hmm, is the latest alsa-utils broke? Jan 21 17:40:37 broonie: ping Jan 21 17:40:55 alsa-utils failed building because it can't find the command xmlto Jan 21 17:44:42 hi, I'm building openmoko with openembedded (mokomakefile). I also need libglade to build another application but it seems it is not anymore in the build environment? Jan 21 17:44:58 Is there any way to add it? Jan 21 17:47:28 greetings! has anyone tried building qt-embedded lately? Jan 21 17:49:39 qt-embedded fails for me with this http://pastebin.ca/1314910 Jan 21 17:50:08 broonie: you have at91sam9263ek board? Jan 21 17:55:25 broonie: if yes, then do you have sound working on it? Jan 21 17:55:29 * hrw has to go Jan 21 17:55:50 I am not pleased with alsa-utils needing xmlto Jan 21 17:57:48 Crofton|irssi: for documentation? Jan 21 18:04:49 jimsheldon: I built it not so long ago and it worked for me then Jan 21 18:05:21 erbo: oh good, actually I just saw it's building in the tinderbox so that's good Jan 21 18:05:44 But I think I build r3 and not r4 Jan 21 18:05:51 zecke: not sure why, but we have no OE package, so I had to install on my machine Jan 21 18:05:59 but the r4 changes shouldn't cause your error I guess Jan 21 18:06:01 erbo: I'm trying to build it in an old OE tree though, I'm sure that's where the problem is... Jan 21 18:06:02 I do not like things that break builds Jan 21 18:06:11 erbo: cool, good to know Jan 21 18:06:34 Crofton|irssi: sounds like docbook to me --disable-doc on configure Jan 21 18:06:59 Crofton|irssi: the situation is worse, xmlto/docbook will download stuff from the net. so a offline build will fail Jan 21 18:09:15 urg Jan 21 18:10:19 woglinde, are you there? Jan 21 18:10:31 woglinde, oops hello Jan 21 18:11:37 re Jan 21 18:11:42 gnutoo *g* Jan 21 18:11:48 you are lucky Jan 21 18:12:01 woglinde, why? Jan 21 18:12:17 bescause I just returned in front of my computer Jan 21 18:12:26 woglinde, ah ok lol Jan 21 18:13:04 woglinde, you seem to use uclibc so I wondered if you already had a fix for that: http://rafb.net/p/qTt5d174.html ? Jan 21 18:13:42 hm the gettext error_print_progname? Jan 21 18:13:48 woglinde, yes Jan 21 18:14:16 hm Jan 21 18:14:26 I will took the differten patch from buildroot Jan 21 18:15:04 but we already had the patch in Jan 21 18:15:10 woglinde, yes we had Jan 21 18:15:16 only you and zecke is hitting this Jan 21 18:15:17 woglinde, I saw that Jan 21 18:15:19 dont know why Jan 21 18:15:22 ok Jan 21 18:16:04 woglinde, I did something nasty...I moved my tmpdir from ~/oetmp// to ~/oetmp/ could it be that? Jan 21 18:16:22 perhaps Jan 21 18:16:25 dont know Jan 21 18:16:51 gnutoo have you enabled locales? Jan 21 18:16:53 woglinde, did zecke do the same thing? Jan 21 18:16:58 yes I have Jan 21 18:17:00 gnutoo hm no Jan 21 18:17:04 ok Jan 21 18:17:05 ah okay Jan 21 18:17:07 zecke: looks like --disable-doc does not work in alsa-utils :( Jan 21 18:17:10 try to build without Jan 21 18:17:14 locales Jan 21 18:17:15 ok thanks Jan 21 18:17:21 maybee thats why Jan 21 18:19:38 woglinde, ok thanks a lot Jan 21 18:27:05 * * OE Bug 4579 has been RESOLVED (FIXED) by Jan 21 18:27:07 * * Fix tangogps dependencies Jan 21 18:27:09 * * http://bugs.openembedded.net/show_bug.cgi?id=4579 Jan 21 18:28:51 03Sander van Grieken  07org.openembedded.dev * r2002eb5dd5 10openembedded.git/packages/pingus/ (5 files in 2 dirs): pingus: new recipe; A lemmings-clone based on SDL Jan 21 18:29:06 * * OE Bug 4711 has been RESOLVED (FIXED) by Jan 21 18:29:07 * * new bb recipe for Pingus, the free lemmings clone Jan 21 18:29:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4711 Jan 21 18:34:19 what is the best way to configre/enable my usb dongle; if i reboot the module gets loaded, but the if is not listed in ifconfig Jan 21 18:34:38 sry wrong channel Jan 21 18:36:44 03Koen Kooi  07org.openembedded.dev * rc953deb821 10openembedded.git/packages/alsa/alsa-utils_1.0.18.bb: alsa-utils 1.0.18: def pref -1 till someone fixes it so it actually builds Jan 21 18:39:08 03Koen Kooi  07org.openembedded.dev * r47e9365d31 10openembedded.git/contrib/angstrom/build-feeds.sh: angstrom feed builder: add pingus and frozen-bubble Jan 21 18:39:31 03Elena of Valhalla  07org.openembedded.dev * r29541a1cc5 10openembedded.git/packages/tuxmath/tuxmath_1.6.0.bb: tuxmath: new recipe; A math game based on SDL Jan 21 18:42:05 * * OE Bug 4516 has been RESOLVED (FIXED) by Jan 21 18:42:07 * * new recipe for tuxmath Jan 21 18:42:09 * * http://bugs.openembedded.net/show_bug.cgi?id=4516 Jan 21 18:42:39 03Koen Kooi  07org.openembedded.dev * r00251a2295 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Jan 21 18:42:40 03Koen Kooi  07org.openembedded.dev * r1b350fb987 10openembedded.git/packages/pingus/pingus_0.7.2.bb: Jan 21 18:42:41 pingus: fix style issues and packaging Jan 21 18:42:43 * machine specific start script ?!?!?! Can't it check /proc/cpuinfo? Jan 21 18:47:05 * * OE Bug 4755 has been RESOLVED (WONTFIX) by Jan 21 18:47:07 * * Recipes for vala-0.4.0 Jan 21 18:47:09 * * http://bugs.openembedded.net/show_bug.cgi?id=4755 Jan 21 18:47:40 Would someone know which package ships xmlto? Jan 21 18:49:54 woglinde, same error...strange...i did a ENABLE_BINARY_LOCALE_GENERATION = "0" Jan 21 18:50:42 03Erik Boto  07org.openembedded.dev * rebb2efd204 10openembedded.git/packages/libgpod/libgpod_0.6.0.bb: libgpod: new recipe; A library to access the Apple iPod Jan 21 18:53:02 I'm building an application for Openmoko with "bitbame -b .../app.bb" and everything works except that the .ipk package is not created Jan 21 18:53:06 * * OE Bug 4982 has been RESOLVED (FIXED) by Jan 21 18:53:08 * * new package, libgpod, recipe attached Jan 21 18:53:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4982 Jan 21 18:53:36 Do I need to set a value somewhere to create the ipk package ? Jan 21 18:53:59 In my case it is not failing it is just not trying to create the ipk Jan 21 18:55:59 03Koen Kooi  07org.openembedded.dev * r65a91cae41 10openembedded.git/conf/checksums.ini: checksums.ini: *sigh* add missing checksums for the last 2 commits (pingus and tuxmath) Jan 21 18:56:10 03Koen Kooi  07org.openembedded.dev * rc6b694a234 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Jan 21 18:57:45 03Koen Kooi  07org.openembedded.dev * r513858c459 10openembedded.git/conf/checksums.ini: checksums.ini: Look, another recipe added without checksums: libgpod Jan 21 18:58:45 03Ken  07org.openembedded.dev * r57b3f1b6c9 10openembedded.git/ (2 files in 2 dirs): bootchart-lite: new recipe; bootchart diagnostic tool for embedded systems Jan 21 19:02:06 jmichel_ why you dont use bitbake packagename? Jan 21 19:02:13 hallo... Jan 21 19:02:48 hi timtimred Jan 21 19:02:54 woglinde: It's my own application with my .bb file Jan 21 19:03:27 jmichel_ hm this didnt answer my queston Jan 21 19:04:10 woglinde: well packagename will not be recognised by bitbake Jan 21 19:04:40 woglinde: because it is in another folder not part of my moko folder Jan 21 19:06:06 * * OE Bug 4758 has been RESOLVED (FIXED) by Jan 21 19:06:08 * * Recipe for bootchart-lite Jan 21 19:06:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4758 Jan 21 19:06:54 jmichel_ then put it there Jan 21 19:07:02 or use oeoverlay Jan 21 19:07:33 otherwise you have not support configuartion and so you will hardy get support Jan 21 19:09:06 * * OE Bug 4874 has been RESOLVED (FIXED) by Jan 21 19:09:08 * * Recipe for mumpot-0.3.2 Jan 21 19:09:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4874 Jan 21 19:11:28 woglinde: you mean that I should put my code under the ~/moko folder to build my app? Sorry I guess I misunderstood some part of the proceedure Jan 21 19:12:07 * * OE Bug 4973 has been RESOLVED (FIXED) by Jan 21 19:12:09 * * new recipe - directfb-examples_1.0.1.bb Jan 21 19:12:11 * * http://bugs.openembedded.net/show_bug.cgi?id=4973 Jan 21 19:13:04 recipes should be under oopenembedded/package/packagname/packagefoo_version.bb Jan 21 19:16:34 03Andreas Kemnade  07org.openembedded.dev * r08564690e2 10openembedded.git/packages/mumpot/ (files/mumpot-cyclemap.desktop mumpot_0.3.2.bb): mumpot: add version 0.3.2 Jan 21 19:16:45 03Michael 'Mickey' Lauer  07org.openembedded.dev * rf18b0303e6 10openembedded.git/packages/nonworking/timidity/timidity_2.13.2.bb: timidity: new nonworking recipe; needs host tool Jan 21 19:21:06 * * OE Bug 4387 has been RESOLVED (FIXED) by Jan 21 19:21:09 * * pushd/popd bashism in recipes Jan 21 19:21:11 * * http://bugs.openembedded.net/show_bug.cgi?id=4387 Jan 21 19:27:06 hrw: Yes, but i've not actually powered it up yet. Jan 21 19:27:35 hrw: Support for the audio was contributed by Atmel so I believe it's well tested. Jan 21 19:27:38 broonie: you see my email abobut alsa-utils? Jan 21 19:27:46 Crofton|irssi: not yet Jan 21 19:28:00 it is not building for some of us Jan 21 19:28:12 Oh, bugger. Jan 21 19:28:16 I can make it build by installing xmlto Jan 21 19:28:27 but sakoman does that and gets a filed sf.net fetch Jan 21 19:28:28 Worked for me. Jan 21 19:28:35 you'd have to ask him for the detail Jan 21 19:28:38 heh Jan 21 19:28:42 of course :) Jan 21 19:29:55 Hrm. I'll assume he's running on Beagleboard or somesuch Jan 21 19:30:03 yeah Jan 21 19:30:11 it is a build machine dependency Jan 21 19:30:28 soemthign in the alsa-util build process runs xmlto Jan 21 19:30:38 check your machine, I bet it is installed Jan 21 19:30:46 It is, definitely. Jan 21 19:30:58 rename it and rebuild alsa-utils Jan 21 19:31:11 I'm not on a machine with OE ATM. Jan 21 19:31:14 heh Jan 21 19:31:15 I'll have a look later. Jan 21 19:31:18 thanks Jan 21 19:31:27 koen marked it as def pref = -1 for now Jan 21 19:31:36 don't see your e-mail at all Jan 21 19:31:40 Where did you send it? Jan 21 19:31:48 oe-dev? Jan 21 19:32:02 Oh, right. That's very near the bottom of my mail stack. Jan 21 19:32:29 :) Jan 21 19:34:01 I'll have a look later/tomorrow morning. Jan 21 19:49:52 broonie: The error I get is: xsl:import : unable to load http://docbook.sourceforge.net/release/xsl/current/manpages/docbook.xsl Jan 21 19:50:06 Same error on both build machines Jan 21 19:53:38 03Koen Kooi  07org.openembedded.dev * r69241c097a 10openembedded.git/packages/powervr-drivers/ (4 files in 3 dirs): omap3 sgx drivers: fix initscript, fix S in kernel recipe Jan 21 19:58:39 Hmmm, any recent fix for bluez4 ? Jan 21 19:58:47 Seems it complains about libusb Jan 21 20:48:09 03Tim 'timtim' Ellis  07org.openembedded.dev * r3d36042158 10openembedded.git/packages/linux/linux.inc: linux.inc: Add lsppchg and lsppchd to linux.inc and sort the KERNEL_DEVICETREE devices by name Jan 21 20:52:34 03Tim 'timtim' Ellis  07org.openembedded.dev * reaf5979185 10openembedded.git/conf/checksums.ini: checksums.ini: Add transmission-1.40 Jan 21 20:55:14 03Tim 'timtim' Ellis  07org.openembedded.dev * rbfa4d089bd 10openembedded.git/packages/transmission/ (files/init transmission_1.40.bb): transmission: Add transmission, a bittorrent client with a nice web interface, great on NAS. Thanks dave purdy for this. Jan 21 21:36:20 he ant Jan 21 21:59:22 woglinde: flu under control ? Jan 21 22:01:36 hm yes Jan 21 22:22:58 * timtimred hooks up his home-builder to the stats thingy.... Jan 21 22:25:39 has anyone played got any bitbake recipes for the UBI/UBIFS userspace tools? Jan 21 22:25:57 *...who has played with UBI... Jan 21 22:29:14 oliford: I thought koen may have commited some Jan 21 22:29:45 oh, err, i've just discovered it's part of mtd-utils Jan 21 22:33:51 Yes, that's the case Jan 21 22:34:03 it used to be mtd-utils +mkfs.ubi Jan 21 22:34:09 but that got merged Jan 21 22:34:21 * Tartarus hasn't yet, but will "soon" get to play with the updated versions of all of this Jan 21 22:35:25 ok cool, so will just building mtdutils-native build me all the ubi image creating utils? Jan 21 22:35:45 * oliford tries Jan 21 22:36:20 Yes Jan 21 22:36:31 There's no IMAGE_CMD_ubi stuff tho Jan 21 22:36:33 also will mtd-utils build ubiattach etc for target? Jan 21 22:36:35 So you'll have to do that part yourself Jan 21 22:36:39 yes Jan 21 22:36:45 all of the target stuff is ther Jan 21 22:36:45 ok, i can probably manage that Jan 21 22:39:10 this is my first attempt at running this thing entirely off flash. I was happy with the MMC card but the platform's mailing list are pestering me for it. I did try with JFFS2 but it seems to be too slow and flaky. Jan 21 22:42:00 arg, pain. Jan 21 22:42:01 after all finding and compiling, 'ipkg upgrade mtd-utils' pulled it off the angstrom feed Jan 21 22:42:56 That I can't help with, sorry :) Jan 21 23:23:10 what can I do to get rid of this QA error?? ERROR: QA Issue: No GNU_HASH in the elf binary Jan 21 23:25:35 when you have a file:... in SRC_URI its automatically unpacked into the work dir - is there a way to have that 'not' happen for a specifc file? Jan 21 23:25:53 tharvey, for a .tgz or similar? Jan 21 23:26:02 yes, for a tar.gz Jan 21 23:26:08 AFAICT, no, unless you call it .tar.gz.somethingelse Jan 21 23:26:12 it's in base.bbclass Jan 21 23:26:34 which is also why .zip in SRC_URI can fail with DEPENDS += unzip-native Jan 21 23:26:34 ya Jan 21 23:26:55 hmmm... perhaps override unpack function if there are no other files in SRC_URI that need unpack? Jan 21 23:27:06 sakoman_: Do you have an actual build log for the failure... Jan 21 23:27:10 Perhaps Jan 21 23:28:02 * broonie notes that koen reported a second failure in his commit message but I didn't see that either; I suspect that's just broken XML packaging on his distro though. Jan 21 23:28:23 thx Jan 21 23:30:59 broonie: do you need one? Jan 21 23:32:33 Crofton|irssi: Might help, he's got xmlto installed and is probably seeing whatever subsequent error was reported. Jan 21 23:32:51 Though I think I'm just going to submit a patch bodging out xmlto support. Jan 21 23:33:13 http://rafb.net/p/HyVRkR19.html Jan 21 23:33:28 that is on my fedora machine Jan 21 23:33:40 it works if install xmlto Jan 21 23:33:44 Crofton|irssi: that's not what he's reporting. Jan 21 23:33:55 who? Jan 21 23:34:18 sakoman_: had another problem Jan 21 23:34:30 Steve - he appears to be running xmlto but it flobs later on (like I say, probably the same thing koen swa). Jan 21 23:34:38 I wish we could just turn off doc generation .... Jan 21 23:34:42 yeah Jan 21 23:34:52 something like it tried retrieving a file from sf.net Jan 21 23:34:57 * Crofton|irssi kicks sakoman_ Jan 21 23:35:00 I suspect it's just a flobbed XML toolchain and when I bodge out xmlto it'll be fine. Jan 21 23:35:08 yeah Jan 21 23:35:20 I tried some configure things, but no luck Jan 21 23:35:36 It's not accessible via configure, it's hard coded in Makefile.am Jan 21 23:35:37 we could make an xmlto native package .... Jan 21 23:35:42 dorks Jan 21 23:35:50 I'll fix upstream tomorrow. Jan 21 23:36:04 thanks, that is the best solution Jan 21 23:36:06 But joy! bitbake has decided to rebuild the toolchain! Jan 21 23:36:12 hmmm Jan 21 23:36:17 I wonder why? Jan 21 23:36:32 I did a pull to get the borken stuff in locally, it'll have picked up other stuff. Jan 21 23:38:12 This means I won't be able to test any fixes tonight. Jan 21 23:38:37 broonie: I will post my build output in a moment Jan 21 23:38:50 sakoman_: Thanks, just want to confirm it's from the man page build. Jan 21 23:41:11 anyone know what the NO_GNU_HASH QA error means? or how to go about fixing it? Jan 21 23:41:25 scruggs: I'm wondering that myself Jan 21 23:41:48 google gets 0 hits on NO_GNU_HASH Jan 21 23:41:59 maybe the GNU needs some hash? Jan 21 23:42:06 I see the check in insane.bbclass, and google has me believing its glibc related somehow Jan 21 23:42:53 scruggs make sure the app takes the LDFLAGS from enviroment Jan 21 23:43:07 perl builds doesnt do it yet Jan 21 23:43:21 ncurses has the problem too Jan 21 23:43:27 good nite Jan 21 23:43:39 broonie: http://pastebin.com/m2540ad88 Jan 21 23:43:52 back later . . . Jan 21 23:43:59 woglinde: thx... I'll check that out Jan 21 23:44:37 sakoman_: OK, confirmed it's the same xmlto run. Thanks. Jan 21 23:47:13 Crofton|irssi: Any chance you could test a fix for me? Jan 21 23:47:49 03Tim 'timtim' Ellis  07org.openembedded.dev * r0319c765a9 10openembedded.git/packages/linux/ (3 files in 2 dirs): linux-linkstationppc: Move from 2.6.21-rc5 to 2.6.27.10. Move to using deice tree stuff in linux.inc. Jan 21 23:48:05 broonie: possibly :) Jan 21 23:49:15 03Tim 'timtim' Ellis  07org.openembedded.dev * rfe9aa076a3 10openembedded.git/conf/machine/include/lsppc.inc: lsppc: Add KERNEL_IMAGETYPE, UBOOT_ENTRYPOINT and UBOOT_LOADADDRESS info and kernel-devicetree to MACHINE_EXTRA_RDEPENDS Jan 21 23:49:56 Crofton|irssi: http://cassiel.sirena.org.uk/~broonie/0001-alsa-utils-1.0.18-Remove-xmlto-usage-from-the-build.patch Jan 21 23:49:59 powerpc linkstations build/boot again. Jan 21 23:50:04 woo etc. Jan 21 23:50:08 Completely untested! Jan 21 23:50:18 http://cassiel.sirena.org.uk/~broonie/0001-alsa-utils-1.0.18-Remove-xmlto-usage-from-the-build.patch Jan 21 23:50:25 I can guarantee that it doesn't work. Jan 21 23:50:47 ok, let me see what happens Jan 21 23:52:04 heh Jan 21 23:52:41 building Jan 21 23:55:56 huh, there's a new OpenEmbedded logo? Jan 21 23:56:47 I am having a massive "I am an idiot day" Jan 21 23:56:54 kergoth: yes Jan 21 23:57:12 the buglabs guys logo designer made one for us Jan 21 23:57:22 sheesh, i need to pay closer attention. been busy with work, and going back to school Jan 21 23:57:30 where's the new logo at? Jan 21 23:57:33 school? Jan 21 23:57:37 not sure Jan 21 23:57:44 going back to pick up a CS degree Jan 21 23:57:46 Crofton|irssi: Americans call university school. Jan 21 23:57:54 it looks very "logoish" Jan 21 23:57:57 been 10 years since i was in college, and that was during my senior year of hs Jan 21 23:58:02 * Crofton|irssi is an American Jan 21 23:58:12 talk about rusty. went from 3 semesters of college calc to intermediate algebra Jan 21 23:58:16 * kergoth feels lame, stupid memory.. Jan 21 23:58:20 Crofton|irssi: Oh, duh. Sorry. Jan 21 23:58:23 * Crofton|irssi has been in school over half his life Jan 21 23:58:37 today was the first day of classes, only taking 3 Jan 21 23:58:37 and is old enogh that shoudl should be horrified Jan 21 23:58:45 but the list of how many classes i want to take keeps growing Jan 21 23:58:46 Nobody knows you're a dog and all that. Jan 21 23:58:51 kergoth: math is hard to get back :( Jan 21 23:58:51 debating just continuing to go to school indefinately Jan 21 23:59:05 heh Jan 21 23:59:17 still working? Jan 21 23:59:18 heh, they had like 4 different math placement tests, i had no idea which to take, had no idea what i'd remember Jan 21 23:59:33 yea, full time work, thats why only 3 classes. and one of those is woodworking, just to get a new hobby Jan 21 23:59:47 cool Jan 22 00:00:08 what degree are you going for? Jan 22 00:00:43 me? I finshed a MS in EE a year ago december Jan 22 00:00:47 i don't particularly -want- a cs degree, or -like- my job, but it pays so well that i coudl never leave it.. so my new plan is to do everything i can to increase my wage, so i can retire as soon as is feasable and do part time contracting from there Jan 22 00:00:50 ahh nice Jan 22 00:00:51 but I spent like 8 years on an MS :) Jan 22 00:01:05 good plan :) Jan 22 00:01:58 course, i kind of need to stop spending so much money to make that plan happen :) Jan 22 00:02:07 * kergoth is $2700 over budget this month.. Jan 22 00:02:12 Worked for my dad. Except then he found he got jobs that he liked so the part time bit went out the window. Jan 22 00:02:13 that's is also a good habit to get into Jan 22 00:02:47 better to be in a position where people are asking you to work for them, than begging for work Jan 22 00:02:49 broonie: well, if you find one you love, thatd work too... i'm just skeptical of such a thing ever happening :) Jan 22 00:06:42 I think part of it was that he liked *starting* jobs so the contracting smostly stopped him getting bored Jan 22 00:07:34 heh Jan 22 00:07:49 yeah, the first few weeks are always the best, then it gets tedious :) Jan 22 00:07:52 heh, i can see that. most of my jobs have been interesting for the first month or so.. Jan 22 00:07:54 heh :) Jan 22 00:09:03 wonder if itd be worth trying to get a second job, part time.. guess itd be hard to manage a second computing job, would almost have to go with something nonskilled Jan 22 00:12:03 (K) clerk? :) Jan 22 00:12:37 broonie: it built Jan 22 00:12:42 \o/# Jan 22 00:13:06 guess part time retail could be useful, discount Jan 22 00:13:17 kergoth: don't over do it :) Jan 22 00:13:22 I'll punt it into bugzilla, then. Jan 22 00:13:30 true, i'd hate to burn out Jan 22 00:13:43 you can push right? Jan 22 00:13:49 nope Jan 22 00:18:20 Crofton|irssi: Thanks for testing. Jan 22 00:18:28 night Jan 22 00:18:52 * Jay7 -> sleep() Jan 22 00:19:07 * * OE Bug 4995 has been created by broonie(AT)sirena.org.uk Jan 22 00:19:09 * * Fix alsa-utils 1.0.18 build Jan 22 00:19:11 * * http://bugs.openembedded.net/show_bug.cgi?id=4995 **** BEGIN LOGGING AT Thu Jan 22 01:14:16 2009 **** ENDING LOGGING AT Thu Jan 22 02:59:57 2009