**** BEGIN LOGGING AT Thu Mar 22 02:59:58 2007 Mar 22 03:21:08 arg Mar 22 03:21:48 I'm trying to run bitbake from a sub-shell and it just spews "InvalidTask exception" no matter what I do.. works outside the subshell. Any ideas? Mar 22 03:28:31 oh ignore that Mar 22 03:28:47 I'm a super idiot. The subshell script had "-c clear" :) Mar 22 07:42:41 03rwhitby 07org.oe.dev * r95ffaab0... 10/ (1 conf/distro/include/slugos.inc): slugos.conf: Fix FULL_OPTIMIZATION, cause it broke the wireshark package by using gcc 4 specific options. Mar 22 07:42:46 03rwhitby 07org.oe.dev * r7180c92e... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages.bb: Add wireshark. Mar 22 07:54:00 03koen 07org.oe.dev * rd0c48feb... 10/ (1 packages/geoclue/geoclue_svn.bb): geoclue: don't build panelapplet to prevent pulling in half of gnome Mar 22 08:07:18 03Laibsch 07org.oe.dev * rdef854b5... 10/ (14 files in 2 dirs): Mar 22 08:07:18 glibc: recommit my glibc clean-up (minus the one change that broke things) Mar 22 08:07:18 only semantic change is what is being discussed in bug 1987 Mar 22 09:06:41 <_law_> can i already switch to bitbake 1.8.0 ? does it work with OE ? Mar 22 09:08:43 yes and yes Mar 22 09:08:52 you can even switch to 1.9 if you want Mar 22 09:12:24 <_law_> ok so i change to the new 1.8 stable version Mar 22 10:15:21 morning all Mar 22 10:16:32 good monring Mar 22 10:21:43 ~seen woglinde Mar 22 10:22:19 woglinde was last seen on IRC in channel #oe, 6d 12h 36m 31s ago, saying: 'hi florian'. Mar 22 10:37:59 morning Mar 22 10:42:34 03hrw 07org.oe.dev * r1c69dd15... 10/ (5 files in 2 dirs): contacts: added 0.4, unified recipes, fixed PV in svn version, disabled gnome-vfs (from Poky) Mar 22 10:42:39 03hrw 07org.oe.dev * rf0a5ff5a... 10/ (5 files in 2 dirs): tasks: added 0.4, unified recipes, fixed PV in svn version (from Poky) Mar 22 10:42:43 03hrw 07org.oe.dev * r8b675ccc... 10/ (4 files in 2 dirs): dates: unify recipes, fix PV in svn version (from Poky) Mar 22 10:42:48 03hrw 07org.oe.dev * rb3599341... 10/ (1 packages/tasks/task-base.bb): task-base: ECB is needed to get WEP working Mar 22 10:45:22 good morning all Mar 22 10:51:41 hi koen|lab Mar 22 10:59:17 koen|lab: will we add avahi-daemon into default installation in angstrom? Mar 22 10:59:33 poky do that so ssh root@c7x0.local works Mar 22 11:52:12 moin Mar 22 11:52:23 hi zecke Mar 22 11:52:35 zecke, should we update getting started for bitbake 1.8? Mar 22 11:53:08 Crofton: yes Mar 22 11:55:21 done Mar 22 11:56:27 hmm, I wonder if we should remove the FC2-4 instructions? Mar 22 11:57:02 Crofton: do they hurt? I know RH stopped supporting them but people might still use them Mar 22 11:57:16 I doubt they hurt, just clutter Mar 22 11:57:34 99% of fedora users would have upgraded Mar 22 11:57:47 although, oddly enough I run OE on a FC4 machine Mar 22 12:17:01 hrw: that's the plan Mar 22 12:17:03 (avahi) Mar 22 12:17:13 hrw: if you want to add it, go ahead Mar 22 12:18:11 ok Mar 22 12:18:19 zecke: You got any idea why bitbake is screwy? http://pastebin.ca/405769 Mar 22 12:18:28 later Mar 22 12:18:46 NAiL: no, RP might have a clue Mar 22 12:19:10 NAiL: either build started events gets fired more than once Mar 22 12:19:21 NAiL: or bitbake's forking is... welll I don't know Mar 22 12:19:32 RP: ping? Mar 22 12:19:52 It doesn't happen if I run bitbake from the cmdlin Mar 22 12:20:02 but if I nohup a script that runs bitbake, it goes wonky Mar 22 12:20:21 ah Mar 22 12:20:38 It works, albeit slow and with a lot of output Mar 22 12:21:05 what happens if you create a new process group for your script Mar 22 12:21:28 new process group? Mar 22 12:21:38 yes, new process group Mar 22 12:21:52 I wonder who you SIGHUP and if BitBake gets this and python gets confused Mar 22 12:22:56 http://pastebin.ca/406405 <-- that's the script I nohup Mar 22 12:23:26 so your script is wrong? Mar 22 12:23:32 it starts multiple bitbake sessions Mar 22 12:23:40 no Mar 22 12:23:40 NAiL: e.g. patch bitbake to print the PID Mar 22 12:23:49 NAiL: patch BitBake and show me the PID Mar 22 12:24:17 how do I get BitBake to show me the pid? Mar 22 12:24:25 bitbake is python, right? Mar 22 12:24:34 import os Mar 22 12:24:37 os.getpid() Mar 22 12:26:38 import zecke Mar 22 12:26:45 zecke.fixbugs(all) Mar 22 12:26:56 hey vivijim Mar 22 12:27:21 hrw: does apt build in poky? Mar 22 12:28:36 zecke: ok, exactly what command do I use to get it to print the pid itself? Mar 22 12:29:25 bb.msg.note(something?) Mar 22 12:29:49 NAiL: print os.getpid() Mar 22 12:29:51 koen|lab: will check Mar 22 12:29:59 hrw: thanks Mar 22 12:30:08 * hrw have nfs/avahi/phone day today Mar 22 12:30:25 * koen|lab has a stuck in the lab day today Mar 22 12:30:43 shells: e/poky/b-o/n/z/b-poky/oe Mar 22 12:31:24 ssh-to-ewi/poky-metadata/build-openmoko/neo1973/zaurus-c7x0/build-poky/oe-metadata Mar 22 12:31:42 koen|lab: poky apt does not even patch Mar 22 12:31:46 hrm Mar 22 12:32:22 zecke: ok, trying now Mar 22 12:33:21 hrw: I'm asking because indt wants to use OE to build a bootstrap with .debs Mar 22 12:35:22 hi mickeyl Mar 22 12:35:28 hey mickeyl Mar 22 12:35:29 cheers hrw Mar 22 12:35:31 hi koen Mar 22 12:36:10 ~lart gsmd for /tmp/gsm.log Mar 22 12:36:11 * ibot puts on a hockey mask and jumps out at gsmd for /tmp/gsm.log Mar 22 12:36:45 hrw: it grew to a gazillion MB for you as well? Mar 22 12:36:54 Hi! Mar 22 12:37:17 zecke: I put print os.getpid() in the beginning of bin/bitbake Mar 22 12:37:19 64M Mar 22 12:37:21 root@fic-gta01:/var/tmp$ ls gsm.log -la Mar 22 12:37:21 lrwxrwxrwx 1 root root 9 Mar 22 12:37 gsm.log -> /dev/null Mar 22 12:37:21 it only gets printed once Mar 22 12:37:29 koen|lab, hrw: Well, 200703 is kinda closing out, ant news for Angstrom, gentlemen? Mar 22 12:39:58 psokolovsky: not really Mar 22 12:40:17 koen|away, ;-( So, what to expect in general? Mar 22 12:40:25 koen|lab, ^ Mar 22 12:40:29 psokolovsky: if you (or anyone else) wants to get it ready for release, by mickeyl's and my guest Mar 22 12:41:10 I'll be swamped till the beginning of april and leave for ELC2007 after that Mar 22 12:42:01 koen|lab, is there any criteria for release somewhere? http://linuxtogo.org/gowiki/AngstromRoadmap doesn't seem to be too specific Mar 22 12:42:44 koen|lab, you are becoming a real traveller Mar 22 12:43:01 psokolovsky: basically: make angstrom-console-image and angstrom-x11-image work out of the box on 'supported' devices Mar 22 12:43:06 Crofton: yes :) Mar 22 12:43:21 You are beginning to act like a consultant Mar 22 12:44:24 koen|lab, ok. I'm not really sure I would have exact understanding what that would take, but I for sure know "loose" areas which worth improving Mar 22 12:44:29 13:44 hrw@home:tasks$ ssh root@fic-gta01.local Mar 22 12:44:36 I started to like avahi Mar 22 12:44:48 koen|lab, one concern is services, i.e. /etc/init.d/* stuff Mar 22 12:45:15 koen|lab, do we have clear policies what services should run on startup, and which on demand? Mar 22 12:47:01 I'd like to have avahi-daemon, dropbear and bluez running where appropriate Mar 22 12:47:27 and I think we should release bare-bone images people can install stuff on themselves Mar 22 12:50:36 koen|lab, well, good idea. So, would you OK cleanup of update-rc.d calls in recipes? One vivid example is irda-utils which stuff irattach to be run on startup by default. that keeps IrDA port busy and waste battery. Whereas IrDA should really be controlled by UI (and that's how it is) Mar 22 12:51:00 psokolovsky: sure Mar 22 12:51:16 psokolovsky: send mails about stuff you're not sure about to angstrom dev ml Mar 22 12:51:42 koen|lab, nice! Will do so. Mar 22 12:54:05 03hrw 07org.oe.dev * r9723a9e9... 10/ (1 packages/tasks/task-openmoko.bb): task-openmoko: ship openmoko-today, openmoko-icon-theme-standard and settings-daemon by default Mar 22 12:59:59 03hrw 07org.oe.dev * r54525e4a... 10/ (1 packages/gpe-terminal/gpe-terminal_1.1.bb): gpe-terminal: recommend rxvt-unicode Mar 22 13:00:28 AwayNAiL: what was the result of getpid? Mar 22 13:01:40 hrw: Did you have a chance to look at the nunome stuff yet? http://oz.leggewie.org/wip/nunome-1.0.1-r2/ Mar 22 13:02:18 Laibsch: I think that I will run build for it Mar 22 13:20:30 http://www.hrw.one.pl/2007/03/22/hotpluging-usbnet/ Mar 22 13:28:51 hrw: That would be nice. Mar 22 13:29:03 hrw: But it fails for me. I guess it will fail for you, too. Mar 22 13:58:35 does anyone have a machine conf file for a tomtom ONE (or an opentom distro conf file)? Mar 22 14:07:51 mccarthy: maybe TomTom itself? Mar 22 14:08:59 schurig, not on their website (AFAIK). They have a /gpl directory that has all their patches and some other source code, but no OE related stuff (they *do* have a cross compile toolchain, but that is what OE is for ;) Mar 22 14:14:08 03tmbinc 07org.oe.dreambox * ree64a963... 10/ (1 packages/dreambox/dreambox-secondstage.bb): dreambox-secondstage: use v50 for dm600pvr Mar 22 14:14:14 03tmbinc 07org.oe.dreambox * rc4bd1da9... 10/ (1 packages/dreambox/dreambox-dvb-modules.bb): dreambox-dvb-modules: use 20070227 for dm600pvr Mar 22 14:14:18 03tmbinc 07org.oe.dreambox * r02d92284... 10/ (1 conf/machine/dm600pvr.conf): conf/machine/dm600pvr.conf: use gdb,gdb-cross 6.5 Mar 22 14:14:22 03tmbinc 07org.oe.dreambox * r299a5bba... 10/ (1 packages/enigma/enigma-modem.bb): enigma-modem: update to 0.2 Mar 22 14:18:46 http://www.cyclingnews.com/photos/2006/apr06/roubaix06/index.php?id=s023 Mar 22 14:18:52 Spring is here! Mar 22 14:21:30 hmm, libtool-native-1.5.10-r4 keeps failing while compiling :( Mar 22 14:22:03 Shoragan: what is the error? Mar 22 14:22:16 it tries to configure it again, and then complains that it is already configured Mar 22 14:23:37 Shoragan: you have symlinks in path Mar 22 14:24:10 hrw: maybe bitbake should detect if anything in the path is a symlink and refuse to work if that is true Mar 22 14:24:37 schurig: would be nice Mar 22 14:25:40 schurig: it's a turing test to see if people read GettingStarted ;) Mar 22 14:26:03 zecke: I have a patch for bitbake branch 1.8: http://pastebin.ca/406513 Mar 22 14:26:06 :/ Mar 22 14:26:37 symlink to the oe checkout is ok, but not for work? Mar 22 14:26:43 or tmp Mar 22 14:26:49 zecke: without this patch, a "bitbake -v something" would just say "I cannot build XYZ", but it would not say what it missing. I had to append -D to get this info Mar 22 14:28:29 zecke: it could be bb.msg.note instead of bb.msg.error, but not bb.msg.debug isn't so nice Mar 22 14:34:57 schurig, hrw: I added that idea about symlinks and bitbake as an RFE/RFC to the BTS -> bug 2030 Mar 22 14:39:21 zecke: 13:37 < NAiL> zecke: I put print os.getpid() in the beginning of bin/bitbake Mar 22 14:39:26 zecke: 13:37 < NAiL> it only gets printed once Mar 22 14:39:28 http://www.engadget.com/2007/03/22/tyans-40-cpu-personal-supercomputer-now-shipping/ - will OE get one for buildbox? Mar 22 14:42:15 hrw, thanks, that fixed it :) Mar 22 14:44:41 Laibsch: NOTE: package nunome-1.0.1-r1: task do_compile: completed Mar 22 14:45:52 Laibsch: NOTE: package nunome-1.0.1-r1: task do_build: completed Mar 22 14:46:01 Laibsch: thats with angstrom and qte-mt Mar 22 14:46:09 hi hrw Mar 22 14:47:11 hi woglinde Mar 22 14:47:27 hrw: NICE! Mar 22 14:47:33 hrw: No changes necessary? Mar 22 14:47:39 It did not build for me. Mar 22 14:48:08 hrw: that's a complete replacement for a one-family-house central heating Mar 22 14:48:10 hrw: you said qte-mt. Did you specify that in some conf or bb file (or even on the command line)? Mar 22 14:48:25 woglinde: hi Mar 22 14:49:31 hi mr_nice Mar 22 14:53:44 schurig: thats too Mar 22 14:54:06 Laibsch: I built qte-mt by hand Mar 22 14:54:11 I can test with qte Mar 22 15:01:36 hrw: and bitbake picks that up automatically and builds for qte-mt instead of qt if qte-mt is present? Mar 22 15:01:43 if present Mar 22 15:03:01 if there is no preference set in configs Mar 22 15:04:55 NOTE: package nunome-1.0.1-r1: task do_build: completed Mar 22 15:05:20 Depends: virtual/japanese-font, libqpe1 (>= 1.2.2+cvs20070323), libqte2 (>= 2.3.10), libc6 (>= 2.5), libgcc1 (>= 4.1.1) Mar 22 15:05:50 strange. Mar 22 15:06:26 I will retry. Mar 22 15:06:43 brb Mar 22 15:06:59 or even not brb - will change some drives in desktop Mar 22 15:11:34 Is there a good way to find out if something is writing to a flash disk? I see my flash disk driver using 100% CPU for about 1/4 second every 4 seconds. Is there a known reason for this? Mar 22 15:16:11 woglinde: did you tried to get opie compiled a second? I have tried it and got into this error (http://pastebin.ca/406583). do you know if this patch is neccessary? else I will just comment it out Mar 22 15:16:26 sorry to late *g Mar 22 15:30:07 [Bre Mar 22 15:37:37 zecke: did you see my reply? Mar 22 15:42:02 03justinp 07org.oe.dev * r3103fc69... 10/ (1 packages/gnome/libgnomeui_2.16.1.bb): libgnomeui-2.16.1: add unpackagedd files to -dev and -dbg Mar 22 15:46:41 NAiL: parts of it, you say it is the same process :) Mar 22 15:47:14 yeah Mar 22 15:47:20 weird :) Mar 22 15:47:29 03justinp 07org.oe.dev * r58920574... 10/ (1 packages/gnome/libgnomeui_2.12.0.bb): libgnomeui-2.12.0: add unpackagedd files to -dev and -dbg Mar 22 15:47:40 03justinp 07org.oe.dev * r45ed69e5... 10/ (1 conf/distro/angstrom-2007.1.conf): angstrom: add PREFERRED_VERSION for libgnomeui-2.16.1 (bug #2029) Mar 22 15:54:19 hrw|gone: create a bridge device and bridge all new usbX in this network Mar 22 15:55:21 koen|lab: to build OpenMoko, org.oe.dev is all I need? Mar 22 15:56:41 zecke: yes Mar 22 15:56:55 koen|lab: distro is angstrom? Mar 22 15:57:04 that's what I use Mar 22 15:57:21 koen|lab: It is time to reflash, and to install the RSS reader Mar 22 15:58:13 good luck Mar 22 15:58:28 Collected errors: Mar 22 15:58:28 Package openmoko-dates wants to install file /usr/share/applications/dates.desktop Mar 22 15:58:28 But that file is already provided by package dates Mar 22 15:58:29 koen|lab: this implies I may face majors issues? Mar 22 15:58:32 * koen|lab stabs shlib code Mar 22 15:58:45 zecke: no, any issues is a bug Mar 22 15:58:54 koen|lab: didn't hrw apply his patch? Mar 22 15:59:03 zecke: not that I know off Mar 22 15:59:28 * koen|lab ipkg upgrade -force-overwrite ; ipkg remove dates -force-depends ; ipkg install openmoko-dates -force-overwrite Mar 22 15:59:48 the joy of suboptimal code: many coffee breaks Mar 22 16:02:39 !oebug 2029 Mar 22 16:02:40 * * Bug 2029, Status: RESOLVED, Created: 2007-03-21 13:05 Mar 22 16:02:41 * * papercrane(AT)reversefold.com: libgnomeui-2.16.1 requires gdk-pixbuf-2.0 >= 2.9.0 which is not built Mar 22 16:02:42 * * http://bugs.openembedded.org/show_bug.cgi?id=2029 Mar 22 16:21:19 hi, all! Mar 22 16:21:48 how could I set PREFERRED_PROVIDER for gdk-pixbuf-loader-png? Mar 22 16:22:12 (and why it doesn't by default in Angstrom) Mar 22 16:32:58 something that has driven me crazy for about a month now is a 2.4 kernel + OE. Mar 22 16:33:30 At one time, I don't know how I did it, I built an image that would rmmod / insmod / modprobe that 2.4 kernel. Mar 22 16:33:53 Now, I cannot get the kernel modules to work. Yes, busybox is set for a 2.4 kernel. Mar 22 16:34:16 any suggestions? anyone using a 2.4 kernel with OE? Mar 22 16:41:27 how can I get something to run before do_fetch... in order to set up some environment vars? Mar 22 16:46:16 I tried in do_fetch_prepend, but that doesn't seem to work Mar 22 16:51:42 re Mar 22 16:51:58 zecke: did not applied it - need a bit of work still Mar 22 16:52:06 ah kay Mar 22 16:52:24 someone here use nfs on uclibc system? Mar 22 16:54:15 zecke, hrw : hi Mar 22 16:54:29 Ifaistos: howdy Mar 22 16:55:08 Ifaistos: didn;t you used uclibc on your systems? Mar 22 16:55:16 hrw: yes Mar 22 16:55:32 Ifaistos: did you used nfs-utils on it? Mar 22 16:55:44 Ifaistos: or do you know what provides rpcinfo for uclibc? Mar 22 16:56:04 * hrw want to add glibc-utils to nfs-utils but this will break uclibc ;( Mar 22 16:57:51 hrw: hmm i think i tried once with uclibc 0.9.28 on efika but had some problems with rpc also Mar 22 16:58:39 Ifaistos: I have working nfs server on my Z (client too) Mar 22 16:58:46 will check does it works without glibc-utils Mar 22 17:00:02 hrw: never tested that Mar 22 17:01:07 hrw: uclibc stuff is in the back-burner at the moment. trying to figure out toolchain problems Mar 22 17:02:12 in my case i have 2 apps (apache2 and boa) that compile and run without a problem on efika (ppc603) but sigfault on walnut (ppc405) Mar 22 17:02:32 which is *very* strange INHO Mar 22 17:02:35 IMHO Mar 22 17:11:17 slapin_nb: iirc bitbake 1.8 doesn't have that problem (preffered_providers) Mar 22 17:12:06 koen|lab: quick question: Is switching from glibc to uclibc mainly a question of changing TARGET_OS ? Mar 22 17:12:24 NAiL: yes(*) Mar 22 17:12:28 heh Mar 22 17:12:44 NAiL: see what ANGSTROM_MODE does Mar 22 17:12:51 koen: ok, thanks Mar 22 17:13:01 koen, is bitbake 1.8 ready for building OE.dev? Mar 22 17:13:06 slapin_nb: yes Mar 22 17:13:08 hrw: So you build nunome successfully for both qte and qte-mt? Mar 22 17:13:23 NAiL: the trick is to seperate ipkg built with uclibc and ones with glibc Mar 22 17:13:29 koen, cool, so I just upgrade and all troubles will vanish Mar 22 17:13:35 hrw: I have trouble compiling anything at all. Despite having nuked tmp just about a week ago. Mar 22 17:13:37 slapin_nb: that's the idea Mar 22 17:13:44 * Laibsch curses his slow build host Mar 22 17:14:40 Laibsch: yes Mar 22 17:16:09 koen: Ok, could I use ANGSTROM_MODE as-is, or should I make a separate FOONAS_MODE? :) Mar 22 17:16:31 NAiL: that's up to you :) Mar 22 17:16:47 ok, I'll just use ANGSTROM_MODE Mar 22 17:16:48 NAiL: you'd probably want to include your own *glibc.inc and *uclibc.inc Mar 22 17:17:04 they should be generic enough, though Mar 22 17:17:26 (feel free to rename the .inc to something generic if you want) Mar 22 17:18:34 NAiL: it will alter deploy/ into deploy/glibc/ and deploy/uclibc/ since ipkg aren't marked with the libc version Mar 22 17:18:49 s/version/kind/ Mar 22 17:19:02 yeah, got that bit Mar 22 17:24:04 NOTE: :maximum recursion depth exceeded while evaluating: Mar 22 17:24:07 ${SOV}+cvs${SRCDATE} Mar 22 17:24:08 nice... Mar 22 17:24:56 03hrw 07org.oe.dev * r6b37af53... 10/ (3 files in 3 dirs): Mar 22 17:24:56 nfs-utils: give file with exports for mountd to get it working (from Poky) Mar 22 17:24:56 - mountd should read /etc/exports by default but it did not Mar 22 17:26:32 koen: I can't come up with a good generic name for the .inc files ;) Mar 22 17:27:28 NAiL: join the club :) Mar 22 17:27:47 wow, we have more crappy bb files each and every day Mar 22 17:27:48 hi koen, I'm not finding that documentation that you show me, about how to use an existent local cross toolchain... where are it? Mar 22 17:28:42 vivijim: http://www.openembedded.org/user-manual Mar 22 17:29:09 thanks! Mar 22 17:38:07 hmmm Mar 22 17:38:24 * koen tries to remove various pizza ingredients from his laptop Mar 22 17:38:41 hehe Mar 22 17:42:06 koen : hope it does not get "alive" like the pc on "electric dreams" Mar 22 17:42:26 mine is more of a coffee addict :) Mar 22 17:42:27 indeed Mar 22 17:43:26 man Mar 22 17:43:30 still no uclibc .29 Mar 22 17:44:01 schurig: ping Mar 22 17:44:46 heh: http://uclibc.org/lists/uclibc/2007-March/017512.html Mar 22 17:44:53 Many of my builds for angstrom fail. For example, gcc-cross fails with "configure: error: C preprocessor "gcc -E" fails sanity check" Any idea about why?" I cleared tmp about one week ago. Mar 22 17:45:58 Laibsch: which cpp is used? Mar 22 17:47:53 Are you back from FOSDEM yet? :) Mar 22 17:48:05 zecke: can your sandbox redirect calls to /usr/* to ${STAGING}/usr ? Mar 22 17:50:29 koen: we need to define use cases Mar 22 17:50:46 koen: my 'sandbox' will not be a security tool Mar 22 17:50:55 if people escape it, they deserve it Mar 22 17:51:25 koen: but what should be done, is up for discussion. I have looked at the gentoo box and I will steal canonical.c which they stole from GNU Mar 22 17:51:31 zecke: I also have been thinking about chrooting to ${D} and bindmount STAGING and mangle $PATH Mar 22 17:51:51 for applications that think DESTDIR is overrated Mar 22 17:51:53 koen: the question is, will we require root priviliges to build OE Mar 22 17:52:02 I hope not Mar 22 17:52:07 maybe in OE+ ;) Mar 22 17:52:20 chroot will require at least setuid program :) Mar 22 17:52:38 * koen wonders why the bad guys always lock up the a-team in a room with dynamite and power tools Mar 22 17:54:05 * ade|desk wonders why the bad guy in James Bond films always finds the most stupid way to kill Bond Mar 22 17:55:06 mickey_away: [18:54] tko: hildon-1 and hildon-desktop should be runnable against stock gtk Mar 22 17:55:17 later all Mar 22 18:19:15 is anyone using any of the caching dns servers that can be build with OE ? Mar 22 18:19:38 dnsmasq? Mar 22 18:21:14 * mwester is a user of dnsmasq Mar 22 18:21:34 * NAiL too Mar 22 18:22:06 3 in favour . That's good ;) Mar 22 18:25:47 03florian 07org.oe.dev * r30ed2981... 10/ (3 files in 3 dirs): gpe-session-scripts-phone: Save dbus session bus address to a file. Mar 22 18:30:52 hi Mar 22 18:38:37 hmm... anyone building angstrom for uclibc nowadays? Mar 22 18:39:10 I'm just wondering how long I can expect this to take: Mar 22 18:39:10 NOTE: package uclibc-initial-0.9.28-r9: task do_configure: started Mar 22 18:39:21 Because it's been there for 10-15 minutes now Mar 22 18:40:35 * Laibsch is back from a long telephone conversation Mar 22 18:40:53 NAiL: check does it loop or not Mar 22 18:40:54 zecke: How do I find out about the version of cpp used? Mar 22 18:41:02 Laibsch: config.log :) Mar 22 18:41:11 zecke: OK. I will check Mar 22 18:41:13 Laibsch: config.log contains the data you want Mar 22 18:41:16 NAiL: its 'make oldconfig' which can break^Wloop on broken defconfig Mar 22 18:41:24 ah.. Mar 22 18:41:48 doesn't make oldconfig generate huge logs when looping? Mar 22 18:41:57 because it wasn't eating diskspace Mar 22 18:42:05 (but I'll do it manually anyway) Mar 22 18:42:13 ah. then check log.do_configure Mar 22 18:43:34 cu - killbill2 time Mar 22 18:44:19 hrw|gone: you were right, it was stuck Mar 22 18:45:40 gconvert.c:48:2: error: #error GNU libiconv not in use but included iconv.h is from libiconv Mar 22 18:46:13 and libiconv is build Mar 22 18:58:24 Good evening Mar 22 18:59:48 'evening all Mar 22 19:00:32 Anyone know anything about this? Mar 22 19:00:33 http://linuxdevices.com/news/NS5355237317.html Mar 22 19:00:53 not me Mar 22 19:01:23 Crofton: 32 MB to 1 GB NAND flash ! Yeah ! Mar 22 19:01:25 we need some :) Mar 22 19:04:54 http://www.garz-fricke.com/render.php?sess_pid=357 Mar 22 19:04:59 German company Mar 22 19:05:43 I think linux-rp_2.6.20-r7 is broken after the sharpsl remote patch. Mar 22 19:05:52 Do you have problems building it? Mar 22 19:06:17 RP: Hello. Are you aware of any problem here ? Mar 22 19:07:23 It seems that there are some undefined references to spitzscoop2_device in sharpsl_rc_interrupt and sharpsl_rc_timer_callback, it that have meaning for anyone. Mar 22 19:27:55 does anyone know if something I can add to my .bb file to keep binaries from being stripped Mar 22 19:30:51 tried INHIBIT_PACKAGE_STRIP ? Mar 22 19:31:04 Dunno if that's what it does, it just sounds like it might work ;) Mar 22 19:47:38 NAiL: thanks I'll give it a try Mar 22 19:47:49 np Mar 22 19:48:17 Hmm, I think I've fixed it. I'll submit a bug Mar 22 19:54:32 NAiL: I think it worked! Mar 22 19:54:47 I added INHIBIT_PACKAGE_STRIP = "1" to my .bb file Mar 22 19:54:49 xuumbi: cool Mar 22 19:54:51 yeah Mar 22 19:55:14 I don't see any .debug areas created, or anything in them Mar 22 19:55:22 thanks again Mar 22 19:56:19 is anyone here working on, or worked on an embedded product using OE? Mar 22 19:56:41 do you mean actual hardware? Mar 22 19:57:17 yeah I'm just curious if anyone here is from a company that developed a product using OE Mar 22 19:57:38 cause I'm trying to make the case for OE for a product in my company Mar 22 19:57:45 only product I know that is using OE is the fic neo 1973 (openmoko) Mar 22 19:57:54 right that looks like a cool phone Mar 22 19:57:59 indeed Mar 22 19:58:21 I know cbrake has developed at least one, if not a few devices based on OE Mar 22 19:58:30 ~seen cbrake Mar 22 19:58:33 doh Mar 22 19:58:52 looks like he's not here right now Mar 22 19:58:57 cbrake was last seen on IRC in channel #elinux, 9d 3h 1m 6s ago, saying: 'CosmicPenguin: nod, I still see a lot of stuff based on them. OK, so this is pretty much just incremental stuff. I was just wondering if it was a brand new core or something, but does not look like it.'. Mar 22 19:59:21 xuumbi, we are. Mar 22 19:59:41 ~seen likewise Mar 22 19:59:44 likewise was last seen on IRC in channel #oe, 21h 48m 19s ago, saying: 'good nite'. Mar 22 19:59:56 koen|away: when you get back... I have problems getting the uclibc build to work. Mar 22 20:00:53 HopsNBarley: may I ask how your environment incorporates updates to the OE repository? Mar 22 20:01:25 HopsNBarley: I'm wondering if we should just take a snapshot in a known-good state and not pull in updates Mar 22 20:01:47 xuumbi: taking a known-good snapshot is a *good* idea. Mar 22 20:02:16 org.openembedded.dev is a development branch Mar 22 20:02:27 we've put a good number of things back to OE (apache, modphp, etc) Mar 22 20:02:40 but we do pretty much work from a "known good" snapshot. Mar 22 20:02:44 HopsNBarley: right I would like to do that too Mar 22 20:03:01 i take updates to another directory, diff them, and then use monotone pluck to move things over we need/like. Mar 22 20:03:16 i "owe" OE a bunch of updates (-; Mar 22 20:03:22 so far I've only been customizing packages, but I did create some simple .bb files to pull in some perl modules Mar 22 20:03:43 building angstrom gpe-image, bitbake fails with "QA ran found fatal errors" building librsvg: http://pastebin.ca/406916 Mar 22 20:03:44 ah very interesting, I like that idea Mar 22 20:03:45 any ideas? Mar 22 20:03:57 once I got the mechanics of keeping our own branch, it's worked out pretty smoothly, modulo the horrible fakeroot problem i'm stricken with... Mar 22 20:04:24 so you have your own monotone branch locally? Mar 22 20:04:30 yes. Mar 22 20:07:30 what I've done is create an area with customized packages and added it to BBFILES, and I keep that checked into our repository Mar 22 20:07:32 mreimer_: feel free to fix them :) Mar 22 20:10:29 zecke: ok. where do I start? I'm not sure what the real error is. Is it the "non debug package contains .debug directory", or the cryptic "Exception:exceptions.SystemExit Message:1"? Mar 22 20:10:48 mreimer_: ERROR: QA Issue Mar 22 20:11:00 mreimer_: in line 6 and 9, www.openembedded.org title page describe the issue Mar 22 20:11:10 thanks zecke Mar 22 20:11:16 JustinP: May I ask about the purpose of http://www.openembedded.org/viewmtn/diff.psp?id1=58920574db79c241a2670c25ee28a5be67312de9&id2=45ed69e5efc5ba31b80dc1e00747aeb58424e6c1 Mar 22 20:11:26 That version was being built for me without that setting. Mar 22 20:12:34 mreimer_: normally adjusting FILES or changing the order of -dev and -dbg package can fix it Mar 22 20:13:12 mreimer_: Hello! Any progress with your w3220 ? Mar 22 20:16:31 sirfred: hi! no, I haven't had time, sadly. I'm currently building angstrom gpe-image for my job, so I hope to build XImageon soon and give it a try Mar 22 20:16:53 mreimer_: I released a new version some days ago. Mar 22 20:17:13 sirfred: I did get as far as dumping the registers and making a list of registers with non-zero values that weren't documented, but no further Mar 22 20:17:23 sirfred: I'll check it out. Mar 22 20:17:26 mreimer_: Well,that's something Mar 22 20:17:55 sirfred: I read that it talks to the fb driver now for config--good job Mar 22 20:18:00 mreimer_: Have you tried to ask for specs to ATI (AMD) Mar 22 20:18:20 sirfred: no, I haven't Mar 22 20:18:39 mreimer_: Well, I suspect I've messed up randr orientations. Didn't have time to check it further, but at least, it now support 320x240, and that's great for emulators Mar 22 20:18:58 mreimer_: Perhaps they have changed their policy about that. Mar 22 20:19:06 mreimer_: Or perhaps they are tired of saying no. ;-) Mar 22 20:19:10 sirfred: very good. that could be useful for watching videos without scaling Mar 22 20:19:34 sirfred: I'll email AMD about that when I get a chance. It would be awesome to get specs. Mar 22 20:19:38 bbiab Mar 22 20:19:43