**** BEGIN LOGGING AT Mon Oct 22 02:59:56 2007 Oct 22 04:07:51 morning Oct 22 04:19:45 Greetings Oct 22 04:20:39 morning Oct 22 04:28:58 03mickeyl 07org.oe.dev * r3cb56853... 10/ (3 files in 2 dirs): fic-gta01/stereoout.state: turn off (useless) 3d booster Oct 22 04:29:05 03mickeyl 07org.oe.dev * r56f71aa4... 10/ (6 files in 3 dirs): linux-openmoko 2.6.22.5 use linux.inc Oct 22 06:31:07 good morning Oct 22 07:06:20 good morning all Oct 22 07:08:02 good morning all Oct 22 07:20:43 RP: thanks for fixing bitbake Oct 22 07:28:48 03koen 07org.oe.dev * r1e7260f2... 10/ (1 packages/openobex/openobex_1.3.bb): openobex: use autotools_stage_all, that stages more than we currently do Oct 22 07:43:01 koen: that orgasmic response on the oe list about cleaning up the GettingStarted page was... well, it frigtened me Oct 22 07:43:26 heh Oct 22 07:44:55 it wasn't intended as such Oct 22 07:49:32 mickey-tw: any idea why I get http://scap.linuxtogo.org/files/ecd25929c1aeeec1610ed5a9fe15f302.png and abraxa gets http://scap.linuxtogo.org/files/5cab8cd5467c79dfecf5ac48142419cd.png ? Oct 22 07:54:32 Hello Oct 22 07:57:37 hmm, koen been washing his hair with herbal essences :-) Oct 22 08:03:06 morning all Oct 22 08:03:26 koen: Does the PREFERRED_VERSION ++ problem still exist? Oct 22 08:06:29 Dammit Oct 22 08:06:36 glibc-intermediate fails to build Oct 22 08:06:45 /usr/local/oe/tmp/work/sh3-linux/glibc-intermediate-2.6.1-r1/build-sh3-linux/libc.so.6: undefined reference to `__libc_use_alloca' Oct 22 08:07:02 Is anyone else getting this? Oct 22 08:07:34 I grepped the file (libc.so.6) and it does have that symbol Oct 22 08:08:25 Does it matter much if glibc-intermediate and glibc are different versions? Oct 22 08:08:26 B_Lizzard: Just because it has a reference to that symbol, doesn't mean it contains it... Oct 22 08:08:33 Aha... Oct 22 08:08:56 Searching in google doesn't provide anything meaningful Oct 22 08:09:17 Um, glibc-intermediate 2.6.1 that is Oct 22 08:09:23 2.5 builds great Oct 22 08:09:40 Actually, glibc-initial is also at 2.6.1 and builds OK Oct 22 08:10:07 RP: the '+' thing was a flaw in the sed/awk script Oct 22 08:11:47 koen: ok, great :) Oct 22 08:18:53 One question - is it really important that glibc-intermediate and glibc are the same version? Oct 22 08:19:18 As I understand, glibc-intermediate is the glibc that is used to build glibc :) Oct 22 08:20:07 no, its used to build the compiler that builds glibc Oct 22 08:20:11 aha Oct 22 08:20:52 B_Lizzard: and this approach is needed for NTPL support in glibc... (AFAIK) Oct 22 08:22:35 I see. Oct 22 08:24:57 koen: strange. that can only mean we´re still not using the latest theme svnrev. you sure it builds it? Oct 22 08:24:58 zecke: does qemu build on mac? Oct 22 08:25:11 koen: i have seen bitbake svnrev malfunctioning recently Oct 22 08:25:17 claiming it builds rfoo Oct 22 08:25:27 but when you remove the sources, you see it downloads rev bar Oct 22 08:25:36 with bar less than foo Oct 22 08:25:47 no idea where bar comes from Oct 22 08:26:07 openmoko-theme-standard2 - 0.1.0+svnr20070823-r2 - Standard Gtk+ theme for the OpenMoko distribution Oct 22 08:26:12 ljp: not the neo one Oct 22 08:26:20 * koen spots the problem Oct 22 08:26:47 morning all Oct 22 08:29:30 Well, found the problem Oct 22 08:29:42 It's SuperH-specific or something like that Oct 22 08:29:55 Thank you for your help, dudes :) Oct 22 08:38:31 03koen 07org.oe.dev * r36403374... 10/ (23 files in 2 dirs): openmoko2: SVNREV -> rSRCREV Oct 22 08:38:35 03koen 07org.oe.dev * r60bf9410... 10/ (4 files in 2 dirs): openmoko-theme, openmoko-terminal: repair PV damage by cheating a bit an bumping the micro version instead of PE Oct 22 08:38:55 http://www.linuxdevices.com/news/NS6964769377.html Oct 22 08:39:00 morning Oct 22 08:41:23 hey hrw Oct 22 08:42:56 koen: oooh Oct 22 08:43:17 well spotted Oct 22 08:45:56 mickey-tw: only eds-dbus needs to get SRCREV'ed Oct 22 08:46:13 but I'm a bit too lazy to match SRCDATE to SRCREV before breakfast Oct 22 08:46:38 Not at all related, but are the Nokia Internet Tablets supported by Angstrom? Oct 22 08:46:45 As in userspace and kernel. Oct 22 08:47:14 "it boots" is probably a correct summary Oct 22 08:47:33 hehehe Oct 22 08:47:44 it´s a bit of a pain with all the binary only stuff to make it work really well Oct 22 08:48:13 we can build a working image with our own kernel *if* you have the binary module for wifi Oct 22 08:48:17 Well, I'm guessing that most hardware works with open-source stuff. Oct 22 08:48:21 Aha... Oct 22 08:48:27 but we can;t distribute such an image :( Oct 22 08:49:02 Well, you could distribute an incomplete image and support scripts Oct 22 08:49:07 Install scripts Oct 22 08:49:11 (too much work) Oct 22 08:49:45 well, post-install script should be a doable Oct 22 08:49:55 just cumbersome Oct 22 08:50:06 "please insert an sd card with all binary crap now" Oct 22 08:50:10 "copying..." Oct 22 08:50:13 "rebooting" Oct 22 08:50:41 Yeah, something like that. Oct 22 08:50:44 mickey-tw: You could do it but you need binutils on device Oct 22 08:51:18 mickey-tw: http://scap.linuxtogo.org/files/d390f9f2e290760c7969b9155c167cc7.png Oct 22 08:51:42 RP: are you ok with adding the vmlinux package to the blacklist in kernel.bbclass? Oct 22 08:52:15 koen: that´s better :) Oct 22 08:52:29 (I noticed a 2.6MB file in the /boot dir of my images called vmlinux-something) Oct 22 08:52:49 Apparently from the kernel-vmlinux package Oct 22 08:52:56 koen: yes Oct 22 08:53:05 B_Lizzard: see what I just said about blacklisting :) Oct 22 08:53:14 Yeah :) Oct 22 08:53:20 koen: Thats to stop weird dependencies appearing? Oct 22 08:53:25 yes Oct 22 08:53:27 http://bugs.openembedded.org/show_bug.cgi?id=3165 Oct 22 08:53:43 koen: Those dependencies are unintentional... Oct 22 08:58:57 mickey-tw: I noticed that holger moved to taiwan and is working on gsmd ;) Oct 22 08:59:18 holger? Oct 22 08:59:48 I got and email for 'tick' Oct 22 08:59:59 heh Oct 22 09:00:04 right, but that´s not really holger Oct 22 09:05:07 hi Oct 22 09:06:20 Linux Nokia-N810-41-15 2.6.21-omap1 #2 Mon Oct 15 16:25:41 EEST 2007 armv6l unknown Oct 22 09:06:33 hopefully nokia updates the nxxx to .21 as well Oct 22 09:20:07 Is the N810 any good then? Oct 22 09:22:17 * Bernardo needs to choose a machine to develop for... Maybe the N810... Oct 22 09:22:41 B_Lizzard: the 810 hasn't been released yet :) Oct 22 09:22:44 hey lrg Oct 22 09:22:51 hehehehehe Oct 22 09:22:51 :D Oct 22 09:23:19 hey koen Oct 22 09:24:39 cbrake_away: let me know when you are back Oct 22 09:27:05 any ideas on what machine to choose that has GPS, GSM/3G, and audio? Ah, and the possibility to connect a external antenna. Oct 22 09:27:40 I need to start developing a project in Dec/Jan, and would like something that is supported in OE Oct 22 09:28:27 gps + gsm/2.5g + audio can be done Oct 22 09:30:03 that would be good enough Oct 22 09:30:12 3g isn't a requirement Oct 22 09:30:17 as well as wifi Oct 22 09:30:34 keep your fingers crossed that the fic-gta02 is ready soon :) Oct 22 09:30:41 those are only extras Oct 22 09:30:50 yep, I was thinking on that one Oct 22 09:31:01 but I do need a working gps... :( Oct 22 09:31:13 gta02 has a working gps (atmel based) Oct 22 09:31:20 cool Oct 22 09:31:32 that would solve my problems Oct 22 09:31:43 time to set up a openmoko devel system here, then Oct 22 09:32:18 do you know if the physical specs (dimensions, how to unassemble, etc.) of the gta02 are already available? Oct 22 09:32:36 should be the same as gta01 Oct 22 09:32:52 same case, slightly different PCB Oct 22 09:32:52 One of the things I'll need will be to replace the plastic shell with a steel case Oct 22 09:33:15 hence the external antenna :) Oct 22 09:33:22 exactly Oct 22 09:33:25 :) Oct 22 09:34:37 the device will be part of a movable structure, so the gps and a way to communicate are absolute needs Oct 22 09:40:50 XorA: any updates on Xglamo since OEdem? Oct 22 09:41:04 koen: no Oct 22 09:41:09 ah, too bad Oct 22 09:42:34 03ifaistos 07org.oe.dev * red0c9499... 10/ (1 packages/linux/linux/x86/defconfig): packages/linux/linux/x86/defconfig : add more drivers as modules Oct 22 09:42:38 03ifaistos 07org.oe.dev * r3cad99e0... 10/ (1 conf/machine/i586-generic.conf): conf/machine/i586-generic.conf : Add xserver-kdrive-vesa as the default xserver Oct 22 09:42:42 03ifaistos 07org.oe.dev * r33e66f42... 10/ (1 conf/machine/i686-generic.conf): conf/machine/i686-generic.conf : Add xserver-kdrive-vesa as the default xserver Oct 22 09:42:47 03koen 07org.oe.dev * r95220d9f... 10/ (1 conf/distro/include/sane-srcrevs.inc): sane-srcrevs: bump openmoko-appearance for .desktop, calculator for icon and mediaplayer for video support Oct 22 10:13:50 good morning Oct 22 10:19:04 hey florian Oct 22 10:43:12 hey florian Oct 22 10:43:28 koen: board shipped - should get to you buy Wed - Thu Oct 22 10:43:33 by Oct 22 10:43:47 * lrg cant speel ttoday Oct 22 10:53:57 can someone tell me how to include kernel modules in images ? cbrake told me something about it the othery day but unfortunately I don't recall... Oct 22 10:55:01 check fic-gta01.conf Oct 22 10:55:05 (or any other) Oct 22 10:55:37 gilligan_: IMAGE_INSTALL += "linux-modules" iirc Oct 22 10:57:30 nud: so that would install all modules then? Oct 22 10:58:26 gilligan_: yes Oct 22 10:58:56 nud: ah okay thanks.. i'll try that :) Oct 22 11:00:18 hm.. nope Oct 22 11:00:58 seems like there is no 'linux-modules' Oct 22 11:01:04 gilligan_: my bad Oct 22 11:01:07 kernel-modules Oct 22 11:01:39 ah yeah.. just found it while grepping for modules in packages/images Oct 22 11:11:52 make pdf Oct 22 11:11:53 ops Oct 22 11:12:39 Makefile not found Oct 22 11:12:50 hrw :) Oct 22 11:14:39 * koen is no longer an idiot Oct 22 11:19:27 mickey-tw: what's the canonical place to get libjana? just add a do_stage to openmoko-dates2? Oct 22 11:21:16 koen: let me check whether they are using an svn external or not Oct 22 11:23:44 koen: "ris Oct 22 11:23:44 koen: "12:44 Oct 22 11:23:45 koen: ah, dates2 is in their repo anyways. can´t check then. probably just start with do_stage Oct 22 11:23:46 » thomas: the world clock app won't build in oe until I sort out the staging function for the dates .bb either - which I'll do after I auto-tool it" Oct 22 11:23:59 mickey-tw: if dates2 builds a libjana.a we're screwed Oct 22 11:24:00 koen: that was chris Oct 22 11:24:36 since statically linking doesn't work with multimachine as RP and I pointed out in the ml last week Oct 22 11:29:08 drat Oct 22 11:29:40 hrw: tell chris to build a dynamic libjana*, otherwise it will break multimachine Oct 22 11:37:37 RP: I'm going to test http://rafb.net/p/sdlRT254.html Oct 22 11:38:08 ~ping crofton Oct 22 11:38:08 pong crofton Oct 22 11:38:46 pong Oct 22 11:41:25 hey Crofton|home Oct 22 11:41:47 * Crofton|home has been in NAND flash for DaVinci world Oct 22 11:42:00 hey mallum Oct 22 11:42:12 koen: it will be dynamic - I just told chris why it has to be Oct 22 11:42:26 hey koen Oct 22 11:42:30 and in paralllel I'll try to fix staging :) Oct 22 11:42:36 and yesterday we made a hole bigger and put a pipe in it :) Oct 22 11:42:38 http://www.flickr.com/photos/32615155@N00/ Oct 22 11:43:35 mickey-tw: did the meeting-of-doom finish already? Oct 22 11:43:59 koen: only pre-hell this time, since two important people are missing. real doom probably tomorrow Oct 22 11:44:12 (head of marketing and CEO are sick today) Oct 22 11:46:12 Crofton|home: what is that pipe supposed to do? Oct 22 11:46:26 http://www.flickr.com/photos/32615155@N00/1681415273/ Oct 22 11:46:33 koen: That patch will explode badly :-( Oct 22 11:46:35 stop dirt going in this pipe Oct 22 11:46:38 http://www.flickr.com/photos/32615155@N00/1681415273/ Oct 22 11:47:00 RP: yes, but I want to see where it explodes Oct 22 11:47:03 koen: This about where PACKAGE_ARCH="all" should stage too and how that staging should merge with one where PACKAGE_ARCH="spitz" Oct 22 11:47:07 that pipe goes into a cave, some people dug the entrance in 1948 Oct 22 11:47:41 RP: see the footnote in http://lists.linuxtogo.org/pipermail/openembedded-devel/2007-October/003312.html :) Oct 22 11:48:00 koen: I've given this a bit of thought and the only way it can be fixed is to make the populate_staging task machine specific Oct 22 11:48:24 anyone seen my bug report on building binutils 2.18 / uclibc ? fails to link.. is that something that has been known before ? Oct 22 11:49:28 RP: or by adjusting [C,LD]FLAGS in machine specific packages Oct 22 11:50:01 do we have any 'all' packages that need cross libs? Oct 22 11:52:00 koen: No, but if we're going to fix this, might as well fix it properly... Oct 22 11:52:32 I don't want to loose multimachine, but we would get that back with packaged-staging NG Oct 22 11:52:45 koen: Adjusting C/LD flags is insane given the permutations potentialy involved. Might as well forget changing the staging layout too Oct 22 12:06:04 * * OE Bug 3191 has been created by autobuild(AT)openembedded.org Oct 22 12:06:06 * * openmoko-dates2-0.1.0+svnr640-r0-do_populate_staging Oct 22 12:06:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3191 Oct 22 12:07:56 http://www.youtube.com/watch?v=cjshyT4cZpI Oct 22 12:09:07 hmm Oct 22 12:09:18 isn´t gst-plugins-ugly supposed to put out mpeg2dec plugin? Oct 22 12:09:35 isn't mpeg2 decoding in -good? Oct 22 12:09:52 not according to http://webcvs.freedesktop.org/gstreamer/gst-plugins-ugly/ext/mpeg2dec/ Oct 22 12:10:10 http://www.angstrom-distribution.org/repo/?action=details&pnm=gst-plugin-mpegvideoparse Oct 22 12:10:17 it's in -ugly Oct 22 12:10:32 mpegvideoparse -- is that mpeg2decß Oct 22 12:10:33 ? Oct 22 12:10:40 ehm Oct 22 12:10:43 not sure Oct 22 12:10:46 hmm Oct 22 12:10:52 * mickey-tw no clue bout video Oct 22 12:10:55 XorA: ? Oct 22 12:11:09 in doubt use gst-plugin-ffmpeg :) Oct 22 12:11:25 does that work for you? Oct 22 12:11:27 mpeg2dec: mpeg2dec: mpeg1 and mpeg2 video decoder Oct 22 12:11:27 mpegvideoparse: mpegvideoparse: MPEG video elementary stream parser Oct 22 12:11:31 aha, that´s from my desktop Oct 22 12:11:43 so, checking why does mpeg2dec not build for us Oct 22 12:14:30 http://pastebin.com/d3fe2e15b <-- hm, I am doing something wrong with a /mini/ package here, perhaps someone can help me out. Something is wrong with the paths. It tries to copy to a wrong image/rootfs directory apparently Oct 22 12:15:21 it just copies a single file to ${D}/sbin -- but i suppose it is missing something Oct 22 12:17:49 should probably use the ${sbindir} variable, but don't think that alone would be the issue.. Oct 22 12:21:05 mickey-tw: could you do me a fav and have a short glimpse ? Oct 22 12:21:59 mickey-tw: all mpeg* plugins are in ugly Oct 22 12:25:41 03rpurdie * r979 10/ (4 files in 4 dirs): data.py: Fix unexport handling (#3135) Oct 22 12:26:56 Henryk: did you see http://article.gmane.org/gmane.comp.version-control.monotone.devel/12807 ? Oct 22 12:28:29 who use busybox 1.7.2? Oct 22 12:28:46 I did till I overfilled my flash on the hx Oct 22 12:28:55 hrw: cbrake and pH5 use it Oct 22 12:29:33 hrw: it's the default in angstrom 2008 Oct 22 12:31:05 koen: what is a roster in monotone? Oct 22 12:31:36 Henryk: something they introduced in .26 to make things 10 times faster Oct 22 12:31:42 i *did* notice that in the profile run most cycles were in roster.cc Oct 22 12:31:42 (that's all I know) Oct 22 12:35:37 Henryk: Oct 22 12:35:40 Henryk: http://lists.gnu.org/archive/html/monotone-devel/2005-11/msg00102.html Oct 22 12:37:00 koen: thanks Oct 22 12:37:00 "A roster represents a tree state, like a manifest does; but there's extra stuff added so that it unambiguously records _all_ the Oct 22 12:37:01 information that is needed to reconstruct a revision. (You can't do this with normal manifests, because if you look at two plain Oct 22 12:37:01 manifests you can't figure out whether any renames happened, forinstance.) " Oct 22 13:06:04 * * OE Bug 3191 has been RESOLVED (INVALID) by Oct 22 13:06:06 * * openmoko-dates2-0.1.0+svnr640-r0-do_populate_staging Oct 22 13:06:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3191 Oct 22 13:27:21 morning Oct 22 13:28:09 hey chouimat Oct 22 13:34:20 mickey|away: DEPENDS = "gstreamer libmikmod libmad liboil libogg tremor libvorbis libid3tag esound" Oct 22 13:34:26 mickey|away: no mpeg* in there Oct 22 13:37:16 XorA: openmoko-worldclock fails to build Oct 22 13:37:50 koen http://www.youtube.com/watch?v=1qiGyxPplAw Oct 22 13:38:13 chouimat: hrw showed that earlier on :) Oct 22 13:39:10 at least my piggies are too smal to pick up a baseball bat :) Oct 22 13:44:06 03xora 07org.oe.dev * rdac5061a... 10/ (6 files in 3 dirs): Oct 22 13:44:06 packages/openmoko2/openmoko-worldclock2_svn.bb : add new recipe and Oct 22 13:44:06 fix up revs stuff. Oct 22 13:44:11 03koen 07org.oe.dev * r1331abd4... 10/ (1 packages/openmoko2/openmoko-worldclock2_svn.bb): Oct 22 13:44:11 openmoko-worldclock2: remove SVNREV Oct 22 13:44:11 * Please stop adding recipes with SVNREV Oct 22 13:44:36 XorA: wordclock needs: Oct 22 13:44:38 -SRCREV_pn-openmoko-dates2 ?= "640" Oct 22 13:44:39 +SRCREV_pn-openmoko-dates2 ?= "652" Oct 22 13:44:52 (and dates2 as well, otherwise staging fails) Oct 22 13:45:46 * koen hates studying for exams Oct 22 13:47:25 * Bernardo remembers fondly studying... Probably too many years have passed since Oct 22 13:49:04 * * OE Bug 3192 has been created by tbayne(AT)tbayne.net Oct 22 13:49:06 * * Linksys WCF-11 Card works on OZ Opie, but not Angstrom Oct 22 13:49:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3192 Oct 22 13:49:16 * * OE Bug 3193 has been created by joaquinduran(AT)adtelecom.es Oct 22 13:49:18 * * source code of screeensaver unable to be downloaded Oct 22 13:49:20 * * http://bugs.openembedded.org/show_bug.cgi?id=3193 Oct 22 13:49:30 RP: time for bitbake 1.8.10? Oct 22 14:05:04 * * OE Bug 3191 has been REOPENED by Oct 22 14:05:06 * * openmoko-dates2-0.1.0+svnr640-r0-do_populate_staging Oct 22 14:05:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3191 Oct 22 14:09:38 koen: Probably soon, yes Oct 22 14:09:54 ah, about the bitbake release :) Oct 22 14:09:56 koen: There are a few other gremlins in the bugtracker that should be looked at first Oct 22 14:10:56 * koen switches to irssi Oct 22 14:11:10 there, some more ram freed Oct 22 14:15:26 cbrake: could you change `CMDLINE_cm-x270 =` to a `?=` allowing for cmdline designation in local.conf ? Oct 22 14:28:06 gilligan_: will "CMDLINE_cm-x270 ?=" take precendence over a previous "CMDLINE =" assignment? If you've tested this and it works, it is probably fine. Oct 22 14:30:21 cbrake: i was hoping you would know that for sure ;] i thought ?= would assign if not previously assigned in some global config (such as local.conf) Oct 22 14:31:09 cbrake: i'll try that later Oct 22 14:31:25 gilligan_: the only problem I can see if CMDLINE is assigned somewhere, and then CMDLINE_cm-x270 ?= won't override it Oct 22 14:31:37 gilligan_: if you can test it, then I'll add it Oct 22 14:31:45 ok Oct 22 14:32:00 gilligan_: I've never tried to build binutils in a uclibc image Oct 22 14:32:02 03xora 07org.oe.dev * ra80845bd... 10/ (1 conf/distro/include/sane-srcrevs.inc): conf/distro/include/sane-srcrevs.inc : increment dates2 Oct 22 14:32:25 cbrake: ah ok.. i wanted to in order to get rid of the alignment issue Oct 22 14:33:02 gilligan_: how do you fix alignment issues on the CMDLINE? Oct 22 14:33:23 gilligan_: doh, never mind Oct 22 14:33:27 cbrake: no hehe.. you got me wrong :) Oct 22 14:33:54 ~ping ade|desk Oct 22 14:33:55 pong ade|desk Oct 22 14:34:34 XorA, hrw: should we make libjana* seperate packages? Oct 22 14:35:44 koen|away: forwarded it to chris Oct 22 14:35:52 cbrake: as for the cmdline i am just trying to avoid having to patch around in packages or building exact copies with minor changes (such as just wanting to change the rdinit argument) Oct 22 14:35:56 http://www.busybox.net/lists/busybox/2007-September/028838.html needs to be applied to 1.7.2 busybox Oct 22 14:36:04 * * OE Bug 3191 has been RESOLVED (FIXED) by dp(AT)xora.org.uk Oct 22 14:36:06 * * openmoko-dates2-0.1.0+svnr640-r0-do_populate_staging Oct 22 14:36:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3191 Oct 22 14:36:23 gilligan_: agreed Oct 22 14:39:44 hrw: Things can just depend on openmoko-dates for now Oct 22 14:39:44 16:37 Oct 22 14:39:46 » hrw: If OE or poky wants it to be packaged separately, it could be Oct 22 14:39:54 koen|away: thats from chris Oct 22 14:39:57 * hrw -> afk Oct 22 14:40:19 cbrake: okay.. and now I think I will go nuts... :) changing CMDLINE_cm-x270 in packages/linux/2.6.23.bb and a `bitbake virtual/kernel -crebuild` and (a make image..) should suffice to change the commandline I suppose.. but it doesn't for some reason ;] Oct 22 14:55:46 cbrake: in armmon you might have a command line saved Oct 22 14:55:49 euh Oct 22 14:55:50 gilligan_: Oct 22 14:56:12 so it overrides the default one you provide in the bitbake file Oct 22 14:57:28 stefan_schmidt: The openmoko problem was different this time. If you can give bugzilla another shot, it will hopefully work now. Oct 22 14:57:28 that's a fix, but not a solution ;] Oct 22 14:57:52 gilligan_: I'll try it now Oct 22 14:57:57 koen|away: so go for it (libjana) Oct 22 14:58:02 gilligan_: I meant, perhaps there is already one saved Oct 22 14:58:51 stefan_schmidt: okay, cool Oct 22 14:59:58 cbrake: http://bec-systems.com/web/content/view/79/9/ forgets to mention that DISTRO_* also needs a rebuild of task-base Oct 22 15:00:07 hrw: I don't have build tree atm Oct 22 15:01:00 gilligan_: Test account requested. Oct 22 15:01:19 Cool, that looked like it went through Oct 22 15:02:10 stefan_schmidt: hm? Oct 22 15:02:23 koen|away: it is in the list of bullet points in about the middle of the article Oct 22 15:02:41 gilligan_: Got it. :) Oct 22 15:02:51 gilligan_: Thanks. What was it now? Oct 22 15:03:21 gilligan_: I think he's typing your nick and meaning me. :) Oct 22 15:03:41 ggilbert: i am starting to think so as well, because i have no idea what he is talking about now :) Oct 22 15:03:44 cbrake: ah right, sorry Oct 22 15:03:53 grr, of course. Sorry Oct 22 15:04:01 It's already 23:00 in .tw Oct 22 15:04:03 stefan_schmidt: The bugzilla-daemon alias had disappeared out of my aliases file. That's back, so now openmoko's spam check passes. Oct 22 15:04:29 ggilbert: ok, nice to see it work. Oct 22 15:04:33 yo ggilbert Oct 22 15:04:41 Yo mickey Oct 22 15:05:26 why is x11-image building bluez-gnome? Oct 22 15:07:52 zecke: ./packages/angstrom/angstrom-gpe-task-base.bb:37: ${@base_contains("COMBINED_FEATURES", "bluetooth", "bluez-gnome", "",d)} \ Oct 22 15:08:15 03koen 07org.oe.dev * r9bf73e3c... 10/ (1 packages/busybox/busybox-1.7.2/run_parts.c): busybox 1.7.2: add run_parts.c so Marcin can add it into the recipe :) Oct 22 15:08:17 also what the heck is bluez-gnome anyway? :) /me googles Oct 22 15:08:41 ah, gnome bluetooth support apps Oct 22 15:08:41 koen|away: I'm trying xterminal-image now, but it is kind of odd to get GPE+GNOME when asking for x11 ;) Oct 22 15:08:44 gilligan_: a gtk+ frontend for bluez Oct 22 15:08:54 zecke: bluez-gnome != gnome Oct 22 15:09:23 "gnome" in "bluez-gnome" anyway, thanks for your response Oct 22 15:09:40 zecke: have you actually checked the recipe? Oct 22 15:09:46 it depends on nothing gnome Oct 22 15:09:59 even HAL is optional Oct 22 15:10:15 no, have not checked because my remote connection is SLOOOW :) Oct 22 15:10:49 an OE can't generate a working non-root x11 based image without gpe at the moment Oct 22 15:11:00 feel free to fix that Oct 22 15:11:05 i see Oct 22 15:11:15 instead of bitching about how 'x11' includes gpe and gnome Oct 22 15:11:36 koen|away: I just asked. Oct 22 15:11:40 seems you drank the qtopia kool-aid already Oct 22 15:12:19 koen|away: hehe, not yet I'm just preparing my ELC presentation Oct 22 15:13:48 koen|away: have you witnessed an do_install error with bluez-gnome yet? Oct 22 15:15:14 gilligan_: does the microphone work ok with the latest linux / alsa on cm-x270N Oct 22 15:15:15 ? Oct 22 15:16:13 zecke: not yet (and the autobuilder hasn't either :) ) Oct 22 15:16:14 nud: works for me Oct 22 15:16:30 koen|away: make[3]: execvp: /home/zecke/openembedded/build/angstrom-glibc/tmp/work/armv4t-angstrom-linux-gnueabi/bluez-gnome-0.14-r1/image/usr/share/applications: Permission denied Oct 22 15:16:32 cbrake: using snd-pxa2xx-ac97 ? Oct 22 15:16:35 i don't have the slightest idea because the microphone is really the last thing I will be needing ;] Oct 22 15:16:38 nud: yes Oct 22 15:16:42 cbrake: what version of libalsa ? Oct 22 15:17:03 zecke: mount -o user? Oct 22 15:17:18 zecke pasted "bluez-gnone do_install error" at http://paste.lisp.org/display/49626 Oct 22 15:17:30 * nud considers either downgrading to september's state or upgrading to the top notch distro Oct 22 15:17:48 nud: libasound 1.0.14-r2 Oct 22 15:18:12 ok maybe the bug was fixed that avoided the mic to work Oct 22 15:18:14 koen|away: no, perm's on this dir look sane as well Oct 22 15:18:19 cbrake: no other regressions noticed ? Oct 22 15:18:20 nud: record was broke for me a month or two back Oct 22 15:18:35 nud: no, the HEAD as of last week is working very well Oct 22 15:18:44 a month ago I blacklisted the latest libalsa and it worked with 1.0.13 Oct 22 15:19:01 cbrake: I guess you didn't try V4L tough :p Oct 22 15:19:10 my latest update broke V4L2 with pwc Oct 22 15:19:21 nud: nope :-) Oct 22 15:19:30 this reminds me I have a zillion bb files to push upstream Oct 22 15:22:37 bye Oct 22 15:23:00 steliosk: you ping'ed ? Oct 22 15:23:14 * nud considers using a different mtn checkout for each of his builds Oct 22 15:23:36 cbrake: `exec switch_root -c /dev/console /mnt/nand /sbin/init 5` <-- you've got something like this in your custom init script I assume ? i get 'switch_root: not rootfs` for some reason Oct 22 15:25:16 does OE support arm7? Oct 22 15:25:41 gilligan_: yes, looks similiar to what I have Oct 22 15:25:55 jsheldon: what kind of arm7? Oct 22 15:26:21 oe basically supports anything gcc supports Oct 22 15:26:23 cbrake: do you know how to override the defconfig cleanly ? Oct 22 15:26:52 ie without copying the package/linux directory nor overriding the defconfig file in the upstream dir Oct 22 15:27:28 koen: I believe it's the ARM7TDMI-S Oct 22 15:29:16 nud: no, I just copy the entire recipe. I need to find something better though. Maybe I'll learn how to use MTN branches better. Oct 22 15:29:40 koen: bb 1.7.2 in Poky has run-parts applied Oct 22 15:29:40 copying the entire recipe is ... beurk Oct 22 15:29:50 koen: feel free to copy - I do not have time today for it Oct 22 15:29:59 bbcollection should allow a nicer way to do htat Oct 22 15:30:11 nud: I also think it would be nice if bitbake could find FILES outside the recipe tree Oct 22 15:30:46 nud: so you could localize customized FILES for a number of recipies in one place -- but, I've not thought though this idea yet. Oct 22 15:31:59 cbrake: imho it should try to get the file in both the overlay and the upstream tree Oct 22 15:32:32 nud: that would work Oct 22 15:33:03 so it would try $overlay/linux-2.23.X/defconfig et $upstream/linux-2.23.x/defconfig if the former was not found Oct 22 15:34:15 * nud is so stupid he can't find where to register to the OE mailing-list Oct 22 15:34:28 heh Oct 22 15:35:10 nud: http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel Oct 22 15:35:36 mickey-tw: gst* doens't have any mpeg* in DEPENDS Oct 22 15:36:11 yep Oct 22 15:36:13 seen it Oct 22 15:36:14 fixing atm. Oct 22 15:36:21 together with adding some more missing deps Oct 22 15:37:01 mickey-tw: thanks Oct 22 15:37:22 np Oct 22 15:38:39 cbrake: do you do any other preparations apart from mounting the nand somewhere before you execute switch_root ? I always get "not rootfs" Oct 22 15:40:06 gilligan_: no, not that I know of Oct 22 15:43:14 RP: howdy Oct 22 15:45:04 03koen 07org.oe.dev * r70c98eca... 10/ (1 packages/busybox/busybox_1.7.2.bb): busybox 1.7.2: replace run_part.c with the one we snatched of the mailinglist Oct 22 15:45:06 please, could someone tell me how I can force a package version? Oct 22 15:48:36 look into PREFERRED_VERSION_ Oct 22 15:50:03 well last time I tried it didn't prevent bitbake to build the latest version Oct 22 15:50:29 cbrake: switch_root checks for the existance of a file '/init' (in the current tmpfs) and fails if there is none.. so that is probably why it fails.. testing now Oct 22 15:56:48 03tmbinc 07org.oe.dreambox * r3bd73aaf... 10/ (7 files in 4 dirs): enigma: update Oct 22 16:09:04 * * OE Bug 3194 has been created by joshua(AT)willowisp.net Oct 22 16:09:06 * * 2007-10-20 'bitbake gpe-image' failed on several missing DEPENDS Oct 22 16:09:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3194 Oct 22 16:11:00 hm.. there was some variable to specify global gcc compile flags if I remember correctly? Or maybe there is a specfic varaible to enable debug symbols (-g) ? Oct 22 16:14:01 gilligan_: see TARGET_ vars in bitbake.conf Oct 22 16:14:11 thanks Oct 22 16:17:43 gilligan_: might be as easy as setting DEBUG_BUILD in the recipe you want to debug, and installing the dbg package -- see local.conf.sample Oct 22 16:18:57 cbrake: ah.. that would be nice.. can't say I fancy the idea of patching around in bitbake.conf Oct 22 16:19:33 gilligan_: yeah, any global variable (*.conf) can be overridden in a recipie Oct 22 16:22:59 "Also note that OE now produces -dbg packages which contain debugging symbols" so it always builds packages with debug information as well.. and when DEBUG_BUILD is set it actually installs those isntead of the stripped ones I guess Oct 22 16:24:12 why the hell is openldap compiled for console-image ? Oct 22 16:25:01 cbrake: also the problem with switch_root was indeed the missing /init file. Do you think that's maybe worth mentioning on your page about setting up the compulab board ? angstrom doesn't install an init script to /init per default so some might fall into this "trap" Oct 22 16:27:34 mornin Oct 22 16:29:06 does openldap build for you ? Oct 22 16:29:26 | autoreconf: running: automake --foreign --add-missing --copy --force-missing --warnings=cross Oct 22 16:29:26 | automake: unknown warning category `cross' Oct 22 16:48:12 ouch!!! OE build dirs can get huge If you don't pat attention... Oct 22 16:48:20 s/pat/pay/ Oct 22 16:48:49 INHERIT += "rm_work" Oct 22 16:50:09 koen: I will do that next time :-) Oct 22 16:50:18 just gained 100GB of free space Oct 22 16:50:20 what i need is INHERIT += "magically_power_up_this_PC" Oct 22 16:51:06 INHERIT += "money-for-quad-core" Oct 22 16:51:20 mickey-tw: i'll try that ;) Oct 22 16:51:28 gilligan_: and with the "dont_touch_bank_account" ;) Oct 22 16:51:37 mickey-tw: my build from scratch is nearly up to gstreamer, so I'm eagerly awaiting your fixes :) Oct 22 16:51:53 $ mail myboss -s INHERIT += "money-for-quad-core" Oct 22 16:52:12 koen: everything pushed already Oct 22 16:52:19 * Bernardo is away: Ausente por agora. Oct 22 16:52:26 mtn: dropping packages/gstreamer/gst-plugins-ugly_0.10.5.bb Oct 22 16:52:26 mtn: modifying packages/gstreamer/gst-plugins-ugly_0.10.6.bb Oct 22 16:52:26 mtn: modifying packages/mpeg2dec/mpeg2dec_0.4.0b.bb Oct 22 16:52:29 * Philippe actually noticed that IO is a bigger issue Oct 22 16:52:29 yep, got it Oct 22 16:55:42 03mickeyl 07org.oe.dev * rbede59fb... 10/ (1 packages/mpeg2dec/mpeg2dec_0.4.0b.bb): mpeg2dec: fix individual packaging, reformat and simplify recipe Oct 22 16:55:47 03mickeyl 07org.oe.dev * rcebb5151... 10/ (3 files in 2 dirs): Oct 22 16:55:47 gst-plugins-ugly: DEPENDS on mpeg2dec and libsidplay Oct 22 16:55:47 TODO: add more optional libraries Oct 22 16:55:54 03mickeyl 07org.oe.dev * raf88183f... 10/ (3 files in 3 dirs): gst-plugins-ugly: add gstsid_autofoo_HACK.patch to make gst-plugins-ugly build the sid plugin Oct 22 17:32:53 03likewise 07org.oe.dev * rc6caaa43... 10/ (3 files in 3 dirs): binutils-2.18: Added powerpc e300c2c3 patch. Oct 22 17:35:14 mrxvt 0.5.3 released Oct 22 18:36:04 * * OE Bug 3195 has been created by ar(AT)oszine.de Oct 22 18:36:06 * * can't mount cf card on collie Oct 22 18:36:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3195 Oct 22 18:57:37 grrr Oct 22 18:57:38 19-Oct-2007 Netherlands Pending customs inspection Oct 22 18:57:42 lazy bastards Oct 22 19:00:47 One question - is busybox 1.7.2 borked for you too? Oct 22 19:00:55 For me it's tottally borked Oct 22 19:01:05 in general or is that in mtn now? Oct 22 19:01:21 I get kernel messages on my tty, which means that klogd isn't running Oct 22 19:01:26 Dunno Oct 22 19:01:30 It's in OE Oct 22 19:01:38 i got 1.7.2 to work locally, but i have older metadata for this project Oct 22 19:01:43 had to make that myself, heh Oct 22 19:01:47 hehehe Oct 22 19:01:47 up to date mtn or as usual a stale, month old snapshot you have on disk Oct 22 19:02:12 I just updated OE and built an image Oct 22 19:02:29 Um, actually, built a new tmp folder Oct 22 19:02:46 Everything is completely borked Oct 22 19:02:50 home sweet home Oct 22 19:02:53 evening Oct 22 19:03:16 Instead of a prompt I get "\u@\h:\w$" Oct 22 19:03:22 It doesn't even expand the names Oct 22 19:03:25 Or whatever Oct 22 19:03:26 :/ Oct 22 19:03:58 right, that means you have (as usual) old metadata Oct 22 19:04:59 :D Oct 22 19:06:07 bug #3150 was fixed over ten days ago] Oct 22 19:06:33 hmmm Oct 22 19:06:43 i just updated oe... Oct 22 19:37:04 * * OE Bug 3196 has been created by joshua(AT)willowisp.net Oct 22 19:37:06 * * gpephone-image fails on librecord due to problems in Makefile. am and configure.ac Oct 22 19:37:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3196 Oct 22 19:43:37 * koen OMGs at the tsc2003 mail in his inbox Oct 22 20:02:34 oops :) Oct 22 20:02:54 * Crofton has the impression the gumstix guys are over committed work wise Oct 22 20:03:35 I *juuuuuust* updated OE and busybox does the same Oct 22 20:03:53 I'm guessing I'll have to go back to 1.2.1 Oct 22 20:03:54 :/ Oct 22 20:17:24 B_Lizzard: ah, you aren;'t using angstrom, but some random broken distro Oct 22 20:17:33 :D:D:D:D Oct 22 20:17:33 I suggest you start using angstrom Oct 22 20:17:46 Um, I'm using our own distro :) Oct 22 20:17:49 or rip it off at least Oct 22 20:18:32 I would say something extremely distasteful and offensive, but I will not :) Oct 22 20:19:10 Also, busybox would be broken anywhere Oct 22 20:19:21 Crofton: ouch! that's got to hurt to find an error like that on a shipping product. Oct 22 20:19:27 Unless you only fix *your* bugs :) Oct 22 20:19:47 heh. it depends. if its a newly released version of busybox it will be broken. until several point releases/patches later. Oct 22 20:19:56 :) Oct 22 20:20:32 Yeah, I guess, though I wouldn't think they'd release such an *extremely* broken version Oct 22 20:20:35 yeah Oct 22 20:20:36 Marked stable Oct 22 20:20:49 1.7.2+ patches works great here Oct 22 20:20:53 So it must be me Oct 22 20:20:55 Or something Oct 22 20:21:10 no, i work out of a fork with our own separete bb 1.7.2 Oct 22 20:22:18 fork is possibly the wrong work. semi fork that lags behind dev. Oct 22 20:22:21 heh Oct 22 20:22:32 afk Oct 22 20:26:50 morning Oct 22 20:27:14 B_Lizzard: \u\h problem is with OE defconfig for busybox 1.7.2 - angstrom one is fixed Oct 22 20:27:55 Aha... Oct 22 20:28:05 koen: you should consider updating also bb 1.7.2 normal defconfig - too much distroisms we have.. Oct 22 20:29:44 Sounds like bad practice also Oct 22 20:29:52 Fixing stuff only for a select group Oct 22 20:29:58 hrw: heh Oct 22 20:30:08 hrw: I used to update global defconfigs Oct 22 20:30:14 got lots of hatemail and flames Oct 22 20:30:18 ok Oct 22 20:30:24 so nowadays I update angstrom stuff Oct 22 20:30:48 B_Lizzard, do your own config! then upstream cant break it! :) Oct 22 20:30:51 I'll run a diff between the two files Oct 22 20:30:54 :) Oct 22 20:32:15 03hrw 07org.oe.dev * r639885e0... 10/ (1 packages/linux/linux-rp_2.6.21.bb): linux-rp: make 2.6.21 default for collie Oct 22 20:32:42 hrw: but as I said at OEDEM the largest group of hatemails are people belonging to the hh.org group Oct 22 20:32:58 koen: which we now mostly ignore Oct 22 20:33:00 so it might be safe again to update the global busybox configs :) Oct 22 20:33:17 :D Oct 22 20:34:19 koen: if everyone likes what you are doing, you probably are not doing anything :-) Oct 22 20:34:46 yeah Oct 22 20:34:53 And then there's the chance someone will be annoyed that you're doing nothing. Oct 22 20:35:00 So best go fuck all. Oct 22 20:35:19 :) Oct 22 20:35:28 And do whatever you like. Oct 22 20:37:55 Was the kernel-vmlinux thing solved? Oct 22 20:38:05 I didn't see any changes in kernel.bbclass Oct 22 21:01:07 hm, now gumstix says the pullups are there Oct 22 21:01:43 the pullup is on the verdex board Oct 22 21:05:17 and the cap? Oct 22 21:05:36 clarson@foul org.openembedded.dev$ ls packages/*.bb Oct 22 21:05:36 packages/mkfontscale_1.0.3.bb packages/mkfontscale-native_1.0.3.bb Oct 22 21:05:37 * kergoth rolls eyes Oct 22 21:06:02 OK, with Angstrom defconfig things work fine Oct 22 21:06:13 But I still get the damn kernel messages on my tty Oct 22 21:06:55 kergoth: you get nominated to move those to xorg-apps :) Oct 22 21:08:53 Eh, for some reason syslogd isn't getting started Oct 22 21:11:18 /etc/init.d/syslog wants to start syslogd with -a, which isn't a valid option Oct 22 21:14:05 Oh Oct 22 21:14:16 Well, some argument that's not valid, anyways Oct 22 21:16:45 hi steliosk: I forgot to tell you... I could use icecc using a lot of machines, but I've noticed that the build is slower using icecc then without it... :( Oct 22 21:24:01 hi Oct 22 21:24:24 do someone know how I could list packages that depend on a particular package ? Oct 22 21:25:09 packages or recipes? Oct 22 21:25:20 for recipes you can use bitbake -g Oct 22 21:25:45 ~seen spyro Oct 22 21:25:49 spyro was last seen on IRC in channel #handhelds.org, 278d 23h 28m 35s ago, saying: 'hi folks'. Oct 22 21:26:34 ~curse handhelds.org for forking kernel into who-knows-will-we-ever-merge something Oct 22 21:26:35 May you be reincarnated as a Windows XP administrator, handhelds.org for forking kernel into who-knows-will-we-ever-merge something ! Oct 22 21:27:06 ~seen psokolovsky Oct 22 21:27:08 psokolovsky was last seen on IRC in channel #oe, 15d 20h 50m 15s ago, saying: 'hope it's just me...'. Oct 22 21:27:21 well, everytime someone discusses merging their account gets an 'accident' during 'routine maintenance' Oct 22 21:27:55 hopefully I do not have account there anymore - it was disabled on my request Oct 22 21:28:13 the kernel is nowadays the only reason hh.org exists Oct 22 21:28:20 once it gets merged upstream.... Oct 22 21:30:19 ~curse cvs Oct 22 21:30:20 May the fleas of a thousand camels infest your most sensitive regions, cvs ! Oct 22 21:39:32 bye Oct 22 21:41:00 03koen 07org.oe.dev * r9a0caec6... 10/ (5 files in 3 dirs): mkfontscale*: move to xorg-app Oct 22 21:44:31 is icecc better than distcc Oct 22 21:45:00 khem: yes Oct 22 21:45:21 florian: I have been using distcc so far its been ok Oct 22 21:45:29 does it have a scheduler Oct 22 21:45:39 re icecc Oct 22 21:45:40 khem: yes Oct 22 21:45:43 hmmm Oct 22 21:45:50 thats interesting Oct 22 21:46:15 so it will notice that one of the nodes has a two cpus and distribute 4 jobs to it Oct 22 21:47:22 * khem sees that icecc is not a supported fedora package Oct 22 21:47:28 it will take network latency into account as well Oct 22 21:47:51 khem: icecc is KDE ware, it changes the wire protocol every tuesday Oct 22 21:48:02 koen: haha Oct 22 21:51:52 koen: Do we use icecc on OE Oct 22 21:52:34 INHERIT += "icecc" and ping stelios if it doesn't work Oct 22 21:52:37 'night all Oct 22 21:52:42 k Oct 22 21:53:05 * * OE Bug 3194 has been RESOLVED (WORKSFORME) by philippedeswert(AT)scarlet.be Oct 22 21:53:07 * *  2007-10-20 'bitbake gpe-image' failed on several missing DEPENDS Oct 22 21:53:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3194 Oct 22 22:08:04 * * OE Bug 3197 has been created by zap(AT)homelink.ru Oct 22 22:08:06 * * Proposed keymap for Zaurus spitz and akita Oct 22 22:08:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3197 Oct 22 22:19:02 Crofton: if you're here, want to drop into #gumstix to discuss I2C to the TSC chip? Oct 22 22:19:40 Bah, then the phone rings, and I need to go pick up my wife -- I guess we can do it later Oct 22 22:23:53 other than setting MACHINE to spitz are there any other config settings needed to build angstrom for zaurus C3000? Oct 22 22:28:49 sure, there is a number of them Oct 22 22:29:15 DL_DIR, BBFILES, TMPDIR Oct 22 22:29:17 at least Oct 22 22:29:42 I have all that... I'm familiar with oe :) Oct 22 22:30:00 just wondering about anything for this particular device Oct 22 22:30:31 well other than that it works fine for me ;) Oct 22 22:31:58 thanks! Oct 22 22:33:27 do you still need to build 'pivotboot-image'? That build is failing for me on no provider of task-pivotboot Oct 22 22:36:11 hmm, no Oct 22 22:36:22 I have built the console image and x11 image directly Oct 22 22:37:04 * * OE Bug 3198 has been created by B_Linuz(AT)yahoo.com Oct 22 22:37:06 * * syslogd fails to start at boot with busybox 1.7.2 Oct 22 22:37:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3198 Oct 22 22:38:03 ok - I was going off of something I read in the wiki - I'll give it a shot and if it works I'll update the wiki... thanks for the help! Oct 22 23:19:52 florian: how does icecc find about other clients ready to accept jobs Oct 22 23:20:21 does the scheduler enquire on the network ? Oct 22 23:20:40 khem: yes, no configration necessary :-) Oct 22 23:21:04 florian: OK so any machine on my network which has the iceccd running will be used ? Oct 22 23:21:45 right now I am trying to do a kernel compile but I dont see it sending any loads to the other machines Oct 22 23:23:20 any ideas what could be wrong Oct 22 23:23:23 khem: If you have the same version of icecc on both machines and the same name for the icecc net to use configured. Oct 22 23:23:46 version is same I think 0.7.14 Oct 22 23:23:57 khem: and of course you need to tell "make" to spawn multiple compilers. (-j ) Oct 22 23:24:19 by net you mean subnet in LAN Oct 22 23:24:26 yes I am doing -j 16 Oct 22 23:24:45 I started scheduler with no options other than -d Oct 22 23:24:55 is that the way to invoke it Oct 22 23:25:45 oh secondly I am doing a cross compile I have doe the ICECC_CC and ICE_CXX thing too Oct 22 23:25:49 khem: check /etc/icecc/icecc.conf "ICECC_NETNAME" Oct 22 23:26:05 on the server ? Oct 22 23:26:16 khem: on all machines Oct 22 23:26:21 * florian needs some sleep Oct 22 23:26:35 ICECC_NETNAME is empty Oct 22 23:31:54 good Oct 22 23:32:09 khem: icemon might be useful too Oct 22 23:33:14 florian: ok my ubuntu boxes start both scheduler and iceccd Oct 22 23:33:21 should I turn of scheduler ? Oct 22 23:35:37 khem: yes, there should be only one Oct 22 23:35:46 * florian -> sleep, good night Oct 22 23:36:13 it's been a long time for me since I built an image... can someone please tell me what I type to build an angstrom image for my zaurus, angstrom-x11-image seems to be out dated Oct 22 23:37:06 Zero_Chaos: use x11-image Oct 22 23:47:32 so no more angstrom-specific image? cool thanks! Oct 22 23:47:42 and distro should still be 2007.1 right? Oct 22 23:54:54 Zero_Chaos: you can chose any 2007.1 is stabler I believe Oct 22 23:55:15 thanks, is that what the release intends to be? 2007.1? Oct 22 23:55:35 Zero_Chaos: yes Oct 22 23:55:40 great, thanks **** ENDING LOGGING AT Tue Oct 23 02:59:57 2007