**** BEGIN LOGGING AT Tue Aug 29 02:59:56 2006 Aug 29 03:41:58 T0mW: So you have no makedevs in tmp/staging? If you run "bitbake makedevs-native" does makedevs then appear in tmp/staging? Aug 29 03:43:30 I have a bitbake running now, I will try that when it finishes / fails out. Aug 29 03:44:34 v8jlene: after you edit the local.conf, I assume that doing sometthing like "bitbake gpe-image" will build everything needed, then create the IMAGE_TYPES I defined? In this case: jffs2 + tar Aug 29 03:45:59 T0mW: It depends a bit on what you change. If you only changed IMAGE_TYPES then that's fine. If you changed something like machine or target you may need to delete the tmp dir before starting again. Aug 29 03:46:31 correct, I got that info before, so I started all over again. Aug 29 03:47:54 I removed tmp/ dir, then created a local.conf following the wink demo. Then ran "bitbake task-gpe", didn't get anything other than ipk files in deploy. So, now I'm running "bitbake gpe-image" Aug 29 03:48:24 will work fine then Aug 29 03:48:42 ok, good to know that I'm on the right track then. heh Aug 29 03:49:02 That's sound right, since it's the image you want. Aug 29 03:49:11 did I need to run the "task-gpe" first? or not at all? Aug 29 03:49:57 I'm getting the idea that a "task-" only creates ipkg files, nothing more. Aug 29 03:50:39 oh, don't assume that I know anything. I'm coming from buildroot. Aug 29 03:50:47 no ipkg, etc. Aug 29 03:51:17 I'm assuming all that ipkg is is that it is a mere replacement for apt-get or urpmi? Aug 29 03:51:51 The tasks things just list all the various dependencies, so they won't build any images... the images use the tasks to get everything built before packaging it up. Aug 29 03:52:04 ah Aug 29 03:52:20 normally, I would not use the tasks directly. Aug 29 03:52:25 yeah, ipkg is just a cut down version of apt-get/dpkg Aug 29 03:53:18 so the idea is the XXXX-image file is the basic recipe that I'm concerned with, or would build for customization of packages on my platform ? Aug 29 03:53:26 Yep, don't use the tasks directly. Note that the way tasks works was changed in the past day or two, there's an email to the oe-devel list about how to make use of the new task setup. Aug 29 03:53:46 I just subscribed tonight Aug 29 03:53:54 I'll check the archives Aug 29 03:55:03 T0mW: Theres various ways to control what things the images include. If you find that email you'll see the new, more flexible, method. Aug 29 03:55:20 k Aug 29 03:56:03 You can always create your own images, I use "bitbake titan-image" at the moment, which is my own. Although with the recent changes I'm looking at moving back to using one of the standard image types. Aug 29 03:57:31 ok, "RDEPENDS" Aug 29 03:58:12 gpe-image references gpe-task-base which is found inside task-gpe.bb Aug 29 04:03:13 JoeSchmo: Newark, huh? Aug 29 04:03:26 errr, no Aug 29 04:03:37 in the ball park, but i don't live in newark Aug 29 04:04:07 JoeSchmo: just saw your ip info, I'm in Stroudsburg Aug 29 04:04:50 ah Aug 29 04:05:32 always kind of fun to find someone "local" on an irc channel Aug 29 04:06:56 kergoth: ok, it is building makedevs-native & genextfs-native this time. Aug 29 04:10:35 i'm about 40 mins west of newark, not that far Aug 29 04:11:21 Fairfield? Aug 29 04:12:02 heh, 40 minutes on rt80 won't get you far. Aug 29 04:16:31 hi everyone Aug 29 04:21:25 I found some packages installed file in ".debug" in image, how can I avoid these files from being installed? Aug 29 04:23:25 O_Neil: I added some details on how to do this to tge bug about the .debug dirs Aug 29 04:23:45 O_Neil: Here it is: http://bugs.openembedded.org/show_bug.cgi?id=1211 Aug 29 04:23:58 v8jlene: thanks Aug 29 04:32:28 almost made it, pango wants >= glibc-2.10.0 Aug 29 04:35:25 * v8jlene away Aug 29 06:04:02 hi Aug 29 06:12:43 good morning hrw :) Aug 29 06:13:19 03Carsten 07org.oe.oz354x * r704acabc... 10/ (1 packages/zsafe packages/zsafe/zsafe_2.1.3.bb): zsafe: added 2.1.3 - close #1370 Aug 29 06:13:25 03Carsten 07org.oe.dev * re65bb7a2... 10/ (1 packages/zsafe packages/zsafe/zsafe_2.1.3.bb): Aug 29 06:13:25 zsafe: added 2.1.3 - close #1370 Aug 29 06:13:25 applied changes from 26de4ae1a5a4cc78e31a35514476c8f83392d32e (.oz354x) Aug 29 06:13:25 through 704acabcb81bc6c3e644c6c18bb730c30a8d882c Aug 29 06:20:38 hi hvontres|home Aug 29 06:23:17 03hrw 07org.oe.oz354x * r7c5553db... 10/ (4 files in 2 dirs): libxine: add libpng to DEPENDS - close #1357 Aug 29 06:23:22 03hrw 07org.oe.dev * rc9c7d03b... 10/ (4 files in 2 dirs): libxine: add libpng to DEPENDS - close #1357 Aug 29 07:30:47 good morning all Aug 29 07:32:37 * T0mW is still trying to build something Aug 29 07:37:12 ok, looks like gpe-image has hit a brickwall that I cannot get around. I got around pango needing glib-2.0 > 2.10, but now dbus-native fails. Aug 29 07:37:30 Can I drop back a few weeks to something stable? Aug 29 07:38:14 So far, nothing I've tried to build has gone all the way through without stopping. Aug 29 07:40:12 the frustrating thing is that I've been building code for a long long time now and I cannot see what OE is doing with the sources so I can fix problems. Aug 29 07:40:53 * T0mW decides to blow it all OE away and start over again. Aug 29 07:41:31 morning all Aug 29 07:41:58 T0mW: How do you mean "see what OE is doing with the sources"? Aug 29 07:42:23 re Aug 29 07:43:03 T0mW: look into tmp/work/name-of-recipe*/temp/ for logs which show what OE is doing with the sources Aug 29 07:43:32 what I've been running into is that the bitmake gpe-image will decide to build glib-2-0_2.6.4, then pango 1.14.0 decides that that version of glib-2 is not sufficient. Aug 29 07:43:52 T0mW: Which bitbake are you using? Aug 29 07:44:03 the latest and greatest Aug 29 07:44:16 The one marked unstable? Aug 29 07:44:37 Whatever the wiki told me to do on "getting started" I did Aug 29 07:45:00 ok, that isn't the latest and isn't unstable so that's ok :) Aug 29 07:45:30 You need to convince bitbake you want a later version of glib to build Aug 29 07:45:37 hmm, something is broken then. I've tried a couple of times to build something and I've hit these dependancy errors Aug 29 07:46:12 These kind of things are controlled by the distro. Which distro are you using? Aug 29 07:46:37 Isn't there a "known good" set of files that I can start with? e.g.: use this local.conf, then run "bitbake X" ? Aug 29 07:46:57 There are known good configurations people use often Aug 29 07:47:10 I've tried generic, generic-uClibc, then familiar-unstable. Aug 29 07:47:14 Say MACHINE="spitz" and DISTRO="openzaurus-unstable" Aug 29 07:47:52 do I need to say anything else except BBFILES, MACHINE, DISTRO ? Aug 29 07:48:30 If you want really stable/working, you could try the oz354x branch - that is the most guaranteed to build (for zaurii machines) Aug 29 07:48:47 No, those variables should cover it Aug 29 07:49:04 My ultimate build will be for a h1940, closest machine to my hardware. Aug 29 07:49:40 morning Aug 29 07:49:54 T0mW: That's tricky as its tradionally a familiar machine but familiar in .dev doesn't get much attention Aug 29 07:49:55 Somebody should update the wiki regarding the LocalConfExamples, that conf won't build, it complains that Zaurus version 3.5.4.1 is not valid Aug 29 07:50:34 T0mW: Change it to say openzaurus-unstable ;-) Aug 29 07:50:53 3.5.4.1 is in the oz354x branch now Aug 29 07:51:01 ok, I've got to reinstall the OE as I blew it away in frustration, I'll give the spitz a try. Aug 29 07:51:03 morning XorA Aug 29 07:51:44 hi XorA Aug 29 07:51:53 MACHINE=h1940 with DISTRO=openzaurus-unstable should get you most of the way Aug 29 07:51:58 lack of any docs how to make working build suxx Aug 29 07:52:19 RP: on that spitz machine, what command should I issue to bitbake? "bitbake ? " Aug 29 07:52:30 T0mW: btw nstall ccache to make blowing away tmp less painful Aug 29 07:52:46 I've ccache here Aug 29 07:52:52 T0mW: bitbake and then whatever you want to build. bootstrap-image is a nice simple one, opie-image/gpe-image more complex Aug 29 07:53:17 bitbake bash would build bash Aug 29 07:53:40 RP: respectfully, that is my problem, I don't know what will work and what won't. I'll try the "bitbake bootstrap-image" then. Aug 29 07:54:30 T0mW: I'll set a clean oz spitz build away here too, then if it fails and we're lucky, we'll see the same errors :) Aug 29 07:54:53 RP: spitz won't work on his armv4t hardware Aug 29 07:55:12 IIRC Aug 29 07:56:35 Ah, 1940 is armv4t? That is annoying as most of the Zs are v5 :-( Aug 29 07:56:44 armv5t Aug 29 07:56:49 arm920t Aug 29 07:57:02 arm920t = armv4t Aug 29 07:57:06 k Aug 29 07:57:28 my arm 920 SoC sigills on clz, but not on bx Aug 29 07:57:37 Yes, its v4 Aug 29 07:58:09 T0mW: ok, try MACHINE = "h1840", DISTRO="openzaurus-unstable" as XorA recommended Aug 29 07:58:13 ~lart gcc people for breaking armv4/EABI support Aug 29 07:58:14 * ibot accelerates a free AOL cd to 50,000 rpm and lets gcc people feel it for breaking armv4/EABI support Aug 29 07:58:16 ok Aug 29 07:58:17 s/18/19/ Aug 29 07:58:36 heh Aug 29 07:58:53 I did 'for i in * ; bitbake $i ; done' before I went to bed Aug 29 07:59:10 I return to see a bash prompt expecting me too fix patches Aug 29 08:00:22 pango-1.14 does that too Aug 29 08:03:00 koen: blame kergoth ;-) Aug 29 08:03:12 is the h1940 the super cheapo ipaq? Aug 29 08:03:27 hehe Aug 29 08:04:00 koen: yeah, a patch failure will drop you into a shell in the source tree so you can fix it. i'll check in an alternate resolver class to be used to disable the functionality Aug 29 08:04:15 XorA: that's the 1930 (where they removed the serial port leads) Aug 29 08:05:12 XorA: uber cheap is h1910 Aug 29 08:06:48 "mtn: misuse: branch 'oz354x' is empty" Aug 29 08:07:55 org.openembedded.ox354x Aug 29 08:08:13 s/ox/oz/ Aug 29 08:08:16 yeah, noticed that reverse notation, ok, that worked Aug 29 08:08:18 org.openembedded.oz54x even Aug 29 08:08:35 * RP gives up Aug 29 08:08:39 hehe Aug 29 08:08:52 not enough caffeine yet? ;) Aug 29 08:08:55 org.openembedded.oz354x Aug 29 08:09:01 RP: your stuck on dev forever :-) Aug 29 08:09:14 kergoth: looks a bit like it :) Aug 29 08:09:27 XorA: Could be worse ;-) Aug 29 08:09:41 morning all Aug 29 08:09:59 hi Dirk Aug 29 08:10:12 hey Richard Aug 29 08:10:43 hmm Aug 29 08:10:44 Connecting to www.handhelds.org|128.31.0.25|:80... failed: Connection refused. Aug 29 08:10:46 heh Aug 29 08:33:49 ok, this is what I did, I don't know if the branch checkout was correct? http://rafb.net/paste/results/djMqav12.html Aug 29 08:41:15 T0mW: oz354x only supports zaurus machines. If you want to use h1940, you'll need .dev Aug 29 08:41:43 T0mW: DISTRO=openzaurus-3.5.4.1 is in the branch, DISTRO=openzaurus-unstab;le is .dev Aug 29 08:41:57 I'm using the spitz machine Aug 29 08:42:41 T0mW: ok, it looks ok. Are you seeing an error? Aug 29 08:43:15 1/cr Aug 29 08:43:31 Polish Linux news site noted release of Familiar 0.8.4 Aug 29 08:43:43 article is called 'Familiar vs OpenZaurus' ;D Aug 29 08:43:48 RP: http://rafb.net/paste/results/2DCM5765.html Aug 29 08:44:38 conf is: http://rafb.net/paste/results/fnrq3Z97.html Aug 29 08:45:41 koen: your and florian attempts to use OZ on h2200/hx4700/nbp are noted Aug 29 08:46:09 :) Aug 29 08:47:45 T0mW: I'm not sure about BBFILEs vs BBFILES Aug 29 08:47:52 RP: what is odd is that bootstrap-image.bb is in org.openembedded.oz354x/packages/meta/ Aug 29 08:47:59 03mickeyl 07org.oe.dev * raa2272bf... 10/ (1 packages/libesmtp/libesmtp_1.0.4.bb): libesmtp: correct damage done by previous commit Aug 29 08:48:20 T0mW: I also remember HOME not being recommended for certain variables. I can't remember is BBFILES was one of them... Aug 29 08:48:22 RP: that is it! Aug 29 08:48:46 "Handling BitBake files: " Aug 29 08:49:55 03pH5 07org.oe.dev * r736e3745... 10/ (3 files in 2 dirs): samba: fix -dbg packaging for 3.0.14a and 3.0.20 Aug 29 08:52:14 RP: "TARGET_OS := linux" ? Aug 29 08:53:01 03koen 07org.oe.dev * r5aa5d42f... 10/ (4 files in 3 dirs): linux-ezx: update to ezx7 and add 2 patches for gcc4-isms Aug 29 08:53:49 T0mW: It should default to that Aug 29 08:53:59 RP: it won't Aug 29 08:54:27 T0mW: ? Aug 29 08:56:09 RP: http://rafb.net/paste/results/FGjroT54.html Aug 29 08:56:22 re Aug 29 08:56:51 T0mW: specify your target OS in your 'local.conf' Aug 29 08:56:58 I did now Aug 29 08:57:14 T0mW: mine is specified with just a '=' not a ':=' Aug 29 08:57:24 T0mW: ah, use DISTRO = "openzaurus-3.5.4.2" for that branch, not unstable Aug 29 08:57:42 T0mW: You should not need to set TARGET_OS with a valid distro Aug 29 08:57:52 bigd0g: yeah, I caught that ":=" vs "=" IIRC, := is expanded on each pass, = sets it once Aug 29 08:58:12 * bigd0g nods Aug 29 08:58:31 RP: shrug, bitbake didn't like it. Aug 29 08:58:54 RP: its building ipkg now Aug 29 08:59:07 RP: its building ipkg-native now Aug 29 08:59:14 T0mW: Did you set the DISTRO correctly? Aug 29 08:59:48 DISTRO = "openzaurus-unstable" Aug 29 09:00:15 oh Aug 29 09:00:37 The fact TARGET_OS wasn't set was a sign of a bigger problem Aug 29 09:01:09 Read what I said above as it will make things easier in the long run ;-) Aug 29 09:01:17 03koen 07org.oe.dev * r6f596144... 10/ (10 files in 2 dirs): linux-ezx: remove outdated patches Aug 29 09:01:24 ~seen Crofton Aug 29 09:01:25 RP: you seem to like misspelling DISTRO as DISTO :D Aug 29 09:01:32 I versioned it this time, blew tmp away and re-running Aug 29 09:01:42 crofton is currently on #casualti (9h 22m 1s) #ol (9h 22m 1s) #oe (9h 22m 1s) #elinux (9h 22m 1s), last said: 'Where does version.h come from?'. Aug 29 09:02:06 ibot: re Aug 29 09:02:11 re is probably religious education Aug 29 09:02:16 lol Aug 29 09:02:37 mickeyl: I do? Aug 29 09:03:06 RP: ok, that distro set everything correctly Aug 29 09:03:20 good morning Aug 29 09:03:39 * T0mW thinks ruby on rails is easier Aug 29 09:03:44 hi florian_kc, zecke Aug 29 09:04:29 hi florian_kc Aug 29 09:10:09 can someone list the current org.openembedded.dev database rev ? Aug 29 09:10:12 mine's listed as befec029b0968612722681e4b1bad74128e981d6 Aug 29 09:10:24 i'm quite sure that's out-of-date... but it says there's nothing to update Aug 29 09:10:26 * bigd0g shrugs Aug 29 09:10:56 r6f596144... Aug 29 09:11:36 hrmm Aug 29 09:12:02 sure you don't accidantly have a fixed branch in? try updating org.openembedded.* Aug 29 09:12:03 koen: i did " mtn --db=OE.mtn pull monotone.openembedded.org org.openembedded.dev " Aug 29 09:12:07 koen: and then " cd org.openembedded.dev && mtn update " Aug 29 09:12:12 koen: does that some incorrect? Aug 29 09:12:17 s/some/seem Aug 29 09:12:21 that's correct Aug 29 09:12:26 hrm. Aug 29 09:12:30 * bigd0g ponders Aug 29 09:12:42 you might have noticed that we committed a lot of stuff the past few minutes Aug 29 09:12:56 takes a bit to update? Aug 29 09:13:17 synchronize, i mean Aug 29 09:13:43 03koen 07org.oe.dev * re69ec8c2... 10/ (5 files in 4 dirs): imagemagick: add 6.2.9, closes #336 Aug 29 09:14:17 koen: you got my recipe as is? Aug 29 09:14:21 as soon as CIA reports it, it is in the repo Aug 29 09:14:29 hrw|work: no Aug 29 09:14:59 hrw|work: this recipe only leaves out the .la files Aug 29 09:15:32 koen: and package perl stuff separatelly? Aug 29 09:16:05 no, since that points to i686 staging over here Aug 29 09:17:52 do you have to blow away tmp after updating the database? Aug 29 09:18:05 no Aug 29 09:18:14 (generally, no) Aug 29 09:18:33 hrm. ok. well, i'll try it anyway Aug 29 09:18:40 noted, though :) Aug 29 09:19:49 03pH5 07org.oe.dev * ra7f77b32... 10/ (3 files in 2 dirs): libgnomecanvas: fix -dbg packaging for 2.10.2 and 2.12.0 Aug 29 09:22:19 anyone of you ever gave your account number for international money transfer? Aug 29 09:22:24 what information does the sender need? Aug 29 09:23:15 IBAN & BIC, that's everything? Aug 29 09:23:19 morning all Aug 29 09:23:21 mickeyl: IBAN (and BIC) Aug 29 09:23:22 cheers lrg Aug 29 09:23:26 koen: thanks. Aug 29 09:23:28 what is BIC? Aug 29 09:23:45 hrw|work: the same as IBAN, but different Aug 29 09:23:51 ah Aug 29 09:24:02 some banks charge extra €€€€ if you only provide one Aug 29 09:24:51 s/some/most/ Aug 29 09:25:44 RP: looks like some editing to do on this package: http://rafb.net/paste/results/zJq00754.html Aug 29 09:26:15 T0mW: lart the qemu people for not supported a recent gcc Aug 29 09:26:26 hehhe Aug 29 09:26:36 supporting* Aug 29 09:27:02 I guess they would rather stick with "stable" other than this bleeding edge stuff Aug 29 09:27:08 T0mW: If you provide a gcc 3.x as gcc-3.4 or gcc-3.3, OE will use it and you will avoid that problem Aug 29 09:27:30 hmmm Aug 29 09:27:33 it didn't over here Aug 29 09:27:36 yeah, all my stuff is 4.0x Aug 29 09:27:46 I have to switch my 'gcc' over to 3.3 Aug 29 09:28:03 and I do have these in my path: gcc gcc-3.3 gcc-3.4 gcc-4.0 gcc-4.1 gccbug gccbug-3.3 gccbug-3.4 gccbug-4.0 gccbug-4.1 gccmakedep Aug 29 09:28:05 koen: still using 2.95-3 eh? Aug 29 09:28:28 koen: it does look for gcc-3.4/gcc-3.3 - see the .bb file... Aug 29 09:29:25 ~lart people subscribing to mailing lists plus protecting their account w/ the friends system Aug 29 09:29:26 * ibot whacks people subscribing to mailing lists plus protecting their account w/ the friends system with the cluebat Aug 29 09:29:37 ah, only the cvs version does Aug 29 09:29:45 we need a qemu-native.inc :) Aug 29 09:30:41 ok, I give up, time for sleep Aug 29 09:30:52 RP: thanks for the attempt, I keep trying. Aug 29 09:31:01 RP: or syncing with .oz354x Aug 29 09:32:31 hrm... i did the mtn update of org.oe.dev, but it's still listing an old version when i compile Aug 29 09:32:32 OE_REVISION = "befec029b0968612722681e4b1bad74128e981d6" Aug 29 09:32:40 * bigd0g scratches his head Aug 29 09:32:43 marcan: shout when you're about Aug 29 09:32:51 hi mikearthur Aug 29 09:32:54 hey lrg Aug 29 09:32:59 hey mikearthur Aug 29 09:33:00 bigd0g: what does mtn heads show? Aug 29 09:33:04 or mtn list heads Aug 29 09:33:14 XorA, hrw|work: yo Aug 29 09:33:23 do13_: you around? Aug 29 09:33:26 or 'mtn head' Aug 29 09:33:37 wants me to specify a branch ? Aug 29 09:34:06 mtn: branch 'org.openembedded.oz354x' is currently merged: Aug 29 09:34:06 7c5553dbd2bd2a9bf37d8f5e0db983e7f0075ba6 hrw@openembedded.org 2006-08-29T06:35:34 Aug 29 09:34:08 XorA: free upgrade to 2 * core duo for your PC - (4 cores) Aug 29 09:35:26 does this look right? " mtn head --branch=org.openembedded.dev --db=OE.mtn" Aug 29 09:35:36 mtn: branch 'org.openembedded.dev' is currently merged: Aug 29 09:35:36 befec029b0968612722681e4b1bad74128e981d6 hrw@openembedded.org 2006-08-24T15:28:39 Aug 29 09:35:52 bigd0g: a bit old.. Aug 29 09:36:01 mtn: branch 'org.openembedded.dev' is currently merged: Aug 29 09:36:01 9aa86687e7b05c58ae96324abb450a588180f55e pH5@openembedded.org 2006-08-29T09:32:05 Aug 29 09:36:06 hrw|work: i know... i'm trying to merge the new one ;p Aug 29 09:36:14 hrw|work: but it doesn't seem to be... uh... merging Aug 29 09:36:18 bigd0g: mtn pull;mtn merge;mtn update is my combo Aug 29 09:36:28 hrw|work: ok, i'll give that a whirl. Aug 29 09:36:38 hrw|work: must've forgotten a step. Aug 29 09:38:52 mtn sync ; mtn merge ; mtn up ; mtn sync (if you have write access) Aug 29 09:41:22 koen: months ago we discussed that sync should be changed to pull/push... which way if official then? Aug 29 09:41:45 sync = no local branches Aug 29 09:42:06 but it's semantics Aug 29 09:42:29 I like 'sync', because I know that the other end will have the same data as I have Aug 29 09:43:28 so the commands do the same thing? or 'sync' does something different? Aug 29 09:44:22 lrg: what the hell do you do with all those cores :-) Aug 29 09:44:54 XorA: smoke em with with bitbake-mt Aug 29 09:45:01 4 processor intensive tasks simultaneously :) Aug 29 09:45:28 now a 4 core zaurus, that would be rockin Aug 29 09:45:53 hrw|work: when i do the "mtn merge --branch=org.openembedded.dev --db=OE.mtn" Aug 29 09:46:02 hrw|work: after a successful pull, i get "mtn: branch 'org.openembedded.dev' is already merged" Aug 29 09:46:15 XorA, koen, bigd0g: 1 for mtn, 1 for bb, 1 for make and the other for gcc ;) Aug 29 09:46:32 lrg: haha :) Aug 29 09:46:42 bigd0g: cd 'working copy';mtn merge Aug 29 09:47:17 * XorA bets dell offers that machine with XP home edition as well :-) Aug 29 09:47:27 hrw|work: still says "mtn: branch 'org.openembedded.dev' is already merged" Aug 29 09:47:38 bigd0g: then it is ok - skip to 'mtn update' Aug 29 09:47:46 hrw|work: alrighty. Aug 29 09:47:58 bigd0g: as developer without r/w access you can even skip 'merge' step Aug 29 09:48:10 bigd0g: mtn pull/update would be all you need Aug 29 09:48:17 hrw|work: noted. Aug 29 09:49:01 uh... oh Aug 29 09:49:30 hrm... error... mtn: error: merge failed due to unresolved conflicts Aug 29 09:49:45 hrw|work: No external 3-way merge command found. Aug 29 09:49:51 what does that mean? heh. Aug 29 09:51:08 that you have a local diff and/or a local commit Aug 29 09:51:21 (and no kdiff3/meld/xxdiff/emacs installed) Aug 29 09:51:45 koen: which do you recommend? Aug 29 09:52:00 ~lart freaky autoconf scripts Aug 29 09:52:01 * ibot judo chops freaky autoconf scripts Aug 29 09:55:28 is the cause of the "NOTE: preferred version 2.4 of glibc not available" understood? Aug 29 09:55:45 I'm getting the same for dbus 0.92 now with bitbake trunk Aug 29 09:55:46 pH5: angstrom builds? Aug 29 09:56:18 pH5: probably because dependencies have DEFAULT_PREF -1 Aug 29 09:57:04 hrw|work: my gpe-image build is at libbonoboui currently Aug 29 09:57:14 koen: but dbus doesn't have -1 anymore? Aug 29 09:57:23 pH5: dbus-native? Aug 29 09:57:48 hm. dbus only depends on "expat glib-2.0 virtual/libintl" now.. Aug 29 09:58:03 ok Aug 29 09:58:09 in that case I don't know Aug 29 09:59:41 The glibc 2.4 issue is there as it needs to build glibc-initial/intermediate before it can build glibc itself Aug 29 10:01:43 I think that I will leave tosa powered at home next time to have arm machine available by network for doing tests... Aug 29 10:02:17 morning Aug 29 10:02:34 hi CoreDump|home Aug 29 10:02:56 checking for GCONF_DBUS... configure: error: You need either ORBit or D-BUS for Aug 29 10:02:56 IPC Aug 29 10:03:00 mehhh Aug 29 10:03:08 it works in .dev :) Aug 29 10:04:54 ethereal compilation takes lot of time... Aug 29 10:10:39 mikearthur: yep. I'm around Aug 29 10:12:02 do13_: great Aug 29 10:12:08 do13_: pm Aug 29 10:14:28 ~hail schurig Aug 29 10:14:36 * ibot bows down to schurig and chants, "I'M NOT WORTHY!!" Aug 29 10:19:01 brb Aug 29 10:20:04 looks like intel hired some dec engineers Aug 29 10:20:15 16MB cache on the new xeon Aug 29 10:20:29 too bad it's not L2 or L1 cache Aug 29 10:20:44 sooner or later they will add 256M of L4 Aug 29 10:21:03 aargh Aug 29 10:21:08 still netburst Aug 29 10:24:33 Hey guys Aug 29 10:24:39 hey Laibsch Aug 29 10:24:54 hi Laibsch Aug 29 10:25:54 * CoreDump|home hugs his c2d Aug 29 10:26:27 CoreDump|home: try the new multithreaded bitbake :) Aug 29 10:26:59 koen: you're kidding right? Aug 29 10:27:07 CoreDump|home: no Aug 29 10:27:12 cool Aug 29 10:27:14 CoreDump|home: no Aug 29 10:27:44 I will try it someday ;) Aug 29 10:27:52 NOTE: DBG: Active builds increased to 2 by /home/koen/OE/monotone/org.openembedded.dev/packages/angstrom/angstrom-bootstrap-image.bb doing do_populate_staging Aug 29 10:28:37 Well, I'm running w/ -j3 already =) Aug 29 10:28:47 CoreDump|home: thats not the same Aug 29 10:29:01 CoreDump|home: you are wasting cputime on fetching ;) Aug 29 10:29:17 oh lol Aug 29 10:29:40 CoreDump|home: http://ewi546.ewi.utwente.nl/tmp/hrw/fb/ Aug 29 10:30:13 hrw|work: does that fb use overlays to make it faster? Aug 29 10:30:14 hrw|work: excellent idea Aug 29 10:31:06 koen: it use free cycles of usbhost to decode divx in realtime Aug 29 10:31:07 hrw|work: I strongly suspect the DELETE FROM isn't working when updating Aug 29 10:31:31 koen: yep.. Aug 29 10:31:45 and I know why Aug 29 10:35:21 good news. i convinced my employers to put me on the DMZ Aug 29 10:35:26 no more proxy problems. Aug 29 10:35:35 w00t! Aug 29 10:37:04 RP: I found a bug in RDEPENDS handling Aug 29 10:37:14 03hrw 07org.oe.dev * r8f3da738... 10/ (1 contrib/feed-browser/update.php): feed-browser: make all sections lowercase, really remove packages before update Aug 29 10:38:15 RP: http://pastebin.ca/153047 Aug 29 10:38:25 hrw|work: thanks Aug 29 10:41:15 * hrw|work want top like tool which will show how much mem task really use.. Aug 29 10:42:02 bigd0g, ? Aug 29 10:42:37 Crofton: are you going to be around for a bit? Aug 29 10:42:44 * Crofton really needs some tools to show me actual phys mem usage Aug 29 10:42:52 Crofton: i'm probably going to need help getting that uboot thing working Aug 29 10:42:53 mostly Aug 29 10:43:12 Crofton: ok, cool. i'm updating the database and such right now. it might fix itself Aug 29 10:43:18 mtn pull, mtn update, bitbake bootstrap-image ? Aug 29 10:43:22 :) Aug 29 10:43:53 I need to figure out why I can't get it to run arm-depmod-2.6 Aug 29 10:43:57 in the kernel build Aug 29 10:44:01 where's "bootstrap-image" coming from. should i not be doing "bitbake linux-omap1" ? Aug 29 10:44:10 03pH5 07org.oe.dev * r872617c6... 10/ (1 packages/gnome/libbonoboui_2.10.0.bb): libbonoboui-2.10.0: fix -dbg packaging Aug 29 10:44:22 bootstrap-image is a basic flash image Aug 29 10:44:32 building it will build the kernel Aug 29 10:44:49 and linux-omap1 just sets up the toolchain? Aug 29 10:45:24 well, the dependencies work all that stuff out Aug 29 10:45:46 so linux-omap1 is a completely different flash build than bootstrap-image ? Aug 29 10:46:09 linux-omap1 only builds the kernel image Aug 29 10:46:21 bootstrap-image will build a flash file system Aug 29 10:46:32 with busybox and someother stuff on it Aug 29 10:46:52 ok. i'll eventually have to do the boostrap-image. but one thing at the time, i guess Aug 29 10:47:00 yeah Aug 29 10:47:31 mtn error - "No external 3-way merge command found." Aug 29 10:47:36 i installed emacs as koen suggested Aug 29 10:47:39 do i have to do something else? Aug 29 10:47:57 bigd0g: export MTN_MERGE=emacs will maybe help Aug 29 10:48:04 koen: its a debian.bbclass related bug. Aug 29 10:48:04 * hrw|work use kdiff3 or vim Aug 29 10:48:16 RP: I guessed as much Aug 29 10:48:47 koen: We need to not generate the pkgmaps files for files with DEBIAB_NOAUTONAME set Aug 29 10:48:47 * koen did RPROVIDES_tslib-conf = "libts-0.0-conf" Aug 29 10:48:53 hrw|work: ok, trying the env var Aug 29 10:48:58 koen: fix it properly ;-) Aug 29 10:49:17 although I could do RPROVIDES_pn-tslib-conf = "libts-0.0-conf" in local.conf nowadays Aug 29 10:49:34 hrw|work: the env var didn't work. any other ideas? Aug 29 10:49:43 koen: not quite Aug 29 10:50:09 koen: RPROVIDES_pn-tslib_tslib-conf = "libts-0.0-conf" Aug 29 10:50:36 bigd0g, make the tune-arm ... file look like the one in the repo :) Aug 29 10:50:38 (tslib-conf isn't a PN) Aug 29 10:50:47 morning Aug 29 10:50:56 I think that is the 3 way merge? Aug 29 10:51:27 no... trying to merge linux-omap1_2.6.12-rc2.bb Aug 29 10:51:48 i should probably get 3-way merging working, though Aug 29 10:52:47 where do i set $EDITOR in mtn ? Aug 29 10:53:11 sas "set this explicitly in your get_preferred_merge3_command hook" Aug 29 10:53:17 s/sas/says Aug 29 10:54:13 I don't know how mine knows what to use :) Aug 29 10:54:31 hehe :) Aug 29 10:54:51 are you using kdiff3, Crofton? emacs? Aug 29 10:55:45 meld I think Aug 29 10:56:07 can you back out your local changes to linux-omap1 .... Aug 29 10:56:37 is that directed at me? i shouldn't have the ability to write anything to the db Aug 29 10:56:44 yeah Aug 29 10:56:54 how did it? wha? Aug 29 10:56:56 * bigd0g is confused Aug 29 10:56:59 you have local changes in the checked out copy Aug 29 10:57:29 so you have the version comoing from the server, the version in your local db, and the version on your disk Aug 29 10:57:47 how do i back out the changes? Aug 29 10:58:02 hmmm Aug 29 10:58:32 anyone know how to do monotone revert? Aug 29 10:58:42 mtn revert Aug 29 10:58:52 heh Aug 29 10:58:56 so obvious Aug 29 10:59:32 will that work on the directory? Aug 29 11:00:19 just do it on the files you have changes Aug 29 11:00:19 yes Aug 29 11:00:31 mtn revert . will do it recursively Aug 29 11:00:37 make sure you don't have any changes you really need Aug 29 11:00:50 well, all the changes i made i needed to make it work ;p Aug 29 11:00:55 mtn diff . > shelf.diff ; mtn revert . Aug 29 11:00:56 i have them documented, tho Aug 29 11:01:25 koen: thanks. good idea to log them. Aug 29 11:01:43 Crofton: reverting the db now Aug 29 11:01:53 that's what I used to do before I discovered bitbake collections Aug 29 11:02:17 Crofton: tune-arm926ejs.conf and linux-omap1_2.6.12-rc2.bb reverted Aug 29 11:02:30 hehe Aug 29 11:02:43 you'll needto change the tune file after the update Aug 29 11:02:50 it has better comments now Aug 29 11:02:55 yeah. that damnable hyphen Aug 29 11:03:02 oh, cool. comments ftw! Aug 29 11:03:38 ok... now... 3-way merge... you use meld? where is it specified for mtn? Aug 29 11:03:59 iirc it just checks for its precense Aug 29 11:04:18 I use meld, must be the default Aug 29 11:04:19 * koen can't spell today Aug 29 11:04:33 I am sure the monotone docs can clear this up for you Aug 29 11:05:13 reading them now. just didn't know if you had a quick answer or not Aug 29 11:05:19 thanks. Aug 29 11:05:38 once you revert you should be able to pull/update Aug 29 11:06:14 03koen 07org.oe.dev * rb09ecf88... 10/ (1 conf/machine/include/tune-arm926ejs.conf): tune-arm926ejs: .dev uses gcc 4 by default Aug 29 11:07:10 Crofton: good point. i probably don't even need the 3-way merge Aug 29 11:07:32 you will need to fix it eventually Aug 29 11:08:18 indeed. Aug 29 11:08:24 RP: is it possible to use EXCLUDE_FROM_SHLIBS on -dbg package by default? Aug 29 11:09:10 koen: What does that variable do? :) Aug 29 11:09:35 it stops OE from making that package a provider for its shlibs Aug 29 11:09:40 * koen looks at hrw Aug 29 11:10:03 Is there a kernel.bbclass guru in the house? Aug 29 11:10:07 I *think* that should fix the 'package depends on foo-dbg' stuff people have been seeing Aug 29 11:10:46 RP: it exclude package^Wrecipe from being used for shlibs calculation Aug 29 11:13:15 koen: Based on what hrw said and looking at the code, its too coarse to do what you want :-/ Aug 29 11:13:17 great! successful db pull and update Aug 29 11:13:24 meld installed, too Aug 29 11:13:28 heh Aug 29 11:13:39 it's the small things ;) Aug 29 11:13:47 Crofton: Depends what you mean by expect ;-) Aug 29 11:13:55 expert even Aug 29 11:14:04 * RP goes back to bed Aug 29 11:14:30 bitbake just built install-native in work/iwmmxt-linux-gnueabi/install-native here. why isn't it in work/i686-linux? Aug 29 11:14:33 RP, I am building a kernel from a git tree Aug 29 11:14:41 that lacks version.h Aug 29 11:14:42 RP: to bed? gf waits for you or you are sick? Aug 29 11:14:55 hrw|work: Just can't type ;-) Aug 29 11:15:02 this messes up the calculation of KERNEL_VERSION Aug 29 11:15:13 and the selection of of arm-depmod-2.6 Aug 29 11:15:24 Crofton: insert a version.h :) Aug 29 11:15:36 this is one solution :) Aug 29 11:15:51 hmm, it looks like BBCOLLECTIONS screws up the OE_REVISION code Aug 29 11:16:52 Crofton: Doesn't compiling the kernel generate a version.h? Aug 29 11:17:40 hang on Aug 29 11:17:43 Crofton: As for the depmod issue, what's the name of your .bb? Aug 29 11:18:00 Crofton: comments in the arm9 tune file are much clearer :) Aug 29 11:18:12 Crofton: how is gcc4 support on the omap5912 ? Aug 29 11:18:14 linux-omap1_2.6.17+git.bb Aug 29 11:18:23 bigd0g, I am goign to fins out Aug 29 11:18:42 I have a version.h after I compile Aug 29 11:18:54 Crofton: It started with 2.6 so the depmod business should work Aug 29 11:19:07 RP, it doesn't :) Aug 29 11:20:04 Crofton: It works for linux-openzaurus_2.6.17+git ;-) Aug 29 11:20:11 Crofton: i see linux-omap1_2.6.x+git.bb ... will it specify itself when i 'bitbake linux-omap1' ? Aug 29 11:20:30 I set prefered version to get it Aug 29 11:20:38 RP Ilied about the name Aug 29 11:20:53 linux-omap1_2.6.x+git.bb Aug 29 11:21:00 but I checked the 2.6.17 name also Aug 29 11:21:07 Crofton: that should still work Aug 29 11:21:21 mickeyl, RP, koen: can url to OEDEM page be shared with !OE people? Aug 29 11:21:39 Crofton: What does DEPENDS say for that file? (bitbake -e | grep DEPENDS) Aug 29 11:21:49 hrw|work: Fine with me Aug 29 11:22:25 the depend for arm-dpemod-2.6 seems ok Aug 29 11:22:29 mickeyl: ? Aug 29 11:23:03 http://pastebin.ca/153084 Aug 29 11:24:11 *sigh* Aug 29 11:24:27 compiling busybox on my box makes it kneel Aug 29 11:24:31 Crofton: Add a -b /somepath/somefile.bb to that ;-) Aug 29 11:24:47 hrw|work: fine with me Aug 29 11:25:01 ok Aug 29 11:25:13 yeah Aug 29 11:26:04 you want the DEPENDS= line? Aug 29 11:26:14 DEPENDS="ipkg-utils-native file-native quilt-native virtual/arm-linux-gcc virtual/libc u-boot virtual/arm-linux-depmod-2.6 virtual/arm-linux-gcc update-modules" Aug 29 11:26:25 then it is ok Aug 29 11:27:24 it look like the 2.6 from the bb file name picks up the arm-depmod-2.6 dependency Aug 29 11:27:41 but that is calculated differently from the depmod selection when it is run Aug 29 11:29:50 The depend calculation is based of ${PV} and the selection on what depmod to actually run is based of the string from version.h Aug 29 11:31:30 Crofton: so do i specify the "DISTRO_VERSION" var in my 'local.conf' to use the new bb file? Aug 29 11:31:33 hrw|work: sure Aug 29 11:31:41 Crofton: or does it happen automagically? Aug 29 11:31:57 the version info for the kernel file is in the omap5912.conf machine file Aug 29 11:32:30 I haven't actually flashed that kernel into my OSK yet :) Aug 29 11:32:44 I have flashed the u-boot though Aug 29 11:32:58 ok, no bother. i'll carry on as before. :) Aug 29 11:33:02 03mickeyl 07org.oe.dev * r66f4e0ab... 10/ (1 packages/python/python_2.4.3.bb): python 2.4.3 add PACKAGES_DYNAMIC -- close #1366 Aug 29 11:33:20 RP: two questions: a) do you still have that pure-matchbox image in mind I was pestering you to push when you have a chance? b) do we have an example image-bb that bases on task-base yet? Aug 29 11:34:01 mickeyl: All images can use task-base ;-) Aug 29 11:34:45 mickeyl: I made task-bootstrap a variable so you can change it (MACHINE_TASK_PROVIDER) Aug 29 11:35:19 RP, any thoughts on my problem? Aug 29 11:35:23 mickeyl: I still like the idea of that image. What you want is task-oh-core in poky ;-) Aug 29 11:36:36 mickeyl: I will aim to strip the bits OE needs from that and add it to OE :) Aug 29 11:37:02 RP: ya, that'd be pretty cool, thanks. Aug 29 11:38:12 Crofton: I don't understand your problem - it seems to change... Aug 29 11:38:14 ~lart 24h disconnect Aug 29 11:38:15 * ibot duct-tapes 24h disconnect to the floor and drools on him Aug 29 11:38:28 heh Aug 29 11:38:34 hmm Aug 29 11:38:44 arm-depmod-26 is built fine Aug 29 11:38:54 Can anyone help me figure out why I can't compile busybox on my x86_64 box? Aug 29 11:38:58 Crofton: ok Aug 29 11:38:59 but when the bb file goes to run it, it runs arm-depmod- Aug 29 11:39:23 gcc eats up all my memory, until there's nothing left and I have to kill it Aug 29 11:39:26 because the 2.6 it needs is calculated from version.h, not ${PV} Aug 29 11:39:39 slows my x64 3700+ with 1gb ram to a crawl! :-( Aug 29 11:39:47 NAbyss, busybox will build eventually, you must be patient Aug 29 11:40:05 * NAiL assumes tab usage Aug 29 11:40:32 Crofton: Are you running a config without modules? Aug 29 11:40:41 bonjour Aug 29 11:40:51 NAiL, my x86_64 only has .5G RAM Aug 29 11:41:08 RP, not sure what you mean Aug 29 11:41:10 does your box become unresponsive? Aug 29 11:41:16 yes Aug 29 11:41:23 it does finish eventually Aug 29 11:41:23 I use a couple of minutes just to login to a terminal :-( Aug 29 11:41:32 Crofton: What does CONFIG_MODULES say in your kernel defconfig? Aug 29 11:41:40 ok, /me tries again Aug 29 11:41:55 Crofton: And does your defconfig build anything as a module? Aug 29 11:42:19 I do make omap_osk_5912_defconfig Aug 29 11:42:27 there are a few modules Aug 29 11:42:30 found after compile Aug 29 11:43:12 Crofton: Why are you manually running make? Aug 29 11:43:35 hmm Aug 29 11:44:19 I can change it to oe_runamke, and I think the oldconfig is redundant Aug 29 11:44:41 ok, so this is in the .bb? Aug 29 11:44:52 yeah Aug 29 11:45:10 and where do you see the error - on device? Aug 29 11:45:18 later Aug 29 11:45:25 no, during the build Aug 29 11:45:34 Crofton: During do_rootfs? Aug 29 11:45:57 I think during package modules Aug 29 11:46:46 running the build again Aug 29 11:47:02 I don't understand why it would be running depmod then Aug 29 11:47:22 look in kernel.bbclass Aug 29 11:47:28 and search for arm-depmod Aug 29 11:47:38 well depmod Aug 29 11:48:13 Crofton: Yes, its in a postinst... Aug 29 11:48:14 http://pastebin.ca/153100 Aug 29 11:48:32 I know this... Aug 29 11:48:32 KERNEL_MAJOR_VERSION comes from version.h Aug 29 11:48:51 Crofton: Why would it be running a postinst outside do_rootfs? Aug 29 11:48:57 (of an image) Aug 29 11:48:59 but, when you pull from git, version.h does not exist until after the build Aug 29 11:49:14 It never exists until after a build Aug 29 11:49:20 hmmm Aug 29 11:49:26 It never runs depmod until after a build? Aug 29 11:49:45 even for tarballs? Aug 29 11:49:52 correct Aug 29 11:50:08 Until you've built the modules, you can't run depmod on them... Aug 29 11:51:19 anyway, I do not have any clue why this does not work Aug 29 11:51:28 I'll post the error message in a bit Aug 29 11:51:35 to improve context Aug 29 11:51:50 Crofton: That would be easier Aug 29 11:52:02 and I need to go look at some furniture sizing for my wife for a few minutes Aug 29 11:52:08 will you be around for a bit? Aug 29 11:52:21 I suspect so Aug 29 11:52:22 I really would like to get this resolved soon Aug 29 11:52:36 so I can move on to more exciting problems Aug 29 11:53:32 http://pastebin.ca/153103 Aug 29 11:53:35 there you go Aug 29 11:56:39 Hi! Aug 29 11:57:55 pH5: Recent Angstrom snaphost has issues with XCalibrate (doesn't work) and thus xtscal for me. Did you see anything like that? Aug 29 11:58:16 Crofton: I think you can just ignore that ;-) Aug 29 11:58:45 Crofton: I suspect the kernel build system changed a bit, hence the warning but it all appears to have worked Aug 29 11:59:40 psokolovsky: this snapshot is several months old, right? that should be fixed in .dev Aug 29 12:00:35 pH5: Sorry, I meant a my own build made on ~2006-08-25 Aug 29 12:00:49 pH5: from .dev Aug 29 12:01:35 pH5: Specifically, xtscal cannot detect XCalibrate extension, falls back to /dev/h3600_ts and dies Aug 29 12:02:42 psokolovsky: /dev/h3600_ts in angstrom? weird.. Aug 29 12:03:01 does the bb file need the depmod to succeed? Aug 29 12:03:03 hrw|work: I think that's hardcoded onto Xcalibrate Aug 29 12:03:12 ah Aug 29 12:03:33 hrw|work: Yes, xtscal falls back to it if XCalibrate is not detected (behavior is hardcoded in xtscal, I guess, for 2.4 compatibility) Aug 29 12:03:42 ok Aug 29 12:10:26 psokolovsky: my latest build is older than that. I'm rebuilding just now. Aug 29 12:14:28 koen: Were those tslib changes in a clean tmp? Aug 29 12:14:45 RP: I think so Aug 29 12:15:17 I'm bitbaking world right now, so a clean rebuild will have to wait Aug 29 12:15:22 koen: I'm just wondering if the tslib-conf pkgmap was leftover from a previous build Aug 29 12:15:31 * koen wipes tmp almost daily Aug 29 12:15:41 koen: Can you nuke it manually from staging and rebuild tslib? Aug 29 12:15:47 see if it comes back Aug 29 12:16:40 i've got a new error -> http://rafb.net/paste/results/v5spJr55.html <- any ideas? Aug 29 12:17:10 nptl? Aug 29 12:17:34 weird Aug 29 12:18:08 specifying binutils fixed the problem last time. it's still specified, but glibc-intermediate is failing now Aug 29 12:18:12 * bigd0g scratches his head Aug 29 12:18:24 RP: that fixed it, thanks Aug 29 12:19:05 bigd0g, we are running very similar build env's Aug 29 12:19:11 maybe I should nuke tmp Aug 29 12:20:08 bigd0g, maybe you a specifying something in local.conf? Aug 29 12:20:08 i would say it's good we're running similar dev env's. that way i know i'm not going crazy ;) Aug 29 12:20:30 Crofton: i'm specifying binutils... should i not? Aug 29 12:20:35 try not Aug 29 12:20:41 in local.conf? Aug 29 12:20:44 ok. fixing that, nuking tmp, then trying again Aug 29 12:20:47 koen: When you push that, others are going to see that problem :) Aug 29 12:20:49 yeah, in 'local.conf' Aug 29 12:20:53 1 sec... i'll show you Aug 29 12:20:57 try wo nuking tmp firt Aug 29 12:21:01 save some time Aug 29 12:21:05 koen: Particularly people like me with a 6 month old tmp dir :) Aug 29 12:21:05 ok Aug 29 12:21:23 RP: those people will have been hit by the other changes as well Aug 29 12:21:42 koen: not too badly ;-) Aug 29 12:22:00 preferred binutils and binutils-cross commented out of local.conf ... re-running Aug 29 12:23:15 koen: I just mean its something to watch for when people have builds fail, I don't object to the change Aug 29 12:23:31 RP: maybe we should have a killswitch in OE :) Aug 29 12:23:45 koen: I'd like to make tslib-conf machine independent... Aug 29 12:23:57 RP: yes! Aug 29 12:24:25 RP: a lot of 'machine dependencies' can be cured with a little shell wrapper that reads from /proc Aug 29 12:24:44 koen: agreed Aug 29 12:25:42 In a way I'd like to move a load of them into zaurusd but it would need to go somewhere where others can maintain it too Aug 29 12:25:56 03crofton 07org.oe.dev * r249b7bb8... 10/ (1 packages/linux/linux-omap1_2.6.x+git.bb): linux-omap1_2.6.x+git.bb: Clean up file, eliminate unecessary operations. Aug 29 12:28:07 03hrw 07org.oe.oz354x * rc5b6c8cb... 10/ (1 packages/linux/linux-openzaurus_2.6.17.bb): linux-oz-2.6.17: Add fix to make spectrum_cs work. Upgrade to ASoC 0.11pre12 + some extra fixes which seemingly fixes the poodle sound bugs (static and speaker control issues) Aug 29 12:31:18 pH5: Sorry, was pulled afk. Thanks, wonder if you'll see the same on h2200/hx4700. Before I try to stuff it with printf's, would nice to know it's not my build issue ;-) Aug 29 12:31:38 Crofton: it is getting a new binutils... so maybe that'll fix it ;) Aug 29 12:31:51 heh Aug 29 12:32:04 I pushed some fiddly changes to the kernel file Aug 29 12:32:16 pick them up after you finish the build Aug 29 12:34:04 will do. thanks. Aug 29 12:34:32 Do you have a JTAG for your OSK? Aug 29 12:34:48 jtag ? Aug 29 12:34:59 heh Aug 29 12:35:21 I have flashed a poorly built kernel and wiped u-boot from flash Aug 29 12:35:33 we needed the JTAG guy to reload u-boot Aug 29 12:35:49 I'm not sure if he is around right now Aug 29 12:36:01 oh, i see Aug 29 12:36:13 your board doesn't have the fail-safe uboot back-up? Aug 29 12:36:23 I am going to flash the kernel when I get to my office today Aug 29 12:36:26 apparently not Aug 29 12:36:37 JP2 on my osk. you have the Digital Spectrum one? Aug 29 12:36:42 I didn't have any luck with tthe usb utility Aug 29 12:36:45 yeah Aug 29 12:37:01 JP2 should go from pins 2-3 to pins 1-2 ... that's supposed to reflash uboot Aug 29 12:37:08 (on subsequent power up, obviously) Aug 29 12:37:21 hmmm Aug 29 12:37:35 maybe they changed something Aug 29 12:37:37 i remember reading it in the user manual Aug 29 12:37:49 i might have an earlier rev board than you? mine's 1-year old Aug 29 12:37:53 rev d, i think Aug 29 12:37:56 I tihnk that enables the usb thingy Aug 29 12:38:13 i could have the JP number wrong... 1 sec Aug 29 12:38:16 Hopefully, I do not have to worry about this :) Aug 29 12:41:04 psokolovsky: if you built angstrom--image, you should have strace and gdb installed Aug 29 12:42:42 koen: ok, yet need to update my gdb skills for embedded usage Aug 29 12:43:41 Crofton: it's jp3 ... i just looked at the board Aug 29 12:44:06 pH5: almost forgot: SRCDATE_xcalibrate is define twice in sane-srcdates.inc, likely nees to be fixed (but it didn't change between the dates anyway) Aug 29 12:44:09 Crofton: jp1 is battery v. ac, jp2 is usb, jp3 is a reset for uboot on my board Aug 29 12:45:54 got another error with my build -> http://rafb.net/paste/results/FPJFsl48.html <- any ideas? Aug 29 12:46:23 this time it downloaded binutils 2.17 on its own Aug 29 12:50:03 try nuking tmp and starting over Aug 29 12:50:12 ok Aug 29 12:50:34 once you get a stable system, there is much less tmp nuking Aug 29 12:50:42 * bigd0g hopes so Aug 29 12:51:01 ~hail monotone 0.29 speed Aug 29 12:51:10 * ibot bows down to monotone 0.29 speed and chants, "I'M NOT WORTHY!!" Aug 29 12:51:27 it's been a 2-steps forward, 3-steps back game these past 1.5 weeks Aug 29 12:51:46 bigd0g, I am thinking of nuking tmp, just to make sure I can rebuild :) Aug 29 12:52:00 Crofton: you really know how to install confidence, huh? Aug 29 12:52:06 heh Aug 29 12:52:06 s/install/instill Aug 29 12:52:11 ;) Aug 29 12:52:21 well, I want to make sure the process is repeatable Aug 29 12:52:44 building "bitbake linux-omap1" again. Aug 29 12:52:53 zecke, I think xerces-c won't build on a machine that already has xerces-c installed Aug 29 12:52:54 yeah, no it's definitely something that needs to be accounted for Aug 29 12:53:34 usually I try to do big rebuilds over night Aug 29 12:53:57 with my luck, it would fail just after i left it Aug 29 12:54:19 on kind of a tight schedule, anyway. i have 10-weeks to get this done... Aug 29 12:54:32 ...and i've been wasting time trying to get this working Aug 29 12:54:52 "wasting" is relative, i suppose. Aug 29 12:54:56 yeah Aug 29 12:55:17 it failed? Aug 29 12:55:37 i'm just trying a re-build now. it's still re-building the cache. Aug 29 12:55:41 i nuked tmp, as you suggested Aug 29 12:55:42 heh Aug 29 12:57:48 koen: f-b database will change again Aug 29 13:01:04 * koen prepares Aug 29 13:03:37 Crofton: The second I closed firefox and thunderbird, busybox completed Aug 29 13:03:43 (after about an hour) Aug 29 13:05:20 anyone here familiar with the fb_videomode struct from fb.h? Aug 29 13:07:03 heh Aug 29 13:07:34 lovely, xerces-c build really does die if xerces-c is installed on the build host Aug 29 13:07:39 * Crofton hates xerces-c Aug 29 13:09:34 that sucks Aug 29 13:10:10 yeah, short term fix, nuke the xerces-c libs from the host :) Aug 29 13:10:15 haha Aug 29 13:24:02 koen: can you upgrade php-cli on ewi to php5? Aug 29 13:25:07 hrw|work: done Aug 29 13:29:28 pushed Aug 29 13:29:35 koen: rm feeds.db;update.php Aug 29 13:30:42 morning Aug 29 13:30:51 03hrw 07org.oe.dev * rb7799ca5... 10/ (4 files in 3 dirs): Aug 29 13:30:51 feed-browser: feeds has a type now: normal/upgrades. Aug 29 13:30:51 Type is used by updater to update only 'upgrades' feeds: Aug 29 13:30:51 "php update.php upgrades" Aug 29 13:32:40 hrw|work: how can I make update.php update all feeds? Aug 29 13:33:06 koen: php update.php Aug 29 13:33:11 hrw|work: wouldn't it be easier to md5 the Packages.gz file and store that? Aug 29 13:33:30 koen: would be - do it then Aug 29 13:33:41 * koen opens vi Aug 29 13:33:44 koen: I want to speedup adding rather Aug 29 13:33:51 Added 3485 packages in 0 days, 0 hours, 0 minutes and 20 seconds Aug 29 13:38:34 03mickeyl 07org.oe.dev * rb20c3c5a... 10/ (1 packages/uboot/uboot_cvs.bb): uboot cvs fix SRC_URI Aug 29 13:42:22 hmm, how can i disable this "dropping into a shell thing" during patching? Aug 29 13:43:21 mickeyl: by bribing kergoth Aug 29 13:43:31 what is the difference between armeb-linux and arm-linux in /site? Aug 29 13:43:39 03mickeyl 07org.oe.dev * r4569ccd3... 10/ (5 files in 2 dirs): Aug 29 13:43:39 preferred-versions.inc: fix wrong preference for pango. Aug 29 13:43:39 Guys, when you remove an old version of a package, make sure that Aug 29 13:43:39 a) no recipe directly depends on this version and Aug 29 13:43:39 b) no configuration sets PREFERRED_VERSION to this version. Aug 29 13:43:41 ah. it should be customizable. while being cool, it's a bit annoying i think Aug 29 13:43:44 Crofton: endinanned Aug 29 13:43:50 ok Aug 29 13:44:23 does tosa currently use tslib? Aug 29 13:44:32 mikearthur: all zaurus machines use tslib Aug 29 13:45:01 hrw|work: want me to pm or talk here about tosa status? Aug 29 13:45:14 there's exactly 8000 revisions on org.oe.dev now :) Aug 29 13:45:22 yay! Aug 29 13:45:26 onto the 10k :) Aug 29 13:45:37 ArjanS: 8192 would be nicer Aug 29 13:45:42 mikearthur: talk Aug 29 13:46:12 hrw|work: it would indeed, but I guess there are more people who like 8k :) Aug 29 13:46:23 hrw|work: basically I've massively reduced the amount of glitches and the magnitude of them Aug 29 13:46:29 but they still pop up occasionally Aug 29 13:46:59 mikearthur: got patch ready to add? Aug 29 13:47:03 mikearthur: improvement! Aug 29 13:47:05 almost read Aug 29 13:47:06 ready Aug 29 13:47:14 sharp did some averaging in the kernel Aug 29 13:47:24 but I hear from Liam that tslib is used for that jazz now Aug 29 13:48:13 wheee Aug 29 13:48:23 koen: that to me? Aug 29 13:48:30 mikearthur: yes Aug 29 13:48:43 I just realized I'm not the youngest in the pack anymore :) Aug 29 13:48:47 koen: hehe Aug 29 13:48:51 what age are you? Aug 29 13:48:58 23 Aug 29 13:49:04 22 Aug 29 13:49:11 30 Aug 29 13:49:13 I'll upload some before and after pics Aug 29 13:49:15 28 =) Aug 29 13:49:32 boys :) Aug 29 13:49:45 speak up old man we can't hear you ;) Aug 29 13:49:51 heh Aug 29 13:50:12 mickeyl: you do not have to answer to mikearthur... Aug 29 13:50:23 youngsters :) Aug 29 13:50:36 hrw|work: to what question? I'm not paying much attention to this window atm. Aug 29 13:50:42 hrw|work: not till I'm president of the world anyway Aug 29 13:50:52 hrw|work: ah, nm Aug 29 13:51:10 do13: you're even older than I am, right? Aug 29 13:51:19 * lrg is 32 Aug 29 13:51:52 25 Aug 29 13:52:16 mickeyl: I guess one year Aug 29 13:52:44 mickeyl: you are 35? Aug 29 13:53:56 mickeyl: 29 16:03 < mikearthur> speak up old man we can't hear you ;) Aug 29 13:54:02 do13: 34 Aug 29 13:54:07 hrw|work: hah :D Aug 29 13:55:13 mickeyl: right, only one year difference Aug 29 13:55:52 * lrg feels this will probably turn into a drinking game for next FOSDEM Aug 29 13:56:06 lrg: ? Aug 29 13:56:49 hrw|work: drink your age worth of beer, although mikearthur will probably be refused as he is quite youg ;) Aug 29 13:56:56 young Aug 29 13:57:11 NOTE: preferred version 1.0.1 of libxau not available Aug 29 13:57:18 is that a typo ? Aug 29 13:57:22 libxau 0.1.1 is there Aug 29 13:57:23 or Aug 29 13:57:30 lrg envies my youthful good looks Aug 29 13:57:43 X11R7-1-10.1 Aug 29 13:57:51 which one should be in preferred-x-version-preX11R7 ? Aug 29 13:57:54 the 0.1.1 one? Aug 29 13:58:05 mickeyl: I've seen the same today with dbus-0.92 Aug 29 13:58:15 the preferred versions are a bit fucked up Aug 29 13:58:35 i'm going to try with 0.1.1 Aug 29 14:01:50 i'm see more and more reasons to go with a (G)UI toolk for OE Aug 29 14:01:54 i.e. Aug 29 14:01:59 "remove package" Aug 29 14:02:13 "this package is in use by -- do you really want to remove it?" Aug 29 14:02:24 * mickeyl notes Aug 29 14:03:05 mickeyl: what do you think of my web.py suggestion? Aug 29 14:03:14 mickeyl: lack of gui tool is mentioned few times in survey Aug 29 14:03:23 koen: it's a neat idea -- i need to think more about it Aug 29 14:03:28 hrw|work: excellent. we have a major topic there Aug 29 14:03:53 it's not just for users, also for us seasoned guys it would be a relief Aug 29 14:04:32 koen: has it ever been considered to move oe to portage? Aug 29 14:04:37 I've never noticed a problem but I guess poky shields me from the real world :) Aug 29 14:04:44 the dependency stuff is a lot more powerful Aug 29 14:04:53 startgui, load packages-to-build-set.oes, press build button Aug 29 14:04:54 RP: definitly. you have much less packages and stuff Aug 29 14:04:56 mikearthur: OE moved away from portage Aug 29 14:04:59 the problem would be rewriting some of the bb stuff Aug 29 14:05:00 mikearthur: why should we? we came from portage Aug 29 14:05:17 mickeyl: because, in my humble opinion, portage is a better package manager Aug 29 14:05:21 more powerful Aug 29 14:05:24 mikearthur: Improving what we have would be preferred Aug 29 14:05:41 mikearthur: bitbake+oe is not package manager Aug 29 14:05:56 it is as much so as portage is Aug 29 14:06:06 mikearthur: You just volunteered to write down exactly what portage does better so we can copy it ;-) Aug 29 14:06:15 RP: :D Aug 29 14:06:29 mikearthur: I'm not joking! Aug 29 14:06:32 mikearthur: dont laugh, start writing Aug 29 14:06:34 RP: I know Aug 29 14:06:48 mikearthur: would be best to get it before oedem ;D Aug 29 14:07:06 mikearthur: I expect a report in my inbox within the hour ;-) Aug 29 14:07:13 I could add a bodykit + ugly spoiler to bitbake if you want Aug 29 14:07:17 + big shiny rims Aug 29 14:07:28 would that be rice enough? Aug 29 14:08:13 koen: and 255 tyres Aug 29 14:08:29 koen: yes please Aug 29 14:08:36 koen: and -O12312 by default Aug 29 14:08:49 koen: -fgo-fasta-stripes Aug 29 14:08:58 (I don't use Gentoo) Aug 29 14:09:39 * lrg gives mikearthur his daily medication Aug 29 14:11:42 * koen wonders what's taking the parcel service so long Aug 29 14:11:53 koen: terrorists? Aug 29 14:12:01 which strategy do we prefer? Aug 29 14:12:02 according to their track and trace it was only 3km away this morning Aug 29 14:12:11 a) naming foo_0.0+cvs20060320 Aug 29 14:12:20 b) naming foo_cvs + overriding PV within Aug 29 14:12:27 lrg: The poodle speaker bug is fixed in 0.11pre13, static isn't Aug 29 14:12:42 mickeyl: 0.0 is almost always wrong Aug 29 14:12:54 it was an example. answer the question :D Aug 29 14:13:07 I prefer b Aug 29 14:13:11 mickeyl: PV from within is easier to modify I guess Aug 29 14:13:13 b) Aug 29 14:13:16 hmm Aug 29 14:13:18 but that's just me being lazy Aug 29 14:13:22 don't you think mv is just more clean? Aug 29 14:13:24 to upgrade Aug 29 14:13:32 RP: thanks. I think, I'll go ahead and submit. The static will probably be a lot easier to fix from Sept 25th Aug 29 14:13:32 that way you see the date without having to poke inside Aug 29 14:13:41 hmm.. w8 Aug 29 14:13:42 a) is easier to `find` Aug 29 14:13:51 mickeyl: I don't think we should fix the srcdates that way anyway - cvs packages should be s dsitro matter and why we have san-srcdates Aug 29 14:13:52 * XorA wonders what happens on the 25th :-) Aug 29 14:14:01 foo_0.0+cvs20060320 for FixedDate packages, foo_cvs for floating Aug 29 14:14:24 There is no actual need for fixed date packages that I can see... Aug 29 14:14:26 XorA: that's when you get crushed beneath a 4 core workstation Aug 29 14:14:33 RP: I see them Aug 29 14:14:37 They just increase the amount of metadata we have to maintain... Aug 29 14:14:50 RP: look at esound-gpe which is not maintained upstream Aug 29 14:15:06 hrw|work: What about it? Aug 29 14:15:15 RP: i know you're not a friend of locking down packages, but my use case is a bbfile where i have a patch that only is valid for _one_ day. i don't think fiddling with mindate, maxdate makes a sense here, because without this patch the whole purpose of the package is void Aug 29 14:15:16 RP: its cvs with fixeddate Aug 29 14:16:14 mickeyl: and mindate/maxdate would break thanks to timezones Aug 29 14:16:30 koen: Everything should be using GMT iirc Aug 29 14:16:50 mickeyl: That sounds like a very unusual case. In that case you could argue it, but in general I do dislike them Aug 29 14:16:58 RP: i agree Aug 29 14:17:04 hrw|work: I know they exist, I just don't think they should Aug 29 14:17:23 in the case of esound gpe, someone should just make a tarball Aug 29 14:17:32 glibc showed us how badly broken cvs is Aug 29 14:18:13 lrg: Sounds good to me. I think we have all the main issues sorted. What happens on the 25th btw? Aug 29 14:18:26 koen: right - do it then? Aug 29 14:18:42 OE also lack source archive Aug 29 14:19:28 another point for OEDEM Aug 29 14:21:24 RP: I'll have access to a corgi. Aug 29 14:21:36 lrg: ha, of course :) Aug 29 14:21:50 s/ha/ah/ Aug 29 14:23:17 anyone know what the size of a long double is on an arm? Aug 29 14:23:34 12 parsecs Aug 29 14:23:54 lrg: want poodle? Aug 29 14:24:00 Crofton: ac_cv_sizeof_long_double=${ac_cv_sizeof_long_double=8} Aug 29 14:24:03 that is the sort of answer I would expect from microsoft ... Aug 29 14:24:08 yeah Aug 29 14:24:10 Crofton: from site/arm-linux Aug 29 14:24:16 I am wondering if this is correct :) Aug 29 14:25:23 I need to make a test program Aug 29 14:25:59 * koen is thinking about trying to follow http://venge.net/monotone/wiki/DaggyFixes Aug 29 14:28:43 * koen is pretty happy with the gruesome reply-to munging the OE list is using Aug 29 14:29:08 hrw|work: not atm thanks, the static issue is the same on corgi (I'll have access to one on the 25th Sept), so fixing on corgi will also fix on poodle (I hope). Thanks Aug 29 14:29:45 lrg: we got another one offered but noone wants it Aug 29 14:31:17 oh, okay then. This will be helpfull. thanks :) Aug 29 14:33:11 any progress on the collie SD front? Aug 29 14:34:03 koen: slowly Aug 29 14:37:07 hrw|work: will I send you my address ?, we should be finished with the tosa soon so we can send it to someone who wants to work on it's othe features. Aug 29 14:37:50 lrg: hrw@openembedded.org is ok Aug 29 14:38:05 hrw|work: ok Aug 29 14:41:28 do13_: that's good news Aug 29 14:43:05 lrg: s/corgi/poodle Aug 29 14:44:11 hrw|work: oops my mistake. poodle is good :) Aug 29 14:47:40 lrg: sent Aug 29 14:48:11 hrw|work: that was quick ! Aug 29 14:48:22 lrg: mail sent Aug 29 14:48:33 hrw|work and others: http://img79.imageshack.us/img79/5600/wm97xxlcdnoisepatchvp5.png Aug 29 14:48:39 hrw|work: many thanks Aug 29 14:48:41 thats the before and after shots of the patch Aug 29 14:48:52 mikearthur: awesome Aug 29 14:48:53 mikearthur: cheers Aug 29 14:48:54 drawing straight(ish) lines in gpe-sketchbook Aug 29 14:49:53 that png would look cute in the releasenotes Aug 29 14:49:53 ~hail mikearthur for getting tosa touch in usable state Aug 29 14:50:01 :D Aug 29 14:50:02 * ibot bows down to mikearthur for getting tosa touch in usable state and chants, "I'M NOT WORTHY!!" Aug 29 14:50:08 mikearthur: wow now thats a result Aug 29 14:51:20 that's a mighty fine improvement Aug 29 14:51:37 :) Aug 29 14:52:16 bigd0g, did the build finish? Aug 29 14:52:17 mikearthur: now you have two days to fix spikes in corgi audio :-) Aug 29 14:52:37 Crofton: still compiling... on gcc-cross-4.1.1-r6 Aug 29 14:52:41 XorA: I could remove all the spikes if you like Aug 29 14:52:43 ok Aug 29 14:52:45 turn the volume down Aug 29 14:52:47 fixed Aug 29 14:52:47 :D Aug 29 14:53:00 Crofton: seems to have gotten by glibc. so... yay? Aug 29 14:53:02 mikearthur: where is a patch? Aug 29 14:53:08 :) Aug 29 14:53:13 ;) Aug 29 14:53:25 hrw|work: you need to pass on series of gradually more enduring tests before you can get it Aug 29 14:53:28 ~lart qt for using internal freetype functons Aug 29 14:53:29 * ibot makes a balloon animal out of qt for using internal freetype functons Aug 29 14:53:30 first one involves a dragon Aug 29 14:53:48 mikearthur: lrg isnt that bad is he? Aug 29 14:53:54 zing! Aug 29 14:53:59 * lrg still has to check wm97xx patch for mistakes Aug 29 14:54:06 mikearthur: or I will send OESF to get it from you by using force Aug 29 14:54:14 OESF = OpenEmbedded Special Forces Aug 29 14:54:17 ~oesf Aug 29 14:54:30 hrw|work: I'm sure do13 and RP will also want to check. Aug 29 14:54:36 [oesf] newer name of ZUG - it provide forum for Zaurus/Ipaq/Simpad/etc users (see zug). It has nothing common with OpenEmbedded (except first two words), or http://www.oesf.org/ Aug 29 14:54:37 lrg: I know Aug 29 14:55:15 * chouimat decides to build stuff for his sl-5000d Aug 29 14:55:20 hrw|work: do you have tosa ? Aug 29 14:56:00 lrg: collie, poodle, tosa, husky Aug 29 14:56:23 lack akita or one of spitz devices Aug 29 14:56:43 hrw|work: wow, then you should also check. mikearthur is known to do some hacks ;) Aug 29 14:57:37 I just wrote the whole fix using inline asm, thats fine, right? Aug 29 14:57:48 he actualy fixed it with #include "wince.h" Aug 29 14:57:55 hahaha XorA :D Aug 29 14:58:51 XorA: M$ will be forced to send ibot to kill you for releasing trade secrets :) Aug 29 15:00:57 hrw|work: btw, what audio does collie have ? Aug 29 15:01:14 lrg: moment.. Aug 29 15:01:18 np Aug 29 15:01:37 collie_tc35143af Aug 29 15:01:40 iirc Aug 29 15:01:49 or I will check in one doc which I do not have Aug 29 15:02:00 hrw|work: ah, is it pxa2xx based ? Aug 29 15:02:10 sa11000 Aug 29 15:02:48 hrw|work, koen: thanks Aug 29 15:03:13 lrg: sa1100 and tc35143af for ts and audio Aug 29 15:03:33 koen: dbus / gconf-dbus is somewhat inconsistent Aug 29 15:03:45 sane-srcdates says we use Aug 29 15:03:52 SRCDATE_gconf-dbus ?= "20060719" Aug 29 15:04:12 which says Aug 29 15:04:13 | configure: error: Package requirements (dbus-glib-1 >= 0.60) were not met: Aug 29 15:04:20 however we say Aug 29 15:04:40 PREFERRED_VERSION_dbus ?= "0.23.4" Aug 29 15:04:47 in all preferred-gpe-verions Aug 29 15:04:58 mickeyl: Fwiw, I still can't get dbus-glib-native to build. Fresh angstrom /tmp. Will add bug Aug 29 15:05:05 how to fix that? lower gconf-dbus srcdate or increase dbus preferred version? Aug 29 15:05:07 mickeyl: I only use dbus 0.9x Aug 29 15:05:09 lrg: ok Aug 29 15:05:47 mickeyl: IMO most gpe-preferred- should get dropped Aug 29 15:06:21 koen: ok. in other words... which dbus version should go w/ gconf-dbus from 0719? 0.92 or 0.62 ? Aug 29 15:06:37 cbrake: hi Aug 29 15:06:39 0.92, since that is in API freeze Aug 29 15:06:45 excellent Aug 29 15:06:45 thanks Aug 29 15:07:43 mickeyl: in org.oe.core I'd like to avoid multiple versions of packages Aug 29 15:08:18 koen: then you will have to keep org.oe.DISTRO branches Aug 29 15:08:41 hrw|work: I don't mind org.oe.distro branches Aug 29 15:09:07 I just think we need some guidelines for branches in the org.oe namespace Aug 29 15:09:22 Crofton: any luck with the osk board and uboot? Aug 29 15:10:12 I have tried the u-boot I built in OE Aug 29 15:10:20 I haven't tried the kernel yet Aug 29 15:10:26 your build still runing? Aug 29 15:10:42 yes. i'm quite limited by this machine Aug 29 15:11:06 only an athlonxp (mobile) @ 1.5-ghz, 512-mb ddr266 Aug 29 15:11:14 what target are you building? Aug 29 15:11:20 linux-omap1 Aug 29 15:11:39 current progress: package glibc-2.4-r10: task do_compile: started Aug 29 15:13:34 bigd0g: similar to my machine here: axp 1.4, 768 ddr400 Aug 29 15:14:11 hrw|work: i miss my desktop... a64 3500+ @ 2.4 ghz, 1-gb ddr400... it's 3500 miles away Aug 29 15:14:35 my desktop is not such powerfull Aug 29 15:15:02 my desktop's for gaming, primarily Aug 29 15:15:07 my not Aug 29 15:15:14 so... it runs windows *shudder* Aug 29 15:15:29 mx440 and gaming... or mga400 and gaming.. Aug 29 15:15:48 i've got a 6800gt overclocked to ultra at home. nothing too impressive Aug 29 15:16:08 this lappy just has radeon igp graphics Aug 29 15:16:35 crap Aug 29 15:16:43 is berlios down *again*? Aug 29 15:19:05 koen: yep Aug 29 15:23:24 RP: style question. something like zaurus-updater.sh or a jtag utility for a certain MACHINE. Where should this go in the new scheme? Aug 29 15:23:42 it's not a distro thing, but rather a machine thing imo Aug 29 15:23:53 think native-utilities Aug 29 15:24:11 mickeyl: Its an IMAGE_DEPENDS Aug 29 15:24:22 mickeyl: zaurus-clamshell.conf already sets that iirc Aug 29 15:25:03 ah i see, thanks. (zaurus-clamshell was just an example) Aug 29 15:25:07 i'm working on a custom thingy atm. Aug 29 15:25:34 mickeyl: Is you look at task-base, there are some variables to add essential things for the machine as well Aug 29 15:25:54 nokia770-init is an example I added to a machine.conf file Aug 29 15:26:51 mickeyl: c7x0 and akita/spitz use 16 bytes header? Aug 29 15:27:06 iirc yea Aug 29 15:27:28 They handle things that need to go into the image. Essential ones are ones required to boot like nokia770-init. We also have others like zaurusd which aren't essential but recommended :) Aug 29 15:27:40 hrw|work: yes Aug 29 15:27:41 EXTRA_IMAGEDEPENDS += ? Aug 29 15:27:47 (vs. IMAGE_DEPENDS) Aug 29 15:27:52 That's the one, sorry Aug 29 15:27:56 np, thanks Aug 29 15:28:29 Thats for tools to actually build the image (or at a push use it) Aug 29 15:28:49 ya, makes sense. Aug 29 15:29:09 like, 'sending/booting' over usb or so would qualify as well Aug 29 15:29:21 yes Aug 29 15:42:29 dbus-0.92 no longer provides dbus-glib, is that correct? Aug 29 15:42:29 someone tried to flash clamshell without 16 bytes header? Aug 29 15:42:31 it got split? Aug 29 15:42:34 mickeyl: it got Aug 29 15:42:41 thanks Aug 29 15:42:55 i think someone tried Aug 29 15:42:58 it hang Aug 29 15:43:02 because kernel was corrupt Aug 29 15:43:05 missed 16 bytes in front :D Aug 29 15:43:25 that shitload of dbus versions is killing me Aug 29 15:43:26 * mickeyl sighs Aug 29 15:43:51 mickeyl: kernel does not have header - rootfs has Aug 29 15:43:58 oops Aug 29 15:44:04 darn, it's been too long Aug 29 15:44:27 I think that I will have to try to flash without those 16 bytes Aug 29 15:45:03 heh. any specific reason besides being curious? Aug 29 15:45:21 Do I need to do anything about this message: Aug 29 15:45:22 updater.sh for Z machines have very small changes between them so it can be united Aug 29 15:45:22 NOTE: Multiple libraries (libomniDynamic4.so.0, libomniCodeSets4.so.0, libomnithread.so.3, libCOS4.so.0, libCOSDynamic4.so.0, libomniORB4.so.0) found and LEAD_SONAME not defined Aug 29 15:45:26 Crofton: no Aug 29 15:45:30 thanks Aug 29 15:45:32 hrw|work: cool Aug 29 15:45:35 that'd rock Aug 29 15:46:14 Crofton: for me it means: I'm told to put few libs in one package instead of package per lib and I do not know how to call resulting package so I will call it PN Aug 29 15:46:34 mickeyl: am I right with this meaning? Aug 29 15:46:52 ok Aug 29 15:46:59 mickeyl: I only lack 2/5 platforms for testing updater Aug 29 15:47:06 Crofton: hey... i just got a similar message a few mins ago Aug 29 15:47:14 heh Aug 29 15:47:17 uh Aug 29 15:47:21 wtf happened? Aug 29 15:47:22 different package I bet :) Aug 29 15:47:25 * bigd0g pokes his tab key Aug 29 15:47:36 probably Aug 29 15:48:09 oh, ok... doesn't look like it printed out... just printed every name beginning with C on my screen Aug 29 15:48:20 heh Aug 29 15:48:35 argh, dbus-glib-native is broken because it has two DEPENDS lines and dbus-native added to the first (wrong) one. Aug 29 15:49:37 pH5: I was about to submit a bug for this. You beat me to it :) Aug 29 15:49:38 Crofton: probably was diff pack, but can't scroll far enough up. only have 1-mb framebuffer set Aug 29 15:50:49 pH5: I take it this will fix the problem whereby the headers are not being copied ? Aug 29 15:50:52 mickeyl: that would need testers with cxx00 machines... Aug 29 15:51:00 Crofton: yes! it's using 'git' to get uboot Aug 29 15:51:02 * bigd0g dances Aug 29 15:51:58 mickeyl: http://www.spacezone.de/zaurus/articles/update.sh.txt Aug 29 15:54:11 mickeyl: do you remember why tosa/poodle does not reboot after flashing? Aug 29 15:55:39 lrg: as I understand the problem was that bitbake tried to build dbus-glib-native without having dbus-native staged. this should be fixed now. Aug 29 15:55:53 03crofton 07org.oe.dev * reca6b706... 10/ (4 files in 3 dirs): Aug 29 15:55:53 omniorb-native_4.0.7.bb, omniorb_4.0.7.bb, long_double.patch : Aug 29 15:55:53 * Add bb files for omniorb-4.0.7 Aug 29 15:55:53 * Patch long_double.patch allows compilation when sizeof(long double) Aug 29 15:55:53 * is something other than 12 or 16. Aug 29 15:56:31 Crofton: can you look at omni* bugs in bugtracker? Aug 29 15:56:42 yeah Aug 29 15:57:08 great - they are very old Aug 29 15:57:28 um, how do I find them? Aug 29 15:57:42 03tmbinc 07org.oe.dreambox * r23bd8fce... 10/ (1 packages/linux/linux-dm600pvr.bb): linux-dm600pvr: update kernel to r2 in order to fix IDE Aug 29 15:57:48 03tmbinc 07org.oe.dreambox * r8b0970f1... 10/ (1 packages/dreambox/dreambox-secondstage.bb): dreambox-secondstage: update to 45 to fix IDE reset Aug 29 15:57:53 03pH5 07org.oe.dev * r1f105f36... 10/ (1 packages/dbus/dbus-glib-native_0.71.bb): dbus-glib-native-0.71: fix bogus DEPENDS Aug 29 15:58:10 pH5: thanks Aug 29 15:58:26 Crofton: search for 'omni'? Aug 29 15:58:39 no bugs found :) Aug 29 15:59:02 http://bugs.openembedded.org/show_bug.cgi?id=90 Aug 29 15:59:23 Crofton: probably your bug even ;) Aug 29 15:59:30 yeah Aug 29 15:59:41 you ugly dev! Aug 29 15:59:58 how could you forget own bug... ;D Aug 29 15:59:58 I am avoiding needing omnievents at the moment :) Aug 29 16:00:47 ;) Aug 29 16:00:48 cu Aug 29 16:07:12 03justinp 07org.oe.dev * r4d391acc... 10/ (1 contrib/mtnpatch.py): mtnpatch.py: Remove unneeded script (mtn now has pluck) Aug 29 16:08:10 bitbake gpe-image keeps trying to build packages from gpe-task-sectest and gpe-task-desktopapps, and I don't know how to stop it :( Aug 29 16:08:34 pH5: move those tasks to seperate .bb Aug 29 16:09:03 pH5: still fails. Will create bug Aug 29 16:09:43 pH5: bitbake needs to build everything in (R)DEPENDS/RECOMMENDS to handle shlibs renaming Aug 29 16:10:23 lrg: please do. I just built it from a clean tmp this way. Aug 29 16:10:26 How often are the snapshots updated? Aug 29 16:11:12 koen: thanks! Aug 29 16:11:26 pH5: I was too lazy to do that Aug 29 16:12:59 Someone really needs to fix debian.bbclass... Aug 29 16:13:35 * RP is tempted to fix with rm... Aug 29 16:15:12 pH5: bug #1371 Aug 29 16:16:03 lrg: thanks. Aug 29 16:16:36 hmmm Aug 29 16:16:46 Hey all:) Aug 29 16:16:48 this new toy requires a strange battery Aug 29 16:17:01 * koen orders a 2CR5 battery Aug 29 16:17:04 hey TruPilot Aug 29 16:17:41 Does anyone here connect to the Monotone repository using an HTTP CONNECT proxy? Aug 29 16:18:02 TruPilot: it doesn't use http Aug 29 16:18:06 lrg: hey, that's exactly the same error I got before. can you confirm that dbus-native is not staged at this point? Aug 29 16:18:43 koen - I know - that's the problem - I want to make it connect via a proxy that allows HTTP CONNECT Aug 29 16:18:59 TruPilot: do you happen to have access to a socks proxy as well? Aug 29 16:19:27 I can probably set up a delegate proxy to go via the regular proxy... Aug 29 16:19:36 But I have never done that either. :/ Aug 29 16:19:52 anyway, if you've got a socks proxy, try tsocks, works like a charm Aug 29 16:19:56 lrg: if this is the case, does the dbus-glib-native contain dbus-native in it's DEPENDS? Aug 29 16:20:06 useses LD_PRELOAD to route all tcp traffic through socks Aug 29 16:20:09 -es Aug 29 16:20:51 * koen wonders how far this OE build will get on OSX Aug 29 16:21:17 * TruPilot really wishes that OE/BB sources were available via Subversion... Aug 29 16:21:50 (including devel) Aug 29 16:23:19 TruPilot: hmm... they are not? Aug 29 16:23:36 a+ Aug 29 16:23:38 pH5: the missing dbus headers are in the staging dir (is this what you mean) and their are no dbus-native stamps Aug 29 16:23:40 I don't know - I'm not the OE/BB dev at my co. Aug 29 16:23:56 http://developer.berlios.de/svn/?group_id=2701 Aug 29 16:24:03 bitbake has been in svn since ages Aug 29 16:24:11 Right, but what about OE? Aug 29 16:24:22 you can get a snapshot, right? Aug 29 16:24:22 http://www.openembedded.org/snapshots/OE.mtn.bz2 Aug 29 16:24:32 Is that a daily snapshot of the dev tree? Aug 29 16:25:12 it *is* the dev tree Aug 29 16:25:16 pH5: dbus-glib-native depends on dbus-native in my sys. Aug 29 16:25:23 Right - but how often is the snapshot taken? Aug 29 16:25:40 every morning at 6:00 GMT Aug 29 16:25:49 lrg: that's what I mean. so I'd expect dbus-native to be built and staged. Aug 29 16:25:56 That's what I needed to know - thanks a bunch :) Aug 29 16:25:57 and it isn't a snapshot, it's a copy of the database Aug 29 16:26:05 (the dir name is a bit misleading) Aug 29 16:26:44 pH5: I'll do a little more checking..... Aug 29 16:27:17 l8r all - and thanks again koen :) Aug 29 16:29:38 ~hail do13 for helping mikearthur with tosa touch screen Aug 29 16:29:41 * ibot bows down to do13 for helping mikearthur with tosa touch screen and chants, "I'M NOT WORTHY!!" Aug 29 16:32:31 NOTE: build 200608291406: completed Aug 29 16:32:35 w00t! w00t! w00t! Aug 29 16:32:56 Attempted builds: 23 Aug 29 16:32:59 well, better lat than never Aug 29 16:33:11 too bad Crofton signed off before i could tell him Aug 29 16:33:20 that's all for now, folks. see you tomorrow... Aug 29 16:36:52 lrg: you around? Aug 29 16:37:12 * bigd0g runs to catch the bus Aug 29 16:37:23 mikearthur: about to go in 10 mins Aug 29 16:37:34 lrg: hows the patch? Aug 29 16:38:38 mikearthur: no time to check it yet, Will do on the train Aug 29 16:38:52 lrg: no worries Aug 29 16:38:56 lrg: slaters Aug 29 16:39:16 mikearthur: cu Aug 29 16:58:40 koen are you getting anywhere building on osx? Aug 29 17:29:35 yay! Aug 29 17:29:36 NOTE: package gcc-cross-initial-4.1.1: completed Aug 29 17:29:46 built on OSX :) Aug 29 17:31:10 I am having trouble with some of my packages failing Aug 29 17:31:21 hmm Aug 29 17:31:27 * koen wonders how qemu will run on osx Aug 29 17:31:28 because they have libs present in /usr/local/lib Aug 29 17:34:02 koen just get Q ;) Aug 29 17:34:51 reboot Aug 29 17:35:25 ah, NOTE: package qemu-native-0.8.2-r0: task do_compile: started Aug 29 17:35:51 at times like this I wish my powerbook had a 4xG5 Aug 29 17:42:32 yes Aug 29 17:42:40 dvb-usb: schedule remote query interval to 200 msecs. Aug 29 17:42:42 dvb-usb: MSI Mega Sky 580 DVB-T USB2.0 successfully initialized and connected. Aug 29 17:43:27 woglinde: is that a good sign? Aug 29 17:43:45 koen hehe Aug 29 17:43:49 now I can watch tv Aug 29 17:43:58 I think Aug 29 17:45:33 now I must push "the fix it" button Aug 29 17:46:07 oh my Aug 29 17:46:37 it booted Aug 29 17:54:40 ...and then it bursted into fire Aug 29 17:54:43 hm Aug 29 17:54:50 looks like it dont work Aug 29 17:55:36 ah Aug 29 17:55:37 yes Aug 29 17:55:40 scan Aug 29 18:01:44 yes Aug 29 18:01:46 tagesschau Aug 29 18:02:07 hmmm Aug 29 18:02:09 * woglinde is happy Aug 29 18:02:14 the feared german cipher Aug 29 18:02:23 *g* Aug 29 18:02:33 koen, no fire Aug 29 18:02:43 need a root file system though :) Aug 29 18:09:58 nice only 20% cpu Aug 29 18:12:00 woglinde: I have a 100% cpu in my computer Aug 29 18:12:09 why are you missing 80% of yours? Aug 29 18:14:04 mallum: Hi! Is http://gitweb.freedesktop.org/?p=xorg/lib/libXCalibrate.git;a=summary now authoritive location for libXCalibrate? Aug 29 18:15:27 psokolovsky_: hmm, probably, but it really needs pb_'s blessing Aug 29 18:16:28 mallum: ok, won't add new .bb for it yet then ;-). Also, sent an email cc: you regarding xcalibrate ext issues with latest kdrive. Aug 29 18:18:13 03pH5 07org.oe.dev * r773964e4... 10/ (1 packages/xorg-xserver/xserver-kdrive_git.bb): xserver-kdrive-git: apply the xcalibrate patch from X11R7.1 Aug 29 18:19:44 psokolovsky_: I have to push xcalibrate back into git xorg Aug 29 18:20:19 mallum: cool! Aug 29 18:22:08 mallum: I'm compiling glibc on OSX now using OE Aug 29 18:43:12 * mrz80 is away: Metaphysically I'm still here; physically, I'm elsewhere Aug 29 18:43:58 koen: arg you are still using that apple foo :( Aug 29 18:44:14 koen: hasn't the novelty worn off yet ? Aug 29 18:44:28 hmmm Aug 29 18:44:48 I flashed my boostrap-image into my OSK Aug 29 18:44:55 and it hangs at Aug 29 18:45:04 mallum: I actually tried to migrate to gnome this week Aug 29 18:45:13 Starting the hotplug events dispatcher udevd Aug 29 18:45:17 mallum: I tried importing my photos into fspot Aug 29 18:45:50 mallum: which isn't really possible because the fileselector causes it to hang Aug 29 18:46:15 mallum: put after working around that, I noticed that f-spot doesn't actually do anything usefull Aug 29 18:47:36 so write a good photomanagement app and I'll swith back :) Aug 29 18:49:34 koen, any insight into why my new file system leads to a hang? Aug 29 18:49:59 * koen polishes his crystal ball Aug 29 18:50:05 heh Aug 29 18:50:33 Hangs at Starting the hotplu events dispacther udevd Aug 29 18:50:50 broad sweeping statements might be helpful Aug 29 18:50:52 'hangs' or 'takes forever'? Aug 29 18:51:07 well Aug 29 18:51:21 it hasn't done anything since it stopped there Aug 29 18:51:31 can you ctrl-c? Aug 29 18:51:35 so we will say "hangs" unless something happens Aug 29 18:51:49 I've seen that issue popup in the past Aug 29 18:51:50 no response to ctril-c Aug 29 18:51:55 try rebooting Aug 29 18:52:25 reboting Aug 29 18:52:34 this is for omap5912osk machine Aug 29 18:52:45 I could easily have bad stuff going on Aug 29 18:53:00 it is still taking forever Aug 29 18:53:17 none responsive to ctrl-c (via serial port) Aug 29 18:53:27 ping OK, ssh connection refused Aug 29 18:54:18 hmmm Aug 29 18:56:01 the messgae suggests it has mounted root ok? Aug 29 18:56:20 I think so Aug 29 19:01:28 hmm Aug 29 19:01:41 the message comes from init.d/udev Aug 29 19:02:11 you could try booting into sh directly Aug 29 19:02:20 'init=/bin/busybox sh' Aug 29 19:02:21 hmmm Aug 29 19:02:37 in uboot? Aug 29 19:02:49 yes Aug 29 19:03:02 dunno how you tell uboot how to do that, though Aug 29 19:03:07 never used it Aug 29 19:03:18 bootargs I bet ... Aug 29 19:04:16 sounds like a good bet Aug 29 19:05:52 will init=/bin/sh work? Aug 29 19:06:30 I will attempt to answer my own question ... Aug 29 19:06:39 yes! Aug 29 19:08:04 it does? Aug 29 19:08:11 busybox used to hate that Aug 29 19:09:01 well, since it takes forever to compile bb now, maybe it works :) Aug 29 19:11:26 koen - glad you got builds working on osx - i have a pretty good start on ''build on osx'' on the wiki - would you mind reading and commenting on it? (my writeup is mostly theoretical so far except for the preparation section) Aug 29 19:11:48 it eh... Aug 29 19:11:51 ehm.. Aug 29 19:11:54 stopped on glibc Aug 29 19:12:03 I think zecke had a patch Aug 29 19:12:40 yeah - i believe so - thought he had committed that already - you couldnt get it yet? Aug 29 19:13:19 maybe I'm hitting a ppc problem Aug 29 19:13:54 isnt zecke on ppc too? (im on g4 dual 1ghz ppc) Aug 29 19:14:25 he has a macbook Aug 29 19:14:40 oh so hes mactel Aug 29 19:16:02 well im still interested in getting your feedback on the wiki page - http://www.openembedded.org/buildonosx Aug 29 19:18:30 If I run udevd from the command line, it hangs Aug 29 19:19:03 can't interupt it with ctrl-c Aug 29 19:21:00 * chouimat is away: BEER BEER BEER BEER BEER BEER BURP! Aug 29 19:21:36 Crofton: that doesn't work with init=/bin/sh. run udevd & instead Aug 29 19:21:50 ok Aug 29 19:23:12 that started Aug 29 19:25:18 although I do not know if it is doing anything .. Aug 29 19:32:43 has anyone built bluez-utils lately? I just tried building it and get errors. Aug 29 19:37:30 Gerrath: which flavor? Aug 29 19:37:53 hms Aug 29 19:37:58 the driver isnt perfect Aug 29 19:38:03 oops after unplug and plug Aug 29 19:38:58 * pH5 starts a bluez-utils 3.4 build Aug 29 19:39:24 pH5 I'm trying a rebuild just to make sure.. Aug 29 19:40:37 pH5, I originally just tried bitbake bluez-utils and it crashes (errors during compile) but if I do bitbake bluez-utils-nodbus it works fine.. Aug 29 19:42:23 pH5, also bitbake bluez-utils-dbus has compile errors. Aug 29 19:45:32 I just built bluez-utils w/o problem. Can you post the error log somewhere? Aug 29 19:45:58 03pH5 07org.oe.dev * rb0f19e03... 10/ (3 files in 2 dirs): bluez-libs, bluez-utils: add 3.4 Aug 29 19:46:37 pH5, sure.. Aug 29 19:48:58 pH5, http://pastebin.ca/153720 Aug 29 19:50:25 ah, but that's not an error in bluez-utils, but an error in glib-2.0 (which is needed for dbus which is needed for bluez-utils(-dbus)) Aug 29 19:50:55 "#error GNU libiconv not in use but included iconv.h is from libiconv" Aug 29 19:55:17 pH5, Yes, sorry. Aug 29 19:56:37 hmmm my tmpfs file systems do not mount Aug 29 19:57:00 crofton hm Aug 29 19:57:09 ah, the mount -a problem again :( Aug 29 19:57:44 pH5, I know what dbus is (app. comm bus) but how is it used in the Linux Bluetooth system? Aug 29 19:58:27 koen: oedem page updated Aug 29 19:58:42 Gerrath: the bluetooth daemon has a dbus interface now to communicate with gui tools Aug 29 19:58:43 koen: drupal lack good diffs - mediawiki like preferred Aug 29 19:59:05 hrw: patches accepted Aug 29 19:59:19 ~/website/modules/diff.module iirc Aug 29 19:59:33 pH5, ok that makes sense and is what I thought.. so if you are not really using gui tools then the dbus it not needed. Aug 29 20:00:05 koen: I remind you that I do not use ltg accounts Aug 29 20:00:24 then don't complain Aug 29 20:01:18 heh Aug 29 20:02:24 "Debian - when you have better things to do than fixing systems" Aug 29 20:02:27 ;D Aug 29 20:02:42 koen: read links which I added to oedem page Aug 29 20:05:35 cu Aug 29 20:08:03 nite hrw Aug 29 20:15:26 the tmpfs in etc/fstab won't mount Aug 29 20:15:30 Is that a problem Aug 29 20:16:12 hm Aug 29 20:16:15 depends Aug 29 20:16:22 logread complains it can't find the circular buffer Aug 29 20:16:45 /dev is tmpfs on hosts Aug 29 20:16:57 and /dev/shm Aug 29 20:17:03 and my /tmp Aug 29 20:17:17 but I didnt see such an error Aug 29 20:17:34 which machine do you use? Aug 29 20:18:14 machine.conf Aug 29 20:18:52 hm I spoked about my desktop Aug 29 20:18:59 not the embedded devices Aug 29 20:19:19 heh Aug 29 20:19:48 to illustrate why I isnt always an error not to have tmpfs Aug 29 20:19:55 I = it Aug 29 20:19:57 hrms Aug 29 20:21:22 I am just fishing Aug 29 20:22:40 http://static.flickr.com/60/228542395_2b1b6fcb78.jpg Aug 29 20:38:51 hmm I am pulling in hotplug and udev Aug 29 20:39:06 how do I prevent hotplug from coming in and using udev? Aug 29 20:39:45 or do I need both? Aug 29 20:54:59 koen, psokolovsky_: http://gitweb.freedesktop.org/?p=xorg/xserver.git;a=commit;h=5ddbf4bcd46fe0d3d682668c2748c712fea410ae Aug 29 20:55:36 mallum: yay! Aug 29 20:57:22 mallum: Thanks! Aug 29 20:58:41 Xext/xcalibrate.c | 262 ++++++++++++++++++++++++++++++++++++++++ Aug 29 21:00:09 * koen wonders about the other patches in http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/xorg-xserver/xserver-kdrive Aug 29 21:04:08 no joy Aug 29 21:04:17 tomorrow is another day Aug 29 21:04:42 nite crofton Aug 29 21:04:48 and you need only udev Aug 29 21:04:50 good night Aug 29 21:05:00 yeah, but still having problems Aug 29 21:05:08 need to go though this carefully tomorrow Aug 29 21:06:20 need to get it booting ok tomorrow, bigdog will want to do the same thing very soon I expect Aug 29 21:12:29 * mrz80 is back (gone 02:29:18) Aug 29 21:14:30 mallum: any word on a new kdrive release to celebrate the tslib add-on? Aug 29 21:15:19 koen: I hope it will be part of next xorg release Aug 29 21:15:36 meh Aug 29 21:15:49 if I had a dollar everytime I heard that :S Aug 29 21:16:39 on the bright side, keithp hasn't broken it again, so let's be happy Aug 29 21:17:47 koen: current xorg has kdrive though right Aug 29 21:18:13 yes, we should be pulling from that Aug 29 21:18:36 but I dislike being dependant on a git checkout to get rid of major bugs in the release Aug 29 21:19:34 koen: yeah but most bugs should now be fixed ? Aug 29 21:19:51 koen: too many people use xephyr nowadays Aug 29 21:20:09 mallum: the git version works pretty well Aug 29 21:20:14 koen: that makes kdrive a bit more generally needed I think Aug 29 21:21:47 03koen 07org.oe.dev * r43921a86... 10/ (1 packages/xorg-xserver/xserver-kdrive_git.bb): xserver-kdrive git: remove tslib and xcalibrate patches that have been applied upstream (yay!) and enable tslib in EXTRA_OECONF Aug 29 21:22:25 mallum: it just needs proper evdev and hotplug suppport :) Aug 29 21:23:10 koen: input-otplug branch Aug 29 21:23:59 psokolovsky_: write that down ;) Aug 29 21:24:36 and have a spare h :) (input-hotplug) Aug 29 21:24:45 * koen checks Aug 29 21:24:51 I'm out of those Aug 29 21:25:46 otplug is the east london successor to hotplug Aug 29 21:26:01 its what all the kids are using nowadays Aug 29 21:43:16 mallum: do you know why the epson kdrive server has dropped out of SUBDIR? Aug 29 21:43:23 mallum: it builds fine Aug 29 21:44:54 koen: not a clue Im afraid Aug 29 21:45:07 koen: I know there was a time when there were some issues with it Aug 29 21:45:15 yeah Aug 29 21:45:20 we had patches in OE Aug 29 21:45:28 but I just added it to subdirs and it built Aug 29 21:45:41 Packaged contents of xserver-kdrive-epson into /data/build/koen/OE/build/tmp/angstrom/deploy/ipk/xserver-kdrive-epson_1.1.0+git20060829-r3_armv5te.ipk Aug 29 21:46:15 wierd Aug 29 21:46:23 yep Aug 29 21:46:29 I think there are like 3 people on the planet with that h/w though Aug 29 21:46:45 mallum: everyone with a netbook Aug 29 21:46:53 3 people ;) Aug 29 21:47:07 you, florian and wookey Aug 29 21:47:17 phillipe and pb__ Aug 29 21:47:30 and the people at hard-data.com Aug 29 21:47:42 still not much :) Aug 29 21:48:30 pb could enable it, he must still have commit privs Aug 29 21:48:40 Im too scared though Aug 29 21:49:04 aargh Aug 29 21:49:12 we're missing xcalibrateproto in .dev Aug 29 21:49:22 that's it Aug 29 21:49:27 * koen goes to sleep Aug 29 22:59:17 hell Aug 29 22:59:19 o Aug 29 22:59:24 stupid lag Aug 29 22:59:51 anyone happen to know of any cf 802.11 cards that do not protrude? Aug 29 23:01:16 hm? Aug 29 23:01:21 I have orinocos Aug 30 01:25:15 * chouimat is back. Aug 30 02:43:04 is there a web browser that supports javascript in the metadata? **** ENDING LOGGING AT Wed Aug 30 02:59:57 2006