**** BEGIN LOGGING AT Sat Jun 04 23:59:56 2005 Jun 05 00:11:59 good morning all Jun 05 00:21:21 It's stange I'm looking to package ipatience for openzaurus, i'ts in familiar but not in openembedded it seems stange, guess I'll just have track down the source and create my own bb file Jun 05 00:21:54 the ipkg should have a Source: field Jun 05 00:22:54 ahh ok thank koen Jun 05 00:26:59 koen is that in the control file? Jun 05 00:28:25 yes Jun 05 00:28:42 Interesting maintainer for this ipkg Keith Packard , unfortunatly no Source line there:( Jun 05 00:30:18 it looks like they converted a .deb Jun 05 00:30:31 but if it's gpl, they should have a copy of the source Jun 05 00:31:09 I'm trying to compile links with X, I get the error message | checking for X... disabled Jun 05 00:31:09 | configure: error: No graphics drivers found. Jun 05 00:31:29 do I need to install x using oe somehow? Jun 05 00:31:42 yes Jun 05 00:32:03 ok how do I do that bitbake Xorg? Jun 05 00:32:10 but the 'disabled' leads me to believe someone (probably team opie :)) disabled it Jun 05 00:32:20 ok Jun 05 00:32:28 bitbake virtual/xserver should do the job Jun 05 00:32:40 It was origonally configured to just use the framebuffer, but I'm using gpe Jun 05 00:32:55 so I wanted to configure it to use with-x and without-fb Jun 05 00:33:03 thanks koen Jun 05 00:33:22 put the --with-x in the EXTRA_OECONF in the .bb Jun 05 00:33:30 and add 'x11' to the DEPENDS Jun 05 00:33:51 great thanks again!:) Jun 05 00:35:06 that should do the trick, but I haven't tried it yet Jun 05 00:46:13 ok koen I have to source to ipatience it was at Keith's home page Jun 05 00:46:46 I'll have a go at making a .bb file Jun 05 00:47:10 cool, thanks Jun 05 01:00:20 One thing I'm worried about is that the source is named ipackage.tar.gz but i'ts actually a bz2 file so should be ipackage.tar.bz2 Jun 05 01:00:41 heh Jun 05 01:08:48 Damn I've installed x and everything but on building I get checking for X... disabled Jun 05 01:08:48 | configure: error: No graphics drivers found. Jun 05 01:08:48 | FATAL: oe_runconf failed Jun 05 01:08:57 which is the same error as before this is for links Jun 05 01:09:10 I've added x11 to the depends Jun 05 01:09:31 perhaps the conf is cached somewhere hmm Jun 05 01:11:26 and you removed the 2(!) --without-x from EXTRA_OECONF? Jun 05 01:11:41 yes I changed it to --with-x Jun 05 01:12:41 ahh just checked again there was an extra --without x Jun 05 01:12:45 ok I'll try again Jun 05 01:15:00 03koen 07 * r1.3507 10openembedded/packages/links/links_2.1pre14.bb: links_2.1pre14.bb: one --without-x is enough, it really is. Jun 05 01:22:19 CIA-4 :) agreed Jun 05 01:26:47 Well I've got a bit further with links now with /stuff/tmp/cross/lib/gcc/arm-linux/3.4.3/../../../../arm-linux/bin/ld: cannot find -lfl Jun 05 01:27:43 ah, it probably need fltk too in its DEPENDSs Jun 05 01:28:53 ahh ok thank koen Jun 05 01:33:36 I added fltk to the depends and got the same error:( Jun 05 01:34:40 I'll try flnx Jun 05 01:43:18 flnx did not want to compile sigh:( Jun 05 01:52:01 does anyone know what I need to create an install for a new package? Jun 05 01:52:26 my compile of ipatience is fine but fails on packaging Jun 05 01:53:47 you could write your own do_install() Jun 05 01:54:00 or pllay with FILES_${PN} Jun 05 01:55:12 ok thanks koen, I've just found the log and it did not compile anything so I'll have to rethink this Jun 05 01:55:58 does it enter the right directory for compiling? Jun 05 01:56:32 if not you need to set 'S="${WORKDIR}/ipatience-0.2' or something like that Jun 05 01:59:26 ok thanks Jun 05 02:43:06 anyone alive? Jun 05 02:44:59 sorta Jun 05 02:45:19 looking for the 0.8.4 dillo bb patches Jun 05 02:50:48 morning all Jun 05 02:50:55 hi rp Jun 05 02:51:23 hey RP & pb_ Jun 05 02:51:32 hi koen Jun 05 02:51:47 anyone know what happend to mike 0.8.4 dillo bb patches? Jun 05 02:51:53 I don't suppose anyone's had 2.6.12-rc5 running on a pxa successfully? Jun 05 02:52:10 hi Jun 05 02:52:15 hey CoreDump|home Jun 05 02:53:48 * BigAl hasn't tried. Jun 05 02:54:49 I think 2.6.12-rc5 introduces random segfaults on the pxa :-/ Jun 05 02:54:55 Not good. Jun 05 02:55:04 I'll stick with -rc4 then Jun 05 02:56:09 BigAl: The problem being if everyone does that, the problem will never get fixed... Jun 05 02:56:32 hi Jun 05 02:56:33 RP: I'm still trying to get the LCD to work, so I'd rather have only one problem at a time. :) Jun 05 02:56:40 hey hrw Jun 05 02:56:45 morning Jun 05 02:57:07 hi hrw Jun 05 02:57:14 RP: I dont bump kernel because dont know which part of mm is neede, which dont etc.. no kernel hacker Jun 05 02:57:24 BigAl: I can understand that :) Jun 05 02:57:44 yesterday I found that I'm nearly no programmer even... Jun 05 02:57:52 hrw: We use some of the input and mtd trees and inotify for mickey Jun 05 02:58:57 BigAl: If you send me a dump of the pxa framebuffer registers under 2.4, I can probably get you a working lcd Jun 05 02:59:00 ok. time to trace wine problem here Jun 05 03:00:04 RP: I'll see what I can do. I've done some checking of the registers in 2.6, and they seem to correspond to what sharp has in the 2.4 code, but the display is always blank (white). Jun 05 03:01:07 BigAl: Does the LCD have power? Jun 05 03:02:45 RP: I think it should, if nothing else, the backlight gets brighter for a moment when the lcdtg startup code is called. Jun 05 03:03:09 RP: Although I don't know if that means that it's worked correctly or not. Jun 05 03:04:07 I really don't know... Jun 05 03:04:20 I wish we had some doco on that LCD chip sharp use. Jun 05 03:05:07 IIRC, it uses something similar to the c7x0. Have you looked at the corgi_lcd.c code? Jun 05 03:05:14 I'd guess its the same lcd chip Jun 05 03:05:39 fsck.. wine with full debug = 180M log to parse Jun 05 03:05:52 It seems to be exactly the same, so I'm using the corgi_lcd code with a few minor modifications for testing Jun 05 03:05:52 You may find my code slightly more readable and I know roughly how it works Jun 05 03:06:07 You have the ssp interface setup correctly? Jun 05 03:06:24 The ssp stuff is almost exactly the same too, apart from clock dividers and GPIO differences. Yup, that's working, I can read from the MAX1111 Jun 05 03:07:07 anyone here had much experiance working with PCB? Jun 05 03:07:29 mithro: I've had some... Jun 05 03:07:44 * BigAl has made a couple of small ones before Jun 05 03:07:47 * mithro wants to chat about solder masks :P Jun 05 03:07:53 do i need one? Jun 05 03:08:13 If you're doing very small smd parts, they're handy Jun 05 03:08:23 i'm thinking that if i have tracks running between pins i would? Jun 05 03:08:30 * mithro is mainly using smd Jun 05 03:08:41 you don't _need_ one, but it'll certainly make your life easier. Jun 05 03:08:41 trying not to use through hole stuff atm Jun 05 03:08:45 * BigAl thinks a solder mask would be a good idea. Jun 05 03:08:52 mithro: Yes, in that case it will make your life easier Jun 05 03:09:12 You could so without but you might have problems assembling it without bridges Jun 05 03:09:20 bridges? Jun 05 03:09:39 and how to you solder SMD stuff :P Jun 05 03:09:47 solder bridges - bits of solder accidently joining tracks together Jun 05 03:09:52 if you don't have a mask, it's very easy for the whole board to dissolve into a sheet of molten solder, then you need to spend ages wicking it off. Jun 05 03:10:17 is the solder mask, solder repelent? Jun 05 03:11:03 * mithro is currently thinking or using solder paste for doing the SMD stuff? is that a good idea? Jun 05 03:11:50 I wouldn't generally use solder paste for hand assembly. It's too tedious to apply without a stencil. Jun 05 03:11:57 http://www.amazon.com/exec/obidos/ASIN/B00026P9US/002-8619616-6779266 - irda keyboard for 9.88 USD... Jun 05 03:12:26 so how would you recommend doing it? Jun 05 03:12:54 Just use regular solder. Tin the pads, sit the component on top of them, heat the pins up, voila. Jun 05 03:13:44 You can use a spot of superglue to hold small devices in place prior to soldering Jun 05 03:13:49 so would anyone be able to check over my circuit at some point in the next week or two? Jun 05 03:14:03 mithro: What are you designing? Jun 05 03:14:11 this is my first PCB and it's a bit complex Jun 05 03:14:23 8 Port Asterisk FXS device Jun 05 03:15:05 basically a gloarified 8 channel DAC/ADC :P Jun 05 03:16:42 Fun! Jun 05 03:17:04 it's a really good learning experiance Jun 05 03:17:24 I can imagine.. Good luck with it! Jun 05 03:18:18 mithro: I can look over it. Without knowing exactly what you're doing, I'm not sure how much use it will be though Jun 05 03:19:04 well there are most probably stupid, this is my first PCB mistakes in there :P Jun 05 03:21:55 mithro: What are you drawing it in? A proper pcb program or by hand? Jun 05 03:22:19 mithro: My best tip would be to print off a scale copy of it and check the components match the pad sizes etc Jun 05 03:23:26 i'm currently using Cadsoft Eagle Jun 05 03:24:24 I've heard of it but never used it... Jun 05 03:25:15 it's got a freeware version which is pretty good Jun 05 03:25:33 btw anyone recommend a good PCB prototyping house? Jun 05 03:27:23 mithro: If you were in the UK, I'd recommend pcbtrain... Jun 05 03:39:14 RP: well i'm going to be in sweden soon Jun 05 03:43:38 ~lart sox Jun 05 03:43:39 * ibot takes out a cattle prod and gives sox a good jolt Jun 05 03:44:10 someone know ANY util which will convert WAVE mpeg audio encoded into wave pcm? Jun 05 03:45:39 hi folks Jun 05 03:45:47 hrw: madplay? mpeg2dec? Jun 05 03:46:01 mpg123? gstreamer? Jun 05 03:46:03 lame? Jun 05 03:46:15 (lame can decode too) Jun 05 03:46:28 ~lart Qt/E4 Jun 05 03:46:28 * ibot nukes Qt/E4 with a single large nuke Jun 05 03:46:42 hail qt/x11 Jun 05 03:46:43 thx Jun 05 03:47:11 pb_: They should -no-gfx-all as no fscking graphic driver compiles Jun 05 03:47:17 zecke: heh Jun 05 03:47:34 qt/e with no graphics does sound like it would suck Jun 05 03:49:02 2968 wave files to convert... Jun 05 03:53:19 anyone know of any oe changes in the last two days which would suddenly cause "too many open files" errors when trying to build gcc-cross ? this is on builds which previously worked Jun 05 03:53:56 at first we thought it was a gcc 3.4.4 thing but it is happening on 3.4.3 too (so I am told) Jun 05 03:54:23 can't think of anything. Jun 05 03:54:36 I suggest you look in /proc//fd while bitbake is running, to see if it is leaking file descriptors Jun 05 03:54:53 and/or run it under strace, to see which open is failing Jun 05 03:55:34 under strace? ouch Jun 05 03:55:55 well, don't run the whole thing under strace, just attach it when you get close to the point of failure Jun 05 03:56:14 I know which open is failing, it just doesn't help me much: Jun 05 03:56:18 /home/packages/oe/nslu2/bb_unslung_default/tmp/cross/armeb-linux/bin/ld: cannot open linker script file /home/packages/oe/nslu2/bb_unslung_default/tmp/cross/armeb-linux/lib/libpthread.so.0: Too many open files Jun 05 03:56:30 ah, right Jun 05 03:56:51 okay. well, check the /proc thing to find out where all your files have gone. Jun 05 03:57:09 why is it trying to build libgcjgc anyway ? isn;t that a javac thing? Jun 05 03:57:25 ah, whoops Jun 05 03:57:37 I turned java on for all architectures except arm. I guess armeb doesn't want it either. Jun 05 03:57:57 no, we don't :-) Jun 05 03:58:41 I always forget that big endian is a separate TARGET_ARCH. Jun 05 03:58:41 heh Jun 05 03:58:52 don't feel bad, almost everybody does Jun 05 03:58:52 pb_: libpthread-0.9.so contains "GROUP ( libpthread.so.0 libpthread_nonshared.a )" - I don't think that's right Jun 05 03:59:07 and I think that's causing the orubouris effect Jun 05 03:59:19 rwhitby-away: ah, you think it's causing a loop? Jun 05 03:59:30 that's my first guess Jun 05 03:59:39 only been looking at it a couple of minutes Jun 05 03:59:40 let me have a look at what I have Jun 05 03:59:56 "/home/unslung/tmp/cross/armeb-linux/bin/ld: cannot open linker script file /home/unslung/tmp/cross/armeb-linux/lib/libpthread.so.0: Too many open files" Jun 05 04:00:19 -rwxr-xr-x 1 pb pb 98877 May 5 13:50 libpthread-0.10.so Jun 05 04:00:19 -rw-r--r-- 1 pb pb 143876 May 5 13:50 libpthread.a Jun 05 04:00:19 -rw-r--r-- 1 pb pb 1168 May 5 13:50 libpthread_nonshared.a Jun 05 04:00:19 -rw-rw-r-- 1 pb pb 49 May 5 13:50 libpthread.so Jun 05 04:00:19 lrwxrwxrwx 1 pb pb 18 May 21 16:14 libpthread.so.0 -> libpthread-0.10.so Jun 05 04:00:31 so, yah, libpthread.so is meant to be that script, but libpthread-0.9.so should be the real library Jun 05 04:00:36 On armeb we get: -rwxr-xr-x 1 rwhitby rwhitby 49 2005-06-05 14:19 libpthread-0.9.so Jun 05 04:00:36 lrwxrwxrwx 1 rwhitby rwhitby 15 2005-06-05 14:19 libpthread.so -> libpthread.so.0 Jun 05 04:00:36 lrwxrwxrwx 1 rwhitby rwhitby 17 2005-06-05 14:19 libpthread.so.0 -> libpthread-0.9.so Jun 05 04:00:52 and the -0.9.so file contains that GROUP command which points back to itself Jun 05 04:00:59 right, that's clearly no good Jun 05 04:01:14 I'll run an armeb build here and see if I can find out what's up. Jun 05 04:01:36 `preciate it Jun 05 04:01:36 Note that my paste above is with gcc put back to 3.4.3 - I first thought it was a problem with 3.4.4 Jun 05 04:01:53 what version of glibc are you guys using? Jun 05 04:02:04 2.2.5 - we have to use that version Jun 05 04:02:06 rwhitby-away, that's the exact same error I get with 3.4.4 Jun 05 04:02:08 okay Jun 05 04:02:18 rwhitby-away, you tried openslug yet? I haven't Jun 05 04:02:22 libpthread is part of glibc, so I'm guessing it is to blame Jun 05 04:02:25 (compatibility with Linksys binaries for which we do not have the source) Jun 05 04:02:52 jacques: I get an unrelated error in initscripts-openslug Jun 05 04:03:18 rwhitby-away, so it gets way past glibc then ? Jun 05 04:03:26 what MACHINE and DISTRO should I be setting to reproduce your build? Jun 05 04:03:34 nslu2, unslung Jun 05 04:03:36 thanks Jun 05 04:03:42 no, thank you :-) Jun 05 04:04:01 pb_, and please turn off java for armeb when you get a chance :-D Jun 05 04:04:18 jacques: do we know that's the cause? Jun 05 04:05:04 rwhitby-away, we don't need it, it adds build time, and pb had intended it to be off for arm archs anyway Jun 05 04:05:10 heck I'm not even sure it works on arm Jun 05 04:05:13 yah, it doesn't work on arm Jun 05 04:05:33 I don't think it will even compile, but it certainly doesn't run. Jun 05 04:05:45 there's a whole pile of runtime stuff missing right now. Jun 05 04:05:58 03pb 07 * r1.3508 10openembedded/packages/gcc/gcc3-build.inc: also suppress java on armeb Jun 05 04:05:59 ok, cool Jun 05 04:06:15 (in fact, I just added "make gcj work on arm" to our list of project suggestions for the summer of code) Jun 05 04:06:28 jacques: there you go Jun 05 04:06:44 make dillo conirm the ACID2 test Jun 05 04:06:48 confirm... Jun 05 04:06:59 make dillo use css, javascript, ssl Jun 05 04:07:00 pb_, thankee Jun 05 04:07:02 oh wait Jun 05 04:07:11 improve osb-browser Jun 05 04:07:28 jacques: pulling into nslu2-linux now ... Jun 05 04:07:57 I think I'll sell TT my Tinderbox setup... Jun 05 04:08:00 zecke: dillo is obsolete crap Jun 05 04:08:01 zecke: atomicity, consistency, isolation and durability? sounds like a good idea Jun 05 04:09:50 zecke: they have a nice abstraction design, but refuse to use it Jun 05 04:10:01 hehe Jun 05 04:10:32 did someone try to run opera/gtk on a ipaq? Jun 05 04:10:43 is that opensource? Jun 05 04:10:49 * mithro ponders Jun 05 04:10:49 :} Jun 05 04:11:05 koen: that solely depends on your access rights Jun 05 04:11:10 heh Jun 05 04:12:58 heh.. mpg123, mpg321, madplay, lame create noise ;( Jun 05 04:18:25 libaudiofile error 0: 'WAVE MPEG layer 3 data format is not supported' Jun 05 04:18:28 fsck Jun 05 04:24:03 someone can decode second line of that? http://home.hrw.one.pl/tmp/error.jpg Jun 05 04:34:56 pb_, thanks - unslung builds for us again :-) Jun 05 04:45:46 afternoon Jun 05 05:36:23 Are the kernel-module-* ipkgs depencies set according to depmod ? Jun 05 05:37:06 on 2.4, yes Jun 05 05:37:08 Or do I have to be verbose in the BOOTSTRAP_RDEPENDS ? Jun 05 05:37:18 Ah, not on 2.6 ? Jun 05 05:37:33 there was some talk a few months ago about it Jun 05 05:38:10 so the easiest would be to include kernel-modules ? (which I assume depends on kernel-module-*) Jun 05 05:38:21 yeah Jun 05 05:38:25 thanks Jun 05 05:38:33 that's wasts about 3 mb of flash, though Jun 05 05:40:29 well, hx4700 has 128MB :) Jun 05 05:41:03 in one partition or in 2x64? Jun 05 05:42:17 I think todd's bootloader uses 2x64 Jun 05 05:57:21 pb_: kernel.bbclass is odd Jun 05 05:57:38 it should DEPEND on the kernel gcc -3.3.3 Jun 05 05:57:44 but somehow 3.3.4 gets built Jun 05 05:59:03 zecke: yeah, I notices that too, but forgot to mention it Jun 05 05:59:30 I've changed the sa kernels to use 3.3.4 Jun 05 05:59:37 but that does not look like the right fix now Jun 05 06:04:05 * zecke decided not to be in #opie anymore Jun 05 06:05:33 bad atmosphere? Jun 05 06:06:01 Twiun: no I'm the only one answering questions Jun 05 06:06:08 and nobody cares to even read what I say Jun 05 06:06:31 maybe time to pack up the channel and leave everything to the mailing list :( Jun 05 06:06:48 maybe just leave Opie altogether Jun 05 06:06:58 users make me crazy Jun 05 06:07:03 Whatever you do they're not happy Jun 05 06:07:17 One hacks tslib support and it is broken XYZ works but why does Opie Jun 05 06:07:20 ~users Jun 05 06:07:25 so this is what we get for sharing our unpaid volunteer work with you? complaints and accuses? well done, this clearly supports our motivation to continue working on open source projects. Jun 05 06:07:31 One hacks keyboard support and it is broken XYZ works but why does Opie not work Jun 05 06:07:42 * Twiun nods Jun 05 06:07:45 One hacks a Terminal Emulator, vi backspace does not work but with XYZ it works Jun 05 06:08:02 heh, that's a busybox bug Jun 05 06:08:34 no it works with aterm ;) Jun 05 06:08:43 :) Jun 05 06:09:38 zecke: dropping opie would mean lessening your workload ;) Jun 05 06:10:10 Twiun: I'll continue doing Mobile Stuff Jun 05 06:13:01 CoreDump|home: tell me how to stay calm and friendly with not leaving Opie? Jun 05 06:13:15 killall users Jun 05 06:13:31 just ignore the usual idiots Jun 05 06:13:32 CoreDump|home: I've to talk to people able to spend ~1000 bucks for a device Jun 05 06:13:55 CoreDump|home: I'Ve not thousands bucks to spend within 3/4 a year... Jun 05 06:14:09 few people have Jun 05 06:14:24 all my money is spent for food and drugs Jun 05 06:14:42 mmmmmmm..... food..... Jun 05 06:14:42 luckily I've a shalter Jun 05 06:16:19 * CoreDump|home preps a new release Jun 05 06:17:32 building for 2 devices is a pain, 3 devices are going to be a PITA Jun 05 06:18:28 CoreDump|home: completely new or just moving from do_not_use Jun 05 06:19:10 CoreDump|home: how much is missing to allow building these images in one tree? Jun 05 06:19:16 zecke: new "official" RCs for do_not_use. Now for collie and poodle. Jun 05 06:19:29 zecke: nothing :) Jun 05 06:20:04 I had to use some machine overides in some places but that's it Jun 05 06:23:18 anyways, uploading 100MB+ sucks Jun 05 06:23:19 where does debian store it's postinsts? Jun 05 06:24:44 koen: /var/lib/dpkg/info Jun 05 06:24:50 ah, thanks Jun 05 06:24:57 np Jun 05 06:55:37 zecke: that does sound a bit odd Jun 05 06:57:27 sigh, looks like someone removed collie-kernel-all.bb. Jun 05 06:59:16 CoreDump|home: last I heard of that mickeyl gau Jun 05 06:59:27 guarded that with DISTRO_TYPE Jun 05 06:59:31 guarded that with DISTRO_TYPE Jun 05 06:59:45 * koen gives screen a funny look Jun 05 07:00:04 yeah, but, err the collie-kernel-all _did_ come in handy from time to time Jun 05 07:00:24 if you set DISTRO_TYPE=release it should buid all Jun 05 07:00:57 zecke: oh, heh. here's why: Jun 05 07:01:03 DEBUG: providers for virtual/arm-linux-gcc-3.3.3 are: ['gcc-cross-kernel'] Jun 05 07:01:03 DEBUG: selecting /nexus/home/pb/oe/oe/packages/gcc/gcc-cross-kernel_3.3.4.bb to satisfy virtual/arm-linux-gcc-3.3.3 Jun 05 07:01:11 that's a bit embarrassing. Jun 05 07:01:34 koen: thx, will keep that in mind:) Jun 05 07:05:46 :} Jun 05 07:06:03 zecke: I'll check in a fix^H^H^Hworkaround Jun 05 07:06:17 * pb_ waits for consistency check Jun 05 07:07:08 grr, bitkeeper is no fun on nfs Jun 05 07:11:38 03pb 07 * r1.3509 10openembedded/packages/gcc/ (2 files): disambiguate gcc-cross-kernel packages to work around dim handling of PV in PROVIDES Jun 05 07:11:40 zecke: try that Jun 05 07:15:38 hehe Jun 05 07:15:51 ~lart ETSI Jun 05 07:15:51 * ibot DoSes ETSI Jun 05 07:26:46 hum, my gpe-image seems to be hanging at "Configuring locale-base-de-de" ... Jun 05 07:28:37 zecke: ETSI? Jun 05 07:29:16 koen: www.etsi.org Jun 05 08:50:16 03pb 07 * r1.3510 10openembedded/packages/glibc/glibc_2.2.5.bb: don't clobber glibc 2.2.5 libpthread with a stray linker script Jun 05 08:52:53 03pb 07 * r1.3511 10openembedded/packages/glibc/glibc_2.3.3+cvs20050420.bb: set DEFAULT_PREFERENCE_i586=0 for 20050420 glibc Jun 05 09:52:32 03zecke 07 * r1.3507.1.1 10openembedded/packages/qpe-gaim/ (files/libgaim.patch qpe-libgaim_1.3.0.bb): Jun 05 09:52:32 QPE gaim: Jun 05 09:52:32 Update to gaim 1.3.0 (to fix security issues) Jun 05 09:55:49 zecke: http://www.opiegonebad.com/FRONT/nonflash.html Jun 05 09:56:51 hehe Jun 05 09:57:00 * zecke searches for Lorn Jun 05 11:15:03 what directory do people keep their build stuff in? do most folks use the example /stuff as their build root? Jun 05 11:15:37 I use ~/src/oe2005 Jun 05 11:28:01 ~/code/projects/oe/ Jun 05 11:30:59 hey kergoth Jun 05 11:34:07 ~/oe Jun 05 11:34:11 well, /oe is base dir for all OE stuff. ~/oe/oe is openembedded snapshot Jun 05 11:38:56 hmm... Jun 05 11:40:25 I had my stuff in /home/bitbake/bb, I'd get errors about not finding arm-linux-gcc-2.95... then I moved it to /stuff and now it seems to compile okay (filled my partition) Jun 05 11:45:51 ~nslookup hyperion.kergoth.com Jun 05 11:45:53 hyperion.kergoth.com is 24.118.222.11 Jun 05 11:49:14 hy all how do I force a patch in a bb file Jun 05 11:49:22 -f usually Jun 05 11:52:41 you fix the patch Jun 05 11:53:17 bitbake can't force apply patches. Jun 05 11:53:27 okay Jun 05 11:53:54 I was looking for the dillo .0.8.4 bb patches but they seen to me missing Jun 05 13:00:48 koen|tv: we should rot13 all of OE variable keys Jun 05 13:01:24 ggilbert__: hey, I finally have a kdepim + kdelibs that sort of works with kolab Jun 05 13:01:33 ggilbert__: the alioth debian packages are broken :} Jun 05 13:01:38 zecke: nice Jun 05 13:01:49 ggilbert__: now the shitty part in debugging begins Jun 05 13:01:58 ggilbert__: actually I will prepare some stuff to ease debugging Jun 05 13:02:14 zecke: that would improve readability :) Jun 05 13:02:54 we need a #oe-bofh channel Jun 05 13:03:17 actually I should only join channels with that in the name Jun 05 13:07:46 zecke: new distro name BOFH linux Jun 05 13:08:54 fruiting exec_task is not execing my fruiting task again. Jun 05 13:09:09 I thought openbsd was the bofh linux *d&r* Jun 05 13:11:43 03pb 07 * r1.3513 10openembedded/packages/gcc/ (gcc_csl-arm-20050603.bb gcc_4.0.0.bb): shuffle some gcc DEFAULT_PREFERENCEs around Jun 05 13:12:28 zecke: we should invite theo to join the ML :) Jun 05 13:15:08 aoe: no OpenBSD was about singing hippie songs and drawing fishs Jun 05 13:31:24 Ciao all Jun 05 13:36:43 03koen 07 * r1.3514 10openembedded/ (2 files in 2 dirs): Jun 05 13:36:43 tslib-maemo_0.0.1-15.bb: tweak tslib-maemo a bit, courtesy Florian Boor Jun 05 13:36:43 nokia770.conf: tweak preferred providers, courtesy Florian Boor Jun 05 13:37:13 koen: what is different in their version of tslib? Jun 05 13:38:05 zecke: hacks to make it work with kdrive Jun 05 13:38:42 that's why we didn't patch the 'regualr' tslib Jun 05 13:39:38 It will be obsoleted in a few weeks Jun 05 13:45:03 zecke: here comes one for you Jun 05 13:45:11 03koen 07 * r1.3515 10openembedded/packages/qte/qte_2.3.10.bb: qte_2.3.10.bb: add md5sum to pester zecke Jun 05 13:45:56 heh Jun 05 13:51:43 koen: lol, you start trusting Qt? Jun 05 13:53:01 zecke: just breaking unimportant stuff ;) Jun 05 13:58:38 why is gpe so slow on 5500? Jun 05 13:58:53 ljp: it uses Qt4/Arthur for drawing Jun 05 13:58:55 03koen 07 * r1.3516 10openembedded/packages/maemo/ (xsp_1.0.0-8.bb xpext_1.0-5.bb): Add missing nokia x-extensions as Florian pointed out Jun 05 14:00:12 ha! Jun 05 14:00:14 it is like someone wrote the code to do this correctly, and then did not use that code. Jun 05 14:00:46 i want to try maemo on 5500 Jun 05 14:01:15 can someone tell me why bb.build.exec_task calls exec_task recursively from the callback? and why it doesn't use the 'usecache' option for digraph.walkdown? Jun 05 14:02:57 ljp: maemo on a c700: http://dominion.kabel.utwente.nl/koen/blog/pyblosxom.cgi/Handhelds/maemo-z.html Jun 05 14:03:39 keturn: does it depend on itself? Jun 05 14:04:21 it seems like the whole point of digraph.walkdown is that it does the ordering, and so I don't understand why there's this "if task != item: exec_task(item)" thing Jun 05 14:09:08 koen: oh did we decide on ;md5= now? Jun 05 14:09:22 I thought it was MD5SUM = "" Jun 05 14:09:47 zecke: MD5 = was only for .inc, right? Jun 05 14:09:55 as a fallback Jun 05 14:09:58 MD5SUM doesnt make much sense, given you wouldnt know which of hte items in SRC_URI it applies to Jun 05 14:10:04 heh Jun 05 14:13:41 kergoth: overrides Jun 05 14:13:48 MD5SUM_basename = "sum" Jun 05 14:14:07 and add basename tempoarirly to the overrides Jun 05 14:15:01 any screenshots of maemo w/o apps running? Jun 05 14:16:44 Luke-Jr: only with out the 'desktop' Jun 05 14:17:13 hmm Jun 05 14:18:36 Luke-Jr: http://handhelds.org/scap/port.30330.png Jun 05 14:18:50 the default desktop has small applet windows Jun 05 14:19:01 like a clock and mp3 player Jun 05 14:19:18 similar to gpe-appmgr at all? Jun 05 14:19:39 no Jun 05 14:20:21 they want to imitate the symbian look & feel a bit Jun 05 14:20:29 to be consistent with other nokia stuff Jun 05 14:20:33 I've never used Symbian Jun 05 14:20:41 me neither :) Jun 05 14:21:43 good nite Jun 05 14:21:53 nite zecke Jun 05 14:22:01 'night zecke Jun 05 14:24:15 'night all Jun 05 14:24:23 nite koen Jun 05 14:24:30 time to sleep Jun 05 14:24:32 nite all Jun 05 14:36:45 nite Jun 05 14:57:29 Well, I finally worked out that rmk broke xscale in 2.6.12-rc4-git1... Jun 05 15:18:37 03pb 07 * r1.3517 10openembedded/packages/busybox/ (3 files in 2 dirs): patch busybox bogosities causing compile failure with gcc 4.0.0 Jun 05 15:33:36 03pb 07 * r1.3518 10openembedded/packages/modutils/ (modutils_2.4.27.bb modutils-2.4.27/gcc4.patch): apply patch for modutils compile failure with gcc4 **** ENDING LOGGING AT Sun Jun 05 23:59:56 2005