**** BEGIN LOGGING AT Sat May 12 02:59:56 2007 May 12 03:44:12 Hi, How to convert .dot files to image files using dot? May 12 07:28:49 03koen 07org.oe.dev * r9b851bdc... 10/ (1 packages/angstrom/angstrom-x11-image.bb): angstrom-x11-image: don't overwrite EXTRA_INSTALL May 12 07:30:27 is it a problem expat and expat-native not being the same version? May 12 08:46:43 03ifaistos 07org.oe.dev * r311b05b3... 10/ (1 classes/oplinux-mirrors.bbclass): classes/oplinux-mirrors.bbclass: Add mirrors for OPLinux distro May 12 08:55:34 03ifaistos 07org.oe.dev * rbad764a0... 10/ (1 conf/distro/include/oplinux.inc): conf/distro/include/oplinux.inc:Add the mirrors to oplinux and oplinux-uclibc May 12 09:21:09 03ifaistos 07org.oe.dev * r4b80b299... 10/ (1 packages/linux/linux-magicbox_2.6.19.2.bb): packages/linux/linux-magicbox_2.6.19.2.bb: Remove some comments, bump PR May 12 09:58:17 Hi May 12 09:59:55 I am using angstorm 2007.1 and I am trying to define PREFERED_PROVIDER_virtual/xserver in my local.conf but it does not work, I use n800 as machine May 12 10:02:47 I tried the same trick for defining what libsdl to use and that worked May 12 10:02:52 RP: are you around? May 12 10:03:30 Do I perhaps need to put the PREFERED definitions in a differetn file? May 12 10:04:49 PREFERRED_PROVIDERS May 12 10:04:55 you have a typo May 12 10:08:37 PREFERRED_PROVIDER_virtual/xserver is correct May 12 10:13:23 CONFIG_JFFS2_FS=m argh May 12 10:21:40 hi koen, CoreDump|home May 12 10:22:13 hey Ifaistos May 12 10:23:13 FATAL: This kernel is too big for your PXA Zaurus May 12 10:23:28 this is getting better and better May 12 10:24:24 CoreDump|home, if you compiled jffs2 in, maybe you should make ext module? May 12 10:25:12 polyonymous: I'm trying that right now May 12 10:25:38 well, indeed, I shouldn't waste precious irc bandwidth on trivialities ;-) May 12 10:26:05 polyonymous: your trivialities are appreciated =) May 12 10:26:26 yeah, I actually wanted to show support :) May 12 10:27:02 koen: is there a reason why ipkg-native is not compiled by default for angstrom ( and hence do_rootfs fails )? May 12 10:27:23 yes, it's called 'bitbake 1.6' May 12 10:28:15 right, just found the thread on the ML May 12 10:28:30 CoreDump|home, there's a patch on angstrom ml May 12 10:28:40 hi May 12 10:28:49 polyonymous: thanks, I'll have a look May 12 10:29:05 manually building ipkg-native fixed it for now May 12 10:29:08 I had a strange dependencies problem with bb 1.8 too May 12 10:29:16 hi Bernardo May 12 10:29:26 koen says it slows down bb to crawl, I haven't noticed though. May 12 10:29:55 I did a "bitbake sax-utils-native angstrom-x11-image angstrom-e-image" and it tried to build linux-headers before sax-utils-native May 12 10:30:02 so the qa run failed May 12 10:30:04 polyonymous: exchanging ext2 for jffs2 worked btw =) May 12 10:30:21 good :) May 12 10:30:29 Bernardo: don't put two images in one commandline May 12 10:30:36 if I did a "bitbake sax-utils-native" only, the necessary tools got built May 12 10:30:38 Bernardo: OE doesn't have locking around do_rootfs May 12 10:30:56 koen: ok, I just assumed bitbake would build targets in order May 12 10:31:14 thanks May 12 10:32:12 to avoid all the cache checking phase, I would just do a long bitbake line with all the targets I wanted to build (usually meta-opie opie-kdepim-image and so on) May 12 10:32:35 time to split things up with "&&" and leave it running thru the night May 12 10:34:44 yay it's booting May 12 10:57:06 Bernardo: with 1.8 the cache checking takes ~2 seconds May 12 11:20:29 hello May 12 11:23:31 ~seen PR May 12 11:23:54 ~seen RP May 12 11:24:02 pr was last seen on IRC in channel #debian, 68d 9h 53m 22s ago, saying: 'er'. May 12 11:24:16 rp is currently on #oe #openmoko #openzaurus #handhelds.org. Has said a total of 13 messages. Is idling for 20h 44m 30s, last said: 'Ideally, we need to fix IO redirection properly...'. May 12 11:25:44 RP: need to ask you a few things regarding your lzo patches May 12 11:40:44 03koen 07org.oe.dev * r3dbe1271... 10/ (1 packages/angstrom/angstrom-console-image.bb): angstrom-console-image: only ship psplash-zap if psplash has been included May 12 12:14:16 RP: would you be cool with this patch to linux-rp.inc? http://rafb.net/p/BkmthK33.html May 12 12:14:33 RP: it would allow easier creation of custom kernels May 12 12:24:39 03pfalcon 07org.oe.dev * r2ef201b5... 10/ (4 files in 2 dirs): linux-handhelds-2.6: CVS HEAD is 2.6.21 now, 2.6.20 is in branch. May 12 12:24:43 03pfalcon 07org.oe.dev * r1687f04a... 10/ (1 packages/linux/linux-handhelds-2.6_cvs.bb): linux-handhelds-2.6 cvs: CVS HEAD is 2.6.21-hh2 now. May 12 12:26:51 03pfalcon 07org.oe.dev * rf261cf94... 10/ (4 files in 2 dirs): linux-handhelds-2.6: Drop obsolete recipes. May 12 13:50:18 bugger ! . still not luck . bitbake worl starts complaining about meta-maemo. bitbake gpe-image complains that the requested glibc 2.5 is not aviable for virtual/arm-angstrom-linux-gnueabi-libc-for-gcc . May 12 13:55:46 03ifaistos 07org.oe.dev * ra9b0b2d0... 10/ (1 packages/madwifi/madwifi-ng_r.inc): packages/madwifi/madwifi-ng_r.inc: set magicbox's arch to ppc so madwifi-ng compiles May 12 14:04:21 keesj, the glibc warning is safe to ignore as far as I know. May 12 14:20:24 * koen stabs redboot May 12 14:31:48 what is the recomended way to test multiple images? altboot? May 12 14:32:31 by flashing them May 12 14:32:36 altboot seems to be broken May 12 14:33:10 there have been some people who've hacked it to get it working again, or are using an old version May 12 14:33:16 :( May 12 14:33:27 * polyonymous is some one. May 12 14:33:43 but I only have it on spitz May 12 14:33:59 polyonymous: care to tell me how to? I'd like to keep my opie image for work, but still start testing enlightnemnt May 12 14:34:06 03Laibsch 07org.oe.dev * rc46f8208... 10/ (1 packages/bitbake/bitbake.inc): May 12 14:34:06 bitbake.inc: correct spelling mistake May 12 14:34:06 replace references to OpenZaurus and Familiar with Angstrom May 12 14:34:10 on a akita, so there shouldn't be much diference May 12 14:34:29 Bernardo, does akita have hdd? I think not, and that's what makes a difference here. May 12 14:34:48 no, but I have a 2GB sd I can play with May 12 14:35:10 or is altboot hardcoded for /dev/hda? May 12 14:35:14 basically, I've built and old altboot (the previous version available in OE) and built a chroot script for it. May 12 14:35:55 the altboot now in the dev branch, or the one in the oz354x branch? May 12 14:35:55 well, basically, altboot just replaces init and provides a menu to run chrooted init. May 12 14:36:04 I only use .dev May 12 14:36:12 ok May 12 14:36:16 polyonymous: you should figure out a way to package that... that'd be nice to have May 12 14:36:40 I've built altboot on dev, so I would need your script besides that, no? May 12 14:36:50 summatusmentis, it's not hard to package one, not sure if it should go to feeds, though. It's not hard to put into feeds also, but that's not my realm. May 12 14:37:10 Bernardo, those are quite trivial May 12 14:37:34 polyonymous: I meant the script more so, which version of altboot are you using? May 12 14:37:54 03koen 07org.oe.dev * r44ac9811... 10/ (1 conf/checksums.ini): checksums.ini: fis.c is fetched from svn, so don't checksum it May 12 14:37:54 well, I tried previously to use altboot, but it never liked the images I put on my sd card May 12 14:37:57 even just a .tar.bz2 of the script w/ instructions on how to place it May 12 14:38:16 summatusmentis, I told you last time we've talked... May 12 14:38:32 oh, I guess I do remember that May 12 14:38:36 * summatusmentis goes and searches logs May 12 14:41:35 http://rafb.net/p/bkCTlm49.html http://rafb.net/p/kP9mDg53.html == these are my silly scripts. May 12 14:43:55 polyonymous: cool, thanks a lot May 12 14:44:08 summatusmentis, found version in log? May 12 14:44:14 the OPIE one isn't necessary? May 12 14:44:36 summatusmentis, that's in altboot-menu (note the filenames in 'description') May 12 14:44:39 polyonymous: yeah, the only thing, I'm not seeing the -r0 at the end, but I've got the same .bb file except for that May 12 14:44:41 it doesn't have to be opie. May 12 14:44:58 summatusmentis, that was the version, not the filename of bb May 12 14:45:17 ok, good May 12 14:45:55 so, does if I'm booting from SD, do I need anything like the OPIE script? or will altboot handle picking an image? May 12 14:48:07 summatusmentis, no idea, as I told you - I'm testing chrooted images. May 12 14:48:46 altboot will work again on SL-Cxx00 eventually May 12 14:48:52 right... I guess I don't entirely understand what you're doing. Don't worry about it, I'll figure it out May 12 14:50:29 here comes the altboot master :) May 12 14:50:57 oh right! he's getting a 3200 soon May 12 14:52:01 good. I could've helped with it if he wasn't, but now it doesn't matter. And I'm leaving for two weeks this monday. May 12 14:52:48 It's already here =) backing up the HDD ATM May 12 14:53:56 CoreDump|home, congrats! May 12 14:54:05 thanks! May 12 14:54:34 * summatusmentis wishes CoreDump|home better luck w/ initial angstrom install than summatusmentis had May 12 14:54:56 angstrom-console-image is already booted May 12 14:55:29 * Laibsch has updated May 12 14:55:32 !oebug 2162 May 12 14:55:34 * * Bug 2162, Status: NEW, Created: 2007-04-29 09:33 May 12 14:55:35 * * bugs.openembedded.org(AT)rolf.leggewie.biz: latest altboot packages have packaging issues May 12 14:55:36 * * http://bugs.openembedded.org/show_bug.cgi?id=2162 May 12 14:55:44 with the latest info May 12 14:55:55 Laibsch: thanks May 12 14:55:59 CoreDump|home: Self-compiled? May 12 14:56:03 Bernardo: de nada May 12 14:56:12 Laibsch: yes May 12 14:56:18 hm May 12 14:56:29 a build from ~15hrs ago May 12 14:56:47 Laibsch: that means self-compiled altboot won't even package now right? May 12 14:57:05 * summatusmentis looks at the bug link May 12 14:58:07 oh, nvm May 12 14:58:38 summatusmentis: Not the one that is chosen by default. May 12 14:58:49 But you can change that May 12 14:58:54 I will look into it May 12 14:59:02 but not today May 12 14:59:03 nice, thank you May 12 14:59:11 Laibsch: yeah, I was gonna do what polyonymous is doing May 12 14:59:34 summatusmentis: That is probably the fastest road to success at this point in time May 12 14:59:52 yep, that's what I figured May 12 15:07:10 * wick720 is away: sl33p May 12 15:23:50 morning May 12 16:12:42 i have made a new package for clish -> http://clish.sourceforge.net/ May 12 16:13:21 can anyone figure out which licence i should use on the package ? it seems that parts of it are gpl parts are copyrighted May 12 16:15:17 Ifaistos: seems BSD/GPL combo. As GPL is most strict, use GPL. May 12 16:16:00 likewise : Thanks May 12 16:16:19 Ifaistos: But I'm not a lawyer ;-) May 12 16:16:58 likewise : i'll add a note to the bb file and the commit and maybe someone can clear it out May 12 16:17:19 Ifaistos: I have a question about: http://www.openembedded.org/viewmtn/revision.psp?id=a9b0b2d052fa806e034121432dba032618a0ec99 May 12 16:17:30 Ifaistos: You changed arch from powerpc to ppc? May 12 16:18:22 3333333333333333333333333333333333333333333333333333333333333333333333333333333333333332xxxxxxxxxxx May 12 16:18:26 kitten on keyboard error May 12 16:18:59 yes. mafiwifi recognizes it as ppc (its because amcc 405 is still on the ppc side of the kernel) May 12 16:19:07 :) May 12 16:19:12 m,eeeeeeeeeeeeeeeeeeeeeeeeeeeuuuuuuuuuuus================= May 12 16:19:44 s/mafiwifi/madwifi May 12 16:20:08 Ifaistos: 3EEEEEEEEEEEE?uyh May 12 16:20:27 Ifaistos: seems BSD/GPL combo. As GPL is most strict, use GPL. May 12 16:21:28 SORRY May 12 16:21:29 dejavu May 12 16:21:42 kitten likes clicking keys... May 12 16:21:43 there is something wrong in the matrix :) May 12 16:22:00 yeah, I took the blue pill May 12 16:22:33 Now I have 30+ screenshot windows open... May 12 16:25:16 heh May 12 16:25:32 probably got excited by the mouse :) May 12 16:27:15 ifaistos there is a screen saver for kittens May 12 16:27:22 but its a windows program May 12 16:30:27 likewise: I HATE REDBOOT! May 12 16:30:49 for some reason it doesn't accept ctrl-c to abort the script May 12 16:30:51 koen: haha. what part sucks most? May 12 16:30:59 so it keeps booting into snapgear May 12 16:31:37 koen: how many consoles? May 12 16:31:53 could be redboot pushes to multiple consoles, but reads only from one May 12 16:32:17 I attached the serial cable to the only serial plug coming out May 12 16:32:29 POE rocks, btw May 12 16:32:30 koen: you compiled redboot yourself, or is the on-flash one? May 12 16:32:37 on-flash May 12 16:32:39 koen: true POE (48 Volts etc)? May 12 16:32:51 does anyone remembers how to change the default sell once someone is a busybox enviroment ? May 12 16:32:54 koen: or 5V non-IEEE POE? May 12 16:32:56 it say 'POE' on the box, so I assume 40V May 12 16:33:00 passwd -e does not seem to work May 12 16:33:31 likewise: so now I'm looking for a statically compiled 'fconfig' binary for OABI armv5teb May 12 16:33:58 koen: for linux? May 12 16:34:43 yes: http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/redboot-utils May 12 16:39:37 koen file fconfig: fconfig: ELF 32-bit MSB executable, ARM, version 1, for GNU/Linux 2.4.0, statically linked, stripped, 408 kB May 12 16:40:20 could you but it online somewhere? May 12 16:40:42 koen: http://www.esrac.ele.tue.nl/~leon/linux/fconfig May 12 16:43:43 thanks May 12 16:48:11 hmmm May 12 16:48:17 that doesn't work out as well May 12 16:48:37 ah well May 12 16:48:41 time to watch some tv May 12 17:11:06 03polyonymous 07org.oe.dev * red86bdd4... 10/ (1 packages/git/git_1.4.4.2.bb): git: it turns out that git also relies on a decent sed. May 12 17:19:44 03ifaistos 07org.oe.dev * re78d06b7... 10/ (1 packages/clish packages/clish/clish_0.7.1.bb): May 12 17:19:44 packages/clish/clish_0.7.1.bb: Add new package clish May 12 17:19:44 CLISH is a modular framework for implementing a CISCO-like CLI on a Unix system. May 12 17:19:44 Not sure about the licence. Seems to be a mixed BSD/GPL so i used GPL as more strict. May 12 17:19:44 If anynone things its wrong feel free to update it. May 12 18:06:23 koen : Any problem to rename angstrom-minimal-image to minimal-image so it can by used by other distros and not create confusion ? May 12 18:08:02 it can already be used by other distros May 12 18:09:13 I can move it to packages/angstrom if people get too confused May 12 18:15:01 koen : I don't think you need to move it just rename it to a more "generic" minimal-image May 12 18:15:10 why? May 12 18:16:00 well for example we now issue bitbake bootstrap-image for any distro May 12 18:16:19 people that get confused by such a name shouldn't be using OE IMO May 12 18:16:39 :) May 12 18:17:18 to make it even more confusing, it creates an image called 'minimalist-image' :) May 12 18:17:36 well yeah, this is another matter May 12 18:18:14 basically we need a minimal-image as the difference between bootstrap and minimal is around 800K on a jffs2 fs May 12 18:18:42 and there is such an image: angstrom-minimal-image :) May 12 18:18:50 any distro can use that if they want May 12 18:20:41 http://www.eglibc.org/mission May 12 18:21:00 so i don't think there will be a problem to add minimal-image.bb which created a minimal-image- fs ? May 12 18:22:07 likewise: "freescale" "mvista" "windriver" "CSL" May 12 18:22:14 koen: I know... May 12 18:22:36 koen: more reasons to try uclibc first :-) May 12 18:22:47 yes May 12 18:23:07 * koen wants to make an initramfs installer for angstrom May 12 18:23:13 a la debian-installer, but better :) May 12 18:27:25 * koen wonders how big cherokee + avahi would be in a uclibc rootfs May 12 18:32:29 likewise: do you have your uclibc 0.9.29 tries online somewhere? May 12 18:33:26 koen: no, I deleted them when starting from scratch again on other issues. It was rc1, and I did not really work on uclibc issues. May 12 18:33:55 koen: currently, I am working on the new libtool (and its issues). May 12 18:40:39 likewise, I have been lookig at libtool, but my brain is dead May 12 18:41:38 03ifaistos 07org.oe.dev * r04e66b01... 10/ (1 packages/images/minimal-image.bb): May 12 18:41:38 packages/images/minimal-image.bb : Add a minimal-image that contains only the nessesary progs to boot May 12 18:41:38 and an ssh daemon for systems with limited flash resources May 12 18:46:31 oh fsck, so uclibc 0.9.29 out finally... May 12 18:52:07 Ifaistos: why did you duplicate angstrom-minimal-image?!?! May 12 18:53:57 03koen 07org.oe.dev * rb23341fd... 10/ (1 packages/images/minimal-image.bb): disapproval of revision '04e66b01ee822895efeb76e94c17d1f6605e0f83' May 12 18:54:08 there May 12 18:54:15 * koen curses at senseless duplication May 12 18:56:21 likewise: I'm running into problems with uclibc.inc wanting to symlink (non-existant) kernel headers May 12 19:07:15 Crofton|home: Been at the graduation/memorial ceremony? May 12 19:12:14 bah, uclibc is giving me head-aches May 12 19:19:11 ah May 12 19:19:12 NOTE: Unpacking /home/koen/OE/downloads/uClibc-0.9.29.tar.bz2 to /data/build/koen/OE/build/tmp/angstrom/work/ixp4xxbe-angstrom-linux-uclibcgnueabi/uclibc-0.9.29-r0/ May 12 19:19:13 :) May 12 19:20:58 koen : I think this goes nowhere.Either rename angstrom-minima-image it to something general or let there be a general minimal-image May 12 19:21:30 or we will end up with -minimal-image May 12 19:21:36 Ifaistos: what's the problem exactly? May 12 19:21:48 any distro can do 'bitbake angstrom-minimal-image' May 12 19:22:44 i think you understand what i am saying. May 12 19:23:19 actually, I don't May 12 19:24:30 There should be a minimal-image as there is a bootstrap-image and not an angstrom-bootstrap-image or a distro-name-boostrap-image May 12 19:25:11 likewise, hiding from the crowds :) May 12 19:25:24 summer is great around here, no people, no traffic May 12 19:25:35 Ifaistos: you still haven't said what's the problem May 12 19:25:50 i think its too distro specific May 12 19:25:57 that's nonsense May 12 19:26:02 and it does not make any sense May 12 19:26:19 apart from the name, what's distro specific about it? May 12 19:26:39 nothing. that's why the name should be generic also May 12 19:27:04 if it was doing something specific to angstrom i would have no objection May 12 19:27:15 but it does not May 12 19:27:38 so the only problem is "it has angstrom in the recipe name" May 12 19:29:10 likewise: do you by chance still have the changes to insane.bbclass for uclibc-gnueabi May 12 19:29:20 yes, and that's make it look distro distro specific when its not May 12 19:29:52 unless you want it to be, so let there be a generic one May 12 19:30:07 and how does making a 1:1 copy of it containing ANGSTROM_EXTRA_INSTALL ?= "" solve that? May 12 19:30:07 koen: no... May 12 19:30:59 i just change a few things a commited it. if the ANGSTROM_EXTRA_INSTALL is there it should be removed May 12 19:31:36 Ifaistos: senseless duplication changes nothing May 12 19:31:39 koen: preparing my efika/angstrom build for uclibc again. May 12 19:31:46 i have nothing against angstrom or you, but really this should be generic May 12 19:33:09 in the same sense you shoudl rename bootstrap-image to angstrom-bootstrap-image and i should add a oplinux-bootstrap-image and maybe a nylon-boostrap-image etc May 12 19:33:20 does not make sense May 12 19:33:37 if its is angstrom specific them keep the name May 12 19:33:44 it was angstrom specific May 12 19:33:48 and let also a generic one May 12 19:33:52 I made it generic, but didn't rename it May 12 19:34:13 I'd like to keep the name to give credit which distro added it May 12 19:34:41 koen: That is "suboptimal" to say the least May 12 19:34:44 put it inside, we all know you did it May 12 19:35:10 koen: I remember one uclibc change I had to make regarding using kernel headers built from the kernel sources. May 12 19:35:26 koen: sed related May 12 19:36:29 koen: Does the uclibc reciepe "work"? May 12 19:36:51 NAiL: I'm attempting to build a toolchain right now May 12 19:37:06 koen: You using "old-style" kernel headers? May 12 19:37:19 likewise: 2.6.20, new style May 12 19:37:26 0.29 should be able to handle that May 12 19:37:46 koen: ah ok, I probably fixed 0.28 to cope with that then. May 12 19:37:55 "should" May 12 19:37:56 * likewise has lossy memory May 12 19:38:18 copy that May 12 19:39:10 afaik there are 2 issues: needing kernel headers and changing the var that points to them May 12 19:40:09 with .28 you needed some magic for the first issue May 12 19:40:18 .29 changed some variable names May 12 19:44:24 03koen 07org.oe.dev * r450ce9da... 10/ (1 classes/insane.bbclass): insane.bbclass: add support for ARM EABI/uclibc May 12 20:04:31 hmmm May 12 20:04:55 I could make people's life easier by (ab)using OE vars to autogenerate uclibc configs May 12 20:13:53 koen: we have been mangline .config's for long in OE? May 12 20:14:00 s/mangline/mangling/ May 12 20:29:21 koen: I patched uclibc.inc with one or two variable changes for uclibc, that's all I did before running into other issues. May 12 20:36:22 likewise: NOTE: package uclibc-0.9.29-r0: task do_package: started May 12 20:36:43 koen: :-) May 12 20:41:14 yay May 12 20:42:12 03koen 07org.oe.dev * r443d14e5... 10/ (1 packages/uclibc/uclibc.inc): uclibc.inc: Fix QA issues and prepare for 0.9.29 May 12 20:43:58 :-Ptime for a rm -rf tmp May 12 20:48:25 Hello May 12 20:48:42 I would like to know, does anyone have a (working) bb file for mozilla? May 12 20:48:52 Prefferably the 1.7.12 release May 12 20:49:01 But I can tweak it May 12 20:57:20 likewise, Nail: May 12 20:58:11 03koen 07org.oe.dev * r7aa51798... 10/ (8 files in 5 dirs): May 12 20:58:11 uclibc: add recipe for 0.9.29 May 12 20:58:11 * builds for angstrom/ixp4xxbe May 12 20:58:11 * other distros and machines need to add their own uClibc.machine and uClibc.distro configs May 12 20:58:15 03koen 07org.oe.dev * r18938df9... 10/ (1 packages/uclibc/uclibc-0.9.29/uClibc.machine): uclibc 0.9.29: drop bogus config May 12 21:00:23 koen: tnx, was already working on the efika changes... but I'm a bit slow. May 12 21:07:00 likewise: http://lists.linuxtogo.org/pipermail/openembedded-devel/2007-May/002129.html May 12 21:14:36 koen: Like this? uclibc-0.9.29/{{machine1,machine2,...}/uClibc.machine,{distro1,distro2,...}/uClibc.distro} May 12 21:14:48 yes May 12 21:16:56 03koen 07org.oe.dev * r84a7ae17... 10/ (9 files in 6 dirs): May 12 21:16:56 uclibc 0.9.29: add configs for a780, akita, spitz and hx4700 May 12 21:16:56 * aka "generic pxa270 config" May 12 21:34:55 expat and expat-native are not the same version. simply renaming expat-native_1.95.7.bb to expat-native_2.0.0.bb does the trick. builds fine for me May 12 21:35:21 !oebug 2300 May 12 21:35:22 * * Bug 2300, Status: NEW, Created: 2007-05-10 23:44 May 12 21:35:24 * * xjqian(AT)gmail.com: gnuplot 4.2.0 recipe May 12 21:35:24 * * http://bugs.openembedded.org/show_bug.cgi?id=2300 May 12 21:35:31 they don't need to be the same version, that's why we don't bother updating -native :) May 12 21:35:32 see if I did it right this time May 12 21:37:09 03koen 07org.oe.dev * ra3a91258... 10/ (1 conf/distro/angstrom-2007.1.conf): angstrom: prefer uclibc 0.9.29 May 12 21:39:52 koen: uClibc.config in packages/uclibc/ totally disappears now, or not? May 12 21:40:27 it's not getting used if your machine has support IIRC May 12 21:41:22 koen: OK, but for the new approach (just discussed), uClibc.config is still cat'd in uclibc.inc/do_configure(). It fails here if a uClibc.config cannot be found. May 12 21:41:41 hrm May 12 21:41:56 koen: maybe use an empty one? May 12 21:41:58 I guess I need to be more awake to look at that :) May 12 21:42:04 good idea! May 12 21:43:50 koen: or, I could add an if [ -f file]; then cat here fi; May 12 21:44:51 koen: the test is more robust. May 12 21:45:04 fine by me May 12 22:07:06 koen: angstrom/uClibc.distro does not have a KERNEL_SOURCE, so it is not getting replaced by sed. May 12 22:08:21 * koen scratched head May 12 22:08:36 ./ixp4xxbe/uClibc.machine:68:KERNEL_HEADERS="/usr/include" May 12 22:09:36 so it seems it doesn't need to get replaced May 12 22:09:49 * koen wonders why gcc-cross now fails on locale May 12 22:10:23 is KERNEL_* machine specific? May 12 22:10:54 no idea, I just took the easy way out with splitting the configs May 12 22:11:35 hey likewise May 12 22:11:39 got the board, or still waiting? :> May 12 22:11:58 waiting, distributor has to ask the manufacturer for an OK first. sigh. May 12 22:12:40 doh.. May 12 22:13:25 koen: ok, I now see we should indeed work towards your idea: I could make people's life easier by (ab)using OE vars to autogenerate uclibc configs May 12 22:13:58 Jin^eLD: they have to ask Micronas *IF* we can *LEND* the board... May 12 22:14:18 Jin^eLD: probably to keep the board away from competition... as if. May 12 22:14:47 that sounds really weird May 12 22:15:13 I'd expect them to be happy if someone wants to work with the hardware... May 12 22:16:00 Jin^eLD: Ah well, the last few weeks I have 3 new boards to play with... May 12 22:16:05 heh :) May 12 22:17:11 too bad that they did not release the SDK yet May 12 22:17:17 I mean - publically May 12 22:21:18 Jin^eLD: if you get the binaries from them, you can ask and distribute the sources AFAIK. That's what GPL is all about. May 12 22:21:36 Jin^eLD: there is no SDK at all? May 12 22:21:49 Jin^eLD: there is no SDK at all, I mean with the demo board? May 12 22:22:08 likewise: there is, but its not finished May 12 22:22:23 they have to finish it and release it May 12 22:22:34 so I can't just take the sources and give them away, if you know what I mean :) May 12 22:24:21 rumor goes they have a WinCE customer now, so all effort goes into that direction May 12 22:26:04 Jin^eLD: For GPL it does not matter if something is finished or released, only when it is distributed. May 12 22:28:23 likewise: that may be true, but if you are allowed to have a look at a prerelease, you usually don't give it out behind your business partners back, or something, if you know what I mean May 12 22:29:32 and I can understand that they want to finish it before releasing... May 12 22:31:39 likewise, I think we need to join the libtool list and ask questions May 12 22:32:08 "why does libtool sucks so hard?" May 12 22:32:27 "did you put the million-monkeys theory to the test?" May 12 22:32:35 um May 12 22:32:48 Jin^eLD: sure, but at the moment I would pay for the kit, I would stand in my GPL obtained right May 12 22:32:57 I suspect using a less confrontational apprach would be better :) May 12 22:33:16 I wonder if the gnash guy knows the libtool guy May 12 22:33:36 the gnash guy offered to be helpful with autoconf issues May 12 22:33:54 Crofton: we didnt fix gnash yet :-) May 12 22:33:54 the gnash dude is one of the original autofoo authors :) May 12 22:34:14 yeah May 12 22:34:16 glibc is dead? May 12 22:34:17 koen: yeah he is only 25, look what autofoo did to him :-D May 12 22:34:29 he looked a lot older at fosdem May 12 22:34:39 I thought autofoo is older than that :) May 12 22:34:47 heh May 12 22:34:57 he was a dead head May 12 22:35:10 that lifestyled will age you :) May 12 22:36:16 http://www.welcomehome.org/rob.html May 12 22:43:46 93 bugs fixed over the last week (not too many from me) May 12 22:43:49 Cool! May 12 22:44:17 interesting, Rob hung out a Seneca rocks at one point May 12 22:44:34 He may know some friends of mine, need to ask them next time I see them May 12 22:51:41 whatever, the 'fairly recent (1994) picture' doesn't look like a 12 y.o. at all :) May 13 01:21:38 03pfalcon 07org.oe.dev * re785d0a6... 10/ (1 conf/machine/rx3000.conf): rx3000.conf: Ship BT and WiFi firmware. May 13 01:21:43 03pfalcon 07org.oe.dev * r35a8e968... 10/ (4 files in 2 dirs): h4000.conf, hx4700.conf, rx3000.conf: Ship kernel-module-acx for devices with ACX WiFi. May 13 01:21:52 03pfalcon 07org.oe.dev * r3d46ab3b... 10/ (1 packages/tasks/task-base.bb): May 13 01:21:52 task-base: RRECOMMENDS_task-base-kernel26: Add RTC kernel modules. May 13 01:21:52 * RTC is basic kernel subsys, so here it goes. May 13 01:21:52 * All RTC-related modules are [to be] listed, including CPU-dependent. May 13 01:21:52 Specific machine will obviously build only what it needs, and only that will May 13 01:21:54 be pulled into image. May 13 01:21:57 03pfalcon 07org.oe.dev * r7672b3b8... 10/ (1 conf/machine/htcuniversal.conf): htcuniversal.conf: Remove explicit handling of RTC modules, now done by task-base. May 13 01:22:06 03pfalcon 07org.oe.dev * rac6a4112... 10/ (13 files in 8 dirs): linux-handhelds-2.6: Add 2.6.21-hh2, with defconfigman-generated configs per RFC. **** ENDING LOGGING AT Sun May 13 02:59:59 2007