**** BEGIN LOGGING AT Mon Jan 08 02:59:57 2007 Jan 08 03:14:32 03likewise 07org.oe.dev * r9f8a39fa... 10/ (1 packages/fastcgi packages/fastcgi/fastcgi_2.4.0.bb): fastcgi: New package, version 2.4.0. FastCGI is a library for developing FastCGI webserver-backend gateways. Jan 08 03:14:38 03hrw 07org.oe.dev * r4a4b64f2... 10/ (1 packages/linux/linux-mtx-3_2.6.15.4.bb): linux-mtx-3: DEPEND on u-boot not on uboot Jan 08 03:14:43 03xora 07org.oe.dev * r031a650e... 10/ (3 files in 3 dirs): xserver-common_1.13.bb : add Ximageon support Jan 08 03:14:47 03pfalcon 07org.oe.dev * r94b9aca1... 10/ (1 packages/fbreader/files/fbreader-0.7.4q_buildsys_oe.patch): fbreader 0.7.4q: Add missing patch. Jan 08 04:20:43 hi! can someone please help me - i'm building OE, and it finished building task-base, but i have nothing in /build and it seems as if the tmp dir is huge... is that normal/what should i do? Jan 08 04:27:17 anyone..? Jan 08 04:30:02 shirour: tmp/deploy/images should contain the resultant image(s) Jan 08 04:30:28 v8jlene: thanks..! but why isn't the stuff in the build dir? Jan 08 04:30:44 shirour: err.. what's the "build" dir? Jan 08 04:30:49 i don't have image under tmp/deploy Jan 08 04:31:07 like /stuff/build in the tutorial Jan 08 04:31:18 shirour: Oh, task-base doesn't build images. I build bootstrap-image Jan 08 04:31:32 v8jlene: and where does it put it? Jan 08 04:32:20 The image ends up in tmp/depoly/images, the files that go in it are in tmp/rootfs Jan 08 04:32:31 tmp/deploy/images that is Jan 08 04:32:42 ie, like this: tmp/deploy/images/bootstrap-image-titan-20070105144218.rootfs.tar.gz Jan 08 04:33:16 but i don't have /stuff/tmp/........ Jan 08 04:33:38 another thing strange: when i now hit "bitbake opie-base" i get "ERROR: Nothing provides dependency opie-base" Jan 08 04:34:02 my images are in /tmp/deploy/glibc/images Jan 08 04:34:57 no wait... there's nothing interesting there... Jan 08 04:35:00 what have i done wrong? Jan 08 04:35:04 have a look here: http://pastebin.com/854027 Jan 08 04:35:24 TMPDIR in local.conf determines where everything ends up Jan 08 04:35:53 I have: TMPDIR = /home/lenehan/devel/oe/build/titan-glibc-25/tmp Jan 08 04:36:09 thanks..! but where did my files go then..? and what's the "ERROR: Nothing provides dependency opie-base" message when building opie? Jan 08 04:36:36 As for opie-base, I can't really help with that. None of my devices have a UI... so I don't build any UI stuff. Jan 08 04:37:52 that's strange - i have nothing in /build/tmp, and i haven't used the "TMPDIR" var in local.conf Jan 08 04:38:07 and it looks as if that's the default dir Jan 08 04:38:12 but i just have nothing in there Jan 08 04:38:39 I have no idea where you files went ;) Jan 08 04:38:47 oh wait.... i know what's the problem... i've ran "bitbake" from /oe instead of /oe/build Jan 08 04:39:11 TMPDIR defaults to $TOPDIR/tmp - and I presume $TOPDIR is where you run bitbake from. Jan 08 04:39:27 oh i see Jan 08 04:39:31 so that's the problem Jan 08 04:39:41 but i seems it was able to find the conf file... Jan 08 04:39:49 should i move the tmp dir to /build ..? Jan 08 04:40:45 You need to make sure you always run bitbake from the same place... it's up to you if you want it somewhere else. Moving it won't work though - you'd need to delete it and start again in the new location. Jan 08 04:42:12 i don't get it... it says it finished building but i can't find it... Jan 08 04:42:21 "NOTE: package angstrom-version-test-20070108: completed" Jan 08 04:43:08 Nothing in the tmp/rootfs or tmp/deploy dirs? What did you run? Jan 08 04:44:12 i have "glibc" under "deploy" Jan 08 04:44:32 and there i have "images/omap5912osk/" and Jan 08 04:44:38 u-boot-omap5912osk-LABEL.2006.06.30.2020-r2.bin, vmlinux-2.6.19-omap1-omap5912osk-20070107172910.bin Jan 08 04:44:58 and also a huge amount of ipkg files under glibc/ipk Jan 08 04:45:58 the u-boot-...bin thing looks like the image you create, the vmlinux-... looks like the kernel and the ipk directory contains all the packages you created (what it's doing under glibc I have no idea) Jan 08 04:47:51 very strange... but the uboot image is tiny Jan 08 04:49:06 shirour: BTW, I don't build UI stuff, or angstrom or anything like that. The people who do are sleeping at this time of day - they should start showing up in about 5 hours. Jan 08 04:49:53 what were you building? Jan 08 04:51:41 angstrom.... ok.... thanks ..! Jan 08 04:55:10 still i can't find my images... very strange Jan 08 04:55:13 Heh, yeah but you didn't "bitbake angstrom" though, what did you bitbake? I'm not sure either of those are the root filesystem... do you have IMAGE_FSTYPES set in local.conf or the machine file you are using? Jan 08 04:55:54 angstrom doesn't set IMAGE_FSTYPES, and if it isn't set then I don't think you'll get any images created Jan 08 04:55:57 yes IMAGE_FSTYPES = "jffs2 tar" Jan 08 04:56:19 but all i've built is "task-base" Jan 08 04:57:06 I don't think tar is a valid type, tar.gz is (if you can get to openembedded.org the types are listed in the reference section of the user manual) Jan 08 04:57:32 oh, ok... thanks..! Jan 08 04:57:32 task-base doesn't create any images... you need an image such as bootstrap-image Jan 08 04:57:42 great! thanks a lot!!! Jan 08 04:59:06 Looks like you should use angstrom-bootstrap-image for angstrom.. Jan 08 04:59:21 or one of the other ones in packages/angstrom Jan 08 04:59:52 right..! thanks again! bbl Jan 08 05:49:48 morning Jan 08 06:24:54 morning guys Jan 08 06:32:06 * mickey|earlybird mutters about outside being dark early in the morning Jan 08 07:33:22 hmm. If I move my local.conf to another directory (whose parent is also defined in BBPATH) AFTER I have done some building, do I have to delete the existing bb_cache.dat? For some reason bitbake still tries to read from the ...org.openembedded.dev/conf/local.conf Jan 08 07:39:18 Answering to myself: "yes, it seems that I have to delete it". Jan 08 07:41:26 heh, trying rocks ;) Jan 08 07:42:04 heh , i have some problems with the source format Jan 08 07:42:22 I was just afraid that deleting the cache would somehow cause every depended recepie to be built again Jan 08 07:42:45 the compilation worked but the final error message couldn't be found be tar.gz format Jan 08 07:42:55 no need to worry, the cache is only responsible for parsing Jan 08 07:42:56 is it possible to add more than one format ? Jan 08 07:43:29 anushsh: can you pastebin the complete error? Jan 08 07:43:31 ~pastebin Jan 08 07:43:33 [pastebin] a place to paste your stuff without flooding the channel - try http://pastebin.ca, or http://channels.debian.net/paste Jan 08 07:43:51 yeah cool Jan 08 07:44:20 The current versions of bitbake and monotone are lovely. The performance, that is. Year ago when I did my thesis, it wasn't as sweet as it is now. Jan 08 07:44:49 absolutely Jan 08 07:45:03 we've been through a lot of performance enhancements in the last 12 months Jan 08 07:45:11 i'm going to have a talk about that @ FOSDEM Jan 08 07:45:21 and the best is yet to come Jan 08 07:45:24 can you say GUI? ;) Jan 08 07:45:37 Nice. Jan 08 07:46:07 Though I feel a bit rusty, when I finished my work last spring, I haven't played with OE at all. Jan 08 07:46:31 Now I'm trying to see if it's possible to compile Ã¥ngström for 770, in any level. Jan 08 07:51:46 that'd be cool. I'd love to see the 770 being a target for stuff != maemo as well Jan 08 07:52:05 hmm, my Makefile.am puts include files into a wrong include directory Jan 08 07:52:23 any idea how i can override DESTDIR for includes being the package name directory? Jan 08 07:59:07 morning all Jan 08 07:59:16 mickey|earlybird: you here? so early?? Jan 08 07:59:18 wow Jan 08 07:59:43 indeed. can't quite believe it Jan 08 07:59:44 morning hrw Jan 08 08:01:46 speaking about n770... task-base builds for it Jan 08 08:02:07 mickey|earlybird: expect mail about arm machines status soon ;D Jan 08 08:02:22 goood! Jan 08 08:02:32 and can someone restore topic? Jan 08 08:06:57 something like that? Jan 08 08:07:28 yep Jan 08 08:09:04 mickey|earlybird: what you work on now? Jan 08 08:09:45 building some openmoko apps and trying how they look on the production LCM Jan 08 08:13:19 fscking disconnecto.pl Jan 08 08:13:26 heh Jan 08 08:13:43 we have 100Mbit connection through half of city Jan 08 08:13:53 but it have so many drops that... ARGH Jan 08 08:16:31 morning all Jan 08 08:16:37 hi Richard Jan 08 08:17:15 morning rich Jan 08 08:17:16 ard Jan 08 08:20:21 does someone got hit by hanging do_configure when config is ok? Jan 08 08:20:32 morning Jan 08 08:20:45 omap1610h2 hangs when used with Angstrom Jan 08 08:23:02 btw - all those omapXXXXyy machines other then omap5912osk were added by kergoth when he was working in TI or someone else? Jan 08 08:25:54 brb Jan 08 08:25:57 morning Jan 08 08:35:16 palmld built Jan 08 08:40:49 morning all Jan 08 08:41:16 hey ade|desk Jan 08 08:41:21 hey likewise Jan 08 08:47:39 XorA: hey Jan 08 08:54:34 moin Jan 08 09:24:29 good morning everyone Jan 08 09:24:55 back from snowy vacation Jan 08 09:27:12 moin vlo Jan 08 09:47:33 good morning all Jan 08 09:47:42 hi koen Jan 08 09:47:45 hrw|work: laibsch reported hangs with bitbake 1.6 on configure Jan 08 09:48:02 posokolvsky: rx3000 needs to be added to linux-handhelds-2.6.inc Jan 08 09:49:15 koen: hey Jan 08 09:49:43 posokolvsky: htcblueangel fails like h1910 Jan 08 10:20:08 XorA: cheers on the imageon bit Jan 08 10:20:13 XorA: and http://uwog.net/blog/index.php/archives/63 Jan 08 10:21:10 koen: collab scares me Jan 08 10:21:56 XorA: don't you want your collegues to insert goatse pictures in your report? Jan 08 10:21:56 koen: Xv rocks Jan 08 10:22:51 koen: Ive done things like that myself under windows by accidently pasting into wrong window Jan 08 10:26:01 hrw|work, thanks for the botmail .. will start fixing Jan 08 10:26:31 rob_w: config is fixed now Jan 08 10:26:37 kernel fails anyway Jan 08 10:27:00 yeah the main part is the kernel to move naming totally and then fix oe afterwards .. Jan 08 10:27:19 rob_w: http://ewi546.ewi.utwente.nl/tinderbox/showlog.pl?machine_id=177&logfile=20070108112351.log Jan 08 10:29:02 hmm h4000_set_led argh .. thats when people move to the new led scheme Jan 08 10:29:30 let me grep some time this week and sort his out and fix the naming .. thanks for your intesiv builds anyway ! Jan 08 10:30:37 thx for fixing Jan 08 10:31:35 XorA: you are maintainer of smdk2440 - right? Jan 08 10:32:09 XorA: can you refresh kernel config? Jan 08 10:33:29 .config | 67 ++++++++++++++++++++++++++++++++++++++++++++++++---------------- Jan 08 10:33:29 1 files changed, 51 insertions(+), 16 deletions(-) Jan 08 10:33:38 XorA: quite big change... Jan 08 10:38:10 hrw|work: I dont understand, kernel config is fine for 2.6.19 Jan 08 10:39:12 XorA: here it stopped on IPC Namespaces (IPC_NS) [N/y/?] (NEW) Jan 08 10:39:16 question Jan 08 10:41:25 XorA: when I cd into dir and make 'ARCH=arm make oldconfig' I have to answer to some config questions (ENTER, ENTER, ENTER press is enough) and then build follows Jan 08 10:41:33 hrw|work: pushed Jan 08 10:41:38 thx XorA Jan 08 10:42:06 03xora 07org.oe.dev * rbbbd3918... 10/ (1 packages/linux/linux-smdk2440/defconfig-smdk2440): defconfig-smdk2440 : refresh Jan 08 10:48:00 hrw|work: just confirm yes or no to see if i have understood this line: ${@base_contains("COMBINED_FEATURES", "irda", "opie-irdaapplet", "",d)} it means: Jan 08 10:48:34 hrw|work: if irda is defined in both machine and distro then install opie-irdaapplet. Am i right? Jan 08 10:51:03 no, that doesnt seem right Jan 08 11:01:47 yes Jan 08 11:01:59 if irda is defined in both machine and distro then install opie-irdaapplet Jan 08 11:05:09 but, i cant see "irda" in the angstrom.conf file Jan 08 11:06:37 then this is a bug in distro config Jan 08 11:06:52 is it? Jan 08 11:07:13 03koen 07org.oe.dev * r36c8c4c3... 10/ (1 packages/tasks/task-base.bb): task-base: add 'phone' to machine features Jan 08 11:07:29 goxboxlive: I see irda in angstrom config Jan 08 11:07:33 DISTRO_FEATURES = "nfs smbfs wifi ppp alsa bluetooth ext2 irda pcmcia usbgadget usbhost pci" Jan 08 11:07:36 goxboxlive: look in include/angstrom.inc Jan 08 11:07:47 of course. txh Jan 08 11:10:11 Do you accept "mtn diff file" or do i have to use "diff -uN file file" ? Jan 08 11:11:33 mtn diff is ok Jan 08 11:12:02 good, i'll place some patches today at the bugtracker. Jan 08 11:15:36 goxboxlive: no multibugs! Jan 08 11:15:57 goxboxlive: split them into separate bugs Jan 08 11:16:02 koen: I know, just single ones. Jan 08 11:19:15 goxboxlive: and don't include bogus modules in that patch! Jan 08 11:20:24 koen: I have removed the kerberos stuff, and the ppp stuff. The RTC stills exists, is that ok? Jan 08 11:20:35 yes, that's ok Jan 08 11:31:53 | cc1: error: unrecognized command line option "-mapcs-32" Jan 08 11:31:53 | cc1: error: unrecognized command line option "-mshort-load-bytes" Jan 08 11:32:05 howdy Jan 08 11:32:08 someone familiar with it? gcc 3.3 only? Jan 08 11:32:31 hi noha Jan 08 11:33:15 hrw|work: I read your article about setting up your oe environment. Very nice! Jan 08 11:33:28 noha: part II or part I? Jan 08 11:34:21 As there isn't noted any part it should be part I :) Jan 08 11:35:35 Is it so that some of the variables are just evaluated in specific locations? Let's say DISTRO_* are only effective if the are in a machine target? Jan 08 11:36:31 I did conversion for the simpad machine to feed in the new handling. But if I like to customize my image (not the official distro) I'm quite unsure very to put extra packages Jan 08 11:36:49 I copied conf/machine/simpad.conf and added DISTRO_EXTRA_RDEPENDS Jan 08 11:36:52 noha: create own image Jan 08 11:37:29 Ok, sounds reasonable! Jan 08 11:38:51 found. gcc 2.95 options it was.. Jan 08 11:38:54 Another question. I built the simpad target a few month ago. I know it used kernel 2.4.27. Now I built the image with kernel 2.4.27 and I get a stack trace while booting. Jan 08 11:39:28 hrw|work : i have been getting this http://pastebin.ca/310087 error message Jan 08 11:39:44 late minute compilation failure Jan 08 11:40:26 good morning Jan 08 11:41:17 I assume that the kernel source didn't change much. What would be the best candidate which introduces these problems? I know the question isn't very clever, but. I compiled with gcc 3.4.4 Jan 08 11:43:21 anushsh: please show more of log Jan 08 11:43:35 anushsh: and check does archive which you try to unpack is archive really Jan 08 11:44:06 2.6.9 arm kernel stinks Jan 08 11:47:10 ok Jan 08 11:48:14 hi Jan 08 11:48:37 zecke: Conformist! Jan 08 11:49:02 noha: I started drinking coffee you towel! Jan 08 11:49:13 noha: is your simpad booting yet? Jan 08 11:49:49 zecke: Nope. I started over to use the familiar 0.8.4 package to see if everything else would work. Jan 08 11:50:07 zecke: I'll have a look after the kernel by today evening Jan 08 11:50:26 zecke: the familiar package I use has the same kernel Jan 08 11:50:41 zecke: And you are right. upgrade sucks! Jan 08 11:51:02 noha: ask steve to return my simpad and we cnan work on the 2.6er kernel Jan 08 11:51:19 zecke: How told me that can't even find it anymore :) Jan 08 11:52:25 hi Cliff Jan 08 11:52:52 noha: it was on his desk, last time..... Jan 08 11:53:21 I created a packages/images/ directory under build/ and placed a my-image.bb file there. Shouldn't bitbake find it? Jan 08 11:53:43 noha: yes it should find it Jan 08 11:53:58 ERROR: Nothing provides dependency my-image Jan 08 11:54:26 /opt/oe/build is in BBPATH Jan 08 11:54:35 noha: is it in BBFILES as well? Jan 08 11:54:49 noha: bbpath is used for .inc, .conf and .bbclass Jan 08 11:55:46 my efika boots from the hard drive with the image from OE! Jan 08 11:56:03 zecke: ok, it wasn't in BBFILES Jan 08 12:04:33 noha: I hate compiled languages ;) I hate to wait Jan 08 12:05:20 zecke: There is no way avoiding to compile. It's just when and _how much_!! Jan 08 12:05:59 noha: I want it just in time Jan 08 12:08:40 Can I copy a task file to my own directory and included with some changes? My concerns are about redefinition of RDEPENDS_* stuff Jan 08 12:09:03 zecke: My stuff gets compiled when I press Alt+S :) Jan 08 12:20:57 hi all Jan 08 12:23:24 disaster: hi Jan 08 12:23:30 hi all Jan 08 12:24:16 i have a problem with my embedded system Jan 08 12:25:21 when i boot Jan 08 12:25:24 i have an error Jan 08 12:25:34 cache flushes not supported Jan 08 12:25:46 somebody can help me? please Jan 08 12:27:18 RP: is it possible to do 64bit math inside the kernel on arm? Jan 08 12:27:51 babar: my glass ball is broken Jan 08 12:28:09 kernel is X86 Jan 08 12:28:16 processor is via Jan 08 12:31:02 i use a compact flash for boot system Jan 08 12:31:10 Good morning Jan 08 12:31:23 Is perl not buildable for openzaurus-unstable at the moment? Jan 08 12:32:29 and if I need a driver for ide? Jan 08 12:33:03 babar: if you connect CF as ide then you need ide driver Jan 08 12:34:16 the compact flash is detected like a hard disk Jan 08 12:34:53 the compact flash reader is integrated in motherbord Jan 08 12:39:10 then check how it is integrated Jan 08 12:39:29 it can be ATA, pcmcia, usb and few others Jan 08 12:41:46 ATA Jan 08 12:42:37 compact flash is detected like ATA Jan 08 12:46:33 then you need ata driver in kernel Jan 08 12:48:44 when i do with bitbake? Jan 08 12:50:41 grrr what gives... I've got PREFERRED_PROVIDER_virtual/xserver = "xserver-kdrive" in my local.conf but I still get "NOTE: consider defining a PREFERRED_PROVIDER to match runtime virtual/xserver" Jan 08 13:04:59 Hi! Jan 08 13:05:01 Huh, how come updater.sh is not actually a shell script? Jan 08 13:05:13 it actually is Jan 08 13:05:18 it's just uhm.... Jan 08 13:05:20 encrypted Jan 08 13:05:22 *cough* Jan 08 13:05:24 ~lart Sharp Jan 08 13:05:25 ah Jan 08 13:05:25 * ibot beats Sharp severely about the head and shoulders with a rubber chicken Jan 08 13:05:29 Mmm, obfuscation. Jan 08 13:05:35 ~lart Sharp for a loughable attempt to lock us out Jan 08 13:05:35 * ibot hits Sharp with an anvil and laughs with a contralto voice ... Haha Ha HA Ha for a loughable attempt to lock us out Jan 08 13:05:36 hrw|work, is there any info about how tests is actually performed? some of failures may be due to procedure. Jan 08 13:06:09 And here I was thinking sharp actually supported installing third-party distributions. Jan 08 13:06:14 heh Jan 08 13:06:15 no way Jan 08 13:06:20 the opposite is true Jan 08 13:06:29 Sharp Germany was very supportive Jan 08 13:06:37 but alas, they got canned Jan 08 13:08:45 Are there plans to attend this: http://fosdem.org/2007/beerevent Jan 08 13:09:28 absolutely Jan 08 13:09:39 excellent Jan 08 13:10:05 Astrid i crawling distance? Jan 08 13:10:10 psokolovsky__: echo machine to conf/auto.conf;bitbake task-base Jan 08 13:10:16 somebody know this error? cache flushes not supoported Jan 08 13:10:22 please help me Jan 08 13:10:40 yeah, astrid is like 10 minutes walking distance from roy despagne Jan 08 13:10:40 psokolovsky__: (reverse-i-search)`for': unset MACHINE ;for oemach in mtx-3 omap1510inn omap1610h2 ; do date; echo $oemach; echo "MACHINE='$oemach'">conf/auto.conf; bitbake task-base;echo $oemach;done;date Jan 08 13:10:43 hrw|work, what kernel version is used? Jan 08 13:10:56 psokolovsky__: default angstrom one Jan 08 13:11:33 hrw|work, that's caveat: default ansgtrom's doesn't work for each machine, some needs more fresh one Jan 08 13:12:09 psokolovsky: then submit patch to angstrom config Jan 08 13:12:23 psokolovsky: I will send info to OEML soon Jan 08 13:13:16 somebody can help me??? Jan 08 13:14:22 hrw|work, that's caveat: adding machine to angstrom config would mean it gets supported by angstrom, and that apparently require more administrative actions ;-). I want to raise q of defining more formal procedure of adding machines to angstrom, and to define several levels of support - from alpha to "officially supported" ;-) Jan 08 13:14:31 hrw|work, ok, will be waiting for it Jan 08 13:14:52 Okay, I hope I have an at least semi-working filesystem tarball here... Jan 08 13:15:28 ood break Jan 08 13:15:44 food break Jan 08 13:16:46 babar, try googling for the error Jan 08 13:17:08 it is likely that an expert on your hw is not around at them moment Jan 08 13:17:54 in general we are not "unhelpful" rather we are not all knowing about every piece of hardware :) Jan 08 13:21:45 Hm, won't boot... I wonder what went wrong? Jan 08 13:22:27 "won't boot" as in the screen stays blank when I press the power button after update Jan 08 13:23:10 !oebug 1771 Jan 08 13:23:11 * * Bug 1771, Status: RESOLVED, Created: 2007-01-07 12:13 Jan 08 13:23:12 * * abondarenko(AT)gmail.com: Only ancient version of the oe-adapting patch is avaliable Jan 08 13:23:13 * * http://bugs.openembedded.org/show_bug.cgi?id=1771 Jan 08 13:23:24 !oebug 1776 Jan 08 13:23:26 * * Bug 1776, Status: NEW, Created: 2007-01-07 17:50 Jan 08 13:23:27 * * pmiscml(AT)gmail.com: Setting MACHINE in environment is fatal when building binutils-cross Jan 08 13:23:28 * * http://bugs.openembedded.org/show_bug.cgi?id=1776 Jan 08 13:23:42 RP: nice bug for you ;-) ^^^ (1776) Jan 08 13:34:07 Hm, either compiling a custom kernel from scratch is so trivial that no one has bothered to write a guide about it and I'm just missing something obvious, or it's so hard that no one expects anyone else to want to do it Jan 08 13:34:51 SRC_URI = "blah" ; inherit kernel Jan 08 13:34:57 ah, that's better. VNC broke :-/ Jan 08 13:35:47 koen: Let's say that I just want to know what needs to be done and why, for whatever reason? Jan 08 13:36:07 DataBeaver: have a look at linux-rp.inc Jan 08 13:36:10 k Jan 08 13:36:39 DataBeaver: linux_2.6.19.bb is even simpler ;D Jan 08 13:38:00 Do we need all these different versions of sqlite? I am thinking about how best to deal with bug 1768. Jan 08 13:38:10 Hm, are those supposed to be in packages/linux? Jan 08 13:38:40 It seems sqlite upstream removed a couple of older source tars. There are mirrors which could be used. Jan 08 13:38:49 !oebug 1768 Jan 08 13:38:50 * * Bug 1768, Status: NEW, Created: 2007-01-07 10:05 Jan 08 13:38:51 * * dothedog(AT)comcast.net: sqlite3 build error: fetch http://www.sqlite.org/sqlite-3.2.5.tar.gz not found. Jan 08 13:38:52 * * http://bugs.openembedded.org/show_bug.cgi?id=1768 Jan 08 13:39:17 I am suffering brain rot, I cant remeber how to divide a 64bit number held in two 32bit parts by a 32bit number Jan 08 13:39:27 linux-openzaurus_2.6.18.bb is the closest I can find to the latter and can't find the former at all... Jan 08 13:39:49 ah, you're using the obsolete354x branch ;) Jan 08 13:40:03 Laibsch, just stuff mirror there, and don't start removing spree ;-) Jan 08 13:40:05 That's entirely possible. Jan 08 13:41:56 Gnh, www.openembedded.org is being slow... Jan 08 13:42:21 Laibsch: rather upgrade then using upstream forgotten versions Jan 08 13:43:14 DataBeaver: ! Jan 08 13:43:46 hrw|work, Laibsch: preferrably upgrade GPE to use sqlite3 at the same time, 'cos they're using damn-long forgotten sqlite2 ;-\ Jan 08 13:43:47 DataBeaver: the machine load is low, so i guess its a newtwork issue Jan 08 13:44:03 Let's try from another machine... Jan 08 13:44:35 brb, rebooting Jan 08 13:44:44 psokolovsky: sqlite 2.x is maintained upstream still Jan 08 13:45:23 Nope, doesn't work too well from my home machine either. "Waiting for reply..." Jan 08 13:45:47 DataBeaver: I can't get anything from it either Jan 08 13:46:29 psokolovsky: Which version of bitbake was that bug with? Jan 08 13:47:02 RP: 1.6 branch from svn ;-) Jan 08 13:48:00 psokolovsky: Can you have a look at the generated script in the work/whatever/temp directory and see if they reference MACHINE? There should be an "unset MACHINE" there which I suspect will be missing... Jan 08 13:49:20 Okay, I'm missing kernel commandline apparently. Jan 08 13:49:20 RP: oops, I rebuilt over broken files already. any hint from where that unset should come (bitbake itself/some .bbclass), so I can debug/fix it later? Jan 08 13:50:05 hrw|work: I agree. I wonder if just removing them is not a bit too harsh, though. I don't know if anybody depends on older versions. Jan 08 13:50:49 Laibsch: iirc sqlite3 does not have abi changes which would broke it Jan 08 13:51:01 psokolovsky: base_after_parse in base.bbclass sets the unexport flag. bitbake handles that in data.py's emit_vars Jan 08 13:51:23 hrw|work, they for sure have sql/db behavior changes Jan 08 13:51:37 RP: ok, thanks for hint! Jan 08 13:52:52 hrw|work : I am giving the url of the debian kernel source for kernel compilation Jan 08 13:53:07 with an arm patch Jan 08 13:53:09 anushsh: pastebin recipe please Jan 08 13:54:00 koen, angstrom boots from the hard drive on my efika Jan 08 13:54:22 Crofton: yay! Jan 08 13:54:30 now I should really fix my dhcp server Jan 08 13:55:06 my progear boots angstrom too Jan 08 13:55:14 console=ttyS0,115200n8 <-- What's this for? qemu or something? Jan 08 13:56:29 psokolovsky: A quick test here with bitbake 1.6 says "unset MACHINE" makes it into the script. If that isn't enough, you'll have to find out what it! ;-) Jan 08 13:56:47 s/it/is/ Jan 08 13:57:50 ok ;-) Jan 08 13:58:38 anushsh: yes, I'm around Jan 08 13:58:42 RP: but do you have an idea is bitbake or some bbclass ever uses constract like "${TARGET_ARCH}:${MACHINE}" or it's fully binutils' deed? Jan 08 13:59:10 koen : in http://www.openembedded.org/~koen/amida/linux-amida_2.4.18.bb Jan 08 13:59:31 psokolovsky: binutils Jan 08 14:03:59 to add driver i must add a specifical line in my local.conf or is an diferent operation ? Jan 08 14:04:22 what SRC_URI should i give Jan 08 14:04:22 i was trying with the debian source Jan 08 14:04:34 why not kernel.org? Jan 08 14:05:08 for me ? Jan 08 14:07:50 glibc 2.5 should works with 2.4.19 kernel? Jan 08 14:09:46 univac, ditch 2.4 kernel. start hacking on 2.6 for h3800. Jan 08 14:10:07 You? Jan 08 14:10:34 started Jan 08 14:10:35 univac: glibc 2.5 req 2.6 kernel Jan 08 14:10:47 univac, no. no device. Jan 08 14:10:53 Ah, now i know why my ipaq won't start ;] Jan 08 14:10:55 psokolovsky: ahh, yep i want Jan 08 14:11:02 COMPATIBLE_HOST = "arm.*-linux" Jan 08 14:11:05 its obsolete Jan 08 14:11:21 hwthen, he can't run on 2.4? Jan 08 14:11:21 psokolovsky: I wanted to say "I could send you mine", but then I realised it isn't my h3870 Jan 08 14:11:23 hrw|work: ^ Jan 08 14:12:33 hrw|work: OK, I will remove unfetchable versions, then, after asking the bug reporter why OE did not build the more recent version Jan 08 14:12:34 koen, ;-) np, I have enough to hack on for now, and expect some old stuff like that soon possibly too ;-) Jan 08 14:14:34 koen, I actually want to collect all current cpu's, in this regard I really miss that acer ;-E. Did you hear anything from letux guys since then? Jan 08 14:15:47 psokolovsky: which cpu you already have? Jan 08 14:16:27 hrw|work, only pxa255 & 270 for now only, but negotiating on SA & OMAP ;-) Jan 08 14:16:49 glibc-omitfp hmm Jan 08 14:17:29 * koen has sa, pxa, ep93xx and omap Jan 08 14:17:37 psokolovsky: you also lack nasty pxa250 Jan 08 14:17:52 Do I need to worry about the durability of the flash when updating the kernel? Or does it last like thousands of updates? Jan 08 14:18:19 * hrw|work has sa, pxa255, pxa270 Jan 08 14:18:28 hrw|work, yep, but another guy got h5000 to hack on, so it should be covered ;-) Jan 08 14:19:25 psokolovsky: iirc 5400 was 250, 5500 255 so he could get 5500... Jan 08 14:19:26 my h5000 is pxa255 :) Jan 08 14:19:33 my my c700 pxa250 :( Jan 08 14:21:20 yep, it seems to have been h5555. so, no luck ;-) Jan 08 14:22:47 Hm, cmdline should be in place now, but still nothing on screen... Jan 08 14:25:46 Ha, nope, h3900 is pxa250, so I'm covered! ;-D Jan 08 14:27:01 i have a small question Jan 08 14:27:12 in menu. list for the grub Jan 08 14:27:29 http://pastebin.ca/310182 Jan 08 14:27:52 koen, should I go ahead and push fixes for the missing sizeof's in powerpc-linux? Jan 08 14:28:16 I just added float for omniorb Jan 08 14:28:31 why we have root (hd0,5) Jan 08 14:28:52 disaster: because (hd0,5) is /dev/hda6 Jan 08 14:29:00 Crofton: yes, please do Jan 08 14:29:00 what is the difference with (hd0,0) Jan 08 14:30:11 and (hd0,0) Jan 08 14:31:06 hmm.. linux-gumstix has few NICE patches Jan 08 14:31:18 cpufreq-ondemand by default Jan 08 14:31:35 or more steps for pxa25x Jan 08 14:31:53 100/133/200/266/300/333/400/530 Jan 08 14:32:05 530? Jan 08 14:33:05 intel doc give this freq Jan 08 14:33:23 I haven't seen a pxa25x go to 530 Jan 08 14:33:33 only when I *cough* overclocked it in *cough* wince Jan 08 14:33:42 Referrin to the previous mention about obsolete354x branch... How do I get the non-obsolete one? Jan 08 14:33:46 you did not read pdax guys Jan 08 14:34:10 DataBeaver: .oz354x is nearly year old (with updates) Jan 08 14:34:18 DataBeaver: use .dev for new machines Jan 08 14:34:50 I do have org.openembedded.dev Jan 08 14:35:27 Maybe the mtn db was just old, I did a pull and it got 1636 revisions... Jan 08 14:35:27 DataBeaver: 'mtn pull ; mtn up' Jan 08 14:36:11 Was from last November I think... Jan 08 14:36:16 * mrz80 is away: Metaphysically I'm still here; physically, I'm elsewhere Jan 08 14:37:39 Yup, got stuff now. Jan 08 14:38:42 ok. gumstix task-base built Jan 08 14:39:29 Are the kerne patches documented somewhere? Like a brief description of what they do? Jan 08 14:50:08 ARM guys: GET MAIL :D Jan 08 14:51:25 hrw|work: is xxs1500 not mipsel ? Jan 08 14:53:12 ade|desk: hm. Jan 08 14:53:27 ops. yes Jan 08 14:53:50 I added it to list just because its kernel is not fetchable and then started testing only arms Jan 08 14:55:10 now started mipsel targets Jan 08 15:04:21 hi mickeyl Jan 08 15:04:28 Huh, bitbake tries to fo make install for quilt-native-0.45 first thing after I say build linux? Before even downloading anything? Jan 08 15:05:34 s/fo/do/ Jan 08 15:06:07 DataBeaver: it need to install all needed tools Jan 08 15:06:43 I know that, but how does it expect to install something it didn't build yet? Jan 08 15:07:19 quilt-native do_install() without other steps? Jan 08 15:07:26 http://tdb.fi/~tdb/bb.txt Jan 08 15:07:49 DataBeaver: clean tmp Jan 08 15:08:06 It was clean before doing that Jan 08 15:08:37 i.e. no tmp, start bitbake, parse, build linux Jan 08 15:09:04 DataBeaver: then do 'rm -rf tmp;bitbake virtual/kernel' Jan 08 15:09:38 k, let's see... Jan 08 15:12:30 Is it possible to make bitbake only parse the files needed to build a given target? Jan 08 15:13:04 DataBeaver: only first parse is slow Jan 08 15:13:33 Yeah, but nuking tmp cleared the cache... Jan 08 15:13:54 Oh well, it's almost done Jan 08 15:14:08 I just hope I don't have to nuke it too many times :P Jan 08 15:14:40 Should which virtual/kernel show what variant it's going to build? Jan 08 15:15:25 machine config should have PREF_PROV_virtual/kernel set Jan 08 15:15:31 morning Jan 08 15:15:34 Still getting the same error Jan 08 15:15:51 It tries to start by installing quilt, without building it first Jan 08 15:19:43 Ookay... Did which linux after that failed build and got TypeError: list indices must be integers Jan 08 15:23:26 I probably have a misconfiguration somewhere since which virtual/kernel doesn't show anything that looks like Zaurus... Jan 08 15:23:45 cu Jan 08 15:24:24 DataBeaver: it should build linux-rp Jan 08 15:28:09 When trying to bitbake bootstrap-image I get a compile error on glibc "/stuff/build/tmp/work/i486-linux/glibc-2.3.5+cvs20050627-r13/build-i486-linux/linuxthreads/libpthread_pic.a(pthread.os): In function `pthread_initialize': Jan 08 15:28:09 pthread.c:(.text+0xca3): undefined reference to `_res'" Jan 08 15:28:16 Am i doing something stupid? Jan 08 15:30:44 http://bugs.openembedded.org/show_bug.cgi?id=291 Jan 08 15:30:51 yep, I was doing something stupid Jan 08 15:30:57 not reading the bugtracker beforehand Jan 08 15:31:29 Okay, added distro to conf and it works much better Jan 08 15:40:15 koen: What was that about the corrupt .desktop file you mentioned yesterday? After almost every GUI app which I installed broke opie I just removed the .desktop file from the last one and opie works. Jan 08 15:40:29 Did the .desktop format change for opie in the meantime? Jan 08 15:40:39 Laibsch: no idea, just an educated guess :) Jan 08 15:40:51 VEERRY educated! Jan 08 15:40:55 chapeau! Jan 08 15:41:01 ~hail koen Jan 08 15:41:03 * ibot bows down to koen and chants, "I'M NOT WORTHY!!" Jan 08 15:43:14 * Laibsch inspects older .desktop files and newer .desktop files Jan 08 15:45:19 anyone checked whether our patches still apply to gtk+ 2.10.7 ? Jan 08 15:45:54 is that out already? Jan 08 15:46:08 I could swear it wasn't on gtk.org this morning :O Jan 08 15:46:10 dunno, got a mail on it two days ago Jan 08 15:46:16 via gtk-list Jan 08 15:46:34 not on the frontpage, yeah Jan 08 15:46:45 http://mail.gnome.org/archives/gtk-devel-list/2007-January/msg00052.html Jan 08 15:46:53 we can drop the xft patch IMO Jan 08 15:47:00 it's on the ftp though Jan 08 15:49:23 Hm, none of the patches under linux-rp-2.6.19 seem directly relevant to getting text on screen... Jan 08 15:52:49 03crofton 07org.oe.dev * r43f3239a... 10/ (1 site/powerpc-linux): site/powerpc-linux : Update sizeof for various data types. Jan 08 15:53:19 And there's a lot of patches in RPSRC... I'd very much like to know which of those are needed and why. Jan 08 15:54:02 koen: yeah drop the xft as in certain cases for me it causes corruption Jan 08 15:54:16 anyone has an idea when next version of x.org will be out? it would be nice to drop our patches and use mainline ;-) Jan 08 15:54:17 mickeyl: do you happen to know roughly how long it takes to travel from Frankfurt to Heidelberg? Jan 08 15:54:32 (at lest some of patches ;-)) Jan 08 15:54:50 pb_: ~ 1 h Jan 08 15:54:58 mickeyl: thanks Jan 08 15:55:28 pb_: it's roughly 100km, slightly less probably Jan 08 15:55:57 fast train should make it in 40 minutes or so Jan 08 15:55:58 ok, cool Jan 08 15:56:05 you're in germany soon? Jan 08 15:56:12 in April, I think Jan 08 15:56:13 psokolovsky: i suspect we'd need to fixup xcalibrate first Jan 08 15:56:37 pb: cool. drop me a note when you have time for a beer. we'll meet in Brussels though right? Jan 08 15:56:47 my girlfriend is having an exhibition of her paintings at a gallery in Heidelberg Jan 08 15:56:51 mickeyl: right, I hope so Jan 08 15:56:56 excellent. Jan 08 15:57:03 I haven't actually booked anything for fosdem yet but I still plan to be there Jan 08 15:57:18 koen, but xcalibrate thing went to git. and we use almost the same patch as in git, iirc Jan 08 15:57:35 psokolovsky: the git version didn't work for me :( Jan 08 15:58:13 koen, did you try recently or it was yet then, in autumn? Jan 08 15:58:28 autumn Jan 08 15:59:11 pb_: hi - that's good news :-) Jan 08 15:59:21 ok, worth a fresh try then... ;-) Jan 08 15:59:21 florian_kc: hello :-) Jan 08 15:59:31 re Jan 08 15:59:41 * florian_kc should book for fosdem too Jan 08 16:00:53 fosdem sound fun ... why all interesting stuff is in europe :( Jan 08 16:01:10 chouimat: why aren't you in europe? ;) Jan 08 16:01:37 chouimat: heh. well, you have ols in canada. Jan 08 16:01:38 florian_kc: because I'm in Canada ... the best place in the world ;) Jan 08 16:01:44 pb_: I know :) Jan 08 16:01:50 indeed :-) Jan 08 16:02:18 mickeyl: I have a small assigment for you :) Jan 08 16:02:24 chouimat: ok, right... Canada might be nice Jan 08 16:02:24 pb_: I hope to be able to go this year ... since I live in Ottawa :) Jan 08 16:02:28 koen: uh oh... Jan 08 16:02:48 Vanilla 2.6.19 has apparent support for SL-Cxx00... Does it really need patches to work? Jan 08 16:02:50 03koen 07org.oe.dev * rfe7afda0... 10/ (22 files in 3 dirs): Jan 08 16:02:50 gtk 2.10.7: add WIP version: Jan 08 16:02:50 * created a gtk-2.10.inc with shared metadata Jan 08 16:02:50 * run-iconcache and hardcoded-libtool still need to get rediffed Jan 08 16:02:50 * pangoxft patch has been dropped because cairo is about as fast as xft now Jan 08 16:03:17 mickeyl: rediff those two patches :) Jan 08 16:03:25 are they large? Jan 08 16:04:15 koen@bitbake:~/OE/monotone/org.openembedded.dev/packages/gtk+/gtk+-2.10.7$ cat hardcoded_libtool.patch run-iconcache.patch | wc -l Jan 08 16:04:15 50 Jan 08 16:04:24 i'll have a look Jan 08 16:04:54 in total you need to change ~8 lines :) Jan 08 16:05:17 where ~5 consist of commenting a section Jan 08 16:10:38 Anybody have an idea what it is with these .desktop files that prevent startup of opie for me? bug 1757 Jan 08 16:11:16 Laibsch: maybe charset issues Jan 08 16:11:35 did you remove and locale-data? Jan 08 16:11:44 s/and/any/ Jan 08 16:12:15 I do not remember doing so. Jan 08 16:12:40 And charset? I prepared these .desktop files via vi and still had the same problems. Jan 08 16:13:17 Just "rm zbedic.desktop;touch zbedic.desktop" kills opie. Jan 08 16:15:02 koen: Furthermore, would charset for an ASCII-only .desktop file? Jan 08 16:15:16 hmmm Jan 08 16:15:27 could be a bug in opie itself Jan 08 16:15:34 * koen <- opie n00b Jan 08 16:16:03 * Laibsch asked in #opie, but no reply so far. Jan 08 16:16:07 rm -rf *opie* Jan 08 16:16:13 * Laibsch wonders what is triggering this Jan 08 16:16:35 XorA: not an option. I use my Z not as a toy but for a dictionary. Jan 08 16:16:59 There are still things missing in GPE or any other WM/GUI toolkit for OE to make it usable for me. Jan 08 16:18:07 Laibsch: sounds like you volunteer to become opie developer, or GPE developer to implement missing features Jan 08 16:23:17 hi everybody..! can someone please help me - i've managed to build task-base, and now i'm trying to build opie-base, but i get the following error message FATAL: Configuring qt failed. EXTRA_OECONF was -system-jpeg -system-libpng -system-zlib -no-qvfb -no-xft -no-vnc -gif -xplatform linux-gnueabi-arm-g++ -qconfig qpe -depths 8,16,32 -static -thread"" Jan 08 16:23:41 any idea..?? Jan 08 16:24:07 XorA|gone: Time permitting and skills available, yes. That is what I have always been doing. Jan 08 16:27:02 i get "| No static library support for platform/compiler linux-gnueabi-arm-g++-static" - Jan 08 16:27:26 koen: what packages provide the locale data you were talking about? "ipkg list_installed local*" is empty. But so is the feed browser. Jan 08 16:28:17 *local* Jan 08 16:28:20 does anyone know whether you need to make a specific configuration for the angstrom dist to work..? Jan 08 16:28:35 Okay, got a kernel that displays stuff... Jan 08 16:28:46 koen: any idea about my error..? Jan 08 16:28:47 Much easier to debug now. Jan 08 16:29:04 shirour: I don't use the Obsolete Palmtop Integrated Environment Jan 08 16:30:09 koen: so what do you recommened instead? Jan 08 16:30:28 koen: i'm looking for something that will work well & fast Jan 08 16:31:54 koen: "ipkg list_installed *local*" is empty, too. Jan 08 16:33:52 koen: and another problem i'm having - with getting the patches to work -it seems they are targeted for a specific kernel version... how can i apply them on 2.6.15 ? Jan 08 16:37:45 hrw|gone: hi... ? Jan 08 16:39:23 mickeyl: http://mail.gnome.org/archives/gtk-devel-list/2007-January/msg00078.html Jan 08 16:40:24 neat! let them do the work ;) Jan 08 16:41:17 where is the 770 booting the kernel from? Jan 08 16:42:32 mickeyl: seperate partition iirc Jan 08 16:42:48 When are people arriving/leaving at FOSDEM btw? Jan 08 16:42:55 RP: Friday / Monday Jan 08 16:43:05 Friday is a must :) Jan 08 16:43:06 friday at ~18.00 and monday at ~noon Jan 08 16:43:15 (Roy D'espagne) Jan 08 16:43:24 monday would be cool to have a relaxed breakfast together Jan 08 16:43:31 leaving sunday i found very stressful Jan 08 16:43:38 and this year we even have the booth Jan 08 16:43:48 and have to clean up the booth Jan 08 16:43:51 ya Jan 08 16:44:34 I'll see how the flights work out... Jan 08 16:44:54 NOTE: package gtk+-2.10.7-r0: task do_compile: started Jan 08 16:45:37 anyone experiences with yaffs2 btw? Jan 08 16:45:59 morning Jan 08 16:46:07 hey kergoth Jan 08 16:46:11 how are thingsß Jan 08 16:46:12 ? Jan 08 16:46:20 mickeyl: I'd rather take my chances with logs instead of yaffs(2) Jan 08 16:46:31 mickeyl: NOTE: package gtk+-2.10.7-r0: task do_install: started Jan 08 16:46:35 soso, got my old job back on land warrior after leaving ohand, and now land warrior was killed, so unemployment is imminent again Jan 08 16:46:38 otherwise pretty good Jan 08 16:46:40 how are things with you? Jan 08 16:46:58 ughs. you have a touch with jobs haven't you ::/ Jan 08 16:47:18 pretty well, booked until march atm. Jan 08 16:48:04 going to work on a project for Honda Research Europe that has to do with the Asimo roboter :D Jan 08 16:48:11 exciting stuff Jan 08 16:48:17 although I'm not working directly with the robot Jan 08 16:48:28 using OE? Jan 08 16:48:37 nice, sounds fun Jan 08 16:49:02 nothing OE. writing a Qt/X11 app for controlling objects in their VR simulation Jan 08 16:51:21 can someone please help me building opie ? Jan 08 16:51:34 or telling me what's the best choice for fast & well working gui ? Jan 08 16:52:01 XorA|gone: http://www.openembedded.org/~koen/gtk/ <- feed for gtk 2.10.7 Jan 08 16:52:46 shirour: try opie-devel@handhelds.org, this is where the remaining opie developers hang out Jan 08 16:52:54 or #opie Jan 08 16:54:15 mickeyl: ok, thanks..! what other options do i have? which one is the best? Jan 08 16:54:41 hard to say -- it depends on your needs. GPE seems to work solid Jan 08 16:55:38 but as far as i could tell gpe is a bit slower, isn't it? Jan 08 16:59:13 possible. shouldn't be noticable though with enough memory Jan 08 16:59:34 if you need all the speed you can get, you probably should use EFL and cook apps using Edje Jan 08 16:59:55 or don't use a GUI toolkit at all ;) Jan 08 17:00:43 mm... i have no idea what's efl and edje are... Jan 08 17:00:44 warning: personal statement following Jan 08 17:00:51 the hard truth is there is no convincing handheld environment out there atm. Jan 08 17:01:12 shirour: http://www.enlightenment.org Jan 08 17:02:15 ok, thanks..! still i don't get where is this strange error come from Jan 08 17:04:24 I'll be arriving in Brussels Thursday b4 FOSDEM, leaving Monday Jan 08 17:10:45 cool Jan 08 17:11:31 hmm CIA-4 ? Jan 08 17:12:57 Yay, got to /bin/sh Jan 08 17:13:21 he yho Jan 08 17:13:24 03koen 07org.oe.dev * r6483d4d1... 10/ (3 files in 3 dirs): gtk 2.10.7: rediffed run-iconcache.patch Jan 08 17:13:31 03koen 07org.oe.dev * rd4b1029a... 10/ (3 files in 3 dirs): gtk 2.10.7: rediffed hardcoded-libtool.patch Jan 08 17:13:34 mickeyl: tnx for mail, lets talk @fosdem... Jan 08 17:14:42 Hm, kbd layout is a bit weird... Jan 08 17:14:57 DataBeaver: which device? Jan 08 17:15:03 SL-C3100 Jan 08 17:15:28 DataBeaver: http://www.angstrom-distribution.org/unstable/images/20070104/spitz/ Jan 08 17:15:42 Running self-compiled kernel in Debian, so I don't really expect help from here, though tips are certainly appreciated... Jan 08 17:16:06 Might try Ã…ngström at some point too Jan 08 17:16:48 g'day kergoth Jan 08 17:17:15 I just like to know exactly how my computers work Jan 08 17:17:51 hey pb_ Jan 08 17:18:33 DataBeaver: angstrom is debian based, and on the same arm hardware a lot faster Jan 08 17:20:23 koen: how come? Jan 08 17:20:44 DataBeaver: but I'm glad you didn't go the 'debian in chroot with X over vnc' route :) Jan 08 17:20:58 rw: better floating point emulation Jan 08 17:21:26 That would kinda make me dependant on having a network connection Jan 08 17:21:34 could this stuff be "ported" to oz? Jan 08 17:21:40 I'm planning to enable the faster floating point support Jan 08 17:22:09 DataBeaver: you mean recompile everything? Jan 08 17:22:27 Umm, no, there's an option in the kernel Jan 08 17:22:42 DataBeaver: that won't gain you much Jan 08 17:22:53 Oh, gcc does its own stuff? Jan 08 17:23:01 kergoth: doh, unemployed again? that's unfortunate. Jan 08 17:24:50 kergoth: :( I was kind of wondering what was going on since the land warrior thing was in the news a few weeks ago Jan 08 17:24:52 Is Ã…ngströn compatible enough that I can use Debian's ARM packages on it? Jan 08 17:25:04 DataBeaver: it isn't Jan 08 17:25:32 DataBeaver: angstrom uses EABI: http://wiki.debian.org/ArmEabiPort Jan 08 17:26:03 Hm... Jan 08 17:26:47 The person doing the official eabi port isn't the sharpest tool in the shed, so it will take debian another year Jan 08 17:27:03 the person doing the unofficial port is nearly done :) Jan 08 17:27:46 So far my experiences with OpenZaurus are pretty bad with regards to software being available and working... Jan 08 17:28:16 angstrom isn't openzaurus Jan 08 17:28:22 is there a way to add a user to another group with busybox, or do I have to install coreutils for that? Jan 08 17:28:51 vi /etc/group Jan 08 17:28:54 ta da Jan 08 17:28:55 Does it have stuff like firefox and xmms2? Jan 08 17:29:09 Although Debian probably doesn't have xmms2 for ARM either... Jan 08 17:29:16 kergoth: :) Jan 08 17:29:17 DataBeaver: http://www.angstrom-distribution.org/repo/ Jan 08 17:30:02 No nethack :/ Jan 08 17:31:57 No gnome-terminal either, but xterm at least... Jan 08 17:31:58 gooood morning Jan 08 17:32:40 moin likewise Jan 08 17:32:43 Got rather fed up with certain stuff about OpenZaurus' rxvt, like not supporting bold attribute... Jan 08 17:32:55 hey likewise Jan 08 17:33:02 Jin^eLD: hey! Jan 08 17:33:21 Jin^eLD: I can spend some minutes looking at your MediaVault (name correct?) Jan 08 17:33:28 tomb? Jan 08 17:33:31 mediatomb Jan 08 17:33:38 likewise: MediaTomb :) Jan 08 17:33:46 likewise: there's an updated .bb in mediatomb svn Jan 08 17:33:47 koen: does the updated .bb file work btw? Jan 08 17:33:59 likewise: feel free to commit that :) Jan 08 17:34:05 Jin^eLD: haven't had time yet, sorry Jan 08 17:34:14 I'll try it now Jan 08 17:34:16 koen: what can I commit? Jan 08 17:34:25 ah, the bb from svn Jan 08 17:34:41 DataBeaver: rxvt-unicode does support bold... or did you mean the opie terminal? Jan 08 17:34:47 likewise, I can boot my EFIKA from net and from hd Jan 08 17:34:54 Crofton: tops! Jan 08 17:35:17 yep Jan 08 17:35:18 Crofton: now you can help debug the few packages that do not build! (You knew there was a plan behind all this :-) ) Jan 08 17:35:39 after netboot I formatted hd then untered tar ball on it Jan 08 17:35:41 heh Jan 08 17:35:45 omniorb builds now Jan 08 17:35:59 Crofton: seems especially libstc++ packages are hard to compile. I would really like the musicpd/mpd (headless MP3 player) etc. Jan 08 17:36:01 I committed some updates to site file to set some missing type sizes Jan 08 17:36:08 kergoth: Nope, the one installed by OZ 3.5.4.1. I think it was rxvt. \033[1m resulted in inverted text Jan 08 17:36:20 likewise, omniorb uses libstdc++ Jan 08 17:36:21 Crofton: ah, that's the site/powerpc stuff I just saw being updated then. Jan 08 17:36:29 yep Jan 08 17:36:51 Crofton: did you get failing omniorb builds before? Jan 08 17:37:15 DataBeaver: as far as i know, oz has never used ordinary 'rxvt' Jan 08 17:39:11 Crofton: I am thinking, could you bring some picopsu's to FOSDEM if I have them delivered at your place? Jan 08 17:39:20 Hm, is fbcon=rotate:1,font:VGA8x8 correct syntax? Jan 08 17:40:09 hmm Jan 08 17:40:20 need to check importing rules Jan 08 17:41:18 $50 each? Jan 08 17:41:21 how many? Jan 08 17:41:29 Crofton__: no, never mind then, don't want to burden you with that. Jan 08 17:41:41 I suspect I could call them gifts Jan 08 17:42:10 and they are smal Jan 08 17:42:21 Crofton__: or samples. AFAIK, merchandise samples are not subject to tax (but I will check beforehand). Jan 08 17:42:29 I'll lok Jan 08 17:42:32 look also Jan 08 17:42:34 that is Jan 08 17:44:17 koen: cool, though I was having problems sharing my feed with Angstrom feeds :-( Jan 08 17:46:19 looks like I can easily bring in 175euro of "gifts" Jan 08 17:47:30 possibly I can purchase them and you can pay me in euros on arrival? Jan 08 17:48:34 Crofton: Well, I could order/pay and have them deliver at your place. But I'll check with some colleagues first, they're visiting a partner company in the US in two weeks or so. Jan 08 17:48:44 ok Jan 08 17:49:16 There may be some beneift having me buy them in dollars and you pay me in euros, so I can avoid using an atm :) Jan 08 17:49:32 Crofton: ah sure, that would be an option of course Jan 08 17:51:19 Crofton: my broken efika packages are not type-size related, it seems. 1754, 1696, 1758, 1775 and musicpd/mpd. I am now checking if they build for ARM, or are broken only for powerpc. Jan 08 18:11:01 Jin^eLD: hmm, we need to move to sqlite3 3.3.9 first: http://www.sqlite.org/cvstrac/wiki?p=DatabaseCorruption Jan 08 18:13:10 philippe: and what about image with 2.6 for h3800? Jan 08 18:16:42 likewise, I guess 1754 is why reboot fails :) Jan 08 18:16:56 uh Jan 08 18:17:13 likewise: we support mysql as well, so you could try that if sqlite is broken Jan 08 18:17:40 mysql is swahili for 'data corruption' Jan 08 18:17:43 Jin^eLD: no, committing new version. Jan 08 18:17:48 koen: :-) Jan 08 18:17:51 ah, k :> Jan 08 18:19:31 koen: resetting the server by any chance? Jan 08 18:19:42 likewise: nope Jan 08 18:19:49 koen: what is the difference between (soft)vfp and (soft)fpa? Jan 08 18:20:05 cbrake: floatingpoint word order Jan 08 18:20:16 koen: hmm, suspiciously long time-out during my pull. Jan 08 18:20:28 cbrake: fpa uses bigendian, vfp the same endiannes as cpu Jan 08 18:20:43 cbrake: and fpa clashes with iwmmxt instructions Jan 08 18:20:58 likewise: try pusing to ewi546.ewi.utwente.nl Jan 08 18:21:16 pushing* Jan 08 18:23:48 Good evening Jan 08 18:23:55 hey sirfred Jan 08 18:23:56 evening sirfred Jan 08 18:29:54 bbl Jan 08 18:36:53 03likewise 07org.oe.dev * r4502961e... 10/ (3 files in 2 dirs): sqlite: Added 3.3.9. This release fixes database corruption in rare cases. Jan 08 18:38:17 ? strange. that were 4 files in 2 dirs... Jan 08 18:38:34 likewise: missed one file? Jan 08 18:38:47 mnt add is recursive, right? Jan 08 18:39:22 iirc not Jan 08 18:39:30 mtn commit is however Jan 08 18:39:43 hi - can someone help me - i'm trying to find out how far i'm in the building process... how can i tell? Jan 08 18:40:21 koen: ah, yes. Jan 08 18:40:42 hi univac : still trying to get it too boot completely... getting near. Jan 08 18:49:17 can someone please help me.. how do i know the order of compilation/how far along the build process am i..? Jan 08 18:51:18 shirour: there is no hard and fast order Jan 08 18:53:26 XorA|gone: mmm... so no way of telling ..?? Jan 08 18:55:03 shirour: if glibc built Id estimate you are approx 1/3 through time wise Jan 08 18:55:59 XorA|gone: well, the last thing i saw was "NOTE: package perl-native-5.8.7-r3: task do_build: completed" Jan 08 18:58:24 03likewise 07org.oe.dev * r913931cf... 10/ (10 files in 6 dirs): sqlite: Fix previous incomplete commit of 3.3.9. Jan 08 19:08:58 likewise: psssst, that commit has too much Jan 08 19:09:23 koen: I don't trust the numbers anymore :-) Jan 08 19:09:56 koen: they include the manifest files etc, I assume. Jan 08 19:09:56 likewise: http://lists.linuxtogo.org/pipermail/openembedded-commits/2007-January/002908.html Jan 08 19:10:15 Jan 08 19:10:50 mtn disapprove 913931cff89c154822360bc667ef1aaa2b60ae28 ; mtn commit packages/sqlite Jan 08 19:10:51 whoops. Jan 08 19:10:56 or something similar Jan 08 19:15:25 koen: I should really remember that mtn doesn't care what the pwd is, where cvs does care. Jan 08 19:15:40 in some instances Jan 08 19:15:56 every now and then there is a discussion on monotone-devel about it Jan 08 19:17:42 likewise: I usually wake up when typing in the commit message and seeing a lot of files Jan 08 19:19:59 koen: mtn should issue a kick in the stones for every file, that way you'd learn to be conservatrive :-) Jan 08 19:20:38 * koen retrieves the protector from storage Jan 08 19:21:41 XorA|gone: having to buy a round at fosdem would be a better incentive :) Jan 08 19:22:46 likewise org.oe.dev * r913931cf... / (10 beers for 6 people): sqlite: Further unshiting my noobness. Jan 08 19:23:16 That's two soda's for myself... Jan 08 19:24:31 -ENOCHILISAUS Jan 08 19:26:38 [FOSDEM hotel reservation for sale] [highest bidder] Jan 08 19:28:46 * koen discovers the chilisaus bottle Jan 08 19:28:53 03likewise 07org.oe.dev * r753b1a3c... 10/ (10 files in 6 dirs): disapproval of revision '913931cff89c154822360bc667ef1aaa2b60ae28' Jan 08 19:29:00 03likewise 07org.oe.dev * r74dfff88... 10/ (4 files in 2 dirs): sqlite: Fix previous incomplete commit. Try 2. Jan 08 19:29:56 speaking of fosdem.... Jan 08 19:30:03 beer time! Jan 08 19:30:15 mtn commit does not tell me what I am committing locally. That's where I would wake up (normally, not today) Jan 08 19:30:27 * XorA|gone will expect to be introduced to all the decent .eu beers Jan 08 19:30:44 beer time followed by sleep time for me. Still fully out-of-sync with our sun. Jan 08 19:31:21 XorA|gone: leffe (brun|radieuse), duvel, de koninck, guinness Jan 08 19:31:41 koen: Leffe is already huge in the UK Jan 08 19:31:47 radieuse?? Jan 08 19:31:49 koen: as is guiness :-) Jan 08 19:32:14 * likewise expects to get reacquinted with some Belgium beers (and his head with the wooden table) Jan 08 19:32:34 likewise: sounds like the WWE Jan 08 19:33:39 chouimat: Im guessing Red, as UK has Brown, Blonde and Red Leffe Jan 08 19:34:14 XorA|gone: WWE? Jan 08 19:34:20 if you like blond beers, try leffe 9 degrees Jan 08 19:34:30 XorA|gone: I only have Brown and Blonde here ... Jan 08 19:34:31 koen: Im into Real Ale Jan 08 19:34:32 that was my quest last year, to buy a 6pack of 9 degrees Jan 08 19:34:42 XorA|gone: bring some :) Jan 08 19:34:46 chouimat: where in uk is that Jan 08 19:34:49 hi Jan 08 19:34:54 hrw: wb Jan 08 19:35:04 likewise: World Wrestling Entertainment Jan 08 19:35:29 XorA|gone: I'm in Canada :) Jan 08 19:35:39 XorA|gone: good news on Ximageon Jan 08 19:35:40 chouimat: ah, maybe the red doesnt swim as well Jan 08 19:35:46 koen: yeah, it works well Jan 08 19:36:00 ~hail sirfred Jan 08 19:36:11 * ibot bows down to sirfred and chants, "I'M NOT WORTHY!!" Jan 08 19:36:40 XorA|gone: Hello. I've found the cause of the offset between the overlay and the colorkey rect Jan 08 19:36:40 doesn't the dreambox use imageon as well? Jan 08 19:36:43 XorA|gone: you know this kind of beers doesn't go well with most north american where a great beer == coors light :) Jan 08 19:37:06 chouimat: When I was in USA in 2001, I found some good beers from micro breweries Jan 08 19:37:21 chouimat: but the general american stuff is pish water Jan 08 19:37:36 likewise: 'imageon' is general ati marketing speak iirc Jan 08 19:37:43 Humm, isn't this weird? Jan 08 19:37:48 root@c7x0:~$ apm Jan 08 19:37:48 Off-line, battery status high: 0% Jan 08 19:37:57 JustinP: thx for reminding me about zaurus machines - building them now Jan 08 19:38:01 XorA|gone: yup ... and the most drinken beer in Quebec City is now Coors Light .... *sigh* I think it's the same thing for Ottawa Jan 08 19:38:24 koen: hmm, I see, I think there is a xilleon on the DB. Jan 08 19:38:38 sirfred: BTW I can even suspend during playing, a quick press on F twice (fullscreen switch) and display comes back Jan 08 19:39:06 sirfred: I am mightilly impressed by your work Jan 08 19:39:15 XorA|gone: I didn't dare to test that Jan 08 19:39:28 sirfred: I tested by accident, after 10 mins my Z suspend itself Jan 08 19:39:32 XorA, sirfred: what do you think about giving Xw100 for OZ user in oz 3.5.5? Or will it require too many changes? Jan 08 19:39:49 hrw: wait a couple of days and give Ximageon Jan 08 19:39:58 XorA|gone: I meant Ximageon Jan 08 19:40:11 hrw: Well, to say the truth , it's fresh work. Jan 08 19:40:12 hrw: it looks stable so far and works bloody well with mplayer Jan 08 19:40:22 hrw: I'm surprised it's this stable. Jan 08 19:40:48 But well, all the work made here was tested in libw100 before Jan 08 19:41:23 sirfred: question though, would it be faster if you could xrandr -s 320x240 before playing a 320x240 video? Jan 08 19:41:33 XorA|gone: Probably not. Jan 08 19:41:46 XorA|gone: The great problem is bandwidth. We are sending the same data to the card. Jan 08 19:41:54 sirfred: ok Jan 08 19:41:56 XorA|gone: cant mplayer do videomode change automatically? Jan 08 19:42:12 hrw: it can, but Ximageon doesnt support that yet Jan 08 19:42:17 ok Jan 08 19:42:29 XorA|gone: How does mplayer to that using XV ? Jan 08 19:42:32 XorA|gone: xrandr ? Jan 08 19:42:34 "OE bug 1841: matchbox panel loses applets after playing a video" Jan 08 19:42:43 sirfred: VidMode extension I think Jan 08 19:42:52 sirfred: but that might be x11 only Jan 08 19:43:06 XorA|gone: Probably. I don't think xv is supporting that. Jan 08 19:43:18 At least, I didn't see any hook to implement that. Jan 08 19:43:20 sirfred: Its been years since I didnt have a card with hardware scaler Jan 08 19:43:46 XorA|gone: Most of modern cards are using the 3d scaler to do Xvideo Jan 08 19:44:04 sirfred: but it would be sweet if you could get xrandr to work to change size for stuff like UAE Jan 08 19:44:09 (amiga emul) Jan 08 19:44:17 XorA|gone: Yes, I'm working into that. Jan 08 19:44:25 sirfred: you at FOSDEM? Jan 08 19:44:46 XorA|gone: What's that? Jan 08 19:44:52 Some kind of meeting? Jan 08 19:45:03 ~fosdem Jan 08 19:45:03 extra, extra, read all about it, fosdem is the Free and Open source Software Developers' European Meeting, a 2 day event, organized by volunteers, to promote the widespread use of Free and Open Source software. http://www.fosdem.org, or done for 2005 and no dates have been published yet for 2006 Jan 08 19:45:36 sirfred: I think you are owed beers Jan 08 19:46:08 XorA|gone: Sorry, but my english level is far from desirable, what is owed ? Jan 08 19:46:20 XorA|gone: I know what beers are, of course, but... Jan 08 19:46:21 ibot: fosdem is also in Brussels, Belgium Jan 08 19:46:22 koen: okay Jan 08 19:46:26 sirfred: your work is good so I feel I should buy you beer Jan 08 19:46:51 XorA|gone: A pair of Leffe for me, please. Or some Grinbergen Jan 08 19:47:02 :-D Jan 08 19:47:12 XorA|gone: Erdinger Pikantus won't be rejected, neither. Jan 08 19:47:13 ;-) Jan 08 19:47:17 anyway I must go eat dinner :-D Jan 08 19:47:23 ~bon apetit Jan 08 19:47:36 ~bon appetit Jan 08 19:47:37 well, bon appetit is smacznego. Guten Appetit. Eet Smakelijk. God Appetitt. Buon Appetito. Buen apetito Bom Apetite. buen apetito Smaklig måltid!. Hyvää ruokahalua Jan 08 19:47:56 #openmoko users added some translations Jan 08 19:48:19 ibot: bon appetit is also Bo Proveito Jan 08 19:48:20 okay, sirfred Jan 08 19:50:51 sirfred: Bo Proveito is portugese? Jan 08 19:50:59 koen: Galician Jan 08 19:51:12 dialect of spanish? Jan 08 19:51:15 koen: A language spoken in the North West of Spain. Jan 08 19:51:21 ah, ok Jan 08 19:51:36 koen: No, another latin derivative, it shares common roots with Portuguese Jan 08 19:51:57 It's older than Spanish, actually Jan 08 19:52:16 * koen stores that in his mental random fact database Jan 08 19:52:23 koen: http://en.wikipedia.org/wiki/Galician_language Jan 08 19:52:52 "normative wars" Jan 08 19:53:41 koen: Yes, Galician was forbiden for centuries, so, written norms get lost. Jan 08 19:54:02 XorA|dinner: what was the best time to visit scotland? june, july, august? Jan 08 19:54:05 koen: In the process of re-normalization, there were different opinions. Jan 08 19:54:30 sirfred: remeind me of the spanish rulers trying to ban the 'x' from being used in mexico Jan 08 19:54:37 s/reimind/reminds/ Jan 08 19:54:54 koen: The funny thing is that it's pronounced as 'j' Jan 08 19:55:04 koen: It must be pronounced as 'j' in spanish Jan 08 19:55:23 yep Jan 08 19:55:41 although in .mx it's 'tsj' or 'xs' depending on the region Jan 08 19:56:01 but most people there pronounce it as mejico Jan 08 19:56:56 koen: x itself is pronounced something like 'ks' in Spanish, in Galician, the sound is totally different. Jan 08 19:58:08 gerwinin: good evening Jan 08 19:58:38 gerwinin: did you send me mail? Jan 08 19:59:22 Crofton: http://cgi.ebay.com/picoPSU-120-120w-12v-DC-to-DC-Power-Supply-VIA-Mini-ITX_W0QQitemZ170068115099QQihZ007QQcategoryZ80173QQcmdZViewItem Jan 08 20:00:18 looks like areasoanble price Jan 08 20:00:45 hmm... n800.. thats tempting Jan 08 20:00:57 s/n/nokia n/ Jan 08 20:01:03 * kergoth hmms Jan 08 20:01:04 kergoth: indeed Jan 08 20:01:13 kergoth: get noticed by nokia and get a dev discount :) Jan 08 20:01:19 I figure if I could resist the 770, then I was safe from temptation Jan 08 20:02:10 koen: hehe, thatd be nice. its quite reasonably priced as is though, i'm pleasantly surprised Jan 08 20:02:31 Crofton: 50% of what we pay here... Jan 08 20:02:42 heh Jan 08 20:03:48 kergoth: $400 in usa, €400 in .eu Jan 08 20:03:54 ~change 400 usd to eur Jan 08 20:03:59 400.00 United States Dollar (USD) makes 307.222 Euro (EUR) (from http://www.xe.com/) Jan 08 20:04:01 ~change 400 eur to usd Jan 08 20:04:07 huh Jan 08 20:04:15 EXCHANGE: 500 Can't connect to www.xe.net:80 (connect: timeout) Jan 08 20:04:24 ~change 400 eur to usd Jan 08 20:04:29 400.00 Euro (EUR) makes 520.767 United States Dollar (USD) (from http://www.xe.com/) Jan 08 20:04:51 likewise, how many are you interested in? Jan 08 20:04:54 sounds like they're ripping off the europeans Jan 08 20:05:13 Crofton: three pieces Jan 08 20:05:13 you american guys always had cheaper electronics Jan 08 20:05:28 we should start a shipping channel :-) Jan 08 20:05:43 we could always export beer. Jan 08 20:05:57 there are a few benefits to living here :) Jan 08 20:06:02 cheaper electronics, cheaper cars, cheaper tax - whatever Jan 08 20:06:16 everything is so new though Jan 08 20:06:20 cheaper privacy Jan 08 20:06:48 and you don't have to learn Dutch Jan 08 20:07:41 here you don't even have to learn English ... Jan 08 20:07:41 likewise: move to friesland ;) Jan 08 20:08:14 btw, sealand is for sale Jan 08 20:08:14 http://www.sealandgov.org/ Jan 08 20:08:14 koen: Leeuwarden was nice. (I would have to learn Danish in order to understand the Frisian ppl). Jan 08 20:09:45 i would prefer scotland - it's evil as well, but they have scotch Jan 08 20:10:36 i would prefer - it's rough, but they have nice girls Jan 08 20:10:40 i would prefer Sweden - it's rough, but they have nice girls Jan 08 20:10:50 o.. .dev tslib is no longer machine specific? Jan 08 20:10:52 and bikini teams Jan 08 20:11:00 Packaged contents of tslib-conf into /a/home/hrw/devel/build/angstrom/tmp/deploy/glibc/c7x0//ipk/tslib-conf_1.0-r7_armv5te.ipk Jan 08 20:11:00 without bikinis? Jan 08 20:11:09 hrw: only tslib-conf was for some machines Jan 08 20:11:24 hrw: but with /dev/input/touchscreen0 ... Jan 08 20:11:57 NOTE: Removing failed build target virtual/arm-angstrom-linux-gnueabi-gcc-2.95 Jan 08 20:12:08 this is funny each time Jan 08 20:13:06 likewise: in finland too...oO(and good metal) Jan 08 20:13:31 ~seen [g2] Jan 08 20:13:44 rw: yes. only been in finland for a few days, but the nice girl over day ratio was good Jan 08 20:13:48 rw: Finland has Elakelaiset Jan 08 20:13:55 [g2] was last seen on IRC in channel #openmoko, 6h 19m 14s ago, saying: 'there's an option to specify the number of points in the track to save'. Jan 08 20:14:36 likewise: ;) Jan 08 20:15:20 hrw: you mean the band? Jan 08 20:15:54 rw: yes Jan 08 20:16:31 i don't know them ;) Jan 08 20:16:57 but i know finntroll ;) Jan 08 20:28:26 hi... i've built my system but the kernel freezes after "[42949376.520000] Freeing init memory: 96K"... can anyone help me..? how can i debug it? Jan 08 20:29:31 cu guys Jan 08 20:29:44 (*i've managed to build bootstrap and opie images... i'm proud of oe, that makes it possible for a lame user like me could have done that) Jan 08 20:30:33 good night Jan 08 20:30:38 so does anyone here have a clue how can i debug what's going on after the freeing init memory message? i think that that's the stange where the boot is moved to the hd scripts... any idea..? Jan 08 20:31:20 hrw: do you know what happens after the kernel "Freeing init memory:" message? how can i debug the boot process? Jan 08 20:31:40 lowlevel dbug in the kernel? Jan 08 20:31:52 shirour: you run eabi kernel with eabi rootfs? Jan 08 20:32:05 shirour: or maybe oabi kernel with eabi rootfs? Jan 08 20:32:14 hrw: both eabi i think Jan 08 20:32:23 shirour: then check it again Jan 08 20:32:47 hrw: i've patched the kernel and put "CONFIG_AEABI = y" in Makefile Jan 08 20:32:54 and my image is familiar.1 Jan 08 20:33:07 i mean my dist Jan 08 20:33:24 shirour: did you enabled OABI compatibility in kernel? Jan 08 20:33:36 hrw: no... Jan 08 20:33:45 hrw: how can i check the the AEABI is turned on? Jan 08 20:35:06 truly speaking? dont know Jan 08 20:35:40 hrw: hmmm... and how can i debug the boot process? Jan 08 20:36:15 use lowlevel debug like Crofton suggested? Jan 08 20:36:36 well, even before - inittab is the first thing to run right..? Jan 08 20:36:53 init not inittab Jan 08 20:37:42 what's init? Jan 08 20:37:48 i know that inittab defines the first scripts to run Jan 08 20:37:49 /sbin/init Jan 08 20:37:58 /sbin/init use inittab Jan 08 20:37:59 init reads inittab Jan 08 20:38:05 oh i see Jan 08 20:38:14 if init is sysvinit ofcourse Jan 08 20:38:15 and how can i tell whether it ran or not? Jan 08 20:38:47 I think you are hanging (or losing the console) before it runs Jan 08 20:38:57 why? Jan 08 20:39:02 look at something boot linux and see where the freeing message occurs Jan 08 20:39:07 What happens after "Freeing init memory" ? Jan 08 20:39:23 what was your DISTRO again? Jan 08 20:39:25 look at something else that boots successfully Jan 08 20:39:37 [42949376.500000] VFS: Mounted root (ext2 filesystem). Jan 08 20:39:37 <6>Freeing init memory: 96K Jan 08 20:39:37 [42949376.520000] Freeing init memory: 96K Jan 08 20:39:47 o fsck.. 41G of source already... Jan 08 20:39:49 these are my last 3 lines Jan 08 20:41:02 but i don't know what freeing the memory means... Jan 08 20:41:14 google the message Jan 08 20:42:03 My monotone is segfaulting all the time Jan 08 20:42:05 :( Jan 08 20:44:01 mmm seems to be libstdc++ issue :( Jan 08 20:44:11 hrw, Crofton: looks like it the line right before /sbin/init indeed Jan 08 20:44:35 hrw, Crofton: any idea how can i tell whether my eabi patch was successful? Jan 08 20:44:42 shirour: what's your DISTRO again? Jan 08 20:44:48 shirour: I would try oabi kernel/rootfs first Jan 08 20:44:55 gerwinin, grab a statically linked mtn binary? Jan 08 20:46:08 koen: the angstrom 2007.1 Jan 08 20:46:20 Crofton: I am trying to do that now I am running on ia64 system Jan 08 20:46:21 that should work Jan 08 20:46:41 koen: i'm afraid the patches were not enough.... is there anyplace where i can patch the whole thing at once? Jan 08 20:46:51 apply the 2.6.16 patch Jan 08 20:47:35 can you please tell me how to find it..? Jan 08 20:47:44 on kernel.org Jan 08 20:48:38 do i need anything special? Jan 08 20:49:11 ftp://ftp.kernel.org/pub/linux/kernel/v2.6/patch-2.6.16.bz2 Jan 08 20:49:49 shirour: you are engineer or marketing guy? Jan 08 20:50:22 hrw: electrical eng. programming is just a hobby for me Jan 08 20:50:27 koen: thanks..! Jan 08 20:50:49 anyone know how to fix "Terminal entry no found in terminfo" messages? Jan 08 20:50:50 EE rocks Jan 08 20:51:12 koen: i agree. Jan 08 20:51:32 grub does not like my terminal one bit :-\ Jan 08 20:54:49 ahh, install ncurses-terminfo and everyone is happy :-) Jan 08 21:00:28 guys what do you think about switching from foobar_1.2-r0_arm.ipk to foobar_1.2-r0_oz.ipk? another asshole^Wuser asked why somesharpromapp_arm.ipk cant be installed under OZ Jan 08 21:03:26 hmm.. looks like time to stop working on celf machine for today.. US citizens are at work so bandwidth is taken.. Jan 08 21:04:06 koen,shirour : fehh, Mech, eng is soo much more fun :) Especially when dealing with servo driven mechanisms.. failures can be spectacular Jan 08 21:04:26 hvontres|poodle: failures? Jan 08 21:04:43 sounds like those people failed the control engineering course Jan 08 21:04:51 (like I did) Jan 08 21:05:05 Once I knew control stuff Jan 08 21:05:36 koen: Controls class != actual system...:) Having a servo axis run away can be lots of fun :) Jan 08 21:06:09 koen: Plus in class they NEVER teach you about mechanical resonance problems either :) Jan 08 21:06:50 :) Jan 08 21:08:21 It's also fun to watch CS people 'insulate' their project with silicone gel Jan 08 21:08:21 *conductive* silicone gel Jan 08 21:08:21 hehe... now i feel like i'm at work Jan 08 21:11:22 who want to guess how much time does machine with one sata disk need to copy 25K files (41G total) into same partition? Jan 08 21:13:30 hrw: 3 hrs :) Jan 08 21:13:55 hrw: are you starting a pool? Jan 08 21:15:18 hvontres|poodle: no. creating backup copy before start some dangerous scripts Jan 08 21:18:09 hrw: hehe... reminds me of a script I wrote once that had rm ($) in it... and I passed it * as the argument :) Jan 08 21:18:44 [hrw@misc1 cron]$ for f in sources/*;do f=`basename $f`; char1=`echo $f|cut -c1`;char2=`echo $f|cut -c-2`; mkdir -p mirror/$char1/$char2;done Jan 08 21:18:55 hvontres|poodle: add mv and ln into it... Jan 08 21:32:48 koen: August is Edin festival so its very expensive Jan 08 21:33:02 koen: June July weather is good normally Jan 08 21:33:19 koen: not coming to debconf? Jan 08 21:40:43 re Jan 08 21:40:49 wb florian_ Jan 08 21:48:40 XorA|sleep: probly not to debconf Jan 08 21:48:51 XorA|sleep: generic holiday with the gf Jan 08 21:50:32 cu Jan 08 22:07:52 NOTE: Running task 678 of 864 (ID: 645, /home/koen/OE/monotone/org.openembedded.dev/packages/coreutils/coreutils-native_5.3.0.bb, do_configure) Jan 08 22:10:15 RP: nice new progress bar :) Jan 08 22:20:06 anyone knows an arm/ppc with 4-nics? Jan 08 22:22:27 zecke: uh, what do you need that for?! i have a ppc with pci and an arm machine too Jan 08 22:22:36 anyone hae a good reference to oss versus alsa? Jan 08 22:23:04 florian: oh right, I have a ppc with pci as well :} Jan 08 22:23:31 i would have add only one ;) Jan 08 22:24:28 the arm machine suffers from the fact that someone would need to finish the linux port to it first Jan 08 22:24:50 florian: these folks use a P4 as gateway, I would like to replace it with something else/more light weight (energy wise) Jan 08 22:25:33 zecke: that's what i use my ppc for :-) its an old motorola powerstack... Jan 08 22:25:50 and no, it does not run aix ;) Jan 08 22:27:01 Night Jan 08 22:30:02 zecke: loft + 2x minipci ethernet? Jan 08 22:30:41 koen: got a link to the loft site? Jan 08 22:31:06 http://www.giantshoulderinc.com/ Jan 08 22:31:23 or get one of those ppc NAS thingies Jan 08 22:31:33 those are getting multiple nics as well Jan 08 22:31:45 well, multiple ports + hardware vlan tagging Jan 08 22:32:06 network outage ahead... Jan 08 22:33:48 koen outage ahead Jan 08 22:33:49 ibot: test Jan 08 22:33:51 Test Failed! Jan 08 22:33:56 'night all Jan 08 22:34:07 setting up the net in the new office... Jan 08 22:34:13 I swear I saw an add for an ARM board with mulitple NIC's somewhere Jan 08 22:35:37 iirc the Manga machines (ARM) can have quite a lot nics Jan 08 22:41:25 later, I'm sent home Jan 08 22:50:52 florian: usb checked? Jan 08 22:51:32 greentux: not yet, sorry Jan 08 23:45:12 koen, my lightning talk was accepted Jan 08 23:47:39 Crofton|home: cool Jan 08 23:48:17 now I have to invent some slides :) Jan 08 23:48:37 I think I left i vague enough :) Jan 08 23:48:58 sounds like I will need to import some of these small power supplies also Jan 08 23:49:10 picopsu's Jan 08 23:49:27 Will you be at the "beer event"? Jan 08 23:51:45 Of course I'll be at the beer event :) Jan 08 23:52:14 and I'll also be at the way-too-early-breakfast the next day :) Jan 08 23:54:33 03koen 07org.oe.dev * r28548b0e... 10/ (1 conf/distro/angstrom-2007.1-oabi.conf): angstrom-oabi: correct FEED_ARCHs Jan 08 23:54:57 I hate it when I can't sleep and think of stupid bugs Jan 08 23:55:09 I hate it even more that I can't remember them in the morning Jan 09 00:02:56 koen: Hopefully the progress bar will give the users some idea of what bitbake actually does too :) Jan 09 00:05:51 RP: indeed Jan 09 00:07:28 koen: Poky just saw a bit of activity :) Jan 09 00:08:05 cool Jan 09 00:08:23 I just wished people at ohand would use clearer commit messages: http://www.openembedded.org/bonsai/view/scm/6/ Jan 09 00:09:26 RP: good news poky is now building angstrom-style stuff :) Jan 09 00:09:38 koen: Since the SCM knows which files you touched, arguably listing the package name is duplicate info... Jan 09 00:10:00 right Jan 09 00:10:11 but it makes 2cl a bit useless Jan 09 00:11:06 grrr Jan 09 00:11:11 maemo.org is being broken Jan 09 00:11:13 assuming a capable 2cl script, you should be able to extract the common path of the files in a given revision and prepend that to the log.. Jan 09 00:11:35 true Jan 09 00:12:00 we'll figure something out when we start releasing tarball with changelogs :) Jan 09 00:12:01 koen: Lets hope angstrom style pays off :) Jan 09 00:13:05 If the autobuilder survives that set of changes, I'll be impressed... Jan 09 00:13:47 heh: http://www.oesf.org/forums/index.php?showtopic=22707&pid=150486&st=0&#entry150486 Jan 09 00:14:02 MACHINE=foo bitbake bar-image Jan 09 00:14:49 i'd think itd be handy to convert the scm log to xml, then use xslt to make something sane out of it from there.. svn log --xml is our friend Jan 09 00:14:51 * kergoth wonders if monotone has something similar Jan 09 00:15:38 "sort of" Jan 09 00:15:49 the ciabot sends xml to cia Jan 09 00:15:51 kergoth: If not, the monotone devs would probably consider it... Jan 09 00:16:07 koen: re pdax builder, no comment :) Jan 09 00:16:11 its nice being able to take the log and easily generate html, rss, gnu changelog, out of it Jan 09 00:16:51 it always amuses me that opensource tools assume you are working on source Jan 09 00:17:01 and OE breaks all those assumptions Jan 09 00:19:28 kergoth: any thoughts on http://mail.gnome.org/archives/gtk-devel-list/2007-January/msg00081.html ? Jan 09 00:22:00 'night all Jan 09 00:23:37 koen: it really doesnt matter what the libtool in gtk is called, since its explicitly generated inside the source tree by configure and is run explicitly via LIBTOOL. it won't be obeying PATH, so whats available in the system is a non-issue, what is an issue, for us, is just ensuring that the correct m4 macros are used. having a prefixed libtool binary in /usr/bin makes perfect sense.. for anything that needs to use the system libtool directly.. Jan 09 00:24:03 in oe, as far as i know, only the jpeg build uses the staging libtool rather than generating its own from the .m4, and thats only because we cant rebuild its configure Jan 09 00:24:46 so it should use $(LIBTOOL) Jan 09 00:25:02 since OE generate foo-bar-linux-libtool in ${S} Jan 09 00:25:50 oe sets LIBTOOL to our prefixed filename in ${S}, and upstream will be setting it to the non-prefixed version.. the buildsystem is safe for both versions of libtool.m4 as long as it uses $(LIBTOOL) in the makefiles. Jan 09 00:27:21 so http://www.openembedded.org/repo/org.openembedded.dev/packages/gtk+/gtk+-2.10.7/hardcoded_libtool.patch isn't needed Jan 09 00:28:09 export for building from svn when ./libtool isn't provided Jan 09 00:28:50 that should probably be ${LIBTOOL} in each place where ./libtool is, so it obeys the var that AC_PROG_LIBTOOL sets Jan 09 00:29:11 ok Jan 09 00:29:24 will send the gtk folks a patch with $(LIBTOOL) Jan 09 00:29:25 assuming that LIBTOOL was defined before that point, of course Jan 09 00:29:46 ${LIBTOOL}, not $(LIBTOOL). those spots in configure.in are being run at configure time, not make time, afaict Jan 09 00:29:54 right Jan 09 00:30:26 now I'll got to sleep before I start even making less sense Jan 09 00:30:31 'night all Jan 09 00:30:45 hehe Jan 09 00:31:39 i'm just curious, i've been really fighting with libtool is my own builds lately, i was going through OE but i can't really find what/if anything is changed, any hints on where i can start poking? Jan 09 00:32:03 look in packages/libtool/.. there are a few patches there that are necessary for crosscompilation with it Jan 09 00:32:33 mainly it ensures that the 'libdir' read out of .la files (normally paths like /usr/lib) dont end up as -L's on the link line Jan 09 00:33:16 yep, that's the biggest problem Jan 09 00:34:08 ld should be looking in $sysroot/usr/lib anyway Jan 09 00:34:43 i guess its for specific circumstances Jan 09 00:34:55 say you've got an uninstalled libtool lib (in the source tree, src/.libs) Jan 09 00:35:02 and it depends upon libs in the installed path (/usr/lib) Jan 09 00:35:16 well, say you've got for example gnome installed to /opt/gnome Jan 09 00:35:19 linking against its .la would ensure that the search path for the dependent libs is in the linker search path Jan 09 00:35:29 iguess it's supposed to help with -L/opt/gnome/lib Jan 09 00:35:44 but the pkg-config output from the .pc files usualy takes care of that anyway Jan 09 00:35:57 right, but only for a lib thats uninstalled.. since if it was installed, you'd already have hte -L/opt/gnome/lib to have found its .la in the first place Jan 09 00:36:01 has to be to find dependent stuff Jan 09 00:36:10 yeah, it often comes in from other stuff anyway Jan 09 00:36:34 libtool is a nightmare in general Jan 09 00:36:57 have you seen jlibtool? seemed worth looking into.. a C reimplementation of libtool Jan 09 00:37:09 i've been fighting it since i started the sysroot branch in last june Jan 09 00:37:17 cut build times on apr in half or something Jan 09 00:37:19 finally got locale generation without needing a emulator anyway Jan 09 00:37:21 less feature rich than the gnu one of course Jan 09 00:37:26 heh, i dont envy you Jan 09 00:37:28 jlibtool? will look Jan 09 00:37:43 its in the apr source tree, they have a --enable-experimental-libtool configure opt Jan 09 00:38:10 unfortunately its .la format isnt the same as libtools, so they wouldnt be able to interact on a system, you'd have to use one or the other for the whole machine Jan 09 00:38:16 i was actualy going to re-write libtool this weekend then i saw you guys talking about it, figured i'd ask Jan 09 00:38:32 i dont care that jlibtool is C, but it being a reimplementation means it /might/ be more maintainable Jan 09 00:38:44 having to patch libtool in multiple places due to the duplicate code is bad Jan 09 00:38:49 heh Jan 09 00:38:56 that thing is a mess Jan 09 00:39:14 looked to me like they were afraid to use shell functions for portability reasons Jan 09 00:39:31 well, i buess they wanted compatibility with ash Jan 09 00:39:35 s/buess/guess/ Jan 09 00:40:04 either way, multiple scripts would have been a lot cleaner then one repetative one Jan 09 00:40:10 doesnt SuSv3 have shell functions? probably, they mustve wanted more compatibility than the simple unix specification even Jan 09 00:40:14 yeah, true Jan 09 00:40:42 and they had m4, they still couldve removed duplication in the pre-m4'd script via macros Jan 09 00:40:51 yet there's still duplication, its sad Jan 09 00:41:07 i didn't even think of that, but that's too true Jan 09 00:42:11 ah well, enough ranting Jan 09 00:42:13 * kergoth heads home Jan 09 00:42:19 later **** ENDING LOGGING AT Tue Jan 09 02:59:59 2007