**** BEGIN LOGGING AT Tue Jul 08 02:59:57 2008 Jul 08 04:42:36 hi Jul 08 04:44:24 i couldn't setting for sigma em8620l Jul 08 04:44:34 ? Jul 08 04:59:43 * wick100ct is away: I'm AFK Jul 08 06:05:06 * * OE Bug 4412 has been created by likewise(AT)gmx.net Jul 08 06:05:07 * * Race conditition in concurrent bitbake of .dev Jul 08 06:05:09 * * http://bugs.openembedded.net/show_bug.cgi?id=4412 Jul 08 06:12:47 gm Jul 08 06:35:50 morning Jul 08 08:07:55 bonjour \o/ Jul 08 08:08:47 03koen 07org.oe.dev * r01c449d5... 10/ (1 conf/machine/include/neuros.inc): neuros.inc: use raw oob mode for yaffs2 Jul 08 08:45:21 03koen 07org.oe.dev * r37d09a61... 10/ (12 files in 3 dirs): qt4: add first stab at qt-embedded 4.4.0, the *$%(*$%(#% people at trolltech renamed it *again*, so we probably need to copy all the qtopia core stuff and do s/qtopia-core/qt-embedded/g Jul 08 08:45:43 morning Genesis Jul 08 08:47:16 03koen 07org.oe.dev * r8870dbe8... 10/ (3 files in 3 dirs): qte 4.4.0: don't build tools, use the ones in staging Jul 08 08:49:44 likewize : ok. thanks. will have a look Jul 08 09:04:00 03koen 07org.oe.dev * r46f64322... 10/ (1 packages/uicmoc/uicmoc4-native_4.4.0.bb): uicmoc-native: add 4.4.0 Jul 08 09:10:00 good morning Jul 08 09:10:46 lovely messages :) Jul 08 09:22:03 03mickeyl 07org.oe.dev * r46607c54... 10/ (6 files in 4 dirs): add python-opendir, an extension module using POSIX opendir semantics Jul 08 09:22:07 03koen 07org.oe.dev * r26f1976a... 10/ (1 packages/linux/linux-omap2_git.bb): linux-omap2: bump to rc9 Jul 08 09:32:50 03koen 07org.oe.dev * rf48f9ab8... 10/ (1 packages/qt4/qt-embedded-4.4.0/0003-no-tools.patch): qte 4.4.0: disable some more tools Jul 08 09:36:16 03koen 07org.oe.dev * r6125661e... 10/ (3 files in 3 dirs): linux-neuros: bump SRCREV and update defconfig Jul 08 10:24:38 03koen 07org.oe.dev * rd8e61e9a... 10/ (3 files in 2 dirs): qt4: add webkit, help and clucene packaging Jul 08 10:37:09 03koen 07org.oe.dev * r3a280aa4... 10/ (1 packages/qt4/qt_packaging.inc): qt4: fix help and clucene packaging Jul 08 10:50:08 03koen 07org.oe.dev * r1766954e... 10/ (1 packages/gtk-doc/gtk-doc.bb): gtk-doc: generate packages for sdk and native toolchains Jul 08 10:50:13 03koen 07org.oe.dev * r3d446522... 10/ (3 files in 2 dirs): qt4 packaging: since not all packages exist recommend them instead of depending on them Jul 08 11:10:38 03koen 07org.oe.dev * r7a907665... 10/ (1 packages/qt4/qt_staging.inc): qt staging: fix up more .pc files Jul 08 11:14:12 03koen 07org.oe.dev * r560558ea... 10/ (1 packages/meta/meta-toolchain-neuros.bb): meta-toolchain-neuros: add qte and dbus Jul 08 12:00:14 Hello. I am having this problem on the final step of building a base image for angstrom: http://rafb.net/p/YCfJ3U31.html Jul 08 12:47:43 03mickeyl 07org.oe.dev * rfd2e0873... 10/ (1 packages/images/fso-image.bb): fso-image: ship mickeydbus Jul 08 13:22:05 * * OE Bug 4413 has been created by  Jul 08 13:22:07 * * gmp-native-4.2.2-autobuild Jul 08 13:22:09 * * http://bugs.openembedded.net/show_bug.cgi?id=4413 Jul 08 13:22:17 * * OE Bug 4415 has been created by  Jul 08 13:22:20 * * gcc-native-3.4.6-autobuild Jul 08 13:22:21 Well crud. Anyone have any ideas on how to fix that? Jul 08 13:22:22 * * http://bugs.openembedded.net/show_bug.cgi?id=4415 Jul 08 13:22:28 mickeydbus? Jul 08 13:22:30 * * OE Bug 4414 has been created by  Jul 08 13:22:32 * * binutils-cross-2.18.50.0.7-autobuild Jul 08 13:22:34 * * http://bugs.openembedded.net/show_bug.cgi?id=4414 Jul 08 13:22:38 dbus reimplementation in python? Jul 08 13:22:48 i hope not ^^ Jul 08 13:23:59 piroko: how to fix what? Jul 08 13:24:54 diego_at_work: The error I posted earlier Jul 08 13:25:05 With the double free corruption on creating a base-image for angstrom Jul 08 13:25:12 Kind of an important step. Haha Jul 08 13:25:49 A friend of mine reported the same problem, but it works on my pc... Jul 08 13:25:58 Hmm.... Jul 08 13:26:06 so i don't have any clue atm Jul 08 13:26:10 Maybe my version of glibc? Jul 08 13:33:59 dunno, they got the problem after adding some new package at the image Jul 08 13:35:41 I don't know more, becouse he can't do any clue about which program does a double free Jul 08 13:36:01 ... he can't give me any clue ... Jul 08 13:37:02 have you open a ticket about this problem? Jul 08 13:40:18 piroko, i don't think a glibc version problem, couse we use the same version Jul 08 13:41:35 piroko, looks like a bug in opkg-cl? Jul 08 13:41:55 03mickeyl 07org.oe.dev * rbf12ea84... 10/ (9 files in 2 dirs): python-{pyrex|cython}[-native] 0.9.x update to 0.9.8 Jul 08 13:42:00 03koen 07org.oe.dev * r2b87c80f... 10/ (3 files in 3 dirs): linux-omap2 git: enable uvc for beagleboard Jul 08 13:43:49 zecke: a convenient utility to use dbus Jul 08 13:43:51 modelled after qdbus Jul 08 13:43:55 i can't stand dbus-send Jul 08 13:44:06 so we have something with a much nicer syntax now Jul 08 13:44:12 + introspection Jul 08 13:45:23 mickey|shopping: hehe Jul 08 13:50:32 Crofton: Apparently. But how to fix it? No idea :( Jul 08 13:50:48 sorry, no idea here Jul 08 13:51:04 I have some builds going, so I'll see if I can reproduce Jul 08 13:51:13 what distro are you running OE on Jul 08 13:51:14 Crofton: Angstrom builds? Jul 08 13:51:18 yeah Jul 08 13:51:19 archlinux Jul 08 14:00:02 Crofton: You building angstrom 2008 or 2007? Jul 08 14:00:21 2008 Jul 08 14:01:04 Alright me too Jul 08 14:02:20 I'm wondering if there is an issue in the archlinunx glibc ... Jul 08 14:02:32 That's what I was thinking Jul 08 14:02:37 What are you running? Jul 08 14:02:41 Fedora Jul 08 14:02:47 8 and 9 Jul 08 14:02:56 Ah. I have glibc 2.8(-2) Jul 08 14:03:18 double free problem, usually are a program problem, not a glibc problem Jul 08 14:03:18 glibc-2.8-3.x86_64 Jul 08 14:04:02 diego_at_work: Then why does mine not work but other do? Jul 08 14:04:44 *other's do Jul 08 14:04:52 *does. Jesus it's early Jul 08 14:05:29 diego_at_work, we know, just trying to identify all the variables Jul 08 14:08:29 have you tried to track down which program give the double free? Jul 08 14:09:17 it looks like opkg-cl Jul 08 14:11:50 Crofton, hav you tried to run it with malloc debug on? Jul 08 14:12:25 http://linux.die.net/man/3/malloc Jul 08 14:13:25 export MALLOC_CHECK_=2 should do the trick Jul 08 14:15:08 diego_at_work, I haven't reporoduced the problem Jul 08 14:19:23 piroko, have you tried with MALLOC_CHECK_=2? Jul 08 14:26:50 piroko: are you building stable or devel branch? Jul 08 14:37:29 , Jul 08 14:37:48 What's the new location for http://www.openembedded.org/snapshots/OE.mtn.bz2 ? Jul 08 14:56:54 anyone building for lite5200? Jul 08 15:07:26 DuckFault : try this http://wiki.openembedded.net/snapshots/ Jul 08 15:07:41 thanks! Jul 08 15:08:07 DuckFault : u r welocome Jul 08 15:08:13 *welcome Jul 08 16:51:42 diego_at_work: Sorry, I was away Jul 08 16:51:52 diego_at_work: I have not tried that. I am building the dev branch Jul 08 16:52:38 diego_at_work: Is that an environment variable or a bitbake variable? Jul 08 17:09:18 Well with MALLOC_CHECK_=2, I get this: http://rafb.net/p/aledvm34.html Jul 08 17:10:27 I'll be back Jul 08 17:25:09 03koen 07org.oe.dev * r6fbe937c... 10/ (1 packages/tar/tar_1.20.bb): tar: add 1.20 Jul 08 17:29:13 Well I got the base-image to build after doing the bitbake from an ubuntu live-cd chrooted to my archlinux partition Jul 08 17:29:29 So I'm guessing it's my glibc version. Is there anything I can do to get it working under archlinux/ Jul 08 17:29:32 ? Jul 08 17:36:18 03koen 07org.oe.dev * r20d2ef79... 10/ (8 files in 5 dirs): webkit-gtk: bump SRCREV Jul 08 17:41:10 Maybe an easier problem now Jul 08 17:41:27 After the image was written, I get the same errors I was getting before on boot, as seen here: http://rafb.net/p/rzRGiC96.html Jul 08 17:41:54 I wrote the jffs2 image to NAND flash using the SAM-BA linux tool Jul 08 17:44:00 I must've done something wrong, but I haven't the slightest clue what Jul 08 17:50:04 Different question. Let's say I have a custom defconfig for the linux kernel that I want to use Jul 08 17:50:29 What is the recommended way to get it copied to .config so that linux will be built with that config? Jul 08 17:51:12 piroko: My method is to use user.collection/packages/linux///defconfig Jul 08 17:51:58 DuckFault: Where is user.collection? Jul 08 17:52:11 piroko: create it as an overlay parallel to the oe repo Jul 08 17:52:49 :( Jul 08 17:52:57 build/conf/site.conf should be updated to indicate the location of additional BBFILES and the collection names Jul 08 17:55:08 It's so difficult to keep track of everything that bitbake looks at... Jul 08 17:55:58 Especially since angstrom does some nonstandard stuff with the kernel Jul 08 17:56:17 Is there a guide I could read that teaches the ins and outs of how all this stuff gets handled? Jul 08 17:56:31 There seems to be a million and one ways to do the same task, and no standards to go by :( Jul 08 17:56:52 http://pastebin.com/d7df3cc2b Jul 08 17:57:05 piroko: well, it has to be that way to be as flexible as it is Jul 08 17:57:26 piroko: http://www.openembedded.info/usermanual.htm Jul 08 17:57:30 That is the best manual I've seen. Jul 08 17:58:30 The requested URL /usermanual.htm was not found on this server. Jul 08 17:58:55 whoops Jul 08 17:58:58 add an l at the end Jul 08 17:58:59 its .html Jul 08 17:59:05 AH ok Jul 08 18:00:19 DuckFault: Thank you. I will definitely read up on this Jul 08 18:08:11 hi Jul 08 18:12:23 who * Jul 08 18:17:18 *? Jul 08 18:27:18 It appears that the base-image I'm building is only allocating a 4M partition. Any idea why this might be? Jul 08 19:26:25 Hello there Jul 08 19:26:39 How are you Jul 08 19:27:19 About the modules configuration for OE, what loads when init starts is configured where? Jul 08 19:27:27 I tried the modules.conf and modutils.. nothing . Jul 08 19:27:44 they are all blank files and empty directories Jul 08 19:29:13 anyone know where may I look for the modules configuration, is it even configurable via a recipe? Jul 08 19:43:21 hernan_martinez: kernel modules -- thinking Jul 08 19:44:44 hernan_martinez: lines like the following will cause the modules to autoload and entries to be placed in modules.conf Jul 08 19:44:48 module_autoload_snd-soc-phreedom = "snd-soc-phreedom" Jul 08 19:44:50 module_autoload_pxa27x-keypad = "pxa27x-keypad" Jul 08 19:44:58 put the above in your kernel recipe Jul 08 19:45:28 hum ok Jul 08 19:45:47 this might not be a oe related question, now that i am thinking well about it Jul 08 19:46:09 my problem is that I have two devices using profilic usb-serial converter Jul 08 19:46:19 and one ftdi ft232 usb-serial converter Jul 08 19:46:57 i wanted the ft232 driver to load before pl2303 so my ftdi chip is always ttyUSB0 Jul 08 19:48:12 hernan_martinez: stuff like the ftdi drivers are autoloaded by the kernel, so I'm not sure how offhand to order the loading Jul 08 19:48:42 before using oe i was using mknod,, but thats was with kernel 2.4 Jul 08 19:49:02 hernan_martinez: yeah, all this autostuff is not without tradeoffs Jul 08 19:49:02 i know that now hutplugging usb devices are controlled by hotplug and udev Jul 08 19:49:20 yes Jul 08 19:49:21 hernan_martinez: I would probably try to convince udev to give each device a unique name Jul 08 19:49:48 yes i thought it was something like that.. with a devid Jul 08 19:49:56 hernan_martinez: I do that with USB flash devices -- I always want them to show up as /dev/usbdisk Jul 08 19:50:21 ok Jul 08 19:50:43 ill try to see what i can do. butthat well need some reading before hehe, also i need to tell the guy on charge of the programming that Jul 08 19:51:16 i dont even know if acessing hardware via /dev/ttyUSBx is the nowadays way Jul 08 19:51:52 ill check the unique name stuff Jul 08 19:51:52 thanks Jul 08 19:52:29 hernan_martinez: sure. I can't find an example off hand, but there should be some info in the udev docs Jul 08 19:53:12 cbrake Jul 08 19:53:22 are you developing with pxa270? Jul 08 19:53:25 what device? Jul 08 19:53:45 methril|laptop: several cm-x270 based designs and one full custom PXA270 design Jul 08 19:54:21 oh!! i have the em-x270 but i'm not able to find the JTAG pins (i'm sure that it's possible) Jul 08 19:54:30 maybe i have to ask compulab people Jul 08 19:54:41 but really nice Jul 08 19:55:02 methril|laptop: heh, I have a em-x270 as well :-) Jul 08 19:55:44 yes? nice!! do you check the EB-X270? has a connector that looks like a JTAG, but looking the EB datasheet it's not a JTAG Jul 08 19:55:59 i have to look some PXA270 docs to check the signals ;) Jul 08 19:56:43 well, the device it's not mine, but i'm with the hands on it : Jul 08 19:57:09 well, time to cook Jul 08 19:57:40 methril|laptop: I don't see anything about jtag offhand in the docs, but have never looked too closely Jul 08 19:57:52 ok Jul 08 19:57:57 methril|laptop: the cm-x270 has pads you can solder to, so the em-x270 may have someth similiar Jul 08 19:57:58 i'll try it later Jul 08 19:58:06 i supousse Jul 08 20:07:06 * * OE Bug 4416 has been created by  Jul 08 20:07:08 * * qtopia-core-4.3.3-autobuild Jul 08 20:07:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4416 Jul 08 20:58:06 * * OE Bug 4417 has been created by  Jul 08 20:58:08 * * glib-2.0-2.16.3-autobuild Jul 08 20:58:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4417 Jul 08 21:00:06 * * OE Bug 4418 has been created by  Jul 08 21:00:09 * * qt-embedded-4.4.0-autobuild Jul 08 21:00:11 * * http://bugs.openembedded.net/show_bug.cgi?id=4418 Jul 08 21:23:27 anyone know what I may be missing in my kernel to try to boot with a USB Filesystem as rootfs? my kernel console msgs show that usb host is detected, and usb mass storage support is registered, but I never see the usb device detected - same kernel booted from ATA disk will detect USB dev /dev/sda 'after' rootfs is mounted from hda Jul 08 23:45:11 what is the recommended way to add a custom defconfig for a new x86 based machine? I've created a custom machine conf, but I'm not sure if I should create a new kernel package or modify linux-x86_*.bb **** ENDING LOGGING AT Wed Jul 09 02:59:56 2008