**** BEGIN LOGGING AT Fri Dec 08 02:59:59 2006 Dec 08 09:01:37 * XorA wishes switching kernels on Z was easier Dec 08 09:02:09 or that ALSA EABI/OABI were compatible Dec 08 09:05:19 ah right, the cse problem Dec 08 09:19:33 cse? Dec 08 09:19:49 nah thats fixed Dec 08 09:20:08 problem is the EABI/OBIA syscall mapping for alsa is fuxored Dec 08 10:34:54 gnash on OE is on the road :)) Dec 08 10:39:34 Genesis: cool Dec 08 10:42:19 you'll not commit it but nevermind Dec 08 10:48:42 Ill commit anything for some good alcohol Dec 08 10:55:48 i don't know the oE policy in matter of commit , should everyone like me inexperimented with SCM could merge some file without any advise on the dev tree ? Dec 08 11:12:56 koen: You about? Dec 08 11:13:05 yes Dec 08 11:13:30 do you remember my complaining a while back about the location of sl811 modules being wrong for 2.6.x kernels? Dec 08 11:13:49 vaguely Dec 08 11:13:53 and you saying that it couldn't be changed due to something ipaq specific Dec 08 11:13:59 isn;t sl811 included in the kernel nowadyas? Dec 08 11:14:21 not in openzaurus, it's a separate module, but perhaps that was the iPaq link Dec 08 11:14:51 do you know how the location and extension of a module is determined at run time after a card is inserted? Dec 08 11:16:03 I meant that it's build in tree, instead of building as an external module Dec 08 11:16:33 ah yes, that may be true, I've not checked for a while Dec 08 11:17:22 if it's built in-tree the location is correct Dec 08 11:17:26 http://pastebin.ca/271702 is my problem Dec 08 11:17:49 probably an easy fix - the path to the modules is wrong as are the extensions. Just wondering what I need to patch to fix it? Dec 08 11:18:12 it looks like it's using the old driver Dec 08 11:18:17 hc_sl811 Dec 08 11:18:25 Yep, is that something that will be in config? Dec 08 11:19:05 no idea Dec 08 11:19:22 ah, I think I see it Dec 08 11:19:32 /etc/pcmcia/config Dec 08 11:20:41 It's being bound to the old driver. I'll have to look at the recipe and add some kernel version dependent stuff to it Dec 08 11:53:02 have a nice weekend all Dec 08 11:53:41 actually, before I go, XorA, did you see this thread?: http://www.oesf.org/forums/index.php?showtopic=22280&hl= Dec 08 11:53:51 bye Dec 08 15:41:00 has someone a little experience with libltdl and OE packaging ? Dec 08 15:41:45 i tried to grep the dev tree , seems to noone use that Dec 08 16:58:32 RP: ever had jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x01280000: 0x0001 instead Dec 08 16:59:13 and similar lines with diff 0xblobty-blob-blob Dec 08 16:59:23 ade|desk: On akita? Dec 08 16:59:47 yup Dec 08 17:04:09 RP: fails to find init but does mount root ... so it says ... i don't believe it though Dec 08 17:04:52 ade|desk: Sounds like you corrupted the flash :-/ Dec 08 17:05:01 Its possible the file is no longer in the corrupted rootfs Dec 08 17:05:39 this is a fresh flash of angstrom built about 20 mins ago Dec 08 17:06:43 i had the same problem until I noticed the rootfs image didn't have a 16-byte header that other images i used have Dec 08 17:08:02 i added the header and the jffs2_scan_eraseblock didn't appear anymore and the kernel booted but everything went wrong that can go wrong so I disposed off that image... Dec 08 17:08:20 btw, was also on akita Dec 08 17:09:28 indeed looks like i have that missing also Dec 08 17:26:24 i wonder why the 16 byte header is not added atm Dec 08 17:28:20 flashing the initrd.bin with 16 byte header added .... lets see Dec 08 17:29:48 well it mounts and boots ... segfault heaven .... oh dear Dec 08 17:40:28 might try to build a uclibc rootfs with the arm NPTL+thumb patch this weekend while i am at work **** ENDING LOGGING AT Sat Dec 09 02:59:58 2006