**** BEGIN LOGGING AT Thu Aug 05 02:59:57 2010
Aug 05 07:39:22 <Seppoz> * opkg_install_cmd: Cannot install package locale-base-en-us.
Aug 05 07:39:29 <Seppoz> anybody ever got that be4
Aug 05 08:16:45 <eltazar> hi all, I am building Angstrom for a DM365 device
Aug 05 08:17:12 <eltazar> is there someone who can suggest me a suitable oe-devel stable commit?
Aug 05 11:34:41 <raubvogel> Update in my Zaurus adventures: I grabbed another sd card -- the one the PO used -- and untarred what narcissus created. Tried to boot. It froze on the same location. Booted using kexecboot and told it to scan for bootable partitions. It just hanged there seemingly doingnothing.
Aug 05 11:36:26 <ant_work> hi
Aug 05 11:36:42 <raubvogel> ant_work: Greetings
Aug 05 11:36:45 <ant_work> I can only guess it gets stuck during the scan
Aug 05 11:37:03 <ant_work> that's very strange, as I said that old version is reported as working
Aug 05 11:37:27 <raubvogel> I wonder if something boink happened when I installed kexecboot in it
Aug 05 11:37:41 <ant_work> not there, in case is the jffs2 partition
Aug 05 11:37:49 <ant_work> mtd partition
Aug 05 11:38:01 <raubvogel> Any way to check it
Aug 05 11:38:02 <raubvogel> '
Aug 05 11:38:03 <ant_work> I don't know exactly the nand layout of collie
Aug 05 11:38:19 <ant_work> iirc you showed 3 partitions yesterday
Aug 05 11:38:26 <ant_work> I found 5 in the net
Aug 05 11:38:27 <ant_work> :/
Aug 05 11:38:56 <ant_work> I'll have to see how it is set in the kernel sources
Aug 05 11:39:27 <raubvogel> Whenever you have the time
Aug 05 11:39:32 <raubvogel> No hurry
Aug 05 11:39:33 <ant_work> for other zaurus ( c5600 included) we read the exact partitioning from the nand
Aug 05 11:39:59 <ant_work> can you try reflashing the original sharp rom and then installing kexecboot zImage over that?
Aug 05 11:40:33 <raubvogel> I could; where would I get the sharp rom?
Aug 05 11:41:32 <ant_work> http://osdir.com/ml/drivers.mtd/linux.drivers.mtd/2002-04/msg00148.html
Aug 05 11:41:39 <ant_work> ^ this shows 5 partitions
Aug 05 11:41:49 <ant_work> 2002 = original I suppose
Aug 05 11:42:09 <ant_work> about images, there was smthg on the mailing list
Aug 05 11:42:31 <ant_work> try this link
Aug 05 11:42:33 <ant_work> http://forum.zaurusfr.org/viewtopic.php?f=8&t=4
Aug 05 11:44:36 <ant_work> ah ha
Aug 05 11:44:38 <ant_work> http://lxr.free-electrons.com/source/arch/arm/mach-sa1100/collie.c?v=2.6.26
Aug 05 11:45:03 <ant_work> ^^ it seems the exposed partitons now are 3
Aug 05 11:45:48 <raubvogel> So, my sl-5500 may not be lying after all?
Aug 05 11:46:28 <ant_work> same layout in 2.6.35
Aug 05 11:46:32 <ant_work> good news
Aug 05 11:48:03 <raubvogel> indeed!
Aug 05 11:52:03 <raubvogel> Well, it means I did not screw things but but it does not put my any closer to solving the mystery
Aug 05 11:54:11 <ant_work> I'm a bit confused because on collie there exist different kind of images...
Aug 05 11:54:14 <ant_work> zImage-collie-Xmemory-Ystorage.bin
Aug 05 11:54:25 <ant_work> Where Xmemory is the amount of RAM devoted to system memory, and Ystorage is the amount of RAM devoted to storage space. Xmemory+Ystorage should equal 64MB for SL5500 owners and 32MB for SL5000D owners
Aug 05 11:54:46 <ant_work> I don't know what could happen to the mtd after using one of those
Aug 05 11:55:34 <ant_work> 5600, 700, 750, .. have a different 8Mb nand
Aug 05 11:56:04 <ant_work> I see our kexecboot kernel uses ram=64m
Aug 05 11:56:15 <ant_work> so I suspect the split is 64:0
Aug 05 11:56:52 <ant_work> but really it's the first time I hear about issues on collie
Aug 05 11:56:57 <raubvogel> Weird
Aug 05 11:57:03 <raubvogel> Maybe I am the chosen one ;)
Aug 05 11:57:04 <ant_work> after all, was the prototypalmachine
Aug 05 11:57:23 <raubvogel> Hey, I need to run away for an hour. BBL
Aug 05 11:57:27 <ant_work> np
Aug 05 13:32:26 <jamuraa> I've been trying to get the 'iw' tool to compile, and when I run it the first time I get 'fatal: no names found, can't describe anything' which looks like a git error, and the second time I get a ld error related to 'iw_version' (which I assume is caused by the first error).  Is there a way to get a better idea what's going wrong? 
Aug 05 13:36:55 <raubvogel> Back
Aug 05 13:37:23 <raubvogel> ant_work: What shall we try next?
Aug 05 13:37:46 <ant_work> you should revert to a known state
Aug 05 13:38:40 <ant_work> anyway, you could test an even older kexecboot just  to rule out some doubts
Aug 05 13:38:44 <ant_work> http://www.angstrom-distribution.org/demo/kexecboot/
Aug 05 13:39:08 <raubvogel> Ok
Aug 05 13:39:14 <raubvogel> BTW, myzaurus.com is no longer
Aug 05 13:39:16 <ant_work> iirc this one could not boot from CF but only from SD... can't remember
Aug 05 13:39:27 <ant_work> some missing driver in kernel
Aug 05 13:39:31 <ant_work> config
Aug 05 13:40:00 <raubvogel> Don't you mean the other way around? I thought by default the sl-550 coul donly boot off the CF
Aug 05 13:41:37 <ant_work> I cant remember on collie, on c7x0 and spitz was CF the issue
Aug 05 13:42:39 <raubvogel> I was just thinking that it would not go to flash mode if I used a sd but had no problems with my cf
Aug 05 13:43:06 <ant_work> I mean after, linux-kexecboot had an issue
Aug 05 13:43:16 <ant_work> not the embedded flasher
Aug 05 13:43:27 <raubvogel> Oh, i see now :)
Aug 05 16:25:27 <jamuraa> So I have a patch to a oe recipe, which lets you compile iw inside the git repo that is setup by the oebb.sh script..  I'm not sure where to put it. It doesn't seem right in the oe repository because iw (should) compile fine if you haven't set up your environment as oebb.sh from Angstrom does. 
Aug 05 20:09:00 <andyj> man the do_rootfs  build stage for angstrom-gnome-image-big is a beast
Aug 05 20:14:45 <andyj> anyone have a decent estimate for time it takes the last stage of that image to do_rootfs?
Aug 05 20:17:51 <Seppoz> umm some min
Aug 05 20:18:10 <Seppoz> depends on what u set
Aug 05 20:18:14 <Seppoz> as type
Aug 05 20:18:28 <andyj> angstrom-gnome-image-big
Aug 05 20:18:43 <andyj> is there some other type I can set?
Aug 05 20:19:21 <Seppoz> umm i mean the type
Aug 05 20:19:24 <Seppoz> either jffs2
Aug 05 20:19:29 <Seppoz> or tar.bz
Aug 05 20:19:32 <Seppoz> or ext3
Aug 05 20:19:39 <andyj> oh let me check my conf real quick
Aug 05 20:19:43 <Seppoz> gnome images can get bix
Aug 05 20:19:47 <Seppoz> *big
Aug 05 20:20:36 <andyj> IMAGE_FSTYPES="jffs2 tar"
Aug 05 20:21:01 <Seppoz> so check your deploy/glibc/board folder
Aug 05 20:21:06 <Seppoz> and see how big the images get
Aug 05 20:22:21 <andyj> tar = 647895040 in ls
Aug 05 20:22:27 <Seppoz> there ya go
Aug 05 20:22:40 <andyj> so I'm taking that as 647MB or Mb?
Aug 05 20:22:45 <Seppoz> did the jffs stated yet?
Aug 05 20:22:51 <Seppoz> yes
Aug 05 20:23:30 <andyj> been stuck at this point for quite sometime now:   NOTE: Running task 7492 of 7493 (ID: 10, /home/andy/oe/ODT/openembedded/recipes/images/angstrom-gnome-image-big.bb, do_rootfs)
Aug 05 20:23:30 <andyj> NOTE: package angstrom-gnome-image-big-1.0-r0: task do_rootfs: Started
Aug 05 20:23:35 <Seppoz> see if its still growing by typing the command again
Aug 05 20:23:42 <Seppoz> that looks just fine
Aug 05 20:23:57 <Seppoz> the bigger your rootfs is the longer it will take to build
Aug 05 20:24:18 <andyj> it is still the same size
Aug 05 20:24:44 <Seppoz> df
Aug 05 20:25:14 <andyj> ya df is changing
Aug 05 20:26:24 <andyj> but I have two other packages running compiles so can't be certain it's still bitbaking proper I guess..
Aug 05 20:33:22 <Seppoz> i guess it just take a while
Aug 05 21:05:35 <andyj> I'm really stumped, even after cleaning my tmp and rebuilding angstrom-gnome-image-big I still have no mcookie in the rootfs...  ?
Aug 05 21:06:09 <andyj> somehow util-linux-ng isn't being installed I guess?
Aug 05 22:43:29 <andyj> how can I check that my OE build is setup properly to install util-linux-ng instead of using my build systems native tools for that package?
Aug 05 22:47:33 <Seppoz> mcookie? is that a hamburger from mcdonals? xD
Aug 05 23:02:31 <Seppoz> where is the cross compiler located after build?
**** ENDING LOGGING AT Fri Aug 06 02:59:58 2010