**** BEGIN LOGGING AT Fri Jul 15 02:59:57 2011
Jul 15 07:54:33 <ant_work> I was about writing an RFC for image-device-tables. Then I tested with very few device nodes and again, without input/events kexecboot crashes
Jul 15 07:55:29 <ant_work> iirc once kexecboot was surviving in this case
Jul 15 07:56:55 <Jay7> iirc, it is exiting when no input devices was found
Jul 15 07:57:03 <Jay7> so it does panic
Jul 15 07:57:04 <ant_work> http://git.linuxtogo.org/?p=groups/kexecboot/kexecboot.git;a=blobdiff;f=kexecboot.c;h=1772bf78ca23eb31ce52bd07a443347b55482594;hp=06a68ac80ddb2e40f0f8877eb711e155c9fd1871;hb=51940a6ce14e8cfc8d1068858075ee19bacd8493;hpb=469416a2afc10605d96379e7230153736731f673
Jul 15 07:57:13 <ant_work> probably before this
Jul 15 08:04:36 <ant_work> btw blueighthning almost promised he'll start commiting meta-handhelds today
Jul 15 08:05:13 <ant_work> I've almost finished with the new meta-zaurus , just minor klibc stuff
Jul 15 08:05:24 <ant_work> after that I'll have 2 weeks for kexecboot
Jul 15 08:05:50 <ant_work> well, from 11pm to 02am as usual :p
Jul 15 08:07:19 <Jay7> nice :)
Jul 15 08:10:24 <ant_work> Jay7: as far I could se, it 'should' be possile to goback to the old / fast event interface
Jul 15 08:10:44 <ant_work> but we loose key[0-9] and other stuff
Jul 15 08:11:20 <ant_work> porbably now there are too many commits onthe top of it
Jul 15 08:11:31 <ant_work> no other chanche todebug it
Jul 15 08:11:56 <Jay7> it's easier to compare kexecboot with and w/o that commit
Jul 15 08:13:04 <Jay7> text ui is fast
Jul 15 08:13:23 <Jay7> so I'm not sure that problem is in events processing code..
Jul 15 08:15:02 <ant_work> we'll have to print menu-id in the menu-label to see the ordering
Jul 15 08:22:51 <ant_work> bluelightning: hey
Jul 15 08:23:03 <ant_work> to finish the do_clean story,
Jul 15 08:23:05 <bluelightning> hi ant_work
Jul 15 08:23:22 <ant_work> bitbake -c clean linux (or linux-kexecboot)
Jul 15 08:23:39 <ant_work> actually removes the kernel and the symlink from DEPLOY_DIR_IMAGE
Jul 15 08:23:53 <ant_work> I tested on oe-core
Jul 15 08:24:23 <ant_work> the same DOES not happen for initramfs-images and e.g. for zaurus-updater / zaurus-installer
Jul 15 08:24:35 <bluelightning> right... maybe this is an sstate thing?
Jul 15 08:25:02 <bluelightning> you might need to add the files to a list somewhere, I can't remember how it works but there should be some examples
Jul 15 08:25:24 <ant_work> well, adding the deploy class ?
Jul 15 08:25:46 <ant_work> afaik this is the modern package_stagefile_shell
Jul 15 18:59:49 <Jay7> so much questions in angstrom-devel/users about images for zauruses..
**** ENDING LOGGING AT Sat Jul 16 02:59:57 2011