**** BEGIN LOGGING AT Wed Aug 03 02:59:57 2011 **** BEGIN LOGGING AT Wed Aug 03 14:05:28 2011 Aug 03 14:26:36 eranm: it was http://www.angstrom-distribution.org/recent-kexecboot-ipaq-hx4700 Aug 03 14:28:29 I got to the boot screen, but couldn't get kexecboot to actually boot from my sd card. Aug 03 14:31:00 have you checked the .config used for hx4700 ? Aug 03 14:36:15 Nope, where could I find it? I pulled the git tree for kexecboot and built it, but I've noticed it only builds the kexecboot binary (not the zImage + initramfs + kexecboot) Aug 03 14:41:45 yes, this latter is built alone (we use OpenEmbedded) Aug 03 14:43:00 is this kernel Aug 03 14:43:02 http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/linux/linux-kexecboot_2.6.21-hh20.bb Aug 03 14:43:18 http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/linux/linux-kexecboot-2.6.21-hh20/hx4700/defconfig Aug 03 14:46:35 great, thanks Aug 03 15:17:17 Jay7: [17:02] koen ant_work: create a fat loopfs in ram, export that over usb gadget as filesystem. use host computer to copy over kernel and boot.cfg, eject and kexecboot will pick it up as option Aug 03 15:18:08 [17:04] av500 koen: we do that for SW updates, the FAT in RAM Aug 03 15:19:35 as far as I know, roots can be stacked so we can mount kexecboot-initramfs on the top of another initramfs Aug 03 16:39:28 hm.. Aug 03 16:39:57 eranm: alexxy is our hx4700 tester :) Aug 03 16:40:11 but not sure he is anknowledged yet ;) Aug 03 16:41:26 interesting idea about loop in ram Aug 03 16:41:40 If he comes around, I'll ask him about using it on the hx4700. Is there a place where the nightly builds (or equivalent) is available? all I could find is Zaurus images Aug 03 16:42:02 right now we have no such builds Aug 03 16:42:33 we are building it with OE every time when testing is needed Aug 03 21:33:51 Jay7: think about setting a stable tag (before new ui, with null check for sysmenu) Aug 03 21:34:09 so I'll build some kernels for ltg Aug 03 21:34:25 ant__: I'm thinking everytime :) Aug 03 21:34:38 just need some free time to do.. Aug 03 21:34:59 well, if you have time then debug it instead :) Aug 03 21:35:35 try to build with rev before merging fb functions with upstream Aug 03 21:35:36 or I could take an old git SRCREV and add the patch for sysmenu for the moment Aug 03 21:35:43 yes, this Aug 03 21:35:51 there was old menu Aug 03 21:35:53 we loose the 0-9 keys, though Aug 03 21:35:55 w/o icons Aug 03 21:36:05 I think 0-9 isn't very frequently used Aug 03 21:36:24 it wasnėt so back, iirc I had icons and old gui code Aug 03 21:36:42 do I remember wrong maybe? Aug 03 21:36:58 w/o icons in sysmenu I mean Aug 03 21:37:09 ah Aug 03 21:37:36 I'l'm sure that if you 'unroll' that code the bug will be evident Aug 03 21:37:41 is too optimized ;) Aug 03 21:37:55 it was just rewritten.. Aug 03 21:39:43 now two words about losetup stuff Aug 03 21:40:12 iirc the issue was you cannot 'pivot', you are rootfs Aug 03 21:40:42 but there are commits in klibc for the new switch_root (old run_init) Aug 03 21:40:57 yes, switch_root is in klibc Aug 03 21:41:08 well, run_init it was Aug 03 21:41:16 you shoul read some ML msgs. maybe Aug 03 21:42:33 what ML? Aug 03 21:43:21 http://www.zytor.com/mailman/listinfo/klibc Aug 03 21:43:49 btw, I'm mumbling around multiple initramfs Aug 03 21:43:51 http://en.gentoo-wiki.com/wiki/Initramfs#Multiple_initramfs Aug 03 21:49:59 but it seems kexec_load can accept a single initramfs Aug 03 21:50:03 I'll check the code Aug 03 23:09:06 hi bluelightning **** ENDING LOGGING AT Thu Aug 04 02:59:57 2011