**** BEGIN LOGGING AT Mon Jan 07 02:59:59 2013
Jan 07 10:48:34 <ant_work> ah ha... eglibc.inc has  ARM_INSTRUCTION_SET = "arm"
Jan 07 10:53:27 <ant_work> my bad: http://lists.linuxtogo.org/pipermail/openembedded-commits/2011-June/058198.html
Jan 07 10:53:42 <ant_work> [oe-commits] Andrea Adami : klibc: don't force ARM_INSTRUCTION_SET = "arm"
Jan 07 11:46:38 <ant_work> Jay7: have you read zeldin's answer? OFC we'll keep 16 bit
Jan 07 12:29:01 <nareshbhat> ant_work: ping
Jan 07 12:29:15 <ant_work> pong
Jan 07 12:29:17 <ant_work> hello
Jan 07 12:30:09 <nareshbhat> ant_work: Hi, good news..I am successful to boot the Torvalds kernel with latest kexec-tools :) on vexpress cortex-a15 target
Jan 07 12:30:24 <ant_work> \°/
Jan 07 12:33:20 <ant_work> we have old kexec-klibc-static in OpenEmbedded, needing updates
Jan 07 12:36:22 <nareshbhat> ok
Jan 07 12:38:47 <ant_work> kexecboot should work, though
Jan 07 13:00:51 <nareshbhat> yep
Jan 07 13:40:41 <ant_work> nice, please post a picture somewhere
Jan 07 13:40:56 <ant_work> to add it to the screenshots
Jan 07 14:18:49 <nareshbhat> ant_work: I have not yet tested
Jan 07 14:19:16 <nareshbhat> will test and let you know
Jan 07 16:31:19 <Jay7> ant_work: so, do you found issue with collie?
Jan 07 16:31:27 <Jay7> *did
Jan 07 16:33:37 <ant_work> the problem is in the metadata
Jan 07 16:33:54 <ant_work> hopefully can be fixed in klibc recipe
Jan 07 16:34:28 <ant_work> binutils are also patched, eglibc-binaries do run on collie
Jan 07 16:40:25 <ant_work> Jay7: I've seen in kexecboot log the ioctl gives the color offsets
Jan 07 16:48:13 <Jay7> hm..
Jan 07 16:48:22 <Jay7> btw, yes
Jan 07 16:49:38 <Jay7> but anyway we need place bytes in proper order (BE/LE)
Jan 07 17:08:01 <ant_work> see this http://stackoverflow.com/questions/3629205/what-makes-this-code-endian-independent
Jan 07 17:28:57 <ant_work> oh, and this is also good
Jan 07 17:29:05 <ant_work> http://comments.gmane.org/gmane.comp.gcc.help/43248
Jan 07 17:29:22 <ant_work> "iirc, the C language gives no guarantee of the ordering of fields within machine words,
Jan 07 17:29:22 <ant_work> so if you use them for the h/w register modeling, you program will not only be nonportable,
Jan 07 17:29:22 <ant_work> it will be compiler-dependent too. you should really read suitable byte/word/doubleword
Jan 07 17:29:22 <ant_work> from mapped h/w memory, convert it with le/be_to_cpu() method and shift/mask to access bits."
Jan 07 20:01:14 <ant_home> Jay7: will you have some holidays?
Jan 07 20:01:31 <ant_home> (finally :)
Jan 07 20:07:17 <bluelightning> hi ant_home - sounds like you had some success?
Jan 07 20:07:44 <ant_home> yes and not ...
Jan 07 20:07:51 <ant_home> kernel is bad
Jan 07 20:08:00 <ant_home> klibc userspace is broken
Jan 07 20:08:08 <ant_home> at least we know the situation
Jan 07 20:08:41 <ant_home> kernel can only boot from SD atm
Jan 07 20:08:45 <ant_home> on collie
Jan 07 20:09:01 <ant_home> kexecbooted
Jan 07 20:13:34 <bluelightning> ok
**** ENDING LOGGING AT Tue Jan 08 02:59:58 2013