**** BEGIN LOGGING AT Sat Jan 26 03:00:00 2013
Jan 26 08:07:33 <Jay7> "I came across a bug in gcc (4.6 or 4.7) that causes it to produce invalid code for some accesses through volatile pointers."
Jan 26 08:07:39 <Jay7> from qi-hardware ML
Jan 26 08:07:53 <Jay7> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56098
Jan 26 08:08:04 <Jay7> this may hit us as well
Jan 26 08:08:32 <Jay7> ah, -O2
Jan 26 08:08:41 <Jay7> we are using -Os iirc
Jan 26 22:30:58 <ant_home> bluelightning: I think I've got al least collie mtd :)
Jan 26 22:31:11 <bluelightning> ant_home: nice :)
Jan 26 22:32:00 <ant_home> fb still broken :/
Jan 26 22:32:29 <ant_home> imagine after resume sometimes fb-gui appears but misplaced
Jan 26 22:32:40 <ant_home> on first boot just twxt
Jan 26 22:32:46 <ant_home> *text
Jan 26 23:51:01 <ant_home> Jay7: somehow devtmps seems failing on collie
Jan 26 23:51:18 <ant_home> I have to populate /dev then kexecboot starts
Jan 26 23:59:48 <ant_home> anyway, here is scanning cramfs :)
Jan 26 23:59:49 <ant_home> http://paste.debian.net/229120/
Jan 27 00:23:21 <ant_home> bluelightning: years lost for this bug on sa1100 :/
Jan 27 00:39:32 <bluelightning> :(
Jan 27 00:47:01 <ant_home> bluelightning: what about two overrides ?
Jan 27 00:47:09 <ant_home> export CC_armv4_linux-gnueabi = "${TARGET_PREFIX}klcc -march=armv4 -mthumb-interwork"
Jan 27 00:47:13 <ant_home> seems to work
Jan 27 00:47:55 <bluelightning> it's always supposed to have worked, but I noticed it was broken a while ago... Richard fixed it shortly afterwards
Jan 27 00:48:14 <bluelightning> I believe that fix went into danny FWIW
Jan 27 00:48:32 <ant_home> pls correct me: we want this in case of eabi builds for armv4
Jan 27 00:49:34 <ant_home> linux-uclibceabi as well
Jan 27 00:50:01 <ant_home> interworking is mandatory for eabi conformance
Jan 27 00:50:50 <ant_home> CC_armv4 =  would be easy, dunno if acceptable
Jan 27 00:51:43 <ant_home> I see this override only in klibc.inc
Jan 27 00:51:52 <ant_home> put by Khem long ago
Jan 27 00:54:14 <ant_home> in fact in siteinfo.bbclass there are only two *eabi
Jan 27 00:56:26 <ant_home> zZZ now
Jan 27 00:56:27 <ant_home> gn
**** ENDING LOGGING AT Sun Jan 27 02:59:59 2013