[00:23] <erappleman> mlankhorst, re the hybrid spec. does airlie still plan to have power management ready for 3.8? i haven't seen him prep it for -next or anything that should've been done already
[00:26] <mlankhorst> seems to be delayed
[00:27] <mlankhorst> I'm reviving my synchronization work though
[00:29] <erappleman> your tree is very promising
[00:29] <mlankhorst> it's slow work though, I never touched radeon before, the i915 commits *might* work, but didn't test it much
[00:30] <erappleman> i could try rolling the kernel over the next few days
[00:31] <mlankhorst> it's not ready yet, the first half deals with getting ttm locking changed in a more useful way, after that some changes for debuggability
[00:34] <erappleman> thanks, i'll keep an eye on developments
[00:35] <mlankhorst> it would be useful if someone could review the mutex patch I posted though
[00:39] <mlankhorst> part of the changes were making ttm use an extended version of mutexes for implementing reservations, but without that patch in I can't get the rest upstreamed
[00:41] <erappleman> mutex: add support for reservation style locks?
[00:42] <erappleman> couldn't hurt to let dri-devel look at it
[00:42] <erappleman> or would that be an lkml thing
[00:44] <mlankhorst> suppose i could repost and cc dri-devel
[00:44] <mlankhorst> and now that I actually got the preliminary ttm patches in elaborate a bit more in the commit on how to use it
=== soren_ is now known as soren
=== Quintasan_ is now known as Quintasan
=== shadeslayer is now known as buildslayer