05:31hakzsam: imirkin, could you push these two branches ? http://cgit.freedesktop.org/~hakzsam/mesa/log/?h=nvc0_mp_counters_nvf0 and http://cgit.freedesktop.org/~hakzsam/mesa/log/?h=nouveau_clang_warnings
10:26imirkin_: hakzsam: you have a fd.o account but no mesa push access?
10:30imirkin_: hakzsam: pushed
10:30imirkin_: thanks for doing the fixes ;)
10:30imirkin_: next -- perf counters on GK110/GK208 ;)
10:31imirkin_: or figure out why it's destroying the universe on fermi
11:02hakzsam: imirkin_, yeah, next is global perf counters for tesla, fermi and kepler :)
22:32imirkin: skeggsb: have you had a chance to look at the PDISP hangs bug which was bisected to a relevant-seeming commit?
22:32imirkin: skeggsb: https://bugs.freedesktop.org/show_bug.cgi?id=90276
22:39skeggsb: imirkin: no immediate ideas, except that for some reason, something is trying to turn the cursor on on head2/3 without a mode set..
22:41imirkin: would it be easy to prevent that?
22:41skeggsb: i don't really believe that bisect to have possibly caused that issue
22:43skeggsb: yes, but i'd like to know what/how it got caused to begin with if it really is a regression
22:44imirkin: a different user also seeing hangs misunderstood my suggestion of trying an older kernel and instead tried an older gnome
22:44imirkin: and... it helped
22:46skeggsb: that'd make more sense, probably
22:47skeggsb: - if (show && nv_crtc->cursor.nvbo)
22:47skeggsb: + if (show && nv_crtc->base.enabled && nv_crtc->cursor.nvbo)
22:47skeggsb: that, or something like it, would do it
23:16imirkin: skeggsb: any idea why stk 0.9 dies on start in libdrm?
23:16imirkin: i traced it down to a pushbuf pointer list corruption
23:16imirkin: but couldn't make out why it was happening