08:23 imirkin: mogorva: can you try http://hastebin.com/evetumezor.py instead of the other fix for the crash-on-start issue? i think it may be better, if it works. unfortunately i can't repro the situation with a piglit test :(
08:24 mogorva: kk
08:38 heath_: nouveau doesn't have a released driver currently correct?
08:40 mlankhorst: no
08:42 imirkin: i think the first "stable" release (i.e. inclusion in the kernel, stable ABI, etc) happened in 2012
08:43 imirkin: but it was in the kernel's staging for much longer than that
08:43 imirkin: mogorva: on second thought, i think i like the original patch
08:44 mogorva: imirkin: the patch works very well, fixes the crash-on-start issue in all 4 games that I know of, thank you :)
08:44 mogorva: the previous patch didn't work with all the games
08:45 imirkin: mogorva: oh?
08:46 mogorva: can you explain what exactly the problem is?
08:46 imirkin: no. and that's the problem.
08:46 imirkin: i'm stumbling in the dark
08:46 imirkin: i thought i knew, but then i tried to make a test case, and it didn't crash
08:47 mogorva: did you manage to reproduce the crash in any of the demos that I showed you?
08:47 imirkin: i haven't tried that yet :)
08:48 imirkin: my next step is to make an apitrace and then replay it
08:48 imirkin: but i gtg shortly
08:48 imirkin: so wait, did my second patch fix the AA thing too?
08:48 mogorva: thx again and let me know if can help with testing or logs
08:48 imirkin: with naval whatever
08:48 mogorva: yep
08:49 imirkin: interesting.
08:49 mogorva: without the patch that game crashes if anti-aliasing is enabled
08:49 imirkin: and that's the second patch, to main/context.c right?
08:50 mogorva: yep
08:50 imirkin: hrmph
08:50 imirkin: i guess that might quasi-make sense.
08:50 mogorva: only one game remained that still crashes, but that could be a different problem because it occurs mid-game
08:51 imirkin: well, despite fixing things, i still think it's the wrong solution...
09:29 imirkin_: buhman: really, 50% of nvidia blob? that seems awfully good for that perf level
09:29 buhman: yeah
09:29 imirkin_: or do you mean nvidia at a comparable perf level?
09:29 buhman: no
09:30 buhman: I tried uningine again, and it actually only does like 6fps contrast to what I remembered
09:30 buhman: (on nvidia)
09:30 imirkin_: ah :)
09:30 imirkin_: but it does tess
09:30 imirkin_: but you can try my tess branch
09:30 buhman: after turning it off
09:30 imirkin_: but it doesn't really work
09:30 imirkin_: it only sorta works :)
09:31 buhman: I think it sounds like you're asking for more scientific comparisons ;p
09:31 imirkin_: not really
14:22 hakzsam_: imirkin, I just tried to remove the double query thing, works fine as expected :)
14:22 imirkin_: great
14:26 hakzsam_: do you plan to review more patches of the series or do you prefer to see the v2 before continuing the review?
14:26 imirkin_: mmmm
14:26 imirkin_: no, you don't have to resend
14:26 imirkin_: [at least not yet]
14:27 hakzsam_: okay
14:31 imirkin_: gnurou: care to provide odds of me getting a response to my latest question to nvidia?
14:34 hakzsam_: imirkin_, in case you want to take a look, the v2 can be found here http://cgit.freedesktop.org/~hakzsam/mesa/log/?h=nouveau_perfmon_v2
18:07 gnurou: imirkin_: errr which question was it?
18:08 imirkin: gnurou: the one about which ops it's ok to do c0[$a0+4] with
18:08 imirkin: sent earlier today
18:08 imirkin: [i'm not expecting an answer *today*, i know better... but say in the next 30 days]
18:42 gnurou: imirkin: mmm cannot see it in my inbox - which list did you send it to?
18:42 gnurou: I am on gpu-public-documentation but maybe you used another one?
18:42 imirkin: gpu-public-documentation@nvidia.com
18:42 imirkin: subject: "What are the restrictions around loading indirect constbuf values"
18:43 imirkin: if my emails are coming up as spam, that certainly answers why i'm not getting replies ;)
18:43 airlied:won't mention the external security list that only accepted mails from internally.... oops I just did :-P
18:44 skeggsb_: airlied: hehe, well, that address definitely *usually* works
18:45 gnurou: imirkin: oh wait, looks like I am *not* on that list
18:46 imirkin: i also sent it to nouveau@ :)
18:46 gnurou: but strangely I got an email that skeggsb_ sent to it earlier this week...
18:46 gnurou: ah, I have it @nouveau :)
18:46 imirkin: perhaps someone just fixed the glitch? :)
18:46 skeggsb_:feels special
18:47 skeggsb_:feels even more special without an _
18:47 skeggsb_: wtf.. ok, that didn't work
18:47 gnurou: ok, it got sorted into my Nouveau folder instead of Inbox
18:47 gnurou: stupid Outlook
18:48 gnurou: but that doesn't tell me whether I will see the internal replies