03:32 rhyskidd: Lyude: Do you have access to one of those Chamelium testing boards?
10:53 RSpliet: Integer dot product (aka. the neural net op)?
11:53 karolherbst: fun, doing a piglit run and wondering why the machine crashes. Then I noticed, Nouveau wasn't even loaded
11:57 pmoreau: :-p
11:57 pmoreau: Good job!
11:57 pmoreau: Though, to be honest, similar things happened to me as well.
11:59 karolherbst: well, it is a Coffee Lake CPU, so the intel driver seems to be still a bit buggy
12:04 karolherbst: pmoreau: is there some review you still need for some clover patches?
12:04 karolherbst: wondering when we can actually merge those
12:04 pmoreau: Good questions, I need to have a look at it. :-D
12:05 karolherbst: would be good to merge those things first, so that the other stuff makes more sense
12:05 pmoreau: I was wondering whether to put the SPIR-V validation thing directly in clover/spirv/invocation.cpp, rather than in clover/api/program.cpp and move them to clover/spirv/invocation.cpp in the second series.
12:05 pmoreau: Agreed
12:06 karolherbst: I think the spirv specific code should stay in clover/spirv
12:07 karolherbst: I don't really like the fact, that the interface is different for each backend anyway :/ but this mistake was made befor your spirv work already anyway
12:07 karolherbst: *before
12:07 karolherbst: when I have some time I might even cleanup clover generally
12:08 pmoreau: I’ll change patch 16 then.
12:09 pmoreau: Anyway, I need to go back to classes, ttyl
16:46 Lyude: rhyskidd: mhm!
16:47 Lyude: I wrote the igt support for them
16:47 Lyude: well, the initial stuff. an intel intern helped a lot as well :)
17:38 karolherbst: imirkin: for atomics I have to set the mask to 0x1 though
17:42 imirkin_: mmmmmmmaybe
17:42 imirkin_: dunno
17:42 imirkin_: probably a good idea.
17:42 imirkin_: i don't think it matters tho
17:42 karolherbst: and it seems the hardware is unhappy about 0xf in a few other places
17:42 karolherbst: well, I got hangs
17:43 imirkin_: ok
17:43 karolherbst: I coulg imagine that for textures it actually is important
17:43 karolherbst: I mean tex instructions
17:43 imirkin_: for tex it's definitely extremely important
17:43 imirkin_: i meant for atomics tho
17:43 karolherbst: right
17:44 karolherbst: I will take a deeper look tomorrow
17:44 imirkin_: as always... just do what tgsi does.
17:44 karolherbst: yeah
17:45 karolherbst: the spilling fails I saw with kepler1 should be real bugs though. I remember seeing them somewhere else
17:46 karolherbst: we just fail spilling at values of 64 bits or more
17:46 karolherbst: *for
17:48 karolherbst: it is usually okay if your access starts at the .x component, but if it starts at something else, it breaks
17:48 karolherbst: anyway, gtg
21:33 rhyskidd: Lyude: oh great! i think i will take you up on that. that hardware looks really helpful for validation. was Paul K the intern?
21:39 Lyude: rhyskidd: yep! :), keep in mind you might want to make sure what you need to test will probably need to be supported in IGT since I don't really have much else setup
22:10 mupuf: rhyskidd: before working on the chamelium support and testing for nouveau, we need to have a way to get a CRC and/or frames out of our crtcs
22:11 mupuf: this would unlock a lot of IGT tests