#asahi-gpu · 2021-07-10 17:27
xerpi[m]Let's hope this was not the issue and that the GPU also snoops the L1/L2, otherwise it will get hairy, you would also need to invalidate BOs when reading data the GPU has written..#asahi-gpu · 2021-07-10 17:22
jix(I was asking because I was wondering whether the windowserver error has anything to do with how you get stuff on screen in the end (not knowing how you do that/how that works on osx), but just everything starting to fault makes sense, especially after seeing OBS hang before)#asahi-gpu · 2021-07-10 17:17
jixis the windowserver log msg just forwarding errors from kernel, or is it having an additional error when this happens?#asahi-gpu · 2021-07-10 17:17
xerpi[m]Regarding this random hang/HW issue: are the CPU and GPU cache coherent? If so, only LLC or also L1/L2? Maybe you need to flush some shared stuff/descriptors up to LLC#asahi-gpu · 2021-07-10 17:11
bloomwondering how dirty tracking of user uniforms was ever supposed to work...#asahi-gpu · 2021-07-10 16:57
jixah yeah that was what I was missing, thanks :)#asahi-gpu · 2021-07-10 16:57
jix(I might be totaly confused though)#asahi-gpu · 2021-07-10 16:56
jixwouldn't this also change the thing you were testing last?#asahi-gpu · 2021-07-10 16:56
jixat least that's what I had expected#asahi-gpu · 2021-07-10 16:55
jixI'm a bit confused by having DIRTY_ANY defined to be a specific bit, or do you set it whenever you set another bit? (might have missed that)#asahi-gpu · 2021-07-10 16:10
jixmaybe that code got culled at some point ;)#asahi-gpu · 2021-07-10 16:07
jixno still fine#asahi-gpu · 2021-07-10 15:58
jixi wonder, are the handles at the end of the header and at the end of each entry, or are they actually at the start of each entry? (might be obvious with more context, but isn't to me)#asahi-gpu · 2021-07-10 15:52
jixyeah#asahi-gpu · 2021-07-10 15:52
bloomshould be back now#asahi-gpu · 2021-07-10 15:51
jixyeah can hear you#asahi-gpu · 2021-07-10 15:51
jixaudio is still fine but it only shows the broken demo#asahi-gpu · 2021-07-10 15:51
jixbloom: I think the stream freezed#asahi-gpu · 2021-07-10 15:16
jixbloom: yeah your change of zeroing everything was what I was thinking of when I wrote that#asahi-gpu · 2021-07-10 15:11
jixbloom: do you need to exclude already freed entries in track_free?#asahi-gpu · 2021-07-10 14:44
jixfound it again, it was https://www.realworldtech.com/tile-based-rasterization-nvidia-gpus/#asahi-gpu · 2021-07-10 14:40
jixI've heard that even recent-ish discrete desktop GPUs use tiling (need to dig out the blog post which had some code to actually demonstrate that)#asahi-gpu · 2021-07-10 14:35
jixcan hear you :)#asahi-gpu · 2021-07-10 14:34
bloomon air#asahi-gpu · 2021-07-10 14:30
bloomJust getting setu now#asahi-gpu · 2021-07-10 14:30
bloom:F#asahi-gpu · 2021-07-10 14:29
chadmedbloom: hence my dilemma, its almost 1am so i should sleep, but i also want to watch#asahi-gpu · 2021-07-10 14:18
jixworks for me, it's afternoon here, so me not sleeping is probably a good thing ^^#asahi-gpu · 2021-07-10 14:11
bloomAlright. Sure. Let me get a bite to eat first. 20 minutes, then? (Starting at ~10:30am?)#asahi-gpu · 2021-07-10 14:10
bloomI don't want to be the reason someone's not sleeping!#asahi-gpu · 2021-07-10 13:53
chadmednow im conflicted as to whether to fall asleep to such a stream or stay up and learn something #asahi-gpu · 2021-07-10 13:50
jixwould probably watch if you stream... might be too distracted to actually follow what's going on today though#asahi-gpu · 2021-07-10 13:39
bloommay or may not stream, thinking about it though#asahi-gpu · 2021-07-10 13:38
bloomthinking of doing some housekeeping on the mesa driver#asahi-gpu · 2021-07-07 03:04
alyssa(The rest are a mix: discard(), points, one-off bugs)#asahi-gpu · 2021-07-07 03:03
alyssaFor remaining fails, about half are mipmap and cubemap related and about a quarter are depth/stencil related#asahi-gpu · 2021-07-07 02:33
alyssaDo we not have a Twitter bot? ok#asahi-gpu · 2021-07-07 02:33
alyssahttps://twitter.com/alyssarzg/status/1412600093505589251#asahi-gpu · 2021-07-06 02:43
chadmed68k around then reminds me of whats going on with x86 now only marginally less bad #asahi-gpu · 2021-07-06 02:42
krbtgtnever let the amiga fanboys tell you differently, 68k sucks#asahi-gpu · 2021-07-06 02:42
krbtgtalso as someone who fixed emulation bugs in 68k FPUs they are hell and motorola just kept adding to the tire fire#asahi-gpu · 2021-07-06 02:42
krbtgtkind of funny#asahi-gpu · 2021-07-06 02:42
krbtgt486/68040 integrated it... the 487 was basically another 486DX to replace your existing 486SX#asahi-gpu · 2021-07-06 02:41
chadmedah right#asahi-gpu · 2021-07-06 02:41
Yuzu486, not pentium#asahi-gpu · 2021-07-06 02:41
chadmedone of the reasons the fast inverse sqrt function exists#asahi-gpu · 2021-07-06 02:41
chadmedoutboard were once the norm. intel only started integrating real floating point hardware with the pentiums iirc #asahi-gpu · 2021-07-06 02:41
krbtgtbecause there's an amiga like cult around them. oh, and glide#asahi-gpu · 2021-07-06 02:41
krbtgti have a voodoo! i got it for 20$ which is a steal compared to the insane prices you see on like ebay#asahi-gpu · 2021-07-06 02:40
chadmedeven with hindsight, what engineers and developers managed to achieve in such a no mans land with so many hard limitations is still extremely impressive to me. on some levels, even more impressive than a lot of software and hardware development happening even now