#asahi-gpu · 2022-10-27 02:12
alyssaHm.#asahi-gpu · 2022-10-27 02:10
alyssalina: sorry what?#asahi-gpu · 2022-10-27 02:10
alyssa02:10 < lina> Keep in mind that the depth buffer, unless the app really needs it, is only written to on TVB overflows#asahi-gpu · 2022-10-27 01:34
alyssaPresumably#asahi-gpu · 2022-10-27 01:22
alyssathe PBE just writes images#asahi-gpu · 2022-10-27 01:22
alyssabut the ZLS knows what depth buffers and stencil buffers are,#asahi-gpu · 2022-10-27 01:22
alyssathe closest equivalent is the PBE which handles image writes in general#asahi-gpu · 2022-10-27 01:22
alyssathere is not dedicated colour buffer hardware#asahi-gpu · 2022-10-27 01:22
alyssaNote that there is dedicated depth/stencil hardware (the ZLS)#asahi-gpu · 2022-10-27 01:10
alyssaand that's entirely possible, there are lots of funny limitations here#asahi-gpu · 2022-10-27 01:10
alyssas/framebuffer/depth stencil/, lina's patch doesn't touch the colour attachments#asahi-gpu · 2022-10-27 00:05
alyssaIs there any Linux userspace that can use YUV overlays these days? :(#asahi-gpu · 2022-10-27 00:04
alyssaAh lol nice#asahi-gpu · 2022-10-26 23:56
alyssaHow'd you discover the flag?#asahi-gpu · 2022-10-26 23:56
alyssajannau: Woof#asahi-gpu · 2022-10-26 22:16
alyssaYes, I think so#asahi-gpu · 2022-10-26 21:59
alyssaIt's probably not a *big* deal to support properly in mesa but that patch is not the way to do it#asahi-gpu · 2022-10-26 21:57
alyssaMetal really doesn't like rendering to linear#asahi-gpu · 2022-10-26 21:57
alyssas/linear/uncompressed/, IIRC it's stil twiddled#asahi-gpu · 2022-10-26 16:45
alyssafinding the next allocation in the same BO and using that as an upper bound#asahi-gpu · 2022-10-26 16:45
alyssano, adjacency as well#asahi-gpu · 2022-10-26 16:45
alyssaor maybe not#asahi-gpu · 2022-10-26 16:45
alyssalina: guess#asahi-gpu · 2022-10-26 15:24
alyssaOK#asahi-gpu · 2022-10-26 15:24
alyssao#asahi-gpu · 2022-10-26 15:20
alyssa(the diff from t8103 mesa I mean)#asahi-gpu · 2022-10-26 15:20
alyssaI'd love to see the Mesa patch if you've pushed#asahi-gpu · 2022-10-26 15:20
alyssaVery curious#asahi-gpu · 2022-10-26 15:09
alyssait does visually look like hier-z but I'm not convinced it actually is#asahi-gpu · 2022-10-26 15:09
alyssalina: as I texted you, the "Z acceleration" buffer is probably just from depth compression, which Apple's driver is aggressive about enabling and took me a lot of time to figure out how to disable#asahi-gpu · 2022-10-26 02:21
alyssaone answer is "I did a better job at register allocation for AGX than Mali"#asahi-gpu · 2022-10-26 02:20
alyssawhy would the same program use so many fewer registers on AGX?#asahi-gpu · 2022-10-26 02:20
alyssawhereas AGX is using 50/2 = 25 dword registers#asahi-gpu · 2022-10-26 02:20
alyssait's not explicitly printed for Mali-G57, but "1 thread" means that it's using at least 33 dword registers#asahi-gpu · 2022-10-26 02:20
alyssaSimilarly, register pressure#asahi-gpu · 2022-10-26 02:18
alyssaAGX uses variable length instructions, in this shader instructions are an average of 6.65 bytes#asahi-gpu · 2022-10-26 02:18
alyssaWhy? mali-g57 uses fixed-length instructions, all are 8 bytes, and many are sparse#asahi-gpu · 2022-10-26 02:18
alyssathere are fewer of them (1249 vs 1294), but they're bigger#asahi-gpu · 2022-10-26 02:17
alyssaThis is interesting, a Mali-G57 instruction can actually do *more* than an AGX instruction#asahi-gpu · 2022-10-26 02:17
alyssa(quadwords = 16 bytes)#asahi-gpu · 2022-10-26 02:17
alyssaMali is ~10,000 bytes#asahi-gpu · 2022-10-26 02:17
alyssaAGX is 8600 bytes#asahi-gpu · 2022-10-26 02:16
alyssabut here's an obvious one:#asahi-gpu · 2022-10-26 02:16
alyssa(AGX will get similar stuff in time~)#asahi-gpu · 2022-10-26 02:16
alyssaoff the bat you'll notice that the statistics for Mali are a lot more advanced, because we actually understand the uarch there#asahi-gpu · 2022-10-26 02:16
alyssatop is Apple M1, bottom is Mali-G57#asahi-gpu · 2022-10-26 02:16
alyssashaders/glmark/1-22.shader_test - MESA_SHADER_FRAGMENT shader: 1249 inst, 16.015625 cycles, 16.015625 fma, 3.453125 cvt, 0.000000 sfu, 0.125000 v, 0.000000 t, 0.000000 ls, 632 quadwords, 1 threads, 0 loops, 0:0 spills:fills#asahi-gpu · 2022-10-26 02:15
alyssashaders/glmark/1-22.shader_test - MESA_SHADER_FRAGMENT shader: 1294 inst, 8600 bytes, 50 halfregs, 1 threads, 0 loops, 0:0 spills:fills#asahi-gpu · 2022-10-26 02:15
alyssahere's a chunky shader from glmark2 -bterrain#asahi-gpu · 2022-10-26 02:15
alyssaIt's really interesting comparing statistics for a given compiler shader *across* instruction sets