#asahi-gpu · 2023-04-08 11:45
jannauunfortunately still no useable debug information in mesa for the following segfault but that's probably just misguided error handling#asahi-gpu · 2023-04-08 11:43
jannauout of memory appears unlikely on a 32GB M1 Max system#asahi-gpu · 2023-04-08 11:41
jannaustill falling into the 2-3 days window. not sure if it is related to the buffer leak. last time I looked before the GEM_BIND error it was around 105 buffers with ~800MB#asahi-gpu · 2023-04-08 11:39
jannauI'm still seeing GEM_BIND errors after flipping the allocations over to the GPU#asahi-gpu · 2023-04-08 11:37
jannauwould it make sense to have a different debug flag for submit? that is at least 2 orders of magnitude noisier than other ioctls#asahi-gpu · 2023-04-08 02:38
i509vcbper metal agx seems to support mesh shading but I assume that is quite low priority to reverse engineer at the moment#asahi-gpu · 2023-04-08 01:39
alyssareplaying on t8103 you'll see black rectangles around the gift boxes in the background#asahi-gpu · 2023-04-08 01:39
alyssasent you an apitrace of it#asahi-gpu · 2023-04-08 01:38
alyssathink I'm going to stop looking at this for now, maybe you can take a try on Wednesday#asahi-gpu · 2023-04-08 01:37
alyssalevel 0 seems to be (so not a mipmapping issue)#asahi-gpu · 2023-04-08 01:33
alyssaskipping all transfers doesn't make a difference#asahi-gpu · 2023-04-08 01:32
alyssazeroing the BO doesn't help#asahi-gpu · 2023-04-08 01:28
alyssaspecifically Z32F textures#asahi-gpu · 2023-04-08 01:23
alyssaunclear what's up.#asahi-gpu · 2023-04-08 01:23
alyssalina: I'm seeing some artefacts with supertuxkart's advanced renderer when depth/stencil compression is enabled#asahi-gpu · 2023-04-07 20:22
alyssadEQP-GLES3.functional.texture.specification.texsubimage3d_pbo.rgba4_2d_array i mean#asahi-gpu · 2023-04-07 20:22
alyssapossibly preexisting bug#asahi-gpu · 2023-04-07 20:22
alyssabut I see it using linear 2D arrays so I'm ?? about the whole thing#asahi-gpu · 2023-04-07 20:22
alyssasomething from enabling RGBA4#asahi-gpu · 2023-04-07 20:13
alyssaone regression in there..#asahi-gpu · 2023-04-07 20:11
alyssaPass: 44150, Fail: 9, Crash: 8, Warn: 38, Skip: 184, Duration: 6:24, Remaining: 0#asahi-gpu · 2023-04-07 20:11
alyssadeqp-gles3 is coming along#asahi-gpu · 2023-04-07 17:54
jannauso possibly a kde bug#asahi-gpu · 2023-04-07 17:54
jannaulina: reproducer for the buffer leak: 2 konsole windows tiled left and right + kate maximized. Switch between the windows via activation on the panel. I suspect it might be related to the window preview shown after a while#asahi-gpu · 2023-04-07 00:27
alyssaPretty straightforward#asahi-gpu · 2023-04-07 00:27
alyssalina: I've rebased asahi/mesa on upstream and opened an upstream MR accordingly#asahi-gpu · 2023-04-04 20:51
TriVoxeloh, read the desc. just ignore this#asahi-gpu · 2023-04-04 20:50
TriVoxelHey, sorry if this has already been answered. I was re-installing Fedora using the Asahi Fedora Remix instead of the SW Alternative listed in the docs. I have no GPU acceleration, just software rendering. How can I fix this?#asahi-gpu · 2023-04-04 18:53
TriVoxelI am currently using OpenGL 2.1 or something.#asahi-gpu · 2023-04-04 18:52
TriVoxelHello. Amazing to see all the awesome work being done here. I am using the Fedora alternative distro, and I'm wondering how I can test the new GPU drivers. Thanks!#asahi-gpu · 2023-03-30 20:46
alyssacargo, chill#asahi-gpu · 2023-03-30 03:10
alyssameh. it's a silly case. Not super worried tbh#asahi-gpu · 2023-03-30 03:10
alyssaI know power-of-two alignment was needed to avoid a bunch of pathologies, forget why I rounded up size too though..#asahi-gpu · 2023-03-30 03:09
alyssais not recalling why she did that#asahi-gpu · 2023-03-30 03:08
alyssaoh.. because I'm forcing my vectors to be power of two size... right#asahi-gpu · 2023-03-30 03:07
alyssa(I think I see what's happening but still need to figure out why)#asahi-gpu · 2023-03-30 03:03
alyssa(regression on agx/next from live range splitting)#asahi-gpu · 2023-03-30 03:03
alyssatodo: revisit shaders/android/angle/manhattan_10/70.shader_test, it's generating moves that it shouldn't be#asahi-gpu · 2023-03-30 02:24
alyssanot the weirdest thing I guess#asahi-gpu · 2023-03-30 02:24
alyssamaybe we're supposed to force translucent punchthrough for multisampling and punchthrough is a single non-msaa thing#asahi-gpu · 2023-03-30 02:23
alyssatranslucent passthrough is fine, though?#asahi-gpu · 2023-03-30 02:23
alyssawat#asahi-gpu · 2023-03-30 02:23
alyssabut punch through is doing fine#asahi-gpu · 2023-03-30 02:23
alyssaok, it's the writes_sample_mask that's causing bad#asahi-gpu · 2023-03-30 02:21
alyssaok, it's the writes_sample_mask that's causing bad#asahi-gpu · 2023-03-30 02:15
alyssapowered by AI [citation needed - failed verification]#asahi-gpu · 2023-03-30 02:15
alyssabut if the MR is non-Draft status, there's a bot that'll do it for you#asahi-gpu · 2023-03-30 02:15
alyssayeah you need commit rights to add labels#asahi-gpu · 2023-03-30 02:14
i509vcbI submitted the MR but I can't seem to add labels via the UI?#asahi-gpu · 2023-03-30 01:51
alyssaHm still faulting.