#
…

2026-02-08

00:01alicela1n has joined #asahi-gpu
00:06XeR has quit [Remote host closed the connection]
00:06XeR has joined #asahi-gpu
00:48c10l has quit [Quit: Bye o/]
00:51alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
00:55pb17 has quit [Ping timeout: 480 seconds]
00:56c10l has joined #asahi-gpu
00:59pb17 has joined #asahi-gpu
01:39alicela1n has joined #asahi-gpu
02:51pb17 has quit [Ping timeout: 480 seconds]
03:07pb17 has joined #asahi-gpu
03:14alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
03:16jnn has joined #asahi-gpu
03:20jn has quit [Ping timeout: 480 seconds]
04:52alicela1n has joined #asahi-gpu
04:56integralpilot has joined #asahi-gpu
04:57alicela1n has quit []
05:05alicela1n has joined #asahi-gpu
05:49alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
06:15pb17 has quit [Ping timeout: 480 seconds]
06:16pb17 has joined #asahi-gpu
08:08pb17 has quit [Ping timeout: 480 seconds]
08:14pb17 has joined #asahi-gpu
08:23mattia013 has quit [Quit: mattia013]
08:24mattia013 has joined #asahi-gpu
09:18HerbY_NL has joined #asahi-gpu
09:30nela has quit [Ping timeout: 480 seconds]
09:48integralpilot has quit [Ping timeout: 480 seconds]
09:48pb17 has quit [Ping timeout: 480 seconds]
09:48integralpilot has joined #asahi-gpu
09:48pb17 has joined #asahi-gpu
10:16HerbY_NL has quit [Quit: Leaving]
10:31jannauCTS runs have the potential to crash the GPU so do not run this with unsaved work in your desktop session
10:37oliverperfect, thank you
10:37oliveri have one question: the way i understood mesa, INTERCHANGE should only be used for surface textures with my change, like with the "window" or the "desktop", everything else should still use GPU_TILED as before. is this covered when running those tests?
10:41jannauthe tests test everything. your mesa changes are responsible to allow only working use cases
10:42noopwafelI mean, the question is, are you going to even hit this case with CTS
10:43jannaufor development it'll probably more comfortable to have a local CTS build so you can run single tests under a debugger
10:46noopwafel(certainly not with vulkan CTS I guess, since this seems gallium-only?)
10:46jannaudue to the placement in the best_modifier array I expect interchange to be used even when it shouldn't be. but yes when CTS passes it'll a good idea to break interchange on purpose to verify that it is exercised
10:47jannauduh. I guess I'll finish the asahi OpenGL CI expectations now
10:54oliverbest_modifier seems to be only used in the context of the "screen". when checking with ASAHI_MESA_DEBUG i see it only used for the surface, all the other textures allocated are either LINEAR or GPU_TILED ... which is fine imho
10:56c10l has quit [Quit: Bye o/]
11:04c10l has joined #asahi-gpu
11:14integralpilot has quit [Ping timeout: 480 seconds]
11:32c10l has quit [Quit: Bye o/]
11:34integralpilot has joined #asahi-gpu
11:38T4ForFun has joined #asahi-gpu
11:41c10l has joined #asahi-gpu
11:45T4ForFun has quit [Remote host closed the connection]
11:47T4ForFun has joined #asahi-gpu
12:00c10l has quit [Quit: Bye o/]
12:07olivertest_vk finished with some 42 failures on mesa 26.0.0-rc3. i have a g13s according to glxinfo. also my changes did not affect vulkan at all. i'll run it again with the changes applied to the vk driver too.
12:09HerbY_NL has joined #asahi-gpu
12:10oliver(on mesa main, not rc)
12:11jannauit finished on g14s without unexpected failures
12:17HerbY_NL has quit [Quit: Leaving]
12:21c10l has joined #asahi-gpu
12:27oliverbut, like i said, the changes do not affect vulkan at all, it is just gl.
12:27oliverthat one leads to more failures, it will crash for yuv formats.
12:27oliver(or all non 32bbp formats with a non one block size)
12:53c10l has quit [Quit: Bye o/]
13:02c10l has joined #asahi-gpu
13:03jannauoliver: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39763 and update lamci.sh. test_gl will then execute the OpenGL CTS
13:04alyssajannau: serves me right for not running dEQP-EGL in years. oops :p
13:06alyssajannau: you can make the toml a lot more compact because `glcts` secretly contains all the deqp-gles tests too. whether that's actually better is unclear. see https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34960
13:07alyssaperso I use one big merged caselist and a single invocation for all of gles3.2+gl4.6 cts (except for WSI)
13:07alyssaMESA_SHADER_CACHE_DIR=/dev/shm EGL_PLATFORM=surfaceless deqp-runner run --deqp ~/GLCTS/glcts --output output --timeout 60 --caselist /dev/shm/caselist.txt --skips /dev/shm/skips.txt --tests-per-group=1000 -- --deqp-surface-type=pbuffer --deqp-gl-config-name=rgba8888d24s8ms0 --deqp-surface-width=400 --deqp-surface-height=300 --deqp-log-images=disable
13:08alyssa--deqp-log-shader-sources=disable
13:08alyssawith manual caselist - cat gles/aosp_mustpass/main/gles2-main.txt gles/aosp_mustpass/main/gles3-main.txt gles/aosp_mustpass/main/gles31-main.txt gles/khronos_mustpass/main/gles2-khr-main.txt gles/khronos_mustpass/main/gles3-khr-main.txt gles/khronos_mustpass/main/gles31-khr-main.txt gl/khronos_mustpass/4.6.1.x/gl46-main.txt gl/khronos_mustpass_single/4.6.1.x/gl46-khr-single.txt >
13:08alyssa/dev/shm/caselist.txt
13:10pb17 has quit [Ping timeout: 480 seconds]
13:14jannaulet's do that. any specific reason for the non-pot surface width/height? CI config in mesa uses 256/256 consistently
13:20c10l has quit [Quit: Bye o/]
13:21integralpilot has quit [Quit: integralpilot]
13:24nela has joined #asahi-gpu
13:29c10l has joined #asahi-gpu
13:39alyssajannau: working around a broken test
13:39alyssalet me see if I can remember which one
13:44alyssajannau: KHR-GL46.texture_query_lod.* tests are broken at small surface sizes
13:44pb17 has joined #asahi-gpu
13:44alyssawhich I guess got fixed in CTS commit 84604d777 ("Skip small viewports in textureQueryLod tests")
13:44alyssaso 256x256 should be ok now again
13:45jannauI saw with the combined list Fails in KHR-GL46.texture_query_lod.sampler{1,2,3}D_test but not with the original suite toml
13:45alyssathose would be the broken tests in question yes ..
13:46alyssathey were added late Mar 2025 and skipped late Oct 2025
13:46alyssaand I changed jobs during that 7 month window which is why my script was out of sync :p
13:50c10l has quit [Quit: Bye o/]
13:52jannaucommented in !34960 that using glcts for all tests at least doesn't slow the tests down on asahi
13:59c10l has joined #asahi-gpu
14:07T4ForFun has quit [Ping timeout: 480 seconds]
14:10c10l has quit [Quit: Bye o/]
14:12T4ForFun has joined #asahi-gpu
14:14yuka has quit [Remote host closed the connection]
14:15yuka has joined #asahi-gpu
14:37c10l has joined #asahi-gpu
15:33pb17 has quit [Ping timeout: 480 seconds]
15:36c10l has quit [Quit: Bye o/]
15:44pb17 has joined #asahi-gpu
15:57c10l has joined #asahi-gpu
16:34alicela1n has joined #asahi-gpu
16:52alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
16:53c10l has quit [Quit: Bye o/]
16:58alicela1n has joined #asahi-gpu
17:17c10l has joined #asahi-gpu
17:26alicela1n has quit [Remote host closed the connection]
17:26alicela1n has joined #asahi-gpu
17:29alicela1n has quit []
17:29alicela1n has joined #asahi-gpu
17:37c10l has quit [Quit: Bye o/]
17:46alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
17:53c10l has joined #asahi-gpu
18:03alicela1n has joined #asahi-gpu
18:34alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
18:51alicela1n has joined #asahi-gpu
19:21oliverI realized I can just tell mesa not reject a modifier for a specific format, did that. now the cts seem to run fine for !39755.
19:21oliverfun fact: rgb with 16 bit per channel is also chunked into 16x16px tiles.
19:23jannauso is interleaved CbCr for NV10 and P010
19:31oliverBut that I'll exclude for now.
20:05HerbY_NL has joined #asahi-gpu
20:07HerbY_NL has quit []
20:27pb17 has quit [Ping timeout: 480 seconds]
20:50T4ForFun has quit [Remote host closed the connection]
20:51pb17 has joined #asahi-gpu
20:56alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
21:05alicela1n has joined #asahi-gpu
23:16alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
23:47rainbyte has joined #asahi-gpu
23:53pb17 has quit [Ping timeout: 480 seconds]
23:54alicela1n has joined #asahi-gpu
23:56pb17 has joined #asahi-gpu
…