#asahi-gpu · 2024-01-06 10:56
knedlikOkay, now it should be complete#asahi-gpu · 2024-01-06 10:52
knedlikOh wait... forgot to include failed tests, one sec#asahi-gpu · 2024-01-06 10:50
knedlikWhich of these would be a good thing to get started with?#asahi-gpu · 2024-01-06 10:50
knedlikSo I have successfully run the tests, and made a deduped text file of the non-passing ones: https://gist.github.com/KnedlikMCPE/4c4fffa220a668d0bb48851c77bb2062#asahi-gpu · 2024-01-06 01:23
alyssaread a bit more carefully#asahi-gpu · 2024-01-06 00:28
nullspacei believe this isn't true for m2 and m3? hardware atomics are supported, right?#asahi-gpu · 2024-01-06 00:27
nullspace> The M1 lacks hardware instructions for image atomics, even though it has non-image atomics and non-atomic images#asahi-gpu · 2024-01-06 00:21
nullspacehi#asahi-gpu · 2024-01-05 19:55
i509vcbOk#asahi-gpu · 2024-01-05 19:19
knedlikthere I have cloned the mesa and CTS repos#asahi-gpu · 2024-01-05 19:19
knedlikI already closed the terminal, but it was in the root dir of my external SSD, so /run/media/knedlik/LinuxEXT/#asahi-gpu · 2024-01-05 19:17
i509vcbwhat does pwd say?#asahi-gpu · 2024-01-05 19:17
i509vcbAnd you did ./path/to/caselist#asahi-gpu · 2024-01-05 19:14
knedlikI specified the absolute path#asahi-gpu · 2024-01-05 19:13
knedlikIt isn't, should it be?#asahi-gpu · 2024-01-05 19:13
i509vcbCheck --help#asahi-gpu · 2024-01-05 19:13
i509vcbAlthough last I recall caselist files aren't just specified that way?#asahi-gpu · 2024-01-05 19:13
i509vcbknedlik: your current directory is the same location as dEQP-vk?#asahi-gpu · 2024-01-05 18:57
alyssaagx_remap_varyings_vs control that#asahi-gpu · 2024-01-05 18:56
alyssaassume it's after clip distance I guess#asahi-gpu · 2024-01-05 18:56
alyssadoesn't remember the ordering off hand#asahi-gpu · 2024-01-05 18:55
alyssabut viewport tgt + render tgt (== layer) work like clip distance as described there#asahi-gpu · 2024-01-05 18:55
alyssaalso would appreciate a doc patch to https://docs.mesa3d.org/drivers/asahi.html#vertex-shader with what you've learned :)#asahi-gpu · 2024-01-05 18:54
alyssasee the GL driver logic#asahi-gpu · 2024-01-05 18:53
alyssa(there might be other bugs, but that's the only obvious one)#asahi-gpu · 2024-01-05 18:52
alyssa`Output select` is missing the viewport target + render target bits to consume the fixed-function layer#asahi-gpu · 2024-01-05 18:51
alyssathe VS is writing but it's not getting routed #asahi-gpu · 2024-01-05 18:51
alyssaer, rather#asahi-gpu · 2024-01-05 18:51
alyssai509vcb: the FS is reading the LAYER, but the VS isn't writing it#asahi-gpu · 2024-01-05 18:46
knedlikSo considering this is pretty new territory to me, I'm gonna need some help - when I try to run deqp-vk with the params listed in the README, I get 'FATAL ERROR: Failed to initialize dEQP: Failed to open file: './vk-default/api.txt' at tcuResource.cpp:53'...#asahi-gpu · 2024-01-05 18:39
i509vcbAlthough I've found if you start too many jobs with deqp-runner your gpu hangs and the kernel can't recover it without a hard reboot#asahi-gpu · 2024-01-05 18:38
i509vcbIt could also be useful to grab deqp-runner#asahi-gpu · 2024-01-05 18:38
knedlikAlright, thanks#asahi-gpu · 2024-01-05 18:38
i509vcbit should be under external/vulkancts/modules/vulkan#asahi-gpu · 2024-01-05 18:38
i509vcbdEQP-VK#asahi-gpu · 2024-01-05 18:36
knedlikSo I have successfully compiled the VK Mesa and the test suite... now, could someone help me with which binary I should run for the CTS?#asahi-gpu · 2024-01-05 18:18
i509vcbI think .64 is the one where we start seeing uses of layer reads: https://usercontent.irccloud-cdn.com/file/EgxJlqzk/3d_to_3d.a2r10g10b10_sint_pack32.a2r10g10b10_sint_pack32.zip#asahi-gpu · 2024-01-05 18:16
i509vcbhttps://usercontent.irccloud-cdn.com/file/knju3cGX/image.png#asahi-gpu · 2024-01-05 18:16
i509vcbalthough the output in the viewer shows something odd#asahi-gpu · 2024-01-05 18:16
i509vcbalyssa: yes, it is an int, did the change and still no dice#asahi-gpu · 2024-01-05 17:32
knedlikWill try to compile it now, I'll get back if I need any help#asahi-gpu · 2024-01-05 17:31
alyssaknedlik: yep, branch agxv/main#asahi-gpu · 2024-01-05 17:20
knedlikIs it the asahi/mesa repo under freedesktop?#asahi-gpu · 2024-01-05 17:14
knedlikAlright, just to make sure, where can I find the current Vulkan implementation? Last time I looked in the upstream Mesa repo, the Asahi dir didn't contain a Vulkan dir#asahi-gpu · 2024-01-05 17:12
i509vcbAreas to avoid: the entire vkCmdCopy family of commands, I am currently working on those#asahi-gpu · 2024-01-05 17:11
i509vcbOr you could try to implement extensions, although there is a lot of core API that these need to make that truly work#asahi-gpu · 2024-01-05 17:11
i509vcbknedlik: build the vulkan cts, find a group of failing tests and try to fix those#asahi-gpu · 2024-01-05 13:56
knedlikHey there! Is there any way I could contribute to the Vulkan drivers becoming a thing? I'd say I'm a pretty good programmer, although I don't have much knowledge in low-level stuff...#asahi-gpu · 2024-01-05 11:06
alyssanir_load_input(b, nir_imm_int(b, 0), .component = 0, .base = /* you get the idea */)#asahi-gpu · 2024-01-05 11:05
alyssanir_intrinsic_instr_create is the "old" way, nowadays we prefer the much more concise helpers