#asahi-gpu · 2024-10-28 14:13
alyssaover on Linux we use the uapi structs for the same effect#asahi-gpu · 2024-10-28 14:13
alyssaoliverb: the content within the "cmdbuf" is kernel defined, it's a "kernel command buffer" in macOS land#asahi-gpu · 2024-10-28 13:08
alyssa(also fair warning - I don't know if I'd merge the Mesa patches, they were a huge maintenance burden... virglrenderer patches might be easier to persuade me on though for the "Linux guest + macOS host" use case)#asahi-gpu · 2024-10-28 13:07
alyssa:)#asahi-gpu · 2024-10-28 13:07
alyssabut as a way to prove the haters wrong? go get 'em#asahi-gpu · 2024-10-28 13:07
alyssa(that is significantly less stable than Asahi Linux... where upgrading the dual booted macOS is generally safe for the Linux)#asahi-gpu · 2024-10-28 13:06
alyssaAs I'm sure you know, none of this is viable for shipping to end-users because macOS updates will break it#asahi-gpu · 2024-10-28 13:06
alyssahttps://youtu.be/pDsksRBLXPk?t=22621#asahi-gpu · 2024-10-28 13:05
alyssahttps://indico.freedesktop.org/event/6/contributions/296/attachments/227/307/2024-10-10%20XDC%202024%20-%20A%20Little%20Windows%20with%20your%20Mesa.pdf#asahi-gpu · 2024-10-28 13:04
alyssathis is what Faith did for Windows, effectively#asahi-gpu · 2024-10-28 13:04
alyssa(and if that works, dropping the VM layer would be the obvious extension)#asahi-gpu · 2024-10-28 13:04
alyssathat lets you hit all the interesting problems of UAPI translation, without hitting any of the really annoying problems of WSI or fixing Mesa macOS build breakages#asahi-gpu · 2024-10-28 13:03
alyssain Linux#asahi-gpu · 2024-10-28 13:03
alyssathat way you don't have to touch Mesa at all yet, you can just run stock asahi/mesa in a UTM guest and then try to get "gl4.6+vk1.3+cl3.0+gles3.2 working inn inux guest on a macOS host"#asahi-gpu · 2024-10-28 13:02
alyssa4. native context is by far the spiciest question. I think that depends how you implement things. Honestly, that might be the *easiest* place to start though - trying to hack up virglrenderer to translate Linux virtualized UAPI into macOS UAPI#asahi-gpu · 2024-10-28 13:01
alyssaIronically Vulkan would be easier since the MoltenVK and lavapipe interfaces already exist.#asahi-gpu · 2024-10-28 13:01
alyssa3. Effectively. There'd still be a pile of WSI code needed to get OpenGL working in Cocoa windows. For initial driver bring-up, I was just using XQuartz which is... not really what people want from their macOS graphics drivers!#asahi-gpu · 2024-10-28 13:00
alyssa2. More or less, yeah. Should be possible. Timestamps are the one place that the uapi already/will diverge but you don't have to worry about that right now. I also assume sparse will be different but that only matters for DX12, also not a right now problem. (That's all the VM_BIND calls. I don't actually know if macOS has an equivalent. It might?)#asahi-gpu · 2024-10-28 12:58
alyssa1. I don't know. I would assume it's similar, but it's entirely posible added a field in g13x or something like that. Apple doesn't guarantee stability on this interface.#asahi-gpu · 2024-10-28 12:57
alyssaoliverb: great questiosn =D#asahi-gpu · 2024-09-03 10:26
alyssasascha willems terraintessellation going at 120fps on m1 at 4k#asahi-gpu · 2024-09-03 10:26
alyssait's pretty fast shrug#asahi-gpu · 2024-09-03 10:23
alyssachadmed: also anything .cl is on the gpu :p#asahi-gpu · 2024-09-03 10:17
alyssaMary: no fbfetch wired up in honeykrisp yet so that'd actually be much worse ;P#asahi-gpu · 2024-09-03 10:17
alyssachadmed: yeah, hk definitely lower cpu overhead than gl#asahi-gpu · 2024-09-03 01:58
alyssadunno what to make of that#asahi-gpu · 2024-09-03 01:58
alyssainteresting. dolphin is measurably faster with honeykrisp than asahi gl, despite hk having higher intrinsic gpu load#asahi-gpu · 2024-08-23 11:33
alyssathis is why we conformance test...#asahi-gpu · 2024-08-23 11:33
alyssachadmed: umm. it shouldn't. o_o#asahi-gpu · 2024-08-22 18:55
alyssaand if msaa 8x + rgba32 is actually compressible on m3#asahi-gpu · 2024-08-22 18:54
alyssait's unclear to me if apple actually ships msaa 8x on m3 though.#asahi-gpu · 2024-08-22 18:53
alyssaLol#asahi-gpu · 2024-08-22 18:48
alyssaso something had to be reworked there. guess ail is going to get more complicated. bah.#asahi-gpu · 2024-08-22 18:48
alyssasince it would require 16x8 tiles, but those are smaller than the 16x16 compression tiles#asahi-gpu · 2024-08-22 18:47
alyssawhich would conflict with how image layouts work on m1 and m2#asahi-gpu · 2024-08-22 18:47
alyssaThat would require 128 bytes per pixel#asahi-gpu · 2024-08-22 18:47
alyssa...I wonder how MSAA 8x of RGBA32 images works on M3#asahi-gpu · 2024-08-09 13:43
alyssayeehaw#asahi-gpu · 2024-08-09 02:19
alyssacopies are... trickier#asahi-gpu · 2024-08-09 02:19
alyssakarolherbst: pushed a patch to agx/blockblit fixing idiv#asahi-gpu · 2024-08-08 16:06
alyssayeah i need a break too many things happening right now#asahi-gpu · 2024-08-08 16:02
alyssai was busy reading french wikipedia's page on newton's method :P#asahi-gpu · 2024-08-08 16:02
alyssaor I can just give a patch in a bit#asahi-gpu · 2024-08-08 16:01
alyssaI think I know what's happening but I'd like to confirm#asahi-gpu · 2024-08-08 16:01
alyssawhich test case is that?#asahi-gpu · 2024-08-08 16:00
alyssaok#asahi-gpu · 2024-08-08 15:49
alyssayeehaw#asahi-gpu · 2024-08-08 15:47
alyssai'll get to it soon#asahi-gpu · 2024-08-08 15:47
alyssathough not right now I have a big todo list this week#asahi-gpu · 2024-08-08 15:47
alyssai can work with this :+1: