←back to thread

OpenGL 3.1 on Asahi Linux

(asahilinux.org)
512 points simjue | 2 comments | 06 Jun 23 13:46 UTC | HN request time: 0.496s | source
Show context
kytazo ◴[06 Jun 23 14:18 UTC] No.36213178[source]▶
Its been more than a year I'm running asahi on my macbook air and I can't stress how grateful I feel for enjoying such wonderful freedom.

I don't feel like ever going back to x86 to be honest, at this point there is nothing lacking or unable to run and when the neural engine drivers come online now that the GPU is starting to mature people will be able to juice out every last bit of computation this machine is capable of.

For the record, I've switched to the edge branch a couple of months ago and honestly I noticed no actual difference in my day-to-day tasks which is really telling about how powerful even the M1 is when it can handle software rendering in such an effortless manner coupled with anything else running.

Really thank god for asahi being a thing.

replies(8): >>36213250 #>>36213626 #>>36213905 #>>36214314 #>>36214545 #>>36215750 #>>36217933 #>>36218411 #
1. aseipp ◴[06 Jun 23 15:03 UTC] No.36213905[source]▶
The compute accelerator story on mainstream, non-patched Linux, with upstream software isn't that good at the moment. You're going to be waiting a while before you can do fun stuff like organize layers across the Neural Engine and GPU for ML models, something CoreML can do today. Compute using graphics APIs exists, but it isn't really the same and loses out on many features people practically want and are used to, and it moves forward much more quickly than graphics APIs e.g. Nvidia just released Heterogeneous Memory Management as stable in the open source GPU driver for x86. The Linux accelerator ecosystem in practice is just held together by Nvidia's effort, honestly.

We really need something like Mesa, but for compute accelerator APIs. I'm really hoping that IREE helps smooth out parts of the software stack and can fill in part of this, but the pieces aren't all put in place yet. You'll need the GPU for a substantial amount of accelerator work regardless of Neural Engine support.

I disagree that there is nothing lacking on these machines with Asahi, I still run into small nits all the time (from 16k page sizes biting back to software missing features). But my M2 Air is 100%, no-questions-asked usable as a daily driver and on-the-go hacking machine, it is fast as hell and quiet, it has nested virtualization and is the only modern ARM machine on the market, and I love it for that.

replies(1): >>36219456 #
2. themulticaster ◴[06 Jun 23 21:41 UTC] No.36219456[source]▶
Is the Neural Engine/CoreML used in "normal" desktop apps on macOS, or is it limited to specialized ML centered apps? In other words, where should I expect performance improvements if there was a hypothetical Mesa for compute accelerators? Spontaneously, I can only think of image editors like Photoshop offering AI-based tools.