kompute VS HIP

The two I know of are IREE and Kompute[1]. I'm not sure how much momentum the latter has, I don't see it referenced much. There's also a growing body of work that uses Vulkan indirectly through WebGPU. This is currently lagging in performance due to lack of subgroups and cooperative matrix mult, but I see that gap closing. There I think wonnx[2] has the most momentum, but I am aware of other efforts.

[1]: https://kompute.cc/

[2]: https://github.com/webonnx/wonnx

Great to see the positive momentum of this framework! Best wishes and upvotes from the Vulkan Kompute team :)

To a first approximation, Kompute[1] is that. It doesn't seem to be catching on, I'm seeing more buzz around WebGPU solutions, including wonnx[2] and more hand-rolled approaches, and IREE[3], the latter of which has a Vulkan back-end.

[1]: https://kompute.cc/

[2]: https://github.com/webonnx/wonnx

[3]: https://github.com/openxla/iree

I look in an example and see similar instructions, stating that the build should be quite simple. But again, it doesn't work. It generates a bunch of folders with Visual Studio stuff, but no executables, no libraries, or anything like that.

I can't figure out how, and there are no tutorials. According to https://kompute.cc/overview/build-system.html I should simply run "cmake -Bbuild". But this doesn't output what I need, and when I look in the Makefile I get the sense that this is more an example Makefile... but then that contradicts the above tutorial.

Kompute is in my opinion good example to take inspiration for abstractions.

Currently, there's a few Vulkan compute frameworks floating around (like Kompute). I would work with those. Kompute simplifies a lot of the biolerplate and seems like you could benefit from using it.

Try Kompute, a project from the Linux foundation. It is quite simple to use, and does not require deep knowledge of graphics API. It’s a bit painful to setup, but it kinda works well (and I have a project going on on it)

This so much.

This makes alot of sense considering the Linux Foundation is also in charge of Kompute which is likely to be the basis of vendor agnostic GPGPU, and thus the basis of vendor agnostic GPU-based machine learning.

Is it perhaps because they want people to use HIP?

> HIP is very thin and has little or no performance impact over coding directly in CUDA mode.

> The HIPIFY tools automatically convert source from CUDA to HIP.

1. https://github.com/ROCm/HIP

AMD has released HIP and a tool called HIPIFY which kind of behaves like this but at the source level¹. Rather than try and just translate CUDA to work on AMD compute they are more focused on higher level tooling.

Currently they seem to have a particular focus on AI frameworks and tools like PyTorch/Tensorflow/ONNX. They have sponsored and helped with a lot of PyTorch development for example, so PyTorch support for AMD is much better than it was this time last year².

¹(https://github.com/ROCm/HIP)

²(https://pytorch.org/blog/experience-power-pytorch-2.0/)

> what would be the point for someone to add ROCm support to various pieces of software which currently require CUDA

It isn't just old cards though, CUDA is a point of centralization on a single provider during a time when access to that providers higher end cards isn't even available and that is causing people to look elsewhere.

ROCm supports CUDA through the included HIP projects...

https://github.com/ROCm/HIP

https://github.com/ROCm/HIPCC

https://github.com/ROCm/HIPIFY

The later will regex replace your CUDA methods with HIP methods. If it is as easy as running hipify on your codebase (or just coding to HIP apis), it certainly makes sense to do so.

AMD's equivalent is HIP [1], for sufficiently flexible definitions of "equivalent". I can't speak to how complete/correct/performant it is (I'm just a guy running tutorial/toy-level ML stuff on an RDNA1 card), but part of AMD's problem is that it might not practically matter how well they do this because the broader ecosystem support specifically for the CUDA stack is so entrenched.

[1] https://github.com/ROCm-Developer-Tools/HIP

From what I know, modern GPUs are still programmed with C++ exclusively. See CUDA [0] for Nvidia and ROCm [1] for AMD.

Why is this? Why Rust is not loved there?

[0] https://docs.nvidia.com/cuda/

[1] https://github.com/ROCm-Developer-Tools/HIP

Go ahead and try to ship ROCm code that works on multiple consumer graphics cards on Linux, MacOS, and Windows. As an example of how much AMD cares about it, the installation notes linked to in the readme returns a 404.

The big whoop for ROCm is that AMD invested a considerable amount of engineering time and talent into a tool they call hip. Basically, it's an analysis tool that does its best to port proprietary Nvidia CUDA-style code - which due to various smelly reasons rules the roost - to code that can happily run on AMD graphics cards, and presumably others. Intel has a similar thing going with OneAPI. They've done this whilst working on porting a lot of their code base to the linux AMGPU open source kernel driver, as well.