HIP VS rocm-arch

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.

I never could get rocm/hip to work on mint after going round and round several times over months. I finally switched distros and had it running in minutes under an arch based distro using https://github.com/rocm-arch/rocm-arch

I struggled for some time trying to make AMD's Rocm/HIP work in linux so I could use my GPU with blender. Never could get it to work across several distros (mint, ubuntu, fedora). I finally dropped it and just ran GPUless for months. I have since found myself on Manjaro and gave it another try after finding this github. Using this on manjaro I was able to install rocm/hip in NO time and am now using blender with my AMD GPU in linux.

To do this, you need to use the repository: https://github.com/rocm-arch/rocm-arch and checkout to commit 135aa026557d7c11d5f446d6c01521535e9e5747

To make it brief. Trying to install rocm-opencl-runtime and rocm-hip-runtime from rocm-arch on github.

rocm-arch README

Then why doesn't it work unless I have this installed?

Yeah, I had to install rocm using this method https://github.com/rocm-arch/rocm-arch. Afaik, it doesn't replace mesa, but provides OpenCL and HIP.