hip

> There's also HIP, which converts CUDA to something that can run on AMD GPUs. Kinda making your point, HIP is actually an (open source) one-to-one replacement for the CUDA API. While there are tools (https://github.com/ROCm-Developer-Tools/HIP/tree/master/hipi... and https://github.com/ROCm-Developer-Tools/HIP/blob/master/bin/...) that convert CUDA to HIP, these are meant to be run once and perhaps touched up manually. HIP source can then be compiled (no human interaction) for AMD (ROCm) devices and NVIDIA (CUDA) devices. The CUDA brand refers to both the API and the device. AMD certainly could have done a better job, but wanted to distinguish these, presumably in hopes that existing users of the CUDA API would consider porting their code to HIP even when targeting CUDA devices.

I'd say it's likely because NVIDA's CUDA compiler produces code for NVIDIA GPUs, and AMD would have a lot of (questionably legal) reverse engineering ahead of them in order to support the same code on their own GPUs. If you're asking why AMD doesn't make a compiler for CUDA source code that targets their own GPUs--that's basically what ROCm currently does. They're pushing their CUDA alternative, called "HIP", which is essentially just CUDA code with a find-and-replace of "cuda" with "hip". (And other similarly minor changes.) They have an open source "hipify" program that does this automatically (https://github.com/ROCm-Developer-Tools/HIP/tree/master/hipi...). So, basically, AMD GPUs are already sort of CUDA compatible: just run your CUDA code through hipify, then compile it using the HIP compiler, and run it on a ROCm-supported syste

>... so library-writers will have to basically start from scratch if this AMD cluster is confirmed. I doubt library writers will care beyond anyone involved in the project. That said, it's possible to port CUDA code already with a bit of work, so not quite from scratch in any case: https://github.com/ROCm-Developer-Tools/HIP

HIP (https://rocm-documentation.readthedocs.io/en/latest/Programm...) is basically 1-1 with CUDA, and can be almost automatically generated from CUDA (https://github.com/ROCm-Developer-Tools/HIP/tree/master/hipi...). The datasheet doesn't mention OpenCL so evidently it isn't a priority any more.