[Mlir-commits] [mlir] [mlir][gpu][RFC] Add a source language enum attribute to gpu.func and gpu.launch ops (PR #133163)

[Guray Ozen]

grypp wrote:

> This is the current situation in one compiler. For upstream support the OpenACC offloading is likely to use the offload runtime (tgt API) while the CUDA Fortran kernel will be using a dedicated API.

Right. If OpenACC is enabled, you can run the tgt api lowering pass. If CUDA Fortran is enabled, you can run another pass. For these cases, you don't need this language info on the op. When both are enabled, do you want to lower `gpu.launch` differently depending on the language? 

https://github.com/llvm/llvm-project/pull/133163