[Openmp-commits] [openmp] [OpenMP][AMDGPU] Adapt dynamic callstack sizes to HIP behavior (PR #74080)

[Michael Halkenhäuser via Openmp-commits]

mhalk wrote:

> > @jayfoad @b-sumner any comments on moving the duplicated code above to a header in llvm/lib/Target/AMDGPU?
> 
> No strong opinions from me. Stuff like that that needs to be shared between different parts of the backend (e.g. between codegen and the assembler/disassembler) often ends up in `lib/Target/AMDGPU/Utils/`, so maybe you could put this functionality in e.g. `lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h`?

Thanks for the input!
This looks like a reasonable way to resolve this situation -- I will start to investigate.

https://github.com/llvm/llvm-project/pull/74080