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

[Michael Halkenhäuser via Openmp-commits]

mhalk wrote:

> You can encode arbitrary values in a subtarget feature. The current main example would be FeatureLocalMemorySize*. Not sure if we directly expose those in the TargetParser

Great, thank you @arsenm! That helped a lot.

Now I've added the corresponding three `SubtargetFeatures` and added the getters similar to `(Feature)LocalMemorySize`.
TBH it is still not entirely clear to me how to access the newly available `(get)MaxWaveScratchSize ` via the `TargetParser` or `AMDGPUBaseInfo`.

When using the `TargetParser` I was able to construct the FeatureMap corresponding to my present device(s) but that did not contain the info (and from what it contained, MaxWaveScratchSize does not really seem to 'fit' in that group of device capabilities).

Do you have another pointer for me on how to proceed?

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