[llvm] AMDGPU: Remove flat/global fmin/fmax intrinsics (PR #105642)

[Gheorghe-Teodor Bercea via llvm-commits]

doru1004 wrote:

> There won't be a new function, but @yxsamliu is working on source level controls for these fields for the atomic builtins

I need to be able to call this from Clang code gen. The reason I insist on this is because I don't think the selection of attributes to pass to the current atomicRMW function is clear enough i.e. if I was relying on the intrinsic up until now, it is not clear what set of additional options to pass to atomicRMW() to make it behave like the intrinsic. So I would like that knowledge to live somewhere in the compiler i.e. in the shape of a function atomicRMWFastFAdd or something like that which calls atomicRMW under the hood BUT ALSO provides the set of additional options that make it behave like the intrinsic.

https://github.com/llvm/llvm-project/pull/105642