[PATCH] D48573: [AMDGPU] Add llvm.amdgcn.fmad.ftz intrinsic

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

Actually according to the selection code, f16 mad does not support denormals the same, so the intrinsic should work with f16 if that is correct


Repository:
  rL LLVM

https://reviews.llvm.org/D48573