[PATCH] D84517: [AMDGPU] Lower frem f16

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/test/CodeGen/AMDGPU/frem.ll:6
 
+define amdgpu_kernel void @frem_f16(half addrspace(1)* %out, half addrspace(1)* %in1,
+; SI-LABEL: frem_f16:
----------------
Can you also add a function with f32 denormals flushed? In that case the fdiv expansion could theoretically skip the mode switch


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D84517/new/

https://reviews.llvm.org/D84517