[PATCH] D157388: [AMDGPU] Support FMin/FMax in AMDGPUAtomicOptimizer.

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

Also should have some end to end codegen tests, just this won't catch the interaction between the atomic expand and atomic optimizer



================
Comment at: llvm/test/CodeGen/AMDGPU/global_atomics_iterative_scan_fp.ll:543
+  %divValue = bitcast i32 %id.x to float
+  %result = atomicrmw fmax ptr addrspace(1) %ptr, float %divValue seq_cst
+  ret void
----------------
Should also test with different scopes, at least default system and agent


================
Comment at: llvm/test/CodeGen/AMDGPU/global_atomics_iterative_scan_fp.ll:547
+
 attributes #0 = {"target-cpu"="gfx906"}
----------------
Should take this from the command line


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D157388/new/

https://reviews.llvm.org/D157388