[PATCH] D80804: [AMDGPU] Introduce Clang builtins to be mapped to AMDGCN atomic inc/dec intrinsics

[Matt Arsenault via Phabricator via cfe-commits]

arsenm added inline comments.


================
Comment at: clang/include/clang/Basic/BuiltinsAMDGPU.def:62-63
 BUILTIN(__builtin_amdgcn_fence, "vUicC*", "n")
+BUILTIN(__builtin_amdgcn_atomic_inc, "iiD*iUicC*", "n")
+BUILTIN(__builtin_amdgcn_atomic_dec, "iiD*iUicC*", "n")
 
----------------
My main concern is we should probably have both 32 and 64-bit variants


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D80804/new/

https://reviews.llvm.org/D80804