[llvm-branch-commits] [clang] [AMDGPU] Add builtins for wave reduction intrinsics (PR #150170)

[Matt Arsenault via llvm-branch-commits]

================
@@ -351,6 +351,31 @@ BUILTIN(__builtin_amdgcn_endpgm, "v", "nr")
 BUILTIN(__builtin_amdgcn_get_fpenv, "WUi", "n")
 BUILTIN(__builtin_amdgcn_set_fpenv, "vWUi", "n")
 
+//===----------------------------------------------------------------------===//
+
+// Wave Reduction builtins.
+
+//===----------------------------------------------------------------------===//
+
+BUILTIN(__builtin_amdgcn_wave_reduce_add_i32, "iii", "nc")
----------------
arsenm wrote:

Should these be "u32" instead of "i32"? Also we should probably start using whatever magic character sequence gives int32_t/uint32_t instead of i/u 

https://github.com/llvm/llvm-project/pull/150170