[llvm] [AMDGPU] Constant folding for wave-reduce intrinsics (PR #212755)

[Madhur Kumar via llvm-commits]

================
@@ -149,6 +149,242 @@ define amdgpu_cs void @atomic_sub(<4 x i32> inreg %arg)  {
   ret void
 }
 
+define amdgpu_cs void @atomic_add_constant_0(<4 x i32> inreg %arg)  {
----------------
MadhurKumar004 wrote:

Maybe the test cases could be in their own commit.

https://github.com/llvm/llvm-project/pull/212755