[llvm] [AMDGPU] Improve codegen for copysign(x, fneg(y)) (PR #207178)

[Matt Arsenault via llvm-commits]

================
@@ -2938,6 +2951,11 @@ def : AMDGPUPat <
   (S_AND_B32 (S_MOV_B32 (i32 0x80000000)), $src1)
 >;
 
+def : AMDGPUPat <
+  (fcopysign (f32 fpimm_zero), (fneg f32:$src1)),
----------------
arsenm wrote:

Maybe should be marking all of these with DivergentBinOp. We probably don't want to be forcing the VALU op for SALU 

https://github.com/llvm/llvm-project/pull/207178