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

[Matt Arsenault via llvm-commits]

================
@@ -1223,6 +1223,62 @@ define float @v_copysign_f32_0_f16(half %sign) {
   ret float %result
 }
 
+define float @v_copysign_f32_0_fneg(float %sign) {
----------------
arsenm wrote:

Test SALU variants of these patterns 

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