[llvm] [AMDGPU] Add pattern to select scalar ops for fshr with uniform operands (PR #165295)

[Jay Foad via llvm-commits]

================
@@ -2763,6 +2763,21 @@ def : GCNPat<(fshr i32:$src0, i32:$src1, i32:$src2),
                                 /* src2_modifiers */ 0,
                                 $src2, /* clamp */ 0, /* op_sel */ 0)
 >;
+
+// The commented out code has been left intentionally to aid the review process, if needed.
+// Will delete before landing.
----------------
jayfoad wrote:

Please just remove it. We can always look at older commits in this PR if we want to see it.

https://github.com/llvm/llvm-project/pull/165295