[llvm] X86: make VBMI2 funnel shifts use VSHLD/VSHRD for const splats (PR #169401)

[Phoebe Wang via llvm-commits]

================
@@ -57622,6 +57626,49 @@ static SDValue combineFP_TO_xINT_SAT(SDNode *N, SelectionDAG &DAG,
   return SDValue();
 }
 
+// Combiner: turn uniform-constant splat funnel shifts into VSHLD/VSHRD
----------------
phoebewang wrote:

Why do we need the combine? Can we use patten match instead?

https://github.com/llvm/llvm-project/pull/169401