[llvm] [X86] Adding lowerings for vector ISD::LRINT and ISD::LLRINT (PR #90065)

[Phoebe Wang via llvm-commits]

================
@@ -51556,6 +51583,22 @@ static SDValue combineFaddFsub(SDNode *N, SelectionDAG &DAG,
   return SDValue();
 }
 
+static SDValue combineLRINT_LLRINT(SDNode *N, SelectionDAG &DAG,
+                                   const X86Subtarget &Subtarget) {
+  EVT VT = N->getValueType(0);
+  SDValue Src = N->getOperand(0);
+  EVT SrcVT = Src.getValueType();
+  SDLoc DL(N);
+
+  if (!Subtarget.hasDQI() || !Subtarget.hasVLX() || VT != MVT::v2i64 ||
+      SrcVT != MVT::v2f32)
+    return SDValue();
+
+  return DAG.getNode(X86ISD::CVTP2SI, DL, VT,
+                     DAG.getNode(ISD::CONCAT_VECTORS, DL, MVT::v4f32, Src,
+                                 DAG.getUNDEF(SrcVT)));
----------------
phoebewang wrote:

No, the result type is legal.

https://github.com/llvm/llvm-project/pull/90065