diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index e65a4f9eb1ec..0f71dbda120f 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -16432,21 +16432,12 @@ static SDValue lowerV2X128Shuffle(const SDLoc &DL, MVT VT, SDValue V1,
     bool SplatHi = isShuffleEquivalent(Mask, {2, 3, 2, 3}, V1);
     if ((SplatLo || SplatHi) && !Subtarget.hasAVX512() && V1.hasOneUse() &&
         X86::mayFoldLoad(peekThroughOneUseBitcasts(V1), Subtarget)) {
+      MVT MemVT = VT.getHalfNumVectorElementsVT();
+      unsigned Ofs = SplatLo ? 0 : MemVT.getStoreSize();
       auto *Ld = cast<LoadSDNode>(peekThroughOneUseBitcasts(V1));
-      if (!Ld->isNonTemporal()) {
-        MVT MemVT = VT.getHalfNumVectorElementsVT();
-        unsigned Ofs = SplatLo ? 0 : MemVT.getStoreSize();
-        SDVTList Tys = DAG.getVTList(VT, MVT::Other);
-        SDValue Ptr = DAG.getMemBasePlusOffset(Ld->getBasePtr(),
-                                               TypeSize::Fixed(Ofs), DL);
-        SDValue Ops[] = {Ld->getChain(), Ptr};
-        SDValue BcastLd = DAG.getMemIntrinsicNode(
-            X86ISD::SUBV_BROADCAST_LOAD, DL, Tys, Ops, MemVT,
-            DAG.getMachineFunction().getMachineMemOperand(
-                Ld->getMemOperand(), Ofs, MemVT.getStoreSize()));
-        DAG.ReplaceAllUsesOfValueWith(SDValue(Ld, 1), BcastLd.getValue(1));
-        return BcastLd;
-      }
+      if (SDValue BcstLd = getBROADCAST_LOAD(X86ISD::SUBV_BROADCAST_LOAD, DL,
+                                             VT, MemVT, Ld, Ofs, DAG))
+        return BcstLd;
     }
 
     // With AVX2, use VPERMQ/VPERMPD for unary shuffles to allow memory folding.