Index: llvm/trunk/include/llvm/CodeGen/TargetLowering.h
===================================================================
--- llvm/trunk/include/llvm/CodeGen/TargetLowering.h
+++ llvm/trunk/include/llvm/CodeGen/TargetLowering.h
@@ -2846,7 +2846,8 @@
                             unsigned Depth = 0,
                             bool AssumeSingleUse = false) const;
 
-  /// Helper wrapper around SimplifyDemandedBits
+  /// Helper wrapper around SimplifyDemandedBits.
+  /// Adds Op back to the worklist upon success.
   bool SimplifyDemandedBits(SDValue Op, const APInt &DemandedMask,
                             DAGCombinerInfo &DCI) const;
 
@@ -2869,7 +2870,8 @@
                                   TargetLoweringOpt &TLO, unsigned Depth = 0,
                                   bool AssumeSingleUse = false) const;
 
-  /// Helper wrapper around SimplifyDemandedVectorElts
+  /// Helper wrapper around SimplifyDemandedVectorElts.
+  /// Adds Op back to the worklist upon success.
   bool SimplifyDemandedVectorElts(SDValue Op, const APInt &DemandedElts,
                                   APInt &KnownUndef, APInt &KnownZero,
                                   DAGCombinerInfo &DCI) const;
Index: llvm/trunk/lib/CodeGen/SelectionDAG/TargetLowering.cpp
===================================================================
--- llvm/trunk/lib/CodeGen/SelectionDAG/TargetLowering.cpp
+++ llvm/trunk/lib/CodeGen/SelectionDAG/TargetLowering.cpp
@@ -490,8 +490,10 @@
   KnownBits Known;
 
   bool Simplified = SimplifyDemandedBits(Op, DemandedMask, Known, TLO);
-  if (Simplified)
+  if (Simplified) {
+    DCI.AddToWorklist(Op.getNode());
     DCI.CommitTargetLoweringOpt(TLO);
+  }
   return Simplified;
 }
 
@@ -1359,8 +1361,10 @@
 
   bool Simplified =
       SimplifyDemandedVectorElts(Op, DemandedElts, KnownUndef, KnownZero, TLO);
-  if (Simplified)
+  if (Simplified) {
+    DCI.AddToWorklist(Op.getNode());
     DCI.CommitTargetLoweringOpt(TLO);
+  }
   return Simplified;
 }
 
Index: llvm/trunk/lib/Target/X86/X86ISelLowering.cpp
===================================================================
--- llvm/trunk/lib/Target/X86/X86ISelLowering.cpp
+++ llvm/trunk/lib/Target/X86/X86ISelLowering.cpp
@@ -36534,16 +36534,10 @@
     // simplify ops leading up to it. We only demand the MSB of each lane.
     SDValue Mask = Mst->getMask();
     if (Mask.getScalarValueSizeInBits() != 1) {
-      TargetLowering::TargetLoweringOpt TLO(DAG, !DCI.isBeforeLegalize(),
-                                            !DCI.isBeforeLegalizeOps());
       const TargetLowering &TLI = DAG.getTargetLoweringInfo();
       APInt DemandedMask(APInt::getSignMask(VT.getScalarSizeInBits()));
-      KnownBits Known;
-      if (TLI.SimplifyDemandedBits(Mask, DemandedMask, Known, TLO)) {
-        DCI.AddToWorklist(Mask.getNode());
-        DCI.CommitTargetLoweringOpt(TLO);
+      if (TLI.SimplifyDemandedBits(Mask, DemandedMask, DCI))
         return SDValue(N, 0);
-      }
     }
 
     // TODO: AVX512 targets should also be able to simplify something like the
@@ -38962,16 +38956,10 @@
   // With AVX2 we only demand the upper bit of the mask.
   if (!Subtarget.hasAVX512()) {
     const TargetLowering &TLI = DAG.getTargetLoweringInfo();
-    TargetLowering::TargetLoweringOpt TLO(DAG, !DCI.isBeforeLegalize(),
-                                          !DCI.isBeforeLegalizeOps());
     SDValue Mask = N->getOperand(2);
-    KnownBits Known;
     APInt DemandedMask(APInt::getSignMask(Mask.getScalarValueSizeInBits()));
-    if (TLI.SimplifyDemandedBits(Mask, DemandedMask, Known, TLO)) {
-      DCI.AddToWorklist(Mask.getNode());
-      DCI.CommitTargetLoweringOpt(TLO);
+    if (TLI.SimplifyDemandedBits(Mask, DemandedMask, DCI))
       return SDValue(N, 0);
-    }
   }
 
   return SDValue();