diff --git a/llvm/lib/Transforms/Vectorize/VPlan.h b/llvm/lib/Transforms/Vectorize/VPlan.h
--- a/llvm/lib/Transforms/Vectorize/VPlan.h
+++ b/llvm/lib/Transforms/Vectorize/VPlan.h
@@ -806,6 +806,16 @@
     // Mask is optional and therefore the last operand.
     return isMasked() ? getOperand(getNumOperands() - 1) : nullptr;
   }
+
+#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+  /// Print the operands to \p O.
+  void printOperands(raw_ostream &O, VPSlotTracker &SlotTracker) const;
+#endif
+
+  const_operand_range operands_without_mask() const {
+    auto End = isMasked() ? op_end() - 1 : op_end();
+    return const_operand_range(op_begin(), End);
+  }
 };
 
 // Helper macro to define common classof implementations for recipes.
diff --git a/llvm/lib/Transforms/Vectorize/VPlan.cpp b/llvm/lib/Transforms/Vectorize/VPlan.cpp
--- a/llvm/lib/Transforms/Vectorize/VPlan.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlan.cpp
@@ -1148,12 +1148,6 @@
   else
     OS << "vp<%" << Tracker.getSlot(this) << ">";
 }
-
-void VPUser::printOperands(raw_ostream &O, VPSlotTracker &SlotTracker) const {
-  interleaveComma(operands(), O, [&O, &SlotTracker](VPValue *Op) {
-    Op->printAsOperand(O, SlotTracker);
-  });
-}
 #endif
 
 void VPInterleavedAccessInfo::visitRegion(VPRegionBlock *Region,
diff --git a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
--- a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
@@ -227,6 +227,20 @@
   insertBefore(BB, I);
 }
 
+#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+void VPRecipeBase::printOperands(raw_ostream &O,
+                                 VPSlotTracker &SlotTracker) const {
+  interleaveComma(operands_without_mask(), O, [&O, &SlotTracker](VPValue *Op) {
+    Op->printAsOperand(O, SlotTracker);
+  });
+  if (isMasked()) {
+    O << " (mask ";
+    getMask()->printAsOperand(O, SlotTracker);
+    O << ")";
+  }
+}
+#endif
+
 FastMathFlags VPRecipeWithIRFlags::getFastMathFlags() const {
   assert(OpType == OperationType::FPMathOp &&
          "recipe doesn't have fast math flags");
diff --git a/llvm/lib/Transforms/Vectorize/VPlanValue.h b/llvm/lib/Transforms/Vectorize/VPlanValue.h
--- a/llvm/lib/Transforms/Vectorize/VPlanValue.h
+++ b/llvm/lib/Transforms/Vectorize/VPlanValue.h
@@ -215,11 +215,6 @@
   VPUserID ID;
 
 protected:
-#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
-  /// Print the operands to \p O.
-  void printOperands(raw_ostream &O, VPSlotTracker &SlotTracker) const;
-#endif
-
   VPUser(ArrayRef<VPValue *> Operands, VPUserID ID) : ID(ID) {
     for (VPValue *Operand : Operands)
       addOperand(Operand);
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding-forced.ll b/llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding-forced.ll
--- a/llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding-forced.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding-forced.ll
@@ -28,7 +28,7 @@
 ; VPLANS-NEXT:     ACTIVE-LANE-MASK-PHI vp<[[LANEMASK_PHI:%[0-9]+]]> = phi vp<[[LANEMASK_ENTRY]]>, vp<[[LANEMASK_LOOP:%[0-9]+]]>
 ; VPLANS-NEXT:     vp<[[STEP:%[0-9]+]]>    = SCALAR-STEPS vp<[[INDV]]>, ir<1>
 ; VPLANS-NEXT:     CLONE ir<%gep> = getelementptr ir<%ptr>, vp<[[STEP]]>
-; VPLANS-NEXT:     WIDEN store ir<%gep>, ir<%val>, vp<[[LANEMASK_PHI]]>
+; VPLANS-NEXT:     WIDEN store ir<%gep>, ir<%val> (mask vp<[[LANEMASK_PHI]]>)
 ; VPLANS-NEXT:     EMIT vp<[[INC:%[0-9]+]]> = VF * Part + vp<[[INDV]]>
 ; VPLANS-NEXT:     EMIT vp<[[LANEMASK_LOOP]]> = active lane mask vp<[[INC]]>, vp<[[NEWTC]]>
 ; VPLANS-NEXT:     EMIT vp<[[INDV_UPDATE:%[0-9]+]]> = VF * UF + vp<[[INDV]]>