diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -8157,40 +8157,9 @@
   // load/store/gather/scatter. Initialize BlockMask to no-mask.
   VPValue *BlockMask = nullptr;
 
-  if (OrigLoop->getHeader() == BB) {
     if (!CM.blockNeedsPredicationForAnyReason(BB))
       return BlockMaskCache[BB] = BlockMask; // Loop incoming mask is all-one.
-
-    assert(CM.foldTailByMasking() && "must fold the tail");
-
-    // If we're using the active lane mask for control flow, then we get the
-    // mask from the active lane mask PHI that is cached in the VPlan.
-    TailFoldingStyle TFStyle = CM.getTailFoldingStyle();
-    if (useActiveLaneMaskForControlFlow(TFStyle))
-      return BlockMaskCache[BB] = Plan.getActiveLaneMaskPhi();
-
-    // Introduce the early-exit compare IV <= BTC to form header block mask.
-    // This is used instead of IV < TC because TC may wrap, unlike BTC. Start by
-    // constructing the desired canonical IV in the header block as its first
-    // non-phi instructions.
-
-    VPBasicBlock *HeaderVPBB = Plan.getVectorLoopRegion()->getEntryBasicBlock();
-    auto NewInsertionPoint = HeaderVPBB->getFirstNonPhi();
-    auto *IV = new VPWidenCanonicalIVRecipe(Plan.getCanonicalIV());
-    HeaderVPBB->insert(IV, HeaderVPBB->getFirstNonPhi());
-
-    VPBuilder::InsertPointGuard Guard(Builder);
-    Builder.setInsertPoint(HeaderVPBB, NewInsertionPoint);
-    if (useActiveLaneMask(TFStyle)) {
-      VPValue *TC = Plan.getTripCount();
-      BlockMask = Builder.createNaryOp(VPInstruction::ActiveLaneMask, {IV, TC},
-                                       nullptr, "active.lane.mask");
-    } else {
-      VPValue *BTC = Plan.getOrCreateBackedgeTakenCount();
-      BlockMask = Builder.createNaryOp(VPInstruction::ICmpULE, {IV, BTC});
-    }
-    return BlockMaskCache[BB] = BlockMask;
-  }
+  assert (OrigLoop->getHeader() != BB);
 
   // This is the block mask. We OR all incoming edges.
   for (auto *Predecessor : predecessors(BB)) {
@@ -8209,6 +8178,48 @@
   return BlockMaskCache[BB] = BlockMask;
 }
 
+VPValue *VPRecipeBuilder::createTailFoldHeaderMask(VPlan &Plan) {
+  BasicBlock *Header = OrigLoop->getHeader();
+  // Look for cached value.
+  BlockMaskCacheTy::iterator BCEntryIt = BlockMaskCache.find(Header);
+  if (BCEntryIt != BlockMaskCache.end())
+    return BCEntryIt->second;
+
+  // All-one mask is modelled as no-mask following the convention for masked
+  // load/store/gather/scatter. Initialize BlockMask to no-mask.
+  VPValue *BlockMask = nullptr;
+
+  assert(CM.foldTailByMasking() && "must fold the tail");
+
+  // If we're using the active lane mask for control flow, then we get the
+  // mask from the active lane mask PHI that is cached in the VPlan.
+  TailFoldingStyle TFStyle = CM.getTailFoldingStyle();
+  if (useActiveLaneMaskForControlFlow(TFStyle))
+    return BlockMaskCache[Header] = Plan.getActiveLaneMaskPhi();
+
+  // Introduce the early-exit compare IV <= BTC to form header block mask.
+  // This is used instead of IV < TC because TC may wrap, unlike BTC. Start by
+  // constructing the desired canonical IV in the header block as its first
+  // non-phi instructions.
+
+  VPBasicBlock *HeaderVPBB = Plan.getVectorLoopRegion()->getEntryBasicBlock();
+  auto NewInsertionPoint = HeaderVPBB->getFirstNonPhi();
+  auto *IV = new VPWidenCanonicalIVRecipe(Plan.getCanonicalIV());
+  HeaderVPBB->insert(IV, HeaderVPBB->getFirstNonPhi());
+
+  VPBuilder::InsertPointGuard Guard(Builder);
+  Builder.setInsertPoint(HeaderVPBB, NewInsertionPoint);
+  if (useActiveLaneMask(TFStyle)) {
+    VPValue *TC = Plan.getTripCount();
+    BlockMask = Builder.createNaryOp(VPInstruction::ActiveLaneMask, {IV, TC},
+                                     nullptr, "active.lane.mask");
+  } else {
+    VPValue *BTC = Plan.getOrCreateBackedgeTakenCount();
+    BlockMask = Builder.createNaryOp(VPInstruction::ICmpULE, {IV, BTC});
+  }
+  return BlockMaskCache[Header] = BlockMask;
+}
+
 VPRecipeBase *VPRecipeBuilder::tryToWidenMemory(Instruction *I,
                                                 ArrayRef<VPValue *> Operands,
                                                 VFRange &Range,
@@ -8903,6 +8914,9 @@
                         DLInst ? DLInst->getDebugLoc() : DebugLoc(),
                         CM.getTailFoldingStyle(IVUpdateMayOverflow));
 
+  if (CM.foldTailByMasking())
+    RecipeBuilder.createTailFoldHeaderMask(*Plan);
+
   // Scan the body of the loop in a topological order to visit each basic block
   // after having visited its predecessor basic blocks.
   LoopBlocksDFS DFS(OrigLoop);
@@ -8965,11 +8979,11 @@
       }
 
       RecipeBuilder.setRecipe(Instr, Recipe);
-      if (isa<VPWidenIntOrFpInductionRecipe>(Recipe) &&
+      if (isa<VPHeaderPHIRecipe>(Recipe) &&
           HeaderVPBB->getFirstNonPhi() != VPBB->end()) {
         // Move VPWidenIntOrFpInductionRecipes for optimized truncates to the
         // phi section of HeaderVPBB.
-        assert(isa<TruncInst>(Instr));
+        assert(CM.foldTailByMasking() || isa<TruncInst>(Instr));
         Recipe->insertBefore(*HeaderVPBB, HeaderVPBB->getFirstNonPhi());
       } else
         VPBB->appendRecipe(Recipe);
diff --git a/llvm/lib/Transforms/Vectorize/VPRecipeBuilder.h b/llvm/lib/Transforms/Vectorize/VPRecipeBuilder.h
--- a/llvm/lib/Transforms/Vectorize/VPRecipeBuilder.h
+++ b/llvm/lib/Transforms/Vectorize/VPRecipeBuilder.h
@@ -138,6 +138,9 @@
   /// mask for the block BB.
   VPValue *createBlockInMask(BasicBlock *BB, VPlan &Plan);
 
+  /// Create the mask for the vector loop header block when tail-folding.
+  VPValue *createTailFoldHeaderMask(VPlan &Plan);
+
   /// A helper function that computes the predicate of the edge between SRC
   /// and DST.
   VPValue *createEdgeMask(BasicBlock *Src, BasicBlock *Dst, VPlan &Plan);