1.diff

diff --git a/lib/Transforms/Scalar/CallSiteSplitting.cpp b/lib/Transforms/Scalar/CallSiteSplitting.cpp
index 2e841b2ac94..6ea74218886 100644
--- a/lib/Transforms/Scalar/CallSiteSplitting.cpp
+++ b/lib/Transforms/Scalar/CallSiteSplitting.cpp
@@ -209,8 +209,6 @@ static bool canSplitCallSite(CallSite CS, TargetTransformInfo &TTI) {
   return CallSiteBB->canSplitPredecessors();
 }
 
-/// Return true if the CS is split into its new predecessors.
-///
 /// For each (predecessor, conditions from predecessors) pair, it will split the
 /// basic block containing the call site, hook it up to the predecessor and
 /// replace the call instruction with new call instructions, which contain
@@ -255,12 +253,25 @@ static bool canSplitCallSite(CallSite CS, TargetTransformInfo &TTI) {
 static void splitCallSite(
     CallSite CS,
     const SmallVectorImpl<std::pair<BasicBlock *, ConditionsTy>> &Preds) {
+  bool IsMustTail = CS.isMustTailCall();
   Instruction *Instr = CS.getInstruction();
   BasicBlock *TailBB = Instr->getParent();
 
   PHINode *CallPN = nullptr;
-  if (Instr->getNumUses())
+
+  Instruction *StopAt;
+  if (IsMustTail) {
+    // We need to copy everything including the terminator of the Tail block,
+    // since musttail must be followed by optional bitcast and ret.
+    //
+    StopAt = new UnreachableInst(TailBB->getContext(), TailBB);
+
+    // musttail call results are used just once - no need to create phi.
+  } else if (Instr->getNumUses()) {
     CallPN = PHINode::Create(Instr->getType(), Preds.size(), "phi.call");
+    StopAt = &*std::next(Instr->getIterator());
+  }
+
 
   DEBUG(dbgs() << "split call-site : " << *Instr << " into \n");
 
@@ -271,11 +282,17 @@ static void splitCallSite(
   for (unsigned i = 0; i < Preds.size(); i++) {
     BasicBlock *PredBB = Preds[i].first;
     BasicBlock *SplitBlock = DuplicateInstructionsInSplitBetween(
-        TailBB, PredBB, &*std::next(Instr->getIterator()), ValueToValueMaps[i]);
+        TailBB, PredBB, StopAt, ValueToValueMaps[i]);
     assert(SplitBlock && "Unexpected new basic block split.");
 
-    Instruction *NewCI =
-        &*std::prev(SplitBlock->getTerminator()->getIterator());
+    Instruction *NewCI;
+
+    if (IsMustTail) {
+      NewCI = dyn_cast<Instruction>(&*ValueToValueMaps[i][Instr]);
+      SplitBlock->getTerminator()->removeFromParent();
+    } else {
+      NewCI = &*std::prev(SplitBlock->getTerminator()->getIterator());
+    }
     CallSite NewCS(NewCI);
     addConditions(NewCS, Preds[i].second);
 
@@ -309,7 +326,7 @@ static void splitCallSite(
   // instruction, so we do not end up deleting them. By using reverse-order, we
   // do not introduce unnecessary PHI nodes for def-use chains from the call
   // instruction to the beginning of the block.
-  auto I = Instr->getReverseIterator();
+  auto I = (&*std::prev(StopAt->getIterator()))->getReverseIterator();
   while (I != TailBB->rend()) {
     Instruction *CurrentI = &*I++;
     if (!CurrentI->use_empty()) {
@@ -330,6 +347,9 @@ static void splitCallSite(
       break;
   }
 
+  if (IsMustTail)
+    TailBB->eraseFromParent();
+
   NumCallSiteSplit++;
 }