bool llvm::isInTailCallPosition(SelectionDAG &DAG, SDNode *Node,
SDValue &Chain, const TargetLowering &TLI) {
const Function *F = DAG.getMachineFunction().getFunction();
// Conservatively require the attributes of the call to match those of
// the return. Ignore noalias because it doesn't affect the call sequence.
Attributes CallerRetAttr = F->getAttributes().getRetAttributes();
if (AttrBuilder(CallerRetAttr)
.removeAttribute(Attributes::NoAlias).hasAttributes())
return false;
// It's not safe to eliminate the sign / zero extension of the return value.
if (CallerRetAttr.hasAttribute(Attributes::ZExt) ||
CallerRetAttr.hasAttribute(Attributes::SExt))
return false;
// Check if the only use is a function return node.
return TLI.isUsedByReturnOnly(Node, Chain);
}
/// DupRetToEnableTailCallOpts - Look for opportunities to duplicate return
/// instructions to the predecessor to enable tail call optimizations. The
/// case it is currently looking for is:
/// @code
/// bb0:
/// %tmp0 = tail call i32 @f0()
/// br label %return
/// bb1:
/// %tmp1 = tail call i32 @f1()
/// br label %return
/// bb2:
/// %tmp2 = tail call i32 @f2()
/// br label %return
/// return:
/// %retval = phi i32 [ %tmp0, %bb0 ], [ %tmp1, %bb1 ], [ %tmp2, %bb2 ]
/// ret i32 %retval
/// @endcode
///
/// =>
///
/// @code
/// bb0:
/// %tmp0 = tail call i32 @f0()
/// ret i32 %tmp0
/// bb1:
/// %tmp1 = tail call i32 @f1()
/// ret i32 %tmp1
/// bb2:
/// %tmp2 = tail call i32 @f2()
/// ret i32 %tmp2
/// @endcode
bool CodeGenPrepare::DupRetToEnableTailCallOpts(ReturnInst *RI) {
if (!TLI)
return false;
PHINode *PN = 0;
BitCastInst *BCI = 0;
Value *V = RI->getReturnValue();
if (V) {
BCI = dyn_cast<BitCastInst>(V);
if (BCI)
V = BCI->getOperand(0);
PN = dyn_cast<PHINode>(V);
if (!PN)
return false;
}
BasicBlock *BB = RI->getParent();
if (PN && PN->getParent() != BB)
return false;
// It's not safe to eliminate the sign / zero extension of the return value.
// See llvm::isInTailCallPosition().
const Function *F = BB->getParent();
Attributes CallerRetAttr = F->getAttributes().getRetAttributes();
if (CallerRetAttr.hasAttribute(Attributes::ZExt) ||
CallerRetAttr.hasAttribute(Attributes::SExt))
return false;
// Make sure there are no instructions between the PHI and return, or that the
// return is the first instruction in the block.
if (PN) {
BasicBlock::iterator BI = BB->begin();
do { ++BI; } while (isa<DbgInfoIntrinsic>(BI));
if (&*BI == BCI)
// Also skip over the bitcast.
++BI;
if (&*BI != RI)
return false;
} else {
BasicBlock::iterator BI = BB->begin();
while (isa<DbgInfoIntrinsic>(BI)) ++BI;
if (&*BI != RI)
return false;
}
/// Only dup the ReturnInst if the CallInst is likely to be emitted as a tail
/// call.
SmallVector<CallInst*, 4> TailCalls;
if (PN) {
for (unsigned I = 0, E = PN->getNumIncomingValues(); I != E; ++I) {
CallInst *CI = dyn_cast<CallInst>(PN->getIncomingValue(I));
// Make sure the phi value is indeed produced by the tail call.
if (CI && CI->hasOneUse() && CI->getParent() == PN->getIncomingBlock(I) &&
TLI->mayBeEmittedAsTailCall(CI))
TailCalls.push_back(CI);
}
} else {
SmallPtrSet<BasicBlock*, 4> VisitedBBs;
for (pred_iterator PI = pred_begin(BB), PE = pred_end(BB); PI != PE; ++PI) {
if (!VisitedBBs.insert(*PI))
continue;
BasicBlock::InstListType &InstList = (*PI)->getInstList();
BasicBlock::InstListType::reverse_iterator RI = InstList.rbegin();
BasicBlock::InstListType::reverse_iterator RE = InstList.rend();
do { ++RI; } while (RI != RE && isa<DbgInfoIntrinsic>(&*RI));
if (RI == RE)
continue;
CallInst *CI = dyn_cast<CallInst>(&*RI);
if (CI && CI->use_empty() && TLI->mayBeEmittedAsTailCall(CI))
TailCalls.push_back(CI);
}
}
bool Changed = false;
for (unsigned i = 0, e = TailCalls.size(); i != e; ++i) {
CallInst *CI = TailCalls[i];
CallSite CS(CI);
// Conservatively require the attributes of the call to match those of the
// return. Ignore noalias because it doesn't affect the call sequence.
Attributes CalleeRetAttr = CS.getAttributes().getRetAttributes();
if (AttrBuilder(CalleeRetAttr).
removeAttribute(Attributes::NoAlias) !=
AttrBuilder(CallerRetAttr).
removeAttribute(Attributes::NoAlias))
continue;
// Make sure the call instruction is followed by an unconditional branch to
// the return block.
BasicBlock *CallBB = CI->getParent();
BranchInst *BI = dyn_cast<BranchInst>(CallBB->getTerminator());
if (!BI || !BI->isUnconditional() || BI->getSuccessor(0) != BB)
continue;
// Duplicate the return into CallBB.
(void)FoldReturnIntoUncondBranch(RI, BB, CallBB);
ModifiedDT = Changed = true;
++NumRetsDup;
}
// If we eliminated all predecessors of the block, delete the block now.
if (Changed && !BB->hasAddressTaken() && pred_begin(BB) == pred_end(BB))
BB->eraseFromParent();
return Changed;
}
/// Test if the given instruction is in a position to be optimized
/// with a tail-call. This roughly means that it's in a block with
/// a return and there's nothing that needs to be scheduled
/// between it and the return.
///
/// This function only tests target-independent requirements.
bool llvm::isInTailCallPosition(ImmutableCallSite CS, Attributes CalleeRetAttr,
const TargetLowering &TLI) {
const Instruction *I = CS.getInstruction();
const BasicBlock *ExitBB = I->getParent();
const TerminatorInst *Term = ExitBB->getTerminator();
const ReturnInst *Ret = dyn_cast<ReturnInst>(Term);
// The block must end in a return statement or unreachable.
//
// FIXME: Decline tailcall if it's not guaranteed and if the block ends in
// an unreachable, for now. The way tailcall optimization is currently
// implemented means it will add an epilogue followed by a jump. That is
// not profitable. Also, if the callee is a special function (e.g.
// longjmp on x86), it can end up causing miscompilation that has not
// been fully understood.
if (!Ret &&
(!TLI.getTargetMachine().Options.GuaranteedTailCallOpt ||
!isa<UnreachableInst>(Term)))
return false;
// If I will have a chain, make sure no other instruction that will have a
// chain interposes between I and the return.
if (I->mayHaveSideEffects() || I->mayReadFromMemory() ||
!isSafeToSpeculativelyExecute(I))
for (BasicBlock::const_iterator BBI = prior(prior(ExitBB->end())); ;
--BBI) {
if (&*BBI == I)
break;
// Debug info intrinsics do not get in the way of tail call optimization.
if (isa<DbgInfoIntrinsic>(BBI))
continue;
if (BBI->mayHaveSideEffects() || BBI->mayReadFromMemory() ||
!isSafeToSpeculativelyExecute(BBI))
return false;
}
// If the block ends with a void return or unreachable, it doesn't matter
// what the call's return type is.
if (!Ret || Ret->getNumOperands() == 0) return true;
// If the return value is undef, it doesn't matter what the call's
// return type is.
if (isa<UndefValue>(Ret->getOperand(0))) return true;
// Conservatively require the attributes of the call to match those of
// the return. Ignore noalias because it doesn't affect the call sequence.
const Function *F = ExitBB->getParent();
Attributes CallerRetAttr = F->getAttributes().getRetAttributes();
if (AttrBuilder(CalleeRetAttr).removeAttribute(Attributes::NoAlias) !=
AttrBuilder(CallerRetAttr).removeAttribute(Attributes::NoAlias))
return false;
// It's not safe to eliminate the sign / zero extension of the return value.
if (CallerRetAttr.hasAttribute(Attributes::ZExt) ||
CallerRetAttr.hasAttribute(Attributes::SExt))
return false;
// Otherwise, make sure the unmodified return value of I is the return value.
// We handle two cases: multiple return values + scalars.
Value *RetVal = Ret->getOperand(0);
if (!isa<InsertValueInst>(RetVal) || !isa<StructType>(RetVal->getType()))
// Handle scalars first.
return getNoopInput(Ret->getOperand(0), TLI) == I;
// If this is an aggregate return, look through the insert/extract values and
// see if each is transparent.
for (unsigned i = 0, e =cast<StructType>(RetVal->getType())->getNumElements();
i != e; ++i) {
const Value *InScalar = FindInsertedValue(RetVal, i);
if (InScalar == 0) return false;
InScalar = getNoopInput(InScalar, TLI);
// If the scalar value being inserted is an extractvalue of the right index
// from the call, then everything is good.
const ExtractValueInst *EVI = dyn_cast<ExtractValueInst>(InScalar);
if (EVI == 0 || EVI->getOperand(0) != I || EVI->getNumIndices() != 1 ||
EVI->getIndices()[0] != i)
return false;
}
return true;
}
