llvm-project/llvm/lib/Transforms/Utils/SimplifyCFG.cpp

576 lines
25 KiB
C++

//===- SimplifyCFG.cpp - Code to perform CFG simplification ---------------===//
//
// The LLVM Compiler Infrastructure
//
// This file was developed by the LLVM research group and is distributed under
// the University of Illinois Open Source License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// Peephole optimize the CFG.
//
//===----------------------------------------------------------------------===//
#include "llvm/Transforms/Utils/Local.h"
#include "llvm/Constants.h"
#include "llvm/Instructions.h"
#include "llvm/Support/CFG.h"
#include <algorithm>
#include <functional>
using namespace llvm;
// PropagatePredecessors - This gets "Succ" ready to have the predecessors from
// "BB". This is a little tricky because "Succ" has PHI nodes, which need to
// have extra slots added to them to hold the merge edges from BB's
// predecessors, and BB itself might have had PHI nodes in it. This function
// returns true (failure) if the Succ BB already has a predecessor that is a
// predecessor of BB and incoming PHI arguments would not be discernible.
//
// Assumption: Succ is the single successor for BB.
//
static bool PropagatePredecessorsForPHIs(BasicBlock *BB, BasicBlock *Succ) {
assert(*succ_begin(BB) == Succ && "Succ is not successor of BB!");
if (!isa<PHINode>(Succ->front()))
return false; // We can make the transformation, no problem.
// If there is more than one predecessor, and there are PHI nodes in
// the successor, then we need to add incoming edges for the PHI nodes
//
const std::vector<BasicBlock*> BBPreds(pred_begin(BB), pred_end(BB));
// Check to see if one of the predecessors of BB is already a predecessor of
// Succ. If so, we cannot do the transformation if there are any PHI nodes
// with incompatible values coming in from the two edges!
//
for (pred_iterator PI = pred_begin(Succ), PE = pred_end(Succ); PI != PE; ++PI)
if (find(BBPreds.begin(), BBPreds.end(), *PI) != BBPreds.end()) {
// Loop over all of the PHI nodes checking to see if there are
// incompatible values coming in.
for (BasicBlock::iterator I = Succ->begin();
PHINode *PN = dyn_cast<PHINode>(I); ++I) {
// Loop up the entries in the PHI node for BB and for *PI if the values
// coming in are non-equal, we cannot merge these two blocks (instead we
// should insert a conditional move or something, then merge the
// blocks).
int Idx1 = PN->getBasicBlockIndex(BB);
int Idx2 = PN->getBasicBlockIndex(*PI);
assert(Idx1 != -1 && Idx2 != -1 &&
"Didn't have entries for my predecessors??");
if (PN->getIncomingValue(Idx1) != PN->getIncomingValue(Idx2))
return true; // Values are not equal...
}
}
// Loop over all of the PHI nodes in the successor BB
for (BasicBlock::iterator I = Succ->begin();
PHINode *PN = dyn_cast<PHINode>(I); ++I) {
Value *OldVal = PN->removeIncomingValue(BB, false);
assert(OldVal && "No entry in PHI for Pred BB!");
// If this incoming value is one of the PHI nodes in BB...
if (isa<PHINode>(OldVal) && cast<PHINode>(OldVal)->getParent() == BB) {
PHINode *OldValPN = cast<PHINode>(OldVal);
for (std::vector<BasicBlock*>::const_iterator PredI = BBPreds.begin(),
End = BBPreds.end(); PredI != End; ++PredI) {
PN->addIncoming(OldValPN->getIncomingValueForBlock(*PredI), *PredI);
}
} else {
for (std::vector<BasicBlock*>::const_iterator PredI = BBPreds.begin(),
End = BBPreds.end(); PredI != End; ++PredI) {
// Add an incoming value for each of the new incoming values...
PN->addIncoming(OldVal, *PredI);
}
}
}
return false;
}
/// GetIfCondition - Given a basic block (BB) with two predecessors (and
/// presumably PHI nodes in it), check to see if the merge at this block is due
/// to an "if condition". If so, return the boolean condition that determines
/// which entry into BB will be taken. Also, return by references the block
/// that will be entered from if the condition is true, and the block that will
/// be entered if the condition is false.
///
///
static Value *GetIfCondition(BasicBlock *BB,
BasicBlock *&IfTrue, BasicBlock *&IfFalse) {
assert(std::distance(pred_begin(BB), pred_end(BB)) == 2 &&
"Function can only handle blocks with 2 predecessors!");
BasicBlock *Pred1 = *pred_begin(BB);
BasicBlock *Pred2 = *++pred_begin(BB);
// We can only handle branches. Other control flow will be lowered to
// branches if possible anyway.
if (!isa<BranchInst>(Pred1->getTerminator()) ||
!isa<BranchInst>(Pred2->getTerminator()))
return 0;
BranchInst *Pred1Br = cast<BranchInst>(Pred1->getTerminator());
BranchInst *Pred2Br = cast<BranchInst>(Pred2->getTerminator());
// Eliminate code duplication by ensuring that Pred1Br is conditional if
// either are.
if (Pred2Br->isConditional()) {
// If both branches are conditional, we don't have an "if statement". In
// reality, we could transform this case, but since the condition will be
// required anyway, we stand no chance of eliminating it, so the xform is
// probably not profitable.
if (Pred1Br->isConditional())
return 0;
std::swap(Pred1, Pred2);
std::swap(Pred1Br, Pred2Br);
}
if (Pred1Br->isConditional()) {
// If we found a conditional branch predecessor, make sure that it branches
// to BB and Pred2Br. If it doesn't, this isn't an "if statement".
if (Pred1Br->getSuccessor(0) == BB &&
Pred1Br->getSuccessor(1) == Pred2) {
IfTrue = Pred1;
IfFalse = Pred2;
} else if (Pred1Br->getSuccessor(0) == Pred2 &&
Pred1Br->getSuccessor(1) == BB) {
IfTrue = Pred2;
IfFalse = Pred1;
} else {
// We know that one arm of the conditional goes to BB, so the other must
// go somewhere unrelated, and this must not be an "if statement".
return 0;
}
// The only thing we have to watch out for here is to make sure that Pred2
// doesn't have incoming edges from other blocks. If it does, the condition
// doesn't dominate BB.
if (++pred_begin(Pred2) != pred_end(Pred2))
return 0;
return Pred1Br->getCondition();
}
// Ok, if we got here, both predecessors end with an unconditional branch to
// BB. Don't panic! If both blocks only have a single (identical)
// predecessor, and THAT is a conditional branch, then we're all ok!
if (pred_begin(Pred1) == pred_end(Pred1) ||
++pred_begin(Pred1) != pred_end(Pred1) ||
pred_begin(Pred2) == pred_end(Pred2) ||
++pred_begin(Pred2) != pred_end(Pred2) ||
*pred_begin(Pred1) != *pred_begin(Pred2))
return 0;
// Otherwise, if this is a conditional branch, then we can use it!
BasicBlock *CommonPred = *pred_begin(Pred1);
if (BranchInst *BI = dyn_cast<BranchInst>(CommonPred->getTerminator())) {
assert(BI->isConditional() && "Two successors but not conditional?");
if (BI->getSuccessor(0) == Pred1) {
IfTrue = Pred1;
IfFalse = Pred2;
} else {
IfTrue = Pred2;
IfFalse = Pred1;
}
return BI->getCondition();
}
return 0;
}
// If we have a merge point of an "if condition" as accepted above, return true
// if the specified value dominates the block. We don't handle the true
// generality of domination here, just a special case which works well enough
// for us.
static bool DominatesMergePoint(Value *V, BasicBlock *BB) {
if (Instruction *I = dyn_cast<Instruction>(V)) {
BasicBlock *PBB = I->getParent();
// If this instruction is defined in a block that contains an unconditional
// branch to BB, then it must be in the 'conditional' part of the "if
// statement".
if (isa<BranchInst>(PBB->getTerminator()) &&
cast<BranchInst>(PBB->getTerminator())->isUnconditional() &&
cast<BranchInst>(PBB->getTerminator())->getSuccessor(0) == BB)
return false;
// We also don't want to allow wierd loops that might have the "if
// condition" in the bottom of this block.
if (PBB == BB) return false;
}
// Non-instructions all dominate instructions.
return true;
}
// SimplifyCFG - This function is used to do simplification of a CFG. For
// example, it adjusts branches to branches to eliminate the extra hop, it
// eliminates unreachable basic blocks, and does other "peephole" optimization
// of the CFG. It returns true if a modification was made.
//
// WARNING: The entry node of a function may not be simplified.
//
bool llvm::SimplifyCFG(BasicBlock *BB) {
bool Changed = false;
Function *M = BB->getParent();
assert(BB && BB->getParent() && "Block not embedded in function!");
assert(BB->getTerminator() && "Degenerate basic block encountered!");
assert(&BB->getParent()->front() != BB && "Can't Simplify entry block!");
// Check to see if the first instruction in this block is just an unwind. If
// so, replace any invoke instructions which use this as an exception
// destination with call instructions.
//
if (UnwindInst *UI = dyn_cast<UnwindInst>(BB->getTerminator()))
if (BB->begin() == BasicBlock::iterator(UI)) { // Empty block?
std::vector<BasicBlock*> Preds(pred_begin(BB), pred_end(BB));
while (!Preds.empty()) {
BasicBlock *Pred = Preds.back();
if (InvokeInst *II = dyn_cast<InvokeInst>(Pred->getTerminator()))
if (II->getUnwindDest() == BB) {
// Insert a new branch instruction before the invoke, because this
// is now a fall through...
BranchInst *BI = new BranchInst(II->getNormalDest(), II);
Pred->getInstList().remove(II); // Take out of symbol table
// Insert the call now...
std::vector<Value*> Args(II->op_begin()+3, II->op_end());
CallInst *CI = new CallInst(II->getCalledValue(), Args,
II->getName(), BI);
// If the invoke produced a value, the Call now does instead
II->replaceAllUsesWith(CI);
delete II;
Changed = true;
}
Preds.pop_back();
}
}
// Remove basic blocks that have no predecessors... which are unreachable.
if (pred_begin(BB) == pred_end(BB)) {
//cerr << "Removing BB: \n" << BB;
// Loop through all of our successors and make sure they know that one
// of their predecessors is going away.
for_each(succ_begin(BB), succ_end(BB),
std::bind2nd(std::mem_fun(&BasicBlock::removePredecessor), BB));
while (!BB->empty()) {
Instruction &I = BB->back();
// If this instruction is used, replace uses with an arbitrary
// constant value. Because control flow can't get here, we don't care
// what we replace the value with. Note that since this block is
// unreachable, and all values contained within it must dominate their
// uses, that all uses will eventually be removed.
if (!I.use_empty())
// Make all users of this instruction reference the constant instead
I.replaceAllUsesWith(Constant::getNullValue(I.getType()));
// Remove the instruction from the basic block
BB->getInstList().pop_back();
}
M->getBasicBlockList().erase(BB);
return true;
}
// Check to see if we can constant propagate this terminator instruction
// away...
Changed |= ConstantFoldTerminator(BB);
// Check to see if this block has no non-phi instructions and only a single
// successor. If so, replace references to this basic block with references
// to the successor.
succ_iterator SI(succ_begin(BB));
if (SI != succ_end(BB) && ++SI == succ_end(BB)) { // One succ?
BasicBlock::iterator BBI = BB->begin(); // Skip over phi nodes...
while (isa<PHINode>(*BBI)) ++BBI;
if (BBI->isTerminator()) { // Terminator is the only non-phi instruction!
BasicBlock *Succ = *succ_begin(BB); // There is exactly one successor
if (Succ != BB) { // Arg, don't hurt infinite loops!
// If our successor has PHI nodes, then we need to update them to
// include entries for BB's predecessors, not for BB itself.
// Be careful though, if this transformation fails (returns true) then
// we cannot do this transformation!
//
if (!PropagatePredecessorsForPHIs(BB, Succ)) {
//cerr << "Killing Trivial BB: \n" << BB;
std::string OldName = BB->getName();
std::vector<BasicBlock*>
OldSuccPreds(pred_begin(Succ), pred_end(Succ));
// Move all PHI nodes in BB to Succ if they are alive, otherwise
// delete them.
while (PHINode *PN = dyn_cast<PHINode>(&BB->front()))
if (PN->use_empty())
BB->getInstList().erase(BB->begin()); // Nuke instruction...
else {
// The instruction is alive, so this means that Succ must have
// *ONLY* had BB as a predecessor, and the PHI node is still valid
// now. Simply move it into Succ, because we know that BB
// strictly dominated Succ.
BB->getInstList().remove(BB->begin());
Succ->getInstList().push_front(PN);
// We need to add new entries for the PHI node to account for
// predecessors of Succ that the PHI node does not take into
// account. At this point, since we know that BB dominated succ,
// this means that we should any newly added incoming edges should
// use the PHI node as the value for these edges, because they are
// loop back edges.
for (unsigned i = 0, e = OldSuccPreds.size(); i != e; ++i)
if (OldSuccPreds[i] != BB)
PN->addIncoming(PN, OldSuccPreds[i]);
}
// Everything that jumped to BB now goes to Succ...
BB->replaceAllUsesWith(Succ);
// Delete the old basic block...
M->getBasicBlockList().erase(BB);
if (!OldName.empty() && !Succ->hasName()) // Transfer name if we can
Succ->setName(OldName);
//cerr << "Function after removal: \n" << M;
return true;
}
}
}
}
// If this is a returning block with only PHI nodes in it, fold the return
// instruction into any unconditional branch predecessors.
if (ReturnInst *RI = dyn_cast<ReturnInst>(BB->getTerminator())) {
BasicBlock::iterator BBI = BB->getTerminator();
if (BBI == BB->begin() || isa<PHINode>(--BBI)) {
// Find predecessors that end with unconditional branches.
std::vector<BasicBlock*> UncondBranchPreds;
for (pred_iterator PI = pred_begin(BB), E = pred_end(BB); PI != E; ++PI) {
TerminatorInst *PTI = (*PI)->getTerminator();
if (BranchInst *BI = dyn_cast<BranchInst>(PTI))
if (BI->isUnconditional())
UncondBranchPreds.push_back(*PI);
}
// If we found some, do the transformation!
if (!UncondBranchPreds.empty()) {
while (!UncondBranchPreds.empty()) {
BasicBlock *Pred = UncondBranchPreds.back();
UncondBranchPreds.pop_back();
Instruction *UncondBranch = Pred->getTerminator();
// Clone the return and add it to the end of the predecessor.
Instruction *NewRet = RI->clone();
Pred->getInstList().push_back(NewRet);
// If the return instruction returns a value, and if the value was a
// PHI node in "BB", propagate the right value into the return.
if (NewRet->getNumOperands() == 1)
if (PHINode *PN = dyn_cast<PHINode>(NewRet->getOperand(0)))
if (PN->getParent() == BB)
NewRet->setOperand(0, PN->getIncomingValueForBlock(Pred));
// Update any PHI nodes in the returning block to realize that we no
// longer branch to them.
BB->removePredecessor(Pred);
Pred->getInstList().erase(UncondBranch);
}
// If we eliminated all predecessors of the block, delete the block now.
if (pred_begin(BB) == pred_end(BB))
// We know there are no successors, so just nuke the block.
M->getBasicBlockList().erase(BB);
return true;
}
}
}
// Merge basic blocks into their predecessor if there is only one distinct
// pred, and if there is only one distinct successor of the predecessor, and
// if there are no PHI nodes.
//
pred_iterator PI(pred_begin(BB)), PE(pred_end(BB));
BasicBlock *OnlyPred = *PI++;
for (; PI != PE; ++PI) // Search all predecessors, see if they are all same
if (*PI != OnlyPred) {
OnlyPred = 0; // There are multiple different predecessors...
break;
}
BasicBlock *OnlySucc = 0;
if (OnlyPred && OnlyPred != BB && // Don't break self loops
OnlyPred->getTerminator()->getOpcode() != Instruction::Invoke) {
// Check to see if there is only one distinct successor...
succ_iterator SI(succ_begin(OnlyPred)), SE(succ_end(OnlyPred));
OnlySucc = BB;
for (; SI != SE; ++SI)
if (*SI != OnlySucc) {
OnlySucc = 0; // There are multiple distinct successors!
break;
}
}
if (OnlySucc) {
//cerr << "Merging: " << BB << "into: " << OnlyPred;
TerminatorInst *Term = OnlyPred->getTerminator();
// Resolve any PHI nodes at the start of the block. They are all
// guaranteed to have exactly one entry if they exist, unless there are
// multiple duplicate (but guaranteed to be equal) entries for the
// incoming edges. This occurs when there are multiple edges from
// OnlyPred to OnlySucc.
//
while (PHINode *PN = dyn_cast<PHINode>(&BB->front())) {
PN->replaceAllUsesWith(PN->getIncomingValue(0));
BB->getInstList().pop_front(); // Delete the phi node...
}
// Delete the unconditional branch from the predecessor...
OnlyPred->getInstList().pop_back();
// Move all definitions in the successor to the predecessor...
OnlyPred->getInstList().splice(OnlyPred->end(), BB->getInstList());
// Make all PHI nodes that referred to BB now refer to Pred as their
// source...
BB->replaceAllUsesWith(OnlyPred);
std::string OldName = BB->getName();
// Erase basic block from the function...
M->getBasicBlockList().erase(BB);
// Inherit predecessors name if it exists...
if (!OldName.empty() && !OnlyPred->hasName())
OnlyPred->setName(OldName);
return true;
}
// If there is a trivial two-entry PHI node in this basic block, and we can
// eliminate it, do so now.
if (PHINode *PN = dyn_cast<PHINode>(BB->begin()))
if (PN->getNumIncomingValues() == 2) {
// Ok, this is a two entry PHI node. Check to see if this is a simple "if
// statement", which has a very simple dominance structure. Basically, we
// are trying to find the condition that is being branched on, which
// subsequently causes this merge to happen. We really want control
// dependence information for this check, but simplifycfg can't keep it up
// to date, and this catches most of the cases we care about anyway.
//
BasicBlock *IfTrue, *IfFalse;
if (Value *IfCond = GetIfCondition(BB, IfTrue, IfFalse)) {
//std::cerr << "FOUND IF CONDITION! " << *IfCond << " T: "
// << IfTrue->getName() << " F: " << IfFalse->getName() << "\n";
// Figure out where to insert instructions as necessary.
BasicBlock::iterator AfterPHIIt = BB->begin();
while (isa<PHINode>(AfterPHIIt)) ++AfterPHIIt;
BasicBlock::iterator I = BB->begin();
while (PHINode *PN = dyn_cast<PHINode>(I)) {
++I;
// If we can eliminate this PHI by directly computing it based on the
// condition, do so now. We can't eliminate PHI nodes where the
// incoming values are defined in the conditional parts of the branch,
// so check for this.
//
if (DominatesMergePoint(PN->getIncomingValue(0), BB) &&
DominatesMergePoint(PN->getIncomingValue(1), BB)) {
Value *TrueVal =
PN->getIncomingValue(PN->getIncomingBlock(0) == IfFalse);
Value *FalseVal =
PN->getIncomingValue(PN->getIncomingBlock(0) == IfTrue);
// FIXME: when we have a 'select' statement, we can be completely
// generic and clean here and let the instcombine pass clean up
// after us, by folding the select instructions away when possible.
//
if (TrueVal == FalseVal) {
// Degenerate case...
PN->replaceAllUsesWith(TrueVal);
BB->getInstList().erase(PN);
Changed = true;
} else if (isa<ConstantBool>(TrueVal) &&
isa<ConstantBool>(FalseVal)) {
if (TrueVal == ConstantBool::True) {
// The PHI node produces the same thing as the condition.
PN->replaceAllUsesWith(IfCond);
} else {
// The PHI node produces the inverse of the condition. Insert a
// "NOT" instruction, which is really a XOR.
Value *InverseCond =
BinaryOperator::createNot(IfCond, IfCond->getName()+".inv",
AfterPHIIt);
PN->replaceAllUsesWith(InverseCond);
}
BB->getInstList().erase(PN);
Changed = true;
} else if (isa<ConstantInt>(TrueVal) && isa<ConstantInt>(FalseVal)){
// If this is a PHI of two constant integers, we insert a cast of
// the boolean to the integer type in question, giving us 0 or 1.
// Then we multiply this by the difference of the two constants,
// giving us 0 if false, and the difference if true. We add this
// result to the base constant, giving us our final value. We
// rely on the instruction combiner to eliminate many special
// cases, like turning multiplies into shifts when possible.
std::string Name = PN->getName(); PN->setName("");
Value *TheCast = new CastInst(IfCond, TrueVal->getType(),
Name, AfterPHIIt);
Constant *TheDiff = ConstantExpr::get(Instruction::Sub,
cast<Constant>(TrueVal),
cast<Constant>(FalseVal));
Value *V = TheCast;
if (TheDiff != ConstantInt::get(TrueVal->getType(), 1))
V = BinaryOperator::create(Instruction::Mul, TheCast,
TheDiff, TheCast->getName()+".scale",
AfterPHIIt);
if (!cast<Constant>(FalseVal)->isNullValue())
V = BinaryOperator::create(Instruction::Add, V, FalseVal,
V->getName()+".offs", AfterPHIIt);
PN->replaceAllUsesWith(V);
BB->getInstList().erase(PN);
Changed = true;
} else if (isa<ConstantInt>(FalseVal) &&
cast<Constant>(FalseVal)->isNullValue()) {
// If the false condition is an integral zero value, we can
// compute the PHI by multiplying the condition by the other
// value.
std::string Name = PN->getName(); PN->setName("");
Value *TheCast = new CastInst(IfCond, TrueVal->getType(),
Name+".c", AfterPHIIt);
Value *V = BinaryOperator::create(Instruction::Mul, TrueVal,
TheCast, Name, AfterPHIIt);
PN->replaceAllUsesWith(V);
BB->getInstList().erase(PN);
Changed = true;
} else if (isa<ConstantInt>(TrueVal) &&
cast<Constant>(TrueVal)->isNullValue()) {
// If the true condition is an integral zero value, we can compute
// the PHI by multiplying the inverse condition by the other
// value.
std::string Name = PN->getName(); PN->setName("");
Value *NotCond = BinaryOperator::createNot(IfCond, Name+".inv",
AfterPHIIt);
Value *TheCast = new CastInst(NotCond, TrueVal->getType(),
Name+".inv", AfterPHIIt);
Value *V = BinaryOperator::create(Instruction::Mul, FalseVal,
TheCast, Name, AfterPHIIt);
PN->replaceAllUsesWith(V);
BB->getInstList().erase(PN);
Changed = true;
}
}
}
}
}
return Changed;
}