//
//===----------------------------------------------------------------------===//
-#include "llvm/Transforms/Scalar.h"
+#include "llvm/Transforms/Scalar/Reassociate.h"
+#include "llvm/ADT/APFloat.h"
+#include "llvm/ADT/APInt.h"
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/PostOrderIterator.h"
-#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/SetVector.h"
+#include "llvm/ADT/SmallPtrSet.h"
+#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
+#include "llvm/Analysis/GlobalsModRef.h"
+#include "llvm/Analysis/ValueTracking.h"
+#include "llvm/IR/Argument.h"
+#include "llvm/IR/BasicBlock.h"
#include "llvm/IR/CFG.h"
+#include "llvm/IR/Constant.h"
#include "llvm/IR/Constants.h"
-#include "llvm/IR/DerivedTypes.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/IRBuilder.h"
+#include "llvm/IR/InstrTypes.h"
+#include "llvm/IR/Instruction.h"
#include "llvm/IR/Instructions.h"
-#include "llvm/IR/IntrinsicInst.h"
+#include "llvm/IR/Operator.h"
+#include "llvm/IR/PassManager.h"
+#include "llvm/IR/PatternMatch.h"
+#include "llvm/IR/Type.h"
+#include "llvm/IR/User.h"
+#include "llvm/IR/Value.h"
#include "llvm/IR/ValueHandle.h"
#include "llvm/Pass.h"
+#include "llvm/Support/Casting.h"
#include "llvm/Support/Debug.h"
+#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/raw_ostream.h"
+#include "llvm/Transforms/Scalar.h"
#include "llvm/Transforms/Utils/Local.h"
#include <algorithm>
+#include <cassert>
+#include <utility>
+
using namespace llvm;
+using namespace reassociate;
#define DEBUG_TYPE "reassociate"
STATISTIC(NumAnnihil, "Number of expr tree annihilated");
STATISTIC(NumFactor , "Number of multiplies factored");
-namespace {
- struct ValueEntry {
- unsigned Rank;
- Value *Op;
- ValueEntry(unsigned R, Value *O) : Rank(R), Op(O) {}
- };
- inline bool operator<(const ValueEntry &LHS, const ValueEntry &RHS) {
- return LHS.Rank > RHS.Rank; // Sort so that highest rank goes to start.
- }
-}
-
#ifndef NDEBUG
-/// PrintOps - Print out the expression identified in the Ops list.
-///
+/// Print out the expression identified in the Ops list.
static void PrintOps(Instruction *I, const SmallVectorImpl<ValueEntry> &Ops) {
- Module *M = I->getParent()->getParent()->getParent();
+ Module *M = I->getModule();
dbgs() << Instruction::getOpcodeName(I->getOpcode()) << " "
<< *Ops[0].Op->getType() << '\t';
for (unsigned i = 0, e = Ops.size(); i != e; ++i) {
}
#endif
-namespace {
- /// \brief Utility class representing a base and exponent pair which form one
- /// factor of some product.
- struct Factor {
- Value *Base;
- unsigned Power;
-
- Factor(Value *Base, unsigned Power) : Base(Base), Power(Power) {}
-
- /// \brief Sort factors by their Base.
- struct BaseSorter {
- bool operator()(const Factor &LHS, const Factor &RHS) {
- return LHS.Base < RHS.Base;
- }
- };
-
- /// \brief Compare factors for equal bases.
- struct BaseEqual {
- bool operator()(const Factor &LHS, const Factor &RHS) {
- return LHS.Base == RHS.Base;
- }
- };
-
- /// \brief Sort factors in descending order by their power.
- struct PowerDescendingSorter {
- bool operator()(const Factor &LHS, const Factor &RHS) {
- return LHS.Power > RHS.Power;
- }
- };
-
- /// \brief Compare factors for equal powers.
- struct PowerEqual {
- bool operator()(const Factor &LHS, const Factor &RHS) {
- return LHS.Power == RHS.Power;
- }
- };
- };
-
- /// Utility class representing a non-constant Xor-operand. We classify
- /// non-constant Xor-Operands into two categories:
- /// C1) The operand is in the form "X & C", where C is a constant and C != ~0
- /// C2)
- /// C2.1) The operand is in the form of "X | C", where C is a non-zero
- /// constant.
- /// C2.2) Any operand E which doesn't fall into C1 and C2.1, we view this
- /// operand as "E | 0"
- class XorOpnd {
- public:
- XorOpnd(Value *V);
-
- bool isInvalid() const { return SymbolicPart == nullptr; }
- bool isOrExpr() const { return isOr; }
- Value *getValue() const { return OrigVal; }
- Value *getSymbolicPart() const { return SymbolicPart; }
- unsigned getSymbolicRank() const { return SymbolicRank; }
- const APInt &getConstPart() const { return ConstPart; }
-
- void Invalidate() { SymbolicPart = OrigVal = nullptr; }
- void setSymbolicRank(unsigned R) { SymbolicRank = R; }
-
- // Sort the XorOpnd-Pointer in ascending order of symbolic-value-rank.
- // The purpose is twofold:
- // 1) Cluster together the operands sharing the same symbolic-value.
- // 2) Operand having smaller symbolic-value-rank is permuted earlier, which
- // could potentially shorten crital path, and expose more loop-invariants.
- // Note that values' rank are basically defined in RPO order (FIXME).
- // So, if Rank(X) < Rank(Y) < Rank(Z), it means X is defined earlier
- // than Y which is defined earlier than Z. Permute "x | 1", "Y & 2",
- // "z" in the order of X-Y-Z is better than any other orders.
- struct PtrSortFunctor {
- bool operator()(XorOpnd * const &LHS, XorOpnd * const &RHS) {
- return LHS->getSymbolicRank() < RHS->getSymbolicRank();
- }
- };
- private:
- Value *OrigVal;
- Value *SymbolicPart;
- APInt ConstPart;
- unsigned SymbolicRank;
- bool isOr;
- };
-}
-
-namespace {
- class Reassociate : public FunctionPass {
- DenseMap<BasicBlock*, unsigned> RankMap;
- DenseMap<AssertingVH<Value>, unsigned> ValueRankMap;
- SetVector<AssertingVH<Instruction> > RedoInsts;
- bool MadeChange;
- public:
- static char ID; // Pass identification, replacement for typeid
- Reassociate() : FunctionPass(ID) {
- initializeReassociatePass(*PassRegistry::getPassRegistry());
- }
-
- bool runOnFunction(Function &F) override;
-
- void getAnalysisUsage(AnalysisUsage &AU) const override {
- AU.setPreservesCFG();
- }
- private:
- void BuildRankMap(Function &F);
- unsigned getRank(Value *V);
- void ReassociateExpression(BinaryOperator *I);
- void RewriteExprTree(BinaryOperator *I, SmallVectorImpl<ValueEntry> &Ops);
- Value *OptimizeExpression(BinaryOperator *I,
- SmallVectorImpl<ValueEntry> &Ops);
- Value *OptimizeAdd(Instruction *I, SmallVectorImpl<ValueEntry> &Ops);
- Value *OptimizeXor(Instruction *I, SmallVectorImpl<ValueEntry> &Ops);
- bool CombineXorOpnd(Instruction *I, XorOpnd *Opnd1, APInt &ConstOpnd,
- Value *&Res);
- bool CombineXorOpnd(Instruction *I, XorOpnd *Opnd1, XorOpnd *Opnd2,
- APInt &ConstOpnd, Value *&Res);
- bool collectMultiplyFactors(SmallVectorImpl<ValueEntry> &Ops,
- SmallVectorImpl<Factor> &Factors);
- Value *buildMinimalMultiplyDAG(IRBuilder<> &Builder,
- SmallVectorImpl<Factor> &Factors);
- Value *OptimizeMul(BinaryOperator *I, SmallVectorImpl<ValueEntry> &Ops);
- Value *RemoveFactorFromExpression(Value *V, Value *Factor);
- void EraseInst(Instruction *I);
- void OptimizeInst(Instruction *I);
- };
-}
+/// Utility class representing a non-constant Xor-operand. We classify
+/// non-constant Xor-Operands into two categories:
+/// C1) The operand is in the form "X & C", where C is a constant and C != ~0
+/// C2)
+/// C2.1) The operand is in the form of "X | C", where C is a non-zero
+/// constant.
+/// C2.2) Any operand E which doesn't fall into C1 and C2.1, we view this
+/// operand as "E | 0"
+class llvm::reassociate::XorOpnd {
+public:
+ XorOpnd(Value *V);
+
+ bool isInvalid() const { return SymbolicPart == nullptr; }
+ bool isOrExpr() const { return isOr; }
+ Value *getValue() const { return OrigVal; }
+ Value *getSymbolicPart() const { return SymbolicPart; }
+ unsigned getSymbolicRank() const { return SymbolicRank; }
+ const APInt &getConstPart() const { return ConstPart; }
+
+ void Invalidate() { SymbolicPart = OrigVal = nullptr; }
+ void setSymbolicRank(unsigned R) { SymbolicRank = R; }
+
+private:
+ Value *OrigVal;
+ Value *SymbolicPart;
+ APInt ConstPart;
+ unsigned SymbolicRank;
+ bool isOr;
+};
XorOpnd::XorOpnd(Value *V) {
assert(!isa<ConstantInt>(V) && "No ConstantInt");
I->getOpcode() == Instruction::And)) {
Value *V0 = I->getOperand(0);
Value *V1 = I->getOperand(1);
- if (isa<ConstantInt>(V0))
+ const APInt *C;
+ if (match(V0, PatternMatch::m_APInt(C)))
std::swap(V0, V1);
- if (ConstantInt *C = dyn_cast<ConstantInt>(V1)) {
- ConstPart = C->getValue();
+ if (match(V1, PatternMatch::m_APInt(C))) {
+ ConstPart = *C;
SymbolicPart = V0;
isOr = (I->getOpcode() == Instruction::Or);
return;
// view the operand as "V | 0"
SymbolicPart = V;
- ConstPart = APInt::getNullValue(V->getType()->getIntegerBitWidth());
+ ConstPart = APInt::getNullValue(V->getType()->getScalarSizeInBits());
isOr = true;
}
-char Reassociate::ID = 0;
-INITIALIZE_PASS(Reassociate, "reassociate",
- "Reassociate expressions", false, false)
-
-// Public interface to the Reassociate pass
-FunctionPass *llvm::createReassociatePass() { return new Reassociate(); }
-
-/// isReassociableOp - Return true if V is an instruction of the specified
-/// opcode and if it only has one use.
+/// Return true if V is an instruction of the specified opcode and if it
+/// only has one use.
static BinaryOperator *isReassociableOp(Value *V, unsigned Opcode) {
if (V->hasOneUse() && isa<Instruction>(V) &&
- cast<Instruction>(V)->getOpcode() == Opcode)
+ cast<Instruction>(V)->getOpcode() == Opcode &&
+ (!isa<FPMathOperator>(V) || cast<Instruction>(V)->isFast()))
return cast<BinaryOperator>(V);
return nullptr;
}
unsigned Opcode2) {
if (V->hasOneUse() && isa<Instruction>(V) &&
(cast<Instruction>(V)->getOpcode() == Opcode1 ||
- cast<Instruction>(V)->getOpcode() == Opcode2))
+ cast<Instruction>(V)->getOpcode() == Opcode2) &&
+ (!isa<FPMathOperator>(V) || cast<Instruction>(V)->isFast()))
return cast<BinaryOperator>(V);
return nullptr;
}
-static bool isUnmovableInstruction(Instruction *I) {
- switch (I->getOpcode()) {
- case Instruction::PHI:
- case Instruction::LandingPad:
- case Instruction::Alloca:
- case Instruction::Load:
- case Instruction::Invoke:
- case Instruction::UDiv:
- case Instruction::SDiv:
- case Instruction::FDiv:
- case Instruction::URem:
- case Instruction::SRem:
- case Instruction::FRem:
- return true;
- case Instruction::Call:
- return !isa<DbgInfoIntrinsic>(I);
- default:
- return false;
- }
-}
-
-void Reassociate::BuildRankMap(Function &F) {
- unsigned i = 2;
+void ReassociatePass::BuildRankMap(Function &F,
+ ReversePostOrderTraversal<Function*> &RPOT) {
+ unsigned Rank = 2;
- // Assign distinct ranks to function arguments
- for (Function::arg_iterator I = F.arg_begin(), E = F.arg_end(); I != E; ++I)
- ValueRankMap[&*I] = ++i;
+ // Assign distinct ranks to function arguments.
+ for (auto &Arg : F.args()) {
+ ValueRankMap[&Arg] = ++Rank;
+ DEBUG(dbgs() << "Calculated Rank[" << Arg.getName() << "] = " << Rank
+ << "\n");
+ }
- ReversePostOrderTraversal<Function*> RPOT(&F);
- for (ReversePostOrderTraversal<Function*>::rpo_iterator I = RPOT.begin(),
- E = RPOT.end(); I != E; ++I) {
- BasicBlock *BB = *I;
- unsigned BBRank = RankMap[BB] = ++i << 16;
+ // Traverse basic blocks in ReversePostOrder
+ for (BasicBlock *BB : RPOT) {
+ unsigned BBRank = RankMap[BB] = ++Rank << 16;
// Walk the basic block, adding precomputed ranks for any instructions that
// we cannot move. This ensures that the ranks for these instructions are
// all different in the block.
- for (BasicBlock::iterator I = BB->begin(), E = BB->end(); I != E; ++I)
- if (isUnmovableInstruction(I))
- ValueRankMap[&*I] = ++BBRank;
+ for (Instruction &I : *BB)
+ if (mayBeMemoryDependent(I))
+ ValueRankMap[&I] = ++BBRank;
}
}
-unsigned Reassociate::getRank(Value *V) {
+unsigned ReassociatePass::getRank(Value *V) {
Instruction *I = dyn_cast<Instruction>(V);
if (!I) {
if (isa<Argument>(V)) return ValueRankMap[V]; // Function argument.
// If this is a not or neg instruction, do not count it for rank. This
// assures us that X and ~X will have the same rank.
- Type *Ty = V->getType();
- if ((!Ty->isIntegerTy() && !Ty->isFloatingPointTy()) ||
- (!BinaryOperator::isNot(I) && !BinaryOperator::isNeg(I) &&
- !BinaryOperator::isFNeg(I)))
+ if (!BinaryOperator::isNot(I) && !BinaryOperator::isNeg(I) &&
+ !BinaryOperator::isFNeg(I))
++Rank;
- //DEBUG(dbgs() << "Calculated Rank[" << V->getName() << "] = "
- // << Rank << "\n");
+ DEBUG(dbgs() << "Calculated Rank[" << V->getName() << "] = " << Rank << "\n");
return ValueRankMap[I] = Rank;
}
+// Canonicalize constants to RHS. Otherwise, sort the operands by rank.
+void ReassociatePass::canonicalizeOperands(Instruction *I) {
+ assert(isa<BinaryOperator>(I) && "Expected binary operator.");
+ assert(I->isCommutative() && "Expected commutative operator.");
+
+ Value *LHS = I->getOperand(0);
+ Value *RHS = I->getOperand(1);
+ if (LHS == RHS || isa<Constant>(RHS))
+ return;
+ if (isa<Constant>(LHS) || getRank(RHS) < getRank(LHS))
+ cast<BinaryOperator>(I)->swapOperands();
+}
+
static BinaryOperator *CreateAdd(Value *S1, Value *S2, const Twine &Name,
Instruction *InsertBefore, Value *FlagsOp) {
- if (S1->getType()->isIntegerTy())
+ if (S1->getType()->isIntOrIntVectorTy())
return BinaryOperator::CreateAdd(S1, S2, Name, InsertBefore);
else {
BinaryOperator *Res =
static BinaryOperator *CreateMul(Value *S1, Value *S2, const Twine &Name,
Instruction *InsertBefore, Value *FlagsOp) {
- if (S1->getType()->isIntegerTy())
+ if (S1->getType()->isIntOrIntVectorTy())
return BinaryOperator::CreateMul(S1, S2, Name, InsertBefore);
else {
BinaryOperator *Res =
static BinaryOperator *CreateNeg(Value *S1, const Twine &Name,
Instruction *InsertBefore, Value *FlagsOp) {
- if (S1->getType()->isIntegerTy())
+ if (S1->getType()->isIntOrIntVectorTy())
return BinaryOperator::CreateNeg(S1, Name, InsertBefore);
else {
BinaryOperator *Res = BinaryOperator::CreateFNeg(S1, Name, InsertBefore);
}
}
-/// LowerNegateToMultiply - Replace 0-X with X*-1.
-///
+/// Replace 0-X with X*-1.
static BinaryOperator *LowerNegateToMultiply(Instruction *Neg) {
Type *Ty = Neg->getType();
- Constant *NegOne = Ty->isIntegerTy() ? ConstantInt::getAllOnesValue(Ty)
- : ConstantFP::get(Ty, -1.0);
+ Constant *NegOne = Ty->isIntOrIntVectorTy() ?
+ ConstantInt::getAllOnesValue(Ty) : ConstantFP::get(Ty, -1.0);
BinaryOperator *Res = CreateMul(Neg->getOperand(1), NegOne, "", Neg, Neg);
Neg->setOperand(1, Constant::getNullValue(Ty)); // Drop use of op.
return Res;
}
-/// CarmichaelShift - Returns k such that lambda(2^Bitwidth) = 2^k, where lambda
-/// is the Carmichael function. This means that x^(2^k) === 1 mod 2^Bitwidth for
+/// Returns k such that lambda(2^Bitwidth) = 2^k, where lambda is the Carmichael
+/// function. This means that x^(2^k) === 1 mod 2^Bitwidth for
/// every odd x, i.e. x^(2^k) = 1 for every odd x in Bitwidth-bit arithmetic.
/// Note that 0 <= k < Bitwidth, and if Bitwidth > 3 then x^(2^k) = 0 for every
/// even x in Bitwidth-bit arithmetic.
return Bitwidth - 2;
}
-/// IncorporateWeight - Add the extra weight 'RHS' to the existing weight 'LHS',
+/// Add the extra weight 'RHS' to the existing weight 'LHS',
/// reducing the combined weight using any special properties of the operation.
/// The existing weight LHS represents the computation X op X op ... op X where
/// X occurs LHS times. The combined weight represents X op X op ... op X with
}
}
-typedef std::pair<Value*, APInt> RepeatedValue;
+using RepeatedValue = std::pair<Value*, APInt>;
-/// LinearizeExprTree - Given an associative binary expression, return the leaf
+/// Given an associative binary expression, return the leaf
/// nodes in Ops along with their weights (how many times the leaf occurs). The
/// original expression is the same as
/// (Ops[0].first op Ops[0].first op ... Ops[0].first) <- Ops[0].second times
/// that have all uses inside the expression (i.e. only used by non-leaf nodes
/// of the expression) if it can turn them into binary operators of the right
/// type and thus make the expression bigger.
-
static bool LinearizeExprTree(BinaryOperator *I,
SmallVectorImpl<RepeatedValue> &Ops) {
DEBUG(dbgs() << "LINEARIZE: " << *I << '\n');
// ways to get to it.
SmallVector<std::pair<BinaryOperator*, APInt>, 8> Worklist; // (Op, Weight)
Worklist.push_back(std::make_pair(I, APInt(Bitwidth, 1)));
- bool MadeChange = false;
+ bool Changed = false;
// Leaves of the expression are values that either aren't the right kind of
// operation (eg: a constant, or a multiply in an add tree), or are, but have
// Leaves - Keeps track of the set of putative leaves as well as the number of
// paths to each leaf seen so far.
- typedef DenseMap<Value*, APInt> LeafMap;
+ using LeafMap = DenseMap<Value *, APInt>;
LeafMap Leaves; // Leaf -> Total weight so far.
- SmallVector<Value*, 8> LeafOrder; // Ensure deterministic leaf output order.
+ SmallVector<Value *, 8> LeafOrder; // Ensure deterministic leaf output order.
#ifndef NDEBUG
- SmallPtrSet<Value*, 8> Visited; // For sanity checking the iteration scheme.
+ SmallPtrSet<Value *, 8> Visited; // For sanity checking the iteration scheme.
#endif
while (!Worklist.empty()) {
std::pair<BinaryOperator*, APInt> P = Worklist.pop_back_val();
// If this is a binary operation of the right kind with only one use then
// add its operands to the expression.
if (BinaryOperator *BO = isReassociableOp(Op, Opcode)) {
- assert(Visited.insert(Op) && "Not first visit!");
+ assert(Visited.insert(Op).second && "Not first visit!");
DEBUG(dbgs() << "DIRECT ADD: " << *Op << " (" << Weight << ")\n");
Worklist.push_back(std::make_pair(BO, Weight));
continue;
LeafMap::iterator It = Leaves.find(Op);
if (It == Leaves.end()) {
// Not in the leaf map. Must be the first time we saw this operand.
- assert(Visited.insert(Op) && "Not first visit!");
+ assert(Visited.insert(Op).second && "Not first visit!");
if (!Op->hasOneUse()) {
// This value has uses not accounted for by the expression, so it is
// not safe to modify. Mark it as being a leaf.
continue;
}
// No uses outside the expression, try morphing it.
- } else if (It != Leaves.end()) {
+ } else {
// Already in the leaf map.
- assert(Visited.count(Op) && "In leaf map but not visited!");
+ assert(It != Leaves.end() && Visited.count(Op) &&
+ "In leaf map but not visited!");
// Update the number of paths to the leaf.
IncorporateWeight(It->second, Weight, Opcode);
// exactly one such use, drop this new use of the leaf.
assert(!Op->hasOneUse() && "Only one use, but we got here twice!");
I->setOperand(OpIdx, UndefValue::get(I->getType()));
- MadeChange = true;
+ Changed = true;
// If the leaf is a binary operation of the right kind and we now see
// that its multiple original uses were in fact all by nodes belonging
// expression. This means that it can safely be modified. See if we
// can usefully morph it into an expression of the right kind.
assert((!isa<Instruction>(Op) ||
- cast<Instruction>(Op)->getOpcode() != Opcode) &&
+ cast<Instruction>(Op)->getOpcode() != Opcode
+ || (isa<FPMathOperator>(Op) &&
+ !cast<Instruction>(Op)->isFast())) &&
"Should have been handled above!");
assert(Op->hasOneUse() && "Has uses outside the expression tree!");
BO = LowerNegateToMultiply(BO);
DEBUG(dbgs() << *BO << '\n');
Worklist.push_back(std::make_pair(BO, Weight));
- MadeChange = true;
+ Changed = true;
continue;
}
if (Ops.empty()) {
Constant *Identity = ConstantExpr::getBinOpIdentity(Opcode, I->getType());
assert(Identity && "Associative operation without identity!");
- Ops.push_back(std::make_pair(Identity, APInt(Bitwidth, 1)));
+ Ops.emplace_back(Identity, APInt(Bitwidth, 1));
}
- return MadeChange;
+ return Changed;
}
-// RewriteExprTree - Now that the operands for this expression tree are
-// linearized and optimized, emit them in-order.
-void Reassociate::RewriteExprTree(BinaryOperator *I,
- SmallVectorImpl<ValueEntry> &Ops) {
+/// Now that the operands for this expression tree are
+/// linearized and optimized, emit them in-order.
+void ReassociatePass::RewriteExprTree(BinaryOperator *I,
+ SmallVectorImpl<ValueEntry> &Ops) {
assert(Ops.size() > 1 && "Single values should be used directly!");
// Since our optimizations should never increase the number of operations, the
Constant *Undef = UndefValue::get(I->getType());
NewOp = BinaryOperator::Create(Instruction::BinaryOps(Opcode),
Undef, Undef, "", I);
- if (NewOp->getType()->isFloatingPointTy())
+ if (NewOp->getType()->isFPOrFPVectorTy())
NewOp->setFastMathFlags(I->getFastMathFlags());
} else {
NewOp = NodesToRewrite.pop_back_val();
break;
ExpressionChanged->moveBefore(I);
ExpressionChanged = cast<BinaryOperator>(*ExpressionChanged->user_begin());
- } while (1);
+ } while (true);
// Throw away any left over nodes from the original expression.
for (unsigned i = 0, e = NodesToRewrite.size(); i != e; ++i)
RedoInsts.insert(NodesToRewrite[i]);
}
-/// NegateValue - Insert instructions before the instruction pointed to by BI,
+/// Insert instructions before the instruction pointed to by BI,
/// that computes the negative version of the value specified. The negative
/// version of the value is returned, and BI is left pointing at the instruction
/// that should be processed next by the reassociation pass.
-static Value *NegateValue(Value *V, Instruction *BI) {
- if (ConstantFP *C = dyn_cast<ConstantFP>(V))
- return ConstantExpr::getFNeg(C);
- if (Constant *C = dyn_cast<Constant>(V))
+/// Also add intermediate instructions to the redo list that are modified while
+/// pushing the negates through adds. These will be revisited to see if
+/// additional opportunities have been exposed.
+static Value *NegateValue(Value *V, Instruction *BI,
+ SetVector<AssertingVH<Instruction>> &ToRedo) {
+ if (Constant *C = dyn_cast<Constant>(V)) {
+ if (C->getType()->isFPOrFPVectorTy()) {
+ return ConstantExpr::getFNeg(C);
+ }
return ConstantExpr::getNeg(C);
+ }
// We are trying to expose opportunity for reassociation. One of the things
// that we want to do to achieve this is to push a negation as deep into an
if (BinaryOperator *I =
isReassociableOp(V, Instruction::Add, Instruction::FAdd)) {
// Push the negates through the add.
- I->setOperand(0, NegateValue(I->getOperand(0), BI));
- I->setOperand(1, NegateValue(I->getOperand(1), BI));
+ I->setOperand(0, NegateValue(I->getOperand(0), BI, ToRedo));
+ I->setOperand(1, NegateValue(I->getOperand(1), BI, ToRedo));
+ if (I->getOpcode() == Instruction::Add) {
+ I->setHasNoUnsignedWrap(false);
+ I->setHasNoSignedWrap(false);
+ }
// We must move the add instruction here, because the neg instructions do
// not dominate the old add instruction in general. By moving it, we are
//
I->moveBefore(BI);
I->setName(I->getName()+".neg");
+
+ // Add the intermediate negates to the redo list as processing them later
+ // could expose more reassociating opportunities.
+ ToRedo.insert(I);
return I;
}
if (InvokeInst *II = dyn_cast<InvokeInst>(InstInput)) {
InsertPt = II->getNormalDest()->begin();
} else {
- InsertPt = InstInput;
- ++InsertPt;
+ InsertPt = ++InstInput->getIterator();
}
while (isa<PHINode>(InsertPt)) ++InsertPt;
} else {
InsertPt = TheNeg->getParent()->getParent()->getEntryBlock().begin();
}
- TheNeg->moveBefore(InsertPt);
+ TheNeg->moveBefore(&*InsertPt);
+ if (TheNeg->getOpcode() == Instruction::Sub) {
+ TheNeg->setHasNoUnsignedWrap(false);
+ TheNeg->setHasNoSignedWrap(false);
+ } else {
+ TheNeg->andIRFlags(BI);
+ }
+ ToRedo.insert(TheNeg);
return TheNeg;
}
// Insert a 'neg' instruction that subtracts the value from zero to get the
// negation.
- return CreateNeg(V, V->getName() + ".neg", BI, BI);
+ BinaryOperator *NewNeg = CreateNeg(V, V->getName() + ".neg", BI, BI);
+ ToRedo.insert(NewNeg);
+ return NewNeg;
}
-/// ShouldBreakUpSubtract - Return true if we should break up this subtract of
-/// X-Y into (X + -Y).
+/// Return true if we should break up this subtract of X-Y into (X + -Y).
static bool ShouldBreakUpSubtract(Instruction *Sub) {
// If this is a negation, we can't split it up!
if (BinaryOperator::isNeg(Sub) || BinaryOperator::isFNeg(Sub))
return false;
+ // Don't breakup X - undef.
+ if (isa<UndefValue>(Sub->getOperand(1)))
+ return false;
+
// Don't bother to break this up unless either the LHS is an associable add or
// subtract or if this is only used by one.
Value *V0 = Sub->getOperand(0);
return false;
}
-/// BreakUpSubtract - If we have (X-Y), and if either X is an add, or if this is
-/// only used by an add, transform this into (X+(0-Y)) to promote better
-/// reassociation.
-static BinaryOperator *BreakUpSubtract(Instruction *Sub) {
+/// If we have (X-Y), and if either X is an add, or if this is only used by an
+/// add, transform this into (X+(0-Y)) to promote better reassociation.
+static BinaryOperator *
+BreakUpSubtract(Instruction *Sub, SetVector<AssertingVH<Instruction>> &ToRedo) {
// Convert a subtract into an add and a neg instruction. This allows sub
// instructions to be commuted with other add instructions.
//
// Calculate the negative value of Operand 1 of the sub instruction,
// and set it as the RHS of the add instruction we just made.
- //
- Value *NegVal = NegateValue(Sub->getOperand(1), Sub);
+ Value *NegVal = NegateValue(Sub->getOperand(1), Sub, ToRedo);
BinaryOperator *New = CreateAdd(Sub->getOperand(0), NegVal, "", Sub, Sub);
Sub->setOperand(0, Constant::getNullValue(Sub->getType())); // Drop use of op.
Sub->setOperand(1, Constant::getNullValue(Sub->getType())); // Drop use of op.
return New;
}
-/// ConvertShiftToMul - If this is a shift of a reassociable multiply or is used
-/// by one, change this into a multiply by a constant to assist with further
-/// reassociation.
+/// If this is a shift of a reassociable multiply or is used by one, change
+/// this into a multiply by a constant to assist with further reassociation.
static BinaryOperator *ConvertShiftToMul(Instruction *Shl) {
Constant *MulCst = ConstantInt::get(Shl->getType(), 1);
MulCst = ConstantExpr::getShl(MulCst, cast<Constant>(Shl->getOperand(1)));
BinaryOperator::CreateMul(Shl->getOperand(0), MulCst, "", Shl);
Shl->setOperand(0, UndefValue::get(Shl->getType())); // Drop use of op.
Mul->takeName(Shl);
+
+ // Everyone now refers to the mul instruction.
Shl->replaceAllUsesWith(Mul);
Mul->setDebugLoc(Shl->getDebugLoc());
+
+ // We can safely preserve the nuw flag in all cases. It's also safe to turn a
+ // nuw nsw shl into a nuw nsw mul. However, nsw in isolation requires special
+ // handling.
+ bool NSW = cast<BinaryOperator>(Shl)->hasNoSignedWrap();
+ bool NUW = cast<BinaryOperator>(Shl)->hasNoUnsignedWrap();
+ if (NSW && NUW)
+ Mul->setHasNoSignedWrap(true);
+ Mul->setHasNoUnsignedWrap(NUW);
return Mul;
}
-/// FindInOperandList - Scan backwards and forwards among values with the same
-/// rank as element i to see if X exists. If X does not exist, return i. This
-/// is useful when scanning for 'x' when we see '-x' because they both get the
-/// same rank.
-static unsigned FindInOperandList(SmallVectorImpl<ValueEntry> &Ops, unsigned i,
- Value *X) {
+/// Scan backwards and forwards among values with the same rank as element i
+/// to see if X exists. If X does not exist, return i. This is useful when
+/// scanning for 'x' when we see '-x' because they both get the same rank.
+static unsigned FindInOperandList(const SmallVectorImpl<ValueEntry> &Ops,
+ unsigned i, Value *X) {
unsigned XRank = Ops[i].Rank;
unsigned e = Ops.size();
- for (unsigned j = i+1; j != e && Ops[j].Rank == XRank; ++j)
+ for (unsigned j = i+1; j != e && Ops[j].Rank == XRank; ++j) {
if (Ops[j].Op == X)
return j;
+ if (Instruction *I1 = dyn_cast<Instruction>(Ops[j].Op))
+ if (Instruction *I2 = dyn_cast<Instruction>(X))
+ if (I1->isIdenticalTo(I2))
+ return j;
+ }
// Scan backwards.
- for (unsigned j = i-1; j != ~0U && Ops[j].Rank == XRank; --j)
+ for (unsigned j = i-1; j != ~0U && Ops[j].Rank == XRank; --j) {
if (Ops[j].Op == X)
return j;
+ if (Instruction *I1 = dyn_cast<Instruction>(Ops[j].Op))
+ if (Instruction *I2 = dyn_cast<Instruction>(X))
+ if (I1->isIdenticalTo(I2))
+ return j;
+ }
return i;
}
-/// EmitAddTreeOfValues - Emit a tree of add instructions, summing Ops together
+/// Emit a tree of add instructions, summing Ops together
/// and returning the result. Insert the tree before I.
static Value *EmitAddTreeOfValues(Instruction *I,
- SmallVectorImpl<WeakVH> &Ops){
+ SmallVectorImpl<WeakTrackingVH> &Ops) {
if (Ops.size() == 1) return Ops.back();
Value *V1 = Ops.back();
return CreateAdd(V2, V1, "tmp", I, I);
}
-/// RemoveFactorFromExpression - If V is an expression tree that is a
-/// multiplication sequence, and if this sequence contains a multiply by Factor,
+/// If V is an expression tree that is a multiplication sequence,
+/// and if this sequence contains a multiply by Factor,
/// remove Factor from the tree and return the new tree.
-Value *Reassociate::RemoveFactorFromExpression(Value *V, Value *Factor) {
+Value *ReassociatePass::RemoveFactorFromExpression(Value *V, Value *Factor) {
BinaryOperator *BO = isReassociableOp(V, Instruction::Mul, Instruction::FMul);
if (!BO)
return nullptr;
}
} else if (ConstantFP *FC1 = dyn_cast<ConstantFP>(Factor)) {
if (ConstantFP *FC2 = dyn_cast<ConstantFP>(Factors[i].Op)) {
- APFloat F1(FC1->getValueAPF());
+ const APFloat &F1 = FC1->getValueAPF();
APFloat F2(FC2->getValueAPF());
F2.changeSign();
if (F1.compare(F2) == APFloat::cmpEqual) {
return nullptr;
}
- BasicBlock::iterator InsertPt = BO; ++InsertPt;
+ BasicBlock::iterator InsertPt = ++BO->getIterator();
// If this was just a single multiply, remove the multiply and return the only
// remaining operand.
}
if (NeedsNegate)
- V = CreateNeg(V, "neg", InsertPt, BO);
+ V = CreateNeg(V, "neg", &*InsertPt, BO);
return V;
}
-/// FindSingleUseMultiplyFactors - If V is a single-use multiply, recursively
-/// add its operands as factors, otherwise add V to the list of factors.
+/// If V is a single-use multiply, recursively add its operands as factors,
+/// otherwise add V to the list of factors.
///
/// Ops is the top-level list of add operands we're trying to factor.
static void FindSingleUseMultiplyFactors(Value *V,
- SmallVectorImpl<Value*> &Factors,
- const SmallVectorImpl<ValueEntry> &Ops) {
+ SmallVectorImpl<Value*> &Factors) {
BinaryOperator *BO = isReassociableOp(V, Instruction::Mul, Instruction::FMul);
if (!BO) {
Factors.push_back(V);
}
// Otherwise, add the LHS and RHS to the list of factors.
- FindSingleUseMultiplyFactors(BO->getOperand(1), Factors, Ops);
- FindSingleUseMultiplyFactors(BO->getOperand(0), Factors, Ops);
+ FindSingleUseMultiplyFactors(BO->getOperand(1), Factors);
+ FindSingleUseMultiplyFactors(BO->getOperand(0), Factors);
}
-/// OptimizeAndOrXor - Optimize a series of operands to an 'and', 'or', or 'xor'
-/// instruction. This optimizes based on identities. If it can be reduced to
-/// a single Value, it is returned, otherwise the Ops list is mutated as
-/// necessary.
+/// Optimize a series of operands to an 'and', 'or', or 'xor' instruction.
+/// This optimizes based on identities. If it can be reduced to a single Value,
+/// it is returned, otherwise the Ops list is mutated as necessary.
static Value *OptimizeAndOrXor(unsigned Opcode,
SmallVectorImpl<ValueEntry> &Ops) {
// Scan the operand lists looking for X and ~X pairs, along with X,X pairs.
return nullptr;
}
-/// Helper funciton of CombineXorOpnd(). It creates a bitwise-and
+/// Helper function of CombineXorOpnd(). It creates a bitwise-and
/// instruction with the given two operands, and return the resulting
/// instruction. There are two special cases: 1) if the constant operand is 0,
/// it will return NULL. 2) if the constant is ~0, the symbolic operand will
/// be returned.
-static Value *createAndInstr(Instruction *InsertBefore, Value *Opnd,
+static Value *createAndInstr(Instruction *InsertBefore, Value *Opnd,
const APInt &ConstOpnd) {
- if (ConstOpnd != 0) {
- if (!ConstOpnd.isAllOnesValue()) {
- LLVMContext &Ctx = Opnd->getType()->getContext();
- Instruction *I;
- I = BinaryOperator::CreateAnd(Opnd, ConstantInt::get(Ctx, ConstOpnd),
- "and.ra", InsertBefore);
- I->setDebugLoc(InsertBefore->getDebugLoc());
- return I;
- }
+ if (ConstOpnd.isNullValue())
+ return nullptr;
+
+ if (ConstOpnd.isAllOnesValue())
return Opnd;
- }
- return nullptr;
+
+ Instruction *I = BinaryOperator::CreateAnd(
+ Opnd, ConstantInt::get(Opnd->getType(), ConstOpnd), "and.ra",
+ InsertBefore);
+ I->setDebugLoc(InsertBefore->getDebugLoc());
+ return I;
}
// Helper function of OptimizeXor(). It tries to simplify "Opnd1 ^ ConstOpnd"
// If it was successful, true is returned, and the "R" and "C" is returned
// via "Res" and "ConstOpnd", respectively; otherwise, false is returned,
// and both "Res" and "ConstOpnd" remain unchanged.
-//
-bool Reassociate::CombineXorOpnd(Instruction *I, XorOpnd *Opnd1,
- APInt &ConstOpnd, Value *&Res) {
+bool ReassociatePass::CombineXorOpnd(Instruction *I, XorOpnd *Opnd1,
+ APInt &ConstOpnd, Value *&Res) {
// Xor-Rule 1: (x | c1) ^ c2 = (x | c1) ^ (c1 ^ c1) ^ c2
// = ((x | c1) ^ c1) ^ (c1 ^ c2)
// = (x & ~c1) ^ (c1 ^ c2)
// It is useful only when c1 == c2.
- if (Opnd1->isOrExpr() && Opnd1->getConstPart() != 0) {
- if (!Opnd1->getValue()->hasOneUse())
- return false;
+ if (!Opnd1->isOrExpr() || Opnd1->getConstPart().isNullValue())
+ return false;
- const APInt &C1 = Opnd1->getConstPart();
- if (C1 != ConstOpnd)
- return false;
+ if (!Opnd1->getValue()->hasOneUse())
+ return false;
- Value *X = Opnd1->getSymbolicPart();
- Res = createAndInstr(I, X, ~C1);
- // ConstOpnd was C2, now C1 ^ C2.
- ConstOpnd ^= C1;
+ const APInt &C1 = Opnd1->getConstPart();
+ if (C1 != ConstOpnd)
+ return false;
- if (Instruction *T = dyn_cast<Instruction>(Opnd1->getValue()))
- RedoInsts.insert(T);
- return true;
- }
- return false;
-}
+ Value *X = Opnd1->getSymbolicPart();
+ Res = createAndInstr(I, X, ~C1);
+ // ConstOpnd was C2, now C1 ^ C2.
+ ConstOpnd ^= C1;
+ if (Instruction *T = dyn_cast<Instruction>(Opnd1->getValue()))
+ RedoInsts.insert(T);
+ return true;
+}
// Helper function of OptimizeXor(). It tries to simplify
// "Opnd1 ^ Opnd2 ^ ConstOpnd" into "R ^ C", where C would be 0, and R is a
// via "Res" and "ConstOpnd", respectively (If the entire expression is
// evaluated to a constant, the Res is set to NULL); otherwise, false is
// returned, and both "Res" and "ConstOpnd" remain unchanged.
-bool Reassociate::CombineXorOpnd(Instruction *I, XorOpnd *Opnd1, XorOpnd *Opnd2,
- APInt &ConstOpnd, Value *&Res) {
+bool ReassociatePass::CombineXorOpnd(Instruction *I, XorOpnd *Opnd1,
+ XorOpnd *Opnd2, APInt &ConstOpnd,
+ Value *&Res) {
Value *X = Opnd1->getSymbolicPart();
if (X != Opnd2->getSymbolicPart())
return false;
APInt C3((~C1) ^ C2);
// Do not increase code size!
- if (C3 != 0 && !C3.isAllOnesValue()) {
- int NewInstNum = ConstOpnd != 0 ? 1 : 2;
+ if (!C3.isNullValue() && !C3.isAllOnesValue()) {
+ int NewInstNum = ConstOpnd.getBoolValue() ? 1 : 2;
if (NewInstNum > DeadInstNum)
return false;
}
Res = createAndInstr(I, X, C3);
ConstOpnd ^= C1;
-
} else if (Opnd1->isOrExpr()) {
// Xor-Rule 3: (x | c1) ^ (x | c2) = (x & c3) ^ c3 where c3 = c1 ^ c2
//
APInt C3 = C1 ^ C2;
// Do not increase code size
- if (C3 != 0 && !C3.isAllOnesValue()) {
- int NewInstNum = ConstOpnd != 0 ? 1 : 2;
+ if (!C3.isNullValue() && !C3.isAllOnesValue()) {
+ int NewInstNum = ConstOpnd.getBoolValue() ? 1 : 2;
if (NewInstNum > DeadInstNum)
return false;
}
/// Optimize a series of operands to an 'xor' instruction. If it can be reduced
/// to a single Value, it is returned, otherwise the Ops list is mutated as
/// necessary.
-Value *Reassociate::OptimizeXor(Instruction *I,
- SmallVectorImpl<ValueEntry> &Ops) {
+Value *ReassociatePass::OptimizeXor(Instruction *I,
+ SmallVectorImpl<ValueEntry> &Ops) {
if (Value *V = OptimizeAndOrXor(Instruction::Xor, Ops))
return V;
SmallVector<XorOpnd, 8> Opnds;
SmallVector<XorOpnd*, 8> OpndPtrs;
Type *Ty = Ops[0].Op->getType();
- APInt ConstOpnd(Ty->getIntegerBitWidth(), 0);
+ APInt ConstOpnd(Ty->getScalarSizeInBits(), 0);
// Step 1: Convert ValueEntry to XorOpnd
for (unsigned i = 0, e = Ops.size(); i != e; ++i) {
Value *V = Ops[i].Op;
- if (!isa<ConstantInt>(V)) {
+ const APInt *C;
+ // TODO: Support non-splat vectors.
+ if (match(V, PatternMatch::m_APInt(C))) {
+ ConstOpnd ^= *C;
+ } else {
XorOpnd O(V);
O.setSymbolicRank(getRank(O.getSymbolicPart()));
Opnds.push_back(O);
- } else
- ConstOpnd ^= cast<ConstantInt>(V)->getValue();
+ }
}
// NOTE: From this point on, do *NOT* add/delete element to/from "Opnds".
// the same symbolic value cluster together. For instance, the input operand
// sequence ("x | 123", "y & 456", "x & 789") will be sorted into:
// ("x | 123", "x & 789", "y & 456").
- std::stable_sort(OpndPtrs.begin(), OpndPtrs.end(), XorOpnd::PtrSortFunctor());
+ //
+ // The purpose is twofold:
+ // 1) Cluster together the operands sharing the same symbolic-value.
+ // 2) Operand having smaller symbolic-value-rank is permuted earlier, which
+ // could potentially shorten crital path, and expose more loop-invariants.
+ // Note that values' rank are basically defined in RPO order (FIXME).
+ // So, if Rank(X) < Rank(Y) < Rank(Z), it means X is defined earlier
+ // than Y which is defined earlier than Z. Permute "x | 1", "Y & 2",
+ // "z" in the order of X-Y-Z is better than any other orders.
+ std::stable_sort(OpndPtrs.begin(), OpndPtrs.end(),
+ [](XorOpnd *LHS, XorOpnd *RHS) {
+ return LHS->getSymbolicRank() < RHS->getSymbolicRank();
+ });
// Step 3: Combine adjacent operands
XorOpnd *PrevOpnd = nullptr;
Value *CV;
// Step 3.1: Try simplifying "CurrOpnd ^ ConstOpnd"
- if (ConstOpnd != 0 && CombineXorOpnd(I, CurrOpnd, ConstOpnd, CV)) {
+ if (!ConstOpnd.isNullValue() &&
+ CombineXorOpnd(I, CurrOpnd, ConstOpnd, CV)) {
Changed = true;
if (CV)
*CurrOpnd = XorOpnd(CV);
// step 3.2: When previous and current operands share the same symbolic
// value, try to simplify "PrevOpnd ^ CurrOpnd ^ ConstOpnd"
- //
if (CombineXorOpnd(I, CurrOpnd, PrevOpnd, ConstOpnd, CV)) {
// Remove previous operand
PrevOpnd->Invalidate();
ValueEntry VE(getRank(O.getValue()), O.getValue());
Ops.push_back(VE);
}
- if (ConstOpnd != 0) {
- Value *C = ConstantInt::get(Ty->getContext(), ConstOpnd);
+ if (!ConstOpnd.isNullValue()) {
+ Value *C = ConstantInt::get(Ty, ConstOpnd);
ValueEntry VE(getRank(C), C);
Ops.push_back(VE);
}
- int Sz = Ops.size();
+ unsigned Sz = Ops.size();
if (Sz == 1)
return Ops.back().Op;
- else if (Sz == 0) {
- assert(ConstOpnd == 0);
- return ConstantInt::get(Ty->getContext(), ConstOpnd);
+ if (Sz == 0) {
+ assert(ConstOpnd.isNullValue());
+ return ConstantInt::get(Ty, ConstOpnd);
}
}
return nullptr;
}
-/// OptimizeAdd - Optimize a series of operands to an 'add' instruction. This
+/// Optimize a series of operands to an 'add' instruction. This
/// optimizes based on identities. If it can be reduced to a single Value, it
/// is returned, otherwise the Ops list is mutated as necessary.
-Value *Reassociate::OptimizeAdd(Instruction *I,
- SmallVectorImpl<ValueEntry> &Ops) {
+Value *ReassociatePass::OptimizeAdd(Instruction *I,
+ SmallVectorImpl<ValueEntry> &Ops) {
// Scan the operand lists looking for X and -X pairs. If we find any, we
// can simplify expressions like X+-X == 0 and X+~X ==-1. While we're at it,
// scan for any
++NumFound;
} while (i != Ops.size() && Ops[i].Op == TheOp);
- DEBUG(errs() << "\nFACTORING [" << NumFound << "]: " << *TheOp << '\n');
+ DEBUG(dbgs() << "\nFACTORING [" << NumFound << "]: " << *TheOp << '\n');
++NumFactor;
// Insert a new multiply.
Type *Ty = TheOp->getType();
- Constant *C = Ty->isIntegerTy() ? ConstantInt::get(Ty, NumFound)
- : ConstantFP::get(Ty, NumFound);
+ Constant *C = Ty->isIntOrIntVectorTy() ?
+ ConstantInt::get(Ty, NumFound) : ConstantFP::get(Ty, NumFound);
Instruction *Mul = CreateMul(TheOp, C, "factor", I, I);
// Now that we have inserted a multiply, optimize it. This allows us to
// Compute all of the factors of this added value.
SmallVector<Value*, 8> Factors;
- FindSingleUseMultiplyFactors(BOp, Factors, Ops);
+ FindSingleUseMultiplyFactors(BOp, Factors);
assert(Factors.size() > 1 && "Bad linearize!");
// Add one to FactorOccurrences for each unique factor in this op.
SmallPtrSet<Value*, 8> Duplicates;
for (unsigned i = 0, e = Factors.size(); i != e; ++i) {
Value *Factor = Factors[i];
- if (!Duplicates.insert(Factor))
+ if (!Duplicates.insert(Factor).second)
continue;
unsigned Occ = ++FactorOccurrences[Factor];
if (ConstantInt *CI = dyn_cast<ConstantInt>(Factor)) {
if (CI->isNegative() && !CI->isMinValue(true)) {
Factor = ConstantInt::get(CI->getContext(), -CI->getValue());
- assert(!Duplicates.count(Factor) &&
- "Shouldn't have two constant factors, missed a canonicalize");
+ if (!Duplicates.insert(Factor).second)
+ continue;
unsigned Occ = ++FactorOccurrences[Factor];
if (Occ > MaxOcc) {
MaxOcc = Occ;
APFloat F(CF->getValueAPF());
F.changeSign();
Factor = ConstantFP::get(CF->getContext(), F);
- assert(!Duplicates.count(Factor) &&
- "Shouldn't have two constant factors, missed a canonicalize");
+ if (!Duplicates.insert(Factor).second)
+ continue;
unsigned Occ = ++FactorOccurrences[Factor];
if (Occ > MaxOcc) {
MaxOcc = Occ;
// If any factor occurred more than one time, we can pull it out.
if (MaxOcc > 1) {
- DEBUG(errs() << "\nFACTORING [" << MaxOcc << "]: " << *MaxOccVal << '\n');
+ DEBUG(dbgs() << "\nFACTORING [" << MaxOcc << "]: " << *MaxOccVal << '\n');
++NumFactor;
// Create a new instruction that uses the MaxOccVal twice. If we don't do
// from an expression will drop a use of maxocc, and this can cause
// RemoveFactorFromExpression on successive values to behave differently.
Instruction *DummyInst =
- I->getType()->isIntegerTy()
+ I->getType()->isIntOrIntVectorTy()
? BinaryOperator::CreateAdd(MaxOccVal, MaxOccVal)
: BinaryOperator::CreateFAdd(MaxOccVal, MaxOccVal);
- SmallVector<WeakVH, 4> NewMulOps;
+ SmallVector<WeakTrackingVH, 4> NewMulOps;
for (unsigned i = 0; i != Ops.size(); ++i) {
// Only try to remove factors from expressions we're allowed to.
BinaryOperator *BOp =
}
// No need for extra uses anymore.
- delete DummyInst;
+ DummyInst->deleteValue();
unsigned NumAddedValues = NewMulOps.size();
Value *V = EmitAddTreeOfValues(I, NewMulOps);
/// ((((x*y)*x)*y)*x) -> [(x, 3), (y, 2)]
///
/// \returns Whether any factors have a power greater than one.
-bool Reassociate::collectMultiplyFactors(SmallVectorImpl<ValueEntry> &Ops,
- SmallVectorImpl<Factor> &Factors) {
+static bool collectMultiplyFactors(SmallVectorImpl<ValueEntry> &Ops,
+ SmallVectorImpl<Factor> &Factors) {
// FIXME: Have Ops be (ValueEntry, Multiplicity) pairs, simplifying this.
// Compute the sum of powers of simplifiable factors.
unsigned FactorPowerSum = 0;
// below our mininum of '4'.
assert(FactorPowerSum >= 4);
- std::stable_sort(Factors.begin(), Factors.end(), Factor::PowerDescendingSorter());
+ std::stable_sort(Factors.begin(), Factors.end(),
+ [](const Factor &LHS, const Factor &RHS) {
+ return LHS.Power > RHS.Power;
+ });
return true;
}
Value *LHS = Ops.pop_back_val();
do {
- if (LHS->getType()->isIntegerTy())
+ if (LHS->getType()->isIntOrIntVectorTy())
LHS = Builder.CreateMul(LHS, Ops.pop_back_val());
else
LHS = Builder.CreateFMul(LHS, Ops.pop_back_val());
/// equal and the powers are sorted in decreasing order, compute the minimal
/// DAG of multiplies to compute the final product, and return that product
/// value.
-Value *Reassociate::buildMinimalMultiplyDAG(IRBuilder<> &Builder,
- SmallVectorImpl<Factor> &Factors) {
+Value *
+ReassociatePass::buildMinimalMultiplyDAG(IRBuilder<> &Builder,
+ SmallVectorImpl<Factor> &Factors) {
assert(Factors[0].Power);
SmallVector<Value *, 4> OuterProduct;
for (unsigned LastIdx = 0, Idx = 1, Size = Factors.size();
// Unique factors with equal powers -- we've folded them into the first one's
// base.
Factors.erase(std::unique(Factors.begin(), Factors.end(),
- Factor::PowerEqual()),
+ [](const Factor &LHS, const Factor &RHS) {
+ return LHS.Power == RHS.Power;
+ }),
Factors.end());
// Iteratively collect the base of each factor with an add power into the
return V;
}
-Value *Reassociate::OptimizeMul(BinaryOperator *I,
- SmallVectorImpl<ValueEntry> &Ops) {
+Value *ReassociatePass::OptimizeMul(BinaryOperator *I,
+ SmallVectorImpl<ValueEntry> &Ops) {
// We can only optimize the multiplies when there is a chain of more than
// three, such that a balanced tree might require fewer total multiplies.
if (Ops.size() < 4)
return nullptr; // All distinct factors, so nothing left for us to do.
IRBuilder<> Builder(I);
+ // The reassociate transformation for FP operations is performed only
+ // if unsafe algebra is permitted by FastMathFlags. Propagate those flags
+ // to the newly generated operations.
+ if (auto FPI = dyn_cast<FPMathOperator>(I))
+ Builder.setFastMathFlags(FPI->getFastMathFlags());
+
Value *V = buildMinimalMultiplyDAG(Builder, Factors);
if (Ops.empty())
return V;
return nullptr;
}
-Value *Reassociate::OptimizeExpression(BinaryOperator *I,
- SmallVectorImpl<ValueEntry> &Ops) {
+Value *ReassociatePass::OptimizeExpression(BinaryOperator *I,
+ SmallVectorImpl<ValueEntry> &Ops) {
// Now that we have the linearized expression tree, try to optimize it.
// Start by folding any constants that we found.
Constant *Cst = nullptr;
return nullptr;
}
-/// EraseInst - Zap the given instruction, adding interesting operands to the
-/// work list.
-void Reassociate::EraseInst(Instruction *I) {
+// Remove dead instructions and if any operands are trivially dead add them to
+// Insts so they will be removed as well.
+void ReassociatePass::RecursivelyEraseDeadInsts(
+ Instruction *I, SetVector<AssertingVH<Instruction>> &Insts) {
+ assert(isInstructionTriviallyDead(I) && "Trivially dead instructions only!");
+ SmallVector<Value *, 4> Ops(I->op_begin(), I->op_end());
+ ValueRankMap.erase(I);
+ Insts.remove(I);
+ RedoInsts.remove(I);
+ I->eraseFromParent();
+ for (auto Op : Ops)
+ if (Instruction *OpInst = dyn_cast<Instruction>(Op))
+ if (OpInst->use_empty())
+ Insts.insert(OpInst);
+}
+
+/// Zap the given instruction, adding interesting operands to the work list.
+void ReassociatePass::EraseInst(Instruction *I) {
assert(isInstructionTriviallyDead(I) && "Trivially dead instructions only!");
+ DEBUG(dbgs() << "Erasing dead inst: "; I->dump());
+
SmallVector<Value*, 8> Ops(I->op_begin(), I->op_end());
// Erase the dead instruction.
ValueRankMap.erase(I);
// and add that since that's where optimization actually happens.
unsigned Opcode = Op->getOpcode();
while (Op->hasOneUse() && Op->user_back()->getOpcode() == Opcode &&
- Visited.insert(Op))
+ Visited.insert(Op).second)
Op = Op->user_back();
RedoInsts.insert(Op);
}
+
+ MadeChange = true;
}
-/// OptimizeInst - Inspect and optimize the given instruction. Note that erasing
+// Canonicalize expressions of the following form:
+// x + (-Constant * y) -> x - (Constant * y)
+// x - (-Constant * y) -> x + (Constant * y)
+Instruction *ReassociatePass::canonicalizeNegConstExpr(Instruction *I) {
+ if (!I->hasOneUse() || I->getType()->isVectorTy())
+ return nullptr;
+
+ // Must be a fmul or fdiv instruction.
+ unsigned Opcode = I->getOpcode();
+ if (Opcode != Instruction::FMul && Opcode != Instruction::FDiv)
+ return nullptr;
+
+ auto *C0 = dyn_cast<ConstantFP>(I->getOperand(0));
+ auto *C1 = dyn_cast<ConstantFP>(I->getOperand(1));
+
+ // Both operands are constant, let it get constant folded away.
+ if (C0 && C1)
+ return nullptr;
+
+ ConstantFP *CF = C0 ? C0 : C1;
+
+ // Must have one constant operand.
+ if (!CF)
+ return nullptr;
+
+ // Must be a negative ConstantFP.
+ if (!CF->isNegative())
+ return nullptr;
+
+ // User must be a binary operator with one or more uses.
+ Instruction *User = I->user_back();
+ if (!isa<BinaryOperator>(User) || User->use_empty())
+ return nullptr;
+
+ unsigned UserOpcode = User->getOpcode();
+ if (UserOpcode != Instruction::FAdd && UserOpcode != Instruction::FSub)
+ return nullptr;
+
+ // Subtraction is not commutative. Explicitly, the following transform is
+ // not valid: (-Constant * y) - x -> x + (Constant * y)
+ if (!User->isCommutative() && User->getOperand(1) != I)
+ return nullptr;
+
+ // Don't canonicalize x + (-Constant * y) -> x - (Constant * y), if the
+ // resulting subtract will be broken up later. This can get us into an
+ // infinite loop during reassociation.
+ if (UserOpcode == Instruction::FAdd && ShouldBreakUpSubtract(User))
+ return nullptr;
+
+ // Change the sign of the constant.
+ APFloat Val = CF->getValueAPF();
+ Val.changeSign();
+ I->setOperand(C0 ? 0 : 1, ConstantFP::get(CF->getContext(), Val));
+
+ // Canonicalize I to RHS to simplify the next bit of logic. E.g.,
+ // ((-Const*y) + x) -> (x + (-Const*y)).
+ if (User->getOperand(0) == I && User->isCommutative())
+ cast<BinaryOperator>(User)->swapOperands();
+
+ Value *Op0 = User->getOperand(0);
+ Value *Op1 = User->getOperand(1);
+ BinaryOperator *NI;
+ switch (UserOpcode) {
+ default:
+ llvm_unreachable("Unexpected Opcode!");
+ case Instruction::FAdd:
+ NI = BinaryOperator::CreateFSub(Op0, Op1);
+ NI->setFastMathFlags(cast<FPMathOperator>(User)->getFastMathFlags());
+ break;
+ case Instruction::FSub:
+ NI = BinaryOperator::CreateFAdd(Op0, Op1);
+ NI->setFastMathFlags(cast<FPMathOperator>(User)->getFastMathFlags());
+ break;
+ }
+
+ NI->insertBefore(User);
+ NI->setName(User->getName());
+ User->replaceAllUsesWith(NI);
+ NI->setDebugLoc(I->getDebugLoc());
+ RedoInsts.insert(I);
+ MadeChange = true;
+ return NI;
+}
+
+/// Inspect and optimize the given instruction. Note that erasing
/// instructions is not allowed.
-void Reassociate::OptimizeInst(Instruction *I) {
+void ReassociatePass::OptimizeInst(Instruction *I) {
// Only consider operations that we understand.
if (!isa<BinaryOperator>(I))
return;
I = NI;
}
- // Commute floating point binary operators, to canonicalize the order of their
- // operands. This can potentially expose more CSE opportunities, and makes
- // writing other transformations simpler.
- if (I->getType()->isFloatingPointTy() || I->getType()->isVectorTy()) {
-
- // FAdd and FMul can be commuted.
- if (I->getOpcode() == Instruction::FMul ||
- I->getOpcode() == Instruction::FAdd) {
- Value *LHS = I->getOperand(0);
- Value *RHS = I->getOperand(1);
- unsigned LHSRank = getRank(LHS);
- unsigned RHSRank = getRank(RHS);
-
- // Sort the operands by rank.
- if (RHSRank < LHSRank) {
- I->setOperand(0, RHS);
- I->setOperand(1, LHS);
- }
- }
+ // Canonicalize negative constants out of expressions.
+ if (Instruction *Res = canonicalizeNegConstExpr(I))
+ I = Res;
- // FIXME: We should commute vector instructions as well. However, this
- // requires further analysis to determine the effect on later passes.
+ // Commute binary operators, to canonicalize the order of their operands.
+ // This can potentially expose more CSE opportunities, and makes writing other
+ // transformations simpler.
+ if (I->isCommutative())
+ canonicalizeOperands(I);
- // Don't try to optimize vector instructions or anything that doesn't have
- // unsafe algebra.
- if (I->getType()->isVectorTy() || !I->hasUnsafeAlgebra())
- return;
- }
+ // Don't optimize floating-point instructions unless they are 'fast'.
+ if (I->getType()->isFPOrFPVectorTy() && !I->isFast())
+ return;
// Do not reassociate boolean (i1) expressions. We want to preserve the
// original order of evaluation for short-circuited comparisons that
// see if we can convert it to X+-Y.
if (I->getOpcode() == Instruction::Sub) {
if (ShouldBreakUpSubtract(I)) {
- Instruction *NI = BreakUpSubtract(I);
+ Instruction *NI = BreakUpSubtract(I, RedoInsts);
RedoInsts.insert(I);
MadeChange = true;
I = NI;
(!I->hasOneUse() ||
!isReassociableOp(I->user_back(), Instruction::Mul))) {
Instruction *NI = LowerNegateToMultiply(I);
+ // If the negate was simplified, revisit the users to see if we can
+ // reassociate further.
+ for (User *U : NI->users()) {
+ if (BinaryOperator *Tmp = dyn_cast<BinaryOperator>(U))
+ RedoInsts.insert(Tmp);
+ }
RedoInsts.insert(I);
MadeChange = true;
I = NI;
}
} else if (I->getOpcode() == Instruction::FSub) {
if (ShouldBreakUpSubtract(I)) {
- Instruction *NI = BreakUpSubtract(I);
+ Instruction *NI = BreakUpSubtract(I, RedoInsts);
RedoInsts.insert(I);
MadeChange = true;
I = NI;
if (isReassociableOp(I->getOperand(1), Instruction::FMul) &&
(!I->hasOneUse() ||
!isReassociableOp(I->user_back(), Instruction::FMul))) {
+ // If the negate was simplified, revisit the users to see if we can
+ // reassociate further.
Instruction *NI = LowerNegateToMultiply(I);
+ for (User *U : NI->users()) {
+ if (BinaryOperator *Tmp = dyn_cast<BinaryOperator>(U))
+ RedoInsts.insert(Tmp);
+ }
RedoInsts.insert(I);
MadeChange = true;
I = NI;
// If this is an interior node of a reassociable tree, ignore it until we
// get to the root of the tree, to avoid N^2 analysis.
unsigned Opcode = BO->getOpcode();
- if (BO->hasOneUse() && BO->user_back()->getOpcode() == Opcode)
+ if (BO->hasOneUse() && BO->user_back()->getOpcode() == Opcode) {
+ // During the initial run we will get to the root of the tree.
+ // But if we get here while we are redoing instructions, there is no
+ // guarantee that the root will be visited. So Redo later
+ if (BO->user_back() != BO &&
+ BO->getParent() == BO->user_back()->getParent())
+ RedoInsts.insert(BO->user_back());
return;
+ }
// If this is an add tree that is used by a sub instruction, ignore it
// until we process the subtract.
ReassociateExpression(BO);
}
-void Reassociate::ReassociateExpression(BinaryOperator *I) {
- assert(!I->getType()->isVectorTy() &&
- "Reassociation of vector instructions is not supported.");
-
+void ReassociatePass::ReassociateExpression(BinaryOperator *I) {
// First, walk the expression tree, linearizing the tree, collecting the
// operand information.
SmallVector<RepeatedValue, 8> Tree;
// the vector.
std::stable_sort(Ops.begin(), Ops.end());
- // OptimizeExpression - Now that we have the expression tree in a convenient
+ // Now that we have the expression tree in a convenient
// sorted form, optimize it globally if possible.
if (Value *V = OptimizeExpression(I, Ops)) {
if (V == I)
DEBUG(dbgs() << "Reassoc to scalar: " << *V << '\n');
I->replaceAllUsesWith(V);
if (Instruction *VI = dyn_cast<Instruction>(V))
- VI->setDebugLoc(I->getDebugLoc());
+ if (I->getDebugLoc())
+ VI->setDebugLoc(I->getDebugLoc());
RedoInsts.insert(I);
++NumAnnihil;
return;
if (I->getOpcode() == Instruction::Mul &&
cast<Instruction>(I->user_back())->getOpcode() == Instruction::Add &&
isa<ConstantInt>(Ops.back().Op) &&
- cast<ConstantInt>(Ops.back().Op)->isAllOnesValue()) {
+ cast<ConstantInt>(Ops.back().Op)->isMinusOne()) {
ValueEntry Tmp = Ops.pop_back_val();
Ops.insert(Ops.begin(), Tmp);
} else if (I->getOpcode() == Instruction::FMul &&
RewriteExprTree(I, Ops);
}
-bool Reassociate::runOnFunction(Function &F) {
- if (skipOptnoneFunction(F))
- return false;
+PreservedAnalyses ReassociatePass::run(Function &F, FunctionAnalysisManager &) {
+ // Get the functions basic blocks in Reverse Post Order. This order is used by
+ // BuildRankMap to pre calculate ranks correctly. It also excludes dead basic
+ // blocks (it has been seen that the analysis in this pass could hang when
+ // analysing dead basic blocks).
+ ReversePostOrderTraversal<Function *> RPOT(&F);
- // Calculate the rank map for F
- BuildRankMap(F);
+ // Calculate the rank map for F.
+ BuildRankMap(F, RPOT);
MadeChange = false;
- for (Function::iterator BI = F.begin(), BE = F.end(); BI != BE; ++BI) {
+ // Traverse the same blocks that was analysed by BuildRankMap.
+ for (BasicBlock *BI : RPOT) {
+ assert(RankMap.count(&*BI) && "BB should be ranked.");
// Optimize every instruction in the basic block.
- for (BasicBlock::iterator II = BI->begin(), IE = BI->end(); II != IE; )
- if (isInstructionTriviallyDead(II)) {
- EraseInst(II++);
+ for (BasicBlock::iterator II = BI->begin(), IE = BI->end(); II != IE;)
+ if (isInstructionTriviallyDead(&*II)) {
+ EraseInst(&*II++);
} else {
- OptimizeInst(II);
- assert(II->getParent() == BI && "Moved to a different block!");
+ OptimizeInst(&*II);
+ assert(II->getParent() == &*BI && "Moved to a different block!");
++II;
}
- // If this produced extra instructions to optimize, handle them now.
+ // Make a copy of all the instructions to be redone so we can remove dead
+ // instructions.
+ SetVector<AssertingVH<Instruction>> ToRedo(RedoInsts);
+ // Iterate over all instructions to be reevaluated and remove trivially dead
+ // instructions. If any operand of the trivially dead instruction becomes
+ // dead mark it for deletion as well. Continue this process until all
+ // trivially dead instructions have been removed.
+ while (!ToRedo.empty()) {
+ Instruction *I = ToRedo.pop_back_val();
+ if (isInstructionTriviallyDead(I)) {
+ RecursivelyEraseDeadInsts(I, ToRedo);
+ MadeChange = true;
+ }
+ }
+
+ // Now that we have removed dead instructions, we can reoptimize the
+ // remaining instructions.
while (!RedoInsts.empty()) {
Instruction *I = RedoInsts.pop_back_val();
if (isInstructionTriviallyDead(I))
RankMap.clear();
ValueRankMap.clear();
- return MadeChange;
+ if (MadeChange) {
+ PreservedAnalyses PA;
+ PA.preserveSet<CFGAnalyses>();
+ PA.preserve<GlobalsAA>();
+ return PA;
+ }
+
+ return PreservedAnalyses::all();
+}
+
+namespace {
+
+ class ReassociateLegacyPass : public FunctionPass {
+ ReassociatePass Impl;
+
+ public:
+ static char ID; // Pass identification, replacement for typeid
+
+ ReassociateLegacyPass() : FunctionPass(ID) {
+ initializeReassociateLegacyPassPass(*PassRegistry::getPassRegistry());
+ }
+
+ bool runOnFunction(Function &F) override {
+ if (skipFunction(F))
+ return false;
+
+ FunctionAnalysisManager DummyFAM;
+ auto PA = Impl.run(F, DummyFAM);
+ return !PA.areAllPreserved();
+ }
+
+ void getAnalysisUsage(AnalysisUsage &AU) const override {
+ AU.setPreservesCFG();
+ AU.addPreserved<GlobalsAAWrapperPass>();
+ }
+ };
+
+} // end anonymous namespace
+
+char ReassociateLegacyPass::ID = 0;
+
+INITIALIZE_PASS(ReassociateLegacyPass, "reassociate",
+ "Reassociate expressions", false, false)
+
+// Public interface to the Reassociate pass
+FunctionPass *llvm::createReassociatePass() {
+ return new ReassociateLegacyPass();
}