#define DEBUG_TYPE "basictti"
#include "llvm/CodeGen/Passes.h"
+#include "llvm/Analysis/TargetTransformInfo.h"
#include "llvm/Target/TargetLowering.h"
-#include "llvm/TargetTransformInfo.h"
#include <utility>
using namespace llvm;
namespace {
class BasicTTI : public ImmutablePass, public TargetTransformInfo {
- const TargetLowering *TLI;
+ const TargetLoweringBase *TLI;
/// Estimate the overhead of scalarizing an instruction. Insert and Extract
/// are set if the result needs to be inserted and/or extracted from vectors.
llvm_unreachable("This pass cannot be directly constructed");
}
- BasicTTI(const TargetLowering *TLI) : ImmutablePass(ID), TLI(TLI) {
+ BasicTTI(const TargetLoweringBase *TLI) : ImmutablePass(ID), TLI(TLI) {
initializeBasicTTIPass(*PassRegistry::getPassRegistry());
}
/// @{
virtual unsigned getNumberOfRegisters(bool Vector) const;
+ virtual unsigned getMaximumUnrollFactor() const;
+ virtual unsigned getRegisterBitWidth(bool Vector) const;
virtual unsigned getArithmeticInstrCost(unsigned Opcode, Type *Ty) const;
virtual unsigned getShuffleCost(ShuffleKind Kind, Type *Tp,
int Index, Type *SubTp) const;
virtual unsigned getIntrinsicInstrCost(Intrinsic::ID, Type *RetTy,
ArrayRef<Type*> Tys) const;
virtual unsigned getNumberOfParts(Type *Tp) const;
+ virtual unsigned getAddressComputationCost(Type *Ty) const;
/// @}
};
char BasicTTI::ID = 0;
ImmutablePass *
-llvm::createBasicTargetTransformInfoPass(const TargetLowering *TLI) {
+llvm::createBasicTargetTransformInfoPass(const TargetLoweringBase *TLI) {
return new BasicTTI(TLI);
}
bool BasicTTI::isLegalAddressingMode(Type *Ty, GlobalValue *BaseGV,
int64_t BaseOffset, bool HasBaseReg,
int64_t Scale) const {
- AddrMode AM;
+ TargetLoweringBase::AddrMode AM;
AM.BaseGV = BaseGV;
AM.BaseOffs = BaseOffset;
AM.HasBaseReg = HasBaseReg;
return 1;
}
+unsigned BasicTTI::getRegisterBitWidth(bool Vector) const {
+ return 32;
+}
+
+unsigned BasicTTI::getMaximumUnrollFactor() const {
+ return 1;
+}
+
unsigned BasicTTI::getArithmeticInstrCost(unsigned Opcode, Type *Ty) const {
// Check if any of the operands are vector operands.
int ISD = TLI->InstructionOpcodeToISD(Opcode);
std::pair<unsigned, MVT> SrcLT = TLI->getTypeLegalizationCost(Src);
std::pair<unsigned, MVT> DstLT = TLI->getTypeLegalizationCost(Dst);
+ // Check for NOOP conversions.
+ if (SrcLT.first == DstLT.first &&
+ SrcLT.second.getSizeInBits() == DstLT.second.getSizeInBits()) {
+
+ // Bitcast between types that are legalized to the same type are free.
+ if (Opcode == Instruction::BitCast || Opcode == Instruction::Trunc)
+ return 0;
+ }
+
+ if (Opcode == Instruction::Trunc &&
+ TLI->isTruncateFree(SrcLT.second, DstLT.second))
+ return 0;
+
+ if (Opcode == Instruction::ZExt &&
+ TLI->isZExtFree(SrcLT.second, DstLT.second))
+ return 0;
+
+ // If the cast is marked as legal (or promote) then assume low cost.
+ if (TLI->isOperationLegalOrPromote(ISD, DstLT.second))
+ return 1;
+
// Handle scalar conversions.
if (!Src->isVectorTy() && !Dst->isVectorTy()) {
if (Opcode == Instruction::BitCast)
return 0;
- if (Opcode == Instruction::Trunc &&
- TLI->isTruncateFree(SrcLT.second, DstLT.second))
- return 0;
-
- if (Opcode == Instruction::ZExt &&
- TLI->isZExtFree(SrcLT.second, DstLT.second))
- return 0;
-
// Just check the op cost. If the operation is legal then assume it costs 1.
if (!TLI->isOperationExpand(ISD, DstLT.second))
return 1;
if (SrcLT.first == DstLT.first &&
SrcLT.second.getSizeInBits() == DstLT.second.getSizeInBits()) {
- // Bitcast between types that are legalized to the same type are free.
- if (Opcode == Instruction::BitCast || Opcode == Instruction::Trunc)
- return 0;
-
// Assume that Zext is done using AND.
if (Opcode == Instruction::ZExt)
return 1;
return LT.first;
}
-unsigned BasicTTI::getIntrinsicInstrCost(Intrinsic::ID, Type *RetTy,
+unsigned BasicTTI::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
ArrayRef<Type *> Tys) const {
- // assume that we need to scalarize this intrinsic.
- unsigned ScalarizationCost = 0;
- unsigned ScalarCalls = 1;
- if (RetTy->isVectorTy()) {
- ScalarizationCost = getScalarizationOverhead(RetTy, true, false);
- ScalarCalls = std::max(ScalarCalls, RetTy->getVectorNumElements());
- }
- for (unsigned i = 0, ie = Tys.size(); i != ie; ++i) {
- if (Tys[i]->isVectorTy()) {
- ScalarizationCost += getScalarizationOverhead(Tys[i], false, true);
+ unsigned ISD = 0;
+ switch (IID) {
+ default: {
+ // Assume that we need to scalarize this intrinsic.
+ unsigned ScalarizationCost = 0;
+ unsigned ScalarCalls = 1;
+ if (RetTy->isVectorTy()) {
+ ScalarizationCost = getScalarizationOverhead(RetTy, true, false);
ScalarCalls = std::max(ScalarCalls, RetTy->getVectorNumElements());
}
+ for (unsigned i = 0, ie = Tys.size(); i != ie; ++i) {
+ if (Tys[i]->isVectorTy()) {
+ ScalarizationCost += getScalarizationOverhead(Tys[i], false, true);
+ ScalarCalls = std::max(ScalarCalls, RetTy->getVectorNumElements());
+ }
+ }
+
+ return ScalarCalls + ScalarizationCost;
}
- return ScalarCalls + ScalarizationCost;
+ // Look for intrinsics that can be lowered directly or turned into a scalar
+ // intrinsic call.
+ case Intrinsic::sqrt: ISD = ISD::FSQRT; break;
+ case Intrinsic::sin: ISD = ISD::FSIN; break;
+ case Intrinsic::cos: ISD = ISD::FCOS; break;
+ case Intrinsic::exp: ISD = ISD::FEXP; break;
+ case Intrinsic::exp2: ISD = ISD::FEXP2; break;
+ case Intrinsic::log: ISD = ISD::FLOG; break;
+ case Intrinsic::log10: ISD = ISD::FLOG10; break;
+ case Intrinsic::log2: ISD = ISD::FLOG2; break;
+ case Intrinsic::fabs: ISD = ISD::FABS; break;
+ case Intrinsic::floor: ISD = ISD::FFLOOR; break;
+ case Intrinsic::ceil: ISD = ISD::FCEIL; break;
+ case Intrinsic::trunc: ISD = ISD::FTRUNC; break;
+ case Intrinsic::rint: ISD = ISD::FRINT; break;
+ case Intrinsic::pow: ISD = ISD::FPOW; break;
+ case Intrinsic::fma: ISD = ISD::FMA; break;
+ case Intrinsic::fmuladd: ISD = ISD::FMA; break; // FIXME: mul + add?
+ }
+
+ std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(RetTy);
+
+ if (TLI->isOperationLegalOrPromote(ISD, LT.second)) {
+ // The operation is legal. Assume it costs 1.
+ // If the type is split to multiple registers, assume that thre is some
+ // overhead to this.
+ // TODO: Once we have extract/insert subvector cost we need to use them.
+ if (LT.first > 1)
+ return LT.first * 2;
+ return LT.first * 1;
+ }
+
+ if (!TLI->isOperationExpand(ISD, LT.second)) {
+ // If the operation is custom lowered then assume
+ // thare the code is twice as expensive.
+ return LT.first * 2;
+ }
+
+ // Else, assume that we need to scalarize this intrinsic. For math builtins
+ // this will emit a costly libcall, adding call overhead and spills. Make it
+ // very expensive.
+ if (RetTy->isVectorTy()) {
+ unsigned Num = RetTy->getVectorNumElements();
+ unsigned Cost = TopTTI->getIntrinsicInstrCost(IID, RetTy->getScalarType(),
+ Tys);
+ return 10 * Cost * Num;
+ }
+
+ // This is going to be turned into a library call, make it expensive.
+ return 10;
}
unsigned BasicTTI::getNumberOfParts(Type *Tp) const {
std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(Tp);
return LT.first;
}
+
+unsigned BasicTTI::getAddressComputationCost(Type *Ty) const {
+ return 0;
+}