ARMTargetTransformInfo.cpp [plain text]
#include "ARM.h"
#include "ARMTargetMachine.h"
#include "llvm/Analysis/TargetTransformInfo.h"
#include "llvm/Support/Debug.h"
#include "llvm/Target/CostTable.h"
#include "llvm/Target/TargetLowering.h"
using namespace llvm;
#define DEBUG_TYPE "armtti"
namespace llvm {
void initializeARMTTIPass(PassRegistry &);
}
namespace {
class ARMTTI final : public ImmutablePass, public TargetTransformInfo {
const ARMBaseTargetMachine *TM;
const ARMSubtarget *ST;
const ARMTargetLowering *TLI;
unsigned getScalarizationOverhead(Type *Ty, bool Insert, bool Extract) const;
public:
ARMTTI() : ImmutablePass(ID), TM(nullptr), ST(nullptr), TLI(nullptr) {
llvm_unreachable("This pass cannot be directly constructed");
}
ARMTTI(const ARMBaseTargetMachine *TM)
: ImmutablePass(ID), TM(TM), ST(TM->getSubtargetImpl()),
TLI(TM->getSubtargetImpl()->getTargetLowering()) {
initializeARMTTIPass(*PassRegistry::getPassRegistry());
}
void initializePass() override {
pushTTIStack(this);
}
void getAnalysisUsage(AnalysisUsage &AU) const override {
TargetTransformInfo::getAnalysisUsage(AU);
}
static char ID;
void *getAdjustedAnalysisPointer(const void *ID) override {
if (ID == &TargetTransformInfo::ID)
return (TargetTransformInfo*)this;
return this;
}
using TargetTransformInfo::getIntImmCost;
unsigned getIntImmCost(const APInt &Imm, Type *Ty) const override;
unsigned getNumberOfRegisters(bool Vector) const override {
if (Vector) {
if (ST->hasNEON())
return 16;
return 0;
}
if (ST->isThumb1Only())
return 8;
return 13;
}
unsigned getRegisterBitWidth(bool Vector) const override {
if (Vector) {
if (ST->hasNEON())
return 128;
return 0;
}
return 32;
}
unsigned getMaximumUnrollFactor() const override {
if (ST->isCortexA15() || ST->isSwift())
return 2;
return 1;
}
unsigned getShuffleCost(ShuffleKind Kind, Type *Tp,
int Index, Type *SubTp) const override;
unsigned getCastInstrCost(unsigned Opcode, Type *Dst,
Type *Src) const override;
unsigned getCmpSelInstrCost(unsigned Opcode, Type *ValTy,
Type *CondTy) const override;
unsigned getVectorInstrCost(unsigned Opcode, Type *Val,
unsigned Index) const override;
unsigned getAddressComputationCost(Type *Val,
bool IsComplex) const override;
unsigned
getArithmeticInstrCost(unsigned Opcode, Type *Ty,
OperandValueKind Op1Info = OK_AnyValue,
OperandValueKind Op2Info = OK_AnyValue) const override;
unsigned getMemoryOpCost(unsigned Opcode, Type *Src, unsigned Alignment,
unsigned AddressSpace) const override;
};
}
INITIALIZE_AG_PASS(ARMTTI, TargetTransformInfo, "armtti",
"ARM Target Transform Info", true, true, false)
char ARMTTI::ID = 0;
ImmutablePass *
llvm::createARMTargetTransformInfoPass(const ARMBaseTargetMachine *TM) {
return new ARMTTI(TM);
}
unsigned ARMTTI::getIntImmCost(const APInt &Imm, Type *Ty) const {
assert(Ty->isIntegerTy());
unsigned Bits = Ty->getPrimitiveSizeInBits();
if (Bits == 0 || Bits > 32)
return 4;
int32_t SImmVal = Imm.getSExtValue();
uint32_t ZImmVal = Imm.getZExtValue();
if (!ST->isThumb()) {
if ((SImmVal >= 0 && SImmVal < 65536) ||
(ARM_AM::getSOImmVal(ZImmVal) != -1) ||
(ARM_AM::getSOImmVal(~ZImmVal) != -1))
return 1;
return ST->hasV6T2Ops() ? 2 : 3;
}
if (ST->isThumb2()) {
if ((SImmVal >= 0 && SImmVal < 65536) ||
(ARM_AM::getT2SOImmVal(ZImmVal) != -1) ||
(ARM_AM::getT2SOImmVal(~ZImmVal) != -1))
return 1;
return ST->hasV6T2Ops() ? 2 : 3;
}
if (SImmVal >= 0 && SImmVal < 256)
return 1;
if ((~ZImmVal < 256) || ARM_AM::isThumbImmShiftedVal(ZImmVal))
return 2;
return 3;
}
unsigned ARMTTI::getCastInstrCost(unsigned Opcode, Type *Dst,
Type *Src) const {
int ISD = TLI->InstructionOpcodeToISD(Opcode);
assert(ISD && "Invalid opcode");
static const CostTblEntry<MVT::SimpleValueType> NEONFltDblTbl[] = {
{ ISD::FP_ROUND, MVT::v2f64, 2 },
{ ISD::FP_EXTEND, MVT::v2f32, 2 },
{ ISD::FP_EXTEND, MVT::v4f32, 4 }
};
if (Src->isVectorTy() && ST->hasNEON() && (ISD == ISD::FP_ROUND ||
ISD == ISD::FP_EXTEND)) {
std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(Src);
int Idx = CostTableLookup(NEONFltDblTbl, ISD, LT.second);
if (Idx != -1)
return LT.first * NEONFltDblTbl[Idx].Cost;
}
EVT SrcTy = TLI->getValueType(Src);
EVT DstTy = TLI->getValueType(Dst);
if (!SrcTy.isSimple() || !DstTy.isSimple())
return TargetTransformInfo::getCastInstrCost(Opcode, Dst, Src);
static const TypeConversionCostTblEntry<MVT::SimpleValueType>
NEONVectorConversionTbl[] = {
{ ISD::SIGN_EXTEND, MVT::v4i32, MVT::v4i16, 0 },
{ ISD::ZERO_EXTEND, MVT::v4i32, MVT::v4i16, 0 },
{ ISD::SIGN_EXTEND, MVT::v2i64, MVT::v2i32, 1 },
{ ISD::ZERO_EXTEND, MVT::v2i64, MVT::v2i32, 1 },
{ ISD::TRUNCATE, MVT::v4i32, MVT::v4i64, 0 },
{ ISD::TRUNCATE, MVT::v4i16, MVT::v4i32, 1 },
{ ISD::SIGN_EXTEND, MVT::v4i64, MVT::v4i16, 3 },
{ ISD::ZERO_EXTEND, MVT::v4i64, MVT::v4i16, 3 },
{ ISD::SIGN_EXTEND, MVT::v8i32, MVT::v8i8, 3 },
{ ISD::ZERO_EXTEND, MVT::v8i32, MVT::v8i8, 3 },
{ ISD::SIGN_EXTEND, MVT::v8i64, MVT::v8i8, 7 },
{ ISD::ZERO_EXTEND, MVT::v8i64, MVT::v8i8, 7 },
{ ISD::SIGN_EXTEND, MVT::v8i64, MVT::v8i16, 6 },
{ ISD::ZERO_EXTEND, MVT::v8i64, MVT::v8i16, 6 },
{ ISD::SIGN_EXTEND, MVT::v16i32, MVT::v16i8, 6 },
{ ISD::ZERO_EXTEND, MVT::v16i32, MVT::v16i8, 6 },
{ ISD::TRUNCATE, MVT::v16i8, MVT::v16i32, 6 },
{ ISD::TRUNCATE, MVT::v8i8, MVT::v8i32, 3 },
{ ISD::SINT_TO_FP, MVT::v4f32, MVT::v4i32, 1 },
{ ISD::UINT_TO_FP, MVT::v4f32, MVT::v4i32, 1 },
{ ISD::SINT_TO_FP, MVT::v2f32, MVT::v2i8, 3 },
{ ISD::UINT_TO_FP, MVT::v2f32, MVT::v2i8, 3 },
{ ISD::SINT_TO_FP, MVT::v2f32, MVT::v2i16, 2 },
{ ISD::UINT_TO_FP, MVT::v2f32, MVT::v2i16, 2 },
{ ISD::SINT_TO_FP, MVT::v2f32, MVT::v2i32, 1 },
{ ISD::UINT_TO_FP, MVT::v2f32, MVT::v2i32, 1 },
{ ISD::SINT_TO_FP, MVT::v4f32, MVT::v4i1, 3 },
{ ISD::UINT_TO_FP, MVT::v4f32, MVT::v4i1, 3 },
{ ISD::SINT_TO_FP, MVT::v4f32, MVT::v4i8, 3 },
{ ISD::UINT_TO_FP, MVT::v4f32, MVT::v4i8, 3 },
{ ISD::SINT_TO_FP, MVT::v4f32, MVT::v4i16, 2 },
{ ISD::UINT_TO_FP, MVT::v4f32, MVT::v4i16, 2 },
{ ISD::SINT_TO_FP, MVT::v8f32, MVT::v8i16, 4 },
{ ISD::UINT_TO_FP, MVT::v8f32, MVT::v8i16, 4 },
{ ISD::SINT_TO_FP, MVT::v8f32, MVT::v8i32, 2 },
{ ISD::UINT_TO_FP, MVT::v8f32, MVT::v8i32, 2 },
{ ISD::SINT_TO_FP, MVT::v16f32, MVT::v16i16, 8 },
{ ISD::UINT_TO_FP, MVT::v16f32, MVT::v16i16, 8 },
{ ISD::SINT_TO_FP, MVT::v16f32, MVT::v16i32, 4 },
{ ISD::UINT_TO_FP, MVT::v16f32, MVT::v16i32, 4 },
{ ISD::FP_TO_SINT, MVT::v4i32, MVT::v4f32, 1 },
{ ISD::FP_TO_UINT, MVT::v4i32, MVT::v4f32, 1 },
{ ISD::FP_TO_SINT, MVT::v4i8, MVT::v4f32, 3 },
{ ISD::FP_TO_UINT, MVT::v4i8, MVT::v4f32, 3 },
{ ISD::FP_TO_SINT, MVT::v4i16, MVT::v4f32, 2 },
{ ISD::FP_TO_UINT, MVT::v4i16, MVT::v4f32, 2 },
{ ISD::SINT_TO_FP, MVT::v2f64, MVT::v2i32, 2 },
{ ISD::UINT_TO_FP, MVT::v2f64, MVT::v2i32, 2 },
{ ISD::SINT_TO_FP, MVT::v2f64, MVT::v2i8, 4 },
{ ISD::UINT_TO_FP, MVT::v2f64, MVT::v2i8, 4 },
{ ISD::SINT_TO_FP, MVT::v2f64, MVT::v2i16, 3 },
{ ISD::UINT_TO_FP, MVT::v2f64, MVT::v2i16, 3 },
{ ISD::SINT_TO_FP, MVT::v2f64, MVT::v2i32, 2 },
{ ISD::UINT_TO_FP, MVT::v2f64, MVT::v2i32, 2 },
{ ISD::FP_TO_SINT, MVT::v2i32, MVT::v2f64, 2 },
{ ISD::FP_TO_UINT, MVT::v2i32, MVT::v2f64, 2 },
{ ISD::FP_TO_SINT, MVT::v8i16, MVT::v8f32, 4 },
{ ISD::FP_TO_UINT, MVT::v8i16, MVT::v8f32, 4 },
{ ISD::FP_TO_SINT, MVT::v16i16, MVT::v16f32, 8 },
{ ISD::FP_TO_UINT, MVT::v16i16, MVT::v16f32, 8 }
};
if (SrcTy.isVector() && ST->hasNEON()) {
int Idx = ConvertCostTableLookup(NEONVectorConversionTbl, ISD,
DstTy.getSimpleVT(), SrcTy.getSimpleVT());
if (Idx != -1)
return NEONVectorConversionTbl[Idx].Cost;
}
static const TypeConversionCostTblEntry<MVT::SimpleValueType>
NEONFloatConversionTbl[] = {
{ ISD::FP_TO_SINT, MVT::i1, MVT::f32, 2 },
{ ISD::FP_TO_UINT, MVT::i1, MVT::f32, 2 },
{ ISD::FP_TO_SINT, MVT::i1, MVT::f64, 2 },
{ ISD::FP_TO_UINT, MVT::i1, MVT::f64, 2 },
{ ISD::FP_TO_SINT, MVT::i8, MVT::f32, 2 },
{ ISD::FP_TO_UINT, MVT::i8, MVT::f32, 2 },
{ ISD::FP_TO_SINT, MVT::i8, MVT::f64, 2 },
{ ISD::FP_TO_UINT, MVT::i8, MVT::f64, 2 },
{ ISD::FP_TO_SINT, MVT::i16, MVT::f32, 2 },
{ ISD::FP_TO_UINT, MVT::i16, MVT::f32, 2 },
{ ISD::FP_TO_SINT, MVT::i16, MVT::f64, 2 },
{ ISD::FP_TO_UINT, MVT::i16, MVT::f64, 2 },
{ ISD::FP_TO_SINT, MVT::i32, MVT::f32, 2 },
{ ISD::FP_TO_UINT, MVT::i32, MVT::f32, 2 },
{ ISD::FP_TO_SINT, MVT::i32, MVT::f64, 2 },
{ ISD::FP_TO_UINT, MVT::i32, MVT::f64, 2 },
{ ISD::FP_TO_SINT, MVT::i64, MVT::f32, 10 },
{ ISD::FP_TO_UINT, MVT::i64, MVT::f32, 10 },
{ ISD::FP_TO_SINT, MVT::i64, MVT::f64, 10 },
{ ISD::FP_TO_UINT, MVT::i64, MVT::f64, 10 }
};
if (SrcTy.isFloatingPoint() && ST->hasNEON()) {
int Idx = ConvertCostTableLookup(NEONFloatConversionTbl, ISD,
DstTy.getSimpleVT(), SrcTy.getSimpleVT());
if (Idx != -1)
return NEONFloatConversionTbl[Idx].Cost;
}
static const TypeConversionCostTblEntry<MVT::SimpleValueType>
NEONIntegerConversionTbl[] = {
{ ISD::SINT_TO_FP, MVT::f32, MVT::i1, 2 },
{ ISD::UINT_TO_FP, MVT::f32, MVT::i1, 2 },
{ ISD::SINT_TO_FP, MVT::f64, MVT::i1, 2 },
{ ISD::UINT_TO_FP, MVT::f64, MVT::i1, 2 },
{ ISD::SINT_TO_FP, MVT::f32, MVT::i8, 2 },
{ ISD::UINT_TO_FP, MVT::f32, MVT::i8, 2 },
{ ISD::SINT_TO_FP, MVT::f64, MVT::i8, 2 },
{ ISD::UINT_TO_FP, MVT::f64, MVT::i8, 2 },
{ ISD::SINT_TO_FP, MVT::f32, MVT::i16, 2 },
{ ISD::UINT_TO_FP, MVT::f32, MVT::i16, 2 },
{ ISD::SINT_TO_FP, MVT::f64, MVT::i16, 2 },
{ ISD::UINT_TO_FP, MVT::f64, MVT::i16, 2 },
{ ISD::SINT_TO_FP, MVT::f32, MVT::i32, 2 },
{ ISD::UINT_TO_FP, MVT::f32, MVT::i32, 2 },
{ ISD::SINT_TO_FP, MVT::f64, MVT::i32, 2 },
{ ISD::UINT_TO_FP, MVT::f64, MVT::i32, 2 },
{ ISD::SINT_TO_FP, MVT::f32, MVT::i64, 10 },
{ ISD::UINT_TO_FP, MVT::f32, MVT::i64, 10 },
{ ISD::SINT_TO_FP, MVT::f64, MVT::i64, 10 },
{ ISD::UINT_TO_FP, MVT::f64, MVT::i64, 10 }
};
if (SrcTy.isInteger() && ST->hasNEON()) {
int Idx = ConvertCostTableLookup(NEONIntegerConversionTbl, ISD,
DstTy.getSimpleVT(), SrcTy.getSimpleVT());
if (Idx != -1)
return NEONIntegerConversionTbl[Idx].Cost;
}
static const TypeConversionCostTblEntry<MVT::SimpleValueType>
ARMIntegerConversionTbl[] = {
{ ISD::SIGN_EXTEND, MVT::i64, MVT::i16, 2 },
{ ISD::TRUNCATE, MVT::i32, MVT::i64, 0 },
{ ISD::TRUNCATE, MVT::i16, MVT::i64, 0 },
{ ISD::TRUNCATE, MVT::i8, MVT::i64, 0 },
{ ISD::TRUNCATE, MVT::i1, MVT::i64, 0 }
};
if (SrcTy.isInteger()) {
int Idx = ConvertCostTableLookup(ARMIntegerConversionTbl, ISD,
DstTy.getSimpleVT(), SrcTy.getSimpleVT());
if (Idx != -1)
return ARMIntegerConversionTbl[Idx].Cost;
}
return TargetTransformInfo::getCastInstrCost(Opcode, Dst, Src);
}
unsigned ARMTTI::getVectorInstrCost(unsigned Opcode, Type *ValTy,
unsigned Index) const {
if (ST->isSwift() &&
Opcode == Instruction::InsertElement &&
ValTy->isVectorTy() &&
ValTy->getScalarSizeInBits() <= 32)
return 3;
return TargetTransformInfo::getVectorInstrCost(Opcode, ValTy, Index);
}
unsigned ARMTTI::getCmpSelInstrCost(unsigned Opcode, Type *ValTy,
Type *CondTy) const {
int ISD = TLI->InstructionOpcodeToISD(Opcode);
if (ST->hasNEON() && ValTy->isVectorTy() && ISD == ISD::SELECT) {
static const TypeConversionCostTblEntry<MVT::SimpleValueType>
NEONVectorSelectTbl[] = {
{ ISD::SELECT, MVT::v16i1, MVT::v16i16, 2*16 + 1 + 3*1 + 4*1 },
{ ISD::SELECT, MVT::v8i1, MVT::v8i32, 4*8 + 1*3 + 1*4 + 1*2 },
{ ISD::SELECT, MVT::v16i1, MVT::v16i32, 4*16 + 1*6 + 1*8 + 1*4 },
{ ISD::SELECT, MVT::v4i1, MVT::v4i64, 4*4 + 1*2 + 1 },
{ ISD::SELECT, MVT::v8i1, MVT::v8i64, 50 },
{ ISD::SELECT, MVT::v16i1, MVT::v16i64, 100 }
};
EVT SelCondTy = TLI->getValueType(CondTy);
EVT SelValTy = TLI->getValueType(ValTy);
if (SelCondTy.isSimple() && SelValTy.isSimple()) {
int Idx = ConvertCostTableLookup(NEONVectorSelectTbl, ISD,
SelCondTy.getSimpleVT(),
SelValTy.getSimpleVT());
if (Idx != -1)
return NEONVectorSelectTbl[Idx].Cost;
}
std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(ValTy);
return LT.first;
}
return TargetTransformInfo::getCmpSelInstrCost(Opcode, ValTy, CondTy);
}
unsigned ARMTTI::getAddressComputationCost(Type *Ty, bool IsComplex) const {
unsigned NumVectorInstToHideOverhead = 10;
if (Ty->isVectorTy() && IsComplex)
return NumVectorInstToHideOverhead;
return 1;
}
unsigned ARMTTI::getShuffleCost(ShuffleKind Kind, Type *Tp, int Index,
Type *SubTp) const {
if (Kind != SK_Reverse && Kind != SK_Alternate)
return TargetTransformInfo::getShuffleCost(Kind, Tp, Index, SubTp);
if (Kind == SK_Reverse) {
static const CostTblEntry<MVT::SimpleValueType> NEONShuffleTbl[] = {
{ISD::VECTOR_SHUFFLE, MVT::v2i32, 1},
{ISD::VECTOR_SHUFFLE, MVT::v2f32, 1},
{ISD::VECTOR_SHUFFLE, MVT::v2i64, 1},
{ISD::VECTOR_SHUFFLE, MVT::v2f64, 1},
{ISD::VECTOR_SHUFFLE, MVT::v4i32, 2},
{ISD::VECTOR_SHUFFLE, MVT::v4f32, 2},
{ISD::VECTOR_SHUFFLE, MVT::v8i16, 2},
{ISD::VECTOR_SHUFFLE, MVT::v16i8, 2}};
std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(Tp);
int Idx = CostTableLookup(NEONShuffleTbl, ISD::VECTOR_SHUFFLE, LT.second);
if (Idx == -1)
return TargetTransformInfo::getShuffleCost(Kind, Tp, Index, SubTp);
return LT.first * NEONShuffleTbl[Idx].Cost;
}
if (Kind == SK_Alternate) {
static const CostTblEntry<MVT::SimpleValueType> NEONAltShuffleTbl[] = {
{ISD::VECTOR_SHUFFLE, MVT::v2f32, 1},
{ISD::VECTOR_SHUFFLE, MVT::v2i64, 1},
{ISD::VECTOR_SHUFFLE, MVT::v2f64, 1},
{ISD::VECTOR_SHUFFLE, MVT::v2i32, 1},
{ISD::VECTOR_SHUFFLE, MVT::v4i32, 2},
{ISD::VECTOR_SHUFFLE, MVT::v4f32, 2},
{ISD::VECTOR_SHUFFLE, MVT::v4i16, 2},
{ISD::VECTOR_SHUFFLE, MVT::v8i16, 16},
{ISD::VECTOR_SHUFFLE, MVT::v16i8, 32}};
std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(Tp);
int Idx =
CostTableLookup(NEONAltShuffleTbl, ISD::VECTOR_SHUFFLE, LT.second);
if (Idx == -1)
return TargetTransformInfo::getShuffleCost(Kind, Tp, Index, SubTp);
return LT.first * NEONAltShuffleTbl[Idx].Cost;
}
return TargetTransformInfo::getShuffleCost(Kind, Tp, Index, SubTp);
}
unsigned ARMTTI::getArithmeticInstrCost(unsigned Opcode, Type *Ty,
OperandValueKind Op1Info,
OperandValueKind Op2Info) const {
int ISDOpcode = TLI->InstructionOpcodeToISD(Opcode);
std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(Ty);
const unsigned FunctionCallDivCost = 20;
const unsigned ReciprocalDivCost = 10;
static const CostTblEntry<MVT::SimpleValueType> CostTbl[] = {
{ ISD::SDIV, MVT::v1i64, 1 * FunctionCallDivCost},
{ ISD::UDIV, MVT::v1i64, 1 * FunctionCallDivCost},
{ ISD::SREM, MVT::v1i64, 1 * FunctionCallDivCost},
{ ISD::UREM, MVT::v1i64, 1 * FunctionCallDivCost},
{ ISD::SDIV, MVT::v2i32, 2 * FunctionCallDivCost},
{ ISD::UDIV, MVT::v2i32, 2 * FunctionCallDivCost},
{ ISD::SREM, MVT::v2i32, 2 * FunctionCallDivCost},
{ ISD::UREM, MVT::v2i32, 2 * FunctionCallDivCost},
{ ISD::SDIV, MVT::v4i16, ReciprocalDivCost},
{ ISD::UDIV, MVT::v4i16, ReciprocalDivCost},
{ ISD::SREM, MVT::v4i16, 4 * FunctionCallDivCost},
{ ISD::UREM, MVT::v4i16, 4 * FunctionCallDivCost},
{ ISD::SDIV, MVT::v8i8, ReciprocalDivCost},
{ ISD::UDIV, MVT::v8i8, ReciprocalDivCost},
{ ISD::SREM, MVT::v8i8, 8 * FunctionCallDivCost},
{ ISD::UREM, MVT::v8i8, 8 * FunctionCallDivCost},
{ ISD::SDIV, MVT::v2i64, 2 * FunctionCallDivCost},
{ ISD::UDIV, MVT::v2i64, 2 * FunctionCallDivCost},
{ ISD::SREM, MVT::v2i64, 2 * FunctionCallDivCost},
{ ISD::UREM, MVT::v2i64, 2 * FunctionCallDivCost},
{ ISD::SDIV, MVT::v4i32, 4 * FunctionCallDivCost},
{ ISD::UDIV, MVT::v4i32, 4 * FunctionCallDivCost},
{ ISD::SREM, MVT::v4i32, 4 * FunctionCallDivCost},
{ ISD::UREM, MVT::v4i32, 4 * FunctionCallDivCost},
{ ISD::SDIV, MVT::v8i16, 8 * FunctionCallDivCost},
{ ISD::UDIV, MVT::v8i16, 8 * FunctionCallDivCost},
{ ISD::SREM, MVT::v8i16, 8 * FunctionCallDivCost},
{ ISD::UREM, MVT::v8i16, 8 * FunctionCallDivCost},
{ ISD::SDIV, MVT::v16i8, 16 * FunctionCallDivCost},
{ ISD::UDIV, MVT::v16i8, 16 * FunctionCallDivCost},
{ ISD::SREM, MVT::v16i8, 16 * FunctionCallDivCost},
{ ISD::UREM, MVT::v16i8, 16 * FunctionCallDivCost},
};
int Idx = -1;
if (ST->hasNEON())
Idx = CostTableLookup(CostTbl, ISDOpcode, LT.second);
if (Idx != -1)
return LT.first * CostTbl[Idx].Cost;
unsigned Cost =
TargetTransformInfo::getArithmeticInstrCost(Opcode, Ty, Op1Info, Op2Info);
if (LT.second == MVT::v2i64 &&
Op2Info == TargetTransformInfo::OK_UniformConstantValue)
Cost += 4;
return Cost;
}
unsigned ARMTTI::getMemoryOpCost(unsigned Opcode, Type *Src, unsigned Alignment,
unsigned AddressSpace) const {
std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(Src);
if (Src->isVectorTy() && Alignment != 16 &&
Src->getVectorElementType()->isDoubleTy()) {
return LT.first * 4;
}
return LT.first;
}