This pass provides access to the codegen interfaces that are needed for IR-level transformations. More...

#include "llvm/Analysis/TargetTransformInfo.h"

Classes
class	Concept

struct	LSRCost

struct	MemCmpExpansionOptions
	If not nullptr, enable inline expansion of memcmp. More...

struct	ReductionFlags
	Flags describing the kind of vector reduction. More...

struct	UnrollingPreferences
	Parameters that control the generic loop unrolling transformation. More...

Public Member Functions
template<typename T >
	TargetTransformInfo (T Impl)
	Construct a TTI object using a type implementing the `Concept` API below. More...

	TargetTransformInfo (const DataLayout &DL)
	Construct a baseline TTI object using a minimal implementation of the `Concept` API below. More...

	TargetTransformInfo (TargetTransformInfo &&Arg)

TargetTransformInfo &	operator= (TargetTransformInfo &&RHS)

	~TargetTransformInfo ()

bool	invalidate (Function &, const PreservedAnalyses &, FunctionAnalysisManager::Invalidator &)
	Handle the invalidation of this information. More...

Generic Target Information
enum	TargetCostKind { TCK_RecipThroughput, TCK_Latency, TCK_CodeSize }
	The kind of cost model. More...

enum	TargetCostConstants { TCC_Free = 0, TCC_Basic = 1, TCC_Expensive = 4 }
	Underlying constants for 'cost' values in this interface. More...

int	getInstructionCost (const Instruction *I, enum TargetCostKind kind) const
	Query the cost of a specified instruction. More...

int	getOperationCost (unsigned Opcode, Type Ty, Type OpTy=nullptr) const
	Estimate the cost of a specific operation when lowered. More...

int	getGEPCost (Type PointeeType, const Value Ptr, ArrayRef< const Value *> Operands) const
	Estimate the cost of a GEP operation when lowered. More...

int	getExtCost (const Instruction I, const Value Src) const
	Estimate the cost of a EXT operation when lowered. More...

int	getCallCost (FunctionType *FTy, int NumArgs=-1) const
	Estimate the cost of a function call when lowered. More...

int	getCallCost (const Function *F, int NumArgs=-1) const
	Estimate the cost of calling a specific function when lowered. More...

int	getCallCost (const Function F, ArrayRef< const Value > Arguments) const
	Estimate the cost of calling a specific function when lowered. More...

unsigned	getInliningThresholdMultiplier () const

int	getIntrinsicCost (Intrinsic::ID IID, Type RetTy, ArrayRef< Type > ParamTys) const
	Estimate the cost of an intrinsic when lowered. More...

int	getIntrinsicCost (Intrinsic::ID IID, Type RetTy, ArrayRef< const Value > Arguments) const
	Estimate the cost of an intrinsic when lowered. More...

unsigned	getEstimatedNumberOfCaseClusters (const SwitchInst &SI, unsigned &JTSize) const

int	getUserCost (const User U, ArrayRef< const Value > Operands) const
	Estimate the cost of a given IR user when lowered. More...

int	getUserCost (const User *U) const
	This is a helper function which calls the two-argument getUserCost with `Operands` which are the current operands U has. More...

bool	hasBranchDivergence () const
	Return true if branch divergence exists. More...

bool	isSourceOfDivergence (const Value *V) const
	Returns whether V is a source of divergence. More...

bool	isAlwaysUniform (const Value *V) const

unsigned	getFlatAddressSpace () const
	Returns the address space ID for a target's 'flat' address space. More...

bool	isLoweredToCall (const Function *F) const
	Test whether calls to a function lower to actual program function calls. More...

void	getUnrollingPreferences (Loop *L, ScalarEvolution &, UnrollingPreferences &UP) const
	Get target-customized preferences for the generic loop unrolling transformation. More...

Scalar Target Information
enum	PopcntSupportKind { PSK_Software, PSK_SlowHardware, PSK_FastHardware }
	Flags indicating the kind of support for population count. More...

bool	isLegalAddImmediate (int64_t Imm) const
	Return true if the specified immediate is legal add immediate, that is the target has add instructions which can add a register with the immediate without having to materialize the immediate into a register. More...

bool	isLegalICmpImmediate (int64_t Imm) const
	Return true if the specified immediate is legal icmp immediate, that is the target has icmp instructions which can compare a register against the immediate without having to materialize the immediate into a register. More...

bool	isLegalAddressingMode (Type Ty, GlobalValue BaseGV, int64_t BaseOffset, bool HasBaseReg, int64_t Scale, unsigned AddrSpace=0, Instruction *I=nullptr) const
	Return true if the addressing mode represented by AM is legal for this target, for a load/store of the specified type. More...

bool	isLSRCostLess (TargetTransformInfo::LSRCost &C1, TargetTransformInfo::LSRCost &C2) const
	Return true if LSR cost of C1 is lower than C1. More...

bool	canMacroFuseCmp () const
	Return true if the target can fuse a compare and branch. More...

bool	shouldFavorPostInc () const

bool	isLegalMaskedStore (Type *DataType) const
	Return true if the target supports masked load/store AVX2 and AVX-512 targets allow masks for consecutive load and store. More...

bool	isLegalMaskedLoad (Type *DataType) const

bool	isLegalMaskedScatter (Type *DataType) const
	Return true if the target supports masked gather/scatter AVX-512 fully supports gather and scatter for vectors with 32 and 64 bits scalar type. More...

bool	isLegalMaskedGather (Type *DataType) const

bool	hasDivRemOp (Type *DataType, bool IsSigned) const
	Return true if the target has a unified operation to calculate division and remainder. More...

bool	hasVolatileVariant (Instruction *I, unsigned AddrSpace) const
	Return true if the given instruction (assumed to be a memory access instruction) has a volatile variant. More...

bool	prefersVectorizedAddressing () const
	Return true if target doesn't mind addresses in vectors. More...

int	getScalingFactorCost (Type Ty, GlobalValue BaseGV, int64_t BaseOffset, bool HasBaseReg, int64_t Scale, unsigned AddrSpace=0) const
	Return the cost of the scaling factor used in the addressing mode represented by AM for this target, for a load/store of the specified type. More...

bool	LSRWithInstrQueries () const
	Return true if the loop strength reduce pass should make Instruction* based TTI queries to isLegalAddressingMode(). More...

bool	isTruncateFree (Type Ty1, Type Ty2) const
	Return true if it's free to truncate a value of type Ty1 to type Ty2. More...

bool	isProfitableToHoist (Instruction *I) const
	Return true if it is profitable to hoist instruction in the then/else to before if. More...

bool	useAA () const

bool	isTypeLegal (Type *Ty) const
	Return true if this type is legal. More...

unsigned	getJumpBufAlignment () const
	Returns the target's jmp_buf alignment in bytes. More...

unsigned	getJumpBufSize () const
	Returns the target's jmp_buf size in bytes. More...

bool	shouldBuildLookupTables () const
	Return true if switches should be turned into lookup tables for the target. More...

bool	shouldBuildLookupTablesForConstant (Constant *C) const
	Return true if switches should be turned into lookup tables containing this constant value for the target. More...

bool	useColdCCForColdCall (Function &F) const
	Return true if the input function which is cold at all call sites, should use coldcc calling convention. More...

unsigned	getScalarizationOverhead (Type *Ty, bool Insert, bool Extract) const

unsigned	getOperandsScalarizationOverhead (ArrayRef< const Value *> Args, unsigned VF) const

bool	supportsEfficientVectorElementLoadStore () const
	If target has efficient vector element load/store instructions, it can return true here so that insertion/extraction costs are not added to the scalarization cost of a load/store. More...

bool	enableAggressiveInterleaving (bool LoopHasReductions) const
	Don't restrict interleaved unrolling to small loops. More...

const MemCmpExpansionOptions *	enableMemCmpExpansion (bool IsZeroCmp) const

bool	enableInterleavedAccessVectorization () const
	Enable matching of interleaved access groups. More...

bool	enableMaskedInterleavedAccessVectorization () const
	Enable matching of interleaved access groups that contain predicated accesses or gaps and therefore vectorized using masked vector loads/stores. More...

bool	isFPVectorizationPotentiallyUnsafe () const
	Indicate that it is potentially unsafe to automatically vectorize floating-point operations because the semantics of vector and scalar floating-point semantics may differ. More...

bool	allowsMisalignedMemoryAccesses (LLVMContext &Context, unsigned BitWidth, unsigned AddressSpace=0, unsigned Alignment=1, bool *Fast=nullptr) const
	Determine if the target supports unaligned memory accesses. More...

PopcntSupportKind	getPopcntSupport (unsigned IntTyWidthInBit) const
	Return hardware support for population count. More...

bool	haveFastSqrt (Type *Ty) const
	Return true if the hardware has a fast square-root instruction. More...

bool	isFCmpOrdCheaperThanFCmpZero (Type *Ty) const
	Return true if it is faster to check if a floating-point value is NaN (or not-NaN) versus a comparison against a constant FP zero value. More...

int	getFPOpCost (Type *Ty) const
	Return the expected cost of supporting the floating point operation of the specified type. More...

int	getIntImmCost (const APInt &Imm, Type *Ty) const
	Return the expected cost of materializing for the given integer immediate of the specified type. More...

int	getIntImmCost (unsigned Opc, unsigned Idx, const APInt &Imm, Type *Ty) const
	Return the expected cost of materialization for the given integer immediate of the specified type for a given instruction. More...

int	getIntImmCost (Intrinsic::ID IID, unsigned Idx, const APInt &Imm, Type *Ty) const

int	getIntImmCodeSizeCost (unsigned Opc, unsigned Idx, const APInt &Imm, Type *Ty) const
	Return the expected cost for the given integer when optimising for size. More...

Vector Target Information
enum	ShuffleKind { SK_Broadcast, SK_Reverse, SK_Select, SK_Transpose, SK_InsertSubvector, SK_ExtractSubvector, SK_PermuteTwoSrc, SK_PermuteSingleSrc }
	The various kinds of shuffle patterns for vector queries. More...

enum	OperandValueKind { OK_AnyValue, OK_UniformValue, OK_UniformConstantValue, OK_NonUniformConstantValue }
	Additional information about an operand's possible values. More...

enum	OperandValueProperties { OP_None = 0, OP_PowerOf2 = 1 }
	Additional properties of an operand's values. More...

enum	CacheLevel { CacheLevel::L1D, CacheLevel::L2D }
	The possible cache levels. More...

enum	MemIndexedMode { MIM_Unindexed, MIM_PreInc, MIM_PreDec, MIM_PostInc, MIM_PostDec }
	The type of load/store indexing. More...

unsigned	getNumberOfRegisters (bool Vector) const

unsigned	getRegisterBitWidth (bool Vector) const

unsigned	getMinVectorRegisterBitWidth () const

bool	shouldMaximizeVectorBandwidth (bool OptSize) const

unsigned	getMinimumVF (unsigned ElemWidth) const

bool	shouldConsiderAddressTypePromotion (const Instruction &I, bool &AllowPromotionWithoutCommonHeader) const

unsigned	getCacheLineSize () const

llvm::Optional< unsigned >	getCacheSize (CacheLevel Level) const

llvm::Optional< unsigned >	getCacheAssociativity (CacheLevel Level) const

unsigned	getPrefetchDistance () const

unsigned	getMinPrefetchStride () const

unsigned	getMaxPrefetchIterationsAhead () const

unsigned	getMaxInterleaveFactor (unsigned VF) const

int	getArithmeticInstrCost (unsigned Opcode, Type Ty, OperandValueKind Opd1Info=OK_AnyValue, OperandValueKind Opd2Info=OK_AnyValue, OperandValueProperties Opd1PropInfo=OP_None, OperandValueProperties Opd2PropInfo=OP_None, ArrayRef< const Value > Args=ArrayRef< const Value *>()) const
	This is an approximation of reciprocal throughput of a math/logic op. More...

int	getShuffleCost (ShuffleKind Kind, Type Tp, int Index=0, Type SubTp=nullptr) const

int	getCastInstrCost (unsigned Opcode, Type Dst, Type Src, const Instruction *I=nullptr) const

int	getExtractWithExtendCost (unsigned Opcode, Type Dst, VectorType VecTy, unsigned Index=-1) const

int	getCFInstrCost (unsigned Opcode) const

int	getCmpSelInstrCost (unsigned Opcode, Type ValTy, Type CondTy=nullptr, const Instruction *I=nullptr) const

int	getVectorInstrCost (unsigned Opcode, Type *Val, unsigned Index=-1) const

int	getMemoryOpCost (unsigned Opcode, Type Src, unsigned Alignment, unsigned AddressSpace, const Instruction I=nullptr) const

int	getMaskedMemoryOpCost (unsigned Opcode, Type *Src, unsigned Alignment, unsigned AddressSpace) const

int	getGatherScatterOpCost (unsigned Opcode, Type DataTy, Value Ptr, bool VariableMask, unsigned Alignment) const

int	getInterleavedMemoryOpCost (unsigned Opcode, Type *VecTy, unsigned Factor, ArrayRef< unsigned > Indices, unsigned Alignment, unsigned AddressSpace, bool UseMaskForCond=false, bool UseMaskForGaps=false) const

int	getArithmeticReductionCost (unsigned Opcode, Type *Ty, bool IsPairwiseForm) const
	Calculate the cost of performing a vector reduction. More...

int	getMinMaxReductionCost (Type Ty, Type CondTy, bool IsPairwiseForm, bool IsUnsigned) const

int	getIntrinsicInstrCost (Intrinsic::ID ID, Type RetTy, ArrayRef< Value > Args, FastMathFlags FMF, unsigned VF=1) const

int	getIntrinsicInstrCost (Intrinsic::ID ID, Type RetTy, ArrayRef< Type > Tys, FastMathFlags FMF, unsigned ScalarizationCostPassed=UINT_MAX) const

int	getCallInstrCost (Function F, Type RetTy, ArrayRef< Type *> Tys) const

unsigned	getNumberOfParts (Type *Tp) const

int	getAddressComputationCost (Type Ty, ScalarEvolution SE=nullptr, const SCEV *Ptr=nullptr) const

unsigned	getCostOfKeepingLiveOverCall (ArrayRef< Type *> Tys) const

bool	getTgtMemIntrinsic (IntrinsicInst *Inst, MemIntrinsicInfo &Info) const

unsigned	getAtomicMemIntrinsicMaxElementSize () const

Value *	getOrCreateResultFromMemIntrinsic (IntrinsicInst Inst, Type ExpectedType) const

Type *	getMemcpyLoopLoweringType (LLVMContext &Context, Value *Length, unsigned SrcAlign, unsigned DestAlign) const

void	getMemcpyLoopResidualLoweringType (SmallVectorImpl< Type *> &OpsOut, LLVMContext &Context, unsigned RemainingBytes, unsigned SrcAlign, unsigned DestAlign) const

bool	areInlineCompatible (const Function Caller, const Function Callee) const

bool	areFunctionArgsABICompatible (const Function Caller, const Function Callee, SmallPtrSetImpl< Argument *> &Args) const

bool	isIndexedLoadLegal (enum MemIndexedMode Mode, Type *Ty) const

bool	isIndexedStoreLegal (enum MemIndexedMode Mode, Type *Ty) const

unsigned	getLoadStoreVecRegBitWidth (unsigned AddrSpace) const

bool	isLegalToVectorizeLoad (LoadInst *LI) const

bool	isLegalToVectorizeStore (StoreInst *SI) const

bool	isLegalToVectorizeLoadChain (unsigned ChainSizeInBytes, unsigned Alignment, unsigned AddrSpace) const

bool	isLegalToVectorizeStoreChain (unsigned ChainSizeInBytes, unsigned Alignment, unsigned AddrSpace) const

unsigned	getLoadVectorFactor (unsigned VF, unsigned LoadSize, unsigned ChainSizeInBytes, VectorType *VecTy) const

unsigned	getStoreVectorFactor (unsigned VF, unsigned StoreSize, unsigned ChainSizeInBytes, VectorType *VecTy) const

bool	useReductionIntrinsic (unsigned Opcode, Type *Ty, ReductionFlags Flags) const

bool	shouldExpandReduction (const IntrinsicInst *II) const

static OperandValueKind	getOperandInfo (Value *V, OperandValueProperties &OpProps)
	Collect properties of V used in cost analysis, e.g. OP_PowerOf2. More...

Detailed Description

This pass provides access to the codegen interfaces that are needed for IR-level transformations.

Definition at line 78 of file TargetTransformInfo.h.

Member Enumeration Documentation

◆ CacheLevel

enum llvm::TargetTransformInfo::CacheLevel

strong

The possible cache levels.

Enumerator
L1D
L2D

Definition at line 717 of file TargetTransformInfo.h.

◆ MemIndexedMode

enum llvm::TargetTransformInfo::MemIndexedMode

The type of load/store indexing.

Enumerator
MIM_Unindexed	No indexing.
MIM_PreInc	Pre-incrementing.
MIM_PreDec	Pre-decrementing.
MIM_PostInc	Post-incrementing.
MIM_PostDec	Post-decrementing.

Definition at line 946 of file TargetTransformInfo.h.

◆ OperandValueKind

enum llvm::TargetTransformInfo::OperandValueKind

Additional information about an operand's possible values.

Enumerator
OK_AnyValue
OK_UniformValue
OK_UniformConstantValue
OK_NonUniformConstantValue

Definition at line 673 of file TargetTransformInfo.h.

◆ OperandValueProperties

enum llvm::TargetTransformInfo::OperandValueProperties

Additional properties of an operand's values.

Enumerator
OP_None
OP_PowerOf2

Definition at line 681 of file TargetTransformInfo.h.

◆ PopcntSupportKind

enum llvm::TargetTransformInfo::PopcntSupportKind

Flags indicating the kind of support for population count.

Compared to the SW implementation, HW support is supposed to significantly boost the performance when the population is dense, and it may or may not degrade performance if the population is sparse. A HW support is considered as "Fast" if it can outperform, or is on a par with, SW implementation when the population is sparse; otherwise, it is considered as "Slow".

Enumerator
PSK_Software
PSK_SlowHardware
PSK_FastHardware

Definition at line 453 of file TargetTransformInfo.h.

◆ ShuffleKind

enum llvm::TargetTransformInfo::ShuffleKind

The various kinds of shuffle patterns for vector queries.

Enumerator
SK_Broadcast	Broadcast element 0 to all other elements.
SK_Reverse	Reverse the order of the vector.
SK_Select	Selects elements from the corresponding lane of either source operand. This is equivalent to a vector select with a constant condition operand.
SK_Transpose	Transpose two vectors.
SK_InsertSubvector	InsertSubvector. Index indicates start offset.
SK_ExtractSubvector	ExtractSubvector Index indicates start offset.
SK_PermuteTwoSrc	Merge elements from two source vectors into one with any shuffle mask.
SK_PermuteSingleSrc	Shuffle elements of single source vector with any shuffle mask.

Definition at line 657 of file TargetTransformInfo.h.

◆ TargetCostConstants

enum llvm::TargetTransformInfo::TargetCostConstants

Underlying constants for 'cost' values in this interface.

Many APIs in this interface return a cost. This enum defines the fundamental values that should be used to interpret (and produce) those costs. The costs are returned as an int rather than a member of this enumeration because it is expected that the cost of one IR instruction may have a multiplicative factor to it or otherwise won't fit directly into the enum. Moreover, it is common to sum or average costs which works better as simple integral values. Thus this enum only provides constants. Also note that the returned costs are signed integers to make it natural to add, subtract, and test with zero (a common boundary condition). It is not expected that 2^32 is a realistic cost to be modeling at any point.

Note that these costs should usually reflect the intersection of code-size cost and execution cost. A free instruction is typically one that folds into another instruction. For example, reg-to-reg moves can often be skipped by renaming the registers in the CPU, but they still are encoded and thus wouldn't be considered 'free' here.

Enumerator
TCC_Free	Expected to fold away in lowering.
TCC_Basic	The cost of a typical 'add' instruction.
TCC_Expensive	The cost of a 'div' instruction on x86.

Definition at line 166 of file TargetTransformInfo.h.

◆ TargetCostKind

enum llvm::TargetTransformInfo::TargetCostKind

The kind of cost model.

There are several different cost models that can be customized by the target. The normalization of each cost model may be target specific.

Enumerator
TCK_RecipThroughput	Reciprocal throughput.
TCK_Latency	The latency of instruction.
TCK_CodeSize	Instruction code size.

Definition at line 121 of file TargetTransformInfo.h.

Constructor & Destructor Documentation

◆ TargetTransformInfo() [1/3]

template<typename T >

llvm::TargetTransformInfo::TargetTransformInfo ( T Impl )

Construct a TTI object using a type implementing the Concept API below.

This is used by targets to construct a TTI wrapping their target-specific implementaion that encodes appropriate costs for their target.

Definition at line 1626 of file TargetTransformInfo.h.

◆ TargetTransformInfo() [2/3]

TargetTransformInfo::TargetTransformInfo ( const DataLayout & DL )

explicit

Construct a baseline TTI object using a minimal implementation of the Concept API below.

The TTI implementation will reflect the information in the DataLayout provided if non-null.

Definition at line 44 of file TargetTransformInfo.cpp.

◆ TargetTransformInfo() [3/3]

TargetTransformInfo::TargetTransformInfo ( TargetTransformInfo && Arg )

Definition at line 49 of file TargetTransformInfo.cpp.

◆ ~TargetTransformInfo()

TargetTransformInfo::~TargetTransformInfo ( )

Definition at line 47 of file TargetTransformInfo.cpp.

Member Function Documentation

◆ allowsMisalignedMemoryAccesses()

bool TargetTransformInfo::allowsMisalignedMemoryAccesses	(	LLVMContext &	Context,
		unsigned	BitWidth,
		unsigned	AddressSpace = `0`,
		unsigned	Alignment = `1`,
		bool *	Fast = `nullptr`
	)		const

Determine if the target supports unaligned memory accesses.

Definition at line 279 of file TargetTransformInfo.cpp.

◆ areFunctionArgsABICompatible()

bool TargetTransformInfo::areFunctionArgsABICompatible	(	const Function *	Caller,
		const Function *	Callee,
		SmallPtrSetImpl< Argument *> &	Args
	)		const

Returns: True if the caller and callee agree on how Args will be passed to the callee.

Parameters

[out] Args The list of compatible arguments. The implementation may filter out any incompatible args from this list.

Definition at line 628 of file TargetTransformInfo.cpp.

Referenced by areFunctionArgsABICompatible().

◆ areInlineCompatible()

bool TargetTransformInfo::areInlineCompatible	(	const Function *	Caller,
		const Function *	Callee
	)		const

Returns: True if the two functions have compatible attributes for inlining purposes.

Definition at line 623 of file TargetTransformInfo.cpp.

Referenced by functionsHaveCompatibleAttributes(), and isWrapperOnlyCall().

◆ canMacroFuseCmp()

bool TargetTransformInfo::canMacroFuseCmp ( ) const

Return true if the target can fuse a compare and branch.

Loop-strength-reduction (LSR) uses that knowledge to adjust its cost calculation for the instructions in a loop.

Definition at line 158 of file TargetTransformInfo.cpp.

Referenced by DeleteTriviallyDeadInstructions().

◆ enableAggressiveInterleaving()

bool TargetTransformInfo::enableAggressiveInterleaving ( bool LoopHasReductions ) const

Don't restrict interleaved unrolling to small loops.

Definition at line 258 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::selectInterleaveCount().

◆ enableInterleavedAccessVectorization()

bool TargetTransformInfo::enableInterleavedAccessVectorization ( ) const

Enable matching of interleaved access groups.

Definition at line 267 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizePass::processLoop().

◆ enableMaskedInterleavedAccessVectorization()

bool TargetTransformInfo::enableMaskedInterleavedAccessVectorization ( ) const

Enable matching of interleaved access groups that contain predicated accesses or gaps and therefore vectorized using masked vector loads/stores.

Definition at line 271 of file TargetTransformInfo.cpp.

Referenced by useMaskedInterleavedAccesses().

◆ enableMemCmpExpansion()

const TargetTransformInfo::MemCmpExpansionOptions * TargetTransformInfo::enableMemCmpExpansion ( bool IsZeroCmp ) const

Definition at line 263 of file TargetTransformInfo.cpp.

◆ getAddressComputationCost()

int TargetTransformInfo::getAddressComputationCost	(	Type *	Ty,
		ScalarEvolution *	SE = `nullptr`,
		const SCEV *	Ptr = `nullptr`
	)		const

Returns: The cost of the address computation. For most targets this can be merged into the instruction indexing mode. Some targets might want to distinguish between address computation for memory operations on vector types and scalar types. Such targets should override this function. The 'SE' parameter holds pointer for the scalar evolution object which is used in order to get the Ptr step value in case of constant stride. The 'Ptr' parameter holds SCEV of the access pointer.

Definition at line 565 of file TargetTransformInfo.cpp.

References assert().

Referenced by chainToBasePointerCost(), and isStrideMul().

◆ getArithmeticInstrCost()

int TargetTransformInfo::getArithmeticInstrCost	(	unsigned	Opcode,
		Type *	Ty,
		OperandValueKind	Opd1Info = `OK_AnyValue`,
		OperandValueKind	Opd2Info = `OK_AnyValue`,
		OperandValueProperties	Opd1PropInfo = `OP_None`,
		OperandValueProperties	Opd2PropInfo = `OP_None`,
		ArrayRef< const Value *>	Args = `ArrayRef<const Value *>()`
	)		const

This is an approximation of reciprocal throughput of a math/logic op.

A higher cost indicates less expected throughput. From Agner Fog's guides, reciprocal throughput is "the average number of clock cycles per instruction when the instructions are not part of a limiting dependency chain." Therefore, costs should be scaled to account for multiple execution units on the target that can process this type of instruction. For example, if there are 5 scalar integer units and 2 vector integer units that can calculate an 'add' in a single cycle, this model should indicate that the cost of the vector add instruction is 2.5 times the cost of the scalar add instruction. Args is an optional argument which holds the instruction operands values so the TTI can analyze those values searching for special cases or optimizations based on those values.

Definition at line 440 of file TargetTransformInfo.cpp.

References assert().

Referenced by createRdxShuffleMask(), isBroadcastShuffle(), matchVectorSplittingReduction(), llvm::LoopVectorizationCostModel::setCostBasedWideningDecision(), and visitIVCast().

◆ getArithmeticReductionCost()

int TargetTransformInfo::getArithmeticReductionCost	(	unsigned	Opcode,
		Type *	Ty,
		bool	IsPairwiseForm
	)		const

Calculate the cost of performing a vector reduction.

This is the cost of reducing the vector value of type Ty to a scalar value using the operation denoted by Opcode. The form of the reduction can either be a pairwise reduction or a reduction that splits the vector at every reduction level.

Pairwise: (v0, v1, v2, v3) ((v0+v1), (v2+v3), undef, undef) Split: (v0, v1, v2, v3) ((v0+v2), (v1+v3), undef, undef)

Definition at line 573 of file TargetTransformInfo.cpp.

References assert().

Referenced by createRdxShuffleMask(), and matchVectorSplittingReduction().

◆ getAtomicMemIntrinsicMaxElementSize()

unsigned TargetTransformInfo::getAtomicMemIntrinsicMaxElementSize ( ) const

Returns: The maximum element size, in bytes, for an element unordered-atomic memory intrinsic.

Definition at line 599 of file TargetTransformInfo.cpp.

◆ getCacheAssociativity()

llvm::Optional< unsigned > TargetTransformInfo::getCacheAssociativity ( CacheLevel Level ) const

Returns: The associativity of the cache level, if available.

Definition at line 370 of file TargetTransformInfo.cpp.

◆ getCacheLineSize()

unsigned TargetTransformInfo::getCacheLineSize ( ) const

Returns: The size of a cache line in bytes.

Definition at line 361 of file TargetTransformInfo.cpp.

◆ getCacheSize()

llvm::Optional< unsigned > TargetTransformInfo::getCacheSize ( CacheLevel Level ) const

Returns: The size of the cache level in bytes, if available.

Definition at line 365 of file TargetTransformInfo.cpp.

◆ getCallCost() [1/3]

int TargetTransformInfo::getCallCost	(	FunctionType *	FTy,
		int	NumArgs = `-1`
	)		const

Estimate the cost of a function call when lowered.

The contract for this is the same as getOperationCost except that it supports an interface that provides extra information specific to call instructions.

This is the most basic query for estimating call cost: it only knows the function type and (potentially) the number of arguments at the call site. The latter is only interesting for varargs function types.

Definition at line 64 of file TargetTransformInfo.cpp.

References assert().

◆ getCallCost() [2/3]

int llvm::TargetTransformInfo::getCallCost	(	const Function *	F,
		int	NumArgs = `-1`
	)		const

Estimate the cost of calling a specific function when lowered.

This overload adds the ability to reason about the particular function being called in the event it is a library call with special lowering.

◆ getCallCost() [3/3]

int TargetTransformInfo::getCallCost	(	const Function *	F,
		ArrayRef< const Value *>	Arguments
	)		const

Estimate the cost of calling a specific function when lowered.

This overload allows specifying a set of candidate argument values.

Definition at line 70 of file TargetTransformInfo.cpp.

References assert().

◆ getCallInstrCost()

int TargetTransformInfo::getCallInstrCost	(	Function *	F,
		Type *	RetTy,
		ArrayRef< Type *>	Tys
	)		const

Returns: The cost of Call instructions.

Definition at line 554 of file TargetTransformInfo.cpp.

References assert().

Referenced by getVectorCallCost().

◆ getCastInstrCost()

int TargetTransformInfo::getCastInstrCost	(	unsigned	Opcode,
		Type *	Dst,
		Type *	Src,
		const Instruction *	I = `nullptr`
	)		const

Returns: The expected cost of cast instructions, such as bitcast, trunc, zext, etc. If there is an existing instruction that holds Opcode, it may be passed in the 'I' parameter.

Definition at line 458 of file TargetTransformInfo.cpp.

References assert(), and llvm::Instruction::getOpcode().

Referenced by chainToBasePointerCost(), matchVectorSplittingReduction(), and llvm::LoopVectorizationCostModel::setCostBasedWideningDecision().

◆ getCFInstrCost()

int TargetTransformInfo::getCFInstrCost ( unsigned Opcode ) const

Returns: The expected cost of control-flow related instructions such as Phi, Ret, Br.

Definition at line 475 of file TargetTransformInfo.cpp.

References assert().

Referenced by llvm::LoopVectorizationCostModel::collectInstsToScalarize(), matchVectorSplittingReduction(), and llvm::LoopVectorizationCostModel::setCostBasedWideningDecision().

◆ getCmpSelInstrCost()

int TargetTransformInfo::getCmpSelInstrCost	(	unsigned	Opcode,
		Type *	ValTy,
		Type *	CondTy = `nullptr`,
		const Instruction *	I = `nullptr`
	)		const

Returns: The expected cost of compare and select instructions. If there is an existing instruction that holds Opcode, it may be passed in the 'I' parameter.

Definition at line 481 of file TargetTransformInfo.cpp.

References assert(), and llvm::Instruction::getOpcode().

Referenced by createRdxShuffleMask(), matchVectorSplittingReduction(), and llvm::LoopVectorizationCostModel::setCostBasedWideningDecision().

◆ getCostOfKeepingLiveOverCall()

unsigned TargetTransformInfo::getCostOfKeepingLiveOverCall ( ArrayRef< Type *> Tys ) const

Returns: The cost, if any, of keeping values of the given types alive over a callsite.

Some types may require the use of register classes that do not have any callee-saved registers, so would require a spill and fill.

Definition at line 590 of file TargetTransformInfo.cpp.

◆ getEstimatedNumberOfCaseClusters()

unsigned TargetTransformInfo::getEstimatedNumberOfCaseClusters	(	const SwitchInst &	SI,
		unsigned &	JTSize
	)		const

Returns: The estimated number of case clusters when lowering 'SI'. JTSize Set a jump table size only when SI is suitable for a jump table.

Definition at line 99 of file TargetTransformInfo.cpp.

◆ getExtCost()

int TargetTransformInfo::getExtCost	(	const Instruction *	I,
		const Value *	Src
	)		const

Estimate the cost of a EXT operation when lowered.

The contract for this function is the same as getOperationCost except that it supports an interface that provides extra information specific to the EXT operation.

Definition at line 86 of file TargetTransformInfo.cpp.

◆ getExtractWithExtendCost()

int TargetTransformInfo::getExtractWithExtendCost	(	unsigned	Opcode,
		Type *	Dst,
		VectorType *	VecTy,
		unsigned	Index = `-1`
	)		const

Returns: The expected cost of a sign- or zero-extended vector extract. Use -1 to indicate that there is no information about the index value.

Definition at line 467 of file TargetTransformInfo.cpp.

References assert().

◆ getFlatAddressSpace()

unsigned TargetTransformInfo::getFlatAddressSpace ( ) const

Returns the address space ID for a target's 'flat' address space.

Note this is not necessarily the same as addrspace(0), which LLVM sometimes refers to as the generic address space. The flat address space is a generic address space that can be used access multiple segments of memory with different address spaces. Access of a memory location through a pointer with this address space is expected to be legal but slower compared to the same memory location accessed through a pointer with a different address space. This is for targets with different pointer representations which can be converted with the addrspacecast instruction. If a pointer is converted to this address space, optimizations should attempt to replace the access with the source address space.

Returns: ~0u if the target does not have such a flat address space to optimize away.

Definition at line 123 of file TargetTransformInfo.cpp.

Referenced by cloneConstantExprWithNewAddressSpace().

◆ getFPOpCost()

int TargetTransformInfo::getFPOpCost ( Type * Ty ) const

Return the expected cost of supporting the floating point operation of the specified type.

Definition at line 301 of file TargetTransformInfo.cpp.

References assert().

◆ getGatherScatterOpCost()

int TargetTransformInfo::getGatherScatterOpCost	(	unsigned	Opcode,
		Type *	DataTy,
		Value *	Ptr,
		bool	VariableMask,
		unsigned	Alignment
	)		const

Returns: The cost of Gather or Scatter operation Opcode - is a type of memory access Load or Store DataTy - a vector type of the data to be loaded or stored Ptr - pointer [or vector of pointers] - address[es] in memory VariableMask - true when the memory access is predicated with a mask that is not a compile-time constant Alignment - alignment of single element

Definition at line 517 of file TargetTransformInfo.cpp.

References assert().

Referenced by isStrideMul().

◆ getGEPCost()

int TargetTransformInfo::getGEPCost	(	Type *	PointeeType,
		const Value *	Ptr,
		ArrayRef< const Value *>	Operands
	)		const

Estimate the cost of a GEP operation when lowered.

The contract for this function is the same as getOperationCost except that it supports an interface that provides extra information specific to the GEP operation.

Definition at line 81 of file TargetTransformInfo.cpp.

Referenced by isGEPFoldable().

◆ getInliningThresholdMultiplier()

unsigned TargetTransformInfo::getInliningThresholdMultiplier ( ) const

Returns: A value by which our inlining threshold should be multiplied. This is primarily used to bump up the inlining threshold wholesale on targets where calls are unusually expensive.

TODO: This is a rather blunt instrument. Perhaps altering the costs of individual classes of instructions would be better.

Definition at line 77 of file TargetTransformInfo.cpp.

◆ getInstructionCost()

int llvm::TargetTransformInfo::getInstructionCost	(	const Instruction *	I,
		enum TargetCostKind	kind
	)		const

inline

Query the cost of a specified instruction.

Clients should use this interface to query the cost of an existing instruction. The instruction must have a valid parent (basic block).

Note, this method does not cache the cost calculation and it can be expensive in some cases.

Definition at line 134 of file TargetTransformInfo.h.

References llvm_unreachable.

Referenced by canSplitCallSite().

◆ getInterleavedMemoryOpCost()

int TargetTransformInfo::getInterleavedMemoryOpCost	(	unsigned	Opcode,
		Type *	VecTy,
		unsigned	Factor,
		ArrayRef< unsigned >	Indices,
		unsigned	Alignment,
		unsigned	AddressSpace,
		bool	UseMaskForCond = `false`,
		bool	UseMaskForGaps = `false`
	)		const

Returns: The cost of the interleaved memory operation. Opcode is the memory operation code VecTy is the vector type of the interleaved access. Factor is the interleave factor Indices is the indices for interleaved load members (as interleaved load allows gaps) Alignment is the alignment of the memory operation AddressSpace is address space of the pointer. UseMaskForCond indicates if the memory access is predicated. UseMaskForGaps indicates if gaps should be masked.

Definition at line 526 of file TargetTransformInfo.cpp.

References assert().

Referenced by isStrideMul().

◆ getIntImmCodeSizeCost()

int TargetTransformInfo::getIntImmCodeSizeCost	(	unsigned	Opc,
		unsigned	Idx,
		const APInt &	Imm,
		Type *	Ty
	)		const

Return the expected cost for the given integer when optimising for size.

This is different than the other integer immediate cost functions in that it is subtarget agnostic. This is useful when you e.g. target one ISA such as Aarch32 but smaller encodings could be possible with another such as Thumb. This return value is used as a penalty when the total costs for a constant is calculated (the bigger the cost, the more beneficial constant hoisting is).

Definition at line 307 of file TargetTransformInfo.cpp.

References assert().

◆ getIntImmCost() [1/3]

int TargetTransformInfo::getIntImmCost	(	const APInt &	Imm,
		Type *	Ty
	)		const

Return the expected cost of materializing for the given integer immediate of the specified type.

Definition at line 315 of file TargetTransformInfo.cpp.

References assert().

Referenced by isSafeAndProfitableToSpeculateAroundPHI(), and tryUnmergingGEPsAcrossIndirectBr().

◆ getIntImmCost() [2/3]

int TargetTransformInfo::getIntImmCost	(	unsigned	Opc,
		unsigned	Idx,
		const APInt &	Imm,
		Type *	Ty
	)		const

Return the expected cost of materialization for the given integer immediate of the specified type for a given instruction.

The cost can be zero if the immediate can be folded into the specified instruction.

Definition at line 321 of file TargetTransformInfo.cpp.

References assert().

◆ getIntImmCost() [3/3]

int TargetTransformInfo::getIntImmCost	(	Intrinsic::ID	IID,
		unsigned	Idx,
		const APInt &	Imm,
		Type *	Ty
	)		const

Definition at line 328 of file TargetTransformInfo.cpp.

References assert().

◆ getIntrinsicCost() [1/2]

int llvm::TargetTransformInfo::getIntrinsicCost	(	Intrinsic::ID	IID,
		Type *	RetTy,
		ArrayRef< Type *>	ParamTys
	)		const

Estimate the cost of an intrinsic when lowered.

Mirrors the getCallCost method but uses an intrinsic identifier.

Referenced by detectShiftUntilZeroIdiom().

◆ getIntrinsicCost() [2/2]

int TargetTransformInfo::getIntrinsicCost	(	Intrinsic::ID	IID,
		Type *	RetTy,
		ArrayRef< const Value *>	Arguments
	)		const

Estimate the cost of an intrinsic when lowered.

Mirrors the getCallCost method but uses an intrinsic identifier.

Definition at line 91 of file TargetTransformInfo.cpp.

References assert().

◆ getIntrinsicInstrCost() [1/2]

int TargetTransformInfo::getIntrinsicInstrCost	(	Intrinsic::ID	ID,
		Type *	RetTy,
		ArrayRef< Value *>	Args,
		FastMathFlags	FMF,
		unsigned	VF = `1`
	)		const

Returns

The cost of Intrinsic instructions. Analyses the real arguments. Three cases are handled: 1. scalar instruction 2. vector instruction

scalar instruction which is to be vectorized with VF.

Definition at line 547 of file TargetTransformInfo.cpp.

References assert().

Referenced by getVectorIntrinsicCost(), and matchVectorSplittingReduction().

◆ getIntrinsicInstrCost() [2/2]

int TargetTransformInfo::getIntrinsicInstrCost	(	Intrinsic::ID	ID,
		Type *	RetTy,
		ArrayRef< Type *>	Tys,
		FastMathFlags	FMF,
		unsigned	ScalarizationCostPassed = `UINT_MAX`
	)		const

Returns: The cost of Intrinsic instructions. Types analysis only. If ScalarizationCostPassed is UINT_MAX, the cost of scalarizing the arguments and the return value will be computed based on types.

Definition at line 538 of file TargetTransformInfo.cpp.

References assert().

◆ getJumpBufAlignment()

unsigned TargetTransformInfo::getJumpBufAlignment ( ) const

Returns the target's jmp_buf alignment in bytes.

Definition at line 224 of file TargetTransformInfo.cpp.

◆ getJumpBufSize()

unsigned TargetTransformInfo::getJumpBufSize ( ) const

Returns the target's jmp_buf size in bytes.

Definition at line 228 of file TargetTransformInfo.cpp.

◆ getLoadStoreVecRegBitWidth()

unsigned TargetTransformInfo::getLoadStoreVecRegBitWidth ( unsigned AddrSpace ) const

Returns: The bitwidth of the largest vector type that should be used to load/store in the given address space.

Definition at line 644 of file TargetTransformInfo.cpp.

◆ getLoadVectorFactor()

unsigned TargetTransformInfo::getLoadVectorFactor	(	unsigned	VF,
		unsigned	LoadSize,
		unsigned	ChainSizeInBytes,
		VectorType *	VecTy
	)		const

Returns: The new vector factor value if the target doesn't support SizeInBytes loads or has a better vector factor.

Definition at line 668 of file TargetTransformInfo.cpp.

◆ getMaskedMemoryOpCost()

int TargetTransformInfo::getMaskedMemoryOpCost	(	unsigned	Opcode,
		Type *	Src,
		unsigned	Alignment,
		unsigned	AddressSpace
	)		const

Returns: The cost of masked Load and Store instructions.

Definition at line 508 of file TargetTransformInfo.cpp.

References assert().

Referenced by isStrideMul().

◆ getMaxInterleaveFactor()

unsigned TargetTransformInfo::getMaxInterleaveFactor ( unsigned VF ) const

Returns: The maximum interleave factor that any transform should try to perform for this target. This number depends on the level of parallelism and the number of execution units in the CPU.

Definition at line 387 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizePass::runImpl(), and llvm::LoopVectorizationCostModel::selectInterleaveCount().

◆ getMaxPrefetchIterationsAhead()

unsigned TargetTransformInfo::getMaxPrefetchIterationsAhead ( ) const

Returns: The maximum number of iterations to prefetch ahead. If the required number of iterations is more than this number, no prefetching is performed.

Definition at line 383 of file TargetTransformInfo.cpp.

◆ getMemcpyLoopLoweringType()

Type * TargetTransformInfo::getMemcpyLoopLoweringType	(	LLVMContext &	Context,
		Value *	Length,
		unsigned	SrcAlign,
		unsigned	DestAlign
	)		const

Returns: The type to use in a loop expansion of a memcpy call.

Definition at line 608 of file TargetTransformInfo.cpp.

Referenced by llvm::createMemCpyLoopKnownSize(), and llvm::createMemCpyLoopUnknownSize().

◆ getMemcpyLoopResidualLoweringType()

void TargetTransformInfo::getMemcpyLoopResidualLoweringType	(	SmallVectorImpl< Type *> &	OpsOut,
		LLVMContext &	Context,
		unsigned	RemainingBytes,
		unsigned	SrcAlign,
		unsigned	DestAlign
	)		const

Parameters

[out]	OpsOut	The operand types to copy RemainingBytes of memory.
	RemainingBytes	The number of bytes to copy.

Calculates the operand types to use when copying RemainingBytes of memory, where source and destination alignments are SrcAlign and DestAlign respectively.

Definition at line 616 of file TargetTransformInfo.cpp.

Referenced by llvm::createMemCpyLoopKnownSize().

◆ getMemoryOpCost()

int TargetTransformInfo::getMemoryOpCost	(	unsigned	Opcode,
		Type *	Src,
		unsigned	Alignment,
		unsigned	AddressSpace,
		const Instruction *	I = `nullptr`
	)		const

Returns: The cost of Load and Store instructions.

Definition at line 497 of file TargetTransformInfo.cpp.

References assert(), and llvm::Instruction::getOpcode().

Referenced by isStrideMul(), and matchVectorSplittingReduction().

◆ getMinimumVF()

unsigned TargetTransformInfo::getMinimumVF ( unsigned ElemWidth ) const

Returns: The minimum vectorization factor for types of given element bit width, or 0 if there is no mimimum VF. The returned value only applies when shouldMaximizeVectorBandwidth returns true.

Definition at line 351 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::computeMaxVF().

◆ getMinMaxReductionCost()

int TargetTransformInfo::getMinMaxReductionCost	(	Type *	Ty,
		Type *	CondTy,
		bool	IsPairwiseForm,
		bool	IsUnsigned
	)		const

Definition at line 580 of file TargetTransformInfo.cpp.

References assert().

Referenced by createRdxShuffleMask(), and matchVectorSplittingReduction().

◆ getMinPrefetchStride()

unsigned TargetTransformInfo::getMinPrefetchStride ( ) const

Returns: Some HW prefetchers can handle accesses up to a certain constant stride. This is the minimum stride in bytes where it makes sense to start adding SW prefetches. The default is 1, i.e. prefetch with any stride.

Definition at line 379 of file TargetTransformInfo.cpp.

◆ getMinVectorRegisterBitWidth()

unsigned TargetTransformInfo::getMinVectorRegisterBitWidth ( ) const

Returns: The width of the smallest vector register type.

Definition at line 343 of file TargetTransformInfo.cpp.

◆ getNumberOfParts()

unsigned TargetTransformInfo::getNumberOfParts ( Type * Tp ) const

Returns: The number of pieces into which the provided type must be split during legalization. Zero is returned when the answer is unknown.

Definition at line 561 of file TargetTransformInfo.cpp.

Referenced by isStrideMul().

◆ getNumberOfRegisters()

unsigned TargetTransformInfo::getNumberOfRegisters ( bool Vector ) const

Returns: The number of scalar or vector registers that the target has. If 'Vectors' is true, it returns the number of vector registers. If it is set to false, it returns the number of scalar registers.

Definition at line 335 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::computeMaxVF(), DeleteTriviallyDeadInstructions(), llvm::LoopVectorizePass::runImpl(), and llvm::LoopVectorizationCostModel::selectInterleaveCount().

◆ getOperandInfo()

TargetTransformInfo::OperandValueKind TargetTransformInfo::getOperandInfo	(	Value *	V,
		OperandValueProperties &	OpProps
	)

static

Collect properties of V used in cost analysis, e.g. OP_PowerOf2.

Definition at line 392 of file TargetTransformInfo.cpp.

References E, llvm::getSplatValue(), I, OK_AnyValue, OK_NonUniformConstantValue, OK_UniformConstantValue, OK_UniformValue, OP_None, and OP_PowerOf2.

Referenced by llvm::BasicTTIImplBase< AMDGPUTTIImpl >::getIntrinsicInstrCost(), matchVectorSplittingReduction(), and llvm::LoopVectorizationCostModel::setCostBasedWideningDecision().

◆ getOperandsScalarizationOverhead()

unsigned TargetTransformInfo::getOperandsScalarizationOverhead	(	ArrayRef< const Value *>	Args,
		unsigned	VF
	)		const

Definition at line 249 of file TargetTransformInfo.cpp.

Referenced by getScalarizationOverhead(), and getScalarizationOverhead().

◆ getOperationCost()

int TargetTransformInfo::getOperationCost	(	unsigned	Opcode,
		Type *	Ty,
		Type *	OpTy = `nullptr`
	)		const

Estimate the cost of a specific operation when lowered.

Note that this is designed to work on an arbitrary synthetic opcode, and thus work for hypothetical queries before an instruction has even been formed. However, this does not work for GEPs, and must not be called for a GEP instruction. Instead, use the dedicated getGEPCost interface as analyzing a GEP's cost required more information.

Typically only the result type is required, and the operand type can be omitted. However, if the opcode is one of the cast instructions, the operand type is required.

The returned cost is defined in terms of TargetCostConstants, see its comments for a detailed explanation of the cost values.

Definition at line 57 of file TargetTransformInfo.cpp.

References assert().

◆ getOrCreateResultFromMemIntrinsic()

Value * TargetTransformInfo::getOrCreateResultFromMemIntrinsic	(	IntrinsicInst *	Inst,
		Type *	ExpectedType
	)		const

Returns: A value which is the result of the given memory intrinsic. New instructions may be created to extract the result from the given intrinsic memory operation. Returns nullptr if the target cannot create a result from the given intrinsic.

Definition at line 603 of file TargetTransformInfo.cpp.

◆ getPopcntSupport()

TargetTransformInfo::PopcntSupportKind TargetTransformInfo::getPopcntSupport ( unsigned IntTyWidthInBit ) const

Return hardware support for population count.

Definition at line 289 of file TargetTransformInfo.cpp.

Referenced by detectShiftUntilZeroIdiom().

◆ getPrefetchDistance()

unsigned TargetTransformInfo::getPrefetchDistance ( ) const

Returns: How much before a load we should place the prefetch instruction. This is currently measured in number of instructions.

Definition at line 375 of file TargetTransformInfo.cpp.

◆ getRegisterBitWidth()

unsigned TargetTransformInfo::getRegisterBitWidth ( bool Vector ) const

Returns: The width of the largest scalar or vector register type.

Definition at line 339 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::calculateRegisterUsage(), and llvm::LoopVectorizationCostModel::computeMaxVF().

◆ getScalarizationOverhead()

unsigned TargetTransformInfo::getScalarizationOverhead	(	Type *	Ty,
		bool	Insert,
		bool	Extract
	)		const

Definition at line 244 of file TargetTransformInfo.cpp.

References getOperandsScalarizationOverhead().

Referenced by llvm::LoopVectorizationCostModel::collectInstsToScalarize(), getScalarizationOverhead(), llvm::LoopVectorizationCostModel::setCostBasedWideningDecision(), and useColdCCForColdCall().

◆ getScalingFactorCost()

int TargetTransformInfo::getScalingFactorCost	(	Type *	Ty,
		GlobalValue *	BaseGV,
		int64_t	BaseOffset,
		bool	HasBaseReg,
		int64_t	Scale,
		unsigned	AddrSpace = `0`
	)		const

Return the cost of the scaling factor used in the addressing mode represented by AM for this target, for a load/store of the specified type.

If the AM is supported, the return value must be >= 0. If the AM is not supported, it returns a negative value. TODO: Handle pre/postinc as well.

Definition at line 195 of file TargetTransformInfo.cpp.

References assert().

Referenced by getScalingFactorCost().

◆ getShuffleCost()

int TargetTransformInfo::getShuffleCost	(	ShuffleKind	Kind,
		Type *	Tp,
		int	Index = `0`,
		Type *	SubTp = `nullptr`
	)		const

Returns: The cost of a shuffle instruction of kind Kind and of type Tp. The index and subtype parameters are used by the subvector insertion and extraction shuffle kinds to show the insert/extract point and the type of the subvector being inserted/extracted. NOTE: For subvector extractions Tp represents the source type.

Definition at line 451 of file TargetTransformInfo.cpp.

References assert().

Referenced by isStrideMul(), and llvm::LoopVectorizationCostModel::setCostBasedWideningDecision().

◆ getStoreVectorFactor()

unsigned TargetTransformInfo::getStoreVectorFactor	(	unsigned	VF,
		unsigned	StoreSize,
		unsigned	ChainSizeInBytes,
		VectorType *	VecTy
	)		const

Returns: The new vector factor value if the target doesn't support SizeInBytes stores or has a better vector factor.

Definition at line 675 of file TargetTransformInfo.cpp.

◆ getTgtMemIntrinsic()

bool TargetTransformInfo::getTgtMemIntrinsic	(	IntrinsicInst *	Inst,
		MemIntrinsicInfo &	Info
	)		const

Returns: True if the intrinsic is a supported memory intrinsic. Info will contain additional information - whether the intrinsic may write or read to memory, volatility and the pointer. Info is undefined if false is returned.

Definition at line 594 of file TargetTransformInfo.cpp.

Referenced by getAccessType(), and isAddressUse().

◆ getUnrollingPreferences()

void TargetTransformInfo::getUnrollingPreferences	(	Loop *	L,
		ScalarEvolution &	SE,
		UnrollingPreferences &	UP
	)		const

Get target-customized preferences for the generic loop unrolling transformation.

The caller will initialize UP with the current target-independent defaults.

Definition at line 131 of file TargetTransformInfo.cpp.

Referenced by llvm::gatherUnrollingPreferences().

◆ getUserCost() [1/2]

int TargetTransformInfo::getUserCost	(	const User *	U,
		ArrayRef< const Value *>	Operands
	)		const

Estimate the cost of a given IR user when lowered.

This can estimate the cost of either a ConstantExpr or Instruction when lowered. It has two primary advantages over the getOperationCost and getGEPCost above, and one significant disadvantage: it can only be used when the IR construct has already been formed.

The advantages are that it can inspect the SSA use graph to reason more accurately about the cost. For example, all-constant-GEPs can often be folded into a load or other instruction, but if they are used in some other context they may not be folded. This routine can distinguish such cases.

Operands is a list of operands which can be a result of transformations of the current operands. The number of the operands on the list must equal to the number of the current operands the IR user has. Their order on the list must be the same as the order of the current operands the IR user has.

The returned cost is defined in terms of TargetCostConstants, see its comments for a detailed explanation of the cost values.

Definition at line 104 of file TargetTransformInfo.cpp.

References assert().

Referenced by llvm::CodeMetrics::analyzeBasicBlock(), analyzeLoopUnrollCost(), llvm::ComputeSpeculationCost(), ComputeSpeculationCost(), findProfitablePHIs(), isFreeInLoop(), matchVectorSplittingReduction(), sinkSelectOperand(), and unswitchBestCondition().

◆ getUserCost() [2/2]

int llvm::TargetTransformInfo::getUserCost ( const User * U ) const

inline

This is a helper function which calls the two-argument getUserCost with Operands which are the current operands U has.

Definition at line 276 of file TargetTransformInfo.h.

References F(), llvm::User::value_op_begin(), and llvm::User::value_op_end().

◆ getVectorInstrCost()

int TargetTransformInfo::getVectorInstrCost	(	unsigned	Opcode,
		Type *	Val,
		unsigned	Index = `-1`
	)		const

Returns: The expected cost of vector Insert and Extract. Use -1 to indicate that there is no information on the index value.

Definition at line 490 of file TargetTransformInfo.cpp.

References assert().

Referenced by findBuildVector(), isBroadcastShuffle(), isStrideMul(), and matchVectorSplittingReduction().

◆ hasBranchDivergence()

bool TargetTransformInfo::hasBranchDivergence ( ) const

Return true if branch divergence exists.

Branch divergence has a significantly negative impact on GPU performance when threads in the same wavefront take different paths due to conditional branches.

Definition at line 111 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::computeMaxVF(), llvm::SpeculativeExecutionPass::runImpl(), and llvm::LegacyDivergenceAnalysis::runOnFunction().

◆ hasDivRemOp()

bool TargetTransformInfo::hasDivRemOp	(	Type *	DataType,
		bool	IsSigned
	)		const

Return true if the target has a unified operation to calculate division and remainder.

If so, the additional implicit multiplication and subtraction required to calculate a remainder from division are free. This can enable more aggressive transformations for division and remainder than would typically be allowed using throughput or size cost models.

Definition at line 182 of file TargetTransformInfo.cpp.

Referenced by optimizeDivRem().

◆ hasVolatileVariant()

bool TargetTransformInfo::hasVolatileVariant	(	Instruction *	I,
		unsigned	AddrSpace
	)		const

Return true if the given instruction (assumed to be a memory access instruction) has a volatile variant.

If that's the case then we can avoid addrspacecast to generic AS for volatile loads/stores. Default implementation returns false, which prevents address space inference for volatile loads/stores.

Definition at line 186 of file TargetTransformInfo.cpp.

Referenced by isSimplePointerUseValidToReplace().

◆ haveFastSqrt()

bool TargetTransformInfo::haveFastSqrt ( Type * Ty ) const

Return true if the hardware has a fast square-root instruction.

Definition at line 293 of file TargetTransformInfo.cpp.

Referenced by runPartiallyInlineLibCalls().

◆ invalidate()

bool llvm::TargetTransformInfo::invalidate	(	Function &	,
		const PreservedAnalyses &	,
		FunctionAnalysisManager::Invalidator &
	)

inline

Handle the invalidation of this information.

When used as a result of TargetIRAnalysis this method will be called when the function this was computed for changes. When it returns false, the information is preserved across those changes.

Definition at line 107 of file TargetTransformInfo.h.

◆ isAlwaysUniform()

bool llvm::TargetTransformInfo::isAlwaysUniform ( const Value * V ) const

Definition at line 119 of file TargetTransformInfo.cpp.

Referenced by llvm::GPUDivergenceAnalysis::GPUDivergenceAnalysis().

◆ isFCmpOrdCheaperThanFCmpZero()

bool TargetTransformInfo::isFCmpOrdCheaperThanFCmpZero ( Type * Ty ) const

Return true if it is faster to check if a floating-point value is NaN (or not-NaN) versus a comparison against a constant FP zero value.

Targets should override this if materializing a 0.0 for comparison is generally as cheap as checking for ordered/unordered.

Definition at line 297 of file TargetTransformInfo.cpp.

Referenced by optimizeSQRT().

◆ isFPVectorizationPotentiallyUnsafe()

bool TargetTransformInfo::isFPVectorizationPotentiallyUnsafe ( ) const

Indicate that it is potentially unsafe to automatically vectorize floating-point operations because the semantics of vector and scalar floating-point semantics may differ.

For example, ARM NEON v7 SIMD math does not support IEEE-754 denormal numbers, while depending on the platform, scalar floating-point math does. This applies to floating-point math operations and calls, not memory operations, shuffles, or casts.

Definition at line 275 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizePass::processLoop().

◆ isIndexedLoadLegal()

bool TargetTransformInfo::isIndexedLoadLegal	(	enum MemIndexedMode	Mode,
		Type *	Ty
	)		const

Returns: True if the specified indexed load for the given type is legal.

Definition at line 634 of file TargetTransformInfo.cpp.

Referenced by DeleteTriviallyDeadInstructions(), and mayUsePostIncMode().

◆ isIndexedStoreLegal()

bool TargetTransformInfo::isIndexedStoreLegal	(	enum MemIndexedMode	Mode,
		Type *	Ty
	)		const

Returns: True if the specified indexed store for the given type is legal.

Definition at line 639 of file TargetTransformInfo.cpp.

Referenced by DeleteTriviallyDeadInstructions(), and mayUsePostIncMode().

◆ isLegalAddImmediate()

bool TargetTransformInfo::isLegalAddImmediate ( int64_t Imm ) const

Return true if the specified immediate is legal add immediate, that is the target has add instructions which can add a register with the immediate without having to materialize the immediate into a register.

Definition at line 136 of file TargetTransformInfo.cpp.

Referenced by mayUsePostIncMode().

◆ isLegalAddressingMode()

bool TargetTransformInfo::isLegalAddressingMode	(	Type *	Ty,
		GlobalValue *	BaseGV,
		int64_t	BaseOffset,
		bool	HasBaseReg,
		int64_t	Scale,
		unsigned	AddrSpace = `0`,
		Instruction *	I = `nullptr`
	)		const

Return true if the addressing mode represented by AM is legal for this target, for a load/store of the specified type.

The type may be VoidTy, in which case only return true if the addressing mode is legal for a load/store of any legal type. If target returns true in LSRWithInstrQueries(), I may be valid. TODO: Handle pre/postinc as well.

Definition at line 144 of file TargetTransformInfo.cpp.

Referenced by llvm::createSeparateConstOffsetFromGEPPass(), isAddFoldable(), isAlwaysFoldable(), and isAMCompletelyFolded().

◆ isLegalICmpImmediate()

bool TargetTransformInfo::isLegalICmpImmediate ( int64_t Imm ) const

Return true if the specified immediate is legal icmp immediate, that is the target has icmp instructions which can compare a register against the immediate without having to materialize the immediate into a register.

Definition at line 140 of file TargetTransformInfo.cpp.

Referenced by isAMCompletelyFolded().

◆ isLegalMaskedGather()

bool TargetTransformInfo::isLegalMaskedGather ( Type * DataType ) const

Definition at line 174 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::isLegalMaskedGather().

◆ isLegalMaskedLoad()

bool TargetTransformInfo::isLegalMaskedLoad ( Type * DataType ) const

Definition at line 170 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::interleavedAccessCanBeWidened(), and llvm::LoopVectorizationCostModel::isLegalMaskedLoad().

◆ isLegalMaskedScatter()

bool TargetTransformInfo::isLegalMaskedScatter ( Type * DataType ) const

Return true if the target supports masked gather/scatter AVX-512 fully supports gather and scatter for vectors with 32 and 64 bits scalar type.

Definition at line 178 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::isLegalMaskedScatter().

◆ isLegalMaskedStore()

bool TargetTransformInfo::isLegalMaskedStore ( Type * DataType ) const

Return true if the target supports masked load/store AVX2 and AVX-512 targets allow masks for consecutive load and store.

Definition at line 166 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::interleavedAccessCanBeWidened(), and llvm::LoopVectorizationCostModel::isLegalMaskedStore().

◆ isLegalToVectorizeLoad()

bool TargetTransformInfo::isLegalToVectorizeLoad ( LoadInst * LI ) const

Returns: True if the load instruction is legal to vectorize.

Definition at line 648 of file TargetTransformInfo.cpp.

◆ isLegalToVectorizeLoadChain()

bool TargetTransformInfo::isLegalToVectorizeLoadChain	(	unsigned	ChainSizeInBytes,
		unsigned	Alignment,
		unsigned	AddrSpace
	)		const

Returns: True if it is legal to vectorize the given load chain.

Definition at line 656 of file TargetTransformInfo.cpp.

◆ isLegalToVectorizeStore()

bool TargetTransformInfo::isLegalToVectorizeStore ( StoreInst * SI ) const

Returns: True if the store instruction is legal to vectorize.

Definition at line 652 of file TargetTransformInfo.cpp.

◆ isLegalToVectorizeStoreChain()

bool TargetTransformInfo::isLegalToVectorizeStoreChain	(	unsigned	ChainSizeInBytes,
		unsigned	Alignment,
		unsigned	AddrSpace
	)		const

Returns: True if it is legal to vectorize the given store chain.

Definition at line 662 of file TargetTransformInfo.cpp.

◆ isLoweredToCall()

bool TargetTransformInfo::isLoweredToCall ( const Function * F ) const

Test whether calls to a function lower to actual program function calls.

The idea is to test whether the program is likely to require a 'call' instruction or equivalent in order to call the given function.

FIXME: It's not clear that this is a good or useful query API. Client's should probably move to simpler cost metrics using the above. Alternatively, we could split the cost interface into distinct code-size and execution-speed costs. This would allow modelling the core of this query more accurately as a call is a single small instruction, but incurs significant execution cost.

Definition at line 127 of file TargetTransformInfo.cpp.

Referenced by llvm::CodeMetrics::analyzeBasicBlock(), analyzeLoopUnrollCost(), findTRECandidate(), and llvm::CGProfilePass::run().

◆ isLSRCostLess()

bool TargetTransformInfo::isLSRCostLess	(	TargetTransformInfo::LSRCost &	C1,
		TargetTransformInfo::LSRCost &	C2
	)		const

Return true if LSR cost of C1 is lower than C1.

Definition at line 154 of file TargetTransformInfo.cpp.

Referenced by DeleteTriviallyDeadInstructions().

◆ isProfitableToHoist()

bool TargetTransformInfo::isProfitableToHoist ( Instruction * I ) const

Return true if it is profitable to hoist instruction in the then/else to before if.

Definition at line 214 of file TargetTransformInfo.cpp.

Referenced by HoistThenElseCodeToIf().

◆ isSourceOfDivergence()

bool TargetTransformInfo::isSourceOfDivergence ( const Value * V ) const

Returns whether V is a source of divergence.

This function provides the target-dependent information for the target-independent LegacyDivergenceAnalysis. LegacyDivergenceAnalysis first builds the dependency graph, and then runs the reachability algorithm starting with the sources of divergence.

Definition at line 115 of file TargetTransformInfo.cpp.

Referenced by llvm::GPUDivergenceAnalysis::GPUDivergenceAnalysis().

◆ isTruncateFree()

bool TargetTransformInfo::isTruncateFree	(	Type *	Ty1,
		Type *	Ty2
	)		const

Return true if it's free to truncate a value of type Ty1 to type Ty2.

e.g. On x86 it's free to truncate a i32 value in register EAX to i16 by referencing its sub-register AX.

Definition at line 210 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::isOptimizableIVTruncate(), mayUsePostIncMode(), and llvm::SCEVExpander::replaceCongruentIVs().

◆ isTypeLegal()

bool TargetTransformInfo::isTypeLegal ( Type * Ty ) const

Return true if this type is legal.

Definition at line 220 of file TargetTransformInfo.cpp.

Referenced by llvm::computeMinimumValueSizes(), isAlwaysFoldable(), and ShouldBuildLookupTable().

◆ LSRWithInstrQueries()

bool TargetTransformInfo::LSRWithInstrQueries ( ) const

Return true if the loop strength reduce pass should make Instruction* based TTI queries to isLegalAddressingMode().

This is needed on SystemZ, where e.g. a memcpy can only have a 12 bit unsigned immediate offset and no index register.

Definition at line 206 of file TargetTransformInfo.cpp.

Referenced by isAMCompletelyFolded().

◆ operator=()

TargetTransformInfo & TargetTransformInfo::operator= ( TargetTransformInfo && RHS )

Definition at line 52 of file TargetTransformInfo.cpp.

◆ prefersVectorizedAddressing()

bool TargetTransformInfo::prefersVectorizedAddressing ( ) const

Return true if target doesn't mind addresses in vectors.

Definition at line 191 of file TargetTransformInfo.cpp.

Referenced by getScalarizationOverhead(), and llvm::LoopVectorizationCostModel::setCostBasedWideningDecision().

◆ shouldBuildLookupTables()

bool TargetTransformInfo::shouldBuildLookupTables ( ) const

Return true if switches should be turned into lookup tables for the target.

Definition at line 232 of file TargetTransformInfo.cpp.

Referenced by SwitchToLookupTable().

◆ shouldBuildLookupTablesForConstant()

bool TargetTransformInfo::shouldBuildLookupTablesForConstant ( Constant * C ) const

Return true if switches should be turned into lookup tables containing this constant value for the target.

Definition at line 235 of file TargetTransformInfo.cpp.

Referenced by ValidLookupTableConstant().

◆ shouldConsiderAddressTypePromotion()

bool TargetTransformInfo::shouldConsiderAddressTypePromotion	(	const Instruction &	I,
		bool &	AllowPromotionWithoutCommonHeader
	)		const

Returns: True if it should be considered for address type promotion. AllowPromotionWithoutCommonHeader Set true if promoting I is profitable without finding other extensions fed by the same input.

Definition at line 355 of file TargetTransformInfo.cpp.

◆ shouldExpandReduction()

bool TargetTransformInfo::shouldExpandReduction ( const IntrinsicInst * II ) const

Returns: True if the target wants to expand the given reduction intrinsic into a shuffle sequence.

Definition at line 687 of file TargetTransformInfo.cpp.

References I.

◆ shouldFavorPostInc()

bool TargetTransformInfo::shouldFavorPostInc ( ) const

Returns: True is LSR should make efforts to create/preserve post-inc addressing mode expressions.

Definition at line 162 of file TargetTransformInfo.cpp.

Referenced by DeleteTriviallyDeadInstructions(), and mayUsePostIncMode().

◆ shouldMaximizeVectorBandwidth()

bool TargetTransformInfo::shouldMaximizeVectorBandwidth ( bool OptSize ) const

Returns: True if the vectorization factor should be chosen to make the vector of the smallest element type match the size of a vector register. For wider element types, this could result in creating vectors that span multiple vector registers. If false, the vectorization factor will be chosen based on the size of the widest element type.

Definition at line 347 of file TargetTransformInfo.cpp.

Referenced by llvm::LoopVectorizationCostModel::computeMaxVF().

◆ supportsEfficientVectorElementLoadStore()

bool TargetTransformInfo::supportsEfficientVectorElementLoadStore ( ) const

If target has efficient vector element load/store instructions, it can return true here so that insertion/extraction costs are not added to the scalarization cost of a load/store.

Definition at line 254 of file TargetTransformInfo.cpp.

Referenced by getScalarizationOverhead().

◆ useAA()

bool TargetTransformInfo::useAA ( ) const

Definition at line 218 of file TargetTransformInfo.cpp.

Referenced by llvm::createSeparateConstOffsetFromGEPPass().

◆ useColdCCForColdCall()

bool TargetTransformInfo::useColdCCForColdCall ( Function & F ) const

Return true if the input function which is cold at all call sites, should use coldcc calling convention.

Definition at line 239 of file TargetTransformInfo.cpp.

References getScalarizationOverhead().

Referenced by OptimizeFunctions().

◆ useReductionIntrinsic()

bool TargetTransformInfo::useReductionIntrinsic	(	unsigned	Opcode,
		Type *	Ty,
		ReductionFlags	Flags
	)		const

Returns: True if the target wants to handle the given reduction idiom in the intrinsics form instead of the shuffle form.

Definition at line 682 of file TargetTransformInfo.cpp.

Referenced by llvm::createSimpleTargetReduction().

The documentation for this class was generated from the following files:

include/llvm/Analysis/TargetTransformInfo.h
lib/Analysis/TargetTransformInfo.cpp

Classes

Public Member Functions

Generic Target Information

Scalar Target Information

Vector Target Information

Detailed Description

Member Enumeration Documentation

◆ CacheLevel

◆ MemIndexedMode

◆ OperandValueKind

◆ OperandValueProperties

◆ PopcntSupportKind

◆ ShuffleKind

◆ TargetCostConstants

◆ TargetCostKind

Constructor & Destructor Documentation

◆ TargetTransformInfo() [1/3]

◆ TargetTransformInfo() [2/3]

◆ TargetTransformInfo() [3/3]

◆ ~TargetTransformInfo()

Member Function Documentation

◆ allowsMisalignedMemoryAccesses()

◆ areFunctionArgsABICompatible()

◆ areInlineCompatible()

◆ canMacroFuseCmp()

◆ enableAggressiveInterleaving()

◆ enableInterleavedAccessVectorization()

◆ enableMaskedInterleavedAccessVectorization()

◆ enableMemCmpExpansion()

◆ getAddressComputationCost()

◆ getArithmeticInstrCost()

◆ getArithmeticReductionCost()

◆ getAtomicMemIntrinsicMaxElementSize()

◆ getCacheAssociativity()

◆ getCacheLineSize()

◆ getCacheSize()

◆ getCallCost() [1/3]

◆ getCallCost() [2/3]

◆ getCallCost() [3/3]

◆ getCallInstrCost()

◆ getCastInstrCost()

◆ getCFInstrCost()

◆ getCmpSelInstrCost()

◆ getCostOfKeepingLiveOverCall()

◆ getEstimatedNumberOfCaseClusters()

◆ getExtCost()

◆ getExtractWithExtendCost()

◆ getFlatAddressSpace()

◆ getFPOpCost()

◆ getGatherScatterOpCost()

◆ getGEPCost()

◆ getInliningThresholdMultiplier()

◆ getInstructionCost()

◆ getInterleavedMemoryOpCost()

◆ getIntImmCodeSizeCost()

◆ getIntImmCost() [1/3]

◆ getIntImmCost() [2/3]

◆ getIntImmCost() [3/3]

◆ getIntrinsicCost() [1/2]

◆ getIntrinsicCost() [2/2]

◆ getIntrinsicInstrCost() [1/2]

◆ getIntrinsicInstrCost() [2/2]

◆ getJumpBufAlignment()

◆ getJumpBufSize()

◆ getLoadStoreVecRegBitWidth()

◆ getLoadVectorFactor()

◆ getMaskedMemoryOpCost()

◆ getMaxInterleaveFactor()

◆ getMaxPrefetchIterationsAhead()

◆ getMemcpyLoopLoweringType()

◆ getMemcpyLoopResidualLoweringType()

◆ getMemoryOpCost()

◆ getMinimumVF()

◆ getMinMaxReductionCost()

◆ getMinPrefetchStride()

◆ getMinVectorRegisterBitWidth()

◆ getNumberOfParts()

◆ getNumberOfRegisters()

◆ getOperandInfo()

◆ getOperandsScalarizationOverhead()