llvm::AMDGPUSubtarget Class Reference

#include "Target/AMDGPU/AMDGPUSubtarget.h"

Inheritance diagram for llvm::AMDGPUSubtarget:

Collaboration diagram for llvm::AMDGPUSubtarget:

Public Types
enum	Generation {   INVALID = 0, R600 = 1, R700 = 2, EVERGREEN = 3,   NORTHERN_ISLANDS = 4, SOUTHERN_ISLANDS = 5, SEA_ISLANDS = 6, VOLCANIC_ISLANDS = 7,   GFX9 = 8, GFX10 = 9, GFX11 = 10 }

Public Member Functions
	AMDGPUSubtarget (const Triple &TT)
std::pair< unsigned, unsigned >	getDefaultFlatWorkGroupSize (CallingConv::ID CC) const
std::pair< unsigned, unsigned >	getFlatWorkGroupSizes (const Function &F) const
std::pair< unsigned, unsigned >	getWavesPerEU (const Function &F) const
std::pair< unsigned, unsigned >	getWavesPerEU (const Function &F, std::pair< unsigned, unsigned > FlatWorkGroupSizes) const
	Overload which uses the specified values for the flat work group sizes, rather than querying the function itself. More...
unsigned	getMaxLocalMemSizeWithWaveCount (unsigned WaveCount, const Function &) const
	Return the amount of LDS that can be used that will not restrict the occupancy lower than WaveCount. More...
unsigned	getOccupancyWithLocalMemSize (uint32_t Bytes, const Function &) const
	Inverse of getMaxLocalMemWithWaveCount. More...
unsigned	getOccupancyWithLocalMemSize (const MachineFunction &MF) const
bool	isAmdHsaOS () const
bool	isAmdPalOS () const
bool	isMesa3DOS () const
bool	isMesaKernel (const Function &F) const
bool	isAmdHsaOrMesa (const Function &F) const
bool	isGCN () const
bool	isGCN3Encoding () const
bool	has16BitInsts () const
bool	hasTrue16BitInsts () const
bool	hasMadMixInsts () const
bool	hasMadMacF32Insts () const
bool	hasDsSrc2Insts () const
bool	hasSDWA () const
bool	hasVOP3PInsts () const
bool	hasMulI24 () const
bool	hasMulU24 () const
bool	hasSMulHi () const
bool	hasInv2PiInlineImm () const
bool	hasFminFmaxLegacy () const
bool	hasTrigReducedRange () const
bool	isPromoteAllocaEnabled () const
unsigned	getWavefrontSize () const
unsigned	getWavefrontSizeLog2 () const
unsigned	getLocalMemorySize () const
Align	getAlignmentForImplicitArgPtr () const
unsigned	getExplicitKernelArgOffset (const Function &F) const
	Returns the offset in bytes from the start of the input buffer of the first explicit kernel argument. More...
virtual unsigned	getMaxWorkGroupsPerCU (unsigned FlatWorkGroupSize) const =0
virtual unsigned	getMinFlatWorkGroupSize () const =0
virtual unsigned	getMaxFlatWorkGroupSize () const =0
virtual unsigned	getWavesPerEUForWorkGroup (unsigned FlatWorkGroupSize) const =0
virtual unsigned	getMinWavesPerEU () const =0
unsigned	getMaxWavesPerEU () const
unsigned	getMaxWorkitemID (const Function &Kernel, unsigned Dimension) const
	Return the maximum workitem ID value in the function, for the given (0, 1, 2) dimension. More...
bool	makeLIDRangeMetadata (Instruction *I) const
	Creates value range metadata on an workitemid.* intrinsic call or load. More...
unsigned	getImplicitArgNumBytes (const Function &F) const
uint64_t	getExplicitKernArgSize (const Function &F, Align &MaxAlign) const
unsigned	getKernArgSegmentSize (const Function &F, Align &MaxAlign) const
AMDGPUDwarfFlavour	getAMDGPUDwarfFlavour () const
virtual	~AMDGPUSubtarget ()=default

Static Public Member Functions
static const AMDGPUSubtarget &	get (const MachineFunction &MF)
static const AMDGPUSubtarget &	get (const TargetMachine &TM, const Function &F)

Protected Attributes
bool	GCN3Encoding = false
bool	Has16BitInsts = false
bool	HasTrue16BitInsts = false
bool	HasMadMixInsts = false
bool	HasMadMacF32Insts = false
bool	HasDsSrc2Insts = false
bool	HasSDWA = false
bool	HasVOP3PInsts = false
bool	HasMulI24 = true
bool	HasMulU24 = true
bool	HasSMulHi = false
bool	HasInv2PiInlineImm = false
bool	HasFminFmaxLegacy = true
bool	EnablePromoteAlloca = false
bool	HasTrigReducedRange = false
unsigned	MaxWavesPerEU = 10
unsigned	LocalMemorySize = 0
char	WavefrontSizeLog2 = 0

Detailed Description

Definition at line 29 of file AMDGPUSubtarget.h.

Member Enumeration Documentation

Generation

enum llvm::AMDGPUSubtarget::Generation

Enumerator
INVALID
R600
R700
EVERGREEN
NORTHERN_ISLANDS
SOUTHERN_ISLANDS
SEA_ISLANDS
VOLCANIC_ISLANDS
GFX9
GFX10
GFX11

Definition at line 31 of file AMDGPUSubtarget.h.

Constructor & Destructor Documentation

AMDGPUSubtarget()

AMDGPUSubtarget::AMDGPUSubtarget

(

const Triple &

TT

)

Definition at line 161 of file AMDGPUSubtarget.cpp.

~AMDGPUSubtarget()

virtual llvm::AMDGPUSubtarget::~AMDGPUSubtarget ( )

virtualdefault

Member Function Documentation

get() [1/2]

const AMDGPUSubtarget & AMDGPUSubtarget::get ( const MachineFunction &  MF )

static

Definition at line 968 of file AMDGPUSubtarget.cpp.

References llvm::Triple::amdgcn, llvm::Triple::getArch(), llvm::MachineFunction::getSubtarget(), llvm::MachineFunction::getTarget(), and llvm::TargetMachine::getTargetTriple().

Referenced by llvm::AMDGPUMachineFunction::AMDGPUMachineFunction(), llvm::AMDGPUTargetLowering::analyzeFormalArgumentsCompute(), llvm::AMDGPUTargetLowering::getImplicitParameterOffset(), and promoteAllocasToVector().

get() [2/2]

const AMDGPUSubtarget & AMDGPUSubtarget::get ( const TargetMachine &  TM, const Function &  F  )

static

Definition at line 975 of file AMDGPUSubtarget.cpp.

References llvm::Triple::amdgcn, F, and TM.

getAlignmentForImplicitArgPtr()

Align llvm::AMDGPUSubtarget::getAlignmentForImplicitArgPtr ( ) const

inline

Definition at line 212 of file AMDGPUSubtarget.h.

References Align, and isAmdHsaOS().

Referenced by getKernArgSegmentSize().

getAMDGPUDwarfFlavour()

AMDGPUDwarfFlavour AMDGPUSubtarget::getAMDGPUDwarfFlavour

(

)

const

Returns

Corresponding DWARF register number mapping flavour for the WavefrontSize.

Definition at line 570 of file AMDGPUSubtarget.cpp.

References getWavefrontSize(), llvm::Wave32, and llvm::Wave64.

getDefaultFlatWorkGroupSize()

std::pair< unsigned, unsigned > AMDGPUSubtarget::getDefaultFlatWorkGroupSize

(

CallingConv::ID

CC

)

const

Returns

Default range flat work group size for a calling convention.

Definition at line 362 of file AMDGPUSubtarget.cpp.

References llvm::CallingConv::AMDGPU_ES, llvm::CallingConv::AMDGPU_GS, llvm::CallingConv::AMDGPU_HS, llvm::CallingConv::AMDGPU_LS, llvm::CallingConv::AMDGPU_PS, llvm::CallingConv::AMDGPU_VS, CC, getMaxFlatWorkGroupSize(), and getWavefrontSize().

Referenced by getFlatWorkGroupSizes().

getExplicitKernArgSize()

uint64_t AMDGPUSubtarget::getExplicitKernArgSize	(	const Function &	F,
		Align &	MaxAlign
	)		const

Definition at line 530 of file AMDGPUSubtarget.cpp.

References Align, llvm::alignTo(), llvm::CallingConv::AMDGPU_KERNEL, Arg, assert(), DL, F, llvm::max(), and llvm::CallingConv::SPIR_KERNEL.

Referenced by getKernArgSegmentSize().

getExplicitKernelArgOffset()

unsigned llvm::AMDGPUSubtarget::getExplicitKernelArgOffset ( const Function &  F ) const

inline

Returns the offset in bytes from the start of the input buffer of the first explicit kernel argument.

Definition at line 218 of file AMDGPUSubtarget.h.

References llvm::Triple::AMDHSA, llvm::Triple::AMDPAL, llvm::Triple::getOS(), llvm_unreachable, llvm::Triple::Mesa3D, and llvm::Triple::UnknownOS.

Referenced by getKernArgSegmentSize(), and llvm::AMDGPUCallLowering::lowerFormalArgumentsKernel().

getFlatWorkGroupSizes()

std::pair< unsigned, unsigned > AMDGPUSubtarget::getFlatWorkGroupSizes

(

const Function &

F

)

const

Returns

Subtarget's default pair of minimum/maximum flat work group sizes for function F, or minimum/maximum flat work group sizes explicitly requested using "amdgpu-flat-work-group-size" attribute attached to function F.

Subtarget's default values if explicitly requested values cannot be converted to integer, or violate subtarget's specifications.

Definition at line 376 of file AMDGPUSubtarget.cpp.

References F, getDefaultFlatWorkGroupSize(), llvm::AMDGPU::getIntegerPairAttribute(), getMaxFlatWorkGroupSize(), and getMinFlatWorkGroupSize().

Referenced by getMaxLocalMemSizeWithWaveCount(), getMaxWorkitemID(), getOccupancyWithLocalMemSize(), getWavesPerEU(), and makeLIDRangeMetadata().

getImplicitArgNumBytes()

unsigned AMDGPUSubtarget::getImplicitArgNumBytes

(

const Function &

F

)

const

Returns

Number of bytes of arguments that are passed to a shader or kernel in addition to the explicit ones declared for the function.

Definition at line 514 of file AMDGPUSubtarget.cpp.

References assert(), F, llvm::AMDGPU::getAmdhsaCodeObjectVersion(), llvm::AMDGPU::getIntegerAttribute(), llvm::AMDGPU::isKernel(), and isMesaKernel().

Referenced by getKernArgSegmentSize().

getKernArgSegmentSize()

unsigned AMDGPUSubtarget::getKernArgSegmentSize	(	const Function &	F,
		Align &	MaxAlign
	)		const

Definition at line 552 of file AMDGPUSubtarget.cpp.

References llvm::alignTo(), F, getAlignmentForImplicitArgPtr(), getExplicitKernArgSize(), getExplicitKernelArgOffset(), getImplicitArgNumBytes(), and llvm::max().

Referenced by llvm::AMDGPU::HSAMD::MetadataStreamerMsgPackV3::getHSAKernelProps().

getLocalMemorySize()

unsigned llvm::AMDGPUSubtarget::getLocalMemorySize ( ) const

inline

Definition at line 208 of file AMDGPUSubtarget.h.

References LocalMemorySize.

Referenced by getMaxLocalMemSizeWithWaveCount(), and getOccupancyWithLocalMemSize().

getMaxFlatWorkGroupSize()

virtual unsigned llvm::AMDGPUSubtarget::getMaxFlatWorkGroupSize ( ) const

pure virtual

Returns

Maximum flat work group size supported by the subtarget.

Implemented in llvm::GCNSubtarget, and llvm::R600Subtarget.

Referenced by getDefaultFlatWorkGroupSize(), and getFlatWorkGroupSizes().

getMaxLocalMemSizeWithWaveCount()

unsigned AMDGPUSubtarget::getMaxLocalMemSizeWithWaveCount	(	unsigned	WaveCount,
		const Function &	F
	)		const

Return the amount of LDS that can be used that will not restrict the occupancy lower than WaveCount.

Definition at line 306 of file AMDGPUSubtarget.cpp.

References F, getFlatWorkGroupSizes(), getLocalMemorySize(), getMaxWavesPerEU(), and getMaxWorkGroupsPerCU().

getMaxWavesPerEU()

unsigned llvm::AMDGPUSubtarget::getMaxWavesPerEU ( ) const

inline

Returns

Maximum number of waves per execution unit supported by the subtarget without any kind of limitation.

Definition at line 255 of file AMDGPUSubtarget.h.

References MaxWavesPerEU.

Referenced by llvm::GCNSubtarget::computeOccupancy(), getMaxLocalMemSizeWithWaveCount(), getOccupancyWithLocalMemSize(), llvm::GCNSubtarget::getOccupancyWithNumSGPRs(), llvm::GCNSubtarget::getOccupancyWithNumVGPRs(), and getWavesPerEU().

getMaxWorkGroupsPerCU()

virtual unsigned llvm::AMDGPUSubtarget::getMaxWorkGroupsPerCU ( unsigned  FlatWorkGroupSize ) const

pure virtual

Returns

Maximum number of work groups per compute unit supported by the subtarget and limited by given FlatWorkGroupSize.

Implemented in llvm::GCNSubtarget, and llvm::R600Subtarget.

Referenced by getMaxLocalMemSizeWithWaveCount(), and getOccupancyWithLocalMemSize().

getMaxWorkitemID()

unsigned AMDGPUSubtarget::getMaxWorkitemID	(	const Function &	Kernel,
		unsigned	Dimension
	)		const

Return the maximum workitem ID value in the function, for the given (0, 1, 2) dimension.

Definition at line 444 of file AMDGPUSubtarget.cpp.

References getFlatWorkGroupSizes(), getReqdWorkGroupSize(), and llvm::max().

Referenced by llvm::AMDGPUTargetLowering::computeKnownBitsForTargetNode(), llvm::AMDGPULegalizerInfo::legalizeWorkitemIDIntrinsic(), and llvm::SITargetLowering::passSpecialInputs().

getMinFlatWorkGroupSize()

virtual unsigned llvm::AMDGPUSubtarget::getMinFlatWorkGroupSize ( ) const

pure virtual

Returns

Minimum flat work group size supported by the subtarget.

Implemented in llvm::GCNSubtarget, and llvm::R600Subtarget.

Referenced by getFlatWorkGroupSizes().

getMinWavesPerEU()

virtual unsigned llvm::AMDGPUSubtarget::getMinWavesPerEU ( ) const

pure virtual

Returns

Minimum number of waves per execution unit supported by the subtarget.

Implemented in llvm::GCNSubtarget, and llvm::R600Subtarget.

Referenced by getWavesPerEU().

getOccupancyWithLocalMemSize() [1/2]

unsigned AMDGPUSubtarget::getOccupancyWithLocalMemSize

(

const MachineFunction &

MF

)

const

Definition at line 356 of file AMDGPUSubtarget.cpp.

References llvm::MachineFunction::getFunction(), llvm::MachineFunction::getInfo(), and getOccupancyWithLocalMemSize().

getOccupancyWithLocalMemSize() [2/2]

unsigned AMDGPUSubtarget::getOccupancyWithLocalMemSize	(	uint32_t	Bytes,
		const Function &	F
	)		const

Inverse of getMaxLocalMemWithWaveCount.

Return the maximum wavecount if the given LDS memory size is the only constraint.

Definition at line 319 of file AMDGPUSubtarget.cpp.

References assert(), F, getFlatWorkGroupSizes(), getLocalMemorySize(), getMaxWavesPerEU(), getMaxWorkGroupsPerCU(), getWavefrontSize(), and llvm::min().

Referenced by llvm::GCNSubtarget::computeOccupancy(), getOccupancyWithLocalMemSize(), and llvm::SIRegisterInfo::getRegPressureLimit().

getWavefrontSize()

unsigned llvm::AMDGPUSubtarget::getWavefrontSize ( ) const

inline

Definition at line 200 of file AMDGPUSubtarget.h.

References WavefrontSizeLog2.

Referenced by llvm::SIRegisterInfo::buildSpillLoadStore(), getAMDGPUDwarfFlavour(), getDefaultFlatWorkGroupSize(), llvm::AMDGPU::HSAMD::MetadataStreamerMsgPackV3::getHSAKernelProps(), getOccupancyWithLocalMemSize(), llvm::SITargetLowering::getRegClassFor(), llvm::SIInstrInfo::getScratchRsrcWords23(), llvm::GCNSubtarget::isWave32(), llvm::GCNSubtarget::isWave64(), llvm::AMDGPULegalizerInfo::legalizeIntrinsic(), lowerFCMPIntrinsic(), lowerICMPIntrinsic(), and llvm::SITargetLowering::requiresUniformRegister().

getWavefrontSizeLog2()

unsigned llvm::AMDGPUSubtarget::getWavefrontSizeLog2 ( ) const

inline

Definition at line 204 of file AMDGPUSubtarget.h.

References WavefrontSizeLog2.

Referenced by llvm::SIRegisterInfo::eliminateFrameIndex(), and llvm::GCNSubtarget::getKnownHighZeroBitsForFrameIndex().

getWavesPerEU() [1/2]

std::pair<unsigned, unsigned> llvm::AMDGPUSubtarget::getWavesPerEU ( const Function &  F ) const

inline

Returns

Subtarget's default pair of minimum/maximum number of waves per execution unit for function F, or minimum/maximum number of waves per execution unit explicitly requested using "amdgpu-waves-per-eu" attribute attached to function F.

Subtarget's default values if explicitly requested values cannot be converted to integer, violate subtarget's specifications, or are not compatible with minimum/maximum number of waves limited by flat work group size, register usage, and/or lds usage.

Definition at line 96 of file AMDGPUSubtarget.h.

References F, and getFlatWorkGroupSizes().

Referenced by llvm::GCNSubtarget::getMaxNumSGPRs(), and llvm::GCNSubtarget::getMaxNumVGPRs().

getWavesPerEU() [2/2]

std::pair< unsigned, unsigned > AMDGPUSubtarget::getWavesPerEU	(	const Function &	F,
		std::pair< unsigned, unsigned >	FlatWorkGroupSizes
	)		const

Overload which uses the specified values for the flat work group sizes, rather than querying the function itself.

FlatWorkGroupSizes Should correspond to the function's value for getFlatWorkGroupSizes.

Definition at line 399 of file AMDGPUSubtarget.cpp.

References F, llvm::AMDGPU::getIntegerPairAttribute(), getMaxWavesPerEU(), getMinWavesPerEU(), and getWavesPerEUForWorkGroup().

getWavesPerEUForWorkGroup()

virtual unsigned llvm::AMDGPUSubtarget::getWavesPerEUForWorkGroup ( unsigned  FlatWorkGroupSize ) const

pure virtual

Returns

Number of waves per execution unit required to support the given FlatWorkGroupSize.

Implemented in llvm::GCNSubtarget, and llvm::R600Subtarget.

Referenced by getWavesPerEU().

has16BitInsts()

bool llvm::AMDGPUSubtarget::has16BitInsts ( ) const

inline

Definition at line 146 of file AMDGPUSubtarget.h.

References Has16BitInsts.

Referenced by llvm::AMDGPUTargetLowering::AMDGPUTargetLowering(), llvm::GCNTTIImpl::getArithmeticInstrCost(), llvm::SITargetLowering::getAsmOperandConstVal(), llvm::GCNTTIImpl::getIntrinsicInstrCost(), llvm::GCNTTIImpl::getMaximumVF(), llvm::SITargetLowering::getNumRegistersForCallingConv(), llvm::SITargetLowering::getPreferredShiftAmountTy(), llvm::SITargetLowering::getRegisterTypeForCallingConv(), llvm::GCNTTIImpl::getVectorInstrCost(), llvm::SITargetLowering::getVectorTypeBreakdownForCallingConv(), llvm::AMDGPUTargetLowering::isFAbsFree(), llvm::SITargetLowering::isFMAFasterThanFMulAndFAdd(), llvm::AMDGPUTargetLowering::isFPImmLegal(), llvm::SIInstrInfo::isInlineConstant(), llvm::AMDGPUTargetLowering::isTruncateFree(), llvm::SITargetLowering::isTypeDesirableForOp(), llvm::AMDGPUTargetLowering::isZExtFree(), llvm::AMDGPUTargetLowering::LowerSINT_TO_FP(), llvm::AMDGPUTargetLowering::LowerUINT_TO_FP(), llvm::AMDGPUTargetLowering::performFAbsCombine(), llvm::AMDGPUTargetLowering::performMulCombine(), and llvm::SITargetLowering::SITargetLowering().

hasDsSrc2Insts()

bool llvm::AMDGPUSubtarget::hasDsSrc2Insts ( ) const

inline

Definition at line 160 of file AMDGPUSubtarget.h.

References HasDsSrc2Insts.

hasFminFmaxLegacy()

bool llvm::AMDGPUSubtarget::hasFminFmaxLegacy ( ) const

inline

Definition at line 188 of file AMDGPUSubtarget.h.

References HasFminFmaxLegacy.

Referenced by llvm::AMDGPUTargetLowering::performSelectCombine().

hasInv2PiInlineImm()

bool llvm::AMDGPUSubtarget::hasInv2PiInlineImm ( ) const

inline

Definition at line 184 of file AMDGPUSubtarget.h.

References HasInv2PiInlineImm.

Referenced by llvm::SITargetLowering::checkAsmConstraintValA(), llvm::SIRegisterInfo::eliminateFrameIndex(), llvm::AMDGPUTargetLowering::isConstantCostlierToNegate(), and llvm::SIInstrInfo::isInlineConstant().

hasMadMacF32Insts()

bool llvm::AMDGPUSubtarget::hasMadMacF32Insts ( ) const

inline

Definition at line 156 of file AMDGPUSubtarget.h.

References HasMadMacF32Insts, and isGCN().

Referenced by llvm::GCNTTIImpl::getArithmeticInstrCost(), llvm::SITargetLowering::isFMADLegal(), llvm::SITargetLowering::isFMAFasterThanFMulAndFAdd(), llvm::AMDGPUTargetLowering::LowerDIVREM24(), llvm::AMDGPUTargetLowering::LowerUDIVREM64(), and llvm::SITargetLowering::SITargetLowering().

hasMadMixInsts()

bool llvm::AMDGPUSubtarget::hasMadMixInsts ( ) const

inline

Definition at line 152 of file AMDGPUSubtarget.h.

References HasMadMixInsts.

Referenced by llvm::SITargetLowering::isFPExtFoldable().

hasMulI24()

bool llvm::AMDGPUSubtarget::hasMulI24 ( ) const

inline

Definition at line 172 of file AMDGPUSubtarget.h.

References HasMulI24.

Referenced by llvm::AMDGPUTargetLowering::performMulCombine(), llvm::AMDGPUTargetLowering::performMulhsCombine(), and llvm::AMDGPUTargetLowering::performMulLoHiCombine().

hasMulU24()

bool llvm::AMDGPUSubtarget::hasMulU24 ( ) const

inline

Definition at line 176 of file AMDGPUSubtarget.h.

References HasMulU24.

Referenced by llvm::AMDGPUTargetLowering::performMulCombine(), llvm::AMDGPUTargetLowering::performMulhuCombine(), and llvm::AMDGPUTargetLowering::performMulLoHiCombine().

hasSDWA()

bool llvm::AMDGPUSubtarget::hasSDWA ( ) const

inline

Definition at line 164 of file AMDGPUSubtarget.h.

References HasSDWA.

Referenced by llvm::AMDGPUTargetLowering::PerformDAGCombine(), and llvm::SIInstrInfo::verifyInstruction().

hasSMulHi()

bool llvm::AMDGPUSubtarget::hasSMulHi ( ) const

inline

Definition at line 180 of file AMDGPUSubtarget.h.

References HasSMulHi.

Referenced by llvm::AMDGPURegisterBankInfo::applyMappingMAD_64_32(), llvm::AMDGPUTargetLowering::performMulhsCombine(), and llvm::AMDGPUTargetLowering::performMulhuCombine().

hasTrigReducedRange()

bool llvm::AMDGPUSubtarget::hasTrigReducedRange ( ) const

inline

Definition at line 192 of file AMDGPUSubtarget.h.

References HasTrigReducedRange.

Referenced by llvm::AMDGPULegalizerInfo::legalizeSinCos().

hasTrue16BitInsts()

bool llvm::AMDGPUSubtarget::hasTrue16BitInsts ( ) const

inline

Definition at line 150 of file AMDGPUSubtarget.h.

References HasTrue16BitInsts.

Referenced by llvm::SIInstrInfo::convertToThreeAddress(), and llvm::SIInstrInfo::FoldImmediate().

hasVOP3PInsts()

bool llvm::AMDGPUSubtarget::hasVOP3PInsts ( ) const

inline

Definition at line 168 of file AMDGPUSubtarget.h.

References HasVOP3PInsts.

Referenced by llvm::GCNTTIImpl::getArithmeticReductionCost(), llvm::GCNTTIImpl::getMinMaxReductionCost(), llvm::GCNTTIImpl::getShuffleCost(), and llvm::SITargetLowering::SITargetLowering().

isAmdHsaOrMesa()

bool llvm::AMDGPUSubtarget::isAmdHsaOrMesa ( const Function &  F ) const

inline

Definition at line 134 of file AMDGPUSubtarget.h.

References F, isAmdHsaOS(), and isMesaKernel().

isAmdHsaOS()

bool llvm::AMDGPUSubtarget::isAmdHsaOS ( ) const

inline

Definition at line 120 of file AMDGPUSubtarget.h.

References llvm::Triple::AMDHSA, and llvm::Triple::getOS().

Referenced by getAlignmentForImplicitArgPtr(), llvm::SIInstrInfo::getDefaultRsrcDataFormat(), llvm::GCNSubtarget::getTrapHandlerAbi(), llvm::GCNSubtarget::initializeSubtargetDependencies(), isAmdHsaOrMesa(), llvm::SITargetLowering::LowerFormalArguments(), and llvm::AMDGPUAsmPrinter::runOnMachineFunction().

isAmdPalOS()

bool llvm::AMDGPUSubtarget::isAmdPalOS ( ) const

inline

Definition at line 124 of file AMDGPUSubtarget.h.

References llvm::Triple::AMDPAL, and llvm::Triple::getOS().

Referenced by llvm::SITargetLowering::LowerFormalArguments(), and llvm::AMDGPUAsmPrinter::runOnMachineFunction().

isGCN()

bool llvm::AMDGPUSubtarget::isGCN ( ) const

inline

Definition at line 138 of file AMDGPUSubtarget.h.

References llvm::Triple::amdgcn, and llvm::Triple::getArch().

Referenced by hasMadMacF32Insts(), llvm::AMDGPUTargetLowering::LowerDIVREM24(), and llvm::AMDGPUTargetLowering::LowerINT_TO_FP32().

isGCN3Encoding()

bool llvm::AMDGPUSubtarget::isGCN3Encoding ( ) const

inline

Definition at line 142 of file AMDGPUSubtarget.h.

References GCN3Encoding.

isMesa3DOS()

bool llvm::AMDGPUSubtarget::isMesa3DOS ( ) const

inline

Definition at line 128 of file AMDGPUSubtarget.h.

References llvm::Triple::getOS(), and llvm::Triple::Mesa3D.

Referenced by llvm::GCNSubtarget::isMesaGfxShader(), and isMesaKernel().

isMesaKernel()

bool AMDGPUSubtarget::isMesaKernel

(

const Function &

F

)

const

Definition at line 440 of file AMDGPUSubtarget.cpp.

References F, isMesa3DOS(), and llvm::AMDGPU::isShader().

Referenced by getImplicitArgNumBytes(), and isAmdHsaOrMesa().

isPromoteAllocaEnabled()

bool llvm::AMDGPUSubtarget::isPromoteAllocaEnabled ( ) const

inline

Definition at line 196 of file AMDGPUSubtarget.h.

References EnablePromoteAlloca.

makeLIDRangeMetadata()

bool AMDGPUSubtarget::makeLIDRangeMetadata

(

Instruction *

I

)

const

Creates value range metadata on an workitemid.* intrinsic call or load.

Definition at line 452 of file AMDGPUSubtarget.cpp.

References llvm::MDBuilder::createRange(), F, getFlatWorkGroupSizes(), getReqdWorkGroupSize(), I, if(), and llvm::max().

Member Data Documentation

EnablePromoteAlloca

bool llvm::AMDGPUSubtarget::EnablePromoteAlloca = false

protected

Definition at line 62 of file AMDGPUSubtarget.h.

Referenced by isPromoteAllocaEnabled().

GCN3Encoding

bool llvm::AMDGPUSubtarget::GCN3Encoding = false

protected

Definition at line 49 of file AMDGPUSubtarget.h.

Referenced by isGCN3Encoding().

Has16BitInsts

bool llvm::AMDGPUSubtarget::Has16BitInsts = false

protected

Definition at line 50 of file AMDGPUSubtarget.h.

Referenced by has16BitInsts().

HasDsSrc2Insts

bool llvm::AMDGPUSubtarget::HasDsSrc2Insts = false

protected

Definition at line 54 of file AMDGPUSubtarget.h.

Referenced by hasDsSrc2Insts().

HasFminFmaxLegacy

bool llvm::AMDGPUSubtarget::HasFminFmaxLegacy = true

protected

Definition at line 61 of file AMDGPUSubtarget.h.

Referenced by hasFminFmaxLegacy(), and llvm::GCNSubtarget::initializeSubtargetDependencies().

HasInv2PiInlineImm

bool llvm::AMDGPUSubtarget::HasInv2PiInlineImm = false

protected

Definition at line 60 of file AMDGPUSubtarget.h.

Referenced by hasInv2PiInlineImm().

HasMadMacF32Insts

bool llvm::AMDGPUSubtarget::HasMadMacF32Insts = false

protected

Definition at line 53 of file AMDGPUSubtarget.h.

Referenced by hasMadMacF32Insts().

HasMadMixInsts

bool llvm::AMDGPUSubtarget::HasMadMixInsts = false

protected

Definition at line 52 of file AMDGPUSubtarget.h.

Referenced by hasMadMixInsts().

HasMulI24

bool llvm::AMDGPUSubtarget::HasMulI24 = true

protected

Definition at line 57 of file AMDGPUSubtarget.h.

Referenced by hasMulI24(), and llvm::R600Subtarget::initializeSubtargetDependencies().

HasMulU24

bool llvm::AMDGPUSubtarget::HasMulU24 = true

protected

Definition at line 58 of file AMDGPUSubtarget.h.

Referenced by hasMulU24(), and llvm::R600Subtarget::initializeSubtargetDependencies().

HasSDWA

bool llvm::AMDGPUSubtarget::HasSDWA = false

protected

Definition at line 55 of file AMDGPUSubtarget.h.

Referenced by hasSDWA().

HasSMulHi

bool llvm::AMDGPUSubtarget::HasSMulHi = false

protected

Definition at line 59 of file AMDGPUSubtarget.h.

Referenced by hasSMulHi(), and llvm::GCNSubtarget::initializeSubtargetDependencies().

HasTrigReducedRange

bool llvm::AMDGPUSubtarget::HasTrigReducedRange = false

protected

Definition at line 63 of file AMDGPUSubtarget.h.

Referenced by hasTrigReducedRange().

HasTrue16BitInsts

bool llvm::AMDGPUSubtarget::HasTrue16BitInsts = false

protected

Definition at line 51 of file AMDGPUSubtarget.h.

Referenced by hasTrue16BitInsts().

HasVOP3PInsts

bool llvm::AMDGPUSubtarget::HasVOP3PInsts = false

protected

Definition at line 56 of file AMDGPUSubtarget.h.

Referenced by hasVOP3PInsts().

LocalMemorySize

unsigned llvm::AMDGPUSubtarget::LocalMemorySize = 0

protected

Definition at line 65 of file AMDGPUSubtarget.h.

Referenced by getLocalMemorySize(), and llvm::GCNSubtarget::initializeSubtargetDependencies().

MaxWavesPerEU

unsigned llvm::AMDGPUSubtarget::MaxWavesPerEU = 10

protected

Definition at line 64 of file AMDGPUSubtarget.h.

Referenced by llvm::GCNSubtarget::GCNSubtarget(), and getMaxWavesPerEU().

WavefrontSizeLog2

char llvm::AMDGPUSubtarget::WavefrontSizeLog2 = 0

protected

Definition at line 66 of file AMDGPUSubtarget.h.

Referenced by getWavefrontSize(), getWavefrontSizeLog2(), and llvm::GCNSubtarget::initializeSubtargetDependencies().

---

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

• lib/Target/AMDGPU/AMDGPUSubtarget.h
• lib/Target/AMDGPU/AMDGPUSubtarget.cpp