Managed reference to MutatorNode. More...

#include <mutator.h>

Inheritance diagram for tvm::meta_schedule::Mutator:

Collaboration diagram for tvm::meta_schedule::Mutator:

Public Types
using	FInitializeWithTuneContext = runtime::TypedPackedFunc< void(const TuneContext &)>
	The function type of `InitializeWithTuneContext` method. More...

using	FApply = runtime::TypedPackedFunc< Optional< tir::Trace >(const tir::Trace &, support::LinearCongruentialEngine::TRandState rand_state)>
	Apply the mutator function to the given trace. More...

using	FClone = runtime::TypedPackedFunc< Mutator()>
	Clone the mutator. More...

using	FAsString = runtime::TypedPackedFunc< String()>
	Get the mutator as string with name. More...

Public Types inherited from tvm::runtime::ObjectRef
using	ContainerType = Object
	type indicate the container type. More...

Public Member Functions
	TVM_DEFINE_MUTABLE_OBJECT_REF_METHODS (Mutator, ObjectRef, MutatorNode)

Public Member Functions inherited from tvm::runtime::ObjectRef
	ObjectRef ()=default
	default constructor More...

	ObjectRef (ObjectPtr< Object > data)
	Constructor from existing object ptr. More...

bool	same_as (const ObjectRef &other) const
	Comparator. More...

bool	operator== (const ObjectRef &other) const
	Comparator. More...

bool	operator!= (const ObjectRef &other) const
	Comparator. More...

bool	operator< (const ObjectRef &other) const
	Comparator. More...

bool	defined () const

const Object *	get () const

const Object *	operator-> () const

bool	unique () const

int	use_count () const

template<typename ObjectType , typename = std::enable_if_t<std::is_base_of_v<Object, ObjectType>>>
const ObjectType *	as () const
	Try to downcast the internal Object to a raw pointer of a corresponding type. More...

template<typename ObjectRefType , typename = std::enable_if_t<std::is_base_of_v<ObjectRef, ObjectRefType>>>
Optional< ObjectRefType >	as () const
	Try to downcast the ObjectRef to a Optional<T> of the requested type. More...

Static Public Member Functions
static Mutator	MutateTileSize ()
	Create a Mutator that mutates the decision of instruction Sample-Perfect-Tile. More...

static Mutator	MutateParallel (int64_t max_jobs_per_core)
	Create a Mutator that mutates the parallel extent. More...

static Mutator	MutateUnroll ()
	Create a Mutator that mutates auto unroll step. More...

static Mutator	MutateComputeLocation ()
	Create a Mutator that mutates the outcome of SampleComputeLocation. More...

static Mutator	MutateThreadBinding ()
	Create a Mutator that mutates auto thread binding. More...

static Mutator	PyMutator (FInitializeWithTuneContext f_initialize_with_tune_context, FApply f_apply, FClone f_clone, FAsString f_as_string)
	Create a mutator with customized methods on the python-side. More...

static Map< Mutator, FloatImm, void >	DefaultLLVM ()
	Create default mutators for LLVM. More...

static Map< Mutator, FloatImm, void >	DefaultCUDA ()
	Create default mutators for CUDA. More...

static Map< Mutator, FloatImm, void >	DefaultCUDATensorCore ()
	Create default mutators for CUDA with TensorCore. More...

static Map< Mutator, FloatImm, void >	DefaultHexagon ()
	Create default mutators for Hexagon. More...

Additional Inherited Members
Static Public Attributes inherited from tvm::runtime::ObjectRef
static constexpr bool	_type_is_nullable = true

Protected Member Functions inherited from tvm::runtime::ObjectRef
Object *	get_mutable () const

Static Protected Member Functions inherited from tvm::runtime::ObjectRef
template<typename T >
static T	DowncastNoCheck (ObjectRef ref)
	Internal helper function downcast a ref without check. More...

static void	FFIClearAfterMove (ObjectRef *ref)
	Clear the object ref data field without DecRef after we successfully moved the field. More...

template<typename ObjectType >
static ObjectPtr< ObjectType >	GetDataPtr (const ObjectRef &ref)
	Internal helper function get data_ as ObjectPtr of ObjectType. More...

Protected Attributes inherited from tvm::runtime::ObjectRef
ObjectPtr< Object >	data_
	Internal pointer that backs the reference. More...

Detailed Description

Managed reference to MutatorNode.

See also: MutatorNode

Member Typedef Documentation

◆ FApply

using tvm::meta_schedule::Mutator::FApply = runtime::TypedPackedFunc<Optional<tir::Trace>( const tir::Trace&, support::LinearCongruentialEngine::TRandState rand_state)>

Apply the mutator function to the given trace.

Parameters

trace The given trace for mutation.

Returns: None if mutator failed, otherwise return the mutated trace.

◆ FAsString

using tvm::meta_schedule::Mutator::FAsString = runtime::TypedPackedFunc<String()>

Get the mutator as string with name.

Returns: The string of the mutator.

◆ FClone

using tvm::meta_schedule::Mutator::FClone = runtime::TypedPackedFunc<Mutator()>

Clone the mutator.

Returns: The cloned mutator.

◆ FInitializeWithTuneContext

using tvm::meta_schedule::Mutator::FInitializeWithTuneContext = runtime::TypedPackedFunc<void(const TuneContext&)>

The function type of InitializeWithTuneContext method.

Parameters

context The tuning context for initialization.

Member Function Documentation

◆ DefaultCUDA()

static Map<Mutator, FloatImm, void> tvm::meta_schedule::Mutator::DefaultCUDA ( )

static

Create default mutators for CUDA.

◆ DefaultCUDATensorCore()

static Map<Mutator, FloatImm, void> tvm::meta_schedule::Mutator::DefaultCUDATensorCore ( )

static

Create default mutators for CUDA with TensorCore.

◆ DefaultHexagon()

static Map<Mutator, FloatImm, void> tvm::meta_schedule::Mutator::DefaultHexagon ( )

static

Create default mutators for Hexagon.

◆ DefaultLLVM()

static Map<Mutator, FloatImm, void> tvm::meta_schedule::Mutator::DefaultLLVM ( )

static

Create default mutators for LLVM.

◆ MutateComputeLocation()

static Mutator tvm::meta_schedule::Mutator::MutateComputeLocation ( )

static

Create a Mutator that mutates the outcome of SampleComputeLocation.

Returns: The mutator created

◆ MutateParallel()

static Mutator tvm::meta_schedule::Mutator::MutateParallel ( int64_t max_jobs_per_core )

static

Create a Mutator that mutates the parallel extent.

Parameters

max_jobs_per_core The maximum number of parallel jobs per core.

Returns: The created mutator.

◆ MutateThreadBinding()

static Mutator tvm::meta_schedule::Mutator::MutateThreadBinding ( )

static

Create a Mutator that mutates auto thread binding.

Returns: The mutator created

◆ MutateTileSize()

static Mutator tvm::meta_schedule::Mutator::MutateTileSize ( )

static

Create a Mutator that mutates the decision of instruction Sample-Perfect-Tile.

◆ MutateUnroll()

static Mutator tvm::meta_schedule::Mutator::MutateUnroll ( )

static

Create a Mutator that mutates auto unroll step.

Returns: The mutator created

◆ PyMutator()

static Mutator tvm::meta_schedule::Mutator::PyMutator	(	FInitializeWithTuneContext	f_initialize_with_tune_context,
		FApply	f_apply,
		FClone	f_clone,
		FAsString	f_as_string
	)

static

Create a mutator with customized methods on the python-side.

Parameters

f_initialize_with_tune_context	The packed function of `InitializeWithTuneContext`.
f_apply	The packed function of `Apply`.
f_clone	The packed function of `Clone`.
f_as_string	The packed function of `AsString`.

Returns: The mutator created.

◆ TVM_DEFINE_MUTABLE_OBJECT_REF_METHODS()

tvm::meta_schedule::Mutator::TVM_DEFINE_MUTABLE_OBJECT_REF_METHODS	(	Mutator	,
		ObjectRef	,
		MutatorNode
	)

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

include/tvm/meta_schedule/mutator.h

Public Types

Public Member Functions

Static Public Member Functions

Additional Inherited Members

Detailed Description

Member Typedef Documentation

◆ FApply

◆ FAsString

◆ FClone

◆ FInitializeWithTuneContext

Member Function Documentation

◆ DefaultCUDA()

◆ DefaultCUDATensorCore()

◆ DefaultHexagon()

◆ DefaultLLVM()

◆ MutateComputeLocation()

◆ MutateParallel()

◆ MutateThreadBinding()

◆ MutateTileSize()

◆ MutateUnroll()

◆ PyMutator()

◆ TVM_DEFINE_MUTABLE_OBJECT_REF_METHODS()