RefTensorHandleDecorator Class Reference

#include <RefTensorHandle.hpp>

Inheritance diagram for RefTensorHandleDecorator:

Collaboration diagram for RefTensorHandleDecorator:

Public Member Functions
	RefTensorHandleDecorator (const TensorInfo &tensorInfo, const RefTensorHandle &parent)
	~RefTensorHandleDecorator ()=default
virtual void	Manage () override
	Indicate to the memory manager that this resource is active. More...
virtual void	Allocate () override
	Indicate to the memory manager that this resource is no longer active. More...
virtual ITensorHandle *	GetParent () const override
	Get the parent tensor if this is a subtensor. More...
virtual const void *	Map (bool) const override
	Map the tensor data for access. More...
virtual void	Unmap () const override
	Unmap the tensor data. More...
TensorShape	GetStrides () const override
	Get the strides for each dimension ordered from largest to smallest where the smallest value is the same as the size of a single element in the tensor. More...
TensorShape	GetShape () const override
	Get the number of elements for each dimension ordered from slowest iterating dimension to fastest iterating dimension. More...
const TensorInfo &	GetTensorInfo () const
virtual MemorySourceFlags	GetImportFlags () const override
	Get flags describing supported import sources. More...
virtual bool	Import (void *memory, MemorySource source) override
	Import externally allocated memory. More...
virtual bool	CanBeImported (void *memory, MemorySource source) override
	Implementations must determine if this memory block can be imported. More...
virtual std::shared_ptr< ITensorHandle >	DecorateTensorHandle (const TensorInfo &tensorInfo) override
	Returns a decorated version of this TensorHandle allowing us to override the TensorInfo for it. More...
void *	Map (bool blocking=true)
	Map the tensor data for access. More...
void	Unmap ()
	Unmap the tensor data that was previously mapped with call to Map(). More...
void	CopyOutTo (void *) const override
	Testing support to be able to verify and set tensor data content. More...
void	CopyInFrom (const void *) override
void	Unimport () override
	Unimport externally allocated memory. More...
virtual const void *	Map (bool blocking=true) const=0
	Map the tensor data for access. More...
void *	Map (bool blocking=true)
	Map the tensor data for access. More...
Public Member Functions inherited from RefTensorHandle
	RefTensorHandle (const TensorInfo &tensorInfo, std::shared_ptr< RefMemoryManager > &memoryManager)
	RefTensorHandle (const TensorInfo &tensorInfo)
	RefTensorHandle (const TensorInfo &tensorInfo, const RefTensorHandle &parent)
	~RefTensorHandle ()
const TensorInfo &	GetTensorInfo () const
virtual const void *	Map (bool blocking=true) const=0
	Map the tensor data for access. More...
void *	Map (bool blocking=true)
	Map the tensor data for access. More...
Public Member Functions inherited from ITensorHandle
virtual	~ITensorHandle ()
void *	Map (bool blocking=true)
	Map the tensor data for access. More...
void	Unmap ()
	Unmap the tensor data that was previously mapped with call to Map(). More...

Detailed Description

Definition at line 84 of file RefTensorHandle.hpp.

Constructor & Destructor Documentation

RefTensorHandleDecorator()

RefTensorHandleDecorator	(	const TensorInfo &	tensorInfo,
		const RefTensorHandle &	parent
	)

Definition at line 174 of file RefTensorHandle.cpp.

: RefTensorHandle(tensorInfo)
, m_TensorInfo(tensorInfo)
, m_Parent(parent)
{
}

~RefTensorHandleDecorator()

~RefTensorHandleDecorator ( )

default

Member Function Documentation

Allocate()

void Allocate ( )

overridevirtual

Indicate to the memory manager that this resource is no longer active.

This is used to compute overlapping lifetimes of resources.

Reimplemented from RefTensorHandle.

Definition at line 185 of file RefTensorHandle.cpp.

{
}

CanBeImported()

bool CanBeImported ( void *  memory, MemorySource  source  )

overridevirtual

Implementations must determine if this memory block can be imported.

This might be based on alignment or memory source type.

Returns

true if this memory can be imported.

false by default, cannot be imported.

Reimplemented from RefTensorHandle.

Definition at line 204 of file RefTensorHandle.cpp.

{
    return false;
}

CopyInFrom()

void CopyInFrom ( const void *  )

inlineoverridevirtual

Reimplemented from RefTensorHandle.

Definition at line 146 of file RefTensorHandle.hpp.

    {};

CopyOutTo()

void CopyOutTo ( void *  ) const

inlineoverridevirtual

Testing support to be able to verify and set tensor data content.

Reimplemented from RefTensorHandle.

Definition at line 143 of file RefTensorHandle.hpp.

    {};

DecorateTensorHandle()

std::shared_ptr< ITensorHandle > DecorateTensorHandle ( const TensorInfo &  tensorInfo )

overridevirtual

Returns a decorated version of this TensorHandle allowing us to override the TensorInfo for it.

Parameters

tensorInfo

the overidden TensorInfo.

Reimplemented from RefTensorHandle.

Definition at line 209 of file RefTensorHandle.cpp.

{
    return nullptr;
}

GetImportFlags()

MemorySourceFlags GetImportFlags ( ) const

overridevirtual

Get flags describing supported import sources.

Reimplemented from RefTensorHandle.

Definition at line 194 of file RefTensorHandle.cpp.

{
    return static_cast<MemorySourceFlags>(MemorySource::Malloc);
}

References armnn::Malloc.

GetParent()

virtual ITensorHandle* GetParent ( ) const

inlineoverridevirtual

Get the parent tensor if this is a subtensor.

Returns

a pointer to the parent tensor. Otherwise nullptr if not a subtensor.

Reimplemented from RefTensorHandle.

Definition at line 95 of file RefTensorHandle.hpp.

    {
        return nullptr;
    }

GetShape()

TensorShape GetShape ( ) const

inlineoverridevirtual

Get the number of elements for each dimension ordered from slowest iterating dimension to fastest iterating dimension.

Returns

a TensorShape filled with the number of elements for each dimension.

Reimplemented from RefTensorHandle.

Definition at line 111 of file RefTensorHandle.hpp.

    {
        return m_TensorInfo.GetShape();
    }

References TensorInfo::GetShape().

GetStrides()

TensorShape GetStrides ( ) const

inlineoverridevirtual

Get the strides for each dimension ordered from largest to smallest where the smallest value is the same as the size of a single element in the tensor.

Returns

a TensorShape filled with the strides for each dimension

Reimplemented from RefTensorHandle.

Definition at line 106 of file RefTensorHandle.hpp.

    {
        return GetUnpaddedTensorStrides(m_TensorInfo);
    }

References armnn::GetUnpaddedTensorStrides().

GetTensorInfo()

const TensorInfo& GetTensorInfo ( ) const

inline

Definition at line 116 of file RefTensorHandle.hpp.

    {
        return m_TensorInfo;
    }

Import()

bool Import ( void *  memory, MemorySource  source  )

overridevirtual

Import externally allocated memory.

Parameters

memory	base address of the memory being imported.
source	source of the allocation for the memory being imported.

Returns

true on success or false on failure

Reimplemented from RefTensorHandle.

Definition at line 199 of file RefTensorHandle.cpp.

{
    return false;
}

Manage()

void Manage ( )

overridevirtual

Indicate to the memory manager that this resource is active.

This is used to compute overlapping lifetimes of resources.

Reimplemented from RefTensorHandle.

Definition at line 181 of file RefTensorHandle.cpp.

{
}

Map() [1/4]

void* Map

inline

Map the tensor data for access.

Must be paired with call to Unmap().

Parameters

blocking

hint to block the calling thread until all other accesses are complete. (backend dependent)

Returns

pointer to the first element of the mapped data.

Definition at line 44 of file ITensorHandle.hpp.

    {
        return const_cast<void*>(static_cast<const ITensorHandle*>(this)->Map(blocking));
    }

Map() [2/4]

void* Map ( bool  blocking = true )

inline

Map the tensor data for access.

Must be paired with call to Unmap().

Parameters

blocking

hint to block the calling thread until all other accesses are complete. (backend dependent)

Returns

pointer to the first element of the mapped data.

Definition at line 131 of file RefTensorHandle.hpp.

    {
        return const_cast<void*>(static_cast<const ITensorHandle*>(this)->Map(blocking));
    }

References RefTensorHandleDecorator::Map().

Map() [3/4]

virtual const void* Map

Map the tensor data for access.

Parameters

blocking

hint to block the calling thread until all other accesses are complete. (backend dependent)

Returns

pointer to the first element of the mapped data.

Map() [4/4]

const void * Map ( bool  blocking ) const

overridevirtual

Map the tensor data for access.

Parameters

blocking

hint to block the calling thread until all other accesses are complete. (backend dependent)

Returns

pointer to the first element of the mapped data.

Reimplemented from RefTensorHandle.

Definition at line 189 of file RefTensorHandle.cpp.

{
    return m_Parent.Map(unused);
}

References RefTensorHandle::Map().

Referenced by RefTensorHandleDecorator::Map().

Unimport()

void Unimport ( )

inlineoverridevirtual

Unimport externally allocated memory.

Reimplemented from ITensorHandle.

Definition at line 150 of file RefTensorHandle.hpp.

    {};

Unmap() [1/2]

void Unmap ( )

inline

Unmap the tensor data that was previously mapped with call to Map().

Definition at line 137 of file RefTensorHandle.hpp.

    {
        return static_cast<const ITensorHandle*>(this)->Unmap();
    }

References RefTensorHandleDecorator::Unmap().

Unmap() [2/2]

virtual void Unmap ( ) const

inlineoverridevirtual

Unmap the tensor data.

Reimplemented from RefTensorHandle.

Definition at line 103 of file RefTensorHandle.hpp.

    {}

Referenced by RefTensorHandleDecorator::Unmap().

---

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

• src/backends/reference/RefTensorHandle.hpp
• src/backends/reference/RefTensorHandle.cpp