ITensorHandle Class Reference

#include <ITensorHandle.hpp>

Inheritance diagram for ITensorHandle:

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	Indicate to the memory manager that this resource is no longer active.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

Detailed Description

Definition at line 16 of file ITensorHandle.hpp.

Constructor & Destructor Documentation

~ITensorHandle()

virtual ~ITensorHandle ( )

inlinevirtual

Definition at line 19 of file ITensorHandle.hpp.

{}

Member Function Documentation

Allocate()

virtual void Allocate ( )

pure virtual

Indicate to the memory manager that this resource is no longer active.

This is used to compute overlapping lifetimes of resources.

Implemented in NeonTensorHandleDecorator, ClTensorHandleDecorator, ClImportSubTensorHandle, NeonSubTensorHandle, ClSubTensorHandle, GpuFsaSubTensorHandle, ConstPassthroughTensorHandle, PassthroughTensorHandle, ScopedTensorHandle, RefTensorHandleDecorator, NeonTensorHandle, ClTensorHandle, ClImportTensorHandle, GpuFsaTensorHandle, RefTensorHandle, TosaRefTensorHandle, and SampleTensorHandle.

Referenced by AllocateAndCopyDataToITensorHandle(), and Graph::AllocateDynamicBuffers().

CanBeImported()

virtual bool CanBeImported ( void *  memory, MemorySource  source  )

inlinevirtual

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 in ClTensorHandle, RefTensorHandleDecorator, NeonTensorHandle, RefTensorHandle, TosaRefTensorHandle, GpuFsaTensorHandle, and ClImportTensorHandle.

Definition at line 86 of file ITensorHandle.hpp.

    {
        IgnoreUnused(memory, source);
        return false;
    };

References armnn::IgnoreUnused().

CopyInFrom()

virtual void CopyInFrom ( const void *  memory )

pure virtual

Implemented in RefTensorHandleDecorator.

Referenced by CopyDataToITensorHandle().

CopyOutTo()

virtual void CopyOutTo ( void *  memory ) const

pure virtual

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

Implemented in RefTensorHandleDecorator.

Referenced by CopyDataFromITensorHandle().

DecorateTensorHandle()

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

inlinevirtual

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

Parameters

tensorInfo

the overidden TensorInfo.

Reimplemented in NeonTensorHandle, ClTensorHandle, RefTensorHandleDecorator, RefTensorHandle, NeonTensorHandleDecorator, and NeonSubTensorHandle.

Definition at line 98 of file ITensorHandle.hpp.

    {
        IgnoreUnused(tensorInfo);
        return nullptr;
    }

References armnn::IgnoreUnused().

GetImportFlags()

virtual unsigned int GetImportFlags ( ) const

inlinevirtual

Get flags describing supported import sources.

Reimplemented in RefTensorHandleDecorator, ClTensorHandle, GpuFsaTensorHandle, NeonTensorHandle, ClImportTensorHandle, RefTensorHandle, TosaRefTensorHandle, and SampleTensorHandle.

Definition at line 70 of file ITensorHandle.hpp.

{ return 0; }

Referenced by LoadedNetwork::ImportInputs(), and LoadedNetwork::ImportOutputs().

GetParent()

virtual ITensorHandle* GetParent ( ) const

pure virtual

Get the parent tensor if this is a subtensor.

Returns

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

Implemented in NeonTensorHandleDecorator, ClTensorHandleDecorator, ClImportSubTensorHandle, NeonSubTensorHandle, ClSubTensorHandle, GpuFsaSubTensorHandle, RefTensorHandleDecorator, ClTensorHandle, GpuFsaTensorHandle, NeonTensorHandle, ClImportTensorHandle, ConstTensorHandle, RefTensorHandle, TosaRefTensorHandle, and SampleTensorHandle.

Referenced by Graph::AllocateDynamicBuffers().

GetShape()

virtual TensorShape GetShape ( ) const

pure virtual

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.

Implemented in NeonTensorHandleDecorator, ClTensorHandleDecorator, ClImportSubTensorHandle, NeonSubTensorHandle, ClSubTensorHandle, GpuFsaSubTensorHandle, RefTensorHandleDecorator, ClTensorHandle, GpuFsaTensorHandle, NeonTensorHandle, ClImportTensorHandle, ConstTensorHandle, RefTensorHandle, TosaRefTensorHandle, and SampleTensorHandle.

Referenced by armnn::CopyTensorContentsGeneric(), and armnnUtils::FindMinMax().

GetStrides()

virtual TensorShape GetStrides ( ) const

pure virtual

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

Implemented in NeonTensorHandleDecorator, ClTensorHandleDecorator, ClImportSubTensorHandle, NeonSubTensorHandle, ClSubTensorHandle, GpuFsaSubTensorHandle, RefTensorHandleDecorator, ClTensorHandle, GpuFsaTensorHandle, NeonTensorHandle, ClImportTensorHandle, ConstTensorHandle, RefTensorHandle, TosaRefTensorHandle, and SampleTensorHandle.

Referenced by armnn::CopyTensorContentsGeneric().

Import()

virtual bool Import ( void *  memory, MemorySource  source  )

inlinevirtual

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 in NeonTensorHandle, ClTensorHandle, RefTensorHandleDecorator, ClImportTensorHandle, RefTensorHandle, TosaRefTensorHandle, SampleTensorHandle, and GpuFsaTensorHandle.

Definition at line 76 of file ITensorHandle.hpp.

    {
        IgnoreUnused(memory, source);
        return false;
    };

References armnn::IgnoreUnused().

Referenced by LoadedNetwork::ImportInputs(), and LoadedNetwork::ImportOutputs().

Manage()

virtual void Manage ( )

pure virtual

Indicate to the memory manager that this resource is active.

This is used to compute overlapping lifetimes of resources.

Implemented in NeonTensorHandleDecorator, ClTensorHandleDecorator, ClImportSubTensorHandle, NeonSubTensorHandle, ClSubTensorHandle, GpuFsaSubTensorHandle, RefTensorHandleDecorator, ClTensorHandle, NeonTensorHandle, GpuFsaTensorHandle, ClImportTensorHandle, ConstTensorHandle, RefTensorHandle, TosaRefTensorHandle, and SampleTensorHandle.

Referenced by Graph::AllocateDynamicBuffers().

Map() [1/2]

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 44 of file ITensorHandle.hpp.

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

References ITensorHandle::Map().

Map() [2/2]

virtual const void* Map ( bool  blocking = true ) const

pure virtual

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.

Implemented in NeonTensorHandleDecorator, NeonSubTensorHandle, RefTensorHandleDecorator, NeonTensorHandle, ConstTensorHandle, RefTensorHandle, TosaRefTensorHandle, SampleTensorHandle, ClTensorHandleDecorator, ClImportSubTensorHandle, ClSubTensorHandle, GpuFsaSubTensorHandle, ClTensorHandle, GpuFsaTensorHandle, and ClImportTensorHandle.

Referenced by armnn::CopyTensorContentsGeneric(), SyncMemGenericWorkload::Execute(), armnnUtils::FindMinMax(), sdb::GetInputTensorData(), armnn::GetInputTensorData(), sdb::GetOutputTensorData(), armnn::GetOutputTensorData(), armnn::InitializeArmComputeTensorData(), ITensorHandle::Map(), armnn::MirrorPad(), and armnn::Pad().

Unimport()

virtual void Unimport ( )

inlinevirtual

Unimport externally allocated memory.

Reimplemented in RefTensorHandleDecorator.

Definition at line 93 of file ITensorHandle.hpp.

    {};

Unmap() [1/2]

void Unmap ( )

inline

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

Definition at line 50 of file ITensorHandle.hpp.

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

References ITensorHandle::Unmap().

Unmap() [2/2]

virtual void Unmap ( ) const

pure virtual

Unmap the tensor data.

Implemented in NeonTensorHandleDecorator, ClTensorHandleDecorator, ClImportSubTensorHandle, NeonSubTensorHandle, ClSubTensorHandle, GpuFsaSubTensorHandle, RefTensorHandleDecorator, NeonTensorHandle, ClTensorHandle, GpuFsaTensorHandle, ClImportTensorHandle, ConstTensorHandle, RefTensorHandle, TosaRefTensorHandle, and SampleTensorHandle.

Referenced by armnn::CopyTensorContentsGeneric(), SyncMemGenericWorkload::Execute(), armnnUtils::FindMinMax(), and ITensorHandle::Unmap().

---

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

• include/armnn/backends/ITensorHandle.hpp