CLTensor Class Reference

Basic implementation of the OpenCL tensor interface. More...

#include <CLTensor.h>

Collaboration diagram for CLTensor:

Public Member Functions
	CLTensor (IRuntimeContext *ctx=nullptr)
	Constructor. More...
	~CLTensor ()=default
	Destructor. More...
	CLTensor (const CLTensor &)=delete
	Prevent copying by construction. More...
	CLTensor (CLTensor &&)=default
	Default move constructor. More...
CLTensor &	operator= (const CLTensor &)=delete
	Prevent copaingy by assignment. More...
CLTensor &	operator= (CLTensor &&)=default
	Default move assignment operator. More...
CLTensorAllocator *	allocator ()
	Return a pointer to the tensor's allocator. More...
void	map (bool blocking=true)
	Enqueue a map operation of the allocated buffer. More...
void	unmap ()
	Enqueue an unmap operation of the allocated and mapped buffer. More...
TensorInfo *	info () const override
	Interface to be implemented by the child class to return the tensor's metadata. More...
TensorInfo *	info () override
	Interface to be implemented by the child class to return the tensor's metadata. More...
const cl::Buffer &	cl_buffer () const override
	Interface to be implemented by the child class to return a reference to the OpenCL buffer containing the image's data. More...
CLQuantization	quantization () const override
	Interface to be implemented by the child class to return the wrapped quantization info data. More...
void	associate_memory_group (IMemoryGroup *memory_group) override
	Associates a memory managable object with the memory group that manages it. More...
CLRuntimeContext *	context ()
void	map (cl::CommandQueue &q, bool blocking=true)
	Enqueue a map operation of the allocated buffer on the given queue. More...
void	unmap (cl::CommandQueue &q)
	Enqueue an unmap operation of the allocated and mapped buffer on the given queue. More...
Public Member Functions inherited from ICLTensor
	ICLTensor ()
	Default constructor. More...
	ICLTensor (const ICLTensor &)=delete
	Prevent instances of this class from being copy constructed. More...
ICLTensor &	operator= (const ICLTensor &)=delete
	Prevent instances of this class from being copied. More...
	ICLTensor (ICLTensor &&)=default
	Allow instances of this class to be move constructed. More...
ICLTensor &	operator= (ICLTensor &&)=default
	Allow instances of this class to be copied. More...
virtual	~ICLTensor ()=default
	Default virtual destructor. More...
void	map (cl::CommandQueue &q, bool blocking=true)
	Enqueue a map operation of the allocated buffer on the given queue. More...
void	unmap (cl::CommandQueue &q)
	Enqueue an unmap operation of the allocated and mapped buffer on the given queue. More...
void	clear (cl::CommandQueue &q)
	Clear the contents of the tensor synchronously. More...
uint8_t *	buffer () const override
	Interface to be implemented by the child class to return a pointer to CPU memory. More...
Public Member Functions inherited from ITensor
virtual	~ITensor ()=default
	Default virtual destructor. More...
uint8_t *	ptr_to_element (const Coordinates &id) const
	Return a pointer to the element at the passed coordinates. More...
void	copy_from (const ITensor &src)
	Copy the content of another tensor. More...
void	print (std::ostream &s, IOFormatInfo io_fmt=IOFormatInfo()) const
	Print a tensor to a given stream using user defined formatting information. More...
bool	is_used () const
	Flags if the tensor is used or not. More...
void	mark_as_unused () const
	Marks a tensor as unused. More...
void	mark_as_used () const
	Marks a tensor as used. More...
Public Member Functions inherited from IMemoryManageable
virtual	~IMemoryManageable ()=default
	Default virtual destructor. More...

Detailed Description

Basic implementation of the OpenCL tensor interface.

Definition at line 41 of file CLTensor.h.

Constructor & Destructor Documentation

CLTensor() [1/3]

CLTensor

(

IRuntimeContext *

ctx = nullptr

)

Constructor.

Parameters

[in]

ctx

(Optional) Pointer to a CLRuntimeContext. If nullptr is passed in, the legacy api using the singletons will be used. Otherwise the memory for the tensor will allocate on the context passed in. The singletons legacy api has been deprecated and will be removed in future releases.

Definition at line 31 of file CLTensor.cpp.

    : _allocator(this, static_cast<CLRuntimeContext *>(ctx)), _ctx(static_cast<CLRuntimeContext *>(ctx))
{
}

~CLTensor()

~CLTensor ( )

default

Destructor.

CLTensor() [2/3]

CLTensor ( const CLTensor &  )

delete

Prevent copying by construction.

CLTensor() [3/3]

CLTensor ( CLTensor &&  )

default

Default move constructor.

Member Function Documentation

allocator()

CLTensorAllocator * allocator

(

)

Return a pointer to the tensor's allocator.

Returns

A pointer to the tensor's allocator

Definition at line 61 of file CLTensor.cpp.

{
    return &_allocator;
}

Referenced by CLTensorHandle::allocate(), CLAuxTensorHandler::CLAuxTensorHandler(), CLTensorHandle::CLTensorHandle(), arm_compute::test::validation::compute_float_target_in_place(), CLReduceMean::configure(), CLFFT2D::configure(), CLRNNLayer::configure(), CLFFT1D::configure(), CLDepthwiseConvolutionLayer::configure(), CLL2NormalizeLayer::configure(), CLReductionOperation::configure(), CLArgMinMaxLayer::configure(), CLInstanceNormalizationLayer::configure(), CLFFTConvolutionLayer::configure(), CLGenerateProposalsLayer::configure(), CLGEMMDeconvolutionLayer::configure(), CLDirectDeconvolutionLayer::configure(), CLLSTMLayerQuantized::configure(), CLLSTMLayer::configure(), CLQLSTMLayer::configure(), CLTensorHandle::free(), arm_compute::manage_workspace(), CLDepthwiseConvolutionLayer::prepare(), CLFFTConvolutionLayer::prepare(), CLGEMMDeconvolutionLayer::prepare(), CLDirectDeconvolutionLayer::prepare(), CLLSTMLayerQuantized::prepare(), CLQLSTMLayer::prepare(), CLConvertFullyConnectedWeightsManaged::release(), CLTensorHandle::release_if_unused(), CLConvertFullyConnectedWeightsManaged::run(), CLFFTConvolutionLayer::run(), and arm_compute::test::validation::TEST_CASE().

associate_memory_group()

void associate_memory_group ( IMemoryGroup *  memory_group )

overridevirtual

Associates a memory managable object with the memory group that manages it.

Parameters

[in]

memory_group

Memory group that manages the object.

Implements IMemoryManageable.

Definition at line 86 of file CLTensor.cpp.

{
    _allocator.set_associated_memory_group(memory_group);
}

References CLTensorAllocator::set_associated_memory_group().

cl_buffer()

const cl::Buffer & cl_buffer ( ) const

overridevirtual

Interface to be implemented by the child class to return a reference to the OpenCL buffer containing the image's data.

Returns

A reference to an OpenCL buffer containing the image's data.

Implements ICLTensor.

Definition at line 51 of file CLTensor.cpp.

{
    return _allocator.cl_data();
}

References CLTensorAllocator::cl_data().

Referenced by arm_compute::create_image2d_from_tensor(), CLFFTConvolutionLayer::run(), and arm_compute::test::validation::TEST_CASE().

context()

CLRuntimeContext * context

(

)

Definition at line 36 of file CLTensor.cpp.

{
    return _ctx;
}

info() [1/2]

TensorInfo * info ( ) const

overridevirtual

Interface to be implemented by the child class to return the tensor's metadata.

Returns

A pointer to the tensor's metadata.

Implements ITensor.

Definition at line 41 of file CLTensor.cpp.

{
    return &_allocator.info();
}

References ITensorAllocator::info().

Referenced by CLReduceMean::configure(), CLDepthwiseConvolutionLayer::configure(), CLArgMinMaxLayer::configure(), CLFFTConvolutionLayer::configure(), CLGenerateProposalsLayer::configure(), CLGEMMDeconvolutionLayer::configure(), CLLSTMLayerQuantized::configure(), CLLSTMLayer::configure(), CLQLSTMLayer::configure(), arm_compute::create_image2d_from_tensor(), CLAccessor::data_layout(), CLAccessor::data_type(), CLAccessor::element_size(), arm_compute::error_on_data_layout_not_in(), arm_compute::error_on_data_type_channel_not_in(), arm_compute::error_on_data_type_not_in(), arm_compute::error_on_mismatching_data_layouts(), arm_compute::error_on_mismatching_data_types(), arm_compute::error_on_unsupported_cpu_bf16(), arm_compute::error_on_unsupported_cpu_fp16(), arm_compute::error_on_unsupported_fp16(), CLAccessor::format(), arm_compute::get_padding_info(), CLAccessor::num_channels(), CLAccessor::num_elements(), CLAccessor::padding(), CLQLSTMLayer::prepare(), CLAccessor::quantization_info(), CLAccessor::shape(), CLAccessor::size(), arm_compute::test::validation::TEST_CASE(), and arm_compute::to_string().

info() [2/2]

TensorInfo* info ( )

overridevirtual

Interface to be implemented by the child class to return the tensor's metadata.

Returns

A pointer to the tensor's metadata.

Implements ITensor.

map() [1/2]

void map

(

bool

blocking = true

)

Enqueue a map operation of the allocated buffer.

Parameters

[in]

blocking

If true, then the mapping will be ready to use by the time this method returns, else it is the caller's responsibility to flush the queue and wait for the mapping operation to have completed.

Definition at line 66 of file CLTensor.cpp.

{
    ICLTensor::map(_ctx == nullptr ? CLScheduler::get().queue() : _ctx->gpu_scheduler()->queue(), blocking);
}

References CLScheduler::get(), CLRuntimeContext::gpu_scheduler(), ICLTensor::map(), and CLScheduler::queue().

Referenced by AclMapTensor(), CLFFT1D::configure(), CLFFTConvolutionLayer::configure(), CLDirectDeconvolutionLayer::configure(), CLTensorHandle::map(), CLDepthwiseConvolutionLayer::prepare(), and CLQLSTMLayer::prepare().

map() [2/2]

void map

Enqueue a map operation of the allocated buffer on the given queue.

Parameters

[in,out]	q	The CL command queue to use for the mapping operation.
[in]	blocking	If true, then the mapping will be ready to use by the time this method returns, else it is the caller's responsibility to flush the queue and wait for the mapping operation to have completed before using the returned mapping pointer.

Definition at line 35 of file ICLTensor.cpp.

{
    _mapping = do_map(q, blocking);
}

operator=() [1/2]

CLTensor& operator= ( CLTensor &&  )

default

Default move assignment operator.

operator=() [2/2]

CLTensor& operator= ( const CLTensor &  )

delete

Prevent copaingy by assignment.

quantization()

CLQuantization quantization ( ) const

overridevirtual

Interface to be implemented by the child class to return the wrapped quantization info data.

Returns

A wrapped quantization info object.

Implements ICLTensor.

Definition at line 56 of file CLTensor.cpp.

{
    return _allocator.quantization();
}

References CLTensorAllocator::quantization().

unmap() [1/2]

void unmap

(

)

Enqueue an unmap operation of the allocated and mapped buffer.

Note

This method simply enqueues the unmap operation, it is the caller's responsibility to flush the queue and make sure the unmap is finished before the memory is accessed by the device.

Definition at line 71 of file CLTensor.cpp.

{
    ICLTensor::unmap(_ctx == nullptr ? CLScheduler::get().queue() : _ctx->gpu_scheduler()->queue());
}

References CLScheduler::get(), CLRuntimeContext::gpu_scheduler(), CLScheduler::queue(), and ICLTensor::unmap().

Referenced by AclUnmapTensor(), CLFFT1D::configure(), CLFFTConvolutionLayer::configure(), CLDirectDeconvolutionLayer::configure(), CLDepthwiseConvolutionLayer::prepare(), CLQLSTMLayer::prepare(), CLTensorHandle::unmap(), and CLAccessor::~CLAccessor().

unmap() [2/2]

void unmap

Enqueue an unmap operation of the allocated and mapped buffer on the given queue.

Note

This method simply enqueues the unmap operation, it is the caller's responsibility to flush the queue and make sure the unmap is finished before the memory is accessed by the device.

Parameters

[in,out]

q

The CL command queue to use for the mapping operation.

Definition at line 40 of file ICLTensor.cpp.

{
    do_unmap(q);
    _mapping = nullptr;
}

---

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

• arm_compute/runtime/CL/CLTensor.h
• src/runtime/CL/CLTensor.cpp