Template Numerical Library: TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator

Searching...

No Matches

Data structure for the Adaptive CSR segments format. More...

#include <TNL/Algorithms/Segments/AdaptiveCSR.h>

Inheritance diagram for TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >:

Collaboration diagram for TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >:

Public Types
using	BlocksArray = typename ViewType::BlocksArray
	Type for arrays of descriptor block of AdaptiveCSR.
using	BlocksType = typename ViewType::BlocksType
	Type for descriptor blocks of AdaptiveCSR.
using	BlocksView = typename BlocksType::ViewType
	Type for descriptor blocks view of AdaptiveCSR.
using	BlocksViewArray = typename ViewType::BlocksViewArray
	Type for arrays of descriptor block view of AdaptiveCSR.
using	ConstViewType = AdaptiveCSRView< Device, std::add_const_t< Index > >
	Type of constant segments view.
template<typename Device_ = Device, typename Index_ = Index, typename IndexAllocator_ = typename Allocators::Default< Device_ >::template Allocator< Index_ >>
using	Self = AdaptiveCSR< Device_, Index_, IndexAllocator_ >
	Templated type for creating AdaptiveCSR segments with different template parameters.
template<typename Device_, typename Index_>
using	ViewTemplate = AdaptiveCSRView< Device_, Index_ >
	Templated view type.
using	ViewType = AdaptiveCSRView< Device, Index >
	Type of segments view.
Public Types inherited from TNL::Algorithms::Segments::CSR< Device, Index, typename Allocators::Default< Device >::template Allocator< Index > >
using	ConstViewType
	Type of constant segments view.
using	IndexAllocatorType
using	OffsetsContainer
	Type of container storing offsets of particular segments.
using	Self
	Templated type for creating CSR segments with different template parameters.
using	ViewTemplate
	Templated view type.
using	ViewType
	Type of segments view.
Public Types inherited from TNL::Algorithms::Segments::CSRBase< Device, Index >
using	ConstOffsetsView = typename OffsetsView::ConstViewType
	The type for representing the constant vector view with segment offsets used in the CSR format.
using	DeviceType = Device
	The device where the segments are operating.
using	IndexType = std::remove_const_t< Index >
	The type used for indexing of segments elements.
using	OffsetsView = Containers::VectorView< Index, DeviceType, IndexType >
	The type for representing the vector view with segment offsets used in the CSR format.
using	SegmentViewType = SegmentView< IndexType, RowMajorOrder >
	Accessor type for one particular segment.

Public Member Functions
	AdaptiveCSR ()
	Constructor with no parameters to create empty segments.
	AdaptiveCSR (AdaptiveCSR &&) noexcept=default
	Move constructor.
	AdaptiveCSR (const AdaptiveCSR &segments)
	Copy constructor (makes deep copy).
template<typename SizesContainer, std::enable_if_t< IsArrayType< SizesContainer >::value, bool > = true>
	AdaptiveCSR (const SizesContainer &segmentsSizes)
	Constructor that initializes segments based on their sizes.
template<typename ListIndex>
	AdaptiveCSR (const std::initializer_list< ListIndex > &segmentsSizes)
	Constructor that initializes segments using an initializer list.
const BlocksViewArray &	getBlocks () const
	Returns a view with blocks used in the Adaptive CSR format.
ConstViewType	getConstView () const
	Returns a constant view for this instance of segments which can by used for example in lambda functions running in GPU kernels.
ViewType	getView ()
	Returns a view for this instance of segments which can by used for example in lambda functions running in GPU kernels.
void	load (File &file)
	Method for loading the segments from a file in a binary form.
AdaptiveCSR &	operator= (AdaptiveCSR &&) noexcept(false)
	Move-assignment operator.
AdaptiveCSR &	operator= (const AdaptiveCSR &segments)
	Copy-assignment operator (makes a deep copy).
template<typename Device_, typename Index_, typename IndexAllocator_>
AdaptiveCSR &	operator= (const AdaptiveCSR< Device_, Index_, IndexAllocator_ > &segments)
	Assignment operator for segments with different template parameters.
void	reset ()
	Reset the segments to empty states (it means that there is no segment in the segments).
void	save (File &file) const
	Method for saving the segments to a file in a binary form.
template<typename SizesContainer>
void	setSegmentsSizes (const SizesContainer &segmentsSizes)
	Set sizes of particular segments.
Public Member Functions inherited from TNL::Algorithms::Segments::CSR< Device, Index, typename Allocators::Default< Device >::template Allocator< Index > >
	CSR ()
	Constructor with no parameters to create empty segments.
ConstViewType	getConstView () const
	Returns a constant view for this instance of segments which can by used for example in lambda functions running in GPU kernels.
ViewType	getView ()
	Returns a view for this instance of segments which can by used for example in lambda functions running in GPU kernels.
void	load (File &file)
	Method for loading the segments from a file in a binary form.
CSR &	operator= (const CSR &segments)
	Copy-assignment operator (makes a deep copy).
void	reset ()
	Reset the segments to empty states (it means that there is no segment in the segments).
void	save (File &file) const
	Method for saving the segments to a file in a binary form.
void	setSegmentsSizes (const SizesContainer &segmentsSizes)
	Set sizes of particular segments.
Public Member Functions inherited from TNL::Algorithms::Segments::CSRBase< Device, Index >
__cuda_callable__	CSRBase ()=default
	Default constructor with no parameters to create empty segments view.
__cuda_callable__	CSRBase (const CSRBase &)=default
	Copy constructor.
__cuda_callable__	CSRBase (const OffsetsView &offsets)
	Binds a new CSR view to an offsets vector.
__cuda_callable__	CSRBase (CSRBase &&) noexcept=default
	Move constructor.
__cuda_callable__	CSRBase (OffsetsView &&offsets)
	Binds a new CSR view to an offsets vector.
template<typename Function>
void	forAllElements (Function function) const
	Call TNL::Algorithms::Segments::CSR::forElements for all elements of the segments.
template<typename Condition, typename Function>
void	forAllElementsIf (Condition condition, Function function) const
template<typename Function>
void	forAllSegments (Function &&function) const
	Call TNL::Algorithms::Segments::CSR::forSegments for all segments.
template<typename Array, typename Function>
void	forElements (const Array &segmentIndexes, Function function) const
template<typename Array, typename Function>
void	forElements (const Array &segmentIndexes, Index begin, Index end, Function function) const
template<typename Function>
void	forElements (IndexType begin, IndexType end, Function function) const
	Iterate over all elements of given segments in parallel and call given lambda function.
template<typename Condition, typename Function>
void	forElementsIf (IndexType begin, IndexType end, Condition condition, Function function) const
template<typename Function>
void	forSegments (IndexType begin, IndexType end, Function &&function) const
	Iterate over all segments in parallel and call given lambda function.
__cuda_callable__ IndexType	getElementCount () const
	Returns the number of elements managed by all segments.
__cuda_callable__ IndexType	getGlobalIndex (Index segmentIdx, Index localIdx) const
	Computes the global index of an element managed by the segments.
__cuda_callable__ OffsetsView	getOffsets ()
	Returns a modifiable vector view with segment offsets used in the CSR format.
__cuda_callable__ ConstOffsetsView	getOffsets () const
	Returns a constant vector view with segment offsets used in the CSR format.
__cuda_callable__ IndexType	getSegmentCount () const
	Returns the number of segments.
__cuda_callable__ IndexType	getSegmentsCount () const
	Returns the number of segments. Deprecated, use getSegmentCount().
__cuda_callable__ IndexType	getSegmentSize (IndexType segmentIdx) const
	Returns the size of a particular segment denoted by segmentIdx.
__cuda_callable__ SegmentViewType	getSegmentView (IndexType segmentIdx) const
	Returns a segment view (i.e., a segment accessor) for the specified segment index.
__cuda_callable__ IndexType	getSize () const
	Returns the number of elements managed by all segments.
__cuda_callable__ IndexType	getStorageSize () const
	Returns number of elements that needs to be allocated by a container connected to this segments.
CSRBase &	operator= (const CSRBase &)=delete
	Copy-assignment operator.
CSRBase &	operator= (CSRBase &&)=delete
	Move-assignment operator.
template<typename Function>
void	sequentialForAllSegments (Function &&function) const
	Call TNL::Algorithms::Segments::CSR::sequentialForSegments for all segments.
template<typename Function>
void	sequentialForSegments (IndexType begin, IndexType end, Function &&function) const
	Call TNL::Algorithms::Segments::CSR::forSegments sequentially for particular segments.

Static Public Member Functions
static std::string	getSegmentsType ()
	Returns string with the segments type.
static std::string	getSerializationType ()
	Returns string with the serialization type.
static constexpr int	MaxValueSizeLog ()
Static Public Member Functions inherited from TNL::Algorithms::Segments::CSRBase< Device, Index >
static constexpr ElementsOrganization	getOrganization ()
	Returns the data layout for the CSR format (it is always row-major order).
static std::string	getSegmentsType ()
	Returns string with the segments type.
static std::string	getSerializationType ()
	Returns string with the serialization type.
static constexpr bool	havePadding ()
	This function denotes that the CSR format does not use padding elements.

Protected Member Functions
template<int SizeOfValue, typename Offsets>
Index	findLimit (Index start, const Offsets &offsets, Index size, detail::Type &type)
void	initBlocks ()
	Initializes the blocks on GPUs and resets them on the other devices.
template<int SizeOfValue, typename Offsets>
void	initValueSize (const Offsets &offsets)
Protected Member Functions inherited from TNL::Algorithms::Segments::CSRBase< Device, Index >
__cuda_callable__ void	bind (OffsetsView offsets)
	Re-initializes the internal attributes of the base class.

Static Protected Member Functions
static int	getSizeValueLog (const int &i)

Protected Attributes
BlocksArray	blocksArray
	blocksArray[ i ] stores blocks for sizeof( Value ) == 2^i.
ViewType	view
Protected Attributes inherited from TNL::Algorithms::Segments::CSR< Device, Index, typename Allocators::Default< Device >::template Allocator< Index > >
OffsetsContainer	offsets
Protected Attributes inherited from TNL::Algorithms::Segments::CSRBase< Device, Index >
OffsetsView	offsets
	Vector view with row offsets used in the CSR format.

Detailed Description

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>
class TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >

Data structure for the Adaptive CSR segments format.

Adaptive CSR segments are inspired by the following paper:

[M. Daga and J. L. Greathouse, Structural Agnostic SpMV: Adapting CSR-Adaptive for Irregular Matrices, 2015 IEEE 22nd International Conference on High Performance Computing (HiPC), Bengaluru, India, 2015, pp. 64-74.](https://ieeexplore.ieee.org/document/7397620).

This format extends the Compressed Sparse Row (CSR) format. The data are organized similarly to CSR, but the segments contain additional meta-information to optimize GPU thread mapping. In particular, reduction operations can benefit from this extension. However, reading the meta-information introduces overhead, which may impact performance, especially for small segments. Use this segment type only if you anticipate significant variation in segment sizes.

See TNL::Algorithms::Segments for more details about segments.

Template Parameters

Device	The type of device on which the segments will operate.
Index	The type used for indexing the elements managed by the segments.
IndexAllocator	The allocator used for managing index containers.

Member Typedef Documentation

◆ Self

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

template<typename Device_ = Device, typename Index_ = Index, typename IndexAllocator_ = typename Allocators::Default< Device_ >::template Allocator< Index_ >>

using TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::Self = AdaptiveCSR< Device_, Index_, IndexAllocator_ >

Templated type for creating AdaptiveCSR segments with different template parameters.

Template Parameters

Device_	is alternative device type.
Index_	is alternative index type.
IndexAllocator_	is alternative index allocator type.

◆ ViewTemplate

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

template<typename Device_, typename Index_>

using TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::ViewTemplate = AdaptiveCSRView< Device_, Index_ >

Templated view type.

Template Parameters

Device_	is alternative device type for the view.
Index_	is alternative index type for the view.

Constructor & Destructor Documentation

◆ AdaptiveCSR() [1/2]

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

template<typename SizesContainer, std::enable_if_t< IsArrayType< SizesContainer >::value, bool > = true>

TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::AdaptiveCSR ( const SizesContainer & segmentsSizes )

Constructor that initializes segments based on their sizes.

The number of segments is determined by the size of segmentsSizes. Each element in this container specifies the size of a corresponding segment.

Template Parameters

SizesContainer The type of container used to store segment sizes. It can be, for example, TNL::Containers::Array or TNL::Containers::Vector.

Parameters

segmentsSizes An instance of the container holding the sizes of the segments.

See the following example:

#include <iostream>
#include <TNL/Containers/Vector.h>
#include <TNL/Algorithms/Segments/traverse.h>
#include <TNL/Devices/Host.h>
#include <TNL/Devices/Cuda.h>
 
template< typename Segments >
void
SegmentsExample()
{
   using Device = typename Segments::DeviceType;
 
   /***
    * Create segments with given segments sizes.
    */
   TNL::Containers::Vector< int, Device > segmentsSizes{ 1, 2, 3, 4, 5 };
   Segments segments( segmentsSizes );
   std::cout << "Segments sizes are: " << segments << '\n';
 
   /***
    * Allocate array for the segments;
    */
   TNL::Containers::Array< double, Device > data( segments.getStorageSize(), 0.0 );
 
   /***
    * Insert data into particular segments.
    */
   auto data_view = data.getView();
   TNL::Algorithms::Segments::forAllElements(
      segments,
      [ = ] __cuda_callable__( int segmentIdx, int localIdx, int globalIdx ) mutable
      {
         if( localIdx <= segmentIdx )
            data_view[ globalIdx ] = segmentIdx;
      } );
 
   /***
    * Print the data managed by the segments.
    */
   auto fetch = [ = ] __cuda_callable__( int globalIdx ) -> double
   {
      return data_view[ globalIdx ];
   };
   std::cout << TNL::Algorithms::Segments::print( segments, fetch ) << '\n';
}
 
int
main( int argc, char* argv[] )
{
   std::cout << "Example of CSR segments on host:\n";
   SegmentsExample< TNL::Algorithms::Segments::CSR< TNL::Devices::Host, int > >();
 
   std::cout << "Example of Ellpack segments on host:\n";
   SegmentsExample< TNL::Algorithms::Segments::Ellpack< TNL::Devices::Host, int > >();
 
#ifdef __CUDACC__
   std::cout << "Example of CSR segments on CUDA GPU:\n";
   SegmentsExample< TNL::Algorithms::Segments::CSR< TNL::Devices::Cuda, int > >();
 
   std::cout << "Example of Ellpack segments on CUDA GPU:\n";
   SegmentsExample< TNL::Algorithms::Segments::Ellpack< TNL::Devices::Cuda, int > >();
#endif
   return EXIT_SUCCESS;
}

The expected output is:

Example of CSR segments on host:
Segments sizes are:  [ 1, 2, 3, 4, 5 ] 
Segment 0: [ 0 ]
Segment 1: [ 1, 1 ]
Segment 2: [ 2, 2, 2 ]
Segment 3: [ 3, 3, 3, 3 ]
Segment 4: [ 4, 4, 4, 4, 4 ]
Example of Ellpack segments on host:
Segments sizes are:  [ 5, 5, 5, 5, 5 ] 
Segment 0: [ 0, 0, 0, 0, 0 ]
Segment 1: [ 1, 1, 0, 0, 0 ]
Segment 2: [ 2, 2, 2, 0, 0 ]
Segment 3: [ 3, 3, 3, 3, 0 ]
Segment 4: [ 4, 4, 4, 4, 4 ]
Example of CSR segments on CUDA GPU:
Segments sizes are:  [ 1, 2, 3, 4, 5 ] 
Segment 0: [ 0 ]
Segment 1: [ 1, 1 ]
Segment 2: [ 2, 2, 2 ]
Segment 3: [ 3, 3, 3, 3 ]
Segment 4: [ 4, 4, 4, 4, 4 ]
Example of Ellpack segments on CUDA GPU:
Segments sizes are:  [ 5, 5, 5, 5, 5 ] 
Segment 0: [ 0, 0, 0, 0, 0 ]
Segment 1: [ 1, 1, 0, 0, 0 ]
Segment 2: [ 2, 2, 2, 0, 0 ]
Segment 3: [ 3, 3, 3, 3, 0 ]
Segment 4: [ 4, 4, 4, 4, 4 ]

◆ AdaptiveCSR() [2/2]

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

template<typename ListIndex>

TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::AdaptiveCSR ( const std::initializer_list< ListIndex > & segmentsSizes )

Constructor that initializes segments using an initializer list.

The number of segments is determined by the size of segmentsSizes. Each element in this initializer list specifies the size of a corresponding segment.

Template Parameters

ListIndex The type used for indexing elements in the initializer list.

Parameters

segmentsSizes An initializer list defining the sizes of the segments.

See the following example:

#include <iostream>
#include <TNL/Containers/Vector.h>
#include <TNL/Algorithms/Segments/traverse.h>
#include <TNL/Devices/Host.h>
#include <TNL/Devices/Cuda.h>
 
template< typename Segments >
void
SegmentsExample()
{
   using Device = typename Segments::DeviceType;
 
   /***
    * Create segments with given segments sizes.
    */
   Segments segments{ 1, 2, 3, 4, 5 };
   std::cout << "Segments sizes are: " << segments << '\n';
 
   /***
    * Allocate array for the segments;
    */
   TNL::Containers::Array< double, Device > data( segments.getStorageSize(), 0.0 );
 
   /***
    * Insert data into particular segments.
    */
   auto data_view = data.getView();
   TNL::Algorithms::Segments::forAllElements(
      segments,
      [ = ] __cuda_callable__( int segmentIdx, int localIdx, int globalIdx ) mutable
      {
         if( localIdx <= segmentIdx )
            data_view[ globalIdx ] = segmentIdx;
      } );
 
   /***
    * Print the data managed by the segments.
    */
   auto fetch = [ = ] __cuda_callable__( int globalIdx ) -> double
   {
      return data_view[ globalIdx ];
   };
   std::cout << TNL::Algorithms::Segments::print( segments, fetch ) << '\n';
}
 
int
main( int argc, char* argv[] )
{
   std::cout << "Example of CSR segments on host:\n";
   SegmentsExample< TNL::Algorithms::Segments::CSR< TNL::Devices::Host, int > >();
 
   std::cout << "Example of Ellpack segments on host:\n";
   SegmentsExample< TNL::Algorithms::Segments::Ellpack< TNL::Devices::Host, int > >();
 
#ifdef __CUDACC__
   std::cout << "Example of CSR segments on CUDA GPU:\n";
   SegmentsExample< TNL::Algorithms::Segments::CSR< TNL::Devices::Cuda, int > >();
 
   std::cout << "Example of Ellpack segments on CUDA GPU:\n";
   SegmentsExample< TNL::Algorithms::Segments::Ellpack< TNL::Devices::Cuda, int > >();
#endif
   return EXIT_SUCCESS;
}

The expected output is:

Example of CSR segments on host:
Segments sizes are:  [ 1, 2, 3, 4, 5 ] 
Segment 0: [ 0 ]
Segment 1: [ 1, 1 ]
Segment 2: [ 2, 2, 2 ]
Segment 3: [ 3, 3, 3, 3 ]
Segment 4: [ 4, 4, 4, 4, 4 ]
Example of Ellpack segments on host:
Segments sizes are:  [ 5, 5, 5, 5, 5 ] 
Segment 0: [ 0, 0, 0, 0, 0 ]
Segment 1: [ 1, 1, 0, 0, 0 ]
Segment 2: [ 2, 2, 2, 0, 0 ]
Segment 3: [ 3, 3, 3, 3, 0 ]
Segment 4: [ 4, 4, 4, 4, 4 ]
Example of CSR segments on CUDA GPU:
Segments sizes are:  [ 1, 2, 3, 4, 5 ] 
Segment 0: [ 0 ]
Segment 1: [ 1, 1 ]
Segment 2: [ 2, 2, 2 ]
Segment 3: [ 3, 3, 3, 3 ]
Segment 4: [ 4, 4, 4, 4, 4 ]
Example of Ellpack segments on CUDA GPU:
Segments sizes are:  [ 5, 5, 5, 5, 5 ] 
Segment 0: [ 0, 0, 0, 0, 0 ]
Segment 1: [ 1, 1, 0, 0, 0 ]
Segment 2: [ 2, 2, 2, 0, 0 ]
Segment 3: [ 3, 3, 3, 3, 0 ]
Segment 4: [ 4, 4, 4, 4, 4 ]

Member Function Documentation

◆ getBlocks()

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

const BlocksViewArray & TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::getBlocks ( ) const

nodiscard

Returns a view with blocks used in the Adaptive CSR format.

The blocks are used only by the reduction on GPUs. On the other devices, the reduction falls back to the CSR kernels and so the blocks are empty.

◆ getSegmentsType()

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

std::string TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::getSegmentsType ( )

staticnodiscard

Returns string with the segments type.

Example: #include <iostream>

#include <functional>

#include <TNL/Algorithms/Segments/CSR.h>

#include <TNL/Algorithms/Segments/Ellpack.h>

#include <TNL/Devices/Host.h>

#include <TNL/Devices/Cuda.h>

template< typename Segments >

void

SegmentsExample()

{

/***

* Create segments and print the segments type.

*/

Segments segments;

std::cout << "The segments type is: " << segments.getSegmentsType() << '\n';

}

int

main( int argc, char* argv[] )

{

std::cout << "Example of CSR segments on host:\n";

SegmentsExample< TNL::Algorithms::Segments::CSR< TNL::Devices::Host, int > >();

std::cout << "Example of Ellpack segments on host:\n";

SegmentsExample< TNL::Algorithms::Segments::Ellpack< TNL::Devices::Host, int > >();

#ifdef __CUDACC__

std::cout << "Example of CSR segments on CUDA GPU:\n";

SegmentsExample< TNL::Algorithms::Segments::CSR< TNL::Devices::Cuda, int > >();

std::cout << "Example of Ellpack segments on CUDA GPU:\n";

SegmentsExample< TNL::Algorithms::Segments::Ellpack< TNL::Devices::Cuda, int > >();

#endif

return EXIT_SUCCESS;

}

Output: Example of CSR segments on host:

The segments type is: CSR

Example of Ellpack segments on host:

The segments type is: Ellpack

Example of CSR segments on CUDA GPU:

The segments type is: CSR

Example of Ellpack segments on CUDA GPU:

The segments type is: Ellpack

◆ getSerializationType()

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

std::string TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::getSerializationType ( )

staticnodiscard

Returns string with the serialization type.

Example: #include <iostream>

#include <functional>

#include <TNL/Algorithms/Segments/CSR.h>

#include <TNL/Algorithms/Segments/Ellpack.h>

#include <TNL/Devices/Host.h>

#include <TNL/Devices/Cuda.h>

template< typename Segments >

void

SegmentsExample()

{

/***

* Create segments and print the serialization type.

*/

Segments segments;

std::cout << "The serialization type is: " << segments.getSerializationType() << '\n';

}

int

main( int argc, char* argv[] )

{

std::cout << "Example of CSR segments on host:\n";

SegmentsExample< TNL::Algorithms::Segments::CSR< TNL::Devices::Host, int > >();

std::cout << "Example of Ellpack segments on host:\n";

SegmentsExample< TNL::Algorithms::Segments::Ellpack< TNL::Devices::Host, int > >();

#ifdef __CUDACC__

std::cout << "Example of CSR segments on CUDA GPU:\n";

SegmentsExample< TNL::Algorithms::Segments::CSR< TNL::Devices::Cuda, int > >();

std::cout << "Example of Ellpack segments on CUDA GPU:\n";

SegmentsExample< TNL::Algorithms::Segments::Ellpack< TNL::Devices::Cuda, int > >();

#endif

return EXIT_SUCCESS;

}

Output: Example of CSR segments on host:

The serialization type is: CSR< int >

Example of Ellpack segments on host:

The serialization type is: Ellpack< int, RowMajorOrder, 32 >

Example of CSR segments on CUDA GPU:

The serialization type is: CSR< int >

Example of Ellpack segments on CUDA GPU:

The serialization type is: Ellpack< int, ColumnMajorOrder, 32 >

◆ load()

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

void TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::load ( File & file )

Method for loading the segments from a file in a binary form.

Parameters

file	is the source file.

◆ operator=()

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

template<typename Device_, typename Index_, typename IndexAllocator_>

AdaptiveCSR & TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::operator= ( const AdaptiveCSR< Device_, Index_, IndexAllocator_ > & segments )

Assignment operator for segments with different template parameters.

Performs a deep copy of the source segments.

Template Parameters

Device_	The device type of the source segments.
Index_	The index type of the source segments.
IndexAllocator_	The index allocator type of the source segments.

Parameters

segments The source segments object.

Returns: A reference to this instance.

◆ save()

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

void TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::save ( File & file ) const

Method for saving the segments to a file in a binary form.

Parameters

file	is the target file.

◆ setSegmentsSizes()

template<typename Device, typename Index, typename IndexAllocator = typename Allocators::Default< Device >::template Allocator< Index >>

template<typename SizesContainer>

void TNL::Algorithms::Segments::AdaptiveCSR< Device, Index, IndexAllocator >::setSegmentsSizes ( const SizesContainer & segmentsSizes )

Set sizes of particular segments.

Template Parameters

SizesContainer is a container with segments sizes. It can be TNL::Containers::Array or TNL::Containers::Vector for example.

Parameters

segmentsSizes is an instance of the container with segments sizes.

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

src/TNL/Algorithms/Segments/AdaptiveCSR.h

Public Types

Public Member Functions

Static Public Member Functions

Protected Member Functions

Static Protected Member Functions

Protected Attributes

Detailed Description

Member Typedef Documentation

◆ Self

◆ ViewTemplate

Constructor & Destructor Documentation

◆ AdaptiveCSR() [1/2]

◆ AdaptiveCSR() [2/2]

Member Function Documentation

◆ getBlocks()

◆ getSegmentsType()

◆ getSerializationType()

◆ load()

◆ operator=()

◆ save()

◆ setSegmentsSizes()