cl_algo::RBC::Scan< C, T > Class Template Reference

Interface class for the scan kernel. More...

#include <algorithms.hpp>

Collaboration diagram for cl_algo::RBC::Scan< C, T >:

Public Types
enum	Memory : uint8_t {   Memory::H_IN, Memory::H_OUT, Memory::D_IN, Memory::D_SUMS,   Memory::D_OUT }
	Enumerates the memory objects handled by the class. More...

Public Member Functions
	Scan (clutils::CLEnv &_env, clutils::CLEnvInfo< 1 > _info)
	Configures an OpenCL environment as specified by _info. More...
cl::Memory &	get (Scan::Memory mem)
	Returns a reference to an internal memory object. More...
void	init (unsigned int _cols, unsigned int _rows, Staging _staging=Staging::IO)
	Configures kernel execution parameters. More...
void	write (Scan::Memory mem=Scan::Memory::D_IN, void *ptr=nullptr, bool block=CL_FALSE, const std::vector< cl::Event > *events=nullptr, cl::Event *event=nullptr)
	Performs a data transfer to a device buffer. More...
void *	read (Scan::Memory mem=Scan::Memory::H_OUT, bool block=CL_TRUE, const std::vector< cl::Event > *events=nullptr, cl::Event *event=nullptr)
	Performs a data transfer to a staging buffer. More...
void	run (const std::vector< cl::Event > *events=nullptr, cl::Event *event=nullptr)
	Executes the necessary kernels. More...
template<typename period >
double	run (clutils::GPUTimer< period > &timer, const std::vector< cl::Event > *events=nullptr)
	Executes the necessary kernels. More...
template<>
	Scan (clutils::CLEnv &_env, clutils::CLEnvInfo< 1 > _info)
template<>
	Scan (clutils::CLEnv &_env, clutils::CLEnvInfo< 1 > _info)

Public Attributes
T *	hPtrIn
T *	hPtrOut

Detailed Description

template<ScanConfig C, typename T = cl_int>
class cl_algo::RBC::Scan< C, T >

Interface class for the scan kernel.

scan performs a scan operation on each row in an array. For more details, look at the kernel's documentation.

Note

The scan kernel is available in kernels/scan_kernels.cl.

The class creates its own buffers. If you would like to provide your own buffers, call get to get references to the placeholders within the class and assign them to your buffers. You will have to do this strictly before the call to init. You can also call get (after the call to init) to get a reference to a buffer within the class and assign it to another kernel class instance further down in your task pipeline.

The following input/output OpenCL memory objects are created by a Scan instance:

Name	Type	Placement	I/O	Use	Properties	Size
H_IN	Buffer	Host	I	Staging	CL_MEM_READ_WRITE	\(width*height*sizeof\ (cl\_int)\)
H_OUT	Buffer	Host	O	Staging	CL_MEM_READ_WRITE	\(width*height*sizeof\ (cl\_int)\)
D_IN	Buffer	Device	I	Processing	CL_MEM_READ_ONLY	\(width*height*sizeof\ (cl\_int)\)
D_OUT	Buffer	Device	O	Processing	CL_MEM_READ_WRITE	\(width*height*sizeof\ (cl\_int)\)

Template Parameters

C	configures the class to perform either inclusive or exclusive scan.
T	configures the class to work with different types of data.

Member Enumeration Documentation

template<ScanConfig C, typename T = cl_int>

enum cl_algo::RBC::Scan::Memory : uint8_t

strong

Enumerates the memory objects handled by the class.

Note

H_* names refer to staging buffers on the host.

D_* names refer to buffers on the device.

Enumerator
H_IN	Input staging buffer.
H_OUT	Output staging buffer.
D_IN	Input buffer.
D_SUMS	Buffer of partial group sums.
D_OUT	Output buffer.

Constructor & Destructor Documentation

template<ScanConfig C, typename T = cl_int>

cl_algo::RBC::Scan< C, T >::Scan	(	clutils::CLEnv &	_env,
		clutils::CLEnvInfo< 1 >	_info
	)

Configures an OpenCL environment as specified by _info.

template<>

cl_algo::RBC::Scan< ScanConfig::INCLUSIVE, cl_int >::Scan	(	clutils::CLEnv &	_env,
		clutils::CLEnvInfo< 1 >	_info
	)

Parameters

[in]	_env	opencl environment.
[in]	_info	opencl configuration. Specifies the context, queue, etc, to be used.

template<>

cl_algo::RBC::Scan< ScanConfig::EXCLUSIVE, cl_int >::Scan	(	clutils::CLEnv &	_env,
		clutils::CLEnvInfo< 1 >	_info
	)

Parameters

[in]	_env	opencl environment.
[in]	_info	opencl configuration. Specifies the context, queue, etc, to be used.

Member Function Documentation

template<ScanConfig C, typename T >

cl::Memory & cl_algo::RBC::Scan< C, T >::get

(

Scan< C, T >::Memory

mem

)

Returns a reference to an internal memory object.

This interface exists to allow CL memory sharing between different kernels.

Parameters

[in]

mem

enumeration value specifying the requested memory object.

Returns

A reference to the requested memory object.

template<ScanConfig C, typename T >

void cl_algo::RBC::Scan< C, T >::init	(	unsigned int	_cols,
		unsigned int	_rows,
		Staging	_staging = Staging::IO
	)

Configures kernel execution parameters.

Sets up memory objects as necessary, and defines the kernel workspaces.

Note

If you have assigned a memory object to one member variable of the class before the call to init, then that memory will be maintained. Otherwise, a new memory object will be created.

Working with float elements and having large summations can be problematic. It is advised that a scaling is applied on the elements for better accuracy.

Parameters

[in]	_cols	number of columns in the input array.
[in]	_rows	number of rows in the input array.
[in]	_staging	flag to indicate whether or not to instantiate the staging buffers.

template<ScanConfig C, typename T >

void * cl_algo::RBC::Scan< C, T >::read	(	Scan< C, T >::Memory	mem = Scan< C, T >::Memory::H_OUT,
		bool	block = CL_TRUE,
		const std::vector< cl::Event > *	events = nullptr,
		cl::Event *	event = nullptr
	)

Performs a data transfer to a staging buffer.

The transfer happens from a device buffer to the associated (specified) staging buffer on the host.

Parameters

[in]	mem	enumeration value specifying an output staging buffer.
[in]	block	a flag to indicate whether to perform a blocking or a non-blocking operation.
[in]	events	a wait-list of events.
[out]	event	event associated with the read operation to the staging buffer.

template<ScanConfig C, typename T >

void cl_algo::RBC::Scan< C, T >::run	(	const std::vector< cl::Event > *	events = nullptr,
		cl::Event *	event = nullptr
	)

Executes the necessary kernels.

The function call is non-blocking.

Parameters

[in]	events	a wait-list of events.
[out]	event	event associated with the last kernel execution.

template<ScanConfig C, typename T = cl_int>

template<typename period >

double cl_algo::RBC::Scan< C, T >::run ( clutils::GPUTimer< period > &  timer, const std::vector< cl::Event > *  events = nullptr  )

inline

Executes the necessary kernels.

This run instance is used for profiling.

Parameters

[in]	timer	GPUTimer that does the profiling of the kernel executions.
[in]	events	a wait-list of events.

Returns

Τhe total execution time measured by the timer.

template<ScanConfig C, typename T >

void cl_algo::RBC::Scan< C, T >::write	(	Scan< C, T >::Memory	mem = Scan< C, T >::Memory::D_IN,
		void *	ptr = nullptr,
		bool	block = CL_FALSE,
		const std::vector< cl::Event > *	events = nullptr,
		cl::Event *	event = nullptr
	)

Performs a data transfer to a device buffer.

The transfer happens from a staging buffer on the host to the associated (specified) device buffer.

Parameters

[in]	mem	enumeration value specifying an input device buffer.
[in]	ptr	a pointer to an array holding input data. If not NULL, the data from ptr will be copied to the associated staging buffer.
[in]	block	a flag to indicate whether to perform a blocking or a non-blocking operation.
[in]	events	a wait-list of events.
[out]	event	event associated with the write operation to the device buffer.

Member Data Documentation

template<ScanConfig C, typename T = cl_int>

T* cl_algo::RBC::Scan< C, T >::hPtrIn

Mapping of the input staging buffer.

template<ScanConfig C, typename T = cl_int>

T* cl_algo::RBC::Scan< C, T >::hPtrOut

Mapping of the output staging buffer.

---

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

• algorithms.hpp
• algorithms.cpp