Tools base class. The way that AQUAgpusph compute each problem is set through a set of tools that are computed sequentially. Several tools can be considered, for instance: More...

#include <CalcServer/Tool.h>

Inheritance diagram for Aqua::CalcServer::Tool:

[legend]

Public Member Functions
	Tool (const std::string tool_name, bool once=false)

virtual	~Tool ()

void	name (const std::string tool_name)

const std::string	name ()

virtual void	setup ()

virtual void	execute ()
	Execute the tool measuring the elapsed time.

virtual Tool *	next_tool ()

size_t	allocatedMemory () const

unsigned int	used_times () const

float	elapsedTime (bool averaged=true) const

float	elapsedTimeVariance () const

float	elapsedTimeDeviation () const

virtual const int	scope_modifier ()

Protected Member Functions
int	id_in_pipeline ()

void	next_tool (Tool *tool)

void	allocatedMemory (size_t mem_size)

virtual cl_event	_execute (const std::vector< cl_event > events)

void	addElapsedTime (float elapsed_time)
	Add new data to the average and squared elapsed times.

void	setDependencies (std::vector< std::string > var_names)
	Set the depedencies of the tool.

void	setDependencies (std::vector< InputOutput::Variable * > vars)
	Set the depedencies of the tool.

const std::vector< InputOutput::Variable * >	getDependencies ()
	Get the depedencies of the tool.

Static Protected Member Functions
static std::vector< cl_kernel >	compile (const std::string source, const std::vector< std::string > names, const std::string flags="")
	Compile an OpenCL source code and generate the corresponding kernel.

static cl_kernel	compile_kernel (const std::string source, const std::string kernel_name, const std::string flags="")
	Compile an OpenCL source code and generate the corresponding kernel.

Detailed Description

Tools base class. The way that AQUAgpusph compute each problem is set through a set of tools that are computed sequentially. Several tools can be considered, for instance:

A single OpenCL kernel
1. A more complex OpenCL tool like Reductions or LinkList
2. Python scripts
3. Variables set

Constructor & Destructor Documentation

◆ Tool()

Aqua::CalcServer::Tool::Tool	(	const std::string	tool_name,
		bool	once = `false`
	)

Constructor.

Parameters

tool_name	Name of the tool. Useful to identify errors.
once	Run this tool just once. Useful to make initializations.

◆ ~Tool()

Aqua::CalcServer::Tool::~Tool ( )

virtual

Destructor

Member Function Documentation

◆ _execute()

virtual cl_event Aqua::CalcServer::Tool::_execute ( const std::vector< cl_event > events )

inlineprotectedvirtual

Execute the tool

Parameters

events List of events that shall be waited before safe execution

Returns: OpenCL event to be waited before accessing the dependencies

Reimplemented in Aqua::CalcServer::Assert, Aqua::CalcServer::Conditional, Aqua::CalcServer::If, Aqua::CalcServer::Copy, Aqua::CalcServer::Kernel, Aqua::CalcServer::LinkList, Aqua::CalcServer::MPISync, Aqua::CalcServer::Python, Aqua::CalcServer::RadixSort, Aqua::CalcServer::Reduction, Aqua::CalcServer::Reports::Performance, Aqua::CalcServer::Reports::Screen, Aqua::CalcServer::Reports::SetTabFile, Aqua::CalcServer::Reports::TabFile, Aqua::CalcServer::Set, Aqua::CalcServer::SetScalar, and Aqua::CalcServer::UnSort.

◆ addElapsedTime()

void Aqua::CalcServer::Tool::addElapsedTime ( float elapsed_time )

protected

Add new data to the average and squared elapsed times.

Parameters

elapsed_time Elapsed time

◆ allocatedMemory() [1/2]

size_t Aqua::CalcServer::Tool::allocatedMemory ( ) const

inline

Get the allocated memory for this tool.

Returns: allocated memory by this tool.

◆ allocatedMemory() [2/2]

void Aqua::CalcServer::Tool::allocatedMemory ( size_t mem_size )

inlineprotected

Set the allocated memory for this tool.

Parameters

mem_size allocated memory by this tool.

◆ compile()

std::vector< cl_kernel > Aqua::CalcServer::Tool::compile	(	const std::string	source,
		const std::vector< std::string >	names,
		const std::string	flags = `""`
	)

staticprotected

Compile an OpenCL source code and generate the corresponding kernel.

With this method several operations are carried out at the same time. First the program is compiled and linked. Afterwards, the required kernels are extracted, and the program object is released

Parameters

source	Source code to be compiled
names	Function names to be extracted in the kernel
flags	Additional compilation flags. Some flags are used by default: -DDEBUG/-DNDEBUG depending on whether DEBUG mode is enabled or not -cl-mad-enable -cl-fast-relaxed-math -DHAVE_2D/-DHAVE_3D depending on whether 2D or 3D is considered

Returns: Kernel instances

Here is the call graph for this function:

◆ compile_kernel()

cl_kernel Aqua::CalcServer::Tool::compile_kernel	(	const std::string	source,
		const std::string	kernel_name,
		const std::string	flags = `""`
	)

staticprotected

Compile an OpenCL source code and generate the corresponding kernel.

With this method several operations are carried out at the same time. First the program is compiled and linked. Afterwards, the required kernel is extracted, and the program object is released

Parameters

source	Source code to be compiled
kernel_name	Function name to be extracted in the kernel
flags	Additional compilation flags. Some flags are used by default: -DDEBUG/-DNDEBUG depending on whether DEBUG mode is enabled or not -cl-mad-enable -cl-fast-relaxed-math -DHAVE_2D/-DHAVE_3D depending on whether 2D or 3D is considered

Returns: Kernel instance

Here is the call graph for this function:

◆ elapsedTime()

float Aqua::CalcServer::Tool::elapsedTime ( bool averaged = true ) const

inline

Get the time consumed by the tool.

Parameters

averaged true if the avergaed time step is required, false otherwise.

Returns: time consumed.

◆ elapsedTimeDeviation()

float Aqua::CalcServer::Tool::elapsedTimeDeviation ( ) const

inline

Get the time consumed standard deviation.

Returns: Time consumed standard deviation.

Here is the call graph for this function:

◆ elapsedTimeVariance()

float Aqua::CalcServer::Tool::elapsedTimeVariance ( ) const

inline

Get the time consumed variance.

Returns: Time consumed variance.

◆ execute()

void Aqua::CalcServer::Tool::execute ( )

virtual

Execute the tool measuring the elapsed time.

Actually this method is just ensuring that the tool can be executed, e.g. the tool has been already executed, but it is asked to be ran just once. If the tool can be executed, then _execute() method is called, measuring the time required to carry out the task.

Returns: false if all gone right, true otherwise.

Note: Usually you don't want to overload this method, but the _execute() protected one.

Reimplemented in Aqua::CalcServer::Reports::Performance.

Here is the call graph for this function:

◆ getDependencies()

const std::vector< InputOutput::Variable * > Aqua::CalcServer::Tool::getDependencies ( )

protected

Get the depedencies of the tool.

Returns: Dependencies

◆ id_in_pipeline()

int Aqua::CalcServer::Tool::id_in_pipeline ( )

protected

Get the tool index in the pipeline

Returns: Index of the tool in the pipeline. -1 if the tool cannot be find

Here is the call graph for this function:

◆ name() [1/2]

const std::string Aqua::CalcServer::Tool::name ( )

inline

Get the tool name.

Returns: Tool name.

◆ name() [2/2]

void Aqua::CalcServer::Tool::name ( const std::string tool_name )

inline

Set the tool name.

Parameters

tool_name Tool name.

◆ next_tool() [1/2]

virtual Tool * Aqua::CalcServer::Tool::next_tool ( )

inlinevirtual

Get the next tool to be executed in the pipeline.

Such tool is usually just the next one in the linearized tools chain. However, conditional tools may alter the flow

Returns: Next tool to be executed. NULL if this is the last tool of the pipeline

Reimplemented in Aqua::CalcServer::Conditional, and Aqua::CalcServer::If.

◆ next_tool() [2/2]

void Aqua::CalcServer::Tool::next_tool ( Tool * tool )

inlineprotected

Set the next tool to be executed in the pipeline.

Parameters

tool	Next tool to be executed. NULL if this is the last tool of the pipeline

◆ scope_modifier()

virtual const int Aqua::CalcServer::Tool::scope_modifier ( )

inlinevirtual

Get the scope modifier

Scopes can be used to create groups of tools that can be eventually enabled/disabled in runtime. This is sueful to create conditions.

Returns: 0 if this tool is not modifying the scope, 1 if this tool is creating a new subscope, and -1 if this tool is closing an already created subscope, returning to the previous one

Note: scopes shall be always balanced

Reimplemented in Aqua::CalcServer::Conditional, and Aqua::CalcServer::End.

◆ setDependencies() [1/2]

void Aqua::CalcServer::Tool::setDependencies ( std::vector< InputOutput::Variable * > vars )

protected

Set the depedencies of the tool.

The dependencies are the variables that this tool is either reading or writing.

Parameters

vars	Dependencies

◆ setDependencies() [2/2]

void Aqua::CalcServer::Tool::setDependencies ( std::vector< std::string > var_names )

protected

Set the depedencies of the tool.

The dependencies are the variables that this tool is either reading or writing.

Parameters

var_names Names of the dependencies

Here is the call graph for this function:

◆ setup()

void Aqua::CalcServer::Tool::setup ( )

virtual

◆ used_times()

unsigned int Aqua::CalcServer::Tool::used_times ( ) const

inline

Get the number of times that this tool has been called.

Returns: Number of times this tool has been called.

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

/home/pepe/SPH/Code/aquagpusph/include/CalcServer/Tool.h
/home/pepe/SPH/Code/aquagpusph/src/CalcServer/Tool.cpp

Public Member Functions

Protected Member Functions

Static Protected Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ Tool()

◆ ~Tool()

Member Function Documentation

◆ _execute()

◆ addElapsedTime()

◆ allocatedMemory() [1/2]

◆ allocatedMemory() [2/2]

◆ compile()

◆ compile_kernel()

◆ elapsedTime()

◆ elapsedTimeDeviation()

◆ elapsedTimeVariance()

◆ execute()

◆ getDependencies()

◆ id_in_pipeline()

◆ name() [1/2]

◆ name() [2/2]

◆ next_tool() [1/2]

◆ next_tool() [2/2]

◆ scope_modifier()

◆ setDependencies() [1/2]

◆ setDependencies() [2/2]

◆ setup()

◆ used_times()