Contains all relevant information about a parallel for loop. Contains a collection of Thr objects, the loop's start and end iterations, the chunk size, the number of threads, and the number of threads that have completed their work. More...

Collaboration diagram for DotMP.WorkShare:

[legend]

Package Functions
	WorkShare (uint num_threads, Thread[] threads, int start, int end, uint chunk_size, Operations? op, IScheduler schedule)
	The constructor for a WorkShare object. More...

	WorkShare ()
	Default constructor. More...

void	AddReductionValue (dynamic value)
	Add a value to reduction_list. More...

void	SetLocal< T > (ref T local)
	Sets the local variable to the appropriate value based on the operation for parallel for reduction loops. For addition and subtraction, the initial starting value is 0. For multiplication, the initial starting value is 1. For binary And, the initial starting value is the bitwise negation of 0. For binary Or and Xor, the initial starting value is 0. For boolean And, the initial starting value is true. For boolean Or, the initial starting value is false. For min, the initial starting value is int.MaxValue. For max, the initial starting value is int.MinValue. More...

void	PerformLoop< T > (ForAction< T > forAction)
	Performs a parallel for loop according to the scheduling policy provided. More...

Properties
int	working_iter `[get]`
	Get Thr object based on current thread ID. More...

int	start `[get]`
	Getter and setter for the singleton integer WorkShare.start_pv. More...

int	end `[get, private set]`
	The ending iteration of the parallel for loop, exclusive. More...

uint	chunk_size `[get, private set]`
	Getter and setter for singleton uint WorkShare.chunk_size_pv. More...

uint	num_threads `[get, private set]`
	The number of threads to be used in the parallel for loop. More...

Operations?	op `[get, private set]`
	The operation to be performed if doing a reduction. More...

List< dynamic >	reduction_values `[get]`
	Getter for WorkShare.reduction_list. More...

IScheduler	schedule `[get, private set]`
	Getter and setter for singleton object WorkShare.schedule_pv. More...

bool	in_for `[get, set]`
	Getter and setter for this thread's value in WorkShare.in_for_pv. More...

Static Private Attributes
static Thread[]	threads
	The threads to be used in the parallel for loop. More...

static int[]	working_iters
	The working iterations of each thread. More...

static volatile int	start_pv
	The starting iteration of the parallel for loop, inclusive. More...

static uint	chunk_size_pv
	The chunk size to be used with the selected scheduler. More...

static volatile List< dynamic >	reduction_list
	The list of reduction variables from each thread. More...

static IScheduler	schedule_pv
	The schedule to be used in the parallel for loop. More...

static bool[]	in_for_pv
	Booleans per-thread to check if we're currently in a Parallel.For or Parallel.ForReduction<T>. More...

Detailed Description

Contains all relevant information about a parallel for loop. Contains a collection of Thr objects, the loop's start and end iterations, the chunk size, the number of threads, and the number of threads that have completed their work.

Constructor & Destructor Documentation

◆ WorkShare() [1/2]

DotMP.WorkShare.WorkShare	(	uint	num_threads,
		Thread[]	threads,
		int	start,
		int	end,
		uint	chunk_size,
		Operations?	op,
		IScheduler	schedule
	)

inlinepackage

The constructor for a WorkShare object.

Parameters

num_threads	The number of threads in the WorkShare.
threads	The Thread[] array of threads.
start	Starting iteration, inclusive.
end	Ending iteration, exclusive.
chunk_size	The chunk size to use.
op	The operation for reduction, null if not a reduction.
schedule	The Parallel.Schedule to use.

◆ WorkShare() [2/2]

DotMP.WorkShare.WorkShare ( )

inlinepackage

Default constructor.

Member Function Documentation

◆ AddReductionValue()

void DotMP.WorkShare.AddReductionValue ( dynamic value )

inlinepackage

Add a value to reduction_list.

Parameters

value The value to add to reduction_list.

◆ PerformLoop< T >()

void DotMP.WorkShare.PerformLoop< T > ( ForAction< T > forAction )

inlinepackage

Performs a parallel for loop according to the scheduling policy provided.

Template Parameters

T	The type of reductions, if applicable.

Parameters

forAction The function to be executed.

Exceptions

InternalSchedulerException Thrown if the internal schedulers throw an exception.

◆ SetLocal< T >()

void DotMP.WorkShare.SetLocal< T > ( ref T local )

inlinepackage

Sets the local variable to the appropriate value based on the operation for parallel for reduction loops. For addition and subtraction, the initial starting value is 0. For multiplication, the initial starting value is 1. For binary And, the initial starting value is the bitwise negation of 0. For binary Or and Xor, the initial starting value is 0. For boolean And, the initial starting value is true. For boolean Or, the initial starting value is false. For min, the initial starting value is int.MaxValue. For max, the initial starting value is int.MinValue.

Template Parameters

T	The type of the local variable.

Parameters

local The local variable to be set.

Member Data Documentation

◆ chunk_size_pv

uint DotMP.WorkShare.chunk_size_pv

staticprivate

The chunk size to be used with the selected scheduler.

◆ in_for_pv

bool [] DotMP.WorkShare.in_for_pv

staticprivate

Booleans per-thread to check if we're currently in a Parallel.For or Parallel.ForReduction<T>.

◆ reduction_list

volatile List<dynamic> DotMP.WorkShare.reduction_list

staticprivate

The list of reduction variables from each thread.

◆ schedule_pv

IScheduler DotMP.WorkShare.schedule_pv

staticprivate

The schedule to be used in the parallel for loop.

◆ start_pv

volatile int DotMP.WorkShare.start_pv

staticprivate

The starting iteration of the parallel for loop, inclusive.

◆ threads

Thread [] DotMP.WorkShare.threads

staticprivate

The threads to be used in the parallel for loop.

◆ working_iters

int [] DotMP.WorkShare.working_iters

staticprivate

The working iterations of each thread.

Property Documentation

◆ chunk_size

uint DotMP.WorkShare.chunk_size

getprivate setpackage

Getter and setter for singleton uint WorkShare.chunk_size_pv.

◆ end

int DotMP.WorkShare.end

getprivate setpackage

The ending iteration of the parallel for loop, exclusive.

◆ in_for

bool DotMP.WorkShare.in_for

getsetpackage

Getter and setter for this thread's value in WorkShare.in_for_pv.

◆ num_threads

uint DotMP.WorkShare.num_threads

getprivate setpackage

The number of threads to be used in the parallel for loop.

◆ op

Operations? DotMP.WorkShare.op

getprivate setpackage

The operation to be performed if doing a reduction.

◆ reduction_values

List<dynamic> DotMP.WorkShare.reduction_values

getpackage

Getter for WorkShare.reduction_list.

◆ schedule

IScheduler DotMP.WorkShare.schedule

getprivate setpackage

Getter and setter for singleton object WorkShare.schedule_pv.

◆ start

int DotMP.WorkShare.start

getpackage

Getter and setter for the singleton integer WorkShare.start_pv.

◆ working_iter

int DotMP.WorkShare.working_iter

getpackage

Get Thr object based on current thread ID.

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

DotMP/WorkShare.cs

Package Functions

Properties

Static Private Attributes

Detailed Description

Constructor & Destructor Documentation

◆ WorkShare() [1/2]

◆ WorkShare() [2/2]

Member Function Documentation

◆ AddReductionValue()

◆ PerformLoop< T >()

◆ SetLocal< T >()

Member Data Documentation

◆ chunk_size_pv

◆ in_for_pv

◆ reduction_list

◆ schedule_pv

◆ start_pv

◆ threads

◆ working_iters

Property Documentation

◆ chunk_size

◆ end

◆ in_for

◆ num_threads

◆ op

◆ reduction_values

◆ schedule

◆ start

◆ working_iter