cppuddle::kernel_aggregation::detail::aggregated_executor< Executor > Class Template Reference

Executor Class that aggregates function calls for specific kernels. More...

#include <aggregation_executors_and_allocators.hpp>

Classes
class	executor_slice
	Slice class - meant as a scope interface to the aggregated executor. More...

Public Types
using	Executor_Slice = executor_slice
using	buffer_entry_t = std::tuple< void *, const size_t, std::atomic< size_t >, bool, const size_t, size_t >
	Data entry for a buffer allocation: void* pointer, size_t for buffer-size, atomic for the slice counter, location_id, gpu_id. More...

Public Member Functions
template<typename T , typename Host_Allocator >
T *	get (const size_t size, const size_t slice_alloc_counter)
	Get new buffer OR get buffer already allocated by different slice. More...
template<typename T , typename Host_Allocator >
void	mark_unused (T *p, const size_t size)
	Notify buffer list that one slice is done with the buffer. More...
bool	sync_aggregation_slices (const size_t slice_launch_counter)
	Only meant to be accessed by the slice executors. More...
template<typename F , typename... Ts>
void	post (const size_t slice_launch_counter, F &&f, Ts &&...ts)
	Only meant to be accessed by the slice executors. More...
template<typename F , typename... Ts>
hpx::lcos::future< void >	async (const size_t slice_launch_counter, F &&f, Ts &&...ts)
	Only meant to be accessed by the slice executors. More...
template<typename F , typename... Ts>
hpx::lcos::shared_future< void >	wrap_async (const size_t slice_launch_counter, F &&f, Ts &&...ts)
	Only meant to be accessed by the slice executors. More...
bool	slice_available (void)
std::optional< hpx::lcos::future< executor_slice > >	request_executor_slice ()
void	reduce_usage_counter (void)
	~aggregated_executor (void)
	aggregated_executor (const size_t number_slices, aggregated_executor_modes mode, const size_t gpu_id=0)
	aggregated_executor (const aggregated_executor &other)=delete
aggregated_executor &	operator= (const aggregated_executor &other)=delete
	aggregated_executor (aggregated_executor &&other)=delete
aggregated_executor &	operator= (aggregated_executor &&other)=delete

Public Attributes
size_t	gpu_id
hpx::lcos::local::promise< void >	slices_full_promise
std::vector< hpx::lcos::local::promise< executor_slice > >	executor_slices
	Promises with the slice executors – to be set when the starting criteria is met. More...
std::deque< aggregated_function_call< Executor > >	function_calls
	List of aggregated function calls - function will be launched when all slices have called it. More...
aggregation_mutex_t	mut
	For synchronizing the access to the function calls list. More...
std::deque< buffer_entry_t >	buffer_allocations
	Keeps track of the aggregated buffer allocations done in all the slices. More...
std::unordered_map< void *, size_t >	buffer_allocations_map
	Map pointer to deque index for fast access in the deallocations. More...
aggregation_mutex_t	buffer_mut
	For synchronizing the access to the buffer_allocations. More...
std::atomic< size_t >	buffer_counter = 0
hpx::lcos::future< void >	current_continuation
hpx::lcos::future< void >	last_stream_launch_done
std::atomic< size_t >	overall_launch_counter = 0
size_t	launched_slices

Detailed Description

template<typename Executor>
class cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >

Executor Class that aggregates function calls for specific kernels.

Executor is not meant to be used directly. Instead it yields multiple executor_slice objects. These serve as interfaces. Slices from the same aggregated_executor are meant to execute the same function calls but on different data (i.e. different tasks)

Member Typedef Documentation

buffer_entry_t

template<typename Executor >

using cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::buffer_entry_t = std::tuple<void*, const size_t, std::atomic<size_t>, bool, const size_t, size_t>

Data entry for a buffer allocation: void* pointer, size_t for buffer-size, atomic for the slice counter, location_id, gpu_id.

Executor_Slice

template<typename Executor >

using cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::Executor_Slice = executor_slice

Constructor & Destructor Documentation

~aggregated_executor()

template<typename Executor >

cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::~aggregated_executor ( void  )

inline

aggregated_executor() [1/3]

template<typename Executor >

cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::aggregated_executor ( const size_t  number_slices, aggregated_executor_modes  mode, const size_t  gpu_id = 0  )

inline

aggregated_executor() [2/3]

template<typename Executor >

cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::aggregated_executor ( const aggregated_executor< Executor > &  other )

delete

aggregated_executor() [3/3]

template<typename Executor >

cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::aggregated_executor ( aggregated_executor< Executor > &&  other )

delete

Member Function Documentation

async()

template<typename Executor >

template<typename F , typename... Ts>

hpx::lcos::future<void> cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::async ( const size_t  slice_launch_counter, F &&  f, Ts &&...  ts  )

inline

Only meant to be accessed by the slice executors.

get()

template<typename Executor >

template<typename T , typename Host_Allocator >

T* cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::get ( const size_t  size, const size_t  slice_alloc_counter  )

inline

Get new buffer OR get buffer already allocated by different slice.

mark_unused()

template<typename Executor >

template<typename T , typename Host_Allocator >

void cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::mark_unused ( T *  p, const size_t  size  )

inline

Notify buffer list that one slice is done with the buffer.

operator=() [1/2]

template<typename Executor >

aggregated_executor& cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::operator= ( aggregated_executor< Executor > &&  other )

delete

operator=() [2/2]

template<typename Executor >

aggregated_executor& cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::operator= ( const aggregated_executor< Executor > &  other )

delete

post()

template<typename Executor >

template<typename F , typename... Ts>

void cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::post ( const size_t  slice_launch_counter, F &&  f, Ts &&...  ts  )

inline

Only meant to be accessed by the slice executors.

reduce_usage_counter()

template<typename Executor >

void cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::reduce_usage_counter ( void  )

inline

request_executor_slice()

template<typename Executor >

std::optional<hpx::lcos::future<executor_slice> > cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::request_executor_slice ( void  )

inline

slice_available()

template<typename Executor >

bool cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::slice_available ( void  )

inline

sync_aggregation_slices()

template<typename Executor >

bool cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::sync_aggregation_slices ( const size_t  slice_launch_counter )

inline

Only meant to be accessed by the slice executors.

wrap_async()

template<typename Executor >

template<typename F , typename... Ts>

hpx::lcos::shared_future<void> cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::wrap_async ( const size_t  slice_launch_counter, F &&  f, Ts &&...  ts  )

inline

Only meant to be accessed by the slice executors.

Member Data Documentation

buffer_allocations

template<typename Executor >

std::deque<buffer_entry_t> cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::buffer_allocations

Keeps track of the aggregated buffer allocations done in all the slices.

buffer_allocations_map

template<typename Executor >

std::unordered_map<void*,size_t> cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::buffer_allocations_map

Map pointer to deque index for fast access in the deallocations.

buffer_counter

template<typename Executor >

std::atomic<size_t> cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::buffer_counter = 0

buffer_mut

template<typename Executor >

aggregation_mutex_t cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::buffer_mut

For synchronizing the access to the buffer_allocations.

current_continuation

template<typename Executor >

hpx::lcos::future<void> cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::current_continuation

executor_slices

template<typename Executor >

std::vector<hpx::lcos::local::promise<executor_slice> > cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::executor_slices

Promises with the slice executors – to be set when the starting criteria is met.

function_calls

template<typename Executor >

std::deque<aggregated_function_call<Executor> > cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::function_calls

List of aggregated function calls - function will be launched when all slices have called it.

gpu_id

template<typename Executor >

size_t cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::gpu_id

last_stream_launch_done

template<typename Executor >

hpx::lcos::future<void> cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::last_stream_launch_done

launched_slices

template<typename Executor >

size_t cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::launched_slices

mut

template<typename Executor >

aggregation_mutex_t cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::mut

For synchronizing the access to the function calls list.

overall_launch_counter

template<typename Executor >

std::atomic<size_t> cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::overall_launch_counter = 0

slices_full_promise

template<typename Executor >

hpx::lcos::local::promise<void> cppuddle::kernel_aggregation::detail::aggregated_executor< Executor >::slices_full_promise

---

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

• include/cppuddle/kernel_aggregation/detail/aggregation_executors_and_allocators.hpp