STAPL API Reference          
Overview   Containers   Algorithms   Views   Skeletons   Run-Time System
Modules     Classes    
Functions
+ Collaboration diagram for Parametric Dependencies of:

Functions

template<typename T , typename Tag >
skeletons_impl::alltoall_pd< T, Tag > stapl::skeletons::alltoall_pd ()
 Creates the simplest alltoall parametric dependency skeleton. More...
 
template<bool pos_aware, bool is_reversed = false, typename Op , typename Filter = skeletons::no_filter>
skeletons_impl::butterfly_pd< Op, Filter, pos_aware, is_reversed > stapl::skeletons::butterfly_pd (Op const &op, Filter const &filter=Filter())
 Creates a butterfly parametric dependency given a op and the direction of the butterfly.
 
template<typename Op >
skeletons_impl::pointer_jumping_pd< Op > stapl::skeletons::pointer_jumping_pd (Op const &op)
 Creates a pointer-jumping parametric dependency given a op. More...
 
template<typename Op >
skeletons_impl::stencil_2d_pd< Op > stapl::skeletons::stencil_2d_pd (Op const &op)
 Creates a 9-point stencil parametric dependency. More...
 

Detailed Description

Data Exchange Skeletons

Function Documentation

◆ alltoall_pd()

template<typename T , typename Tag >
skeletons_impl::alltoall_pd<T, Tag> stapl::skeletons::alltoall_pd ( )

Creates the simplest alltoall parametric dependency skeleton.

◆ pointer_jumping_pd()

template<typename Op >
skeletons_impl::pointer_jumping_pd<Op> stapl::skeletons::pointer_jumping_pd ( Op const &  op)

Creates a pointer-jumping parametric dependency given a op.

A pointer-jumping parametric dependency can be used in various skeletons. One good example is hillis_steele_scan in which scan is done in a non-work optimal but smaller tree.

◆ stencil_2d_pd()

template<typename Op >
skeletons_impl::stencil_2d_pd<Op> stapl::skeletons::stencil_2d_pd ( Op const &  op)

Creates a 9-point stencil parametric dependency.

In a 2D stencil, parametric dependencies are defined in such a way that each node depends on every 9 node in its surroundings including itself from the previous timestep.