Selector for the "sum" reduction implementations. More...

#include "etl/impl/std/sum.hpp"
#include "etl/impl/vec/sum.hpp"
#include "etl/impl/blas/sum.hpp"
#include "etl/impl/cublas/sum.hpp"

Include dependency graph for sum.hpp:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes
struct	etl::detail::sum_impl
	Sum operation implementation. More...

struct	etl::detail::asum_impl
	Absolute Sum operation implementation. More...

Functions
template<typename E >
constexpr etl::sum_impl	etl::detail::select_default_sum_impl (bool no_gpu)
	Select the sum implementation for an expression of type E. More...

template<typename E >
constexpr etl::sum_impl	etl::detail::select_sum_impl ()
	Select the sum implementation for an expression of type E. More...

Detailed Description

Selector for the "sum" reduction implementations.

The functions are responsible for selecting the most efficient implementation for each case, based on what is available. The selection of parallel versus serial is also done at this level. The implementation functions should never be used directly, only functions of this header can be used directly.

Note: In a perfect world (full constexpr function and variable templates), the selection should be done with a template parameter in a variable template full sspecialization (alias for each real functions).

Function Documentation

◆ select_default_sum_impl()

template<typename E >

constexpr etl::sum_impl etl::detail::select_default_sum_impl ( bool no_gpu )

Select the sum implementation for an expression of type E.

This does not consider the local context

Template Parameters

E	The type of expression

Returns: The implementation to use

◆ select_sum_impl()