kamping::ops Namespace Reference

this namespace contains all builtin operations supported by MPI. More...

Classes
struct	null
	builtin null operation (aka MPI_OP_NULL) More...

Typedefs
template<typename T = void>
using	max = kamping::internal::max_impl<T>
	builtin maximum operation (aka MPI_MAX)
template<typename T = void>
using	min = kamping::internal::min_impl<T>
	builtin minimum operation (aka MPI_MIN)
template<typename T = void>
using	plus = std::plus<T>
	builtin summation operation (aka MPI_SUM)
template<typename T = void>
using	multiplies = std::multiplies<T>
	builtin multiplication operation (aka MPI_PROD)
template<typename T = void>
using	logical_and = std::logical_and<T>
	builtin logical and operation (aka MPI_LAND)
template<typename T = void>
using	bit_and = std::bit_and<T>
	builtin bitwise and operation (aka MPI_BAND)
template<typename T = void>
using	logical_or = std::logical_or<T>
	builtin logical or operation (aka MPI_LOR)
template<typename T = void>
using	bit_or = std::bit_or<T>
	builtin bitwise or operation (aka MPI_BOR)
template<typename T = void>
using	logical_xor = kamping::internal::logical_xor_impl<T>
	builtin logical xor operation (aka MPI_LXOR)
template<typename T = void>
using	bit_xor = std::bit_xor<T>
	builtin bitwise xor operation (aka MPI_BXOR)

Variables
constexpr internal::commutative_tag	commutative {}
	global tag for commutativity
constexpr internal::non_commutative_tag	non_commutative {}
	global tag for non-commutativity

Detailed Description

this namespace contains all builtin operations supported by MPI.

You can either use them by passing their STL counterparts like std::plus<>, std::multiplies<> etc. or using the aliases kamping::ops::plus, kamping::ops::multiplies, kamping::ops::max(), ... You can either use them without a template parameter (std::plus<>) or explicitly specify the type (std::plus<int>). In the latter case, the type must match the datatype of the buffer the operation shall be applied to.