olb::gpu::cuda::maximum_and_plus< T > Struct Template Reference

Function object for simulateneously computing maximum and sum in a single thrust::reduce. More...

#include <statistics.hh>

Collaboration diagram for olb::gpu::cuda::maximum_and_plus< T >:

Public Types
using	first_argument_type = pair<T,T>
using	second_argument_type = pair<T,T>
using	result_type = pair<T,T>

Public Member Functions
__device__ result_type	operator() (const first_argument_type &lhs, const second_argument_type &rhs) const

Detailed Description

template<typename T>
struct olb::gpu::cuda::maximum_and_plus< T >

Function object for simulateneously computing maximum and sum in a single thrust::reduce.

Definition at line 63 of file statistics.hh.

Member Typedef Documentation

first_argument_type

template<typename T >

using olb::gpu::cuda::maximum_and_plus< T >::first_argument_type = pair<T,T>

Definition at line 64 of file statistics.hh.

result_type

template<typename T >

using olb::gpu::cuda::maximum_and_plus< T >::result_type = pair<T,T>

Definition at line 66 of file statistics.hh.

second_argument_type

template<typename T >

using olb::gpu::cuda::maximum_and_plus< T >::second_argument_type = pair<T,T>

Definition at line 65 of file statistics.hh.

Member Function Documentation

operator()()

template<typename T >

__device__ result_type olb::gpu::cuda::maximum_and_plus< T >::operator() ( const first_argument_type & lhs, const second_argument_type & rhs ) const

inline

Definition at line 68 of file statistics.hh.

                                                                           {
    return {lhs.first < rhs.first ? rhs.first : lhs.first,
            lhs.second + rhs.second};
  }

References olb::gpu::cuda::pair< T, U >::first, and olb::gpu::cuda::pair< T, U >::second.

---

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

• src/core/platform/gpu/cuda/statistics.hh