CUTLASS
CUDA Templates for Linear Algebra Subroutines and Solvers
|
Defines basic thread level reduction with specializations for Array<T, N>. More...
#include "cutlass/cutlass.h"
#include "cutlass/numeric_types.h"
#include "cutlass/array.h"
#include "cutlass/half.h"
#include "cutlass/functional.h"
Go to the source code of this file.
Classes | |
struct | cutlass::reduction::thread::Reduce< Op, T > |
Structure to compute the thread level reduction. More... | |
struct | cutlass::reduction::thread::Reduce< plus< T >, T > |
Partial Specialization of Reduce for "plus" (a functional operator) More... | |
struct | cutlass::reduction::thread::Reduce< plus< T >, Array< T, N > > |
Partial specialization of Reduce for Array<T, N> More... | |
struct | cutlass::reduction::thread::Reduce< plus< half_t >, Array< half_t, N > > |
Partial specializations of Reduce for Array<half_t, N> More... | |
struct | cutlass::reduction::thread::Reduce< plus< half_t >, AlignedArray< half_t, N > > |
Partial specializations of Reduce for AlignedArray<half_t, N> More... | |
Namespaces | |
cutlass | |
cutlass::reduction | |
cutlass::reduction::thread | |