CUTLASS
CUDA Templates for Linear Algebra Subroutines and Solvers
Public Types | List of all members
cutlass::arch::Mma< gemm::GemmShape< 16, 16, 4 >, 32, half_t, LayoutA, half_t, LayoutB, ElementC, LayoutC, Operator > Struct Template Reference

Matrix multiply-add operation specialized for the entire warp.

#include <mma_sm70.h>

Inheritance diagram for cutlass::arch::Mma< gemm::GemmShape< 16, 16, 4 >, 32, half_t, LayoutA, half_t, LayoutB, ElementC, LayoutC, Operator >:
Inheritance graph
[legend]
Collaboration diagram for cutlass::arch::Mma< gemm::GemmShape< 16, 16, 4 >, 32, half_t, LayoutA, half_t, LayoutB, ElementC, LayoutC, Operator >:
Collaboration graph
[legend]

Public Types

using Shape = gemm::GemmShape< 16, 16, 4 >
 

Member Typedef Documentation

template<typename LayoutA , typename LayoutB , typename ElementC , typename LayoutC , typename Operator >
using cutlass::arch::Mma< gemm::GemmShape< 16, 16, 4 >, 32, half_t, LayoutA, half_t, LayoutB, ElementC, LayoutC, Operator >::Shape = gemm::GemmShape<16, 16, 4>

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