CUTLASS: cutlass::gemm::warp::DefaultMmaTensorOp< WarpShape_, InstructionShape_, ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, Operator_, PartitionsK, AccumulatorsInRowMajor, PartitionsN > Struct Template Reference

| | CUTLASS

CUDA Templates for Linear Algebra Subroutines and Solvers |

• Main Page

• Modules

• Namespaces

• Classes

• Files

• Class List

• Class Index

• Class Hierarchy

• Class Members

• cutlass

• gemm

• warp

• DefaultMmaTensorOp

Public Types | List of all members

cutlass::gemm::warp::DefaultMmaTensorOp< WarpShape_, InstructionShape_, ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, Operator_, PartitionsK, AccumulatorsInRowMajor, PartitionsN > Struct Template Reference

Partial specialization for m-by-n-by-kgroup.

#include <default_mma_tensor_op.h>

|

Public Types

| | using | Policy = cutlass::gemm::warp::MmaTensorOpPolicy< cutlass::arch::Mma< InstructionShape_, 32, ElementA, cutlass::layout::RowMajor, ElementB, cutlass::layout::ColumnMajor, ElementC, cutlass::layout::RowMajor, Operator_ >, cutlass::MatrixShape< 1, 1 > > | | | | using | Type = cutlass::gemm::warp::MmaTensorOp< WarpShape_, ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, Policy, PartitionsK, AccumulatorsInRowMajor, PartitionsN > | | |

Member Typedef Documentation

template<typename WarpShape_ , typename InstructionShape_ , typename ElementA , typename LayoutA , typename ElementB , typename LayoutB , typename ElementC , typename LayoutC , typename Operator_ , int PartitionsK, bool AccumulatorsInRowMajor, int PartitionsN>

| using cutlass::gemm::warp::DefaultMmaTensorOp< WarpShape_, InstructionShape_, ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, Operator_, PartitionsK, AccumulatorsInRowMajor, PartitionsN >::Policy = cutlass::gemm::warp::MmaTensorOpPolicy< cutlass::arch::Mma<InstructionShape_, 32, ElementA, cutlass::layout::RowMajor, ElementB, cutlass::layout::ColumnMajor, ElementC, cutlass::layout::RowMajor, Operator_>, cutlass::MatrixShape<1, 1> > |

template<typename WarpShape_ , typename InstructionShape_ , typename ElementA , typename LayoutA , typename ElementB , typename LayoutB , typename ElementC , typename LayoutC , typename Operator_ , int PartitionsK, bool AccumulatorsInRowMajor, int PartitionsN>

| using cutlass::gemm::warp::DefaultMmaTensorOp< WarpShape_, InstructionShape_, ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, Operator_, PartitionsK, AccumulatorsInRowMajor, PartitionsN >::Type = cutlass::gemm::warp::MmaTensorOp< WarpShape_, ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, Policy, PartitionsK, AccumulatorsInRowMajor, PartitionsN> |

---

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

• [default_mma_tensor_op.h](default mma tensor__op_8h_source.html)

---

Generated by 1.8.11