CUTLASS
CUDA Templates for Linear Algebra Subroutines and Solvers
Public Member Functions | Public Attributes | List of all members
cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments Struct Reference

Argument structure.

#include <gemm_batched.h>

Collaboration diagram for cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments:
Collaboration graph
[legend]

Public Member Functions

CUTLASS_HOST_DEVICE Arguments ()
 Default ctor. More...
 
CUTLASS_HOST_DEVICE Arguments (GemmCoord problem_size_, TensorRef< ElementA const, LayoutA > ref_A_, int64_t stride_A_, TensorRef< ElementB const, LayoutB > ref_B_, int64_t stride_B_, TensorRef< ElementC const, LayoutC > ref_C_, int64_t stride_C_, TensorRef< ElementC, LayoutC > ref_D_, int64_t stride_D_, typename EpilogueOutputOp::Params epilogue_, int batch_count_)
 Constructs an Arguments structure. More...
 

Public Attributes

GemmCoord problem_size
 
TensorRef< ElementA const, LayoutA > ref_A
 
int64_t stride_A
 
TensorRef< ElementB const, LayoutB > ref_B
 
int64_t stride_B
 
TensorRef< ElementC const, LayoutC > ref_C
 
int64_t stride_C
 
TensorRef< ElementC, LayoutC > ref_D
 
int64_t stride_D
 
EpilogueOutputOp::Params epilogue
 
int batch_count
 

Constructor & Destructor Documentation

template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
CUTLASS_HOST_DEVICE cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::Arguments ( )
inline
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
CUTLASS_HOST_DEVICE cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::Arguments ( GemmCoord  problem_size_,
TensorRef< ElementA const, LayoutA >  ref_A_,
int64_t  stride_A_,
TensorRef< ElementB const, LayoutB >  ref_B_,
int64_t  stride_B_,
TensorRef< ElementC const, LayoutC >  ref_C_,
int64_t  stride_C_,
TensorRef< ElementC, LayoutC >  ref_D_,
int64_t  stride_D_,
typename EpilogueOutputOp::Params  epilogue_,
int  batch_count_ 
)
inline

Member Data Documentation

template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
int cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::batch_count
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
EpilogueOutputOp::Params cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::epilogue
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
GemmCoord cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::problem_size
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
TensorRef<ElementA const, LayoutA> cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::ref_A
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
TensorRef<ElementB const, LayoutB> cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::ref_B
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
TensorRef<ElementC const, LayoutC> cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::ref_C
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
TensorRef<ElementC, LayoutC> cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::ref_D
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
int64_t cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::stride_A
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
int64_t cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::stride_B
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
int64_t cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::stride_C
template<typename ElementA_, typename LayoutA_, typename ElementB_, typename LayoutB_, typename ElementC_, typename LayoutC_, typename ElementAccumulator_ = ElementC_, typename OperatorClass_ = arch::OpClassSimt, typename ArchTag_ = arch::Sm70, typename ThreadblockShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::ThreadblockShape, typename WarpShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::WarpShape, typename InstructionShape_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::InstructionShape, typename EpilogueOutputOp_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::EpilogueOutputOp, typename ThreadblockSwizzle_ = threadblock::GemmBatchedIdentityThreadblockSwizzle, int Stages = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kStages, int AlignmentA = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentA, int AlignmentB = DefaultGemmConfiguration<OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::kAlignmentB, typename Operator_ = typename DefaultGemmConfiguration< OperatorClass_, ArchTag_, ElementA_, ElementB_, ElementC_, ElementAccumulator_>::Operator>
int64_t cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::stride_D
The documentation for this struct was generated from the following file:
Generated by   doxygen 1.8.11