Partial specialization for column-major output.  
#include <gemm_splitk_parallel.h>
|  | 
| using | ElementA = ElementA_ | 
|  | 
| using | LayoutA = LayoutA_ | 
|  | 
| using | ElementB = ElementB_ | 
|  | 
| using | LayoutB = LayoutB_ | 
|  | 
| using | ElementC = ElementC_ | 
|  | 
| using | LayoutC = layout::ColumnMajor | 
|  | 
| using | ElementAccumulator = ElementAccumulator_ | 
|  | 
| using | OperatorClass = OperatorClass_ | 
|  | 
| using | ArchTag = ArchTag_ | 
|  | 
| using | ThreadblockShape = ThreadblockShape_ | 
|  | 
| using | WarpShape = WarpShape_ | 
|  | 
| using | InstructionShape = InstructionShape_ | 
|  | 
| using | ConvertScaledOp = ConvertScaledOp_ | 
|  | 
| using | EpilogueOutputOp = EpilogueOutputOp_ | 
|  | 
| using | ReductionOp = ReductionOp_ | 
|  | 
| using | ThreadblockSwizzle = ThreadblockSwizzle_ | 
|  | 
| using | Operator = Operator_ | 
|  | 
| using | UnderlyingOperator = GemmSplitKParallel< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ConvertScaledOp, ReductionOp, ThreadblockSwizzle, Stages, kAlignmentA, kAlignmentB, Operator > | 
|  | 
| using | UnderlyingArguments = typename UnderlyingOperator::Arguments | 
|  | 
| using | GemmKernel = typename UnderlyingOperator::GemmKernel | 
|  | 
| using | ReductionKernel = typename UnderlyingOperator::ReductionKernel | 
|  | 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::ArchTag =  ArchTag_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::ConvertScaledOp =  ConvertScaledOp_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::ElementA =  ElementA_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::ElementAccumulator =  ElementAccumulator_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::ElementB =  ElementB_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::ElementC =  ElementC_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::EpilogueOutputOp =  EpilogueOutputOp_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::GemmKernel =  typename UnderlyingOperator::GemmKernel | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::InstructionShape =  InstructionShape_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::LayoutA =  LayoutA_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::LayoutB =  LayoutB_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::LayoutC =  layout::ColumnMajor | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::Operator =  Operator_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::OperatorClass =  OperatorClass_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::ReductionKernel =  typename UnderlyingOperator::ReductionKernel | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::ReductionOp =  ReductionOp_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::ThreadblockShape =  ThreadblockShape_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::ThreadblockSwizzle =  ThreadblockSwizzle_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::UnderlyingArguments =  typename UnderlyingOperator::Arguments | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::UnderlyingOperator =  GemmSplitKParallel< ElementB, typename layout::LayoutTranspose<LayoutB>::type, ElementA, typename layout::LayoutTranspose<LayoutA>::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ConvertScaledOp, ReductionOp, ThreadblockSwizzle, Stages, kAlignmentA, kAlignmentB, Operator > | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
      
        
          | using cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::WarpShape =  WarpShape_ | 
      
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
  
  | 
        
          | cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::GemmSplitKParallel | ( |  | ) |  |  | inline | 
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
  
  | 
        
          | static Status cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::can_implement | ( | Arguments const & | args | ) |  |  | inlinestatic | 
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
  
  | 
        
          | static size_t cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::get_workspace_size | ( | Arguments const & | args | ) |  |  | inlinestatic | 
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
  
  | 
        
          | Status cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::initialize | ( | Arguments const & | args, |  
          |  |  | void * | workspace |  
          |  | ) |  |  |  | inline | 
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
  
  | 
        
          | Status cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::operator() | ( | cudaStream_t | stream = nullptr | ) |  |  | inline | 
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
  
  | 
        
          | Status cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::operator() | ( | Arguments const & | args, |  
          |  |  | void * | workspace = nullptr, |  
          |  |  | cudaStream_t | stream = nullptr |  
          |  | ) |  |  |  | inline | 
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
  
  | 
        
          | Status cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::run | ( | cudaStream_t | stream = nullptr | ) |  |  | inline | 
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
  
  | 
        
          | static UnderlyingArguments cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::to_underlying_arguments | ( | Arguments const & | args | ) |  |  | inlinestatic | 
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
  
  | 
        
          | Status cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::update | ( | Arguments const & | args, |  
          |  |  | void * | workspace = nullptr |  
          |  | ) |  |  |  | inline | 
 
 
template<typename ElementA_ , typename LayoutA_ , typename ElementB_ , typename LayoutB_ , typename ElementC_ , typename ElementAccumulator_ , typename OperatorClass_ , typename ArchTag_ , typename ThreadblockShape_ , typename WarpShape_ , typename InstructionShape_ , typename EpilogueOutputOp_ , typename ConvertScaledOp_ , typename ReductionOp_ , typename ThreadblockSwizzle_ , int Stages, int kAlignmentA, int kAlignmentB, typename Operator_ > 
  
  | 
        
          | int const cutlass::gemm::device::GemmSplitKParallel< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ >::kStages = Stages |  | static | 
 
 
The documentation for this class was generated from the following file: