CUTLASS
CUDA Templates for Linear Algebra Subroutines and Solvers
Classes | Namespaces
default_epilogue_complex_tensor_op.h File Reference

Epilogue for threadblock scoped complex GEMMs using Tensor Ops. More...

#include "cutlass/cutlass.h"
#include "cutlass/numeric_types.h"
#include "cutlass/array.h"
#include "cutlass/gemm/gemm.h"
#include "cutlass/epilogue/thread/linear_combination.h"
#include "cutlass/epilogue/thread/conversion_op.h"
#include "cutlass/epilogue/thread/reduction_op.h"
#include "cutlass/transform/threadblock/regular_tile_iterator_pitch_linear.h"
#include "cutlass/epilogue/warp/fragment_iterator_complex_tensor_op.h"
#include "cutlass/epilogue/warp/tile_iterator_tensor_op.h"
#include "cutlass/epilogue/threadblock/default_thread_map_tensor_op.h"
#include "cutlass/epilogue/threadblock/predicated_tile_iterator.h"
#include "cutlass/epilogue/threadblock/shared_load_iterator.h"
#include "cutlass/epilogue/threadblock/epilogue.h"
Include dependency graph for default_epilogue_complex_tensor_op.h:

Go to the source code of this file.

Classes

struct  cutlass::epilogue::threadblock::DefaultEpilogueComplexTensorOp< Shape_, WarpMmaTensorOp_, PartitionsK, OutputOp_, ElementsPerAccess >
 Defines sensible defaults for epilogues for TensorOps. More...
 

Namespaces

 cutlass
 
 cutlass::epilogue
 
 cutlass::epilogue::threadblock
 

Detailed Description

The epilogue rearranges the result of a matrix product through shared memory to match canonical tensor layouts in global memory. Epilogues support conversion and reduction operations.