|  | CUTLASS
    CUDA Templates for Linear Algebra Subroutines and Solvers | 
#include <tensor_ref.h>
| Public Types | |
| using | Index = int32_t | 
| Index type used for coordinates.  More... | |
| using | LongIndex = int64_t | 
| Long index type used for offsets.  More... | |
| using | TensorCoord = Coord< kRank, Index > | 
| Logical coordinate.  More... | |
| using | Stride = Coord< kStrideRank, Index > | 
| Stride vector.  More... | |
| Public Member Functions | |
| CUTLASS_HOST_DEVICE | IdentityTensorLayout (Stride const &stride=Stride()) | 
| CUTLASS_HOST_DEVICE LongIndex | operator() (Coord< Rank > const &coord) const | 
| Returns the offset of a coordinate in linear memory.  More... | |
| CUTLASS_HOST_DEVICE Stride | stride () const | 
| Returns the stride of the layout.  More... | |
| CUTLASS_HOST_DEVICE Stride & | stride () | 
| Returns the stride of the layout.  More... | |
| CUTLASS_HOST_DEVICE LongIndex | capacity (TensorCoord const &size) const | 
| Compute the number of contiguous elements needed to store a tensor with the given size.  More... | |
| Static Public Attributes | |
| static int const | kRank = Rank | 
| Logical rank of tensor.  More... | |
| static int const | kStrideRank = Rank | 
| Rank of stride vector.  More... | |
Default layout function from coordinates in a tensor's index space into the n-D array held in memory.
All layout functions must define at least the members shown in IdentityTensorLayout<>.
| using cutlass::IdentityTensorLayout< Rank >::Index = int32_t | 
| using cutlass::IdentityTensorLayout< Rank >::LongIndex = int64_t | 
| using cutlass::IdentityTensorLayout< Rank >::Stride = Coord<kStrideRank, Index> | 
| using cutlass::IdentityTensorLayout< Rank >::TensorCoord = Coord<kRank, Index> | 
| 
 | inline | 
| 
 | inline | 
| 
 | inline | 
| 
 | inline | 
| 
 | inline | 
| 
 | static | 
| 
 | static | 
 1.8.11
 1.8.11