![]() |
CUTLASS
CUDA Templates for Linear Algebra Subroutines and Solvers
|
#include <mma_tensor_op_tile_iterator.h>
Classes | |
| struct | Policy |
| Internal structure of iterator - made public to enable introspection. More... | |
Public Types | |
| using | Shape = Shape_ |
| Shape of tile to load (concept: MatrixShape) More... | |
| using | Element = Element_ |
| Element type. More... | |
| using | Layout = cutlass::layout::ColumnMajor |
| Layout of source tile. More... | |
| using | InstructionShape = InstructionShape_ |
| Shape of one matrix product operation (concept: MatrixShape) More... | |
| using | OpDelta = OpDelta_ |
| Delta between *MMA operations (in units of *MMA operations, concept: MatrixShape) More... | |
| using | TensorRef = TensorRef< Element, Layout > |
| TensorRef type for loading element from a tensor. More... | |
| using | Index = typename TensorRef::Index |
| Index type. More... | |
| using | LongIndex = typename TensorRef::LongIndex |
| Long Index type. More... | |
| using | TensorCoord = typename TensorRef::TensorCoord |
| Coordinate for an element in the tensor. More... | |
| using | Fragment = Array< Element, Shape::kCount/kThreads > |
| Fragment object holding a thread's part of a tile. More... | |
Public Member Functions | |
| CUTLASS_HOST_DEVICE | MmaTensorOpAccumulatorTileIterator () |
| Default ctor constructs null iterator. More... | |
| CUTLASS_HOST_DEVICE | MmaTensorOpAccumulatorTileIterator (TensorRef const &ref, int lane_id) |
| Constructor from TensorRef. More... | |
| CUTLASS_HOST_DEVICE MmaTensorOpAccumulatorTileIterator & | add_pointer_offset (LongIndex offset) |
| Adds a pointer offset to internal pointer(s) to advance through memory. More... | |
| CUTLASS_HOST_DEVICE MmaTensorOpAccumulatorTileIterator & | add_tile_offset (TensorCoord const &tile_offset) |
| Advances an iterator along logical dimensions of matrix in units of whole tiles. More... | |
| CUTLASS_HOST_DEVICE MmaTensorOpAccumulatorTileIterator & | operator++ () |
| Advances the iterator along the advance dimension. More... | |
| CUTLASS_HOST_DEVICE MmaTensorOpAccumulatorTileIterator & | operator-- () |
| Advances the iterator along the advance dimension. More... | |
| CUTLASS_DEVICE MmaTensorOpAccumulatorTileIterator & | operator+= (TensorCoord const &tile_offset) |
| advances in units of whole tiles along the logical coordinate space of the tensor More... | |
| CUTLASS_DEVICE MmaTensorOpAccumulatorTileIterator & | operator-= (TensorCoord const &tile_offset) |
| CUTLASS_HOST_DEVICE void | load (Fragment &frag) const |
| Loads a fragment from memory at the location pointed to by the iterator. More... | |
| CUTLASS_DEVICE void | load_with_pointer_offset (Fragment &frag, Index pointer_offset) const |
| Loads a fragment from memory with additional logical offset. More... | |
| CUTLASS_DEVICE void | load_with_byte_offset (Fragment &frag, Index byte_offset) const |
| Loads a fragment from memory with additional logical offset. More... | |
| CUTLASS_DEVICE void | load (Fragment &frag, TensorCoord const &tile_offset) const |
| Loads a fragment from memory with logical offset in units of whole tiles. More... | |
| CUTLASS_DEVICE void | load (Fragment &frag, TensorCoord const &tile_offset, Index pointer_offset) const |
| Loads a fragment from memory with logical offset in units of whole tiles. More... | |
| CUTLASS_HOST_DEVICE void | store (Fragment const &frag) const |
| Stores a fragment to memory. More... | |
| CUTLASS_DEVICE void | store_with_pointer_offset (Fragment const &frag, Index pointer_offset) const |
| Stores a fragment to memory with additional pointer offset. More... | |
| CUTLASS_DEVICE void | store_with_byte_offset (Fragment const &frag, Index byte_offset) const |
| Stores a fragment to memory with additional pointer offset. More... | |
| CUTLASS_DEVICE void | store (Fragment &frag, TensorCoord const &tile_offset) const |
| Stores a fragment to memory with logical offset in units of whole tiles. More... | |
| CUTLASS_DEVICE void | store (Fragment const &frag, TensorCoord const &tile_offset, Index pointer_offset) const |
| Stores a fragment from memory with logical offset in units of whole tiles. More... | |
Static Public Attributes | |
| static Operand const | kOperand = Operand::kC |
| Operand tag. More... | |
| static int const | kThreads = 32 |
| Number of participating threads. More... | |
This tile iterator is specialized for 32-thread TensorOps. It is used to load or store accumulators from memory and is agnostic to layout. It could be faster if it assumed row-major accumulator layout.
Satisfies: ReadableRandomAccessContiguousTileIteratorConcept | WriteableRandomAccessContiguousTileIteratorConcept
| using cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >::Element = Element_ |
| using cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >::Fragment = Array<Element, Shape::kCount / kThreads> |
| using cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >::Index = typename TensorRef::Index |
| using cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >::InstructionShape = InstructionShape_ |
| using cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >::Layout = cutlass::layout::ColumnMajor |
| using cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >::LongIndex = typename TensorRef::LongIndex |
| using cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >::OpDelta = OpDelta_ |
| using cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >::Shape = Shape_ |
| using cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >::TensorCoord = typename TensorRef::TensorCoord |
| using cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >::TensorRef = TensorRef<Element, Layout> |
|
inline |
|
inline |
|
inline |
|
inline |
|
inline |
|
inline |
< loads a tile with a logical offset in units of whole tiles
| frag | fragment to load from the tensor |
|
inline |
< loads a tile with a logical offset AND a pointer offset
| frag | fragment to load from the tensor |
| tile_offset | loads a tile with a logical offset in units of whole tiles |
|
inline |
< loads a tile with a linear offset
| frag | fragment to load from the tensor |
|
inline |
< loads a tile with a linear offset
| frag | fragment to load from the tensor |
|
inline |
|
inline |
|
inline |
advances in units of whole tiles along the logical coordinate space of the tensor
|
inline |
|
inline |
|
inline |
< stores a tile with a logical offset in units of whole tiles
| frag | fragment to store to the tensor |
|
inline |
| frag | fragment to store to the tensor |
| tile_offset | stores a tile with a logical offset in units of whole tiles |
| pointer_offset | stores a tile with a logical offset AND a pointer offset |
|
inline |
< store a tile with a linear offset
| frag | fragment to store from the tensor |
|
inline |
< store a tile with a linear offset
| frag | fragment to store from the tensor |
|
static |
|
static |
1.8.11