Back to Cutlass

CUTLASS: predicated_tile_iterator.h File Reference

docs/transform_2threadblock_2predicated__tile__iterator_8h.html

4.4.26.1 KB
Original Source

| | CUTLASS

CUDA Templates for Linear Algebra Subroutines and Solvers |

Classes | Namespaces

transform/threadblock/predicated_tile_iterator.h File Reference

Templates implementing loading of tiles from pitch-linear rank=2 tensors. More...

#include "cutlass/arch/memory.h"
#include "cutlass/transform/threadblock/predicated_tile_access_iterator.h"

Include dependency graph for transform/threadblock/predicated_tile_iterator.h:

This graph shows which files directly or indirectly include this file:

[Go to the source code of this file.](transform_2threadblock_2predicated tile iterator_8h_source.html)

|

Classes

| | class | cutlass::transform::threadblock::PredicatedTileIterator< Shape, Element, Layout, AdvanceRank, ThreadMap, AccessSize > | | | | class | [cutlass::transform::threadblock::PredicatedTileIterator< Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessSize >](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 00e7c2c404e7aedfe60ad56bb5571306a1.html) | | | | class | [cutlass::transform::threadblock::PredicatedTileIterator< Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessSize >::Params](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 006a5f2f7a8271031e6cdc5daa5441f2af.html) | | | Parameters object is precomputed state and is host-constructible. [More...](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 006a5f2f7a8271031e6cdc5daa5441f2af.html#details)
| | | | class | [cutlass::transform::threadblock::PredicatedTileIterator< Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessSize >](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 0068b3e874b5d93d11f0fa902c7f1d11d9.html) | | | | class | [cutlass::transform::threadblock::PredicatedTileIterator< Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessSize >::Params](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 00a6b756b1bcfbb35fe4a3e68ff074e380.html) | | | Parameters object is precomputed state and is host-constructible. [More...](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 00a6b756b1bcfbb35fe4a3e68ff074e380.html#details)
| | | | class | [cutlass::transform::threadblock::PredicatedTileIterator< Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessSize >](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 0041ea81994f8af0d4d071fdb9e66b5ff0.html) | | | | class | [cutlass::transform::threadblock::PredicatedTileIterator< Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessSize >::Params](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 004d0f9b5e19c29acc17bcdc360dafebbd.html) | | | Parameters object is precomputed state and is host-constructible. [More...](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 004d0f9b5e19c29acc17bcdc360dafebbd.html#details)
| | | | class | [cutlass::transform::threadblock::PredicatedTileIterator< Shape_, Element_, layout::ColumnMajorInterleaved< InterleavedK >, AdvanceRank, ThreadMap_, AccessSize >](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 00f6b3a9dfab5e7c72d5233f7e5e6e3b9b.html) | | | | class | [cutlass::transform::threadblock::PredicatedTileIterator< Shape_, Element_, layout::ColumnMajorInterleaved< InterleavedK >, AdvanceRank, ThreadMap_, AccessSize >::Params](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 00ebd1a63351e1085d0b718582ec7b06c8.html) | | | Parameters object is precomputed state and is host-constructible. [More...](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 00ebd1a63351e1085d0b718582ec7b06c8.html#details)
| | | | class | [cutlass::transform::threadblock::PredicatedTileIterator< Shape_, Element_, layout::RowMajorInterleaved< InterleavedK >, AdvanceRank, ThreadMap_, AccessSize >](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 00d670f969180a8d182dffb356ebcc957e.html) | | | | class | [cutlass::transform::threadblock::PredicatedTileIterator< Shape_, Element_, layout::RowMajorInterleaved< InterleavedK >, AdvanceRank, ThreadMap_, AccessSize >::Params](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 009fd89f6dad84238fd7d63df0a0c0364f.html) | | | Parameters object is precomputed state and is host-constructible. [More...](classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape 00_01Element 009fd89f6dad84238fd7d63df0a0c0364f.html#details)
| | |

|

Namespaces

| | | cutlass | | | | | cutlass::transform | | | | | cutlass::transform::threadblock | | |

Detailed Description

This iterator uses masks to guard out-of-bounds accesses and visits the last "residue" tile first, with the objective of minimizing predicate mask updates during steady-state operation.

A precomputed "Params" object minimizes the amount of state that must be stored in registers, and integer addition is used to advance the pointer through memory.


Generated by 1.8.11