cutlass::transform::threadblock::RegularTileIterator< Shape_, Element_, layout::VoltaTensorOpMultiplicandCrosswise< sizeof_bits< Element_ >::value, Shape_::kContiguous >, AdvanceRank, ThreadMap_, Alignment >::Detail Struct Reference

Internal details made public to facilitate introspection.

#include <regular_tile_iterator_tensor_op_sm70.h>

Static Public Attributes
static int const	kPointerCount = (ThreadMap::Iterations::kStrided > 1 ? 2 : 1)
	< Number of pointers More...
static int const	kIterarionsPerAccess
	Iterations for the kElementsPerAccess of ThreadMap. More...
static int const	kContiguousElementsPerLine = 4
	Contiguous elements per line. More...

Member Data Documentation

template<typename Shape_ , typename Element_ , int AdvanceRank, typename ThreadMap_ , int Alignment>

int const cutlass::transform::threadblock::RegularTileIterator< Shape_, Element_, layout::VoltaTensorOpMultiplicandCrosswise< sizeof_bits< Element_ >::value, Shape_::kContiguous >, AdvanceRank, ThreadMap_, Alignment >::Detail::kContiguousElementsPerLine = 4

static

template<typename Shape_ , typename Element_ , int AdvanceRank, typename ThreadMap_ , int Alignment>

int const cutlass::transform::threadblock::RegularTileIterator< Shape_, Element_, layout::VoltaTensorOpMultiplicandCrosswise< sizeof_bits< Element_ >::value, Shape_::kContiguous >, AdvanceRank, ThreadMap_, Alignment >::Detail::kIterarionsPerAccess

static

Initial value:

=
        ThreadMap::kElementsPerAccess / Layout::kElementsPerAccess

template<typename Shape_ , typename Element_ , int AdvanceRank, typename ThreadMap_ , int Alignment>

int const cutlass::transform::threadblock::RegularTileIterator< Shape_, Element_, layout::VoltaTensorOpMultiplicandCrosswise< sizeof_bits< Element_ >::value, Shape_::kContiguous >, AdvanceRank, ThreadMap_, Alignment >::Detail::kPointerCount = (ThreadMap::Iterations::kStrided > 1 ? 2 : 1)

static

---

The documentation for this struct was generated from the following file:

• regular_tile_iterator_tensor_op_sm70.h