Cyclops Tensor Framework
parallel arithmetic on multidimensional arrays
|
#include "../tensor/algstrct.h"
Go to the source code of this file.
Namespaces | |
CTF_int | |
Functions | |
void | CTF_int::nosym_transpose (int order, int const *new_order, int const *edge_len, char *data, int dir, algstrct const *sr) |
transposes a non-symmetric (folded) tensor More... | |
double | CTF_int::est_time_transp (int order, int const *new_order, int const *edge_len, int dir, algstrct const *sr) |
estimates time needed to transposes a non-symmetric (folded) tensor based on performance models More... | |
void | CTF_int::nosym_transpose (tensor *A, int all_fdim_A, int const *all_flen_A, int const *new_order, int dir) |
void | CTF_int::nosym_transpose (int order, int const *new_order, int const *edge_len, char const *data, int dir, int max_ntd, char **tswap_data, int64_t *chunk_size, algstrct const *sr) |
transposes a non-symmetric (folded) tensor internal kernel More... | |
bool | CTF_int::hptt_is_applicable (int order, int const *new_order, int elementSize) |
Checks if the HPTT library is applicable. More... | |
void | CTF_int::nosym_transpose_hptt (int order, int const *edge_len, int dir, tensor *&A) |
High-performance implementation of nosym_transpose using HPTT. More... | |