Developer Reference for Intel® oneAPI Math Kernel Library - C

Developer Reference

Contents

mkl_graph_optimize_mxv

Computes internal optimized data for a graph matrix to be used in a matrix-vector product.

Syntax

mkl_graph_status_t mkl_graph_optimize_mxv(mkl_graph_vector_t mask, mkl_graph_semiring_t semiring, mkl_graph_matrix_t A, mkl_graph_vector_t v, mkl_graph_descriptor_t desc, int64_t ncalls);
Include Files
  • mkl_graph.h
Description
The
mkl_graph_optimize_mxv
 routine computes internal optimized data for a graph matrix-vector product
mkl_graph_mxv
. Input arguments match the definitions of the arguments for
mkl_graph_mxv
. You can provide as much information about future calls to
mkl_graph_mxv
 as is known. If mask, input vector or descriptor is not known, NULL can be passed as a corresponding argument.
The last argument,
ncalls
, estimates the number of calls to
mkl_graph_mxv
 with the matrix
A
. More aggressive optimizations may be performed for large values of
ncalls
. If such an estimate is not known, a negative value can be passed to use the default behavior.
Currently the optimizations mainly target the matrix
A
. Optimized data are stored inside the graph matrix and will be deallocated once non-compatible matrix changes are made or when the matrix is destroyed via
mkl_graph_matrix_destroy
.
Input Parameters
mask
A graph vector which will be used as a mask in
mkl_graph_mxv
. If it is not known or will not be present in the operation, NULL can be passed.
semiring
A semiring which will be used in
mkl_graph_mxv
. Refer to Graph API Glossary for more details. If it is not known, any value from
mkl_graph_semiring_t
 can be used.
A
A graph matrix which will be used in
mkl_graph_mxv
. Must be known and not NULL.
V
A graph vector which will be used as input vector in
mkl_graph_mxv
. If it is not known or it will change between multiple calls to
mkl_graph_mxv
, NULL can be passed.
desc
A graph descriptor which will define additional modifiers for
mkl_graph_mxv
. If it is not known, will change between calls to
mkl_graph_mxv
, or will not be used in calls to
mkl_graph_mxv
, NULL can be passed.
ncalls
An estimate on the number of calls to the matrix-vector product which will follow the call to
mkl_graph_optimize_mxv
 with the matrix
A
. If negative, the routine uses the default strategy.
Output Parameters
A
A graph matrix with the optimized data for a matrix-vector product.
Return Values
The function returns a value indicating whether the operation was successful or not and why. Refer to Graph API Glossary for a list of possible options.
Product and Performance Information
Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.
Notice revision #20201201

Product and Performance Information

1

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.