The VML function names are lowercase for Fortran (vsabs) and of mixed (lower and upper) case for C (vsAbs).
The VML mathematical and pack/unpack function names have the following structure:
v[m]<?><name><mod>
where
v is a prefix indicating vector operations.
[m] is an optional prefix for mathematical functions that indicates additional argument to specify a VML mode for a given function call (see SetMode for possible values and their description).
<?> is a precision prefix that indicates one of the following the data types:
REAL for the Fortran interface, or float for the C interface
DOUBLE PRECISION for the Fortran interface, or double for the C interface.
COMPLEX for the Fortran interface, or MKL_Complex8 for the C interface.
DOUBLE COMPLEX for the Fortran interface, or MKL_Complex16 for the C interface.
<name> indicates the function short name, with some of its letters in uppercase for the C interface. See examples in Table "VML Mathematical Functions".
<mod> field (written in uppercase for the C interface) is present only in the pack/unpack functions and indicates the indexing method used:
indexing with a positive increment
indexing with an index vector
indexing with a mask vector.
The VML service function names have the following structure:
vml<name>
where
<name> indicates the function short name, with some of its letters in uppercase for the C interface. See examples in Table "VML Service Functions".
To call VML functions from an application program, use conventional function calls. For example, call the vector single precision real exponential function as
call vsexp ( n, a, y ) for the Fortran interface, or
call vmsexp ( n, a, y, mode ) for the Fortran interface with a specified mode, or
vsExp ( n, a, y ); for the C interface.
Copyright © 1994 - 2010, Intel Corporation. All rights reserved.