VECLIB naming convention

Matching subprograms for real and complex data have been coded to maintain a close correspondence between the two. However, in some areas, the correspondence is necessarily weaker, and this has not been possible.

Subprograms in VECLIB are provided in both 32-bit and 64-bit addressing versions, except in the VECLIB8 library where only 64-bit addressing is supported.

VECLIB naming convention

The name of each VECLIB subprogram is a coded specification of its function within the limits of standard Fortran 77 6-character names. Usually, the first character of a subprogram name, denoted by T, shows the predominant data type according to Table 1-3:

Table 1-3 VECLIB Naming Convention—Data Type

T

Data Type

VECLIB

VECLIB8

 

 

 

 

S

Single Precision

REAL*4

REAL*4

D

Double Precision

REAL*8

REAL*8

I

Integer

INTEGER*4

INTEGER*8

C

Complex

COMPLEX*8

COMPLEX*8

Z

Double Complex

COMPLEX*16

COMPLEX*16

 

 

 

 

Basic Linear Algebra Subprograms, that is, level 1, 2, and 3 BLAS, as well as the Sparse BLAS use this naming convention. For example, subprograms that compute the sum of vector elements are named according to data type: SSUM, DSUM, ISUM, CSUM, and ZSUM. Some function subprograms use two of these letters, the first describing the data type of the function and the second indicating the type of data on which it operates.

Fortran 77 BLAS Standard subprograms use a similar convention, with F_ prepended to each routine name.

For example, the legacy BLAS single-precision, triangular-solve routine is named STRSM and its BLAS Standard counterpart is named F_STRSM. BLAS Standard subprograms that compute the sum of vector elements are named F_SSUM, F_DSUM, F_CSUM, and F_ZSUM. Refer to “Legacy BLAS routines” on page 211 and “BLAS Standard routines” on page 339 for more information.

C BLAS Standard subprograms begin with C_ prepended to each routine name.

24HP MLIB User’s Guide