STPMV/DTPMV/CTPMV/ZTPMV

Matrix-vector multiply

Notes These subprograms conform to specifications of the Level 2 BLAS.

If an error in the arguments is detected, the subprograms call error handler XERBLA, which writes an error message onto the standard error file and terminates execution. The standard version of XERBLA (refer to the end of this chapter) can be replaced with a user-supplied version to change the error procedure. Error conditions are:

uplo ≠ ’L’ or ’l’ or ’U’ or ’u’

trans ≠ ’N’ or ’n’ or ’T’ or ’t’ or ’C’ or ’c’ diag ≠ ’N’ or ’n’ or ’U’ or ’u’

n < 0 incx = 0

Actual character arguments in a subroutine call can be longer than the corresponding dummy arguments. Therefore, readability of the CALL statement can be improved by coding the trans argument as ’NORMAL’ or ’NONTRANS’ for ’N’, ’TRANSPOSE’ for ’T’, or ’CTRANS’ for ’C’. Refer to “Example 2.”

Example 1 Form the REAL*4 matrix-vector product Ax, where A is a 9-by-9 unit-diagonal, lower-triangular real matrix stored in packed form in an array AP of dimension 55 and x is a real vector 9 elements long stored in an array X of dimension 10.

CHARACTER*1 UPLO,TRANS,DIAG

INTEGER*4 N,INCX

REAL*4 AP(55),X(10)

UPLO = ’L’

TRANS = ’N’

DIAG = ’U’

N = 9

INCX = 1

CALL STPMV (UPLO,TRANS,DIAG,N,AP,X,INCX)

Example 2 Form the REAL*8 matrix-vector product ATx, where A is a 9-by-9 nonunit-diagonal, upper-triangular real matrix stored in packed form in an array AP of dimension 55 and x is a real vector 9 elements long stored in an array X of dimension 10.

INTEGER*4

N

REAL*8

AP(55),X(10)

N = 6

 

CALL DTPMV (’UPPER’,’TRANSPOSE’,’NONUNIT’,N,AP,X,1)

312HP MLIB User’s Guide