To call each routine from the FORTRAN 77 program, you can use the LAPACK name.
LAPACK names are listed in Table "Computational Routines for Systems of Equations with Real Matrices" and Table "Computational Routines for Systems of Equations with Complex Matrices", and have the structure ?yyzzz or ?yyzz, which is described below.
The initial symbol ? indicates the data type:
real, single precision
complex, single precision
real, double precision
complex, double precision
Some routines can have combined character codes, such as ds or zc.
The second and third letters yy indicate the matrix type and storage scheme:
general
general band
general tridiagonal
diagonally dominant tridiagonal
symmetric or Hermitian positive-definite
symmetric or Hermitian positive-definite (packed storage)
symmetric or Hermitian positive-definite (RFP storage)
symmetric or Hermitian positive-definite band
symmetric or Hermitian positive-definite tridiagonal
symmetric indefinite
symmetric indefinite (packed storage)
Hermitian indefinite
Hermitian indefinite (packed storage)
triangular
triangular (packed storage)
triangular (RFP storage)
triangular band
The last three letters zzz indicate the computation performed:
perform a triangular matrix factorization
solve the linear system with a factored matrix
estimate the matrix condition number
refine the solution and compute error bounds
refine the solution and compute error bounds using extra-precise iterative refinement
compute the inverse matrix using the factorization
equilibrate a matrix.
For example, the sgetrf routine performs the triangular factorization of general real matrices in single precision; the corresponding routine for complex matrices is cgetrf.
Driver routine names can end with -sv (meaning a simple driver), or with -svx (meaning an expert driver) or with -svxx (meaning an extra-precise iterative refinement expert driver).
The Fortran 95 interfaces to the LAPACK computational and driver routines are the same as the FORTRAN 77 names but without the first letter that indicates the data type. For example, the name of the routine that performs a triangular factorization of general real matrices in Fortran 95 is getrf. Different data types are handled through the definition of a specific internal parameter that refers to a module block with named constants for single and double precision.
Copyright © 1994 - 2010, Intel Corporation. All rights reserved.