This is the mail archive of the
gsl-discuss@sources.redhat.com
mailing list for the GSL project.
gsl interface to LAPACK?
- To: gsl-discuss at sourceware dot cygnus dot com
- Subject: gsl interface to LAPACK?
- From: José Luis Gómez Dans <j dot l dot gomez-dans at sheffield dot ac dot uk>
- Date: Fri, 20 Jul 2001 12:48:28 +0100
- Reply-To: j dot l dot gomez-dans at sheffield dot ac dot uk
Hi,
I am beginning to use GSL, and I am both a newbie in both GSL
and C. So I apologise if I ask something really stupid.
I was playing with GSL to convert some code I have in Fortran
95. Mostly, I got it to work. The problem is that I need to solve a
generalised eiganvalue/vector problem (complex, non-Hermitian).
According to the manuals, you need to use LAPACK for this. I understand
that (I have used LAPACK in F77), but I now have all my matrices and
vectors as gsl_complex_matrix. If I want to use LAPACK, I have to get
the data into a CLAPACK friendly format, which is rather tedious and
error-prone (there might be many an instance when this might apply).
Moreover, the matrices might be Hermitian or not...
I guess my questions is whether someone has written a driver for
LAPACK, for which you pass a matrix and the software decides how to call
the appropriate LAPACK routine.
Thank you for your time,
José
--
José L Gómez Dans PhD student
Tel: +44 114 222 5582 Radar & Communications Group
FAX; +44 870 132 2990 Department of Electronic Engineering
University of Sheffield UK