From: Martin Raspaud
Subject: Lapack ?
Date: 
Message-ID: <c7t6ii$i1h$1@news.u-bordeaux.fr>
Hi all

Does anyone know if there is a 100% Common lisp implementation of Lapack 
routines for matrix inversion ?

Thanx

Martin
From: Nicolas Neuss
Subject: Re: Lapack ?
Date: 
Message-ID: <87u0ylbye2.fsf@ortler.iwr.uni-heidelberg.de>
Martin Raspaud <········@labri.fr> writes:

> Hi all
> 
> Does anyone know if there is a 100% Common lisp implementation of Lapack
> routines for matrix inversion ?
> 
> Thanx
> 
> Martin

My PDE toolbox Femlisp (<http://www.femlisp.org>) contains such code (basic
matrix arithmetic and Gauss elimination with pivoting) in the directory
"femlisp:src;matlisp;".  It does matrix class generation and method
compilation adapted to the matrix element type at runtime and should
therefore be not much slower than ordinary LAPACK.  However, the cache- and
architecture-optimized ATLAS routines (which can be used from CL with
Matlisp, see <http://matlisp.sourceforge.net/>) are much faster for
medium-sized and large matrices.

Yours, Nicolas.