Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!caen!uflorida!bernhold
From: bernhold@red8 (David E. Bernholdt)
Newsgroups: comp.lang.fortran
Subject: Re: vectorization question
Message-ID: <27736@uflorida.cis.ufl.EDU>
Date: 29 Mar 91 17:44:54 GMT
References: <1991Mar29.141313.7418@ariel.unm.edu>
Sender: news@uflorida.cis.ufl.EDU
Organization: Quantum Theory Project, Univ. of Florida
Lines: 29

In article <1991Mar29.141313.7418@ariel.unm.edu> prentice@triton.unm.edu (John Prentice) writes:
>Consider the following loop:
>
>      do 30 k=1,kmax
>	  do 20 j=1,jmax
>	      do 10 i=1,imax
>		  a(i,j,k)=...
>   10         continue
>   20     continue
>   30 continue

Pretty obvious, and possibly not useful in this case, but...

How about passing a in as a 1-d array, dimension kmax*jmax*imax?

Problems:
1) if the physical dimension of the matrix is larger than the
computational dimension.
2) if the rhs has explicit dependence on i, j, k rather than the
meta-index.

If you get any better responses, please cc to me or post 'em.


-- 
David Bernholdt			bernhold@qtp.ufl.edu
Quantum Theory Project		bernhold@ufpine.bitnet
University of Florida
Gainesville, FL  32611		904/392 6365