Xref: utzoo comp.unix.ultrix:7465 comp.lang.fortran:5586
Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!usc!samsung!sol.ctr.columbia.edu!emory!hubcap!elwin
From: elwin@hubcap.clemson.edu (lawrence brown)
Newsgroups: comp.unix.ultrix,comp.lang.fortran
Subject: Re: How to detect NaN's
Message-ID: <1991Jun3.143050.20493@hubcap.clemson.edu>
Date: 3 Jun 91 14:30:50 GMT
References: <1991May30.204332.16506@litwin.com>
Organization: Clemson University
Lines: 18

From article <1991May30.204332.16506@litwin.com>, by vlr@litwin.com (Vic Rice):
> I have encountered a problem on a DECStation 5000 using the Mips Fortran
> compiler. Some operation I am performing is generating a NaN which then
> proceeds to propogate to other dependent variables. Since this is in
> a loop, I am not exactly sure where it starts.
> 
> How can I test a variable for the presence of a NaN or Infinity ???
> -- 
> Dr. Victor L. Rice
> Litwin Process Automation
Our compiler (Fortran for RISC version 2.0 I think) has a barely
documented option that will trap any NaN's and generate an illegal
instruction error.  It also sets all variables to NaN at the beginning
of execution (rather than setting them to zero) so it will test for 
unassigned variable usage (e.g. misspelled variable names). Try
f77 -trapuv
Hope this helps.
Larry Brown