Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!sun-barr!decwrl!sgi!davea@quasar.wpd.sgi.com
From: davea@quasar.wpd.sgi.com (David B. Anderson)
Newsgroups: comp.sys.sgi
Subject: Re: Irix 3.2 (4D) ldexp() function broken
Summary: yes, ldexp(0.0,n) is broken in 3.2
Message-ID: <46455@sgi.sgi.com>
Date: 15 Dec 89 23:17:56 GMT
References: <8912140834.aa03045@VGR.BRL.MIL>
Sender: davea@quasar.wpd.sgi.com
Organization: Silicon Graphics, Inc., Mountain View, CA
Lines: 13

In article <8912140834.aa03045@VGR.BRL.MIL>, gwyn@BRL.MIL (Doug Gwyn, VLD/VMB) writes:
[various lines deleted for brevity]
> 	printf( "%g\n", ldexp( 0.0, 0 ) );
> $ ./foo
> 1.11254e-308
> $ # answer should be "0"; works properly on all other BRL UNIX systems
> $ # I think the problem is that they're treating true-0 as a denormalized
> $ # number and "rounding" it.  This is RONG.

Doug is right on all counts.  We need to fix this.

Regards,
[ David B. Anderson    davea    Bldg-9    MS 9U500     extension 1548 ]