Date: Mon, 20 Mar 2000 18:11:57 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Eric Rudd <rudd AT cyberoptics DOT com>
cc: djgpp-workers AT delorie DOT com
Subject: Re: Unnormals???
In-Reply-To: <38D64B24.CF902909@cyberoptics.com>
Message-ID: <Pine.SUN.3.91.1000320180609.26722H@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: dj-admin AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Mon, 20 Mar 2000, Eric Rudd wrote:

> > > On my Pentium II, I found that -NaN = -NaN * -NaN; all other sign
> > > combinations produced +NaN.
> >
> > This contradicts the Intel manual.  (So what else is new?)
> 
> I have been poring over the Intel manuals and cannot find anything about the sign
> of a product of NaNs.  Where did you find the information?

"Intel Architecture Software Developer's Manual", v.1 "Basic 
Architecture" (I downloaded it from their site as 24319002.pdf, but that 
was quite a while ago), Section 7.6, Table 7-18.  The Result column only 
mentions a real indefinite if neither of the operands is a NaN.  Your 
case appears to be covered by the third possible combination (two QNaNs), 
whose result should be a QNaN, i.e. either with the sign bit reset or 
with a mantissa that doesn't fit the real indefinite description.