Date: Tue, 14 Mar 2000 17:46:07 -0500
Message-Id: <200003142246.RAA20220@envy.delorie.com>
From: DJ Delorie <dj AT delorie DOT com>
To: djgpp-workers AT delorie DOT com
In-reply-to: <200003142156.WAA20972@father.ludd.luth.se> (message from Martin
	Str|mberg on Tue, 14 Mar 2000 22:56:22 +0100 (MET))
Subject: Re: Unnormals???
References:  <200003142156 DOT WAA20972 AT father DOT ludd DOT luth DOT se>
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


> Can NaNs be negative?

Since FP numbers have an explicit sign bit, every FP bit pattern can
be negative or positive.  Whether or not it makes sense to reflect
that is a different story; most software doesn't care about the
difference between +0 and -0.

In the case of NaNs, knowing the sign is probably useful for
debugging, since the sign is often preserved even when the math
over/underflows or becomes undefined (division by zero).