X-Authentication-Warning:	acp3bf.physik.rwth-aachen.de: broeker owned process doing -bs
Date:	Wed, 22 Mar 2000 15:44:33 +0100 (MET)
From:	Hans-Bernhard Broeker <broeker AT physik DOT rwth-aachen DOT de>
X-Sender:	broeker AT acp3bf
To:	djgpp-workers AT delorie DOT com
Subject:	Re: Unnormals???
In-Reply-To:	<200003221407.JAA07300@delorie.com>
Message-ID:	<Pine.LNX.4.10.10003221526130.29657-100000@acp3bf>
MIME-Version:	1.0
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

On Wed, 22 Mar 2000, Dieter Buerssner wrote:
> On 22 Mar 00, Eli Zaretskii wrote:

> > When converted to a long double, these two have the following bit
> > patterns:
> > 
> >  pos_nanshort = 7fff 0001 0000 0000
> >  neg_nanshort = ffff 0001 0000 0000

> > These are indeed unnormals: their mantissa has a zero MSB.
> 
> I think, they are not unnormals. I think this discussion has shown,
> that unnormals must have a finite exponent.

Not necessarily. My literature is not decisive on this, as pointed out
before, but I take these numbers are the so-called 'pseudo-NaNs' mentioned
by the book, but without a definition what that really is. They are not
normalized, obviously . If you do normalize them, you end up with a bit
pattern that is not a NaN any more, but a regular (albeit large) number).

7fff 0001 0000 0000 0000 e.g., would normalize into
7ffc 1000 0000 0000 0000, which is a large, but finite value.

So they are 'Pseudo-NaN', in the sense that they look a bit like NaN, but
aren't, as they do not have the leading '1' bit in the mantissa required
for NaNs, or any number that is not an unnormal.

> I think it should, and it also did this before the unnormal check was
> added. (Yes, it was my error not thinking about the unnormal case when
> adding long double support to _doprint.)

Hans-Bernhard Broeker (broeker AT physik DOT rwth-aachen DOT de)
Even if all the snow were burnt, ashes would remain.

- Raw text -