Date: Tue, 19 Jan 1999 20:12:17 -0500 (EST)
From: Stephen L Moshier <moshier AT mediaone DOT net>
X-Sender: moshier AT moshier DOT ne DOT mediaone DOT net
To: DJ Delorie <dj AT delorie DOT com>
cc: djgpp-workers AT delorie DOT com, robert DOT hoehne AT gmx DOT net
Subject: Re: Bug when printing long doubles
In-Reply-To: <199901191742.MAA10687@envy.delorie.com>
Message-ID: <Pine.LNX.4.05.9901191949300.16986-100000@moshier.ne.mediaone.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com



On Tue, 19 Jan 1999, DJ Delorie wrote:
> I also suspect that there are more NaN patterns than just the one that
> is "the" NaN pattern.

Yes, IEEE says for single and double precision a NaN has the maximum
exponent value and any nonzero fraction.  What the coprocessor
actually generates doesn't limit the set of garbage that printf
must cope with.

Extended formats are required to have at least one quiet and one
signaling NaN, but the bit patterns are not specified.  IEEE further says
"An implementation may also reserve some bit strings for purposes
beyond the scope of this standard" (IEEE 754-1985 para 3.3).