Date: Thu, 7 Jan 1999 08:55:06 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Charles Krug <charles AT mail DOT pentek DOT com>
cc: djgpp AT delorie DOT com
Subject: Re: Looking for IEEE floating point definitions.
In-Reply-To: <3693C86C.E7698013@mail.pentek.com>
Message-ID: <Pine.SUN.3.91.990107085429.16723E-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp AT delorie DOT com


On Wed, 6 Jan 1999, Charles Krug wrote:

> I'm trying to find the header files containing definitions for
> NaN, POS_INF etc.  Do these exist?
> 
> Or are they defined internally somewhere, so that they can simply
> be used?

The math library libm.a supplied with DJGPP v2.02 includes functions
nan(), infinity(), nanf() and infinityf(), which return what you'd
expect (those whose names end in `f' return the float, as opposed to
double, NaN and Inf).  The Info file libm.info documents them (e.g.,
type "info libm math nan" from the DOS prompt).

Is this what you want?

> Of course, if djgpp doesn't use IEEE floating point numbers, I
> guess I need whatever the equivalent "thing" is under djgpp's
> floating point format.

DJGPP cannot do without IEEE FP numbers, since that's what the x87
processors support.