Re: _FINAL_ release MiNTlib PL48

[Yves Pelletier <ypell@cam.org>]

On 10 Jun 1998, Andreas Schwab wrote:

> Here is a complete float.h, all constants have been verified.  This will
> be part of egcs.

Thanks, Andreas.  The values you supplied are in agreement with
what was in PL <= 47 float.h.  But as we know, atof.c (and
possibly other programs) cannot be assembled with gas 2.9 using
these values. So then the question is, why does the new gas
diverge from the IEEE standard, and exactly which values does it
accept?

Yves