Subject: Re: Floating point .....
To: Phil Nelson <phil@cs.wwu.edu>
From: Chris G. Demetriou <cgd@netbsd.org>
List: port-arm32
Date: 10/27/1998 09:59:48
Phil Nelson <phil@cs.wwu.edu> writes:
> Can anyone verify that this is the case.  And if so, is it something that
> should be "fixed" so it looks like IEEE double precision?

Oh yeah:

Does IEEE double precision just specify the fields, field widths, and
meaning, or does it actually specify the memory layout?

my understanding was that it only did the former, i.e. that it
_didn't_ specify the memory layout, but then, i've never seen the
standard.  8-)

In any case, what you've found is an IEEE double, but the way it's
stored in memory on the ARM really isn't the way it's done on either
an x86 or sparc (to pick little and big endian examples).


cgd
-- 
Chris Demetriou - cgd@netbsd.org - http://www.netbsd.org/People/Pages/cgd.html
Disclaimer: Not speaking for NetBSD, just expressing my own opinion.