This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: infinity != infinity
- From: amylaar at spamcop dot net (Joern Rennecke)
- To: schwab at suse dot de (Andreas Schwab)
- Cc: joern dot rennecke at superh dot com (Joern Rennecke), paolo dot bonzini at polimi dot it (Paolo Bonzini), gcc at gcc dot gnu dot org
- Date: Sat, 10 Jul 2004 01:27:02 +0100 (BST)
- Subject: Re: infinity != infinity
> Joern Rennecke <joern.rennecke@superh.com> writes:
>
> > Thus, two infinities with the same sign should compare unordered.
>
> Andreas Schwab:
> The m68881 disagrees, and it is supposed to be IEEE compliant.
I know that there are a number of processors that have this behaviour,
but that is not the point. The issue is what the standard actually says.
By the way, if you don't want to consider infinites as stand-ins for
limit operations like the standard does, but as cardinal numbers in
their own right, you are still stuck with the fact that aleph0 != aleph1
etc. Besides, infinite is also used to express an infinite set (and range)
of finite numbers which are outside of the range representable with
normalized numbers of the selected precision.