This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [www-patch] bugs.html rewrite, part 2: revamp floating point section
- From: Bradley Lucier <lucier at math dot purdue dot edu>
- To: Volker Reichelt <reichelt at igpm dot rwth-aachen dot de>
- Cc: Bradley Lucier <lucier at math dot purdue dot edu>, gcc-patches at gcc dot gnu dot org, gerald at pfeifer dot com
- Date: Tue, 19 Aug 2003 07:40:12 -0500
- Subject: Re: [www-patch] bugs.html rewrite, part 2: revamp floating point section
On Tuesday, August 19, 2003, at 06:40 AM, Volker Reichelt wrote:
On 18 Aug, Bradley Lucier wrote:
I would claim that this *is* a bug in the compiler, but it's one
that's
extremely unlikely to be fixed in the foreseeable future, and we've
rehashed this all before.
I'd claim that the compiler is allowed to round or not to round
floating
point numbers whenever it pleases. This includes comparisons and memory
operations.
Your original example does not illustrate a question of rounding at
all, it's a question of whether the compiler actually respects the
programmer's declarations of the precision of variables and
expressions. See
http://gcc.gnu.org/ml/gcc-patches/2001-03/msg00466.html
and Toon's message that follows it, which contains a link the previous
thread on this issue in 1999. (There are other interesting messages in
these threads, too.)
Unfortunately, I think your understanding of how a compiler can/should
best use the extended precision arithmetic is incomplete. Now that I
(sort of) understand what you're example is trying to show, I think it
should be eliminated completely.
Brad