This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: NO SUBJECT
- To: "egcs at cygnus dot com" <egcs at cygnus dot com>, "hjstein at bfr dot co dot il" <hjstein at bfr dot co dot il>, "moshier at mediaone dot net" <moshier at mediaone dot net>, "tprince at cat dot e-mail dot com" <tprince at cat dot e-mail dot com>
- Subject: Re: NO SUBJECT
- From: "Geert Bosch" <bosch at gnat dot com>
- Date: Thu, 17 Dec 1998 17:44:34 +0100
- Reply-To: "Geert Bosch" <bosch at gnat dot com>
On Thu, 17 Dec 1998 11:29:29 EST, tprince@cat.e-mail.com wrote:
>>Therefore, I think the stuff we'll tend to "fix" by adopting my
proposal will be function return values, offhand, not by what
we think of as "normal" spills -- because it seems that normal
spills in straight-line code are pretty rare.<<
Not so rare that people don't complain about unexpected results
arising from the difference between 80-bit and 64-bit evaluation,
which ought to be corrected by the 80-bit spilling scheme.
In my experience this mainly occurs for non-optimized code. Floating-point
algorithms that give wrong results without optimization, often produce
correct results with optimization. This amazes people and especially those
that are used to (old) compilers that do it the other way around... :-)
Regards,
Geert