This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: patch: rs6000 specific
- From: Dale Johannesen <dalej at apple dot com>
- To: Geoff Keating <geoffk at redhat dot com>
- Cc: Dale Johannesen <dalej at apple dot com>, gcc-patches at gcc dot gnu dot org
- Date: Tue, 4 Dec 2001 15:52:05 -0800
- Subject: Re: patch: rs6000 specific
>> This adds some additional patterns for multiply-add instructions.
>> Bootstrapped and tested on Darwin. (The !POWERPC variants are
>> included, analogous to what was already there, but I was unable
>> to test them.)
>
> I think that some of these are cases that should never occur.
> Perhaps the generic code is not canonifing them properly?
I'm not sure they all do occur. I'll check out the ones you
indicated.
> It might be that some of these are not strictly IEEE compliant, in
> which case the generic code should still fold them... but only when
> -ffast-math is specified.
FMA is not IEEE compliant in general. I was surprised that it is on
by default (not just on Darwin). However, our local floating point
expert, who is much more conservative than I would be about IEEE
compliance, says this is the right thing to do.