This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Dumb register allocation (PPC)
- From: law at redhat dot com
- To: Geoff Keating <geoffk at geoffk dot org>
- Cc: David Edelsohn <dje at watson dot ibm dot com>, gcc at gcc dot gnu dot org
- Date: Wed, 15 May 2002 12:26:39 -0600
- Subject: Re: Dumb register allocation (PPC)
- Reply-to: law at redhat dot com
In message <jmptzxb6m4.fsf@desire.geoffk.org>, Geoff Keating writes:
> David Edelsohn <dje@watson.ibm.com> writes:
>
> > >>>>> Richard Henderson writes:
> >
> > Richard> Eh? My objection is still going to be potential reload
> > Richard> failures. Which "fact" is going to ensure that you
> > Richard> introduce no failures on any target?
> >
> > So maybe this should be enabled on a target-by-target basis
> > instead of SMALL_REGISTER_CLASSES?
>
> Either the change is correct or it is not. If it is correct, it works
> for all targets. If it is not, it'll probably break something on most
> targets, and we don't want it.
So, with that in mind, does this code lengthen the lifetime of the
return register? If it does, then that's a serious correctness
issue.
jeff