This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Loop optimizer issues
- From: Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>
- To: Richard Henderson <rth at redhat dot com>,David Edelsohn <dje at watson dot ibm dot com>,Geoff Keating <geoffk at geoffk dot org>,Diego Novillo <dnovillo at redhat dot com>,Jason Merrill <jason at redhat dot com>, Jan Hubicka <jh at suse dot cz>,Daniel Berlin <dberlin at dberlin dot org>, pop at gauvain dot u-strasbg dot fr,gcc at gcc dot gnu dot org
- Date: Wed, 2 Jul 2003 23:50:48 +0200
- Subject: Re: Loop optimizer issues
- References: <20030702211618.GA26127@atrey.karlin.mff.cuni.cz> <200307022122.RAA25428@makai.watson.ibm.com> <20030702214459.GE2115@redhat.com>
Hello,
> > I'm sorry, I'm still confused. If you are writing an RTL-based
> > loop optimizer, you cannot avoid supporting LIBCALLs. As long as you are
> > working in the RTL space, you cannot make the simplification that Tree-SSA
> > would allow.
>
> Indeed. At least for 3.4, when the tree-ssa optimizations aren't done.
there is no way how I could manage to finish the loop optimizer work
for 3.4, so I don't care.
> It's not clear what loop optimizations should be done at the rtl level
> when tree-ssa is present.
but surely the code motion I'm concerned with will at least partially be
done on tree-ssa (as loop invariant motion or pre or perhaps both).
Zdenek