This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: complete_unrolli / complete_unroll
- From: Richard Guenther <richard dot guenther at gmail dot com>
- To: Dominique Dhumieres <dominiq at lps dot ens dot fr>
- Cc: gcc at gcc dot gnu dot org, Albert dot Cohen at inria dot fr
- Date: Thu, 20 Aug 2009 11:57:10 +0200
- Subject: Re: complete_unrolli / complete_unroll
- References: <20090820094807.C6A5C3BE85@mailhost.lps.ens.fr>
On Thu, Aug 20, 2009 at 11:48 AM, Dominique Dhumieres<dominiq@lps.ens.fr> wrote:
> IIRC another code that is "improved" by complete_unrolli is the polyhedron
> test induct.f90. ?However it gives worse results for some variants
> (see pr34265: induct_v2/3).
>
>> Can't we use graphite to re-roll loops? ...
>
> Is doing and undoing always some kind of work?
Yes it is, but you can only reliably find out if unrolling will enable
further optimizations if you do the unrolling. If it didn't enable anything
re-rolling the loop should be possible. And within the polyhedral
representation it should even be easy (I hope).
Richard.
> Cheers
>
> Dominique
>