This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: -floop-optimize2
- From: Richard Guenther <richard dot guenther at gmail dot com>
- To: Scott Robert Ladd <scott dot ladd at coyotegulch dot com>
- Cc: GCC Mailing List <gcc at gcc dot gnu dot org>
- Date: Tue, 21 Jun 2005 18:18:08 +0200
- Subject: Re: -floop-optimize2
- References: <42B83B95.5090102@coyotegulch.com>
- Reply-to: Richard Guenther <richard dot guenther at gmail dot com>
On 6/21/05, Scott Robert Ladd <scott.ladd@coyotegulch.com> wrote:
> I've noticed that -floop-optimize2 tends to be a pessimism on many
> algorithms.
>
> I'm hesitant to file this as a "bug", given that -floop-optimize2 is a
> "new" replacement for the older loop optimizer.
>
> Is -floop-optimize2 still in development, and not ready yet -- or are
> the problems I'm seeing something that should be analyzed and reported
> as a bug?
The pessimization is that the old loop optimizer is turned off, if you
specify -floop-optimize2. The new loop optimizer is also (additionally)
enabled if you specify either of -fmove_loop_invariants
-funswitch_loops -fpeel_loops -funroll_loops or -fbranch_on_count_reg.
As the old loop optimizer is (hopefully) going away sooner or later
small testcases are surely welcome that identify weak spots of the new
loop optimizer.
Richard.