This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [patch] enabling vectorization by default at -O3
- From: "Daniel Berlin" <dberlin at dberlin dot org>
- To: "H.J. Lu" <hjl at lucon dot org>
- Cc: "Uros Bizjak" <ubizjak at gmail dot com>, "GCC Patches" <gcc-patches at gcc dot gnu dot org>, "Tobias Burnus" <burnus at net-b dot de>, "Dorit Nuzman" <DORIT at il dot ibm dot com>
- Date: Thu, 6 Sep 2007 09:23:31 -0400
- Subject: Re: [patch] enabling vectorization by default at -O3
- References: <5787cf470709060449u1237bb9of84d0b2f93247cb@mail.gmail.com> <20070906131328.GA23040@lucon.org>
On 9/6/07, H.J. Lu <hjl@lucon.org> wrote:
> On Thu, Sep 06, 2007 at 01:49:52PM +0200, Uros Bizjak wrote:
> > >
> > > * Hmm, why is --ffast-math slower? And with vectorization that much
>
> Also see
>
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32183
>
> With -O2 --ffast-math, we turn a faster loop:
>
> float sf;
> ...
> sf = 500 * sf;
> for (i = 0; i < ceplen; i++)
> sum[i] *= sf;
>
> into a slower loop:
>
> for (i = 0; i < ceplen; i++)
> sum[i] = (sum[i]* 500)*sf;
>
> > > slower? I recheck induct (V.F, NV.F) and I could reproduce the timings.
> > >
> >
> > > that is indeed interesting (I'd be happy to look at a testcase)
> >
> > This is PR 32084, http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32084
> >
I still don't remember why we have reassoc2. I'm in favor of removing
it unless someone can show it's producing performance improvements :)