This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Enabling -ftree-slp-vectorize on -O2/Os
- From: Richard Biener <richard dot guenther at gmail dot com>
- To: Allan Sandfeld Jensen <linux at carewolf dot com>
- Cc: GCC Development <gcc at gcc dot gnu dot org>
- Date: Wed, 30 May 2018 11:25:35 +0200
- Subject: Re: Enabling -ftree-slp-vectorize on -O2/Os
- References: <2659301.XPQk3P0qmd@twilight> <CAFiYyc3_XbQ5KHHM44XAQyGMZ1aysmX-o=g-UwBA3RmH1YzVLA@mail.gmail.com> <CAFiYyc3EnfcCGuXJEhK4UNDSTW-Eci3ncMfPmED=XyUWGz9qZw@mail.gmail.com> <5190617.IRreCldYRv@twilight>
On Tue, May 29, 2018 at 5:24 PM Allan Sandfeld Jensen <linux@carewolf.com>
wrote:
> On Dienstag, 29. Mai 2018 16:57:56 CEST Richard Biener wrote:
> >
> > so the situation improves but isn't fully fixed (STLF issues maybe?)
> >
> That raises the question if it helps in these cases even in -O3?
That's a good question indeed. We might end up (partly) BB vectorizing
loop bodies that we'd otherwise loop vectorize with SLP. Benchmarking
with BB vectorization disabled at -O3+ isn't something I've done in the
past. I'm now doing a 2-run with -march=haswell -Ofast
[-fno-tree-slp-vectorize]
for the FP benchmarks.
Note that there were some cases where disabling vectorization wholly
improved things.
> Anyway it doesn't look good for it. Did the binary size at least improve
with
> prefer-avx128, or was that also worse or insignificant?
Similar to the AVX258 results. I guess where AVX256 applied we now simply
do two vector ops with AVX128.
Richard.
> 'Allan