This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH][Aarch64] Fix vec_perm cost for thunderx2t99
- From: Anton Youdkevitch <anton dot youdkevitch at bell-sw dot com>
- To: Kyrylo Tkachov <Kyrylo dot Tkachov at arm dot com>, Andrew Pinski <pinskia at gmail dot com>
- Cc: GCC Patches <gcc-patches at gcc dot gnu dot org>
- Date: Tue, 5 Nov 2019 14:54:39 +0300
- Subject: Re: [PATCH][Aarch64] Fix vec_perm cost for thunderx2t99
- References: <20191101140242.GA7200@bell-sw.com> <CA+=Sn1=bxU_+b+U+00rX7NK83Wd2rUxON3XFL5+ZUdY-8Y_85Q@mail.gmail.com> <7aacefa1-461b-45b4-bfbc-09171745825e@arm.com>
Kyrill,
On 05.11.2019 14:43, Kyrylo Tkachov wrote:
Hi Andrew, Anton,
On 11/1/19 11:22 PM, Andrew Pinski wrote:
On Fri, Nov 1, 2019 at 7:03 AM Anton Youdkevitch
<anton.youdkevitch@bell-sw.com> wrote:
Hello,
Here is the one-liner that fixes the incorrect
vec_perm cost for thunderx2t99 chip.
With the patch applied 526.blender of CPU2017
gets ~5% improvement with no measurable changes
for other benchmarks.
Bootstrapped OK on aarch64-linux-gnu.
OK for trunk?
Maybe the big question is vec_perm used for both 1 input and 2 input
cases? If so maybe splitting the two cases would be important too.
Otherwise this is ok from my point of view but I can't approve it.
I'd be interested to see a testcase/demonstration where this would would
be beneficial.
Well, since I measured this on SPEC 2017, so, the result is
the overall benchmark score. I can try to extract the relevant
pieces of code that get compiled differently to see if they can
be make into a standalone testcase. I didn't try this yet, though.
In the meantime this patch is ok if it helps thunderx2t99 performance.
2019-11-01 Anton Youdkevitch <anton.youdkevitch@bell-sw.com>
* gcc/config/aarch64/aarch64.c (thunderx2t99_vector_cost):
change vec_perm field
ChangeLog nits:
* Two spaces between name and date+email
* No gcc/ prefix as the relevant ChangeLog file lives in gcc/
* End entry with full stop.
Thanks, will do like this next time.
Anton, do you need someone to commit this for you?
Yes, it would be nice if you can do this for me.
Thanks,
Kyrill
Thanks,
Andrew Pinski
2019-11-01 Anton Youdkevitch <anton.youdkevitch@bell-sw.com>
* gcc/config/aarch64/aarch64.c (thunderx2t99_vector_cost):
change vec_perm field
--
Thanks,
Anton