From: "H.J. Lu" <hjl.tools@gmail.com>
To: "Fang, Changpeng" <Changpeng.Fang@amd.com>
Cc: Richard Guenther <richard.guenther@gmail.com>,
"gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH, PR 49089] Don't split AVX256 unaligned loads by default on bdver1 and generic
Date: Fri, 17 Jun 2011 01:10:00 -0000 [thread overview]
Message-ID: <BANLkTin7cnq+SNAA37mSx91soQ2jMLn_EA@mail.gmail.com> (raw)
In-Reply-To: <D4C76825A6780047854A11E93CDE84D005980DC6F8@SAUSEXMBP01.amd.com>
On Thu, Jun 16, 2011 at 4:54 PM, Fang, Changpeng <Changpeng.Fang@amd.com> wrote:
> Hi,
>
> I modify the patch to disable unaligned load splitting only for bdver1 at this moment.
> Unaligned load splitting degrades CFP2006 by 1.3% in geomean for both -mtune=bdver1 and
> -mtune=generic on Bulldozer. However, we agree with H.J's suggestion to determine
> the optimal optimization sets for modern cpus.
>
> Is is OK to commit the attached patch?
>
> Thanks,
Why not just move AVX256_SPLIT_UNALIGNED_STORE
and AVX256_SPLIT_UNALIGNED_LOAD to ix86_tune_indices?
H.J.
> Changpeng
>
>
>
>>> So, is it OK to commit this patch to trunk, and H.J's original patch + this to 4.6 branch?
>>>
>>>
>
>>I have no problems on -mtune=Bulldozer. But I object -mtune=generic
>>change and did suggest a different approach for -mtune=generic.
>
>
> .
>
>
--
H.J.
next prev parent reply other threads:[~2011-06-17 0:53 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-06-14 1:32 Fang, Changpeng
2011-06-14 10:17 ` Richard Guenther
2011-06-14 10:20 ` Jakub Jelinek
2011-06-14 13:24 ` H.J. Lu
2011-06-14 23:22 ` Fang, Changpeng
2011-06-15 0:21 ` H.J. Lu
2011-06-14 22:47 ` Fang, Changpeng
2011-06-14 23:15 ` H.J. Lu
2011-06-15 0:35 ` Fang, Changpeng
2011-06-15 2:54 ` H.J. Lu
2011-06-15 22:07 ` Fang, Changpeng
2011-06-16 7:51 ` Richard Guenther
2011-06-17 0:53 ` Fang, Changpeng
2011-06-17 1:10 ` H.J. Lu [this message]
2011-06-17 18:08 ` Fang, Changpeng
2011-06-17 18:33 ` H.J. Lu
2011-06-17 22:49 ` Fang, Changpeng
2011-06-17 23:12 ` H.J. Lu
2011-06-20 18:44 ` Fang, Changpeng
2011-06-20 18:51 ` Uros Bizjak
2011-06-20 22:17 ` Fang, Changpeng
2011-06-29 16:19 ` Eric Botcazou
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=BANLkTin7cnq+SNAA37mSx91soQ2jMLn_EA@mail.gmail.com \
--to=hjl.tools@gmail.com \
--cc=Changpeng.Fang@amd.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=richard.guenther@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).