From: Al Viro <viro@zeniv.linux.org.uk>
To: lkp@lists.01.org
Subject: Re: [x86] d55564cfc2: will-it-scale.per_thread_ops -5.8% regression
Date: Thu, 07 Jan 2021 18:55:59 +0000 [thread overview]
Message-ID: <20210107185559.GI3579531@ZenIV.linux.org.uk> (raw)
In-Reply-To: <20210107184058.GH3579531@ZenIV.linux.org.uk>
[-- Attachment #1: Type: text/plain, Size: 809 bytes --]
On Thu, Jan 07, 2021 at 06:40:58PM +0000, Al Viro wrote:
> do_sys_poll(): do the wholesale copyout
>
> Don't bother with patching up just one field - 16 bits out of each 64.
> The amount of memory traffic is not going to be greater (might be
> smaller, actually) and the loop in copy_to_user() is optimized for
> bulk copy.
BTW, considering the access pattern, I would expect it to be
considerably cheaper in a lot of cases; basically, we have a copy
of userland array of 64bit values, then we do a non-trivial amount
of work and modify 16 bits out of each 64. Then we want that
propagated back to the original array. I suspect that copying just
those 16bit fields out is going to cost more that a bulk copy of
the entire thing, and not just on s390 and similar oddball cases.
Comments?
WARNING: multiple messages have this Message-ID (diff)
From: Al Viro <viro@zeniv.linux.org.uk>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: kernel test robot <oliver.sang@intel.com>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@kernel.org>, Borislav Petkov <bp@alien8.de>,
Peter Zijlstra <peterz@infradead.org>,
LKML <linux-kernel@vger.kernel.org>,
lkp@lists.01.org, kernel test robot <lkp@intel.com>,
"Huang, Ying" <ying.huang@intel.com>,
Feng Tang <feng.tang@intel.com>,
zhengjun.xing@intel.com
Subject: Re: [x86] d55564cfc2: will-it-scale.per_thread_ops -5.8% regression
Date: Thu, 7 Jan 2021 18:55:59 +0000 [thread overview]
Message-ID: <20210107185559.GI3579531@ZenIV.linux.org.uk> (raw)
In-Reply-To: <20210107184058.GH3579531@ZenIV.linux.org.uk>
On Thu, Jan 07, 2021 at 06:40:58PM +0000, Al Viro wrote:
> do_sys_poll(): do the wholesale copyout
>
> Don't bother with patching up just one field - 16 bits out of each 64.
> The amount of memory traffic is not going to be greater (might be
> smaller, actually) and the loop in copy_to_user() is optimized for
> bulk copy.
BTW, considering the access pattern, I would expect it to be
considerably cheaper in a lot of cases; basically, we have a copy
of userland array of 64bit values, then we do a non-trivial amount
of work and modify 16 bits out of each 64. Then we want that
propagated back to the original array. I suspect that copying just
those 16bit fields out is going to cost more that a bulk copy of
the entire thing, and not just on s390 and similar oddball cases.
Comments?
next prev parent reply other threads:[~2021-01-07 18:55 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-07 13:47 [x86] d55564cfc2: will-it-scale.per_thread_ops -5.8% regression kernel test robot
2021-01-07 13:47 ` kernel test robot
2021-01-07 17:43 ` Linus Torvalds
2021-01-07 17:43 ` Linus Torvalds
2021-01-07 18:33 ` Al Viro
2021-01-07 18:33 ` Al Viro
2021-01-07 18:40 ` Al Viro
2021-01-07 18:40 ` Al Viro
2021-01-07 18:55 ` Al Viro [this message]
2021-01-07 18:55 ` Al Viro
2021-01-07 18:47 ` Linus Torvalds
2021-01-07 18:47 ` Linus Torvalds
2021-01-07 18:58 ` Al Viro
2021-01-07 18:58 ` Al Viro
2021-01-07 19:04 ` Al Viro
2021-01-07 19:04 ` Al Viro
2021-01-07 19:33 ` Linus Torvalds
2021-01-07 19:33 ` Linus Torvalds
2021-01-07 19:57 ` Al Viro
2021-01-07 19:57 ` Al Viro
2021-01-08 9:29 ` David Laight
2021-01-08 9:29 ` David Laight
2021-01-08 9:37 ` David Laight
2021-01-08 9:37 ` David Laight
2021-01-08 9:51 ` Peter Zijlstra
2021-01-08 9:51 ` Peter Zijlstra
2021-01-08 10:44 ` David Laight
2021-01-08 10:44 ` David Laight
2021-01-08 9:49 ` David Laight
2021-01-08 9:49 ` David Laight
2021-01-08 6:13 ` Oliver Sang
2021-01-08 6:13 ` Oliver Sang
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=20210107185559.GI3579531@ZenIV.linux.org.uk \
--to=viro@zeniv.linux.org.uk \
--cc=lkp@lists.01.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.