From: Peter Zijlstra <peterz@infradead.org>
To: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"minchan.kim@gmail.com" <minchan.kim@gmail.com>,
cl@linux-foundation.org
Subject: Re: [RFC PATCH] asynchronous page fault.
Date: Mon, 28 Dec 2009 09:55:33 +0100 [thread overview]
Message-ID: <1261990533.7135.34.camel@laptop> (raw)
In-Reply-To: <20091228093606.9f2e666c.kamezawa.hiroyu@jp.fujitsu.com>
On Mon, 2009-12-28 at 09:36 +0900, KAMEZAWA Hiroyuki wrote:
> Hmm ? for single-thread apps ? This patch's purpose is not for lockless
> lookup, it's just a part of work. My purpose is avoiding false-sharing.
False sharing in the sense of the mmap_sem cacheline containing other
variables? How could that ever be a problem for a single threaded
application?
For multi-threaded apps the contention on that cacheline is the largest
issue, and moving it to a vma cacheline doesn't seem like a big
improvement.
You want something much finer grained than vmas, there's lots of apps
working on a single (or very few) vma(s). Leaving you with pretty much
the same cacheline contention. Only now its a different cacheline.
WARNING: multiple messages have this Message-ID (diff)
From: Peter Zijlstra <peterz@infradead.org>
To: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"minchan.kim@gmail.com" <minchan.kim@gmail.com>,
cl@linux-foundation.org
Subject: Re: [RFC PATCH] asynchronous page fault.
Date: Mon, 28 Dec 2009 09:55:33 +0100 [thread overview]
Message-ID: <1261990533.7135.34.camel@laptop> (raw)
In-Reply-To: <20091228093606.9f2e666c.kamezawa.hiroyu@jp.fujitsu.com>
On Mon, 2009-12-28 at 09:36 +0900, KAMEZAWA Hiroyuki wrote:
> Hmm ? for single-thread apps ? This patch's purpose is not for lockless
> lookup, it's just a part of work. My purpose is avoiding false-sharing.
False sharing in the sense of the mmap_sem cacheline containing other
variables? How could that ever be a problem for a single threaded
application?
For multi-threaded apps the contention on that cacheline is the largest
issue, and moving it to a vma cacheline doesn't seem like a big
improvement.
You want something much finer grained than vmas, there's lots of apps
working on a single (or very few) vma(s). Leaving you with pretty much
the same cacheline contention. Only now its a different cacheline.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2009-12-28 8:56 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-12-25 1:51 [RFC PATCH] asynchronous page fault KAMEZAWA Hiroyuki
2009-12-27 9:47 ` Minchan Kim
2009-12-27 9:47 ` Minchan Kim
2009-12-27 23:59 ` KAMEZAWA Hiroyuki
2009-12-27 23:59 ` KAMEZAWA Hiroyuki
2009-12-27 11:19 ` Peter Zijlstra
2009-12-27 11:19 ` Peter Zijlstra
2009-12-28 0:00 ` KAMEZAWA Hiroyuki
2009-12-28 0:00 ` KAMEZAWA Hiroyuki
2009-12-28 0:57 ` Balbir Singh
2009-12-28 0:57 ` Balbir Singh
2009-12-28 1:05 ` KAMEZAWA Hiroyuki
2009-12-28 1:05 ` KAMEZAWA Hiroyuki
2009-12-28 2:58 ` Balbir Singh
2009-12-28 2:58 ` Balbir Singh
2009-12-28 3:13 ` KAMEZAWA Hiroyuki
2009-12-28 3:13 ` KAMEZAWA Hiroyuki
2009-12-28 8:34 ` Peter Zijlstra
2009-12-28 8:34 ` Peter Zijlstra
2009-12-28 8:32 ` Peter Zijlstra
2009-12-28 8:32 ` Peter Zijlstra
2009-12-29 9:54 ` Balbir Singh
2009-12-29 9:54 ` Balbir Singh
2009-12-27 12:03 ` Peter Zijlstra
2009-12-27 12:03 ` Peter Zijlstra
2009-12-28 0:36 ` KAMEZAWA Hiroyuki
2009-12-28 0:36 ` KAMEZAWA Hiroyuki
2009-12-28 1:19 ` KAMEZAWA Hiroyuki
2009-12-28 1:19 ` KAMEZAWA Hiroyuki
2009-12-28 8:30 ` Peter Zijlstra
2009-12-28 8:30 ` Peter Zijlstra
2009-12-28 9:58 ` KAMEZAWA Hiroyuki
2009-12-28 9:58 ` KAMEZAWA Hiroyuki
2009-12-28 10:30 ` Peter Zijlstra
2009-12-28 10:30 ` Peter Zijlstra
2009-12-28 10:40 ` Peter Zijlstra
2009-12-28 10:40 ` Peter Zijlstra
2010-01-02 16:14 ` Peter Zijlstra
2010-01-02 16:14 ` Peter Zijlstra
2010-01-04 3:02 ` Paul E. McKenney
2010-01-04 3:02 ` Paul E. McKenney
2010-01-04 7:53 ` Peter Zijlstra
2010-01-04 7:53 ` Peter Zijlstra
2010-01-04 15:55 ` Paul E. McKenney
2010-01-04 15:55 ` Paul E. McKenney
2010-01-04 16:02 ` Peter Zijlstra
2010-01-04 16:02 ` Peter Zijlstra
2010-01-04 16:56 ` Paul E. McKenney
2010-01-04 16:56 ` Paul E. McKenney
2010-01-04 13:48 ` [RFC PATCH -v2] speculative " Peter Zijlstra
2010-01-04 13:48 ` Peter Zijlstra
2009-12-28 10:57 ` [RFC PATCH] asynchronous " KAMEZAWA Hiroyuki
2009-12-28 10:57 ` KAMEZAWA Hiroyuki
2009-12-28 11:06 ` Peter Zijlstra
2009-12-28 11:06 ` Peter Zijlstra
2009-12-28 8:55 ` Peter Zijlstra [this message]
2009-12-28 8:55 ` Peter Zijlstra
2009-12-28 10:08 ` KAMEZAWA Hiroyuki
2009-12-28 10:08 ` KAMEZAWA Hiroyuki
2009-12-28 11:43 ` Peter Zijlstra
2009-12-28 11:43 ` Peter Zijlstra
2010-01-02 21:45 ` Benjamin Herrenschmidt
2010-01-02 21:45 ` Benjamin Herrenschmidt
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=1261990533.7135.34.camel@laptop \
--to=peterz@infradead.org \
--cc=cl@linux-foundation.org \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=minchan.kim@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 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.