From: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Subject: Re: [PATCH] Optimize wait_sb_inodes()
Date: Fri, 28 Jun 2013 09:30:17 +0900
Message-ID: <87hagjqdpi.fsf@devron.myhome.or.jp>
References: <20130626231143.GC28426@dastard>
 <87wqpg76ls.fsf@devron.myhome.or.jp> <20130627044705.GB29790@dastard>
 <87y59w5dye.fsf@devron.myhome.or.jp> <20130627063816.GD29790@dastard>
 <87ppv83tnn.fsf@devron.myhome.or.jp> <20130627094042.GZ29338@dastard>
 <87mwqb3m1e.fsf@devron.myhome.or.jp>
 <20130627183631.GB22832@thunk.org>
 <871u7nrupn.fsf@devron.myhome.or.jp>
 <20130627234521.GC22832@thunk.org>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Cc: linux-fsdevel@vger.kernel.org, tux3@tux3.org,
 Dave Chinner <david@fromorbit.com>, Al Viro <viro@zeniv.linux.org.uk>,
 linux-kernel@vger.kernel.org
To: "Theodore Ts'o" <tytso@mit.edu>
Return-path: <tux3-bounces@phunq.net>
In-Reply-To: <20130627234521.GC22832@thunk.org> (Theodore Ts'o's message of
 "Thu, 27 Jun 2013 19:45:21 -0400")
List-Unsubscribe: <http://phunq.net/mailman/options/tux3>,
 <mailto:tux3-request@phunq.net?subject=unsubscribe>
List-Archive: <http://phunq.net/pipermail/tux3>
List-Post: <mailto:tux3@phunq.net>
List-Help: <mailto:tux3-request@phunq.net?subject=help>
List-Subscribe: <http://phunq.net/mailman/listinfo/tux3>,
 <mailto:tux3-request@phunq.net?subject=subscribe>
Errors-To: tux3-bounces@phunq.net
Sender: tux3-bounces@phunq.net
List-Id: linux-fsdevel.vger.kernel.org

Theodore Ts'o <tytso@mit.edu> writes:

> On Fri, Jun 28, 2013 at 08:37:40AM +0900, OGAWA Hirofumi wrote:
>> 
>> Well, anyway, it is simple. This issue was came as the performance
>> regression when I was experimenting to use kernel bdi flusher from own
>> flusher. The issue was sync(2) like I said.  And this was just I
>> couldn't solve this issue by tux3 side unlike other optimizations.
>
> A performance regression using fsstress?  That's not a program
> intended to be a useful benchmark for measuring performance.

Right. fsstress is used as stress tool for me too as part of CI, with
background vmstat 1. Anyway, it is why I noticed this.

I agree it would not be high priority. But I don't think we should stop
to optimize it.

Thanks.
-- 
OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>