From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:49771)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <famz@redhat.com>) id 1ZFJzt-0002SA-S3
	for qemu-devel@nongnu.org; Wed, 15 Jul 2015 06:33:10 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <famz@redhat.com>) id 1ZFJzt-0004qQ-71
	for qemu-devel@nongnu.org; Wed, 15 Jul 2015 06:33:05 -0400
Date: Wed, 15 Jul 2015 18:32:54 +0800
From: Fam Zheng <famz@redhat.com>
Message-ID: <20150715103254.GE2412@ad.nay.redhat.com>
References: <1436413678-7114-1-git-send-email-famz@redhat.com>
	<1436413678-7114-2-git-send-email-famz@redhat.com>
	<20150709125455.GA11166@stefanha-thinkpad.redhat.com>
	<1233193551.9344767.1436499768115.JavaMail.zimbra@oxygem.tv>
	<20150714123131.GM17927@stefanha-thinkpad.redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20150714123131.GM17927@stefanha-thinkpad.redhat.com>
Subject: Re: [Qemu-devel] [Qemu-block] [PATCH 1/3] blockjob: Introduce
 block_job_relax_cpu
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Stefan Hajnoczi <stefanha@gmail.com>
Cc: Kevin Wolf <kwolf@redhat.com>, qemu-block@nongnu.org, qemu-devel <qemu-devel@nongnu.org>, Alexandre DERUMIER <aderumier@odiso.com>

On Tue, 07/14 13:31, Stefan Hajnoczi wrote:
> On Fri, Jul 10, 2015 at 05:42:48AM +0200, Alexandre DERUMIER wrote:
> > >>By the way, why did you choose 10 milliseconds?  That is quite long.
> > >>
> > >>If this function is called once per 10 ms disk I/O operations then we
> > >>lose 50% utilization.  1 ms or less would be reasonable.
> > 
> > From my tests, 1ms is not enough, It still hanging in guest or qmp queries.
> > 10ms give me optimal balance between bitmap scan speed and guest responsiveness.
> 
> Then I don't fully understand the bug.
> 
> Fam: can you explain why 1ms isn't enough?

In Alexandre's case, I suppose it's because the lseek is so slow that sleeping
for 1ms would still let mirror coroutine to occupy, say, 90% of CPU time, so
guest IO stutters. Perhaps we could move lseek to thread pool in the future.

Anyway, 10ms wasn't a deliberate choice, because I didn't have one. I agree in
other cases, 1ms or less should be enough.

Fam