From mboxrd@z Thu Jan  1 00:00:00 1970
From: Avi Kivity <avi@redhat.com>
Subject: Re: high load with usb device
Date: Tue, 14 Sep 2010 11:06:08 +0200
Message-ID: <4C8F3B00.3000703@redhat.com>
References: <4C80E839.50604@msgid.tls.msk.ru> <4C8A1AE3.8030705@redhat.com> <4C8A2187.1020202@msgid.tls.msk.ru> <4C8A59FE.9090400@msgid.tls.msk.ru> <4C8C9CAA.9030203@redhat.com> <4C8F1B54.60604@msgid.tls.msk.ru>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8;
	format=flowed
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: KVM list <kvm@vger.kernel.org>,
	Jes Sorensen <Jes.Sorensen@redhat.com>,
	linux-perf-users@vger.kernel.org
To: Michael Tokarev <mjt@tls.msk.ru>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:50755 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1752421Ab0INJGP (ORCPT <rfc822;kvm@vger.kernel.org>);
	Tue, 14 Sep 2010 05:06:15 -0400
In-Reply-To: <4C8F1B54.60604@msgid.tls.msk.ru>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

  On 09/14/2010 08:51 AM, Michael Tokarev wrote:
> 12.09.2010 13:26, Avi Kivity =D0=BF=D0=B8=D1=88=D0=B5=D1=82:
>>   On 09/10/2010 07:17 PM, Michael Tokarev wrote:
>>> Note the changed subject line.
>>>
>>> I just did a few tests with linux guest (amd64 2.6.35 kernel).
>>>
>>> And it shows the same behavour as win7 (unlike winXP), namely,
>>> high host CPU load when guest is idle.
>> Not for me - F12 idle guest takes 3.5% cpu.
>>
>> What does 'top' in the guest show now?  Connect with ssh to avoid
>> triggering the GUI.
> Guest:
> Cpu(s):  0.0%us,  0.0%sy,  0.0%ni,100.0%id,  0.0%wa,  0.0%hi,  0.0%si=
,  0.0%st
>
> Host:
> Cpu(s):  5.3%us,  6.7%sy,  0.0%ni, 88.0%id,  0.0%wa,  0.0%hi,  0.0%si=
,  0.0%st
>    PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
> 25685 mjt       20   0 1074m 109m 4652 S   19  1.9   0:29.03 qemu-sys=
tem-x86
>
> So it appears it's the qemu process that's doing something,
> after all.  Is there a way to compile it with profiling
> information?  I tried, but it does not compile: it uses
> -fomit-frame-pointer which is not compatible with -p[g],
> but when removing -fo-f-p it complains about assembly.
>
>

Really, 'perf top' should have shown what qemu was doing.  I don't have=
=20
a lot of experience with profiling userspace, copying linux-perf-users=20
for tips.

--=20
error compiling committee.c: too many arguments to function