From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1MqULC-0007PB-O6
	for qemu-devel@nongnu.org; Wed, 23 Sep 2009 12:05:14 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1MqUL7-0007Iy-HM
	for qemu-devel@nongnu.org; Wed, 23 Sep 2009 12:05:14 -0400
Received: from [199.232.76.173] (port=56153 helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1MqUL7-0007Ia-CJ
	for qemu-devel@nongnu.org; Wed, 23 Sep 2009 12:05:09 -0400
Received: from mail.corp.accelance.fr ([213.162.48.15]:58400)
	by monty-python.gnu.org with esmtps
	(TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.60)
	(envelope-from <bcm@accelance.fr>) id 1MqUL6-0006zu-J4
	for qemu-devel@nongnu.org; Wed, 23 Sep 2009 12:05:08 -0400
Received: from [213.162.49.203] (bcm.noc.accelance.net [213.162.49.203])
	(using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits))
	(No client certificate requested)
	by mail.corp.accelance.fr (Postfix) with ESMTP id B11CF1AFA0
	for <qemu-devel@nongnu.org>; Wed, 23 Sep 2009 18:05:05 +0200 (CEST)
Subject: Re: [Qemu-devel] Qemu savevm and CPU soft lockup
From: Benjamin Cleyet-Marrel <bcm@accelance.fr>
In-Reply-To: <1253267801.9686.159.camel@bcm-portable>
References: <1253267801.9686.159.camel@bcm-portable>
Content-Type: multipart/alternative; boundary="=-7jUZDBVg3OpmUV8obj1l"
Date: Wed, 23 Sep 2009 18:05:04 +0200
Message-Id: <1253721904.16114.101.camel@bcm-portable>
Mime-Version: 1.0
List-Id: qemu-devel.nongnu.org
List-Unsubscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: qemu-devel@nongnu.org


--=-7jUZDBVg3OpmUV8obj1l
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi,=20

After further investigation, I figured out that when issuing a savevm
command
the entire qemu process gets stuck on IO wait.
I can't issue any other commands on the monitor the process is shown as
D and the guest is in softlockup state.

Looking at the way migrate (and the -d for detach I presume) is working
I suppose the same behaviour would be expected from the savevm function.

A savevm -d so that the qemu process would not be freezed while saving
the data.

Sorry if I am just talking non sense but my snapshot on iscsi storage
takes about 1 minutes.=20
which means that my guest are down for a minute or so which is not
ideal.

Thanks for your time
Cheers
Ben


-------- Message initial --------
De: Benjamin Cleyet-Marrel <bcm@accelance.fr>
=C3=80: qemu-devel@nongnu.org
Sujet: [Qemu-devel] Qemu savevm and CPU soft lockup
Date: Fri, 18 Sep 2009 11:56:41 +0200

Hi,=20

First post to this list so sorry if I am mistaken.
I am using qemu-kvm-0.11.0-rc2 I've googled quite a lot and could not
find anyone having or reporting the same problem.
The guest is a Centos5.3

The savevm/loadvm seems to be broken:

When issuing a savevm in the monitor,=20
Guest get stuck for a few seconds and I get the infamous soft lockup a
couple of time. (see bellow)

In the end, the snapshot is working.

When doing a loadvm I often get weird behaviour on the console.
this can be completly removed if I issue a stop and cont before and
after the loadvm.

Question is:
Am I doing something wrong or is this a bug ?

Thanks
Cheers
Ben



Sep 18 10:58:48 dhcp155 kernel: BUG: soft lockup - CPU#0 stuck for 10s!
[swapper:0]
Sep 18 10:58:48 dhcp155 kernel:=20
Sep 18 10:58:48 dhcp155 kernel: Pid: 0, comm:              swapper
Sep 18 10:58:48 dhcp155 kernel: EIP: 0060:[<c044d219>] CPU: 0
Sep 18 10:58:48 dhcp155 kernel: EIP is at handle_IRQ_event+0x39/0x8c
Sep 18 10:58:48 dhcp155 kernel:  EFLAGS: 00000246    Not tainted
(2.6.18-128.el5 #1)
Sep 18 10:58:48 dhcp155 kernel: EAX: 00000001 EBX: c06e6f00 ECX:
f7c9bf80 EDX: c0754f9c
Sep 18 10:58:48 dhcp155 kernel: ESI: f7c9bf80 EDI: 00000001 EBP:
00000000 DS: 007b ES: 007b
Sep 18 10:58:48 dhcp155 kernel: CR0: 8005003b CR2: 080c3e80 CR3:
376ad000 CR4: 00000690
Sep 18 10:58:48 dhcp155 kernel:  [<c044d2f0>] __do_IRQ+0x84/0xd6
Sep 18 10:58:48 dhcp155 kernel:  [<c04074e5>] do_IRQ+0xb0/0xc3
Sep 18 10:58:48 dhcp155 kernel:  [<c0405946>] common_interrupt+0x1a/0x20
Sep 18 10:58:48 dhcp155 kernel:  [<c044d219>] handle_IRQ_event+0x39/0x8c
Sep 18 10:58:48 dhcp155 kernel:  [<c044d2f0>] __do_IRQ+0x84/0xd6
Sep 18 10:58:48 dhcp155 kernel:  [<c044d26c>] __do_IRQ+0x0/0xd6
Sep 18 10:58:48 dhcp155 kernel:  [<c04074ce>] do_IRQ+0x99/0xc3
Sep 18 10:58:48 dhcp155 kernel:  [<c0405946>] common_interrupt+0x1a/0x20
Sep 18 10:58:48 dhcp155 kernel:  [<c0428ba7>] __do_softirq+0x57/0x114
Sep 18 10:58:48 dhcp155 kernel:  [<c04073eb>] do_softirq+0x52/0x9c
Sep 18 10:58:48 dhcp155 kernel:  [<c04059d7>] apic_timer_interrupt
+0x1f/0x24
Sep 18 10:58:48 dhcp155 kernel:  [<c0403bb0>] default_idle+0x0/0x59
Sep 18 10:58:48 dhcp155 kernel:  [<c0403be1>] default_idle+0x31/0x59
Sep 18 10:58:48 dhcp155 kernel:  [<c0403ca8>] cpu_idle+0x9f/0xb9
Sep 18 10:58:48 dhcp155 kernel:  [<c06f59ee>] start_kernel+0x379/0x380
Sep 18 10:58:48 dhcp155 kernel:  =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D


in dmesg




--=-7jUZDBVg3OpmUV8obj1l
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 7bit

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN">
<HTML>
<HEAD>
  <META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8">
  <META NAME="GENERATOR" CONTENT="GtkHTML/3.26.0">
</HEAD>
<BODY>
Hi, <BR>
<BR>
After further investigation, I figured out that when issuing a savevm command<BR>
the entire qemu process gets stuck on IO wait.<BR>
I can't issue any other commands on the monitor the process is shown as D and the guest is in softlockup state.<BR>
<BR>
Looking at the way migrate (and the -d for detach I presume) is working I suppose the same behaviour would be expected from the savevm function.<BR>
<BR>
A savevm -d so that the qemu process would not be freezed while saving the data.<BR>
<BR>
Sorry if I am just talking non sense but my snapshot on iscsi storage takes about 1 minutes. <BR>
which means that my guest are down for a minute or so which is not ideal.<BR>
<BR>
Thanks for your time<BR>
Cheers<BR>
Ben<BR>
<BR>
<BR>
-------- Message initial --------<BR>
<B>De</B>: Benjamin Cleyet-Marrel &lt;<A HREF="mailto:Benjamin%20Cleyet-Marrel%20%3cbcm@accelance.fr%3e">bcm@accelance.fr</A>&gt;<BR>
<B>&#192;</B>: <A HREF="mailto:qemu-devel@nongnu.org">qemu-devel@nongnu.org</A><BR>
<B>Sujet</B>: [Qemu-devel] Qemu savevm and CPU soft lockup<BR>
<B>Date</B>: Fri, 18 Sep 2009 11:56:41 +0200<BR>
<BR>
Hi, <BR>
<BR>
First post to this list so sorry if I am mistaken.<BR>
I am using qemu-kvm-0.11.0-rc2 I've googled quite a lot and could not find anyone having or reporting the same problem.<BR>
The guest is a Centos5.3<BR>
<BR>
The savevm/loadvm seems to be broken:<BR>
<BR>
When issuing a savevm in the monitor, <BR>
Guest get stuck for a few seconds and I get the infamous soft lockup a couple of time. (see bellow)<BR>
<BR>
In the end, the snapshot is working.<BR>
<BR>
When doing a loadvm I often get weird behaviour on the console.<BR>
this can be completly removed if I issue a stop and cont before and after the loadvm.<BR>
<BR>
Question is:<BR>
Am I doing something wrong or is this a bug ?<BR>
<BR>
Thanks<BR>
Cheers<BR>
Ben<BR>
<BR>
<BR>
<BR>
Sep 18 10:58:48 dhcp155 kernel: BUG: soft lockup - CPU#0 stuck for 10s! [swapper:0]<BR>
Sep 18 10:58:48 dhcp155 kernel: <BR>
Sep 18 10:58:48 dhcp155 kernel: Pid: 0, comm:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; swapper<BR>
Sep 18 10:58:48 dhcp155 kernel: EIP: 0060:[&lt;c044d219&gt;] CPU: 0<BR>
Sep 18 10:58:48 dhcp155 kernel: EIP is at handle_IRQ_event+0x39/0x8c<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; EFLAGS: 00000246&nbsp;&nbsp;&nbsp; Not tainted&nbsp; (2.6.18-128.el5 #1)<BR>
Sep 18 10:58:48 dhcp155 kernel: EAX: 00000001 EBX: c06e6f00 ECX: f7c9bf80 EDX: c0754f9c<BR>
Sep 18 10:58:48 dhcp155 kernel: ESI: f7c9bf80 EDI: 00000001 EBP: 00000000 DS: 007b ES: 007b<BR>
Sep 18 10:58:48 dhcp155 kernel: CR0: 8005003b CR2: 080c3e80 CR3: 376ad000 CR4: 00000690<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c044d2f0&gt;] __do_IRQ+0x84/0xd6<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c04074e5&gt;] do_IRQ+0xb0/0xc3<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c0405946&gt;] common_interrupt+0x1a/0x20<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c044d219&gt;] handle_IRQ_event+0x39/0x8c<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c044d2f0&gt;] __do_IRQ+0x84/0xd6<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c044d26c&gt;] __do_IRQ+0x0/0xd6<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c04074ce&gt;] do_IRQ+0x99/0xc3<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c0405946&gt;] common_interrupt+0x1a/0x20<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c0428ba7&gt;] __do_softirq+0x57/0x114<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c04073eb&gt;] do_softirq+0x52/0x9c<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c04059d7&gt;] apic_timer_interrupt+0x1f/0x24<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c0403bb0&gt;] default_idle+0x0/0x59<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c0403be1&gt;] default_idle+0x31/0x59<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c0403ca8&gt;] cpu_idle+0x9f/0xb9<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; [&lt;c06f59ee&gt;] start_kernel+0x379/0x380<BR>
Sep 18 10:58:48 dhcp155 kernel:&nbsp; =======================<BR>
<BR>
<BR>
in dmesg<BR>
<BR>
<BR>
<BR>
</BODY>
</HTML>

--=-7jUZDBVg3OpmUV8obj1l--