From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marcelo Tosatti <mtosatti@redhat.com>
Subject: Re: [Qemu-devel] [PATCH] kvmclock: Ensure time in migration never
 goes backward
Date: Wed, 7 May 2014 20:23:27 -0300
Message-ID: <20140507232327.GB25225@amt.cnet>
References: <1399297882-3444-1-git-send-email-agraf@suse.de>
 <20140505232343.GA20638@amt.cnet>
 <20140505233120.GA23113@amt.cnet>
 <53688CC3.7040001@suse.de>
 <53693DFB.3080108@beyond.pl>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: Alexander Graf <agraf@suse.de>, qemu-devel@nongnu.org,
	kvm@vger.kernel.org
To: Marcin =?utf-8?Q?Gibu=C5=82a?= <m.gibula@beyond.pl>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:61090 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751881AbaEGXXq (ORCPT <rfc822;kvm@vger.kernel.org>);
	Wed, 7 May 2014 19:23:46 -0400
Content-Disposition: inline
In-Reply-To: <53693DFB.3080108@beyond.pl>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Tue, May 06, 2014 at 09:54:35PM +0200, Marcin Gibu=C5=82a wrote:
> >Yes, and it isn't. Any ideas why it's not? This patch really just us=
es
> >the guest visible kvmclock time rather than the host view of it on
> >migration.
> >
> >There is definitely something very broken on the host's side since i=
t
> >does return a smaller time than the guest exposed interface indicate=
s.
>=20
> Don't know if helps but here are example values from
> time_at_migration and s->clock from your patch.
>=20
> Tested on 5 restores of saved VM that (used to) hang:
>=20
>    s->clock      time_at_migration
> 157082235125698  157113284546655
> 157082235125698  157113298196976
> 157082235125698  157113284615117
> 157082235125698  157113284486601
> 157082235125698  157113284479740
>=20
> Now, when I compare system time on guest with and without patch:
>=20
> On unpatched qemu vm restores with date: Apr 18 06:56:36
> On patched qemu it says: Apr 18 06:57:06
>=20
> --=20
> mg

Marcin,

Can you turn off the TSC clock on the host and attempt to reproduce the=
=20
issue, without Alexander's patch?=20

Just=20

echo hpet > /sys/devices/system/clocksource/clocksource0/current_clocks=
ource

(or any other clock except tsc if you don't have hpet).
TIA.