From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:54673)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <mtosatti@redhat.com>) id 1XFrwE-0006MR-QX
	for qemu-devel@nongnu.org; Fri, 08 Aug 2014 17:43:08 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <mtosatti@redhat.com>) id 1XFrw8-0004sp-MN
	for qemu-devel@nongnu.org; Fri, 08 Aug 2014 17:43:02 -0400
Received: from mx1.redhat.com ([209.132.183.28]:56207)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <mtosatti@redhat.com>) id 1XFrw8-0004sR-Ex
	for qemu-devel@nongnu.org; Fri, 08 Aug 2014 17:42:56 -0400
Date: Fri, 8 Aug 2014 18:42:26 -0300
From: Marcelo Tosatti <mtosatti@redhat.com>
Message-ID: <20140808214225.GB13545@amt.cnet>
References: <1403734339-14405-1-git-send-email-afaerber@suse.de>
	<1403734339-14405-18-git-send-email-afaerber@suse.de>
	<047A6D07C7C75B4D95D277304F90FDF22198A240@SZXEML506-MBS.china.huawei.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <047A6D07C7C75B4D95D277304F90FDF22198A240@SZXEML506-MBS.china.huawei.com>
Subject: Re: [Qemu-devel] [PULL 17/19] target-i386: Support "invariant tsc"
 flag
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: "chenliang (T)" <chenliang88@huawei.com>
Cc: "Gonglei (Arei)" <arei.gonglei@huawei.com>, "Huangweidong (C)" <weidong.huang@huawei.com>, Andreas =?iso-8859-1?Q?F=E4rber?= <afaerber@suse.de>, Eduardo Habkost <ehabkost@redhat.com>, "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>


Chen Liang,

On Fri, Aug 08, 2014 at 07:07:59AM +0000, chenliang (T) wrote:
> Hi,
> We found that value of host cpus tsc are same to echo other when host has invariant tsc flag.

What you mean exactly?

Do you have a test program?

> But it is not in vm. The problem is that offset of vcpus tsc are not equal. The offset is set when
> vcpu is created. Is it ok to set same offset? And is it reasonable?
> 
> Best regards
> Chen Liang


The code at kvm_write_tsc(), in the kernel (arch/x86/kvm/x86.c), should 
adjust the vcpu tsc offsets properly resulting in cross SMP TSCs being
equal in the VM.