From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ulrich Obergfell <uobergfe@redhat.com>
Subject: Re: [PATCH 1/3] alleviate time drift with HPET periodic timers
Date: Tue, 22 Mar 2011 05:40:21 -0400 (EDT)
Message-ID: <1170032862.455751.1300786821647.JavaMail.root@zmail07.collab.prod.int.phx2.redhat.com>
References: <4D847B1D.3000002@web.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Cc: qemu-devel@nongnu.org, kvm <kvm@vger.kernel.org>,
	gcosta@redhat.com, Anthony Liguori <anthony@codemonkey.ws>,
	aliguori@us.ibm.com
To: Jan Kiszka <jan.kiszka@web.de>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx4-phx2.redhat.com ([209.132.183.25]:40894 "EHLO
	mx4-phx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1752990Ab1CVJk1 (ORCPT <rfc822;kvm@vger.kernel.org>);
	Tue, 22 Mar 2011 05:40:27 -0400
In-Reply-To: <4D847B1D.3000002@web.de>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>


>> Part 1 of the patch implements the following QEMU command line option.
>>
>> -hpet [device=none|present][,driftfix=none|slew]
> 
> Just define driftfix as property of the hpet device. That way it can be
> controlled both globally (-global hpet.driftfix=...) and per hpet block
> (once we support instantiating >1 of them).


Many Thanks Jan,

I started investigating code changes. I'm thinking of ...

- adding a new field to the HPETState structure.

    uint32_t driftfix;

- adding the property 'driftfix' to the DeviceInfo structure.

    DEFINE_PROP_BIT("driftfix", HPETState, driftfix, 0, false)

  Using a single bit so that the option syntax would be, e.g.:

    -global hpet.driftfix=on     (Default is 'off')

- Replace all 'if (hpet_driftfix ...)' by:

    if ((HPETState)s->driftfix ...)


Regards,

Uli