From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from [140.186.70.92] (port=33011 helo=eggs.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1OqQg8-0004iX-Nr
	for qemu-devel@nongnu.org; Tue, 31 Aug 2010 09:15:18 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69)
	(envelope-from <gleb@redhat.com>) id 1OqQfw-0005eB-Od
	for qemu-devel@nongnu.org; Tue, 31 Aug 2010 09:15:04 -0400
Received: from mx1.redhat.com ([209.132.183.28]:58718)
	by eggs.gnu.org with esmtp (Exim 4.69)
	(envelope-from <gleb@redhat.com>) id 1OqQfw-0005dL-IT
	for qemu-devel@nongnu.org; Tue, 31 Aug 2010 09:14:56 -0400
Date: Tue, 31 Aug 2010 16:14:49 +0300
From: Gleb Natapov <gleb@redhat.com>
Subject: Re: [Qemu-devel] Re: [PATCH 5/5] RFC: distinguish warm reset from
	cold reset.
Message-ID: <20100831131449.GG10499@redhat.com>
References: <cover.1283152674.git.yamahata@valinux.co.jp>
	<0a460e01cca4fa24f446c7a715fe6df17d0be9ed.1283152674.git.yamahata@valinux.co.jp>
	<4C7BAB2A.30608@codemonkey.ws>
	<20100831025808.GA19374@valinux.co.jp>
	<4C7CFEC1.8040204@codemonkey.ws>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <4C7CFEC1.8040204@codemonkey.ws>
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.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: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Anthony Liguori <anthony@codemonkey.ws>
Cc: glommer@redhat.com, qemu-devel@nongnu.org, blauwirbel@gmail.com, Isaku Yamahata <yamahata@valinux.co.jp>, alex.williamson@redhat.com, avi@redhat.com

On Tue, Aug 31, 2010 at 08:08:17AM -0500, Anthony Liguori wrote:
> On 08/30/2010 09:58 PM, Isaku Yamahata wrote:
> >>I was thinking that we should stick entirely within the qdev abstraction.
> >>
> >>The patchset I sent out introduced a cold reset as a qdev property on
> >>the devices.
> >>
> >>For warm reset, if I understand correctly, we need two things.  We need
> >>to 1) control propagation order and we need to 2) differentiate
> >>per-device between cold reset and warm reset.
> >>
> >>For (2), I don't know that we truly do need it.  For something like PCI
> >>AER, wouldn't we just move the AER initialization to the qdev init
> >>function and then never change the AER registers during reset?
> >>
> >>IOW, the only way to do a cold reset would be to destroy and recreate
> >>the device.
> >I'm lost here. Then, what should qdev_reset() do?
> 
> I don't know, that's what I'm trying to understand.
> 
> As of this moment, you've convinced me that it should be a warm
> reset.  However, I'm not yet convinced that we need to allow buses
> to change the propagation path of the warm reset.
> 
System_reset should do cold reset like it does now.
 
--
			Gleb.