From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1MHAiJ-0003Zx-Ae
	for qemu-devel@nongnu.org; Thu, 18 Jun 2009 02:03:07 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1MHAiD-0003YQ-Aq
	for qemu-devel@nongnu.org; Thu, 18 Jun 2009 02:03:05 -0400
Received: from [199.232.76.173] (port=41574 helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1MHAiC-0003YN-QN
	for qemu-devel@nongnu.org; Thu, 18 Jun 2009 02:03:00 -0400
Received: from mx2.redhat.com ([66.187.237.31]:44053)
	by monty-python.gnu.org with esmtp (Exim 4.60)
	(envelope-from <gleb@redhat.com>) id 1MHAiC-0006qR-26
	for qemu-devel@nongnu.org; Thu, 18 Jun 2009 02:03:00 -0400
Date: Thu, 18 Jun 2009 09:02:56 +0300
From: Gleb Natapov <gleb@redhat.com>
Subject: Re: [Qemu-devel] [PATCH 2/3] Call piix3_reset() on system reset.
Message-ID: <20090618060255.GA20289@redhat.com>
References: <1245256322-30990-1-git-send-email-gleb@redhat.com>
	<1245256322-30990-2-git-send-email-gleb@redhat.com>
	<f43fc5580906171010j28966a08k4148a8a4ff6461be@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <f43fc5580906171010j28966a08k4148a8a4ff6461be@mail.gmail.com>
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: Blue Swirl <blauwirbel@gmail.com>
Cc: qemu-devel@nongnu.org

On Wed, Jun 17, 2009 at 08:10:23PM +0300, Blue Swirl wrote:
> On 6/17/09, Gleb Natapov <gleb@redhat.com> wrote:
> > Also zero pci_irq_levels on reset to avoid stuck irq after reset.
> >
> >  Signed-off-by: Gleb Natapov <gleb@redhat.com>
> >  Signed-off-by: Yaniv Kamay <ykamay@redhat.com>
> 
> Thanks, applied.
Thanks, but without 1/3 reset is still buggy. Look at hw/pci.c:pci_set_irq()
If pci_dev->irq_state[irq_num] == 1 at reset time it will stay 1 after
reset too. When device will try to race IRQ line after reset it will
fail to do so. And if bus->irq_count[] is not zeroed you have another
set of problems. If you don't like that pci_bus_reset() reset per device
state (why?) it is possible to create pci_device_reset(), but then we
will have to fix 25 devices to call it.  Complication for no good
reason. IMHO pci_device_save/load should be handled in hw/pci.c too.
What are the disadvantages?
 
--
			Gleb.