From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1Mox4x-00061o-T7
	for qemu-devel@nongnu.org; Sat, 19 Sep 2009 06:22:07 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1Mox4s-00060x-1O
	for qemu-devel@nongnu.org; Sat, 19 Sep 2009 06:22:06 -0400
Received: from [199.232.76.173] (port=55965 helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1Mox4r-00060u-RD
	for qemu-devel@nongnu.org; Sat, 19 Sep 2009 06:22:01 -0400
Received: from mail-bw0-f211.google.com ([209.85.218.211]:43162)
	by monty-python.gnu.org with esmtp (Exim 4.60)
	(envelope-from <blauwirbel@gmail.com>) id 1Mox4r-0001ZW-Fp
	for qemu-devel@nongnu.org; Sat, 19 Sep 2009 06:22:01 -0400
Received: by bwz7 with SMTP id 7so1104690bwz.34
	for <qemu-devel@nongnu.org>; Sat, 19 Sep 2009 03:22:00 -0700 (PDT)
MIME-Version: 1.0
In-Reply-To: <20090919042617.GA24944@laped.iglesias.mooo.com>
References: <20090919042617.GA24944@laped.iglesias.mooo.com>
From: Blue Swirl <blauwirbel@gmail.com>
Date: Sat, 19 Sep 2009 13:21:19 +0300
Message-ID: <f43fc5580909190321i3b0f69a5v902b57294244a576@mail.gmail.com>
Subject: Re: [Qemu-devel] device-to-device
Content-Type: text/plain; charset=UTF-8
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: "Edgar E. Iglesias" <edgar.iglesias@gmail.com>
Cc: qemu-devel@nongnu.org

On Sat, Sep 19, 2009 at 7:26 AM, Edgar E. Iglesias
<edgar.iglesias@gmail.com> wrote:
> Hi,
>
> Has anybody done some work on a qdev device-2-device infrastruture?
> For example to model DMA back channels or other device to device
> communication?

We should have a DMA object, with map/unmap methods to translate the
addresses to host pointers. Physical memory should implement the
object as well as any IOMMUs or other address translators like buses
(in some cases).

The interface to qdev could be much like chardev.