From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1G4kuz-0001WV-0G
	for qemu-devel@nongnu.org; Sun, 23 Jul 2006 16:51:17 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1G4kuw-0001S5-KT
	for qemu-devel@nongnu.org; Sun, 23 Jul 2006 16:51:16 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1G4kuw-0001Rj-E3
	for qemu-devel@nongnu.org; Sun, 23 Jul 2006 16:51:14 -0400
Received: from [70.116.9.243] (helo=localhost.localdomain)
	by monty-python.gnu.org with esmtps (TLS-1.0:RSA_AES_256_CBC_SHA:32)
	(Exim 4.52) id 1G4kvs-00055T-EB
	for qemu-devel@nongnu.org; Sun, 23 Jul 2006 16:52:12 -0400
Received: from localhost ([127.0.0.1])
	by localhost.localdomain with esmtp (Exim 4.60)
	(envelope-from <anthony@codemonkey.ws>) id 1G4kuq-0000oS-S4
	for qemu-devel@nongnu.org; Sun, 23 Jul 2006 15:51:08 -0500
Message-ID: <44C3E13A.1070509@codemonkey.ws>
Date: Sun, 23 Jul 2006 15:51:06 -0500
From: Anthony Liguori <anthony@codemonkey.ws>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Subject: [Qemu-devel] Feature request for qcow v2
Reply-To: qemu-devel@nongnu.org
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: qemu-devel@nongnu.org

Hi,

I've been playing with some code to implement an HTTP block device in 
QEMU.  I'll post when I've got something working.  As I am building this 
though, I thought of something that would be nice to have for qcow v2.

What I'd like is essentially a copy-on-read flag.  If this flag is set, 
and there is a base image, then whenever a read occurs to an unallocated 
sector, the sector is copied from the base (similar to a write) and the 
sector is allocated.

The idea is that copy-on-read is useful when read is really slow (for 
instance, with an HTTP block device).  This way, you could essentially 
always use qcow + HTTP so that your block device was locally cached.  It 
seems a bit more elegant than trying to implement a caching mechanism in 
the HTTP block device.

Regards,

Anthony Liguori