From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1NXCjO-0006Qm-3a
	for qemu-devel@nongnu.org; Tue, 19 Jan 2010 06:58:46 -0500
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1NXCjJ-0006QK-IZ
	for qemu-devel@nongnu.org; Tue, 19 Jan 2010 06:58:45 -0500
Received: from [199.232.76.173] (port=33400 helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1NXCjJ-0006QH-CE
	for qemu-devel@nongnu.org; Tue, 19 Jan 2010 06:58:41 -0500
Received: from mx1.redhat.com ([209.132.183.28]:52490)
	by monty-python.gnu.org with esmtp (Exim 4.60)
	(envelope-from <kwolf@redhat.com>) id 1NXCjI-0008M1-Uw
	for qemu-devel@nongnu.org; Tue, 19 Jan 2010 06:58:41 -0500
Message-ID: <4B559E2F.9000309@redhat.com>
Date: Tue, 19 Jan 2010 12:57:35 +0100
From: Kevin Wolf <kwolf@redhat.com>
MIME-Version: 1.0
Subject: Re: [Qemu-devel] [PATCH 04/10] qcow2: Return 0/-errno in
	qcow2_alloc_cluster_offset
References: <1263816696-24122-1-git-send-email-kwolf@redhat.com>
	<1263816696-24122-5-git-send-email-kwolf@redhat.com>
	<20100119113547.GA14641@lst.de>
In-Reply-To: <20100119113547.GA14641@lst.de>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
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: Christoph Hellwig <hch@lst.de>
Cc: qemu-devel@nongnu.org

Am 19.01.2010 12:35, schrieb Christoph Hellwig:
>> @@ -715,6 +721,7 @@ uint64_t qcow2_alloc_cluster_offset(BlockDriverState *bs,
>>  
>>          cluster_offset &= ~QCOW_OFLAG_COPIED;
>>          m->nb_clusters = 0;
>> +        m->depends_on = NULL;
> 
> What does this have to do with the rest?

It's needed to be able to distinguish between the case where the
clusters are already allocated (0/NULL) and the case where the request
depends on another one (0/non-NULL). This check previously used the
return value (cluster_offset for success, 0 for failure) and I didn't
want to overload m->cluster_offset with such a meaning. This is the
change in the caller:

   /* Need to wait for another request? If so, we are done for now. */
-  if (!acb->cluster_offset && acb->l2meta.depends_on != NULL) {
+  if (acb->l2meta.nb_clusters == 0 && acb->l2meta.depends_on != NULL) {

The alternative would have been to keep using the return value and
hijack some errno value. This would possibly conflict with real
read/write errors though, so I decided to leave the return value alone.

Kevin