From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from [140.186.70.92] (port=54465 helo=eggs.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1OIzXz-0000SP-Qk
	for qemu-devel@nongnu.org; Mon, 31 May 2010 03:36:33 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69)
	(envelope-from <Jes.Sorensen@redhat.com>) id 1OIzXw-00016S-Cr
	for qemu-devel@nongnu.org; Mon, 31 May 2010 03:36:30 -0400
Received: from mx1.redhat.com ([209.132.183.28]:53109)
	by eggs.gnu.org with esmtp (Exim 4.69)
	(envelope-from <Jes.Sorensen@redhat.com>) id 1OIzXw-000165-6X
	for qemu-devel@nongnu.org; Mon, 31 May 2010 03:36:28 -0400
Message-ID: <4C0366E0.6040203@redhat.com>
Date: Mon, 31 May 2010 09:36:00 +0200
From: Jes Sorensen <Jes.Sorensen@redhat.com>
MIME-Version: 1.0
References: <cover.1274903328.git.mst@redhat.com> <4BFF9366.5090103@redhat.com>
	<20100530112220.GA27611@redhat.com>
In-Reply-To: <20100530112220.GA27611@redhat.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: [Qemu-devel] Re: [PATCHv2-RFC 0/2] virtio: put last seen used index
 into ring itself
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: "Michael S. Tsirkin" <mst@redhat.com>
Cc: qemu-devel@nongnu.org, Rusty Russell <rusty@rustcorp.com.au>, linux-kernel@vger.kernel.org, kvm@vger.kernel.org, virtualization@lists.linux-foundation.org

On 05/30/10 13:22, Michael S. Tsirkin wrote:
> On Fri, May 28, 2010 at 11:56:54AM +0200, Jes Sorensen wrote:
>> It looks pretty good to me, however one thing I have been thinking of
>> while reading through it:
>>
>> Rather than storing a pointer within the ring struct, pointing into a
>> position within the same struct. How about storing a byte offset instead
>> and using a cast to get to the pointer position? That would avoid the
>> pointer dereference, which is less effective cache wise and harder for
>> the CPU to predict.
>>
>> Not sure whether it really matters performance wise, just a thought.
> 
> I think this won't work: when PUBLUSH_USED_IDX is negotiated,
> the pointer is to within the ring.

Hmmm shame, it would be a nice optimization.

Maybe it's time to introduce the v2 ring format, rather than having
adding more kludges to the existing one?

Cheers,
Jes