From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Vrabel Subject: Re: [PATCH v5] xenbus_client: extend interface to suppurt multi-page ring Date: Tue, 19 Nov 2013 17:23:52 +0000 Message-ID: <528B9EA8.4030400@citrix.com> References: <1384104089-15958-1-git-send-email-wei.liu2@citrix.com> <528B9B49.3080902@citrix.com> <20131119171431.GF13187@zion.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from mail6.bemta5.messagelabs.com ([195.245.231.135]) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1Vip1q-0004lk-56 for xen-devel@lists.xenproject.org; Tue, 19 Nov 2013 17:23:58 +0000 In-Reply-To: <20131119171431.GF13187@zion.uk.xensource.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Wei Liu Cc: xen-devel@lists.xenproject.org, Boris Ostrovsky , mattjd@gmail.com List-Id: xen-devel@lists.xenproject.org On 19/11/13 17:14, Wei Liu wrote: > On Tue, Nov 19, 2013 at 05:09:29PM +0000, David Vrabel wrote: >> On 10/11/13 17:21, Wei Liu wrote: >>> Originally Xen PV drivers only use single-page ring to pass along >>> information. This might limit the throughput between frontend and >>> backend. >>> >>> The patch extends Xenbus driver to support multi-page ring, which in >>> general should improve throughput if ring is the bottleneck. Changes to >>> various frontend / backend to adapt to the new interface are also >>> included. >>> >>> Affected Xen drivers: >>> * blkfront/back >>> * netfront/back >>> * pcifront/back >>> * tpmfront >>> >>> The interface is documented, as before, in xenbus_client.c. >> >> It looks ok to me so I guess: >> >> Reviewed-by: David Vrabel >> >> But, I think this should wait until either: >> >> a) There's an in kernel user that uses multipage rings >> > > The main point I picked this up is there seems to be people interested > in this during Xen developer summit. None of the users is in tree > though. > >> or >> >> b) You demonstrate multi page rings have been tested with some sort of >> unit test. >> > > Wait until I have time to develop multi page ring for network driver? I said unit test. A test driver that creates a multipage ring, maps it and verifies that reads/writes to the two different mappings (i.e, 'front' and 'back') work as expected. This shouldn't be a lot of work. We can't take a new feature that hasn't been tested at all. David