From mboxrd@z Thu Jan  1 00:00:00 1970
From: annie li <annie.li@oracle.com>
Subject: Re: Is: SKB_MAX_LEN bites again. Was: Re: bug disabling
 guest interface
Date: Wed, 13 Mar 2013 14:44:17 +0800
Message-ID: <51402041.9070106@oracle.com>
References: <5139A56D.30107@crc.id.au>
	<00bf01ce1c34$395a4f20$ac0eed60$@jacekowski.org>
	<20130308203636.GA5422@phenom.dumpdata.com>
	<1362797836.8941.189.camel@hastur.hellion.org.uk>
	<20130310191838.GA13011@zion.uk.xensource.com>
	<1363088434.29599.3.camel@zakaz.uk.xensource.com>
	<1363090704.29093.87.camel@zion.uk.xensource.com>
	<513F4499.9080804@oracle.com>
	<1363101940.29093.95.camel@zion.uk.xensource.com>
	<1363119219.29093.112.camel@zion.uk.xensource.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; Format="flowed"
Content-Transfer-Encoding: 7bit
Return-path: <xen-devel-bounces@lists.xen.org>
In-Reply-To: <1363119219.29093.112.camel@zion.uk.xensource.com>
List-Unsubscribe: <http://lists.xen.org/cgi-bin/mailman/options/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xen.org>
List-Help: <mailto:xen-devel-request@lists.xen.org?subject=help>
List-Subscribe: <http://lists.xen.org/cgi-bin/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=subscribe>
Sender: xen-devel-bounces@lists.xen.org
Errors-To: xen-devel-bounces@lists.xen.org
To: Wei Liu <wei.liu2@citrix.com>
Cc: 'Steven Haigh' <netwiz@crc.id.au>, Ian Campbell <Ian.Campbell@citrix.com>, "Palagummi, Siva" <Siva.Palagummi@ca.com>, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>, "xen-devel@lists.xen.org" <xen-devel@lists.xen.org>, "msw@amazon.com" <msw@amazon.com>, Jacek Milewicz <jacekowski@jacekowski.org>
List-Id: xen-devel@lists.xenproject.org


On 2013-3-13 4:13, Wei Liu wrote:
> Actually the copy is done with Hypervisor, I don't know how it is
> possible to do coalesce while copying.
>
My understanding is we do coalescing in netback during grant copy. For 
example,

                  gop->source.u.ref = txreq.gref;
                  gop->source.domid = vif->domid;
                  gop->source.offset = txreq.offset;

                  gop->dest.u.gmfn = virt_to_mfn(page_address(page));  
--  here should be the page where we coalesce data into, it is likely 
the page is the one which previous request used.
                  gop->dest.domid = DOMID_SELF;
                  gop->dest.offset = txreq.offset;  -- dest.offset 
should be changed to something which represents the offset where we 
start coalescing data of this request

                  gop->len = txreq.size;  -- len should be the total len 
of data which we coalesce into a page
                  gop->flags = GNTCOPY_source_gref;

With this coalescing, it is possible that we could decrease the total 
frags number of skb in netback.

Thanks
Annie