From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [RFCv2 00/12] Introduce host-side virtio queue and CAIF Virtio. Date: Mon, 14 Jan 2013 18:54:08 +0200 Message-ID: <20130114165407.GA19207@redhat.com> References: <1354718230-4486-1-git-send-email-sjur@brendeland.net> <20121206102750.GF10837@redhat.com> <877goc0wac.fsf@rustcorp.com.au> <87pq1f2rj0.fsf@rustcorp.com.au> <87wqvl1g9s.fsf@rustcorp.com.au> <20130110111117.GE13451@redhat.com> <87sj681wou.fsf@rustcorp.com.au> <20130111073155.GA13315@redhat.com> <874nin1cc1.fsf@rustcorp.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <874nin1cc1.fsf@rustcorp.com.au> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: virtualization-bounces@lists.linux-foundation.org Errors-To: virtualization-bounces@lists.linux-foundation.org To: Rusty Russell Cc: netdev@vger.kernel.org, Linus Walleij , virtualization@lists.linux-foundation.org List-Id: virtualization@lists.linuxfoundation.org On Sat, Jan 12, 2013 at 10:50:30AM +1030, Rusty Russell wrote: > "Michael S. Tsirkin" writes: > > On Fri, Jan 11, 2013 at 09:18:33AM +1030, Rusty Russell wrote: > >> "Michael S. Tsirkin" writes: > >> > On Thu, Jan 10, 2013 at 09:00:55PM +1030, Rusty Russell wrote: > >> >> Not sure why vhost/net doesn't built a packet and feed it in > >> >> netif_rx_ni(). This is what tun seems to do, and with this code it > >> >> should be fairly optimal. > >> > > >> > Because we want to use NAPI. > >> > >> Not quite what I was asking; it was more a question of why we're using a > >> raw socket, when we trivially have a complete skb already which we > >> should be able to feed to Linux like any network packet. > > > > Oh for some reason I thought you were talking about virtio. > > I don't really understand what you are saying here - vhost > > actually calls out to tun to build and submit the skb. > > Ah, the fd is tun? It can be tun or macvtap. We also support a packet socket backend though I don't know of any users, maybe this can be dropped. > Seems a bit indirect; I wonder if there's room for > more optimization here... > > Cheers, > Rusty. Quite possibly. Using common data structures and code in tun and macvtap would allow calling this code directly from vhost-net.