From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tom Herbert Subject: Re: [RFC PATCH net-next 3/3] virtio-net: Add accelerated RFS support Date: Fri, 17 Jan 2014 20:59:15 -0800 Message-ID: References: <1389795654-28381-1-git-send-email-zwu.kernel@gmail.com> <1389795654-28381-4-git-send-email-zwu.kernel@gmail.com> <1389907887.11912.87.camel@bwh-desktop.uk.level5networks.com> <1389914188.11912.146.camel@bwh-desktop.uk.level5networks.com> <1389979208.27141.11.camel@bwh-desktop.uk.level5networks.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Cc: Zhi Yong Wu , Stefan Hajnoczi , Linux Netdev List , Eric Dumazet , "David S. Miller" , Zhi Yong Wu To: Ben Hutchings Return-path: Received: from mail-oa0-f44.google.com ([209.85.219.44]:56924 "EHLO mail-oa0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752608AbaARE7P (ORCPT ); Fri, 17 Jan 2014 23:59:15 -0500 Received: by mail-oa0-f44.google.com with SMTP id g12so5571234oah.3 for ; Fri, 17 Jan 2014 20:59:15 -0800 (PST) In-Reply-To: <1389979208.27141.11.camel@bwh-desktop.uk.level5networks.com> Sender: netdev-owner@vger.kernel.org List-ID: Ben, I've never quite understood why flow management in aRFS has to be done with separate messages, and if I recall this seems to mitigate performance gains to a large extent. It seems like we should be able to piggyback on a TX descriptor for a connection information about the RX side for that connection, namely the rxhash and queue mapping. State creation should be implicit by just seeing a new rxhash value, tear down might be accomplished with a separate flag on the final TX packet on the connection (this would need some additional logic in the stack). Is this method not feasible in either NICs or virtio-net? On Fri, Jan 17, 2014 at 9:20 AM, Ben Hutchings wrote: > On Sat, 2014-01-18 at 00:54 +0800, Zhi Yong Wu wrote: >> On Fri, Jan 17, 2014 at 7:16 AM, Ben Hutchings >> wrote: > [...] >> > However, to take advantage of ARFS on a physical net driver, it would be >> > necessary to send a control request for part 2. >> aRFS on a physical net driver? What is this physical net driver? I >> thought that in order to enable aRFS, guest virtio_net driver should >> send a control request to its emulated virtio_net NIC. > [...] > > If the backend is connected to a macvlan device on top of a physical net > device that supports ARFS, then there is further potential for improving > performance by steering to the best physical RX queue and CPU as well as > the best virtio_net RX queue and vCPU. > > Ben. > > -- > Ben Hutchings, Staff Engineer, Solarflare > Not speaking for my employer; that's the marketing department's job. > They asked us to note that Solarflare product names are trademarked. >