From mboxrd@z Thu Jan 1 00:00:00 1970 From: Or Gerlitz Subject: Re: [PATCH qemu-kvm] Add raw(af_packet) network backend to qemu Date: Mon, 01 Feb 2010 17:47:40 +0200 Message-ID: <4B66F79C.7040007@voltaire.com> References: <4B5F54E8.3080507@codemonkey.ws> <4B5F5594.6080006@codemonkey.ws> <20100127092451.GC3476@redhat.com> <4B60488F.5020506@codemonkey.ws> <20100127165909.GA13260@redhat.com> <4B6072E1.7030702@codemonkey.ws> <20100127172536.GD13260@redhat.com> <4B60799F.80708@codemonkey.ws> <1264614895.20320.35.camel@w-sridhar.beaverton.ibm.com> <4B607FBA.2070902@codemonkey.ws> <20100127180338.GB13730@redhat.com> <4B6099E0.40101@codemonkey.ws> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: "Michael S. Tsirkin" , Sridhar Samudrala , avi@redhat.com, markmc@redhat.com, kvm@vger.kernel.org, qemu-devel@vger.kernel.org To: Anthony Liguori Return-path: Received: from fwil.voltaire.com ([193.47.165.2]:23620 "EHLO exil.voltaire.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753756Ab0BAPrn (ORCPT ); Mon, 1 Feb 2010 10:47:43 -0500 In-Reply-To: <4B6099E0.40101@codemonkey.ws> Sender: kvm-owner@vger.kernel.org List-ID: Anthony Liguori wrote: > Considering VEPA enabled hardware doesn't exist today and the > standards aren't even finished being defined, I don't think it's a > really strong use case ;-) Anthony, VEPA enabled NIC hardware is live and kicking, maybe even @ your onboard 1Gbs NIC: the intel 82576 (<-- Linux igb network driver) supports SR-IOV && VEPA: 1. register exists which dictates whether the NIC does switching between the different VFs or just send every packet transmitted from the VF to the "uplink" PF 2. a logic exists which makes sure a "downstream" (incoming from the network) packet is never sent to a VF who has the source mac of this packet, which account for multicast support. To learn more about that, see the "Intel 82576 SR-IOV Driver Companion Guide", available on the web. Or.