From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: TODO list for qemu+KVM networking performance v2 Date: Wed, 10 Jun 2009 18:18:01 +0300 Message-ID: <4A2FCEA9.8010604@redhat.com> References: <20090604164320.GB14592@redhat.com> <200906101309.14532.rusty@rustcorp.com.au> <4A2F5217.9090401@redhat.com> <200906110009.34671.rusty@rustcorp.com.au> <20090610145358.GB28601@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Rusty Russell , dlaor@redhat.com, virtualization@lists.linux-foundation.org, Chris Wright , Mark McLoughlin , kvm@vger.kernel.org, Brian Stein , Herbert Xu , Dor Laor , Yaron Haviv , Shahar Klein , Anthony Liguori To: "Michael S. Tsirkin" Return-path: Received: from mx2.redhat.com ([66.187.237.31]:51674 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751968AbZFJPSH (ORCPT ); Wed, 10 Jun 2009 11:18:07 -0400 In-Reply-To: <20090610145358.GB28601@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: Michael S. Tsirkin wrote: > But I don't understand how aio will make implementing it easier - > or are you merely saying that it will make it worthwhile? > If you have aio, the the NIC and the guest proceed in parallel. If the guest is faster (likely), then when it sends the next packet it will see that interrupts are disabled and not notify again. Once aio complete we can recheck the queue; if it's empty we reenable notifications. If there's still stuff in it we submit it with notifications disabled. -- error compiling committee.c: too many arguments to function