From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: Network performance with small packets Date: Tue, 1 Feb 2011 23:42:11 +0200 Message-ID: <20110201214211.GB31105@redhat.com> References: <1296157547.1640.45.camel@localhost.localdomain> <20110127200548.GE5228@redhat.com> <20110127.130240.104065182.davem@davemloft.net> <1296163838.1640.53.camel@localhost.localdomain> <20110128121616.GA8374@redhat.com> <1296523838.30191.39.camel@sridhar.beaverton.ibm.com> <1296594585.26937.817.camel@localhost.localdomain> <20110201212411.GD30770@redhat.com> <1296595955.26937.822.camel@localhost.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Sridhar Samudrala , Steve Dobbelstein , David Miller , kvm@vger.kernel.org, mashirle@linux.vnet.ibm.com, netdev@vger.kernel.org To: Shirley Ma Return-path: Received: from mx1.redhat.com ([209.132.183.28]:57055 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751812Ab1BAVm2 (ORCPT ); Tue, 1 Feb 2011 16:42:28 -0500 Content-Disposition: inline In-Reply-To: <1296595955.26937.822.camel@localhost.localdomain> Sender: netdev-owner@vger.kernel.org List-ID: On Tue, Feb 01, 2011 at 01:32:35PM -0800, Shirley Ma wrote: > On Tue, 2011-02-01 at 23:24 +0200, Michael S. Tsirkin wrote: > > My theory is that the issue is not signalling. > > Rather, our queue fills up, then host handles > > one packet and sends an interrupt, and we > > immediately wake the queue. So the vq > > once it gets full, stays full. > > >From the printk debugging output, it might not be exactly the case. The > ring gets full, run a bit, then gets full, then run a bit, then full... Yes, but does it get even half empty in between? > > If you try my patch with bufs threshold set to e.g. > > half the vq, what we will do is send interrupt after we have processed > > half the vq. So host has half the vq to go, and guest has half the vq > > to fill. > > > > See? > > I am cleaning up my set up to run your patch ... > > Shirley >