From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934043Ab0J2PnV (ORCPT ); Fri, 29 Oct 2010 11:43:21 -0400 Received: from e1.ny.us.ibm.com ([32.97.182.141]:56903 "EHLO e1.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932823Ab0J2PnP (ORCPT ); Fri, 29 Oct 2010 11:43:15 -0400 Subject: Re: [RFC PATCH 1/1] vhost: TX used buffer guest signal accumulation From: Shirley Ma To: "Michael S. Tsirkin" Cc: David Miller , netdev@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org In-Reply-To: <20101029081027.GB22688@redhat.com> References: <1288216693.17571.38.camel@localhost.localdomain> <1288240804.14342.1.camel@localhost.localdomain> <20101028052021.GD5599@redhat.com> <1288286062.11251.15.camel@localhost.localdomain> <20101029081027.GB22688@redhat.com> Content-Type: text/plain; charset="UTF-8" Date: Fri, 29 Oct 2010 08:43:08 -0700 Message-ID: <1288366988.4110.5.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3 (2.28.3-1.fc12) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 2010-10-29 at 10:10 +0200, Michael S. Tsirkin wrote: > Hmm. I don't yet understand. We are still doing copies into the per-vq > buffer, and the data copied is really small. Is it about cache line > bounces? Could you try figuring it out? per-vq buffer is much less expensive than 3 put_copy() call. I will collect the profiling data to show that. > > > 2. How about flushing out queued stuff before we exit > > > the handle_tx loop? That would address most of > > > the spec issue. > > > > The performance is almost as same as the previous patch. I will > resubmit > > the modified one, adding vhost_add_used_and_signal_n after handle_tx > > loop for processing pending queue. > > > > This patch was a part of modified macvtap zero copy which I haven't > > submitted yet. I found this helped vhost TX in general. This pending > > queue will be used by DMA done later, so I put it in vq instead of a > > local variable in handle_tx. > > > > Thanks > > Shirley > > BTW why do we need another array? Isn't heads field exactly what we > need > here? head field is only for up to 32, the more used buffers add and signal accumulated the better performance is from test results. That's was one of the reason I didn't use heads. The other reason was I used these buffer for pending dma done in mavctap zero copy patch. It could be up to vq->num in worse case. Thanks Shirley