From mboxrd@z Thu Jan 1 00:00:00 1970 From: Benjamin LaHaise Subject: Re: [0/14] GRO: Lots of microoptimisations Date: Fri, 12 Jun 2009 12:09:26 -0400 Message-ID: <20090612160926.GA4290@neterion.com> References: <20090527044539.GA32372@gondor.apana.org.au> <20090527175223.GB7804@neterion.com> <20090527230858.GA24278@gondor.apana.org.au> <20090528152143.GA4501@neterion.com> <20090529092821.GA11725@gondor.apana.org.au> <20090529092930.GA11785@gondor.apana.org.au> <20090529162312.GA5191@neterion.com> <20090610054449.GB16984@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: netdev@vger.kernel.org To: Herbert Xu Return-path: Received: from barracuda.s2io.com ([72.1.205.138]:42572 "EHLO barracuda.s2io.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760137AbZFLQJr (ORCPT ); Fri, 12 Jun 2009 12:09:47 -0400 Content-Disposition: inline In-Reply-To: <20090610054449.GB16984@gondor.apana.org.au> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, Jun 10, 2009 at 03:44:49PM +1000, Herbert Xu wrote: > On Fri, May 29, 2009 at 12:23:13PM -0400, Benjamin LaHaise wrote: > > > > It's just being turned off by setting a loadable module option. > > Let me see if I can get my hands on a machine with your NIC. I found at least one reason why: the first skb_shinfo()->frag_list touch in dev_gro_receive() was causing a cache miss. Adding a prefetch in the driver helps that a little bit, but there's still > 500Mbps difference. I'd like to retest on a system with newer CPUs, but it will probably be a few weeks until all the hardware is in the same place. -ben