From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mike Galbraith Subject: Re: [tbench regression fixes]: digging out smelly deadmen. Date: Sat, 25 Oct 2008 07:53:05 +0200 Message-ID: <1224913985.3822.12.camel@marge.simson.net> References: <200810250025.35734.rjw@sisk.pl> <20081024.163111.244324800.davem@davemloft.net> <1224907501.5161.36.camel@marge.simson.net> <20081024.221525.259940836.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: rjw@sisk.pl, mingo@elte.hu, s0mbre@tservice.net.ru, a.p.zijlstra@chello.nl, linux-kernel@vger.kernel.org, netdev@vger.kernel.org To: David Miller Return-path: Received: from mail.gmx.net ([213.165.64.20]:60559 "HELO mail.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1751222AbYJYFxK (ORCPT ); Sat, 25 Oct 2008 01:53:10 -0400 In-Reply-To: <20081024.221525.259940836.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: On Fri, 2008-10-24 at 22:15 -0700, David Miller wrote: > From: Mike Galbraith > Date: Sat, 25 Oct 2008 06:05:01 +0200 > > > My test data indicates (to me anyway) that there is another source of > > localhost throughput loss in .27. In that data, there is no hrtick > > overhead since I didn't have highres timers enabled, and computational > > costs added in .27 were removed. Dunno where it lives, but it does > > appear to exist. > > Disabling TSO on loopback doesn't fix that bit for you? No. Those numbers are with TSO/GSO disabled. I did a manual 100% sched and everything related revert to 26 scheduler, and had ~the same result as these numbers. 27 with 100% revert actually performed a bit _worse_ for me than 27 with it's overhead.. which puzzles me greatly. -Mike