From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754713AbXLFWIJ (ORCPT ); Thu, 6 Dec 2007 17:08:09 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751739AbXLFWH5 (ORCPT ); Thu, 6 Dec 2007 17:07:57 -0500 Received: from one.firstfloor.org ([213.235.205.2]:59201 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751913AbXLFWH4 (ORCPT ); Thu, 6 Dec 2007 17:07:56 -0500 Date: Thu, 6 Dec 2007 23:07:54 +0100 From: Andi Kleen To: Chris Friesen Cc: Andi Kleen , Renzo Davoli , linux-kernel@vger.kernel.org Subject: Re: New Address Family: Inter Process Networking (IPN) Message-ID: <20071206220754.GF20595@one.firstfloor.org> References: <20071205164055.GA2082@cs.unibo.it> <20071206053016.GA1464@cs.unibo.it> <20071206163538.GC20595@one.firstfloor.org> <47585D66.3000404@nortel.com> <20071206212644.GA26725@one.firstfloor.org> <47586E7F.1020903@nortel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <47586E7F.1020903@nortel.com> User-Agent: Mutt/1.4.2.1i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Dec 06, 2007 at 03:49:51PM -0600, Chris Friesen wrote: > Andi Kleen wrote: > >>Latency was very > >>important, so we ended up doing essentially a multicast unix socket > >>rather than taking the extra penalty for UDP multicast. > > > > > >What extra penalty? Local UDP shouldn't be much more expensive than Unix. > > On a 1.4GHz P4 I measured a 44% increase in latency between a unix > datagram and a UDP datagram. That's weird. > > For UDP it has to go down the udp stack, then the ip stack, then through UDP doesn't really have much stack. IP is also very little assuming cached route (connect called first) I would expect the copies to dominate in both cases. -Andi