Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751712AbWCJQCS (ORCPT ); Fri, 10 Mar 2006 11:02:18 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751686AbWCJQCS (ORCPT ); Fri, 10 Mar 2006 11:02:18 -0500 Received: from mx.pathscale.com ([64.160.42.68]:7377 "EHLO mx.pathscale.com") by vger.kernel.org with ESMTP id S1751169AbWCJQCR (ORCPT ); Fri, 10 Mar 2006 11:02:17 -0500 Subject: Re: [openib-general] [PATCH 0 of 20] [RFC] ipath driver - another round for review From: "Bryan O'Sullivan" To: "Michael S. Tsirkin" Cc: rolandd@cisco.com, gregkh@suse.de, akpm@osdl.org, davem@davemloft.net, linux-kernel@vger.kernel.org, openib-general@openib.org In-Reply-To: <20060310153559.GA12778@mellanox.co.il> References: <20060310153559.GA12778@mellanox.co.il> Content-Type: text/plain Organization: PathScale, Inc. Date: Fri, 10 Mar 2006 08:02:16 -0800 Message-Id: <1142006537.29925.13.camel@serpentine.pathscale.com> Mime-Version: 1.0 X-Mailer: Evolution 2.2.3 (2.2.3-2.fc4) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1098 Lines: 36 On Fri, 2006-03-10 at 17:35 +0200, Michael S. Tsirkin wrote: > Two questions on this > 1. It is not standard ethernet nor standard IP over Infiniband either, is it? Correct. > Is there some documentation on the wire protocol that you use? No, but the encapsulation is very simple and easy to figure out from reading the code. > Is it pathscale specific? It doesn't have to be. > 2. Are there practical reasons why you get lower latency and higher > bandwidth with this than with IPoIB? The principal reason is that we haven't had time to pay attention to IPoIB performance yet. The ipath_ether driver was developed before IPoIB was usable for any real work. > If there are optimizations, can they not be added to the standard, common > IP over IB driver? That's certainly our intention. It's simply a matter of finding the time for it.