Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S270463AbTGNAGF (ORCPT ); Sun, 13 Jul 2003 20:06:05 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S270461AbTGNAGF (ORCPT ); Sun, 13 Jul 2003 20:06:05 -0400 Received: from webmail.hamiltonfunding.la ([12.162.17.40]:46225 "EHLO umhlanga.STRATNET.NET") by vger.kernel.org with ESMTP id S270459AbTGNAGC (ORCPT ); Sun, 13 Jul 2003 20:06:02 -0400 To: "David S. Miller" Cc: alan@storlinksemi.com, linux-kernel@vger.kernel.org, linux-net@vger.kernel.org, netdev@oss.sgi.com Subject: Re: TCP IP Offloading Interface References: <20030713004818.4f1895be.davem@redhat.com> <52u19qwg53.fsf@topspin.com> <20030713160200.571716cf.davem@redhat.com> X-Message-Flag: Warning: May contain useful information X-Priority: 1 X-MSMail-Priority: High From: Roland Dreier Date: 13 Jul 2003 17:20:41 -0700 In-Reply-To: <20030713160200.571716cf.davem@redhat.com> Message-ID: <52llv2vu06.fsf@topspin.com> User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.4 (Common Lisp) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-OriginalArrivalTime: 14 Jul 2003 00:20:44.0218 (UTC) FILETIME=[C3EB59A0:01C3499D] Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1529 Lines: 28 David> I didn't say I agree with all of Moguls ideas, just his David> anti-TOE arguments. For example, I also think RDMA sucks David> too yet he thinks it's a good iea. Sure, he talks about some weaknesses of TOE, but his conclusion is that the time has come for OS developers to start working on TCP offload (for storage). David> You obviously don't understand my ideas if you think that David> it matters whether there is some relationship between the David> MTU and the system page size necessary for the scheme to David> work. I was just quoting part of Mogul's paper that seemed to directly contradict your original post. I also said it would be great to see NIC hardware with support for flow classification. Look, I pretty much agree with you about TOE hardware. Every chip I've seen either requires a bunch of dedicated expensive memory (including a giant CAM) or is just firmware running on a low-performance embedded CPU. But I also think Mogul is right: iSCSI HBAs are going to force OS designers to deal with TCP offload. My whole point was just that it doesn't make much sense to dismiss the whole idea by saying "TOE is evil" and then cite as support a paper that explains why TOEs now make sense and need to be supported. - Roland - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/