From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: TCP-MD5 checksum failure on x86_64 SMP Date: Fri, 07 May 2010 02:12:51 -0700 (PDT) Message-ID: <20100507.021251.146344534.davem@davemloft.net> References: <1273147586.2357.63.camel@edumazet-laptop> <70125082-4A0B-4E54-A7F6-B6CDC0375E78@nokia.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: eric.dumazet@gmail.com, bhaskie@gmail.com, shemminger@vyatta.com, bhutchings@solarflare.com, netdev@vger.kernel.org To: lars.eggert@nokia.com Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:58320 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750824Ab0EGJMp (ORCPT ); Fri, 7 May 2010 05:12:45 -0400 In-Reply-To: <70125082-4A0B-4E54-A7F6-B6CDC0375E78@nokia.com> Sender: netdev-owner@vger.kernel.org List-ID: From: Lars Eggert Date: Fri, 7 May 2010 11:46:40 +0300 > You'll obviously still want TCP-MD5 support for talking to existing > equipment, but significant cycles would IMO be better spent on > TCP-AO. Code we have and users use which is unstable and crashes is more important to work on and fix than code we don't have which user's therefore don't use. You're wrong from just about every possible angle. Whoever finds AO useful will work on it, just as was the case with MD5 support. And right now, that "whoever" is definitely not us. :-)