From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rick Jones Subject: Re: [PATCH] CUBIC v2.3 with new improved slow start Date: Wed, 29 Oct 2008 18:08:09 -0700 Message-ID: <490908F9.6070306@hp.com> References: <006001c93a0d$477d4e30$4a580e98@ncsu2cc0c3fa00> <20081029153928.3d47e26f@extreme> <000b01c93a1c$090222c0$6c01a8c0@RHEELAPTOP> <20081029164045.7422b68d@extreme> <4908F775.9040504@hp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: Stephen Hemminger , Injong Rhee , Injong Rhee , netdev@vger.kernel.org To: Rick Jones Return-path: Received: from g4t0015.houston.hp.com ([15.201.24.18]:33387 "EHLO g4t0015.houston.hp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751200AbYJ3BIO (ORCPT ); Wed, 29 Oct 2008 21:08:14 -0400 In-Reply-To: <4908F775.9040504@hp.com> Sender: netdev-owner@vger.kernel.org List-ID: > FWIW, HP-UX and Solaris also have ACK avoidance heursistics - they have > a Mentat history - although both should be pretty good and neither > should hold-off ACKs for two seconds... IIRC the last time I looked at a packet trace from a link with LRO enabled under "Linux," that too caused "ACK avoidance" - that is we stray from the "ACK every other" faith. rick jones