From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH] LRO ack aggregation Date: Mon, 19 Nov 2007 22:22:14 -0800 (PST) Message-ID: <20071119.222214.81458569.davem@davemloft.net> References: <20071119.210919.177660672.davem@davemloft.net> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: gallatin@myri.com, netdev@vger.kernel.org, ossthema@de.ibm.com To: herbert@gondor.apana.org.au Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:39628 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1751983AbXKTGWP (ORCPT ); Tue, 20 Nov 2007 01:22:15 -0500 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org From: Herbert Xu Date: Tue, 20 Nov 2007 14:09:18 +0800 > David Miller wrote: > > > > Fundamentally, I really don't like this change, it batches to the > > point where it begins to erode the natural ACK clocking of TCP, and I > > therefore am very likely to revert it before merging to Linus. > > Perhaps make it a tunable that defaults to off? That's one idea. But if it's there the risk it to have it end up being turned on always by distribution vendors, making our off-default pointless and the internet gets crapped up with misbehaving Linux TCP stacks anyways.