From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759453AbYDOIQl (ORCPT ); Tue, 15 Apr 2008 04:16:41 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753837AbYDOIQ1 (ORCPT ); Tue, 15 Apr 2008 04:16:27 -0400 Received: from [88.44.60.11] ([88.44.60.11]:4434 "EHLO smtp-out01.alice-dsl.net" rhost-flags-FAIL-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1754675AbYDOIQZ (ORCPT ); Tue, 15 Apr 2008 04:16:25 -0400 To: David Miller Cc: vgusev@openvz.org, kuznet@ms2.inr.ac.ru, linux-kernel@vger.kernel.org Subject: Re: [RFC][PATCH][NET] Fix never pruned tcp out-of-order queue From: Andi Kleen References: <200804142121.53860.vgusev@openvz.org> <87mynvtuoj.fsf@basil.nowhere.org> <20080415.010131.156668601.davem@davemloft.net> Date: Tue, 15 Apr 2008 10:14:56 +0200 In-Reply-To: <20080415.010131.156668601.davem@davemloft.net> (David Miller's message of "Tue, 15 Apr 2008 01:01:31 -0700 (PDT)") Message-ID: <87iqyjttz3.fsf@basil.nowhere.org> User-Agent: Gnus/5.1008 (Gnus v5.10.8) Emacs/21.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-OriginalArrivalTime: 15 Apr 2008 08:08:10.0349 (UTC) FILETIME=[D83F11D0:01C89ECF] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org David Miller writes: > From: Andi Kleen > Date: Tue, 15 Apr 2008 09:59:40 +0200 > >> Vitaliy Gusev writes: >> >> > Hello! >> > >> > tcp_prune_queue() doesn't prune an out-of-order queue at all. >> >> Why are you saying this? It has code to prune the ooo queue. You're even >> moving it in your patch. >> >> You're saying that the code doesn't work? If yes why? > > There are inappropriate guards there, and it didn't get invoked > from another important code path. > > That's what his change is fixing. Perhaps I'm dense, but I don't see the inappropiate guards. The guards are pretty much the same as before. The main difference seems to be that sk_rmem_schedule/__sk_mem_schedule is called more often, but it is unclear how this affects the ooo pruning which only checks the queue length anyways. -Andi