Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Sat, 3 Mar 2001 03:38:42 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Sat, 3 Mar 2001 03:38:33 -0500 Received: from [139.134.6.86] ([139.134.6.86]:51694 "EHLO mailin9.bigpond.com") by vger.kernel.org with ESMTP id ; Sat, 3 Mar 2001 03:38:27 -0500 Message-ID: <3AA1D522.7050300@bigpond.net.au> Date: Sat, 03 Mar 2001 19:39:46 -1000 From: Mark Reginald James Organization: WahSounds User-Agent: Mozilla/5.0 (X11; U; Linux 2.4.2 i686; en-US; 0.7) Gecko/20010119 X-Accept-Language: en MIME-Version: 1.0 To: linux-kernel@vger.kernel.org Subject: TCP Congestion Window Bug? Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi: TCP only sends a packet if: tcp_packets_in_flight(tp) < tp->snd_cwnd (function tcp_snd_test in include/net/tcp.h) but regards transmission as application-limited if tp->packets_out < tp->snd_cwnd (function tcp_cwnd_validate in include/net/tcp.h) So the kernel _always_ thinks the connection is application-limited, forcing many un-necessary reductions in the size of the congestion window. I think the condition in tcp_snd_test should be: tcp_packets_in_flight(tp) <= tp->snd_cwnd Mark - 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/