From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH net-2.6 1/2] [TCP]: Fix ratehalving with bidirectional flows Date: Mon, 30 Jul 2007 19:49:46 -0700 (PDT) Message-ID: <20070730.194946.08653937.davem@davemloft.net> References: Mime-Version: 1.0 Content-Type: Text/Plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org To: ilpo.jarvinen@helsinki.fi Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:42862 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1759081AbXGaCtz convert rfc822-to-8bit (ORCPT ); Mon, 30 Jul 2007 22:49:55 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org =46rom: "Ilpo_J=E4rvinen" Date: Mon, 30 Jul 2007 20:18:45 +0300 (EEST) > Actually, the ratehalving seems to work too well, as cwnd is > reduced on every second ACK even though the packets in flight > remains unchanged. Recoveries in a bidirectional flows suffer > quite badly because of this, both NewReno and SACK are affected. >=20 > After this patch, rate halving is performed for ACK only if > packets in flight was supposedly changed too. >=20 > Signed-off-by: Ilpo J=E4rvinen This fix looks great, and I see no potential problems with Reno after doing a bit of auditing. Applied and I'll push to -stable too.