From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH] allow dctcp alpha to drop to zero Date: Fri, 23 Oct 2015 02:47:20 -0700 (PDT) Message-ID: <20151023.024720.1261240812356446593.davem@davemloft.net> References: <20151019045908.GB11368@mininet-vm> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, dborkman@redhat.com, fw@strlen.de, glenn.judd@morganstanley.com To: agshew@gmail.com Return-path: Received: from shards.monkeyblade.net ([149.20.54.216]:33407 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750862AbbJWJau (ORCPT ); Fri, 23 Oct 2015 05:30:50 -0400 In-Reply-To: <20151019045908.GB11368@mininet-vm> Sender: netdev-owner@vger.kernel.org List-ID: From: Andrew Shewmaker Date: Sun, 18 Oct 2015 21:59:08 -0700 > If alpha is strictly reduced by alpha >> dctcp_shift_g and if alpha is less > than 1 << dctcp_shift_g, then alpha may never reach zero. For example, > given shift_g=4 and alpha=15, alpha >> dctcp_shift_g yields 0 and alpha > remains 15. The effect isn't noticeable in this case below cwnd=137, but > could gradually drive uncongested flows with leftover alpha down to > cwnd=137. A larger dctcp_shift_g would have a greater effect. > > This change causes alpha=15 to drop to 0 instead of being decrementing by 1 > as it would when alpha=16. However, it requires one less conditional to > implement since it doesn't have to guard against subtracting 1 from 0U. A > decay of 15 is not unreasonable since an equal or greater amount occurs at > alpha >= 240. > > Signed-off-by: Andrew G. Shewmaker Applied, thanks.