From mboxrd@z Thu Jan 1 00:00:00 1970 From: Lucas Nussbaum Subject: Re: [PATCH] Make CUBIC Hystart more robust to RTT variations Date: Thu, 10 Mar 2011 09:54:33 +0100 Message-ID: <20110310085433.GA11953@xanadu.blop.info> References: <20110310002458.5a94f563.billfink@mindspring.com> <2049243270.17294.1299737824096.JavaMail.root@tahiti.vyatta.com> <20110310021751.2f5fb8bc.billfink@mindspring.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Stephen Hemminger , Injong Rhee , David Miller , xiyou wangcong , netdev@vger.kernel.org, sangtae ha To: Bill Fink Return-path: Received: from xanadu.blop.info ([178.79.145.134]:52687 "EHLO xanadu.blop.info" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751853Ab1CJJAy (ORCPT ); Thu, 10 Mar 2011 04:00:54 -0500 Content-Disposition: inline In-Reply-To: <20110310021751.2f5fb8bc.billfink@mindspring.com> Sender: netdev-owner@vger.kernel.org List-ID: On 10/03/11 at 02:17 -0500, Bill Fink wrote: > On Wed, 9 Mar 2011, Stephen Hemminger wrote: >=20 > > Bill what is the HZ in your kernel config. > > I am concerned hystart doesn't work well with HZ=3D100 >=20 > HZ=3D1000 >=20 > But I did have tcp_timestamps disabled. Should I re-run > the tests with tcp_timestamps enabled? I ran my tests with timestamps enabled and HZ=3D250. If you have the opportunity to run tests in the same config, it would be great. The HZ=3D250 vs HZ=3D1000 difference could explain why it's working. However, enabling or disabling timestamps shouldn't make a difference, since the hystart code doesn't use TCP_CONG_RTT_STAMP. --=20 | Lucas Nussbaum MCF Universit=E9 Nancy 2 | | lucas.nussbaum@loria.fr LORIA / AlGorille | | http://www.loria.fr/~lnussbau/ +33 3 54 95 86 19 |