From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH net-next v2] tcp: force cwnd at least 2 in tcp_cwnd_reduction Date: Wed, 27 Jun 2018 08:04:57 -0700 Message-ID: References: <20180627023403.3395818-1-brakmo@fb.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Cc: Kernel Team , Blake Matheny , Alexei Starovoitov To: Lawrence Brakmo , netdev Return-path: Received: from mail-oi0-f53.google.com ([209.85.218.53]:34732 "EHLO mail-oi0-f53.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753371AbeF0PFB (ORCPT ); Wed, 27 Jun 2018 11:05:01 -0400 Received: by mail-oi0-f53.google.com with SMTP id c2-v6so2173682oic.1 for ; Wed, 27 Jun 2018 08:05:00 -0700 (PDT) In-Reply-To: <20180627023403.3395818-1-brakmo@fb.com> Content-Language: en-US Sender: netdev-owner@vger.kernel.org List-ID: On 06/26/2018 07:34 PM, Lawrence Brakmo wrote: > When using dctcp and doing RPCs, if the last packet of a request is > ECN marked as having seen congestion (CE), the sender can decrease its > cwnd to 1. As a result, it will only send one packet when a new request > is sent. In some instances this results in high tail latencies. > > } > /* Force a fast retransmit upon entering fast recovery */ > sndcnt = max(sndcnt, (tp->prr_out ? 0 : 1)); > - tp->snd_cwnd = tcp_packets_in_flight(tp) + sndcnt; > + tp->snd_cwnd = max((int)tcp_packets_in_flight(tp) + sndcnt, 2); Canonical way is to use min_t(), please respin (no need to explain this trivia in changelog) Thanks.