From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: [PATCH net-next 2/4] tcp: do not force quickack when receiving out-of-order packets Date: Thu, 17 May 2018 05:12:11 -0700 Message-ID: <20180517121213.43559-3-edumazet@google.com> References: <20180517121213.43559-1-edumazet@google.com> Cc: netdev , =?UTF-8?q?Toke=20H=C3=B8iland-J=C3=B8rgensen?= , Neal Cardwell , Yuchung Cheng , Soheil Hassas Yeganeh , Eric Dumazet , Eric Dumazet To: "David S . Miller" Return-path: Received: from mail-pl0-f68.google.com ([209.85.160.68]:39050 "EHLO mail-pl0-f68.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750779AbeEQMM1 (ORCPT ); Thu, 17 May 2018 08:12:27 -0400 Received: by mail-pl0-f68.google.com with SMTP id c19-v6so2438308pls.6 for ; Thu, 17 May 2018 05:12:27 -0700 (PDT) In-Reply-To: <20180517121213.43559-1-edumazet@google.com> Sender: netdev-owner@vger.kernel.org List-ID: As explained in commit 9f9843a751d0 ("tcp: properly handle stretch acks in slow start"), TCP stacks have to consider how many packets are acknowledged in one single ACK, because of GRO, but also because of ACK compression or losses. We plan to add SACK compression in the following patch, we must therefore not call tcp_enter_quickack_mode() Signed-off-by: Eric Dumazet --- net/ipv4/tcp_input.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index b188e0d75edd9e5e1c9f0355818caa932fef7416..99fcab7e6570c8b8758ea4b15cdd26df29fb4fd6 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4708,8 +4708,6 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) if (!before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt + tcp_receive_window(tp))) goto out_of_window; - tcp_enter_quickack_mode(sk); - if (before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) { /* Partial packet, seq < rcv_next < end_seq */ SOCK_DEBUG(sk, "partial packet: rcv_next %X seq %X - %X\n", -- 2.17.0.441.gb46fe60e1d-goog