From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932167AbbA2JE3 (ORCPT ); Thu, 29 Jan 2015 04:04:29 -0500 Received: from ip4-83-240-67-251.cust.nbox.cz ([83.240.67.251]:44818 "EHLO ip4-83-240-18-248.cust.nbox.cz" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S932133AbbA2JEW (ORCPT ); Thu, 29 Jan 2015 04:04:22 -0500 From: Jiri Slaby To: stable@vger.kernel.org Cc: linux-kernel@vger.kernel.org, Herbert Xu , "David S. Miller" , Jiri Slaby Subject: [PATCH 3.12 063/176] tcp: Do not apply TSO segment limit to non-TSO packets Date: Wed, 28 Jan 2015 15:28:13 +0100 Message-Id: X-Mailer: git-send-email 2.2.2 In-Reply-To: References: In-Reply-To: References: Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Herbert Xu 3.12-stable review patch. If anyone has any objections, please let me know. =============== [ Upstream commit 843925f33fcc293d80acf2c5c8a78adf3344d49b ] Thomas Jarosch reported IPsec TCP stalls when a PMTU event occurs. In fact the problem was completely unrelated to IPsec. The bug is also reproducible if you just disable TSO/GSO. The problem is that when the MSS goes down, existing queued packet on the TX queue that have not been transmitted yet all look like TSO packets and get treated as such. This then triggers a bug where tcp_mss_split_point tells us to generate a zero-sized packet on the TX queue. Once that happens we're screwed because the zero-sized packet can never be removed by ACKs. Fixes: 1485348d242 ("tcp: Apply device TSO segment limit earlier") Reported-by: Thomas Jarosch Signed-off-by: Herbert Xu Cheers, Signed-off-by: David S. Miller Signed-off-by: Jiri Slaby --- net/ipv4/tcp_output.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 4c0e55f14f2e..b4435ae4c485 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1871,7 +1871,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, if (unlikely(!tcp_snd_wnd_test(tp, skb, mss_now))) break; - if (tso_segs == 1) { + if (tso_segs == 1 || !sk->sk_gso_max_segs) { if (unlikely(!tcp_nagle_test(tp, skb, mss_now, (tcp_skb_is_last(sk, skb) ? nonagle : TCP_NAGLE_PUSH)))) @@ -1908,7 +1908,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, } limit = mss_now; - if (tso_segs > 1 && !tcp_urg_mode(tp)) + if (tso_segs > 1 && sk->sk_gso_max_segs && !tcp_urg_mode(tp)) limit = tcp_mss_split_point(sk, skb, mss_now, min_t(unsigned int, cwnd_quota, -- 2.2.2