From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: [PATCH net-next] tcp: guarantee forward progress in tcp_sendmsg() Date: Mon, 02 May 2016 21:49:25 -0700 Message-ID: <1462250965.5535.286.camel@edumazet-glaptop3.roam.corp.google.com> References: <1461964613-4872-1-git-send-email-edumazet@google.com> <1461964613-4872-8-git-send-email-edumazet@google.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: netdev , Soheil Hassas Yeganeh , Alexei Starovoitov , Marcelo Ricardo Leitner To: Eric Dumazet , David Miller Return-path: Received: from mail-vk0-f54.google.com ([209.85.213.54]:34940 "EHLO mail-vk0-f54.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750783AbcECEt3 (ORCPT ); Tue, 3 May 2016 00:49:29 -0400 Received: by mail-vk0-f54.google.com with SMTP id b189so10170794vkh.2 for ; Mon, 02 May 2016 21:49:28 -0700 (PDT) In-Reply-To: <1461964613-4872-8-git-send-email-edumazet@google.com> Sender: netdev-owner@vger.kernel.org List-ID: From: Eric Dumazet Under high rx pressure, it is possible tcp_sendmsg() never has a chance to allocate an skb and loop forever as sk_flush_backlog() would always return true. Fix this by calling sk_flush_backlog() only if one skb had been allocated and filled before last backlog check. Fixes: d41a69f1d390 ("tcp: make tcp_sendmsg() aware of socket backlog") Signed-off-by: Eric Dumazet --- net/ipv4/tcp.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index b945c2b046c5ead5503505f250c3c67761b284ae..5c7ed147449c1b7ba029b12e033ad779a631460a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1084,6 +1084,7 @@ int tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) struct sockcm_cookie sockc; int flags, err, copied = 0; int mss_now = 0, size_goal, copied_syn = 0; + bool process_backlog = false; bool sg; long timeo; @@ -1167,9 +1168,10 @@ new_segment: if (!sk_stream_memory_free(sk)) goto wait_for_sndbuf; - if (sk_flush_backlog(sk)) + if (process_backlog && sk_flush_backlog(sk)) { + process_backlog = false; goto restart; - + } skb = sk_stream_alloc_skb(sk, select_size(sk, sg), sk->sk_allocation, @@ -1177,6 +1179,7 @@ new_segment: if (!skb) goto wait_for_memory; + process_backlog = true; /* * Check whether we can use HW checksum. */