From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yuchung Cheng Subject: [PATCH net] tcp: fix tcp_mark_head_lost to check skb len before fragmenting Date: Mon, 25 Jan 2016 14:01:53 -0800 Message-ID: <1453759313-22681-1-git-send-email-ycheng@google.com> Cc: netdev@vger.kernel.org, Neal Cardwell , Yuchung Cheng , Eric Dumazet To: davem@davemloft.net Return-path: Received: from mail-pa0-f46.google.com ([209.85.220.46]:35958 "EHLO mail-pa0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932084AbcAYWDI (ORCPT ); Mon, 25 Jan 2016 17:03:08 -0500 Received: by mail-pa0-f46.google.com with SMTP id yy13so86852385pab.3 for ; Mon, 25 Jan 2016 14:03:07 -0800 (PST) Sender: netdev-owner@vger.kernel.org List-ID: From: Neal Cardwell This commit fixes a corner case in tcp_mark_head_lost() which was causing the WARN_ON(len > skb->len) in tcp_fragment() to fire. tcp_mark_head_lost() was assuming that if a packet has tcp_skb_pcount(skb) of N, then it's safe to fragment off a prefix of M*mss bytes, for any M < N. But with the tricky way TCP pcounts are maintained, this is not always true. For example, suppose the sender sends 4 1-byte packets and have the last 3 packet sacked. It will merge the last 3 packets in the write queue into an skb with pcount = 3 and len = 3 bytes. If another recovery happens after a sack reneging event, tcp_mark_head_lost() may attempt to split the skb assuming it has more than 2*MSS bytes. This sounds very counterintuitive, but as the commit description for the related commit c0638c247f55 ("tcp: don't fragment SACKed skbs in tcp_mark_head_lost()") notes, this is because tcp_shifted_skb() coalesces adjacent regions of SACKed skbs, and when doing this it preserves the sum of their packet counts in order to reflect the real-world dynamics on the wire. The c0638c247f55 commit tried to avoid problems by not fragmenting SACKed skbs, since SACKed skbs are where the non-proportionality between pcount and skb->len/mss is known to be possible. However, that commit did not handle the case where during a reneging event one of these weird SACKed skbs becomes an un-SACKed skb, which tcp_mark_head_lost() can then try to fragment. The fix is to simply mark the entire skb lost when this happens. This makes the recovery slightly more aggressive in such corner cases before we detect reordering. But once we detect reordering this code path is by-passed because FACK is disabled. Signed-off-by: Neal Cardwell Signed-off-by: Yuchung Cheng Signed-off-by: Eric Dumazet --- net/ipv4/tcp_input.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 0003d40..d2ad433 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2164,8 +2164,7 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; - int cnt, oldcnt; - int err; + int cnt, oldcnt, lost; unsigned int mss; /* Use SACK to deduce losses of new sequences sent during recovery */ const u32 loss_high = tcp_is_sack(tp) ? tp->snd_nxt : tp->high_seq; @@ -2205,9 +2204,10 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) break; mss = tcp_skb_mss(skb); - err = tcp_fragment(sk, skb, (packets - oldcnt) * mss, - mss, GFP_ATOMIC); - if (err < 0) + /* If needed, chop off the prefix to mark as lost. */ + lost = (packets - oldcnt) * mss; + if (lost < skb->len && + tcp_fragment(sk, skb, lost, mss, GFP_ATOMIC) < 0) break; cnt = packets; } -- 2.7.0.rc3.207.g0ac5344