5.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Neal Cardwell ncardwell@google.com
[ Upstream commit d0fa59897e049e84432600e86df82aab3dce7aa5 ]
After the following commit from 2024:
commit e37ab7373696 ("tcp: fix to allow timestamp undo if no retransmits were sent")
...there was buggy behavior where TCP connections without SACK support could easily see erroneous undo events at the end of fast recovery or RTO recovery episodes. The erroneous undo events could cause those connections to suffer repeated loss recovery episodes and high retransmit rates.
The problem was an interaction between the non-SACK behavior on these connections and the undo logic. The problem is that, for non-SACK connections at the end of a loss recovery episode, if snd_una == high_seq, then tcp_is_non_sack_preventing_reopen() holds steady in CA_Recovery or CA_Loss, but clears tp->retrans_stamp to 0. Then upon the next ACK the "tcp: fix to allow timestamp undo if no retransmits were sent" logic saw the tp->retrans_stamp at 0 and erroneously concluded that no data was retransmitted, and erroneously performed an undo of the cwnd reduction, restoring cwnd immediately to the value it had before loss recovery. This caused an immediate burst of traffic and build-up of queues and likely another immediate loss recovery episode.
This commit fixes tcp_packet_delayed() to ignore zero retrans_stamp values for non-SACK connections when snd_una is at or above high_seq, because tcp_is_non_sack_preventing_reopen() clears retrans_stamp in this case, so it's not a valid signal that we can undo.
Note that the commit named in the Fixes footer restored long-present behavior from roughly 2005-2019, so apparently this bug was present for a while during that era, and this was simply not caught.
Fixes: e37ab7373696 ("tcp: fix to allow timestamp undo if no retransmits were sent") Reported-by: Eric Wheeler netdev@lists.ewheeler.net Closes: https://lore.kernel.org/netdev/64ea9333-e7f9-0df-b0f2-8d566143acab@ewheeler.... Signed-off-by: Neal Cardwell ncardwell@google.com Co-developed-by: Yuchung Cheng ycheng@google.com Signed-off-by: Yuchung Cheng ycheng@google.com Reviewed-by: Eric Dumazet edumazet@google.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- net/ipv4/tcp_input.c | 37 +++++++++++++++++++++++++------------ 1 file changed, 25 insertions(+), 12 deletions(-)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index a52d5b718e6e1..10f39b2762a74 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2434,20 +2434,33 @@ static inline bool tcp_packet_delayed(const struct tcp_sock *tp) { const struct sock *sk = (const struct sock *)tp;
- if (tp->retrans_stamp && - tcp_tsopt_ecr_before(tp, tp->retrans_stamp)) - return true; /* got echoed TS before first retransmission */ - - /* Check if nothing was retransmitted (retrans_stamp==0), which may - * happen in fast recovery due to TSQ. But we ignore zero retrans_stamp - * in TCP_SYN_SENT, since when we set FLAG_SYN_ACKED we also clear - * retrans_stamp even if we had retransmitted the SYN. + /* Received an echoed timestamp before the first retransmission? */ + if (tp->retrans_stamp) + return tcp_tsopt_ecr_before(tp, tp->retrans_stamp); + + /* We set tp->retrans_stamp upon the first retransmission of a loss + * recovery episode, so normally if tp->retrans_stamp is 0 then no + * retransmission has happened yet (likely due to TSQ, which can cause + * fast retransmits to be delayed). So if snd_una advanced while + * (tp->retrans_stamp is 0 then apparently a packet was merely delayed, + * not lost. But there are exceptions where we retransmit but then + * clear tp->retrans_stamp, so we check for those exceptions. */ - if (!tp->retrans_stamp && /* no record of a retransmit/SYN? */ - sk->sk_state != TCP_SYN_SENT) /* not the FLAG_SYN_ACKED case? */ - return true; /* nothing was retransmitted */
- return false; + /* (1) For non-SACK connections, tcp_is_non_sack_preventing_reopen() + * clears tp->retrans_stamp when snd_una == high_seq. + */ + if (!tcp_is_sack(tp) && !before(tp->snd_una, tp->high_seq)) + return false; + + /* (2) In TCP_SYN_SENT tcp_clean_rtx_queue() clears tp->retrans_stamp + * when setting FLAG_SYN_ACKED is set, even if the SYN was + * retransmitted. + */ + if (sk->sk_state == TCP_SYN_SENT) + return false; + + return true; /* tp->retrans_stamp is zero; no retransmit yet */ }
/* Undo procedures. */