From cadf9df27e7cf40e390e060a1c71bb86ecde798b Mon Sep 17 00:00:00 2001 From: Stephen Mallon Date: Tue, 20 Nov 2018 19:15:02 +1100 Subject: tcp: Fix SOF_TIMESTAMPING_RX_HARDWARE to use the latest timestamp during TCP coalescing During tcp coalescing ensure that the skb hardware timestamp refers to the highest sequence number data. Previously only the software timestamp was updated during coalescing. Signed-off-by: Stephen Mallon Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/ipv4/tcp_input.c') diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 2868ef28ce52..e695584bb33f 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4363,6 +4363,7 @@ static bool tcp_try_coalesce(struct sock *sk, if (TCP_SKB_CB(from)->has_rxtstamp) { TCP_SKB_CB(to)->has_rxtstamp = true; to->tstamp = from->tstamp; + skb_hwtstamps(to)->hwtstamp = skb_hwtstamps(from)->hwtstamp; } return true; -- cgit From 86de5921a3d5dd246df661e09bdd0a6131b39ae3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 20 Nov 2018 05:53:59 -0800 Subject: tcp: defer SACK compression after DupThresh Jean-Louis reported a TCP regression and bisected to recent SACK compression. After a loss episode (receiver not able to keep up and dropping packets because its backlog is full), linux TCP stack is sending a single SACK (DUPACK). Sender waits a full RTO timer before recovering losses. While RFC 6675 says in section 5, "Algorithm Details", (2) If DupAcks < DupThresh but IsLost (HighACK + 1) returns true -- indicating at least three segments have arrived above the current cumulative acknowledgment point, which is taken to indicate loss -- go to step (4). ... (4) Invoke fast retransmit and enter loss recovery as follows: there are old TCP stacks not implementing this strategy, and still counting the dupacks before starting fast retransmit. While these stacks probably perform poorly when receivers implement LRO/GRO, we should be a little more gentle to them. This patch makes sure we do not enable SACK compression unless 3 dupacks have been sent since last rcv_nxt update. Ideally we should even rearm the timer to send one or two more DUPACK if no more packets are coming, but that will be work aiming for linux-4.21. Many thanks to Jean-Louis for bisecting the issue, providing packet captures and testing this patch. Fixes: 5d9f4262b7ea ("tcp: add SACK compression") Reported-by: Jean-Louis Dupond Tested-by: Jean-Louis Dupond Signed-off-by: Eric Dumazet Acked-by: Neal Cardwell Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'net/ipv4/tcp_input.c') diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index e695584bb33f..1e37c1388189 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4268,7 +4268,7 @@ static void tcp_sack_new_ofo_skb(struct sock *sk, u32 seq, u32 end_seq) * If the sack array is full, forget about the last one. */ if (this_sack >= TCP_NUM_SACKS) { - if (tp->compressed_ack) + if (tp->compressed_ack > TCP_FASTRETRANS_THRESH) tcp_send_ack(sk); this_sack--; tp->rx_opt.num_sacks--; @@ -5189,7 +5189,17 @@ send_now: if (!tcp_is_sack(tp) || tp->compressed_ack >= sock_net(sk)->ipv4.sysctl_tcp_comp_sack_nr) goto send_now; - tp->compressed_ack++; + + if (tp->compressed_ack_rcv_nxt != tp->rcv_nxt) { + tp->compressed_ack_rcv_nxt = tp->rcv_nxt; + if (tp->compressed_ack > TCP_FASTRETRANS_THRESH) + NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPACKCOMPRESSED, + tp->compressed_ack - TCP_FASTRETRANS_THRESH); + tp->compressed_ack = 0; + } + + if (++tp->compressed_ack <= TCP_FASTRETRANS_THRESH) + goto send_now; if (hrtimer_is_queued(&tp->compressed_ack_timer)) return; -- cgit