tcp: fix bogus RTT on special retransmission (c84a5711) · Commits · 戴 / test

RTT may be bogus with tall loss probe (TLP) when a packet is retransmitted and latter (s)acked without TCPCB_SACKED_RETRANS flag. For example, TLP calls __tcp_retransmit_skb() instead of tcp_retransmit_skb(). The skb timestamps are updated but the sacked flag is not marked with TCPCB_SACKED_RETRANS. As a result we'll get bogus RTT in tcp_clean_rtx_queue() or in tcp_sacktag_one() on spurious retransmission. The fix is to apply the sticky flag TCP_EVER_RETRANS to enforce Karn's check on RTT sampling. However this will disable F-RTO if timeout occurs after TLP, by resetting undo_marker in tcp_enter_loss(). We relax this check to only if any pending retransmists are still in-flight. Signed-off-by:

Yuchung Cheng <ycheng@google.com> Acked-by:

Eric Dumazet <edumazet@google.com> Acked-by:

Neal Cardwell <ncardwell@google.com> Acked-by:

Nandita Dukkipati <nanditad@google.com> Signed-off-by:

David S. Miller <davem@davemloft.net>

net/ipv4/tcp_input.c

+2 −1

Original line number	Diff line number	Diff line
		@@ -1945,8 +1945,9 @@ void tcp_enter_loss(struct sock *sk, int how)
		if (skb == tcp_send_head(sk))
		break;

		if (TCP_SKB_CB(skb)->sacked & TCPCB_RETRANS)
		if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS)
		tp->undo_marker = 0;

		TCP_SKB_CB(skb)->sacked &= (~TCPCB_TAGBITS)\|TCPCB_SACKED_ACKED;
		if (!(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED) \|\| how) {
		TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_ACKED;

net/ipv4/tcp_output.c

+8 −3

Original line number	Diff line number	Diff line
		@@ -2337,6 +2337,7 @@ int __tcp_retransmit_skb(struct sock sk, struct sk_buff skb)
		struct tcp_sock *tp = tcp_sk(sk);
		struct inet_connection_sock *icsk = inet_csk(sk);
		unsigned int cur_mss;
		int err;

		/* Inconslusive MTU probe */
		if (icsk->icsk_mtup.probe_size) {
		@@ -2400,11 +2401,15 @@ int __tcp_retransmit_skb(struct sock sk, struct sk_buff skb)
		skb_headroom(skb) >= 0xFFFF)) {
		struct sk_buff *nskb = __pskb_copy(skb, MAX_TCP_HEADER,
		GFP_ATOMIC);
		return nskb ? tcp_transmit_skb(sk, nskb, 0, GFP_ATOMIC) :
		err = nskb ? tcp_transmit_skb(sk, nskb, 0, GFP_ATOMIC) :
		-ENOBUFS;
		} else {
		return tcp_transmit_skb(sk, skb, 1, GFP_ATOMIC);
		err = tcp_transmit_skb(sk, skb, 1, GFP_ATOMIC);
		}

		if (likely(!err))
		TCP_SKB_CB(skb)->sacked \|= TCPCB_EVER_RETRANS;
		return err;
		}

		int tcp_retransmit_skb(struct sock sk, struct sk_buff skb)

Admin message