From: Yuchung Cheng <ycheng@google.com>
Date: Wed, 16 Jan 2019 23:05:30 +0000 (-0800)
Subject: tcp: always set retrans_stamp on recovery
X-Git-Url: http://git.cdn.openwrt.org/?a=commitdiff_plain;h=7ae189759cc48cf8b54beebff566e9fd2d4e7d7c;p=openwrt%2Fstaging%2Fblogic.git

tcp: always set retrans_stamp on recovery

Previously TCP socket's retrans_stamp is not set if the
retransmission has failed to send. As a result if a socket is
experiencing local issues to retransmit packets, determining when
to abort a socket is complicated w/o knowning the starting time of
the recovery since retrans_stamp may remain zero.

This complication causes sub-optimal behavior that TCP may use the
latest, instead of the first, retransmission time to compute the
elapsed time of a stalling connection due to local issues. Then TCP
may disrecard TCP retries settings and keep retrying until it finally
succeed: not a good idea when the local host is already strained.

The simple fix is to always timestamp the start of a recovery.
It's worth noting that retrans_stamp is also used to compare echo
timestamp values to detect spurious recovery. This patch does
not break that because retrans_stamp is still later than when the
original packet was sent.

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 57a56e205070..d2d494c74811 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -2963,13 +2963,12 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs)
 #endif
 		TCP_SKB_CB(skb)->sacked |= TCPCB_RETRANS;
 		tp->retrans_out += tcp_skb_pcount(skb);
-
-		/* Save stamp of the first retransmit. */
-		if (!tp->retrans_stamp)
-			tp->retrans_stamp = tcp_skb_timestamp(skb);
-
 	}
 
+	/* Save stamp of the first (attempted) retransmit. */
+	if (!tp->retrans_stamp)
+		tp->retrans_stamp = tcp_skb_timestamp(skb);
+
 	if (tp->undo_retrans < 0)
 		tp->undo_retrans = 0;
 	tp->undo_retrans += tcp_skb_pcount(skb);
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index e7d09e3705b8..1e61f0bd6e24 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -22,28 +22,14 @@
 #include <linux/gfp.h>
 #include <net/tcp.h>
 
-static u32 tcp_retransmit_stamp(const struct sock *sk)
-{
-	u32 start_ts = tcp_sk(sk)->retrans_stamp;
-
-	if (unlikely(!start_ts)) {
-		struct sk_buff *head = tcp_rtx_queue_head(sk);
-
-		if (!head)
-			return 0;
-		start_ts = tcp_skb_timestamp(head);
-	}
-	return start_ts;
-}
-
 static u32 tcp_clamp_rto_to_user_timeout(const struct sock *sk)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	u32 elapsed, start_ts;
 	s32 remaining;
 
-	start_ts = tcp_retransmit_stamp(sk);
-	if (!icsk->icsk_user_timeout || !start_ts)
+	start_ts = tcp_sk(sk)->retrans_stamp;
+	if (!icsk->icsk_user_timeout)
 		return icsk->icsk_rto;
 	elapsed = tcp_time_stamp(tcp_sk(sk)) - start_ts;
 	remaining = icsk->icsk_user_timeout - elapsed;
@@ -197,10 +183,7 @@ static bool retransmits_timed_out(struct sock *sk,
 	if (!inet_csk(sk)->icsk_retransmits)
 		return false;
 
-	start_ts = tcp_retransmit_stamp(sk);
-	if (!start_ts)
-		return false;
-
+	start_ts = tcp_sk(sk)->retrans_stamp;
 	if (likely(timeout == 0)) {
 		linear_backoff_thresh = ilog2(TCP_RTO_MAX/rto_base);