diff mbox series

[net] tcp: fix potential underestimation on rcv_rtt

Message ID 20171213002858.74188-1-tracywwnj@gmail.com
State Accepted, archived
Delegated to: David Miller
Headers show
Series [net] tcp: fix potential underestimation on rcv_rtt | expand

Commit Message

Wei Wang Dec. 13, 2017, 12:28 a.m. UTC 
From: Wei Wang <weiwan@google.com>

When ms timestamp is used, current logic uses 1us in
tcp_rcv_rtt_update() when the real rcv_rtt is within 1 - 999us.
This could cause rcv_rtt underestimation.
Fix it by always using a min value of 1ms if ms timestamp is used.

Fixes: 645f4c6f2ebd ("tcp: switch rcv_rtt_est and rcvq_space to high
resolution timestamps")

Signed-off-by: Wei Wang <weiwan@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 net/ipv4/tcp_input.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

Comments

Neal Cardwell Dec. 13, 2017, 3:04 p.m. UTC | #1 
On Tue, Dec 12, 2017 at 7:28 PM, Wei Wang <weiwan@google.com> wrote:
> From: Wei Wang <weiwan@google.com>
>
> When ms timestamp is used, current logic uses 1us in
> tcp_rcv_rtt_update() when the real rcv_rtt is within 1 - 999us.
> This could cause rcv_rtt underestimation.
> Fix it by always using a min value of 1ms if ms timestamp is used.
>
> Fixes: 645f4c6f2ebd ("tcp: switch rcv_rtt_est and rcvq_space to high
> resolution timestamps")
>
> Signed-off-by: Wei Wang <weiwan@google.com>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> ---

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

Thanks!

neal
David Miller Dec. 13, 2017, 9:03 p.m. UTC | #2 
From: Wei Wang <weiwan@google.com>
Date: Tue, 12 Dec 2017 16:28:58 -0800

> From: Wei Wang <weiwan@google.com>
> 
> When ms timestamp is used, current logic uses 1us in
> tcp_rcv_rtt_update() when the real rcv_rtt is within 1 - 999us.
> This could cause rcv_rtt underestimation.
> Fix it by always using a min value of 1ms if ms timestamp is used.
> 
> Fixes: 645f4c6f2ebd ("tcp: switch rcv_rtt_est and rcvq_space to high
> resolution timestamps")
> 
> Signed-off-by: Wei Wang <weiwan@google.com>
> Signed-off-by: Eric Dumazet <edumazet@google.com>

Please, in the future, do not wrap long Fixes: tag lines.  Also, do
not add an empty line between Fixes: and other tags like Signed-off-by.
That are all tags and belong together as a seamless block of text.

Thank you.
diff mbox series

Patch

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 9550cc42de2d..45f750e85714 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -508,9 +508,6 @@  static void tcp_rcv_rtt_update(struct tcp_sock *tp, u32 sample, int win_dep)
 	u32 new_sample = tp->rcv_rtt_est.rtt_us;
 	long m = sample;
 
-	if (m == 0)
-		m = 1;
-
 	if (new_sample != 0) {
 		/* If we sample in larger samples in the non-timestamp
 		 * case, we could grossly overestimate the RTT especially
@@ -547,6 +544,8 @@  static inline void tcp_rcv_rtt_measure(struct tcp_sock *tp)
 	if (before(tp->rcv_nxt, tp->rcv_rtt_est.seq))
 		return;
 	delta_us = tcp_stamp_us_delta(tp->tcp_mstamp, tp->rcv_rtt_est.time);
+	if (!delta_us)
+		delta_us = 1;
 	tcp_rcv_rtt_update(tp, delta_us, 1);
 
 new_measure:
@@ -563,8 +562,11 @@  static inline void tcp_rcv_rtt_measure_ts(struct sock *sk,
 	    (TCP_SKB_CB(skb)->end_seq -
 	     TCP_SKB_CB(skb)->seq >= inet_csk(sk)->icsk_ack.rcv_mss)) {
 		u32 delta = tcp_time_stamp(tp) - tp->rx_opt.rcv_tsecr;
-		u32 delta_us = delta * (USEC_PER_SEC / TCP_TS_HZ);
+		u32 delta_us;
 
+		if (!delta)
+			delta = 1;
+		delta_us = delta * (USEC_PER_SEC / TCP_TS_HZ);
 		tcp_rcv_rtt_update(tp, delta_us, 0);
 	}
 }