[PATCH 3.16.y-ckt 103/126] tcp: apply Kern's check on RTTs used for congestion control

From: Luis Henriques
Date: Wed Dec 09 2015 - 04:41:50 EST

3.16.7-ckt21 -stable review patch. If anyone has any objections, please let me know.


From: Yuchung Cheng <ycheng@xxxxxxxxxx>

commit 9e45a3e36b363cc4c79c70f2b4f994e66543a219 upstream.

Currently ca_seq_rtt_us does not use Kern's check. Fix that by
checking if any packet acked is a retransmit, for both RTT used
for RTT estimation and congestion control.

Fixes: 5b08e47ca ("tcp: prefer packet timing to TS-ECR for RTT")
Signed-off-by: Yuchung Cheng <ycheng@xxxxxxxxxx>
Signed-off-by: Neal Cardwell <ncardwell@xxxxxxxxxx>
Signed-off-by: Eric Dumazet <edumazet@xxxxxxxxxx>
Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>
[ luis: backported to 3.16: adjusted context ]
Signed-off-by: Luis Henriques <luis.henriques@xxxxxxxxxxxxx>
net/ipv4/tcp_input.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 0cdaa633059b..e3d40a303b8f 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -2891,9 +2891,6 @@ static inline bool tcp_ack_update_rtt(struct sock *sk, const int flag,
* Karn's algorithm forbids taking RTT if some retransmitted data
* is acked (RFC6298).
- seq_rtt_us = -1L;
if (seq_rtt_us < 0)
seq_rtt_us = sack_rtt_us;

@@ -3115,7 +3112,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets,

- if (first_ackt.v64) {
+ if (first_ackt.v64 && !(flag & FLAG_RETRANS_DATA_ACKED)) {
seq_rtt_us = skb_mstamp_us_delta(&now, &first_ackt);
ca_seq_rtt_us = skb_mstamp_us_delta(&now, &last_ackt);
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/