diff options
author | Eric Dumazet <edumazet@google.com> | 2015-05-07 00:26:24 +0300 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2015-05-09 23:42:32 +0300 |
commit | 21c8fe9915276d923f8c1e43434fd6d37a3b9aef (patch) | |
tree | 3bf845d67545bbb3b411b3221cffb37e3ed9ce57 /include/net | |
parent | b063bc5ea77b1c1c0e7798f641f53504d0f64bf8 (diff) | |
download | linux-21c8fe9915276d923f8c1e43434fd6d37a3b9aef.tar.xz |
tcp: adjust window probe timers to safer values
With the advent of small rto timers in datacenter TCP,
(ip route ... rto_min x), the following can happen :
1) Qdisc is full, transmit fails.
TCP sets a timer based on icsk_rto to retry the transmit, without
exponential backoff.
With low icsk_rto, and lot of sockets, all cpus are servicing timer
interrupts like crazy.
Intent of the code was to retry with a timer between 200 (TCP_RTO_MIN)
and 500ms (TCP_RESOURCE_PROBE_INTERVAL)
2) Receivers can send zero windows if they don't drain their receive queue.
TCP sends zero window probes, based on icsk_rto current value, with
exponential backoff.
With /proc/sys/net/ipv4/tcp_retries2 being 15 (or even smaller in
some cases), sender can abort in less than one or two minutes !
If receiver stops the sender, it obviously doesn't care of very tight
rto. Probability of dropping the ACK reopening the window is not
worth the risk.
Lets change the base timer to be at least 200ms (TCP_RTO_MIN) for these
events (but not normal RTO based retransmits)
A followup patch adds a new SNMP counter, as it would have helped a lot
diagnosing this issue.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'include/net')
-rw-r--r-- | include/net/tcp.h | 27 |
1 files changed, 22 insertions, 5 deletions
diff --git a/include/net/tcp.h b/include/net/tcp.h index 6d204f3f9df8..7a2248a35b13 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1043,14 +1043,31 @@ static inline bool tcp_is_cwnd_limited(const struct sock *sk) return tp->is_cwnd_limited; } -static inline void tcp_check_probe_timer(struct sock *sk) +/* Something is really bad, we could not queue an additional packet, + * because qdisc is full or receiver sent a 0 window. + * We do not want to add fuel to the fire, or abort too early, + * so make sure the timer we arm now is at least 200ms in the future, + * regardless of current icsk_rto value (as it could be ~2ms) + */ +static inline unsigned long tcp_probe0_base(const struct sock *sk) { - const struct tcp_sock *tp = tcp_sk(sk); - const struct inet_connection_sock *icsk = inet_csk(sk); + return max_t(unsigned long, inet_csk(sk)->icsk_rto, TCP_RTO_MIN); +} - if (!tp->packets_out && !icsk->icsk_pending) +/* Variant of inet_csk_rto_backoff() used for zero window probes */ +static inline unsigned long tcp_probe0_when(const struct sock *sk, + unsigned long max_when) +{ + u64 when = (u64)tcp_probe0_base(sk) << inet_csk(sk)->icsk_backoff; + + return (unsigned long)min_t(u64, when, max_when); +} + +static inline void tcp_check_probe_timer(struct sock *sk) +{ + if (!tcp_sk(sk)->packets_out && !inet_csk(sk)->icsk_pending) inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0, - icsk->icsk_rto, TCP_RTO_MAX); + tcp_probe0_base(sk), TCP_RTO_MAX); } static inline void tcp_init_wl(struct tcp_sock *tp, u32 seq) |