diff options
| author | Eric Dumazet <edumazet@google.com> | 2025-02-07 15:28:29 +0000 |
|---|---|---|
| committer | Paolo Abeni <pabeni@redhat.com> | 2025-02-11 13:08:00 +0100 |
| commit | 54a378f43425085d0684679d99735696b69165bc (patch) | |
| tree | 637db57f07abee77412a570f3e2028101c8b266c /include | |
| parent | 48b69b4c7e5d74ad66e5214ae8cbdae0b9ea154c (diff) | |
tcp: add the ability to control max RTO
Currently, TCP stack uses a constant (120 seconds)
to limit the RTO value exponential growth.
Some applications want to set a lower value.
Add TCP_RTO_MAX_MS socket option to set a value (in ms)
between 1 and 120 seconds.
It is discouraged to change the socket rto max on a live
socket, as it might lead to unexpected disconnects.
Following patch is adding a netns sysctl to control the
default value at socket creation time.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Diffstat (limited to 'include')
| -rw-r--r-- | include/net/inet_connection_sock.h | 1 | ||||
| -rw-r--r-- | include/net/tcp.h | 16 | ||||
| -rw-r--r-- | include/uapi/linux/tcp.h | 1 |
3 files changed, 13 insertions, 5 deletions
diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 055aa80b05c6..d9978ffacc97 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -90,6 +90,7 @@ struct inet_connection_sock { struct timer_list icsk_delack_timer; __u32 icsk_rto; __u32 icsk_rto_min; + u32 icsk_rto_max; __u32 icsk_delack_max; __u32 icsk_pmtu_cookie; const struct tcp_congestion_ops *icsk_ca_ops; diff --git a/include/net/tcp.h b/include/net/tcp.h index 56557b0104e3..7fd2d7fa4532 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -143,8 +143,9 @@ static_assert((1 << ATO_BITS) > TCP_DELACK_MAX); #define TCP_DELACK_MIN 4U #define TCP_ATO_MIN 4U #endif -#define TCP_RTO_MAX ((unsigned)(120*HZ)) -#define TCP_RTO_MIN ((unsigned)(HZ/5)) +#define TCP_RTO_MAX_SEC 120 +#define TCP_RTO_MAX ((unsigned)(TCP_RTO_MAX_SEC * HZ)) +#define TCP_RTO_MIN ((unsigned)(HZ / 5)) #define TCP_TIMEOUT_MIN (2U) /* Min timeout for TCP timers in jiffies */ #define TCP_TIMEOUT_MIN_US (2*USEC_PER_MSEC) /* Min TCP timeout in microsecs */ @@ -740,10 +741,14 @@ int tcp_mtu_to_mss(struct sock *sk, int pmtu); int tcp_mss_to_mtu(struct sock *sk, int mss); void tcp_mtup_init(struct sock *sk); +static inline unsigned int tcp_rto_max(const struct sock *sk) +{ + return READ_ONCE(inet_csk(sk)->icsk_rto_max); +} + static inline void tcp_bound_rto(struct sock *sk) { - if (inet_csk(sk)->icsk_rto > TCP_RTO_MAX) - inet_csk(sk)->icsk_rto = TCP_RTO_MAX; + inet_csk(sk)->icsk_rto = min(inet_csk(sk)->icsk_rto, tcp_rto_max(sk)); } static inline u32 __tcp_set_rto(const struct tcp_sock *tp) @@ -1428,7 +1433,8 @@ static inline void tcp_reset_xmit_timer(struct sock *sk, { if (pace_delay) when += tcp_pacing_delay(sk); - inet_csk_reset_xmit_timer(sk, what, when, TCP_RTO_MAX); + inet_csk_reset_xmit_timer(sk, what, when, + tcp_rto_max(sk)); } /* Something is really bad, we could not queue an additional packet, diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h index dbf896f3146c..32a27b4a5020 100644 --- a/include/uapi/linux/tcp.h +++ b/include/uapi/linux/tcp.h @@ -136,6 +136,7 @@ enum { #define TCP_AO_REPAIR 42 /* Get/Set SNEs and ISNs */ #define TCP_IS_MPTCP 43 /* Is MPTCP being used? */ +#define TCP_RTO_MAX_MS 44 /* max rto time in ms */ #define TCP_REPAIR_ON 1 #define TCP_REPAIR_OFF 0 |