tcp: Add TCP_USER_TIMEOUT socket option.

[pandora-kernel.git] / net / ipv4 / tcp_timer.c

diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 808bb92..11569de 100644 (file)
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -138,10 +138,10 @@ static void tcp_mtu_probing(struct inet_connection_sock *icsk, struct sock *sk)
  * retransmissions with an initial RTO of TCP_RTO_MIN.
  */
 static bool retransmits_timed_out(struct sock *sk,
-                                 unsigned int boundary)
+                                 unsigned int boundary,
+                                 unsigned int timeout)
 {
-       unsigned int timeout, linear_backoff_thresh;
-       unsigned int start_ts;
+       unsigned int linear_backoff_thresh, start_ts;
 
        if (!inet_csk(sk)->icsk_retransmits)
                return false;
@@ -151,14 +151,15 @@ static bool retransmits_timed_out(struct sock *sk,
        else
                start_ts = tcp_sk(sk)->retrans_stamp;
 
-       linear_backoff_thresh = ilog2(TCP_RTO_MAX/TCP_RTO_MIN);
-
-       if (boundary <= linear_backoff_thresh)
-               timeout = ((2 << boundary) - 1) * TCP_RTO_MIN;
-       else
-               timeout = ((2 << linear_backoff_thresh) - 1) * TCP_RTO_MIN +
-                         (boundary - linear_backoff_thresh) * TCP_RTO_MAX;
+       if (likely(timeout == 0)) {
+               linear_backoff_thresh = ilog2(TCP_RTO_MAX/TCP_RTO_MIN);
 
+               if (boundary <= linear_backoff_thresh)
+                       timeout = ((2 << boundary) - 1) * TCP_RTO_MIN;
+               else
+                       timeout = ((2 << linear_backoff_thresh) - 1) * TCP_RTO_MIN +
+                               (boundary - linear_backoff_thresh) * TCP_RTO_MAX;
+       }
        return (tcp_time_stamp - start_ts) >= timeout;
 }
 
@@ -174,7 +175,7 @@ static int tcp_write_timeout(struct sock *sk)
                        dst_negative_advice(sk);
                retry_until = icsk->icsk_syn_retries ? : sysctl_tcp_syn_retries;
        } else {
-               if (retransmits_timed_out(sk, sysctl_tcp_retries1)) {
+               if (retransmits_timed_out(sk, sysctl_tcp_retries1, 0)) {
                        /* Black hole detection */
                        tcp_mtu_probing(icsk, sk);
 
@@ -187,14 +188,16 @@ static int tcp_write_timeout(struct sock *sk)
 
                        retry_until = tcp_orphan_retries(sk, alive);
                        do_reset = alive ||
-                                  !retransmits_timed_out(sk, retry_until);
+                                  !retransmits_timed_out(sk, retry_until, 0);
 
                        if (tcp_out_of_resources(sk, do_reset))
                                return 1;
                }
        }
 
-       if (retransmits_timed_out(sk, retry_until)) {
+       if (retransmits_timed_out(sk, retry_until,
+           (1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV) ? 0 :
+           icsk->icsk_user_timeout)) {
                /* Has it gone just too far? */
                tcp_write_err(sk);
                return 1;
@@ -436,7 +439,7 @@ out_reset_timer:
                icsk->icsk_rto = min(icsk->icsk_rto << 1, TCP_RTO_MAX);
        }
        inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, icsk->icsk_rto, TCP_RTO_MAX);
-       if (retransmits_timed_out(sk, sysctl_tcp_retries1 + 1))
+       if (retransmits_timed_out(sk, sysctl_tcp_retries1 + 1, 0))
                __sk_dst_reset(sk);
 
 out:;
@@ -556,7 +559,14 @@ static void tcp_keepalive_timer (unsigned long data)
        elapsed = keepalive_time_elapsed(tp);
 
        if (elapsed >= keepalive_time_when(tp)) {
-               if (icsk->icsk_probes_out >= keepalive_probes(tp)) {
+               /* If the TCP_USER_TIMEOUT option is enabled, use that
+                * to determine when to timeout instead.
+                */
+               if ((icsk->icsk_user_timeout != 0 &&
+                   elapsed >= icsk->icsk_user_timeout &&
+                   icsk->icsk_probes_out > 0) ||
+                   (icsk->icsk_user_timeout == 0 &&
+                   icsk->icsk_probes_out >= keepalive_probes(tp))) {
                        tcp_send_active_reset(sk, GFP_ATOMIC);
                        tcp_write_err(sk);
                        goto out;