summaryrefslogtreecommitdiff
path: root/net/ipv4/tcp_timer.c
diff options
context:
space:
mode:
authorYuchung Cheng <ycheng@google.com>2012-05-02 17:30:04 +0400
committerDavid S. Miller <davem@davemloft.net>2012-05-03 04:56:10 +0400
commit750ea2bafa55aaed208b2583470ecd7122225634 (patch)
tree7656d7697566b0cecc7fbbdd8dbae288bca6d7e3 /net/ipv4/tcp_timer.c
parenteed530b6c67624db3f2cf477bac7c4d005d8f7ba (diff)
downloadlinux-750ea2bafa55aaed208b2583470ecd7122225634.tar.xz
tcp: early retransmit: delayed fast retransmit
Implementing the advanced early retransmit (sysctl_tcp_early_retrans==2). Delays the fast retransmit by an interval of RTT/4. We borrow the RTO timer to implement the delay. If we receive another ACK or send a new packet, the timer is cancelled and restored to original RTO value offset by time elapsed. When the delayed-ER timer fires, we enter fast recovery and perform fast retransmit. 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 'net/ipv4/tcp_timer.c')
-rw-r--r--net/ipv4/tcp_timer.c5
1 files changed, 5 insertions, 0 deletions
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 34d4a02c2f16..e911e6c523ec 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -319,6 +319,11 @@ void tcp_retransmit_timer(struct sock *sk)
struct tcp_sock *tp = tcp_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
+ if (tp->early_retrans_delayed) {
+ tcp_resume_early_retransmit(sk);
+ return;
+ }
+
if (!tp->packets_out)
goto out;