diff options
| author | andre <andre@FreeBSD.org> | 2006-02-16 19:38:07 +0000 |
|---|---|---|
| committer | andre <andre@FreeBSD.org> | 2006-02-16 19:38:07 +0000 |
| commit | e83c574f873ca62a647c1d708a65c507c50a3b52 (patch) | |
| tree | 50085f7cd5f6a0c2d314a7e5e7a2100f076b6b44 /sys/netinet/tcp_var.h | |
| parent | ae2214aeaef3707cbd8f1f7a2791b518a8ccef30 (diff) | |
| download | FreeBSD-src-e83c574f873ca62a647c1d708a65c507c50a3b52.zip FreeBSD-src-e83c574f873ca62a647c1d708a65c507c50a3b52.tar.gz | |
Have TCP Inflight disable itself if the RTT is below a certain
threshold. Inflight doesn't make sense on a LAN as it has
trouble figuring out the maximal bandwidth because of the coarse
tick granularity.
The sysctl net.inet.tcp.inflight.rttthresh specifies the threshold
in milliseconds below which inflight will disengage. It defaults
to 10ms.
Tested by: Joao Barros <joao.barros-at-gmail.com>,
Rich Murphey <rich-at-whiteoaklabs.com>
Sponsored by: TCP/IP Optimization Fundraise 2005
Diffstat (limited to 'sys/netinet/tcp_var.h')
| -rw-r--r-- | sys/netinet/tcp_var.h | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/sys/netinet/tcp_var.h b/sys/netinet/tcp_var.h index b1515dd..f9050df 100644 --- a/sys/netinet/tcp_var.h +++ b/sys/netinet/tcp_var.h @@ -199,6 +199,7 @@ struct tcpcb { tcp_seq sack_newdata; /* New data xmitted in this recovery episode starts at this seq number */ struct sackhint sackhint; /* SACK scoreboard hint */ + int t_rttlow; /* smallest observerved RTT */ }; #define IN_FASTRECOVERY(tp) (tp->t_flags & TF_FASTRECOVERY) |
