diff options
| author | David S. Miller <davem@davemloft.net> | 2016-11-21 13:20:17 -0500 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2016-11-21 13:20:17 -0500 |
| commit | 9e36ced6335be42e637a827b64f510f143dbb5a7 (patch) | |
| tree | 870e5c09c37890d286872c9ae22382d6bf336b37 /net/ipv4/tcp_lp.c | |
| parent | 2fcb58ab30deb63e49f238bf95d587740fab59c4 (diff) | |
| parent | e97991832a4ea4a5f47d65f068a4c966a2eb5730 (diff) | |
Merge branch 'tcp-cong-undo_cwnd-mandatory'
Florian Westphal says:
====================
tcp: make undo_cwnd mandatory for congestion modules
highspeed, illinois, scalable, veno and yeah congestion control algorithms
don't provide a 'cwnd_undo' function. This makes the stack default to a
'reno undo' which doubles cwnd. However, the ssthresh implementation of
these algorithms do not halve the slowstart threshold. This causes similar
issue as the one fixed for dctcp in ce6dd23329b1e ("dctcp: avoid bogus
doubling of cwnd after loss").
In light of this it seems better to remove the fallback and make undo_cwnd
mandatory.
First patch fixes those spots where reno undo seems incorrect by providing
.cwnd_undo functions, second patch removes the fallback.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/tcp_lp.c')
| -rw-r--r-- | net/ipv4/tcp_lp.c | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/net/ipv4/tcp_lp.c b/net/ipv4/tcp_lp.c index c67ece1390c2..046fd3910873 100644 --- a/net/ipv4/tcp_lp.c +++ b/net/ipv4/tcp_lp.c @@ -316,6 +316,7 @@ static void tcp_lp_pkts_acked(struct sock *sk, const struct ack_sample *sample) static struct tcp_congestion_ops tcp_lp __read_mostly = { .init = tcp_lp_init, .ssthresh = tcp_reno_ssthresh, + .undo_cwnd = tcp_reno_undo_cwnd, .cong_avoid = tcp_lp_cong_avoid, .pkts_acked = tcp_lp_pkts_acked, |