diff options
| author | Eric Dumazet <edumazet@google.com> | 2016-04-21 20:55:23 +0300 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2016-04-24 21:43:59 +0300 |
| commit | 10d3be569243def8d92ac3722395ef5a59c504e6 (patch) | |
| tree | ac01b70cff99ad2e59c54e80ddcd524eaa9691a8 /net/ipv4/tcp_timer.c | |
| parent | 8cee83dd29dea4e7d27fda3b170381059f628868 (diff) | |
| download | linux-10d3be569243def8d92ac3722395ef5a59c504e6.tar.xz | |
tcp-tso: do not split TSO packets at retransmit time
Linux TCP stack painfully segments all TSO/GSO packets before retransmits.
This was fine back in the days when TSO/GSO were emerging, with their
bugs, but we believe the dark age is over.
Keeping big packets in write queues, but also in stack traversal
has a lot of benefits.
- Less memory overhead, because write queues have less skbs
- Less cpu overhead at ACK processing.
- Better SACK processing, as lot of studies mentioned how
awful linux was at this ;)
- Less cpu overhead to send the rtx packets
(IP stack traversal, netfilter traversal, drivers...)
- Better latencies in presence of losses.
- Smaller spikes in fq like packet schedulers, as retransmits
are not constrained by TCP Small Queues.
1 % packet losses are common today, and at 100Gbit speeds, this
translates to ~80,000 losses per second.
Losses are often correlated, and we see many retransmit events
leading to 1-MSS train of packets, at the time hosts are already
under stress.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Yuchung Cheng <ycheng@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.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 49bc474f8e35..373b03e78aaa 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -404,7 +404,7 @@ void tcp_retransmit_timer(struct sock *sk) goto out; } tcp_enter_loss(sk); - tcp_retransmit_skb(sk, tcp_write_queue_head(sk)); + tcp_retransmit_skb(sk, tcp_write_queue_head(sk), 1); __sk_dst_reset(sk); goto out_reset_timer; } @@ -436,7 +436,7 @@ void tcp_retransmit_timer(struct sock *sk) tcp_enter_loss(sk); - if (tcp_retransmit_skb(sk, tcp_write_queue_head(sk)) > 0) { + if (tcp_retransmit_skb(sk, tcp_write_queue_head(sk), 1) > 0) { /* Retransmission failed because of local congestion, * do not backoff. */ |