Commit c3916ad9 authored by Eric Dumazet's avatar Eric Dumazet Committed by David S. Miller

tcp: smoother receiver autotuning

Back in linux-3.13 (commit b0983d3c ("tcp: fix dynamic right sizing"))
I addressed the pressing issues we had with receiver autotuning.

But DRS suffers from extra latencies caused by rcv_rtt_est.rtt_us
drifts. One common problem happens during slow start, since the
apparent RTT measured by the receiver can be inflated by ~50%,
at the end of one packet train.

Also, a single drop can delay read() calls by one RTT, meaning
tcp_rcv_space_adjust() can be called one RTT too late.

By replacing the tri-modal heuristic with a continuous function,
we can offset the effects of not growing 'at the optimal time'.

The curve of the function matches prior behavior if the space
increased by 25% and 50% exactly.

Cost of added multiply/divide is small, considering a TCP flow
typically would run this part of the code few times in its life.

I tested this patch with 100 ms RTT / 1% loss link, 100 runs
of (netperf -l 5), and got an average throughput of 4600 Mbit
instead of 1700 Mbit.
Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
Acked-by: default avatarSoheil Hassas Yeganeh <soheil@google.com>
Acked-by: default avatarWei Wang <weiwan@google.com>
Acked-by: default avatarNeal Cardwell <ncardwell@google.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent 607065ba
...@@ -601,26 +601,17 @@ void tcp_rcv_space_adjust(struct sock *sk) ...@@ -601,26 +601,17 @@ void tcp_rcv_space_adjust(struct sock *sk)
if (sock_net(sk)->ipv4.sysctl_tcp_moderate_rcvbuf && if (sock_net(sk)->ipv4.sysctl_tcp_moderate_rcvbuf &&
!(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) { !(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) {
int rcvmem, rcvbuf; int rcvmem, rcvbuf;
u64 rcvwin; u64 rcvwin, grow;
/* minimal window to cope with packet losses, assuming /* minimal window to cope with packet losses, assuming
* steady state. Add some cushion because of small variations. * steady state. Add some cushion because of small variations.
*/ */
rcvwin = ((u64)copied << 1) + 16 * tp->advmss; rcvwin = ((u64)copied << 1) + 16 * tp->advmss;
/* If rate increased by 25%, /* Accommodate for sender rate increase (eg. slow start) */
* assume slow start, rcvwin = 3 * copied grow = rcvwin * (copied - tp->rcvq_space.space);
* If rate increased by 50%, do_div(grow, tp->rcvq_space.space);
* assume sender can use 2x growth, rcvwin = 4 * copied rcvwin += (grow << 1);
*/
if (copied >=
tp->rcvq_space.space + (tp->rcvq_space.space >> 2)) {
if (copied >=
tp->rcvq_space.space + (tp->rcvq_space.space >> 1))
rcvwin <<= 1;
else
rcvwin += (rcvwin >> 1);
}
rcvmem = SKB_TRUESIZE(tp->advmss + MAX_TCP_HEADER); rcvmem = SKB_TRUESIZE(tp->advmss + MAX_TCP_HEADER);
while (tcp_win_from_space(sk, rcvmem) < tp->advmss) while (tcp_win_from_space(sk, rcvmem) < tp->advmss)
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment