tcp: sndbuf autotuning improvements
authorEric Dumazet <edumazet@google.com>
Tue, 1 Oct 2013 17:23:44 +0000 (10:23 -0700)
committerDavid S. Miller <davem@davemloft.net>
Wed, 2 Oct 2013 20:45:17 +0000 (16:45 -0400)
commit6ae705323b716ea7a8cc26bee79176398a9b2e89
treee3b90b1519e5b90a6d3ec8b21493c804da185b76
parentbbe34cf8a1a2cc174e6516fc230b91b531da7ddf
tcp: sndbuf autotuning improvements

tcp_fixup_sndbuf() is underestimating initial send buffer requirements.

It was not noticed because big GSO packets were escaping the limitation,
but with smaller TSO packets (or TSO/GSO/SG off), application hits
sk_sndbuf before having a chance to fill enough packets in socket write
queue.

- initial cwnd can be bigger than 10 for specific routes

- SKB_TRUESIZE() is a bit under real needs in some cases,
  because of power-of-two rounding in kmalloc()

- Fast Recovery (RFC 5681 3.2) : Cubic needs 70% factor

- Extra cushion (application might react slowly to POLLOUT)

tcp_v4_conn_req_fastopen() needs to call tcp_init_metrics() before
calling tcp_init_buffer_space()

Then we realize tcp_new_space() should call tcp_fixup_sndbuf()
instead of duplicating this stuff.

Rename tcp_fixup_sndbuf() to tcp_sndbuf_expand() to be more
descriptive.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Acked-by: Maciej Żenczykowski <maze@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ipv4/tcp_input.c
net/ipv4/tcp_ipv4.c