On Tue, Sep 20, 2016 at 10:48 AM, Stephen Hemminger
<step...@networkplumber.org> wrote:
>
> On Mon, 19 Sep 2016 23:39:20 -0400
> Neal Cardwell <ncardw...@google.com> wrote:
>
> > From: Yuchung Cheng <ych...@google.com>
> >
> > Currently the TCP send buffer expands to twice cwnd, in order to allow
> > limited transmits in the CA_Recovery state. This assumes that cwnd
> > does not increase in the CA_Recovery.
> >
> > For some congestion control algorithms, like the upcoming BBR module,
> > if the losses in recovery do not indicate congestion then we may
> > continue to raise cwnd multiplicatively in recovery. In such cases the
> > current multiplier will falsely limit the sending rate, much as if it
> > were limited by the application.
> >
> > This commit adds an optional congestion control callback to use a
> > different multiplier to expand the TCP send buffer. For congestion
> > control modules that do not specificy this callback, TCP continues to
> > use the previous default of 2.
> >
> > Signed-off-by: Van Jacobson <v...@google.com>
> > Signed-off-by: Neal Cardwell <ncardw...@google.com>
> > Signed-off-by: Yuchung Cheng <ych...@google.com>
> > Signed-off-by: Nandita Dukkipati <nandi...@google.com>
> > Signed-off-by: Eric Dumazet <eduma...@google.com>
> > Signed-off-by: Soheil Hassas Yeganeh <soh...@google.com>
> > ---
> > include/net/tcp.h | 2 ++
> > net/ipv4/tcp_input.c | 4 +++-
> > 2 files changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/include/net/tcp.h b/include/net/tcp.h
> > index 3492041..1aa9628 100644
> > --- a/include/net/tcp.h
> > +++ b/include/net/tcp.h
> > @@ -917,6 +917,8 @@ struct tcp_congestion_ops {
> > void (*pkts_acked)(struct sock *sk, const struct ack_sample *sample);
> > /* suggest number of segments for each skb to transmit (optional) */
> > u32 (*tso_segs_goal)(struct sock *sk);
> > + /* returns the multiplier used in tcp_sndbuf_expand (optional) */
> > + u32 (*sndbuf_expand)(struct sock *sk);
> > /* get info for inet_diag (optional) */
> > size_t (*get_info)(struct sock *sk, u32 ext, int *attr,
> > union tcp_cc_info *info);
> > diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
> > index 17de77d..5af0bf3 100644
> > --- a/net/ipv4/tcp_input.c
> > +++ b/net/ipv4/tcp_input.c
> > @@ -289,6 +289,7 @@ static bool tcp_ecn_rcv_ecn_echo(const struct tcp_sock
> > *tp, const struct tcphdr
> > static void tcp_sndbuf_expand(struct sock *sk)
> > {
> > const struct tcp_sock *tp = tcp_sk(sk);
> > + const struct tcp_congestion_ops *ca_ops = inet_csk(sk)->icsk_ca_ops;
> > int sndmem, per_mss;
> > u32 nr_segs;
> >
> > @@ -309,7 +310,8 @@ static void tcp_sndbuf_expand(struct sock *sk)
> > * Cubic needs 1.7 factor, rounded to 2 to include
> > * extra cushion (application might react slowly to POLLOUT)
> > */
> > - sndmem = 2 * nr_segs * per_mss;
> > + sndmem = ca_ops->sndbuf_expand ? ca_ops->sndbuf_expand(sk) : 2;
>
> You could avoid the conditional (if it mattered) by inheriting a default value
> that would mean changing all existing congestion control modules.
> So doing it this way makes life easier.
Good point. But we've not really tested raising it to 3 on other
C.C.s, esp loss-based ones tend to have big cwnd on bloated buffers.
so we like to change locally for now.
>
> Acked-by: Stephen Hemminger <step...@networkplumber.org>
