From mboxrd@z Thu Jan  1 00:00:00 1970
Content-Type: multipart/mixed; boundary="===============4413616108832711735=="
MIME-Version: 1.0
From: Florian Westphal <fw at strlen.de>
To: mptcp at lists.01.org
Subject: [MPTCP] Re: [PATCH mptcp-next] mptcp: adjust mptcp receive buffer
 limit if subflow has larger one
Date: Sat, 15 Aug 2020 00:17:45 +0200
Message-ID: <20200814221745.GH1660@breakpoint.cc>
In-Reply-To: alpine.OSX.2.23.453.2008141255470.20141@cakenne1-mobl1.amr.corp.intel.com
X-Status: 
X-Keywords:                  
X-UID: 5561

--===============4413616108832711735==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable

Mat Martineau <mathew.j.martineau(a)linux.intel.com> wrote:
> > In addition to tcp autotuning during read, it may also increase the
> > receive buffer in tcp_clamp_window().
> > =

> > In this case, mptcp should adjust its receive buffer size as well so
> > it can pull all pending skbs at once.
> > =

> > At this time, when TCP grows its receive buffer, it may have more
> > skbs ready for processing than what mptcp allows.
> > =

> > In the mptcp case, the receive window is derived from free
> > space of the mptcp parent socket instead of the individual subflows.
> > Following the subflow allows mptcp to grow its receive buffer.
> > =

> > This is especially noticable for loopback traffic, when even two
> > skbs are enough to fill the initial receive window.
> > =

> > Signed-off-by: Florian Westphal <fw(a)strlen.de>
> > ---
> > net/mptcp/protocol.c | 15 +++++++++++----
> > 1 file changed, 11 insertions(+), 4 deletions(-)
> > =

> > diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
> > index c800b9147a3c..d9307a3e1e62 100644
> > --- a/net/mptcp/protocol.c
> > +++ b/net/mptcp/protocol.c
> > @@ -603,6 +603,7 @@ void mptcp_data_ready(struct sock *sk, struct sock =
*ssk)
> > {
> > 	struct mptcp_subflow_context *subflow =3D mptcp_subflow_ctx(ssk);
> > 	struct mptcp_sock *msk =3D mptcp_sk(sk);
> > +	int sk_rbuf, ssk_rbuf;
> > 	bool wake;
> > =

> > 	/* move_skbs_to_msk below can legitly clear the data_avail flag,
> > @@ -613,12 +614,18 @@ void mptcp_data_ready(struct sock *sk, struct soc=
k *ssk)
> > 	if (wake)
> > 		set_bit(MPTCP_DATA_READY, &msk->flags);
> > =

> > -	if (atomic_read(&sk->sk_rmem_alloc) < READ_ONCE(sk->sk_rcvbuf) &&
> > -	    move_skbs_to_msk(msk, ssk))
> > +	sk_rbuf =3D READ_ONCE(sk->sk_rcvbuf);
> > +	ssk_rbuf =3D READ_ONCE(ssk->sk_rcvbuf);
> > +	if (ssk_rbuf > sk_rbuf) {
> > +		WRITE_ONCE(sk->sk_rcvbuf, ssk_rbuf);
> =

> Any concern about this racing with updates to sk->sk_rcvbuf assignment (w=
ith
> lock held) in mptcp_rcv_space_adjust() or other subflows concurrently
> calling mptcp_data_ready?

THe only option I see ATM is to move the adjustment to when we move skbs
from subflow to mptcp socket, since thats the only spot where we hold
both mptcp and subflow socket locks.

> > +		sk_rbuf =3D ssk_rbuf;
> > +	}
> > +
> > +	/* over limit? can't append more skbs to msk */
> > +	if (atomic_read(&sk->sk_rmem_alloc) > sk_rbuf)
> > 		goto wake;

To avoid this from short-ciruiting early, what about calculating
the diff of ssk rcvbuf vs. mpcp rcvbuf space and use that as a
"cushion"?

I think that this would avoid any racing access without adding
complexity.

I'll give it a try.
--===============4413616108832711735==--