mbox series

[net-next,0/2] mptcp: add receive buffer auto-tuning

Message ID 20200630192445.18333-1-fw@strlen.de
Headers show
Series mptcp: add receive buffer auto-tuning | expand

Message

Florian Westphal June 30, 2020, 7:24 p.m. UTC
First patch extends the test script to allow for reproducible results.
Second patch adds receive auto-tuning.  Its based on what TCP is doing,
only difference is that we use the largest RTT of any of the subflows
and that we will update all subflows with the new value.

Else, we get spurious packet drops because the mptcp work queue might
not be able to move packets from subflow socket to master socket
fast enough.  Without the adjustment, TCP may drop the packets because
the subflow socket is over its rcvbuffer limit.

Florian Westphal (2):
      selftests: mptcp: add option to specify size of file to transfer
      mptcp: add receive buffer auto-tuning

 net/mptcp/protocol.c                               | 123 +++++++++++++++++++--
 net/mptcp/protocol.h                               |   7 ++
 net/mptcp/subflow.c                                |   5 +-
 tools/testing/selftests/net/mptcp/mptcp_connect.sh |  52 ++++++---
 4 files changed, 166 insertions(+), 21 deletions(-)

Comments

David Miller July 2, 2020, 12:48 a.m. UTC | #1
From: Florian Westphal <fw@strlen.de>
Date: Tue, 30 Jun 2020 21:24:43 +0200

> First patch extends the test script to allow for reproducible results.
> Second patch adds receive auto-tuning.  Its based on what TCP is doing,
> only difference is that we use the largest RTT of any of the subflows
> and that we will update all subflows with the new value.
> 
> Else, we get spurious packet drops because the mptcp work queue might
> not be able to move packets from subflow socket to master socket
> fast enough.  Without the adjustment, TCP may drop the packets because
> the subflow socket is over its rcvbuffer limit.

Series applied, thanks Florian.