[net-next,0/6] mptcp: avoid workqueue usage for data

Message

Paolo Abeni Nov. 27, 2020, 10:10 a.m. UTC

The current locking schema used to protect the MPTCP data-path
requires the usage of the MPTCP workqueue to process the incoming
data, depending on trylock result.

The above poses scalability limits and introduces random delays
in MPTCP-level acks.

With this series we use a single spinlock to protect the MPTCP
data-path, removing the need for workqueue and delayed ack usage.

This additionally reduces the number of atomic operations required
per packet and cleans-up considerably the poll/wake-up code.

Paolo Abeni (6):
  mptcp: open code mptcp variant for lock_sock
  mptcp: implement wmem reservation
  mptcp: protect the rx path with the msk socket spinlock
  mptcp: allocate TX skbs in msk context
  mptcp: avoid a few atomic ops in the rx path
  mptcp: use mptcp release_cb for delayed tasks

 include/net/sock.h     |   1 +
 net/core/sock.c        |   2 +-
 net/mptcp/mptcp_diag.c |   2 +-
 net/mptcp/options.c    |  47 +--
 net/mptcp/protocol.c   | 733 ++++++++++++++++++++++++++++++-----------
 net/mptcp/protocol.h   |  34 +-
 net/mptcp/subflow.c    |  14 +-
 7 files changed, 598 insertions(+), 235 deletions(-)

Comments

Jakub Kicinski Dec. 1, 2020, 2:34 a.m. UTC | #1

On Fri, 27 Nov 2020 11:10:21 +0100 Paolo Abeni wrote:
> The current locking schema used to protect the MPTCP data-path
> requires the usage of the MPTCP workqueue to process the incoming
> data, depending on trylock result.
> 
> The above poses scalability limits and introduces random delays
> in MPTCP-level acks.
> 
> With this series we use a single spinlock to protect the MPTCP
> data-path, removing the need for workqueue and delayed ack usage.
> 
> This additionally reduces the number of atomic operations required
> per packet and cleans-up considerably the poll/wake-up code.

Applied, thanks!