linux-kernel - [PATCH net-next 0/4] mptcp: add TCP_NOTSENT

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20240301-upstream-net-next-20240301-mptcp-tcp_notsent_lowat-v1-0-415f0e8ed0e1@kernel.org>
Date: Fri, 01 Mar 2024 18:43:43 +0100
From: "Matthieu Baerts (NGI0)" <matttbe@...nel.org>
To: mptcp@...ts.linux.dev, Mat Martineau <martineau@...nel.org>, 
 Geliang Tang <geliang@...nel.org>, "David S. Miller" <davem@...emloft.net>, 
 Eric Dumazet <edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>, 
 Paolo Abeni <pabeni@...hat.com>
Cc: netdev@...r.kernel.org, linux-kernel@...r.kernel.org, 
 "Matthieu Baerts (NGI0)" <matttbe@...nel.org>
Subject: [PATCH net-next 0/4] mptcp: add TCP_NOTSENT_LOWAT sockopt support

Patch 3 does the magic of adding TCP_NOTSENT_LOWAT support, all the
other ones are minor cleanup seen along when working on the new feature.

Note that this feature relies on the existing accounting for snd_nxt.
Such accounting is not 110% accurate as it tracks the most recent
sequence number queued to any subflow, and not the actual sequence
number sent on the wire. Paolo experimented a lot, trying to implement
the latter, and in the end it proved to be both "too complex" and "not
necessary".

The complexity raises from the need for additional lock and a lot of
refactoring to introduce such protections without adding significant
overhead. Additionally, snd_nxt is currently used and exposed with the
current semantic by the internal packet scheduling. Introducing a
different tracking will still require us to keep the old one.

More interestingly, a more accurate tracking could be not strictly
necessary: as the MPTCP socket enqueues data to the subflows only up to
the available send window, any enqueue data is sent on the wire
instantly, without any blocking operation short or a drop in the tx path
at the nft or TC layer.

Signed-off-by: Matthieu Baerts (NGI0) <matttbe@...nel.org>
---
Paolo Abeni (4):
      mptcp: cleanup writer wake-up
      mptcp: avoid some duplicate code in socket option handling
      mptcp: implement TCP_NOTSENT_LOWAT support
      mptcp: cleanup SOL_TCP handling

 net/mptcp/protocol.c | 54 ++++++++++++++++++++++++++-------------
 net/mptcp/protocol.h | 42 +++++++++++++++++++++++--------
 net/mptcp/sockopt.c  | 71 +++++++++++++++++++++++-----------------------------
 3 files changed, 101 insertions(+), 66 deletions(-)
---
base-commit: e960825709330cb199d209740326cec37e8c419d
change-id: 20240301-upstream-net-next-20240301-mptcp-tcp_notsent_lowat-770cab93d253

Best regards,
-- 
Matthieu Baerts (NGI0) <matttbe@...nel.org>