[v2] mptcp: address stall under memory pressure

[PATCH v2 mptcp-next 0/7] mptcp: address stall under memory pressure

Paolo Abeni posted 7 patches 1 week, 2 days ago

Diff against v1 v1 v1 v1 v2 v3 v5 v6 v7 v8 v9 v10 v11 v12
Download series mbox

Failed in applying to current master (apply log)

There is a newer version of this series

net/mptcp/mib.c      |   3 +
net/mptcp/mib.h      |   3 +
net/mptcp/options.c  |  32 ++++-
net/mptcp/pm.c       |  42 ++++---
net/mptcp/protocol.c | 290 ++++++++++++++++++++++++++++++-------------
net/mptcp/protocol.h |  11 +-
6 files changed, 273 insertions(+), 108 deletions(-)

Expand all Fold all

[PATCH v2 mptcp-next 0/7] mptcp: address stall under memory pressure

Posted by Paolo Abeni 1 week, 2 days ago

This *should* address opus-5 feedback, as observed upstream.

New iteration takes care of gemini review on v1.

Paolo Abeni (7):
  mptcp: move the retrans loop to a separate helper
  mptcp: move the stale logic out of retrans scheduler
  mptcp: let the retrans scheduler do its job
  mptcp: explicitly drop over memory limits
  mptcp: enforce hard limit on backlog flushing
  mptcp: avoid code duplication in __mptcp_move_skb()
  mptcp: implemented OoO queue pruning

 net/mptcp/mib.c      |   3 +
 net/mptcp/mib.h      |   3 +
 net/mptcp/options.c  |  32 ++++-
 net/mptcp/pm.c       |  42 ++++---
 net/mptcp/protocol.c | 290 ++++++++++++++++++++++++++++++-------------
 net/mptcp/protocol.h |  11 +-
 6 files changed, 273 insertions(+), 108 deletions(-)

-- 
2.55.0

Re: [PATCH v2 mptcp-next 0/7] mptcp: address stall under memory pressure

Posted by Matthieu Baerts 1 week, 1 day ago

Hi Paolo,

On 06/08/2026 16:45, Paolo Abeni wrote:
> This *should* address opus-5 feedback, as observed upstream.
Thank you for the new version, and your patience with AI reviews!
Hopefully opus-5 will not be even more picky!

This new version looks good to me:

Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>


New patches for t/upstream:
- b03efee5f28b: mptcp: move the retrans loop to a separate helper
- 23b698046768: mptcp: move the stale logic out of retrans scheduler
- 1a2614f8a410: mptcp: let the retrans scheduler do its job
- 568cc4fa11f2: mptcp: explicitly drop over memory limits
- c29573ac23e6: mptcp: enforce hard limit on backlog flushing
- e76d982954d3: mptcp: avoid code duplication in __mptcp_move_skb()
- b746103fb892: mptcp: implemented OoO queue pruning
- Results: 775addc43266..7bf94a8339a9 (export)

(+ Geliang's fixes for the BPF selftests)

Tests are now in progress:

- export: https://github.com/multipath-tcp/mptcp_net-next/commit/58dcc35549b1dfcf3d689fc5e163a25fb11a335c/checks

Cheers,
Matt
-- 
Sponsored by the NGI0 Core fund.