MPTCP Linux Development
 help / color / mirror / Atom feed
From: Matthieu Baerts <matttbe@kernel.org>
To: Paolo Abeni <pabeni@redhat.com>, mptcp@lists.linux.dev
Subject: Re: [PATCH mptcp-net] mptcp: fix ack generation for fallback msk
Date: Tue, 11 Nov 2025 16:47:12 +0100	[thread overview]
Message-ID: <db7a5754-c4de-48d2-9cf8-892a0d4db38e@kernel.org> (raw)
In-Reply-To: <fa8316ac-594e-443b-b9dc-00103ee8557a@kernel.org>

Hi Paolo,

On 11/11/2025 16:30, Matthieu Baerts wrote:
> Hi Paolo,
> 
> On 07/11/2025 01:10, Paolo Abeni wrote:
>> mptcp_cleanup_rbuf() need to know the last most recent, mptcp-level
>> rcv_wnd sent, and such information is tracked into the msk->old_wspace
>> field, updated at ack transmission time by mptcp_write_options().
>>
>> Fallback socket do not add any mptcp options, such helper is never
>> invoked, and msk->old_wspace value remain stale. That in turn makes
>> ack generation at recvmsg() time quite randomic.
>>
>> Address the issue ensuring mptcp_write_options() is invoked even
>> for fallback sockets, and just update the needed info in such a
>> case.
>>
>> The issue went unnoticed for a long time, as mptcp overshot the> fallback socket receive buffer autotune significantly.
> 
> Should we say "used to overshot"? :)
> 
>> Fixes: e3859603ba13 ("mptcp: better msk receive window updates")
>> Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/594
> 
> Thank you for the fix!
> 
> (...)
> 
>> @@ -1643,6 +1660,9 @@ void mptcp_write_options(struct tcphdr *th, __be32 *ptr, struct tcp_sock *tp,
>>  				      opts->reset_transient,
>>  				      opts->reset_reason);
>>  		return;
>> +	} else if (unlikely(!opts->suboptions)) {
> 
> Small detail: do you mind if I add /* Fallback to TCP */ just above when
> applying the patch?
> 
>> +		mptcp_track_rwin(tp);
> 
> Should we eventually also call mptcp_track_rwin(tp) in case of MP_FAIL
> without MP_RST? Or is this not needed?
> 
> Other than that, it looks good to me, and I guess we can already apply
> the patch:

That's what I just did: I guess the modification for MP_FAIL, if really
needed, can still be added later on.

New patches for t/upstream-net and t/upstream:
- c3a571ff59cc: mptcp: fix ack generation for fallback msk
- Results: 751a2432b7d8..5b6b4e3aef4f (export-net)
- Results: 247540d77d20..a96affe5f0a1 (export)

Tests are now in progress:

- export-net:
https://github.com/multipath-tcp/mptcp_net-next/commit/7fade8917eeabe227b5ffae540e9cbe48d3e5046/checks
- export:
https://github.com/multipath-tcp/mptcp_net-next/commit/b10c1d3d1e5da09bbc2b81eca8f4339b84499268/checks

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


      reply	other threads:[~2025-11-11 15:47 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-07  0:10 [PATCH mptcp-net] mptcp: fix ack generation for fallback msk Paolo Abeni
2025-11-07  8:41 ` MPTCP CI
2025-11-11  9:09 ` Geliang Tang
2025-11-11 15:30 ` Matthieu Baerts
2025-11-11 15:47   ` Matthieu Baerts [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=db7a5754-c4de-48d2-9cf8-892a0d4db38e@kernel.org \
    --to=matttbe@kernel.org \
    --cc=mptcp@lists.linux.dev \
    --cc=pabeni@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox