Re: [PATCH 7.0.y] mptcp: update window_clamp on subflows when SO_RCVBUF is set

[Matthieu Baerts <matttbe@kernel.org>]

Hi Sasha,

On 01/06/2026 10:39, Sasha Levin wrote:
> From: Gang Yan <yangang@kylinos.cn>
> 
> [ Upstream commit 3a543ae0e2092d5c2085d5f21f7a7dbafdffea3c ]
> 
> Add __mptcp_subflow_set_rcvbuf() helper to write the subflow sk_rcvbuf,
> but also to call the recently added tcp_set_rcvbuf() helper to update
> window_clamp. This is needed because the window clap is updated when
> scaling_ratio changes, in tcp_measure_rcv_mss(). Until scaling_ratio
> changes, the subflow is stuck with the old window clamp which may be
> based on a small initial buffer.
> 
> Use this new helper in both mptcp_sol_socket_sync_intval() (setsockopt
> path) and sync_socket_options() (new subflow creation path).
> 
> Note that this patch depends on commit b025461303d8 ("tcp: update
> window_clamp when SO_RCVBUF is set"): it fixes the issue on TCP side,
> but the same fix is needed on MPTCP side as well.
Thank you for the backport. I guess there is a conflict, because commit
b025461303d8 ("tcp: update window_clamp when SO_RCVBUF is set") has not
been backported to 7.0 (nor 6.18, 6.12 and 6.6). See the note above for
more details.

In other words, MPTCP is imitating TCP's behaviour. So no need to
backport this patch if the fix on TCP side is not backported either.

I think it would be better to take both the TCP fix (b025461303d8), and
this one (3a543ae0e209). But maybe the TCP fix has not been backported
on purpose?

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