diff options
| author | Paolo Abeni <[email protected]> | 2023-10-23 13:44:41 -0700 |
|---|---|---|
| committer | Jakub Kicinski <[email protected]> | 2023-10-25 12:23:34 -0700 |
| commit | 9fdc779331bd3e0b2570c2bced396d53a04a33b9 (patch) | |
| tree | 6bff7ca9d002a84a3c4f727384939ab386ff0d6b | |
| parent | a1ab24e5fc4a3048a1b3a24ab0ddc7b73358baa9 (diff) | |
mptcp: ignore notsent_lowat setting at the subflow level
Any latency related tuning taking action at the subflow level does
not really affect the user-space, as only the main MPTCP socket is
relevant.
Anyway any limiting setting may foul the MPTCP scheduler, not being
able to fully use the subflow-level cwin, leading to very poor b/w
usage.
Enforce notsent_lowat to be a no-op on every subflow.
Note that TCP_NOTSENT_LOWAT is currently not supported, and properly
dealing with that will require more invasive changes.
Reviewed-by: Mat Martineau <[email protected]>
Signed-off-by: Paolo Abeni <[email protected]>
Signed-off-by: Mat Martineau <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>
| -rw-r--r-- | net/mptcp/sockopt.c | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index abf0645cb65d..72858d7d8974 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -1452,6 +1452,12 @@ void mptcp_sockopt_sync_locked(struct mptcp_sock *msk, struct sock *ssk) ssk->sk_rcvlowat = 0; + /* subflows must ignore any latency-related settings: will not affect + * the user-space - only the msk is relevant - but will foul the + * mptcp scheduler + */ + tcp_sk(ssk)->notsent_lowat = UINT_MAX; + if (READ_ONCE(subflow->setsockopt_seq) != msk->setsockopt_seq) { sync_socket_options(msk, ssk); |