On Tue, 2 Nov 2021, Paolo Abeni wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> To compute the rtx timeout schedule_3rdack_retransmission() does multiple
> things in the wrong way: srtt_us is measures in usec/8 and the timeout
> itself is an absolute vale.
>
> Fixes: ec3edaa7ca6ce02f ("mptcp: Add handling of outgoing MP_JOIN requests")
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> ---
> commit message typos added by me
Thanks for the v2 Paolo (both patches). I think they're good for applying
to the export branch, maybe Matthieu can s/measures/measured/ and
s/vale/value/ above to restore Eric's spelling :)
Reviewed-by: Mat Martineau <mathew.j.martineau>@linux.intel.com>
> ---
> net/mptcp/options.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/net/mptcp/options.c b/net/mptcp/options.c
> index 7c3420afb1a0..2ca077fdf10a 100644
> --- a/net/mptcp/options.c
> +++ b/net/mptcp/options.c
> @@ -434,9 +434,10 @@ static void schedule_3rdack_retransmission(struct sock *sk)
>
> /* reschedule with a timeout above RTT, as we must look only for drop */
> if (tp->srtt_us)
> - timeout = tp->srtt_us << 1;
> + timeout = usecs_to_jiffies(tp->srtt_us >> (3-1));
> else
> timeout = TCP_TIMEOUT_INIT;
> + timeout += jiffies;
>
> WARN_ON_ONCE(icsk->icsk_ack.pending & ICSK_ACK_TIMER);
> icsk->icsk_ack.pending |= ICSK_ACK_SCHED | ICSK_ACK_TIMER;
> --
> 2.26.3
>
>
>
--
Mat Martineau
Intel