[ Upstream commit 
ee50e67ba0e17b1a1a8d76691d02eadf9e0f392c ]
To compute the rtx timeout schedule_3rdack_retransmission() does multiple
things in the wrong way: srtt_us is measured in usec/8 and the timeout
itself is an absolute value.
Fixes: ec3edaa7ca6ce02f ("mptcp: Add handling of outgoing MP_JOIN requests")
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Mat Martineau <mathew.j.martineau>@linux.intel.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
 
        /* 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;