[SCTP]: Force update of the rto when processing HB-ACK
Vlad Yasevich [Tue, 30 Jan 2007 22:36:14 +0000 (14:36 -0800)]
When processing a HEARTBEAT-ACK it's possible that the transport rto
timers will not be updated because a prior T3-RTX processing would
have cleared the rto_pending flag on the transport.  However, if
we received a valid HEARTBEAT-ACK, we want to force update the
rto variables, so re-set the rto_pending flag before calling
sctp_transport_update_rto().

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/sctp/sm_sideeffect.c

index 8bd3097..6db77d1 100644 (file)
@@ -621,7 +621,13 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds,
        /* The receiver of the HEARTBEAT ACK should also perform an
         * RTT measurement for that destination transport address
         * using the time value carried in the HEARTBEAT ACK chunk.
+        * If the transport's rto_pending variable has been cleared,
+        * it was most likely due to a retransmit.  However, we want
+        * to re-enable it to properly update the rto.
         */
+       if (t->rto_pending == 0)
+               t->rto_pending = 1;
+
        hbinfo = (sctp_sender_hb_info_t *) chunk->skb->data;
        sctp_transport_update_rto(t, (jiffies - hbinfo->sent_at));