Commit dd929c1b authored Apr 08, 2015 by Eric Dumazet Committed by David S. Miller Apr 08, 2015

tcp: do not rearm rsk_timer on FastOpen requests



FastOpen requests are not like other regular request sockets.

They do not yet use rsk_timer : tcp_fastopen_queue_check()
simply manually removes one expired request from fastopenq->rskq_rst
list.

Therefore, tcp_check_req() must not call mod_timer_pending(),
otherwise we crash because rsk_timer was not initialized.

Fixes: fa76ce73 ("inet: get rid of central tcp/dccp listener timer")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

parent 82740b9a

net/ipv4/tcp_minisocks.c

+10 −4

Original line number	Diff line number	Diff line
		@@ -628,10 +628,16 @@ struct sock tcp_check_req(struct sock sk, struct sk_buff *skb,
		LINUX_MIB_TCPACKSKIPPEDSYNRECV,
		&tcp_rsk(req)->last_oow_ack_time) &&

		!inet_rtx_syn_ack(sk, req))
		mod_timer_pending(&req->rsk_timer, jiffies +
		min(TCP_TIMEOUT_INIT << req->num_timeout,
		TCP_RTO_MAX));
		!inet_rtx_syn_ack(sk, req)) {
		unsigned long expires = jiffies;

		expires += min(TCP_TIMEOUT_INIT << req->num_timeout,
		TCP_RTO_MAX);
		if (!fastopen)
		mod_timer_pending(&req->rsk_timer, expires);
		else
		req->rsk_timer.expires = expires;
		}
		return NULL;
		}

Admin message