TCP reuses t_rxtshift to determine the backoff timer used for both the

persist state and the retransmit timer.  However, the code that implements
"bad retransmit recovery" only checks t_rxtshift to see if an ACK has been
received in during the first retransmit timeout window.  As a result, if
ticks has wrapped over to a negative value and a socket is in the persist
state, it can incorrectly treat an ACK from the remote peer as a
"bad retransmit recovery" and restore saved values such as snd_ssthresh and
snd_cwnd.  However, if the socket has never had a retransmit timeout, then
these saved values will be zero, so snd_ssthresh and snd_cwnd will be set
to 0.

If the socket is in fast recovery (this can be caused by excessive
duplicate ACKs such as those fixed by 220794), then each ACK that arrives
triggers either NewReno or SACK partial ACK handling which clamps snd_cwnd
to be no larger than snd_ssthresh.  In effect, the socket's send window
is permamently stuck at 0 even though the remote peer is advertising a
much larger window and pending data is only sent via TCP window probes
(so one byte every few seconds).

Fix this by adding a new TCP pcb flag (TF_PREVVALID) that indicates that
the various snd_*_prev fields in the pcb are valid and only perform
"bad retransmit recovery" if this flag is set in the pcb.  The flag is set
on the first retransmit timeout that occurs and is cleared on subsequent
retransmit timeouts or when entering the persist state.

Reviewed by:	bz
MFC after:	2 weeks

4 files changed, 9 insertions, 2 deletions

diff --git a/sys/netinet/tcp_input.c b/sys/netinet/tcp_input.c
index cf1daa8..30e8429 100644
--- a/sys/netinet/tcp_input.c
+++ b/sys/netinet/tcp_input.c
@@ -420,6 +420,7 @@ cc_cong_signal(struct tcpcb *tp, struct tcphdr *th, uint32_t type)
 		if (tp->t_flags & TF_WASCRECOVERY)
 			ENTER_CONGRECOVERY(tp->t_flags);
 		tp->snd_nxt = tp->snd_max;
+		tp->t_flags &= ~TF_PREVVALID;
 		tp->t_badrxtwin = 0;
 		break;
 	}
@@ -1575,6 +1576,7 @@ tcp_do_segment(struct mbuf *m, struct tcphdr *th, struct socket *so,
 				 * "bad retransmit" recovery.
 				 */
 				if (tp->t_rxtshift == 1 &&
+				    tp->t_flags & TF_PREVVALID &&
 				    (int)(ticks - tp->t_badrxtwin) < 0) {
 					cc_cong_signal(tp, th, CC_RTO_ERR);
 				}
@@ -2521,7 +2523,8 @@ process_ACK:
 		 * original cwnd and ssthresh, and proceed to transmit where
 		 * we left off.
 		 */
-		if (tp->t_rxtshift == 1 && (int)(ticks - tp->t_badrxtwin) < 0)
+		if (tp->t_rxtshift == 1 && tp->t_flags & TF_PREVVALID &&
+		    (int)(ticks - tp->t_badrxtwin) < 0)
 			cc_cong_signal(tp, th, CC_RTO_ERR);
 
 		/*
diff --git a/sys/netinet/tcp_output.c b/sys/netinet/tcp_output.c
index dc2e5ca..f6488a3 100644
--- a/sys/netinet/tcp_output.c
+++ b/sys/netinet/tcp_output.c
@@ -1346,6 +1346,7 @@ tcp_setpersist(struct tcpcb *tp)
 	int t = ((tp->t_srtt >> 2) + tp->t_rttvar) >> 1;
 	int tt;
 
+	tp->t_flags &= ~TF_PREVVALID;
 	if (tcp_timer_active(tp, TT_REXMT))
 		panic("tcp_setpersist: retransmit pending");
 	/*
diff --git a/sys/netinet/tcp_timer.c b/sys/netinet/tcp_timer.c
index 6d0fa1c..5c2c5c2 100644
--- a/sys/netinet/tcp_timer.c
+++ b/sys/netinet/tcp_timer.c
@@ -524,7 +524,9 @@ tcp_timer_rexmt(void * xtp)
 		else
 			tp->t_flags &= ~TF_WASCRECOVERY;
 		tp->t_badrxtwin = ticks + (tp->t_srtt >> (TCP_RTT_SHIFT + 1));
-	}
+		tp->t_flags |= TF_PREVVALID;
+	} else
+		tp->t_flags &= ~TF_PREVVALID;
 	TCPSTAT_INC(tcps_rexmttimeo);
 	if (tp->t_state == TCPS_SYN_SENT)
 		rexmt = TCP_REXMTVAL(tp) * tcp_syn_backoff[tp->t_rxtshift];
diff --git a/sys/netinet/tcp_var.h b/sys/netinet/tcp_var.h
index 5d92d87..e8189b0 100644
--- a/sys/netinet/tcp_var.h
+++ b/sys/netinet/tcp_var.h
@@ -224,6 +224,7 @@ struct tcpcb {
 #define	TF_NEEDSYN	0x000400	/* send SYN (implicit state) */
 #define	TF_NEEDFIN	0x000800	/* send FIN (implicit state) */
 #define	TF_NOPUSH	0x001000	/* don't push */
+#define	TF_PREVVALID	0x002000	/* saved values for bad rxmit valid */
 #define	TF_MORETOCOME	0x010000	/* More data to be appended to sock */
 #define	TF_LQ_OVERFLOW	0x020000	/* listen queue overflow */
 #define	TF_LASTIDLE	0x040000	/* connection was previously idle */