tcp: detect loss above high_seq in recovery

Correctly implement a loss detection heuristic: New sequences (above
high_seq) sent during the fast recovery are deemed lost when higher
sequences are SACKed.

Current code does not catch these losses, because tcp_mark_head_lost()
does not check packets beyond high_seq. The fix is straight-forward by
checking packets until the highest sacked packet. In addition, all the
FLAG_DATA_LOST logic are in-effective and redundant and can be removed.

Update the loss heuristic comments. The algorithm above is documented
as heuristic B, but it is redundant too because heuristic A already
covers B.

Note that this change only marks some forward-retransmitted packets LOST.
It does NOT forbid TCP performing further CWR on new losses. A potential
follow-up patch under preparation is to perform another CWR on "new"
losses such as
1) sequence above high_seq is lost (by resetting high_seq to snd_nxt)
2) retransmission is lost.

Signed-off-by: Yuchung Cheng <[e-mail]>
Signed-off-by: David S. Miller <[e-mail]>

include/linux/snmp.h

···
 	LINUX_MIB_TCPPARTIALUNDO,		/* TCPPartialUndo */
 	LINUX_MIB_TCPDSACKUNDO,			/* TCPDSACKUndo */
 	LINUX_MIB_TCPLOSSUNDO,			/* TCPLossUndo */
-	LINUX_MIB_TCPLOSS,			/* TCPLoss */
 	LINUX_MIB_TCPLOSTRETRANSMIT,		/* TCPLostRetransmit */
 	LINUX_MIB_TCPRENOFAILURES,		/* TCPRenoFailures */
 	LINUX_MIB_TCPSACKFAILURES,		/* TCPSackFailures */

net/ipv4/proc.c

···
 	SNMP_MIB_ITEM("TCPPartialUndo", LINUX_MIB_TCPPARTIALUNDO),
 	SNMP_MIB_ITEM("TCPDSACKUndo", LINUX_MIB_TCPDSACKUNDO),
 	SNMP_MIB_ITEM("TCPLossUndo", LINUX_MIB_TCPLOSSUNDO),
-	SNMP_MIB_ITEM("TCPLoss", LINUX_MIB_TCPLOSS),
 	SNMP_MIB_ITEM("TCPLostRetransmit", LINUX_MIB_TCPLOSTRETRANSMIT),
 	SNMP_MIB_ITEM("TCPRenoFailures", LINUX_MIB_TCPRENOFAILURES),
 	SNMP_MIB_ITEM("TCPSackFailures", LINUX_MIB_TCPSACKFAILURES),

net/ipv4/tcp_input.c

···
 #define FLAG_SYN_ACKED		0x10 /* This ACK acknowledged SYN.		*/
 #define FLAG_DATA_SACKED	0x20 /* New SACK.				*/
 #define FLAG_ECE		0x40 /* ECE in this ACK				*/
-#define FLAG_DATA_LOST		0x80 /* SACK detected data lossage.		*/
 #define FLAG_SLOWPATH		0x100 /* Do not skip RFC checks for window update.*/
 #define FLAG_ONLY_ORIG_SACKED	0x200 /* SACKs only non-rexmit sent before RTO */
 #define FLAG_SND_UNA_ADVANCED	0x400 /* Snd_una was changed (!= FLAG_DATA_ACKED) */
···
  * These 6 states form finite state machine, controlled by the following events:
  * 1. New ACK (+SACK) arrives. (tcp_sacktag_write_queue())
  * 2. Retransmission. (tcp_retransmit_skb(), tcp_xmit_retransmit_queue())
- * 3. Loss detection event of one of three flavors:
+ * 3. Loss detection event of two flavors:
  *	A. Scoreboard estimator decided the packet is lost.
  *	   A'. Reno "three dupacks" marks head of queue lost.
- *	   A''. Its FACK modfication, head until snd.fack is lost.
- *	B. SACK arrives sacking data transmitted after never retransmitted
- *	   hole was sent out.
- *	C. SACK arrives sacking SND.NXT at the moment, when the
+ *	   A''. Its FACK modification, head until snd.fack is lost.
+ *	B. SACK arrives sacking SND.NXT at the moment, when the
  *	   segment was retransmitted.
  * 4. D-SACK added new rule: D-SACK changes any tag to S.
  *
···
 }
 
 /* Check for lost retransmit. This superb idea is borrowed from "ratehalving".
- * Event "C". Later note: FACK people cheated me again 8), we have to account
+ * Event "B". Later note: FACK people cheated me again 8), we have to account
  * for reordering! Ugly, but should help.
  *
  * Search retransmitted skbs from write_queue that were sent when snd_nxt was
···
 		if (found_dup_sack && ((i + 1) == first_sack_index))
 			next_dup = &sp[i + 1];
 
-		/* Event "B" in the comment above. */
-		if (after(end_seq, tp->high_seq))
-			state.flag |= FLAG_DATA_LOST;
-
 		/* Skip too early cached blocks */
 		while (tcp_sack_cache_ok(tp, cache) &&
 		       !before(start_seq, cache->end_seq))
···
 	tcp_verify_left_out(tp);
 }
 
-/* Mark head of queue up as lost. With RFC3517 SACK, the packets is
- * is against sacked "cnt", otherwise it's against facked "cnt"
+/* Detect loss in event "A" above by marking head of queue up as lost.
+ * For FACK or non-SACK(Reno) senders, the first "packets" number of segments
+ * are considered lost. For RFC3517 SACK, a segment is considered lost if it
+ * has at least tp->reordering SACKed seqments above it; "packets" refers to
+ * the maximum SACKed segments to pass before reaching this limit.
  */
 static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head)
 {
···
 	int cnt, oldcnt;
 	int err;
 	unsigned int mss;
+	/* Use SACK to deduce losses of new sequences sent during recovery */
+	const u32 loss_high = tcp_is_sack(tp) ?  tp->snd_nxt : tp->high_seq;
 
 	WARN_ON(packets > tp->packets_out);
 	if (tp->lost_skb_hint) {
···
 		tp->lost_skb_hint = skb;
 		tp->lost_cnt_hint = cnt;
 
-		if (after(TCP_SKB_CB(skb)->end_seq, tp->high_seq))
+		if (after(TCP_SKB_CB(skb)->end_seq, loss_high))
 			break;
 
 		oldcnt = cnt;
···
 	if (tcp_check_sack_reneging(sk, flag))
 		return;
 
-	/* C. Process data loss notification, provided it is valid. */
-	if (tcp_is_fack(tp) && (flag & FLAG_DATA_LOST) &&
-	    before(tp->snd_una, tp->high_seq) &&
-	    icsk->icsk_ca_state != TCP_CA_Open &&
-	    tp->fackets_out > tp->reordering) {
-		tcp_mark_head_lost(sk, tp->fackets_out - tp->reordering, 0);
-		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSS);
-	}
-
-	/* D. Check consistency of the current state. */
+	/* C. Check consistency of the current state. */
 	tcp_verify_left_out(tp);
 
-	/* E. Check state exit conditions. State can be terminated
+	/* D. Check state exit conditions. State can be terminated
 	 *    when high_seq is ACKed. */
 	if (icsk->icsk_ca_state == TCP_CA_Open) {
 		WARN_ON(tp->retrans_out != 0);
···
 		}
 	}
 
-	/* F. Process state. */
+	/* E. Process state. */
 	switch (icsk->icsk_ca_state) {
 	case TCP_CA_Recovery:
 		if (!(flag & FLAG_SND_UNA_ADVANCED)) {