1 files changed, 44 insertions, 42 deletions

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 2877c3e..53c8ce40 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -105,7 +105,6 @@ int sysctl_tcp_abc __read_mostly;
 #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) */
@@ -1040,13 +1039,11 @@ static void tcp_skb_mark_lost_uncond_verify(struct tcp_sock *tp,
  * 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.
  *
@@ -1153,7 +1150,7 @@ static int tcp_is_sackblock_valid(struct tcp_sock *tp, int is_dsack,
 }
 
 /* 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
@@ -1310,25 +1307,26 @@ static int tcp_match_skb_to_sack(struct sock *sk, struct sk_buff *skb,
 	return in_sack;
 }
 
-static u8 tcp_sacktag_one(const struct sk_buff *skb, struct sock *sk,
-			  struct tcp_sacktag_state *state,
+/* Mark the given newly-SACKed range as such, adjusting counters and hints. */
+static u8 tcp_sacktag_one(struct sock *sk,
+			  struct tcp_sacktag_state *state, u8 sacked,
+			  u32 start_seq, u32 end_seq,
 			  int dup_sack, int pcount)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
-	u8 sacked = TCP_SKB_CB(skb)->sacked;
 	int fack_count = state->fack_count;
 
 	/* Account D-SACK for retransmitted packet. */
 	if (dup_sack && (sacked & TCPCB_RETRANS)) {
 		if (tp->undo_marker && tp->undo_retrans &&
-		    after(TCP_SKB_CB(skb)->end_seq, tp->undo_marker))
+		    after(end_seq, tp->undo_marker))
 			tp->undo_retrans--;
 		if (sacked & TCPCB_SACKED_ACKED)
 			state->reord = min(fack_count, state->reord);
 	}
 
 	/* Nothing to do; acked frame is about to be dropped (was ACKed). */
-	if (!after(TCP_SKB_CB(skb)->end_seq, tp->snd_una))
+	if (!after(end_seq, tp->snd_una))
 		return sacked;
 
 	if (!(sacked & TCPCB_SACKED_ACKED)) {
@@ -1347,13 +1345,13 @@ static u8 tcp_sacktag_one(const struct sk_buff *skb, struct sock *sk,
 				/* New sack for not retransmitted frame,
 				 * which was in hole. It is reordering.
 				 */
-				if (before(TCP_SKB_CB(skb)->seq,
+				if (before(start_seq,
 					   tcp_highest_sack_seq(tp)))
 					state->reord = min(fack_count,
 							   state->reord);
 
 				/* SACK enhanced F-RTO (RFC4138; Appendix B) */
-				if (!after(TCP_SKB_CB(skb)->end_seq, tp->frto_highmark))
+				if (!after(end_seq, tp->frto_highmark))
 					state->flag |= FLAG_ONLY_ORIG_SACKED;
 			}
 
@@ -1371,8 +1369,7 @@ static u8 tcp_sacktag_one(const struct sk_buff *skb, struct sock *sk,
 
 		/* Lost marker hint past SACKed? Tweak RFC3517 cnt */
 		if (!tcp_is_fack(tp) && (tp->lost_skb_hint != NULL) &&
-		    before(TCP_SKB_CB(skb)->seq,
-			   TCP_SKB_CB(tp->lost_skb_hint)->seq))
+		    before(start_seq, TCP_SKB_CB(tp->lost_skb_hint)->seq))
 			tp->lost_cnt_hint += pcount;
 
 		if (fack_count > tp->fackets_out)
@@ -1391,6 +1388,9 @@ static u8 tcp_sacktag_one(const struct sk_buff *skb, struct sock *sk,
 	return sacked;
 }
 
+/* Shift newly-SACKed bytes from this skb to the immediately previous
+ * already-SACKed sk_buff. Mark the newly-SACKed bytes as such.
+ */
 static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb,
 			   struct tcp_sacktag_state *state,
 			   unsigned int pcount, int shifted, int mss,
@@ -1398,10 +1398,13 @@ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb,
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *prev = tcp_write_queue_prev(sk, skb);
+	u32 start_seq = TCP_SKB_CB(skb)->seq;	/* start of newly-SACKed */
+	u32 end_seq = start_seq + shifted;	/* end of newly-SACKed */
 
 	BUG_ON(!pcount);
 
-	if (skb == tp->lost_skb_hint)
+	/* Adjust hint for FACK. Non-FACK is handled in tcp_sacktag_one(). */
+	if (tcp_is_fack(tp) && (skb == tp->lost_skb_hint))
 		tp->lost_cnt_hint += pcount;
 
 	TCP_SKB_CB(prev)->end_seq += shifted;
@@ -1427,8 +1430,11 @@ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb,
 		skb_shinfo(skb)->gso_type = 0;
 	}
 
-	/* We discard results */
-	tcp_sacktag_one(skb, sk, state, dup_sack, pcount);
+	/* Adjust counters and hints for the newly sacked sequence range but
+	 * discard the return value since prev is already marked.
+	 */
+	tcp_sacktag_one(sk, state, TCP_SKB_CB(skb)->sacked,
+			start_seq, end_seq, dup_sack, pcount);
 
 	/* Difference in this won't matter, both ACKed by the same cumul. ACK */
 	TCP_SKB_CB(prev)->sacked |= (TCP_SKB_CB(skb)->sacked & TCPCB_EVER_RETRANS);
@@ -1667,10 +1673,14 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk,
 			break;
 
 		if (in_sack) {
-			TCP_SKB_CB(skb)->sacked = tcp_sacktag_one(skb, sk,
-								  state,
-								  dup_sack,
-								  tcp_skb_pcount(skb));
+			TCP_SKB_CB(skb)->sacked =
+				tcp_sacktag_one(sk,
+						state,
+						TCP_SKB_CB(skb)->sacked,
+						TCP_SKB_CB(skb)->seq,
+						TCP_SKB_CB(skb)->end_seq,
+						dup_sack,
+						tcp_skb_pcount(skb));
 
 			if (!before(TCP_SKB_CB(skb)->seq,
 				    tcp_highest_sack_seq(tp)))
@@ -1844,10 +1854,6 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb,
 		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))
@@ -2515,8 +2521,11 @@ static void tcp_timeout_skbs(struct sock *sk)
 	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)
 {
@@ -2525,6 +2534,8 @@ 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) {
@@ -2546,7 +2557,7 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head)
 		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;
@@ -3033,19 +3044,10 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked,
 	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);
@@ -3077,7 +3079,7 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked,
 		}
 	}
 
-	/* F. Process state. */
+	/* E. Process state. */
 	switch (icsk->icsk_ca_state) {
 	case TCP_CA_Recovery:
 		if (!(flag & FLAG_SND_UNA_ADVANCED)) {