tcp_input.c

		tcp_enter_loss(sk, 1);
		icsk->icsk_retransmits++;
		tcp_retransmit_skb(sk, tcp_write_queue_head(sk));
		inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
					  icsk->icsk_rto, TCP_RTO_MAX);
		return 1;
	}
	return 0;
}

static inline int tcp_fackets_out(struct tcp_sock *tp)
{
	return tcp_is_reno(tp) ? tp->sacked_out + 1 : tp->fackets_out;
}

/* Heurestics to calculate number of duplicate ACKs. There's no dupACKs
 * counter when SACK is enabled (without SACK, sacked_out is used for
 * that purpose).
 *
 * Instead, with FACK TCP uses fackets_out that includes both SACKed
 * segments up to the highest received SACK block so far and holes in
 * between them.
 *
 * With reordering, holes may still be in flight, so RFC3517 recovery
 * uses pure sacked_out (total number of SACKed segments) even though
 * it violates the RFC that uses duplicate ACKs, often these are equal
 * but when e.g. out-of-window ACKs or packet duplication occurs,
 * they differ. Since neither occurs due to loss, TCP should really
 * ignore them.
 */
static inline int tcp_dupack_heurestics(struct tcp_sock *tp)
{
	return tcp_is_fack(tp) ? tp->fackets_out : tp->sacked_out + 1;
}

static inline int tcp_skb_timedout(struct sock *sk, struct sk_buff *skb)
{
	return (tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto);
}

static inline int tcp_head_timedout(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);

	return tp->packets_out &&
	       tcp_skb_timedout(sk, tcp_write_queue_head(sk));
}

/* Linux NewReno/SACK/FACK/ECN state machine.
 * --------------------------------------
 *
 * "Open"	Normal state, no dubious events, fast path.
 * "Disorder"   In all the respects it is "Open",
 *		but requires a bit more attention. It is entered when
 *		we see some SACKs or dupacks. It is split of "Open"
 *		mainly to move some processing from fast path to slow one.
 * "CWR"	CWND was reduced due to some Congestion Notification event.
 *		It can be ECN, ICMP source quench, local device congestion.
 * "Recovery"	CWND was reduced, we are fast-retransmitting.
 * "Loss"	CWND was reduced due to RTO timeout or SACK reneging.
 *
 * tcp_fastretrans_alert() is entered:
 * - each incoming ACK, if state is not "Open"
 * - when arrived ACK is unusual, namely:
 *	* SACK
 *	* Duplicate ACK.
 *	* ECN ECE.
 *
 * Counting packets in flight is pretty simple.
 *
 *	in_flight = packets_out - left_out + retrans_out
 *
 *	packets_out is SND.NXT-SND.UNA counted in packets.
 *
 *	retrans_out is number of retransmitted segments.
 *
 *	left_out is number of segments left network, but not ACKed yet.
 *
 *		left_out = sacked_out + lost_out
 *
 *     sacked_out: Packets, which arrived to receiver out of order
 *		   and hence not ACKed. With SACKs this number is simply
 *		   amount of SACKed data. Even without SACKs
 *		   it is easy to give pretty reliable estimate of this number,
 *		   counting duplicate ACKs.
 *
 *       lost_out: Packets lost by network. TCP has no explicit
 *		   "loss notification" feedback from network (for now).
 *		   It means that this number can be only _guessed_.
 *		   Actually, it is the heuristics to predict lossage that
 *		   distinguishes different algorithms.
 *
 *	F.e. after RTO, when all the queue is considered as lost,
 *	lost_out = packets_out and in_flight = retrans_out.
 *
 *		Essentially, we have now two algorithms counting
 *		lost packets.
 *
 *		FACK: It is the simplest heuristics. As soon as we decided
 *		that something is lost, we decide that _all_ not SACKed
 *		packets until the most forward SACK are lost. I.e.
 *		lost_out = fackets_out - sacked_out and left_out = fackets_out.
 *		It is absolutely correct estimate, if network does not reorder
 *		packets. And it loses any connection to reality when reordering
 *		takes place. We use FACK by default until reordering
 *		is suspected on the path to this destination.
 *
 *		NewReno: when Recovery is entered, we assume that one segment
 *		is lost (classic Reno). While we are in Recovery and
 *		a partial ACK arrives, we assume that one more packet
 *		is lost (NewReno). This heuristics are the same in NewReno
 *		and SACK.
 *
 *  Imagine, that's all! Forget about all this shamanism about CWND inflation
 *  deflation etc. CWND is real congestion window, never inflated, changes
 *  only according to classic VJ rules.
 *
 * Really tricky (and requiring careful tuning) part of algorithm
 * is hidden in functions tcp_time_to_recover() and tcp_xmit_retransmit_queue().
 * The first determines the moment _when_ we should reduce CWND and,
 * hence, slow down forward transmission. In fact, it determines the moment
 * when we decide that hole is caused by loss, rather than by a reorder.
 *
 * tcp_xmit_retransmit_queue() decides, _what_ we should retransmit to fill
 * holes, caused by lost packets.
 *
 * And the most logically complicated part of algorithm is undo
 * heuristics. We detect false retransmits due to both too early
 * fast retransmit (reordering) and underestimated RTO, analyzing
 * timestamps and D-SACKs. When we detect that some segments were
 * retransmitted by mistake and CWND reduction was wrong, we undo
 * window reduction and abort recovery phase. This logic is hidden
 * inside several functions named tcp_try_undo_<something>.
 */

/* This function decides, when we should leave Disordered state
 * and enter Recovery phase, reducing congestion window.
 *
 * Main question: may we further continue forward transmission
 * with the same cwnd?
 */
static int tcp_time_to_recover(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	__u32 packets_out;

	/* Do not perform any recovery during F-RTO algorithm */
	if (tp->frto_counter)
		return 0;

	/* Trick#1: The loss is proven. */
	if (tp->lost_out)
		return 1;

	/* Not-A-Trick#2 : Classic rule... */
	if (tcp_dupack_heurestics(tp) > tp->reordering)
		return 1;

	/* Trick#3 : when we use RFC2988 timer restart, fast
	 * retransmit can be triggered by timeout of queue head.
	 */
	if (tcp_is_fack(tp) && tcp_head_timedout(sk))
		return 1;

	/* Trick#4: It is still not OK... But will it be useful to delay
	 * recovery more?
	 */
	packets_out = tp->packets_out;
	if (packets_out <= tp->reordering &&
	    tp->sacked_out >= max_t(__u32, packets_out/2, sysctl_tcp_reordering) &&
	    !tcp_may_send_now(sk)) {
		/* We have nothing to send. This connection is limited
		 * either by receiver window or by application.
		 */
		return 1;
	}

	return 0;
}

/* Mark head of queue up as lost. With RFC3517 SACK, the packets is
 * is against sacked "cnt", otherwise it's against facked "cnt"
 */
static void tcp_mark_head_lost(struct sock *sk, int packets)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *skb;
	int cnt, oldcnt;
	int err;
	unsigned int mss;

	WARN_ON(packets > tp->packets_out);
	if (tp->lost_skb_hint) {
		skb = tp->lost_skb_hint;
		cnt = tp->lost_cnt_hint;
	} else {
		skb = tcp_write_queue_head(sk);
		cnt = 0;
	}

	tcp_for_write_queue_from(skb, sk) {
		if (skb == tcp_send_head(sk))
			break;
		/* TODO: do this better */
		/* this is not the most efficient way to do this... */
		tp->lost_skb_hint = skb;
		tp->lost_cnt_hint = cnt;

		if (after(TCP_SKB_CB(skb)->end_seq, tp->high_seq))
			break;

		oldcnt = cnt;
		if (tcp_is_fack(tp) || tcp_is_reno(tp) ||
		    (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED))
			cnt += tcp_skb_pcount(skb);

		if (cnt > packets) {
			if (tcp_is_sack(tp) || (oldcnt >= packets))
				break;

			mss = skb_shinfo(skb)->gso_size;
			err = tcp_fragment(sk, skb, (packets - oldcnt) * mss, mss);
			if (err < 0)
				break;
			cnt = packets;
		}

		tcp_skb_mark_lost(tp, skb);
	}
	tcp_verify_left_out(tp);
}

/* Account newly detected lost packet(s) */

static void tcp_update_scoreboard(struct sock *sk, int fast_rexmit)
{
	struct tcp_sock *tp = tcp_sk(sk);

	if (tcp_is_reno(tp)) {
		tcp_mark_head_lost(sk, 1);
	} else if (tcp_is_fack(tp)) {
		int lost = tp->fackets_out - tp->reordering;
		if (lost <= 0)
			lost = 1;
		tcp_mark_head_lost(sk, lost);
	} else {
		int sacked_upto = tp->sacked_out - tp->reordering;
		if (sacked_upto < fast_rexmit)
			sacked_upto = fast_rexmit;
		tcp_mark_head_lost(sk, sacked_upto);
	}

	/* New heuristics: it is possible only after we switched
	 * to restart timer each time when something is ACKed.
	 * Hence, we can detect timed out packets during fast
	 * retransmit without falling to slow start.
	 */
	if (tcp_is_fack(tp) && tcp_head_timedout(sk)) {
		struct sk_buff *skb;

		skb = tp->scoreboard_skb_hint ? tp->scoreboard_skb_hint
			: tcp_write_queue_head(sk);

		tcp_for_write_queue_from(skb, sk) {
			if (skb == tcp_send_head(sk))
				break;
			if (!tcp_skb_timedout(sk, skb))
				break;

			tcp_skb_mark_lost(tp, skb);
		}

		tp->scoreboard_skb_hint = skb;

		tcp_verify_left_out(tp);
	}
}

/* CWND moderation, preventing bursts due to too big ACKs
 * in dubious situations.
 */
static inline void tcp_moderate_cwnd(struct tcp_sock *tp)
{
	tp->snd_cwnd = min(tp->snd_cwnd,
			   tcp_packets_in_flight(tp) + tcp_max_burst(tp));
	tp->snd_cwnd_stamp = tcp_time_stamp;
}

/* Lower bound on congestion window is slow start threshold
 * unless congestion avoidance choice decides to overide it.
 */
static inline u32 tcp_cwnd_min(const struct sock *sk)
{
	const struct tcp_congestion_ops *ca_ops = inet_csk(sk)->icsk_ca_ops;

	return ca_ops->min_cwnd ? ca_ops->min_cwnd(sk) : tcp_sk(sk)->snd_ssthresh;
}

/* Decrease cwnd each second ack. */
static void tcp_cwnd_down(struct sock *sk, int flag)
{
	struct tcp_sock *tp = tcp_sk(sk);
	int decr = tp->snd_cwnd_cnt + 1;

	if ((flag & (FLAG_ANY_PROGRESS | FLAG_DSACKING_ACK)) ||
	    (tcp_is_reno(tp) && !(flag & FLAG_NOT_DUP))) {
		tp->snd_cwnd_cnt = decr & 1;
		decr >>= 1;

		if (decr && tp->snd_cwnd > tcp_cwnd_min(sk))
			tp->snd_cwnd -= decr;

		tp->snd_cwnd = min(tp->snd_cwnd, tcp_packets_in_flight(tp) + 1);
		tp->snd_cwnd_stamp = tcp_time_stamp;
	}
}

/* Nothing was retransmitted or returned timestamp is less
 * than timestamp of the first retransmission.
 */
static inline int tcp_packet_delayed(struct tcp_sock *tp)
{
	return !tp->retrans_stamp ||
		(tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr &&
		 before(tp->rx_opt.rcv_tsecr, tp->retrans_stamp));
}

/* Undo procedures. */

#if FASTRETRANS_DEBUG > 1
static void DBGUNDO(struct sock *sk, const char *msg)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct inet_sock *inet = inet_sk(sk);

	if (sk->sk_family == AF_INET) {
		printk(KERN_DEBUG "Undo %s " NIPQUAD_FMT "/%u c%u l%u ss%u/%u p%u\n",
		       msg,
		       NIPQUAD(inet->daddr), ntohs(inet->dport),
		       tp->snd_cwnd, tcp_left_out(tp),
		       tp->snd_ssthresh, tp->prior_ssthresh,
		       tp->packets_out);
	}
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
	else if (sk->sk_family == AF_INET6) {
		struct ipv6_pinfo *np = inet6_sk(sk);
		printk(KERN_DEBUG "Undo %s " NIP6_FMT "/%u c%u l%u ss%u/%u p%u\n",
		       msg,
		       NIP6(np->daddr), ntohs(inet->dport),
		       tp->snd_cwnd, tcp_left_out(tp),
		       tp->snd_ssthresh, tp->prior_ssthresh,
		       tp->packets_out);
	}
#endif
}
#else
#define DBGUNDO(x...) do { } while (0)
#endif

static void tcp_undo_cwr(struct sock *sk, const int undo)
{
	struct tcp_sock *tp = tcp_sk(sk);

	if (tp->prior_ssthresh) {
		const struct inet_connection_sock *icsk = inet_csk(sk);

		if (icsk->icsk_ca_ops->undo_cwnd)
			tp->snd_cwnd = icsk->icsk_ca_ops->undo_cwnd(sk);
		else
			tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh << 1);

		if (undo && tp->prior_ssthresh > tp->snd_ssthresh) {
			tp->snd_ssthresh = tp->prior_ssthresh;
			TCP_ECN_withdraw_cwr(tp);
		}
	} else {
		tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh);
	}
	tcp_moderate_cwnd(tp);
	tp->snd_cwnd_stamp = tcp_time_stamp;
}

static inline int tcp_may_undo(struct tcp_sock *tp)
{
	return tp->undo_marker && (!tp->undo_retrans || tcp_packet_delayed(tp));
}

/* People celebrate: "We love our President!" */
static int tcp_try_undo_recovery(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);

	if (tcp_may_undo(tp)) {
		int mib_idx;

		/* Happy end! We did not retransmit anything
		 * or our original transmission succeeded.
		 */
		DBGUNDO(sk, inet_csk(sk)->icsk_ca_state == TCP_CA_Loss ? "loss" : "retrans");
		tcp_undo_cwr(sk, 1);
		if (inet_csk(sk)->icsk_ca_state == TCP_CA_Loss)
			mib_idx = LINUX_MIB_TCPLOSSUNDO;
		else
			mib_idx = LINUX_MIB_TCPFULLUNDO;

		NET_INC_STATS_BH(sock_net(sk), mib_idx);
		tp->undo_marker = 0;
	}
	if (tp->snd_una == tp->high_seq && tcp_is_reno(tp)) {
		/* Hold old state until something *above* high_seq
		 * is ACKed. For Reno it is MUST to prevent false
		 * fast retransmits (RFC2582). SACK TCP is safe. */
		tcp_moderate_cwnd(tp);
		return 1;
	}
	tcp_set_ca_state(sk, TCP_CA_Open);
	return 0;
}

/* Try to undo cwnd reduction, because D-SACKs acked all retransmitted data */
static void tcp_try_undo_dsack(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);

	if (tp->undo_marker && !tp->undo_retrans) {
		DBGUNDO(sk, "D-SACK");
		tcp_undo_cwr(sk, 1);
		tp->undo_marker = 0;
		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPDSACKUNDO);
	}
}

/* Undo during fast recovery after partial ACK. */

static int tcp_try_undo_partial(struct sock *sk, int acked)
{
	struct tcp_sock *tp = tcp_sk(sk);
	/* Partial ACK arrived. Force Hoe's retransmit. */
	int failed = tcp_is_reno(tp) || (tcp_fackets_out(tp) > tp->reordering);

	if (tcp_may_undo(tp)) {
		/* Plain luck! Hole if filled with delayed
		 * packet, rather than with a retransmit.
		 */
		if (tp->retrans_out == 0)
			tp->retrans_stamp = 0;

		tcp_update_reordering(sk, tcp_fackets_out(tp) + acked, 1);

		DBGUNDO(sk, "Hoe");
		tcp_undo_cwr(sk, 0);
		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPPARTIALUNDO);

		/* So... Do not make Hoe's retransmit yet.
		 * If the first packet was delayed, the rest
		 * ones are most probably delayed as well.
		 */
		failed = 0;
	}
	return failed;
}

/* Undo during loss recovery after partial ACK. */
static int tcp_try_undo_loss(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);

	if (tcp_may_undo(tp)) {
		struct sk_buff *skb;
		tcp_for_write_queue(skb, sk) {
			if (skb == tcp_send_head(sk))
				break;
			TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST;
		}

		tcp_clear_all_retrans_hints(tp);

		DBGUNDO(sk, "partial loss");
		tp->lost_out = 0;
		tcp_undo_cwr(sk, 1);
		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSSUNDO);
		inet_csk(sk)->icsk_retransmits = 0;
		tp->undo_marker = 0;
		if (tcp_is_sack(tp))
			tcp_set_ca_state(sk, TCP_CA_Open);
		return 1;
	}
	return 0;
}

static inline void tcp_complete_cwr(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh);
	tp->snd_cwnd_stamp = tcp_time_stamp;
	tcp_ca_event(sk, CA_EVENT_COMPLETE_CWR);
}

static void tcp_try_keep_open(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	int state = TCP_CA_Open;

	if (tcp_left_out(tp) || tp->retrans_out || tp->undo_marker)
		state = TCP_CA_Disorder;

	if (inet_csk(sk)->icsk_ca_state != state) {
		tcp_set_ca_state(sk, state);
		tp->high_seq = tp->snd_nxt;
	}
}

static void tcp_try_to_open(struct sock *sk, int flag)
{
	struct tcp_sock *tp = tcp_sk(sk);

	tcp_verify_left_out(tp);

	if (!tp->frto_counter && tp->retrans_out == 0)
		tp->retrans_stamp = 0;

	if (flag & FLAG_ECE)
		tcp_enter_cwr(sk, 1);

	if (inet_csk(sk)->icsk_ca_state != TCP_CA_CWR) {
		tcp_try_keep_open(sk);
		tcp_moderate_cwnd(tp);
	} else {
		tcp_cwnd_down(sk, flag);
	}
}

static void tcp_mtup_probe_failed(struct sock *sk)
{
	struct inet_connection_sock *icsk = inet_csk(sk);

	icsk->icsk_mtup.search_high = icsk->icsk_mtup.probe_size - 1;
	icsk->icsk_mtup.probe_size = 0;
}

static void tcp_mtup_probe_success(struct sock *sk, struct sk_buff *skb)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct inet_connection_sock *icsk = inet_csk(sk);

	/* FIXME: breaks with very large cwnd */
	tp->prior_ssthresh = tcp_current_ssthresh(sk);
	tp->snd_cwnd = tp->snd_cwnd *
		       tcp_mss_to_mtu(sk, tp->mss_cache) /
		       icsk->icsk_mtup.probe_size;
	tp->snd_cwnd_cnt = 0;
	tp->snd_cwnd_stamp = tcp_time_stamp;
	tp->rcv_ssthresh = tcp_current_ssthresh(sk);

	icsk->icsk_mtup.search_low = icsk->icsk_mtup.probe_size;
	icsk->icsk_mtup.probe_size = 0;
	tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
}

/* Process an event, which can update packets-in-flight not trivially.
 * Main goal of this function is to calculate new estimate for left_out,
 * taking into account both packets sitting in receiver's buffer and
 * packets lost by network.
 *
 * Besides that it does CWND reduction, when packet loss is detected
 * and changes state of machine.
 *
 * It does _not_ decide what to send, it is made in function
 * tcp_xmit_retransmit_queue().
 */
static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag)
{
	struct inet_connection_sock *icsk = inet_csk(sk);
	struct tcp_sock *tp = tcp_sk(sk);
	int is_dupack = !(flag & (FLAG_SND_UNA_ADVANCED | FLAG_NOT_DUP));
	int do_lost = is_dupack || ((flag & FLAG_DATA_SACKED) &&
				    (tcp_fackets_out(tp) > tp->reordering));
	int fast_rexmit = 0, mib_idx;

	if (WARN_ON(!tp->packets_out && tp->sacked_out))
		tp->sacked_out = 0;
	if (WARN_ON(!tp->sacked_out && tp->fackets_out))
		tp->fackets_out = 0;

	/* Now state machine starts.
	 * A. ECE, hence prohibit cwnd undoing, the reduction is required. */
	if (flag & FLAG_ECE)
		tp->prior_ssthresh = 0;

	/* B. In all the states check for reneging SACKs. */
	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);
		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSS);
	}

	/* D. Check consistency of the current state. */
	tcp_verify_left_out(tp);

	/* E. 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);
		tp->retrans_stamp = 0;
	} else if (!before(tp->snd_una, tp->high_seq)) {
		switch (icsk->icsk_ca_state) {
		case TCP_CA_Loss:
			icsk->icsk_retransmits = 0;
			if (tcp_try_undo_recovery(sk))
				return;
			break;

		case TCP_CA_CWR:
			/* CWR is to be held something *above* high_seq
			 * is ACKed for CWR bit to reach receiver. */
			if (tp->snd_una != tp->high_seq) {
				tcp_complete_cwr(sk);
				tcp_set_ca_state(sk, TCP_CA_Open);
			}
			break;

		case TCP_CA_Disorder:
			tcp_try_undo_dsack(sk);
			if (!tp->undo_marker ||
			    /* For SACK case do not Open to allow to undo
			     * catching for all duplicate ACKs. */
			    tcp_is_reno(tp) || tp->snd_una != tp->high_seq) {
				tp->undo_marker = 0;
				tcp_set_ca_state(sk, TCP_CA_Open);
			}
			break;

		case TCP_CA_Recovery:
			if (tcp_is_reno(tp))
				tcp_reset_reno_sack(tp);
			if (tcp_try_undo_recovery(sk))
				return;
			tcp_complete_cwr(sk);
			break;
		}
	}

	/* F. Process state. */
	switch (icsk->icsk_ca_state) {
	case TCP_CA_Recovery:
		if (!(flag & FLAG_SND_UNA_ADVANCED)) {
			if (tcp_is_reno(tp) && is_dupack)
				tcp_add_reno_sack(sk);
		} else
			do_lost = tcp_try_undo_partial(sk, pkts_acked);
		break;
	case TCP_CA_Loss:
		if (flag & FLAG_DATA_ACKED)
			icsk->icsk_retransmits = 0;
		if (tcp_is_reno(tp) && flag & FLAG_SND_UNA_ADVANCED)
			tcp_reset_reno_sack(tp);
		if (!tcp_try_undo_loss(sk)) {
			tcp_moderate_cwnd(tp);
			tcp_xmit_retransmit_queue(sk);
			return;
		}
		if (icsk->icsk_ca_state != TCP_CA_Open)
			return;
		/* Loss is undone; fall through to processing in Open state. */
	default:
		if (tcp_is_reno(tp)) {
			if (flag & FLAG_SND_UNA_ADVANCED)
				tcp_reset_reno_sack(tp);
			if (is_dupack)
				tcp_add_reno_sack(sk);
		}

		if (icsk->icsk_ca_state == TCP_CA_Disorder)
			tcp_try_undo_dsack(sk);

		if (!tcp_time_to_recover(sk)) {
			tcp_try_to_open(sk, flag);
			return;
		}

		/* MTU probe failure: don't reduce cwnd */
		if (icsk->icsk_ca_state < TCP_CA_CWR &&
		    icsk->icsk_mtup.probe_size &&
		    tp->snd_una == tp->mtu_probe.probe_seq_start) {
			tcp_mtup_probe_failed(sk);
			/* Restores the reduction we did in tcp_mtup_probe() */
			tp->snd_cwnd++;
			tcp_simple_retransmit(sk);
			return;
		}

		/* Otherwise enter Recovery state */

		if (tcp_is_reno(tp))
			mib_idx = LINUX_MIB_TCPRENORECOVERY;
		else
			mib_idx = LINUX_MIB_TCPSACKRECOVERY;

		NET_INC_STATS_BH(sock_net(sk), mib_idx);

		tp->high_seq = tp->snd_nxt;
		tp->prior_ssthresh = 0;
		tp->undo_marker = tp->snd_una;
		tp->undo_retrans = tp->retrans_out;

		if (icsk->icsk_ca_state < TCP_CA_CWR) {
			if (!(flag & FLAG_ECE))
				tp->prior_ssthresh = tcp_current_ssthresh(sk);
			tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk);
			TCP_ECN_queue_cwr(tp);
		}

		tp->bytes_acked = 0;
		tp->snd_cwnd_cnt = 0;
		tcp_set_ca_state(sk, TCP_CA_Recovery);
		fast_rexmit = 1;
	}

	if (do_lost || (tcp_is_fack(tp) && tcp_head_timedout(sk)))
		tcp_update_scoreboard(sk, fast_rexmit);
	tcp_cwnd_down(sk, flag);
	tcp_xmit_retransmit_queue(sk);
}

/* Read draft-ietf-tcplw-high-performance before mucking
 * with this code. (Supersedes RFC1323)
 */
static void tcp_ack_saw_tstamp(struct sock *sk, int flag)
{
	/* RTTM Rule: A TSecr value received in a segment is used to
	 * update the averaged RTT measurement only if the segment
	 * acknowledges some new data, i.e., only if it advances the
	 * left edge of the send window.
	 *
	 * See draft-ietf-tcplw-high-performance-00, section 3.3.
	 * 1998/04/10 Andrey V. Savochkin <saw@msu.ru>
	 *
	 * Changed: reset backoff as soon as we see the first valid sample.
	 * If we do not, we get strongly overestimated rto. With timestamps
	 * samples are accepted even from very old segments: f.e., when rtt=1
	 * increases to 8, we retransmit 5 times and after 8 seconds delayed
	 * answer arrives rto becomes 120 seconds! If at least one of segments
	 * in window is lost... Voila.	 			--ANK (010210)
	 */
	struct tcp_sock *tp = tcp_sk(sk);
	const __u32 seq_rtt = tcp_time_stamp - tp->rx_opt.rcv_tsecr;
	tcp_rtt_estimator(sk, seq_rtt);
	tcp_set_rto(sk);
	inet_csk(sk)->icsk_backoff = 0;
	tcp_bound_rto(sk);
}

static void tcp_ack_no_tstamp(struct sock *sk, u32 seq_rtt, int flag)
{
	/* We don't have a timestamp. Can only use
	 * packets that are not retransmitted to determine
	 * rtt estimates. Also, we must not reset the
	 * backoff for rto until we get a non-retransmitted
	 * packet. This allows us to deal with a situation
	 * where the network delay has increased suddenly.
	 * I.e. Karn's algorithm. (SIGCOMM '87, p5.)
	 */

	if (flag & FLAG_RETRANS_DATA_ACKED)
		return;

	tcp_rtt_estimator(sk, seq_rtt);
	tcp_set_rto(sk);
	inet_csk(sk)->icsk_backoff = 0;
	tcp_bound_rto(sk);
}

static inline void tcp_ack_update_rtt(struct sock *sk, const int flag,
				      const s32 seq_rtt)
{
	const struct tcp_sock *tp = tcp_sk(sk);
	/* Note that peer MAY send zero echo. In this case it is ignored. (rfc1323) */
	if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr)
		tcp_ack_saw_tstamp(sk, flag);
	else if (seq_rtt >= 0)
		tcp_ack_no_tstamp(sk, seq_rtt, flag);
}

static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
{
	const struct inet_connection_sock *icsk = inet_csk(sk);
	icsk->icsk_ca_ops->cong_avoid(sk, ack, in_flight);
	tcp_sk(sk)->snd_cwnd_stamp = tcp_time_stamp;
}

/* Restart timer after forward progress on connection.
 * RFC2988 recommends to restart timer to now+rto.
 */
static void tcp_rearm_rto(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);

	if (!tp->packets_out) {
		inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS);
	} else {
		inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
					  inet_csk(sk)->icsk_rto, TCP_RTO_MAX);
	}
}

/* If we get here, the whole TSO packet has not been acked. */
static u32 tcp_tso_acked(struct sock *sk, struct sk_buff *skb)
{
	struct tcp_sock *tp = tcp_sk(sk);
	u32 packets_acked;

	BUG_ON(!after(TCP_SKB_CB(skb)->end_seq, tp->snd_una));

	packets_acked = tcp_skb_pcount(skb);
	if (tcp_trim_head(sk, skb, tp->snd_una - TCP_SKB_CB(skb)->seq))
		return 0;
	packets_acked -= tcp_skb_pcount(skb);

	if (packets_acked) {
		BUG_ON(tcp_skb_pcount(skb) == 0);
		BUG_ON(!before(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq));
	}

	return packets_acked;
}

/* Remove acknowledged frames from the retransmission queue. If our packet
 * is before the ack sequence we can discard it as it's confirmed to have
 * arrived at the other end.
 */
static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets)
{
	struct tcp_sock *tp = tcp_sk(sk);
	const struct inet_connection_sock *icsk = inet_csk(sk);
	struct sk_buff *skb;
	u32 now = tcp_time_stamp;
	int fully_acked = 1;
	int flag = 0;
	u32 pkts_acked = 0;
	u32 reord = tp->packets_out;
	u32 prior_sacked = tp->sacked_out;
	s32 seq_rtt = -1;
	s32 ca_seq_rtt = -1;
	ktime_t last_ackt = net_invalid_timestamp();

	while ((skb = tcp_write_queue_head(sk)) && skb != tcp_send_head(sk)) {
		struct tcp_skb_cb *scb = TCP_SKB_CB(skb);
		u32 end_seq;
		u32 acked_pcount;
		u8 sacked = scb->sacked;

		/* Determine how many packets and what bytes were acked, tso and else */
		if (after(scb->end_seq, tp->snd_una)) {
			if (tcp_skb_pcount(skb) == 1 ||
			    !after(tp->snd_una, scb->seq))
				break;

			acked_pcount = tcp_tso_acked(sk, skb);
			if (!acked_pcount)
				break;

			fully_acked = 0;
			end_seq = tp->snd_una;
		} else {
			acked_pcount = tcp_skb_pcount(skb);
			end_seq = scb->end_seq;
		}

		/* MTU probing checks */
		if (fully_acked && icsk->icsk_mtup.probe_size &&
		    !after(tp->mtu_probe.probe_seq_end, scb->end_seq)) {
			tcp_mtup_probe_success(sk, skb);
		}

		if (sacked & TCPCB_RETRANS) {
			if (sacked & TCPCB_SACKED_RETRANS)
				tp->retrans_out -= acked_pcount;
			flag |= FLAG_RETRANS_DATA_ACKED;
			ca_seq_rtt = -1;
			seq_rtt = -1;
			if ((flag & FLAG_DATA_ACKED) || (acked_pcount > 1))
				flag |= FLAG_NONHEAD_RETRANS_ACKED;
		} else {
			ca_seq_rtt = now - scb->when;
			last_ackt = skb->tstamp;
			if (seq_rtt < 0) {
				seq_rtt = ca_seq_rtt;
			}
			if (!(sacked & TCPCB_SACKED_ACKED))
				reord = min(pkts_acked, reord);
		}

		if (sacked & TCPCB_SACKED_ACKED)
			tp->sacked_out -= acked_pcount;
		if (sacked & TCPCB_LOST)
			tp->lost_out -= acked_pcount;

		if (unlikely(tp->urg_mode && !before(end_seq, tp->snd_up)))
			tp->urg_mode = 0;

		tp->packets_out -= acked_pcount;
		pkts_acked += acked_pcount;

		/* Initial outgoing SYN's get put onto the write_queue
		 * just like anything else we transmit.  It is not
		 * true data, and if we misinform our callers that
		 * this ACK acks real data, we will erroneously exit
		 * connection startup slow start one packet too
		 * quickly.  This is severely frowned upon behavior.
		 */
		if (!(scb->flags & TCPCB_FLAG_SYN)) {
			flag |= FLAG_DATA_ACKED;
		} else {
			flag |= FLAG_SYN_ACKED;
			tp->retrans_stamp = 0;
		}

		if (!fully_acked)
			break;

		tcp_unlink_write_queue(skb, sk);
		sk_wmem_free_skb(sk, skb);
		tp->scoreboard_skb_hint = NULL;
		if (skb == tp->retransmit_skb_hint)
			tp->retransmit_skb_hint = NULL;
		if (skb == tp->lost_skb_hint)
			tp->lost_skb_hint = NULL;
	}

	if (skb && (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED))
		flag |= FLAG_SACK_RENEGING;

	if (flag & FLAG_ACKED) {
		const struct tcp_congestion_ops *ca_ops
			= inet_csk(sk)->icsk_ca_ops;

		tcp_ack_update_rtt(sk, flag, seq_rtt);
		tcp_rearm_rto(sk);

		if (tcp_is_reno(tp)) {
			tcp_remove_reno_sacks(sk, pkts_acked);
		} else {
			/* Non-retransmitted hole got filled? That's reordering */
			if (reord < prior_fackets)
				tcp_update_reordering(sk, tp->fackets_out - reord, 0);

			/* No need to care for underflows here because
			 * the lost_skb_hint gets NULLed if we're past it
			 * (or something non-trivial happened)
			 */
			if (tcp_is_fack(tp))
				tp->lost_cnt_hint -= pkts_acked;
			else
				tp->lost_cnt_hint -= prior_sacked - tp->sacked_out;
		}

		tp->fackets_out -= min(pkts_acked, tp->fackets_out);

		if (ca_ops->pkts_acked) {
			s32 rtt_us = -1;

			/* Is the ACK triggering packet unambiguous? */
			if (!(flag & FLAG_RETRANS_DATA_ACKED)) {
				/* High resolution needed and available? */
				if (ca_ops->flags & TCP_CONG_RTT_STAMP &&
				    !ktime_equal(last_ackt,
						 net_invalid_timestamp()))
					rtt_us = ktime_us_delta(ktime_get_real(),
								last_ackt);
				else if (ca_seq_rtt > 0)
					rtt_us = jiffies_to_usecs(ca_seq_rtt);
			}

			ca_ops->pkts_acked(sk, pkts_acked, rtt_us);
		}
	}

#if FASTRETRANS_DEBUG > 0
	WARN_ON((int)tp->sacked_out < 0);
	WARN_ON((int)tp->lost_out < 0);
	WARN_ON((int)tp->retrans_out < 0);
	if (!tp->packets_out && tcp_is_sack(tp)) {
		icsk = inet_csk(sk);
		if (tp->lost_out) {
			printk(KERN_DEBUG "Leak l=%u %d\n",
			       tp->lost_out, icsk->icsk_ca_state);
			tp->lost_out = 0;
		}
		if (tp->sacked_out) {
			printk(KERN_DEBUG "Leak s=%u %d\n",
			       tp->sacked_out, icsk->icsk_ca_state);
			tp->sacked_out = 0;
		}
		if (tp->retrans_out) {