1 files changed, 239 insertions, 244 deletions

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a3d47af0190..88fa2d16068 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -68,12 +68,12 @@
 #include <linux/module.h>
 #include <linux/sysctl.h>
 #include <linux/kernel.h>
+#include <linux/prefetch.h>
 #include <net/dst.h>
 #include <net/tcp.h>
 #include <net/inet_common.h>
 #include <linux/ipsec.h>
 #include <asm/unaligned.h>
-#include <net/netdma.h>
 #include <linux/errqueue.h>
 
 int sysctl_tcp_timestamps __read_mostly = 1;
@@ -201,28 +201,25 @@ static inline bool tcp_in_quickack_mode(const struct sock *sk)
 	return icsk->icsk_ack.quick && !icsk->icsk_ack.pingpong;
 }
 
-static inline void TCP_ECN_queue_cwr(struct tcp_sock *tp)
+static void tcp_ecn_queue_cwr(struct tcp_sock *tp)
 {
 	if (tp->ecn_flags & TCP_ECN_OK)
 		tp->ecn_flags |= TCP_ECN_QUEUE_CWR;
 }
 
-static inline void TCP_ECN_accept_cwr(struct tcp_sock *tp, const struct sk_buff *skb)
+static void tcp_ecn_accept_cwr(struct tcp_sock *tp, const struct sk_buff *skb)
 {
 	if (tcp_hdr(skb)->cwr)
 		tp->ecn_flags &= ~TCP_ECN_DEMAND_CWR;
 }
 
-static inline void TCP_ECN_withdraw_cwr(struct tcp_sock *tp)
+static void tcp_ecn_withdraw_cwr(struct tcp_sock *tp)
 {
 	tp->ecn_flags &= ~TCP_ECN_DEMAND_CWR;
 }
 
-static inline void TCP_ECN_check_ce(struct tcp_sock *tp, const struct sk_buff *skb)
+static void __tcp_ecn_check_ce(struct tcp_sock *tp, const struct sk_buff *skb)
 {
-	if (!(tp->ecn_flags & TCP_ECN_OK))
-		return;
-
 	switch (TCP_SKB_CB(skb)->ip_dsfield & INET_ECN_MASK) {
 	case INET_ECN_NOT_ECT:
 		/* Funny extension: if ECT is not set on a segment,
@@ -233,30 +230,43 @@ static inline void TCP_ECN_check_ce(struct tcp_sock *tp, const struct sk_buff *s
 			tcp_enter_quickack_mode((struct sock *)tp);
 		break;
 	case INET_ECN_CE:
+		if (tcp_ca_needs_ecn((struct sock *)tp))
+			tcp_ca_event((struct sock *)tp, CA_EVENT_ECN_IS_CE);
+
 		if (!(tp->ecn_flags & TCP_ECN_DEMAND_CWR)) {
 			/* Better not delay acks, sender can have a very low cwnd */
 			tcp_enter_quickack_mode((struct sock *)tp);
 			tp->ecn_flags |= TCP_ECN_DEMAND_CWR;
 		}
-		/* fallinto */
+		tp->ecn_flags |= TCP_ECN_SEEN;
+		break;
 	default:
+		if (tcp_ca_needs_ecn((struct sock *)tp))
+			tcp_ca_event((struct sock *)tp, CA_EVENT_ECN_NO_CE);
 		tp->ecn_flags |= TCP_ECN_SEEN;
+		break;
 	}
 }
 
-static inline void TCP_ECN_rcv_synack(struct tcp_sock *tp, const struct tcphdr *th)
+static void tcp_ecn_check_ce(struct tcp_sock *tp, const struct sk_buff *skb)
+{
+	if (tp->ecn_flags & TCP_ECN_OK)
+		__tcp_ecn_check_ce(tp, skb);
+}
+
+static void tcp_ecn_rcv_synack(struct tcp_sock *tp, const struct tcphdr *th)
 {
 	if ((tp->ecn_flags & TCP_ECN_OK) && (!th->ece || th->cwr))
 		tp->ecn_flags &= ~TCP_ECN_OK;
 }
 
-static inline void TCP_ECN_rcv_syn(struct tcp_sock *tp, const struct tcphdr *th)
+static void tcp_ecn_rcv_syn(struct tcp_sock *tp, const struct tcphdr *th)
 {
 	if ((tp->ecn_flags & TCP_ECN_OK) && (!th->ece || !th->cwr))
 		tp->ecn_flags &= ~TCP_ECN_OK;
 }
 
-static bool TCP_ECN_rcv_ecn_echo(const struct tcp_sock *tp, const struct tcphdr *th)
+static bool tcp_ecn_rcv_ecn_echo(const struct tcp_sock *tp, const struct tcphdr *th)
 {
 	if (th->ece && !th->syn && (tp->ecn_flags & TCP_ECN_OK))
 		return true;
@@ -653,7 +663,7 @@ static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb)
 	}
 	icsk->icsk_ack.lrcvtime = now;
 
-	TCP_ECN_check_ce(tp, skb);
+	tcp_ecn_check_ce(tp, skb);
 
 	if (skb->len >= 128)
 		tcp_grow_window(sk, skb);
@@ -1295,9 +1305,9 @@ static bool tcp_shifted_skb(struct sock *sk, struct sk_buff *skb,
 	TCP_SKB_CB(prev)->end_seq += shifted;
 	TCP_SKB_CB(skb)->seq += shifted;
 
-	skb_shinfo(prev)->gso_segs += pcount;
-	BUG_ON(skb_shinfo(skb)->gso_segs < pcount);
-	skb_shinfo(skb)->gso_segs -= pcount;
+	tcp_skb_pcount_add(prev, pcount);
+	BUG_ON(tcp_skb_pcount(skb) < pcount);
+	tcp_skb_pcount_add(skb, -pcount);
 
 	/* When we're adding to gso_segs == 1, gso_size will be zero,
 	 * in theory this shouldn't be necessary but as long as DSACK
@@ -1310,7 +1320,7 @@ static bool tcp_shifted_skb(struct sock *sk, struct sk_buff *skb,
 	}
 
 	/* CHECKME: To clear or not to clear? Mimics normal skb currently */
-	if (skb_shinfo(skb)->gso_segs <= 1) {
+	if (tcp_skb_pcount(skb) <= 1) {
 		skb_shinfo(skb)->gso_size = 0;
 		skb_shinfo(skb)->gso_type = 0;
 	}
@@ -1888,21 +1898,21 @@ static inline void tcp_reset_reno_sack(struct tcp_sock *tp)
 	tp->sacked_out = 0;
 }
 
-static void tcp_clear_retrans_partial(struct tcp_sock *tp)
+void tcp_clear_retrans(struct tcp_sock *tp)
 {
 	tp->retrans_out = 0;
 	tp->lost_out = 0;
-
 	tp->undo_marker = 0;
 	tp->undo_retrans = -1;
+	tp->fackets_out = 0;
+	tp->sacked_out = 0;
 }
 
-void tcp_clear_retrans(struct tcp_sock *tp)
+static inline void tcp_init_undo(struct tcp_sock *tp)
 {
-	tcp_clear_retrans_partial(tp);
-
-	tp->fackets_out = 0;
-	tp->sacked_out = 0;
+	tp->undo_marker = tp->snd_una;
+	/* Retransmission still in flight may cause DSACKs later. */
+	tp->undo_retrans = tp->retrans_out ? : -1;
 }
 
 /* Enter Loss state. If we detect SACK reneging, forget all SACK information
@@ -1925,18 +1935,18 @@ void tcp_enter_loss(struct sock *sk)
 		tp->prior_ssthresh = tcp_current_ssthresh(sk);
 		tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk);
 		tcp_ca_event(sk, CA_EVENT_LOSS);
+		tcp_init_undo(tp);
 	}
 	tp->snd_cwnd	   = 1;
 	tp->snd_cwnd_cnt   = 0;
 	tp->snd_cwnd_stamp = tcp_time_stamp;
 
-	tcp_clear_retrans_partial(tp);
+	tp->retrans_out = 0;
+	tp->lost_out = 0;
 
 	if (tcp_is_reno(tp))
 		tcp_reset_reno_sack(tp);
 
-	tp->undo_marker = tp->snd_una;
-
 	skb = tcp_write_queue_head(sk);
 	is_reneg = skb && (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED);
 	if (is_reneg) {
@@ -1950,9 +1960,6 @@ void tcp_enter_loss(struct sock *sk)
 		if (skb == tcp_send_head(sk))
 			break;
 
-		if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS)
-			tp->undo_marker = 0;
-
 		TCP_SKB_CB(skb)->sacked &= (~TCPCB_TAGBITS)|TCPCB_SACKED_ACKED;
 		if (!(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED) || is_reneg) {
 			TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_ACKED;
@@ -1972,7 +1979,7 @@ void tcp_enter_loss(struct sock *sk)
 				       sysctl_tcp_reordering);
 	tcp_set_ca_state(sk, TCP_CA_Loss);
 	tp->high_seq = tp->snd_nxt;
-	TCP_ECN_queue_cwr(tp);
+	tcp_ecn_queue_cwr(tp);
 
 	/* F-RTO RFC5682 sec 3.1 step 1: retransmit SND.UNA if no previous
 	 * loss recovery is underway except recurring timeout(s) on
@@ -2308,6 +2315,35 @@ static inline bool tcp_packet_delayed(const struct tcp_sock *tp)
 
 /* Undo procedures. */
 
+/* We can clear retrans_stamp when there are no retransmissions in the
+ * window. It would seem that it is trivially available for us in
+ * tp->retrans_out, however, that kind of assumptions doesn't consider
+ * what will happen if errors occur when sending retransmission for the
+ * second time. ...It could the that such segment has only
+ * TCPCB_EVER_RETRANS set at the present time. It seems that checking
+ * the head skb is enough except for some reneging corner cases that
+ * are not worth the effort.
+ *
+ * Main reason for all this complexity is the fact that connection dying
+ * time now depends on the validity of the retrans_stamp, in particular,
+ * that successive retransmissions of a segment must not advance
+ * retrans_stamp under any conditions.
+ */
+static bool tcp_any_retrans_done(const struct sock *sk)
+{
+	const struct tcp_sock *tp = tcp_sk(sk);
+	struct sk_buff *skb;
+
+	if (tp->retrans_out)
+		return true;
+
+	skb = tcp_write_queue_head(sk);
+	if (unlikely(skb && TCP_SKB_CB(skb)->sacked & TCPCB_EVER_RETRANS))
+		return true;
+
+	return false;
+}
+
 #if FASTRETRANS_DEBUG > 1
 static void DBGUNDO(struct sock *sk, const char *msg)
 {
@@ -2364,7 +2400,7 @@ static void tcp_undo_cwnd_reduction(struct sock *sk, bool unmark_loss)
 
 		if (tp->prior_ssthresh > tp->snd_ssthresh) {
 			tp->snd_ssthresh = tp->prior_ssthresh;
-			TCP_ECN_withdraw_cwr(tp);
+			tcp_ecn_withdraw_cwr(tp);
 		}
 	} else {
 		tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh);
@@ -2403,6 +2439,8 @@ static bool tcp_try_undo_recovery(struct sock *sk)
 		 * is ACKed. For Reno it is MUST to prevent false
 		 * fast retransmits (RFC2582). SACK TCP is safe. */
 		tcp_moderate_cwnd(tp);
+		if (!tcp_any_retrans_done(sk))
+			tp->retrans_stamp = 0;
 		return true;
 	}
 	tcp_set_ca_state(sk, TCP_CA_Open);
@@ -2423,35 +2461,6 @@ static bool tcp_try_undo_dsack(struct sock *sk)
 	return false;
 }
 
-/* We can clear retrans_stamp when there are no retransmissions in the
- * window. It would seem that it is trivially available for us in
- * tp->retrans_out, however, that kind of assumptions doesn't consider
- * what will happen if errors occur when sending retransmission for the
- * second time. ...It could the that such segment has only
- * TCPCB_EVER_RETRANS set at the present time. It seems that checking
- * the head skb is enough except for some reneging corner cases that
- * are not worth the effort.
- *
- * Main reason for all this complexity is the fact that connection dying
- * time now depends on the validity of the retrans_stamp, in particular,
- * that successive retransmissions of a segment must not advance
- * retrans_stamp under any conditions.
- */
-static bool tcp_any_retrans_done(const struct sock *sk)
-{
-	const struct tcp_sock *tp = tcp_sk(sk);
-	struct sk_buff *skb;
-
-	if (tp->retrans_out)
-		return true;
-
-	skb = tcp_write_queue_head(sk);
-	if (unlikely(skb && TCP_SKB_CB(skb)->sacked & TCPCB_EVER_RETRANS))
-		return true;
-
-	return false;
-}
-
 /* Undo during loss recovery after partial ACK or using F-RTO. */
 static bool tcp_try_undo_loss(struct sock *sk, bool frto_undo)
 {
@@ -2494,7 +2503,7 @@ static void tcp_init_cwnd_reduction(struct sock *sk)
 	tp->prr_delivered = 0;
 	tp->prr_out = 0;
 	tp->snd_ssthresh = inet_csk(sk)->icsk_ca_ops->ssthresh(sk);
-	TCP_ECN_queue_cwr(tp);
+	tcp_ecn_queue_cwr(tp);
 }
 
 static void tcp_cwnd_reduction(struct sock *sk, const int prior_unsacked,
@@ -2671,8 +2680,7 @@ static void tcp_enter_recovery(struct sock *sk, bool ece_ack)
 	NET_INC_STATS_BH(sock_net(sk), mib_idx);
 
 	tp->prior_ssthresh = 0;
-	tp->undo_marker = tp->snd_una;
-	tp->undo_retrans = tp->retrans_out ? : -1;
+	tcp_init_undo(tp);
 
 	if (inet_csk(sk)->icsk_ca_state < TCP_CA_CWR) {
 		if (!ece_ack)
@@ -2687,7 +2695,6 @@ static void tcp_enter_recovery(struct sock *sk, bool ece_ack)
  */
 static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack)
 {
-	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
 	bool recovered = !before(tp->snd_una, tp->high_seq);
 
@@ -2713,12 +2720,9 @@ static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack)
 
 	if (recovered) {
 		/* F-RTO RFC5682 sec 3.1 step 2.a and 1st part of step 3.a */
-		icsk->icsk_retransmits = 0;
 		tcp_try_undo_recovery(sk);
 		return;
 	}
-	if (flag & FLAG_DATA_ACKED)
-		icsk->icsk_retransmits = 0;
 	if (tcp_is_reno(tp)) {
 		/* A Reno DUPACK means new data in F-RTO step 2.b above are
 		 * delivered. Lower inflight to clock out (re)tranmissions.
@@ -2975,7 +2979,8 @@ void tcp_rearm_rto(struct sock *sk)
 		if (icsk->icsk_pending == ICSK_TIME_EARLY_RETRANS ||
 		    icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) {
 			struct sk_buff *skb = tcp_write_queue_head(sk);
-			const u32 rto_time_stamp = TCP_SKB_CB(skb)->when + rto;
+			const u32 rto_time_stamp =
+				tcp_skb_timestamp(skb) + rto;
 			s32 delta = (s32)(rto_time_stamp - tcp_time_stamp);
 			/* delta may not be positive if the socket is locked
 			 * when the retrans timer fires and is rescheduled.
@@ -3027,6 +3032,21 @@ static u32 tcp_tso_acked(struct sock *sk, struct sk_buff *skb)
 	return packets_acked;
 }
 
+static void tcp_ack_tstamp(struct sock *sk, struct sk_buff *skb,
+			   u32 prior_snd_una)
+{
+	const struct skb_shared_info *shinfo;
+
+	/* Avoid cache line misses to get skb_shinfo() and shinfo->tx_flags */
+	if (likely(!(sk->sk_tsflags & SOF_TIMESTAMPING_TX_ACK)))
+		return;
+
+	shinfo = skb_shinfo(skb);
+	if ((shinfo->tx_flags & SKBTX_ACK_TSTAMP) &&
+	    between(shinfo->tskey, prior_snd_una, tcp_sk(sk)->snd_una - 1))
+		__skb_tstamp_tx(skb, NULL, sk, SCM_TSTAMP_ACK);
+}
+
 /* 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.
@@ -3054,6 +3074,8 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets,
 		u8 sacked = scb->sacked;
 		u32 acked_pcount;
 
+		tcp_ack_tstamp(sk, skb, prior_snd_una);
+
 		/* 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 ||
@@ -3066,10 +3088,12 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets,
 
 			fully_acked = false;
 		} else {
+			/* Speedup tcp_unlink_write_queue() and next loop */
+			prefetchw(skb->next);
 			acked_pcount = tcp_skb_pcount(skb);
 		}
 
-		if (sacked & TCPCB_RETRANS) {
+		if (unlikely(sacked & TCPCB_RETRANS)) {
 			if (sacked & TCPCB_SACKED_RETRANS)
 				tp->retrans_out -= acked_pcount;
 			flag |= FLAG_RETRANS_DATA_ACKED;
@@ -3100,26 +3124,21 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets,
 		 * connection startup slow start one packet too
 		 * quickly.  This is severely frowned upon behavior.
 		 */
-		if (!(scb->tcp_flags & TCPHDR_SYN)) {
+		if (likely(!(scb->tcp_flags & TCPHDR_SYN))) {
 			flag |= FLAG_DATA_ACKED;
 		} else {
 			flag |= FLAG_SYN_ACKED;
 			tp->retrans_stamp = 0;
 		}
 
-		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_ACK_TSTAMP) &&
-		    between(skb_shinfo(skb)->tskey, prior_snd_una,
-			    tp->snd_una + 1))
-			__skb_tstamp_tx(skb, NULL, sk, SCM_TSTAMP_ACK);
-
 		if (!fully_acked)
 			break;
 
 		tcp_unlink_write_queue(skb, sk);
 		sk_wmem_free_skb(sk, skb);
-		if (skb == tp->retransmit_skb_hint)
+		if (unlikely(skb == tp->retransmit_skb_hint))
 			tp->retransmit_skb_hint = NULL;
-		if (skb == tp->lost_skb_hint)
+		if (unlikely(skb == tp->lost_skb_hint))
 			tp->lost_skb_hint = NULL;
 	}
 
@@ -3130,7 +3149,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets,
 		flag |= FLAG_SACK_RENEGING;
 
 	skb_mstamp_get(&now);
-	if (first_ackt.v64) {
+	if (likely(first_ackt.v64)) {
 		seq_rtt_us = skb_mstamp_us_delta(&now, &first_ackt);
 		ca_seq_rtt_us = skb_mstamp_us_delta(&now, &last_ackt);
 	}
@@ -3215,9 +3234,10 @@ static void tcp_ack_probe(struct sock *sk)
 		 * This function is not for random using!
 		 */
 	} else {
+		unsigned long when = inet_csk_rto_backoff(icsk, TCP_RTO_MAX);
+
 		inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0,
-					  min(icsk->icsk_rto << icsk->icsk_backoff, TCP_RTO_MAX),
-					  TCP_RTO_MAX);
+					  when, TCP_RTO_MAX);
 	}
 }
 
@@ -3368,6 +3388,14 @@ static void tcp_process_tlp_ack(struct sock *sk, u32 ack, int flag)
 	}
 }
 
+static inline void tcp_in_ack_event(struct sock *sk, u32 flags)
+{
+	const struct inet_connection_sock *icsk = inet_csk(sk);
+
+	if (icsk->icsk_ca_ops->in_ack_event)
+		icsk->icsk_ca_ops->in_ack_event(sk, flags);
+}
+
 /* This routine deals with incoming acks, but not outgoing ones. */
 static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 {
@@ -3383,6 +3411,9 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	int acked = 0; /* Number of packets newly acked */
 	long sack_rtt_us = -1L;
 
+	/* We very likely will need to access write queue head. */
+	prefetchw(sk->sk_write_queue.next);
+
 	/* If the ack is older than previous acks
 	 * then we can probably ignore it.
 	 */
@@ -3405,8 +3436,10 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	    icsk->icsk_pending == ICSK_TIME_LOSS_PROBE)
 		tcp_rearm_rto(sk);
 
-	if (after(ack, prior_snd_una))
+	if (after(ack, prior_snd_una)) {
 		flag |= FLAG_SND_UNA_ADVANCED;
+		icsk->icsk_retransmits = 0;
+	}
 
 	prior_fackets = tp->fackets_out;
 
@@ -3425,10 +3458,12 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 		tp->snd_una = ack;
 		flag |= FLAG_WIN_UPDATE;
 
-		tcp_ca_event(sk, CA_EVENT_FAST_ACK);
+		tcp_in_ack_event(sk, CA_ACK_WIN_UPDATE);
 
 		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPHPACKS);
 	} else {
+		u32 ack_ev_flags = CA_ACK_SLOWPATH;
+
 		if (ack_seq != TCP_SKB_CB(skb)->end_seq)
 			flag |= FLAG_DATA;
 		else
@@ -3440,10 +3475,15 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 			flag |= tcp_sacktag_write_queue(sk, skb, prior_snd_una,
 							&sack_rtt_us);
 
-		if (TCP_ECN_rcv_ecn_echo(tp, tcp_hdr(skb)))
+		if (tcp_ecn_rcv_ecn_echo(tp, tcp_hdr(skb))) {
 			flag |= FLAG_ECE;
+			ack_ev_flags |= CA_ACK_ECE;
+		}
+
+		if (flag & FLAG_WIN_UPDATE)
+			ack_ev_flags |= CA_ACK_WIN_UPDATE;
 
-		tcp_ca_event(sk, CA_EVENT_SLOW_ACK);
+		tcp_in_ack_event(sk, ack_ev_flags);
 	}
 
 	/* We passed data and got it acked, remove any soft error
@@ -4065,6 +4105,44 @@ static void tcp_sack_remove(struct tcp_sock *tp)
 	tp->rx_opt.num_sacks = num_sacks;
 }
 
+/**
+ * tcp_try_coalesce - try to merge skb to prior one
+ * @sk: socket
+ * @to: prior buffer
+ * @from: buffer to add in queue
+ * @fragstolen: pointer to boolean
+ *
+ * Before queueing skb @from after @to, try to merge them
+ * to reduce overall memory use and queue lengths, if cost is small.
+ * Packets in ofo or receive queues can stay a long time.
+ * Better try to coalesce them right now to avoid future collapses.
+ * Returns true if caller should free @from instead of queueing it
+ */
+static bool tcp_try_coalesce(struct sock *sk,
+			     struct sk_buff *to,
+			     struct sk_buff *from,
+			     bool *fragstolen)
+{
+	int delta;
+
+	*fragstolen = false;
+
+	/* Its possible this segment overlaps with prior segment in queue */
+	if (TCP_SKB_CB(from)->seq != TCP_SKB_CB(to)->end_seq)
+		return false;
+
+	if (!skb_try_coalesce(to, from, fragstolen, &delta))
+		return false;
+
+	atomic_add(delta, &sk->sk_rmem_alloc);
+	sk_mem_charge(sk, delta);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOALESCE);
+	TCP_SKB_CB(to)->end_seq = TCP_SKB_CB(from)->end_seq;
+	TCP_SKB_CB(to)->ack_seq = TCP_SKB_CB(from)->ack_seq;
+	TCP_SKB_CB(to)->tcp_flags |= TCP_SKB_CB(from)->tcp_flags;
+	return true;
+}
+
 /* This one checks to see if we can put data from the
  * out_of_order queue into the receive_queue.
  */
@@ -4072,7 +4150,8 @@ static void tcp_ofo_queue(struct sock *sk)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	__u32 dsack_high = tp->rcv_nxt;
-	struct sk_buff *skb;
+	struct sk_buff *skb, *tail;
+	bool fragstolen, eaten;
 
 	while ((skb = skb_peek(&tp->out_of_order_queue)) != NULL) {
 		if (after(TCP_SKB_CB(skb)->seq, tp->rcv_nxt))
@@ -4085,9 +4164,9 @@ static void tcp_ofo_queue(struct sock *sk)
 			tcp_dsack_extend(sk, TCP_SKB_CB(skb)->seq, dsack);
 		}
 
+		__skb_unlink(skb, &tp->out_of_order_queue);
 		if (!after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt)) {
 			SOCK_DEBUG(sk, "ofo packet was already received\n");
-			__skb_unlink(skb, &tp->out_of_order_queue);
 			__kfree_skb(skb);
 			continue;
 		}
@@ -4095,11 +4174,15 @@ static void tcp_ofo_queue(struct sock *sk)
 			   tp->rcv_nxt, TCP_SKB_CB(skb)->seq,
 			   TCP_SKB_CB(skb)->end_seq);
 
-		__skb_unlink(skb, &tp->out_of_order_queue);
-		__skb_queue_tail(&sk->sk_receive_queue, skb);
+		tail = skb_peek_tail(&sk->sk_receive_queue);
+		eaten = tail && tcp_try_coalesce(sk, tail, skb, &fragstolen);
 		tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
-		if (tcp_hdr(skb)->fin)
+		if (!eaten)
+			__skb_queue_tail(&sk->sk_receive_queue, skb);
+		if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
 			tcp_fin(sk);
+		if (eaten)
+			kfree_skb_partial(skb, fragstolen);
 	}
 }
 
@@ -4126,53 +4209,13 @@ static int tcp_try_rmem_schedule(struct sock *sk, struct sk_buff *skb,
 	return 0;
 }
 
-/**
- * tcp_try_coalesce - try to merge skb to prior one
- * @sk: socket
- * @to: prior buffer
- * @from: buffer to add in queue
- * @fragstolen: pointer to boolean
- *
- * Before queueing skb @from after @to, try to merge them
- * to reduce overall memory use and queue lengths, if cost is small.
- * Packets in ofo or receive queues can stay a long time.
- * Better try to coalesce them right now to avoid future collapses.
- * Returns true if caller should free @from instead of queueing it
- */
-static bool tcp_try_coalesce(struct sock *sk,
-			     struct sk_buff *to,
-			     struct sk_buff *from,
-			     bool *fragstolen)
-{
-	int delta;
-
-	*fragstolen = false;
-
-	if (tcp_hdr(from)->fin)
-		return false;
-
-	/* Its possible this segment overlaps with prior segment in queue */
-	if (TCP_SKB_CB(from)->seq != TCP_SKB_CB(to)->end_seq)
-		return false;
-
-	if (!skb_try_coalesce(to, from, fragstolen, &delta))
-		return false;
-
-	atomic_add(delta, &sk->sk_rmem_alloc);
-	sk_mem_charge(sk, delta);
-	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOALESCE);
-	TCP_SKB_CB(to)->end_seq = TCP_SKB_CB(from)->end_seq;
-	TCP_SKB_CB(to)->ack_seq = TCP_SKB_CB(from)->ack_seq;
-	return true;
-}
-
 static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *skb1;
 	u32 seq, end_seq;
 
-	TCP_ECN_check_ce(tp, skb);
+	tcp_ecn_check_ce(tp, skb);
 
 	if (unlikely(tcp_try_rmem_schedule(sk, skb, skb->truesize))) {
 		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPOFODROP);
@@ -4311,24 +4354,19 @@ static int __must_check tcp_queue_rcv(struct sock *sk, struct sk_buff *skb, int
 
 int tcp_send_rcvq(struct sock *sk, struct msghdr *msg, size_t size)
 {
-	struct sk_buff *skb = NULL;
-	struct tcphdr *th;
+	struct sk_buff *skb;
 	bool fragstolen;
 
 	if (size == 0)
 		return 0;
 
-	skb = alloc_skb(size + sizeof(*th), sk->sk_allocation);
+	skb = alloc_skb(size, sk->sk_allocation);
 	if (!skb)
 		goto err;
 
-	if (tcp_try_rmem_schedule(sk, skb, size + sizeof(*th)))
+	if (tcp_try_rmem_schedule(sk, skb, skb->truesize))
 		goto err_free;
 
-	th = (struct tcphdr *)skb_put(skb, sizeof(*th));
-	skb_reset_transport_header(skb);
-	memset(th, 0, sizeof(*th));
-
 	if (memcpy_fromiovec(skb_put(skb, size), msg->msg_iov, size))
 		goto err_free;
 
@@ -4336,7 +4374,7 @@ int tcp_send_rcvq(struct sock *sk, struct msghdr *msg, size_t size)
 	TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq + size;
 	TCP_SKB_CB(skb)->ack_seq = tcp_sk(sk)->snd_una - 1;
 
-	if (tcp_queue_rcv(sk, skb, sizeof(*th), &fragstolen)) {
+	if (tcp_queue_rcv(sk, skb, 0, &fragstolen)) {
 		WARN_ON_ONCE(fragstolen); /* should not happen */
 		__kfree_skb(skb);
 	}
@@ -4350,7 +4388,6 @@ err:
 
 static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 {
-	const struct tcphdr *th = tcp_hdr(skb);
 	struct tcp_sock *tp = tcp_sk(sk);
 	int eaten = -1;
 	bool fragstolen = false;
@@ -4359,9 +4396,9 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 		goto drop;
 
 	skb_dst_drop(skb);
-	__skb_pull(skb, th->doff * 4);
+	__skb_pull(skb, tcp_hdr(skb)->doff * 4);
 
-	TCP_ECN_accept_cwr(tp, skb);
+	tcp_ecn_accept_cwr(tp, skb);
 
 	tp->rx_opt.dsack = 0;
 
@@ -4403,7 +4440,7 @@ queue_and_out:
 		tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
 		if (skb->len)
 			tcp_event_data_recv(sk, skb);
-		if (th->fin)
+		if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
 			tcp_fin(sk);
 
 		if (!skb_queue_empty(&tp->out_of_order_queue)) {
@@ -4518,7 +4555,7 @@ restart:
 		 * - bloated or contains data before "start" or
 		 *   overlaps to the next one.
 		 */
-		if (!tcp_hdr(skb)->syn && !tcp_hdr(skb)->fin &&
+		if (!(TCP_SKB_CB(skb)->tcp_flags & (TCPHDR_SYN | TCPHDR_FIN)) &&
 		    (tcp_win_from_space(skb->truesize) > skb->len ||
 		     before(TCP_SKB_CB(skb)->seq, start))) {
 			end_of_skbs = false;
@@ -4537,30 +4574,18 @@ restart:
 		/* Decided to skip this, advance start seq. */
 		start = TCP_SKB_CB(skb)->end_seq;
 	}
-	if (end_of_skbs || tcp_hdr(skb)->syn || tcp_hdr(skb)->fin)
+	if (end_of_skbs ||
+	    (TCP_SKB_CB(skb)->tcp_flags & (TCPHDR_SYN | TCPHDR_FIN)))
 		return;
 
 	while (before(start, end)) {
+		int copy = min_t(int, SKB_MAX_ORDER(0, 0), end - start);
 		struct sk_buff *nskb;
-		unsigned int header = skb_headroom(skb);
-		int copy = SKB_MAX_ORDER(header, 0);
 
-		/* Too big header? This can happen with IPv6. */
-		if (copy < 0)
-			return;
-		if (end - start < copy)
-			copy = end - start;
-		nskb = alloc_skb(copy + header, GFP_ATOMIC);
+		nskb = alloc_skb(copy, GFP_ATOMIC);
 		if (!nskb)
 			return;
 
-		skb_set_mac_header(nskb, skb_mac_header(skb) - skb->head);
-		skb_set_network_header(nskb, (skb_network_header(skb) -
-					      skb->head));
-		skb_set_transport_header(nskb, (skb_transport_header(skb) -
-						skb->head));
-		skb_reserve(nskb, header);
-		memcpy(nskb->head, skb->head, header);
 		memcpy(nskb->cb, skb->cb, sizeof(skb->cb));
 		TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(nskb)->end_seq = start;
 		__skb_queue_before(list, skb, nskb);
@@ -4584,8 +4609,7 @@ restart:
 				skb = tcp_collapse_one(sk, skb, list);
 				if (!skb ||
 				    skb == tail ||
-				    tcp_hdr(skb)->syn ||
-				    tcp_hdr(skb)->fin)
+				    (TCP_SKB_CB(skb)->tcp_flags & (TCPHDR_SYN | TCPHDR_FIN)))
 					return;
 			}
 		}
@@ -4953,53 +4977,6 @@ static inline bool tcp_checksum_complete_user(struct sock *sk,
 	       __tcp_checksum_complete_user(sk, skb);
 }
 
-#ifdef CONFIG_NET_DMA
-static bool tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb,
-				  int hlen)
-{
-	struct tcp_sock *tp = tcp_sk(sk);
-	int chunk = skb->len - hlen;
-	int dma_cookie;
-	bool copied_early = false;
-
-	if (tp->ucopy.wakeup)
-		return false;
-
-	if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list)
-		tp->ucopy.dma_chan = net_dma_find_channel();
-
-	if (tp->ucopy.dma_chan && skb_csum_unnecessary(skb)) {
-
-		dma_cookie = dma_skb_copy_datagram_iovec(tp->ucopy.dma_chan,
-							 skb, hlen,
-							 tp->ucopy.iov, chunk,
-							 tp->ucopy.pinned_list);
-
-		if (dma_cookie < 0)
-			goto out;
-
-		tp->ucopy.dma_cookie = dma_cookie;
-		copied_early = true;
-
-		tp->ucopy.len -= chunk;
-		tp->copied_seq += chunk;
-		tcp_rcv_space_adjust(sk);
-
-		if ((tp->ucopy.len == 0) ||
-		    (tcp_flag_word(tcp_hdr(skb)) & TCP_FLAG_PSH) ||
-		    (atomic_read(&sk->sk_rmem_alloc) > (sk->sk_rcvbuf >> 1))) {
-			tp->ucopy.wakeup = 1;
-			sk->sk_data_ready(sk);
-		}
-	} else if (chunk > 0) {
-		tp->ucopy.wakeup = 1;
-		sk->sk_data_ready(sk);
-	}
-out:
-	return copied_early;
-}
-#endif /* CONFIG_NET_DMA */
-
 /* Does PAWS and seqno based validation of an incoming segment, flags will
  * play significant role here.
  */
@@ -5179,27 +5156,15 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
 			}
 		} else {
 			int eaten = 0;
-			int copied_early = 0;
 			bool fragstolen = false;
 
-			if (tp->copied_seq == tp->rcv_nxt &&
-			    len - tcp_header_len <= tp->ucopy.len) {
-#ifdef CONFIG_NET_DMA
-				if (tp->ucopy.task == current &&
-				    sock_owned_by_user(sk) &&
-				    tcp_dma_try_early_copy(sk, skb, tcp_header_len)) {
-					copied_early = 1;
-					eaten = 1;
-				}
-#endif
-				if (tp->ucopy.task == current &&
-				    sock_owned_by_user(sk) && !copied_early) {
-					__set_current_state(TASK_RUNNING);
+			if (tp->ucopy.task == current &&
+			    tp->copied_seq == tp->rcv_nxt &&
+			    len - tcp_header_len <= tp->ucopy.len &&
+			    sock_owned_by_user(sk)) {
+				__set_current_state(TASK_RUNNING);
 
-					if (!tcp_copy_to_iovec(sk, skb, tcp_header_len))
-						eaten = 1;
-				}
-				if (eaten) {
+				if (!tcp_copy_to_iovec(sk, skb, tcp_header_len)) {
 					/* Predicted packet is in window by definition.
 					 * seq == rcv_nxt and rcv_wup <= rcv_nxt.
 					 * Hence, check seq<=rcv_wup reduces to:
@@ -5215,9 +5180,8 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
 					__skb_pull(skb, tcp_header_len);
 					tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
 					NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPHPHITSTOUSER);
+					eaten = 1;
 				}
-				if (copied_early)
-					tcp_cleanup_rbuf(sk, skb->len);
 			}
 			if (!eaten) {
 				if (tcp_checksum_complete_user(sk, skb))
@@ -5254,14 +5218,8 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
 					goto no_ack;
 			}
 
-			if (!copied_early || tp->rcv_nxt != tp->rcv_wup)
-				__tcp_ack_snd_check(sk, 0);
+			__tcp_ack_snd_check(sk, 0);
 no_ack:
-#ifdef CONFIG_NET_DMA
-			if (copied_early)
-				__skb_queue_tail(&sk->sk_async_wait_queue, skb);
-			else
-#endif
 			if (eaten)
 				kfree_skb_partial(skb, fragstolen);
 			sk->sk_data_ready(sk);
@@ -5455,7 +5413,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
 		 *    state to ESTABLISHED..."
 		 */
 
-		TCP_ECN_rcv_synack(tp, th);
+		tcp_ecn_rcv_synack(tp, th);
 
 		tcp_init_wl(tp, TCP_SKB_CB(skb)->seq);
 		tcp_ack(sk, skb, FLAG_SLOWPATH);
@@ -5574,7 +5532,7 @@ discard:
 		tp->snd_wl1    = TCP_SKB_CB(skb)->seq;
 		tp->max_window = tp->snd_wnd;
 
-		TCP_ECN_rcv_syn(tp, th);
+		tcp_ecn_rcv_syn(tp, th);
 
 		tcp_mtup_init(sk);
 		tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
@@ -5904,6 +5862,40 @@ static inline void pr_drop_req(struct request_sock *req, __u16 port, int family)
 #endif
 }
 
+/* RFC3168 : 6.1.1 SYN packets must not have ECT/ECN bits set
+ *
+ * If we receive a SYN packet with these bits set, it means a
+ * network is playing bad games with TOS bits. In order to
+ * avoid possible false congestion notifications, we disable
+ * TCP ECN negociation.
+ *
+ * Exception: tcp_ca wants ECN. This is required for DCTCP
+ * congestion control; it requires setting ECT on all packets,
+ * including SYN. We inverse the test in this case: If our
+ * local socket wants ECN, but peer only set ece/cwr (but not
+ * ECT in IP header) its probably a non-DCTCP aware sender.
+ */
+static void tcp_ecn_create_request(struct request_sock *req,
+				   const struct sk_buff *skb,
+				   const struct sock *listen_sk)
+{
+	const struct tcphdr *th = tcp_hdr(skb);
+	const struct net *net = sock_net(listen_sk);
+	bool th_ecn = th->ece && th->cwr;
+	bool ect, need_ecn;
+
+	if (!th_ecn)
+		return;
+
+	ect = !INET_ECN_is_not_ect(TCP_SKB_CB(skb)->ip_dsfield);
+	need_ecn = tcp_ca_needs_ecn(listen_sk);
+
+	if (!ect && !need_ecn && net->ipv4.sysctl_tcp_ecn)
+		inet_rsk(req)->ecn_ok = 1;
+	else if (ect && need_ecn)
+		inet_rsk(req)->ecn_ok = 1;
+}
+
 int tcp_conn_request(struct request_sock_ops *rsk_ops,
 		     const struct tcp_request_sock_ops *af_ops,
 		     struct sock *sk, struct sk_buff *skb)
@@ -5912,7 +5904,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 	struct request_sock *req;
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct dst_entry *dst = NULL;
-	__u32 isn = TCP_SKB_CB(skb)->when;
+	__u32 isn = TCP_SKB_CB(skb)->tcp_tw_isn;
 	bool want_cookie = false, fastopen;
 	struct flowi fl;
 	struct tcp_fastopen_cookie foc = { .len = -1 };
@@ -5964,7 +5956,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 		goto drop_and_free;
 
 	if (!want_cookie || tmp_opt.tstamp_ok)
-		TCP_ECN_create_request(req, skb, sock_net(sk));
+		tcp_ecn_create_request(req, skb, sk);
 
 	if (want_cookie) {
 		isn = cookie_init_sequence(af_ops, sk, skb, &req->mss);
@@ -5979,12 +5971,14 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 		 * timewait bucket, so that all the necessary checks
 		 * are made in the function processing timewait state.
 		 */
-		if (tmp_opt.saw_tstamp && tcp_death_row.sysctl_tw_recycle) {
+		if (tcp_death_row.sysctl_tw_recycle) {
 			bool strict;
 
 			dst = af_ops->route_req(sk, &fl, req, &strict);
+
 			if (dst && strict &&
-			    !tcp_peer_is_proven(req, dst, true)) {
+			    !tcp_peer_is_proven(req, dst, true,
+						tmp_opt.saw_tstamp)) {
 				NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSPASSIVEREJECTED);
 				goto drop_and_release;
 			}
@@ -5993,7 +5987,8 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 		else if (!sysctl_tcp_syncookies &&
 			 (sysctl_max_syn_backlog - inet_csk_reqsk_queue_len(sk) <
 			  (sysctl_max_syn_backlog >> 2)) &&
-			 !tcp_peer_is_proven(req, dst, false)) {
+			 !tcp_peer_is_proven(req, dst, false,
+					     tmp_opt.saw_tstamp)) {
 			/* Without syncookies last quarter of
 			 * backlog is filled with destinations,
 			 * proven to be alive.