1 files changed, 137 insertions, 90 deletions

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a140d9f7a0a3..e37488d3453f 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -72,6 +72,7 @@
 #include <linux/prefetch.h>
 #include <net/dst.h>
 #include <net/tcp.h>
+#include <net/proto_memory.h>
 #include <net/inet_common.h>
 #include <linux/ipsec.h>
 #include <asm/unaligned.h>
@@ -237,9 +238,14 @@ static void tcp_measure_rcv_mss(struct sock *sk, const struct sk_buff *skb)
 		 */
 		if (unlikely(len != icsk->icsk_ack.rcv_mss)) {
 			u64 val = (u64)skb->len << TCP_RMEM_TO_WIN_SCALE;
+			u8 old_ratio = tcp_sk(sk)->scaling_ratio;
 
 			do_div(val, skb->truesize);
 			tcp_sk(sk)->scaling_ratio = val ? val : 1;
+
+			if (old_ratio != tcp_sk(sk)->scaling_ratio)
+				WRITE_ONCE(tcp_sk(sk)->window_clamp,
+					   tcp_win_from_space(sk, sk->sk_rcvbuf));
 		}
 		icsk->icsk_ack.rcv_mss = min_t(unsigned int, len,
 					       tcp_sk(sk)->advmss);
@@ -563,19 +569,20 @@ static void tcp_init_buffer_space(struct sock *sk)
 	maxwin = tcp_full_space(sk);
 
 	if (tp->window_clamp >= maxwin) {
-		tp->window_clamp = maxwin;
+		WRITE_ONCE(tp->window_clamp, maxwin);
 
 		if (tcp_app_win && maxwin > 4 * tp->advmss)
-			tp->window_clamp = max(maxwin -
-					       (maxwin >> tcp_app_win),
-					       4 * tp->advmss);
+			WRITE_ONCE(tp->window_clamp,
+				   max(maxwin - (maxwin >> tcp_app_win),
+				       4 * tp->advmss));
 	}
 
 	/* Force reservation of one segment. */
 	if (tcp_app_win &&
 	    tp->window_clamp > 2 * tp->advmss &&
 	    tp->window_clamp + tp->advmss > maxwin)
-		tp->window_clamp = max(2 * tp->advmss, maxwin - tp->advmss);
+		WRITE_ONCE(tp->window_clamp,
+			   max(2 * tp->advmss, maxwin - tp->advmss));
 
 	tp->rcv_ssthresh = min(tp->rcv_ssthresh, tp->window_clamp);
 	tp->snd_cwnd_stamp = tcp_jiffies32;
@@ -773,7 +780,8 @@ void tcp_rcv_space_adjust(struct sock *sk)
 			WRITE_ONCE(sk->sk_rcvbuf, rcvbuf);
 
 			/* Make the window clamp follow along.  */
-			tp->window_clamp = tcp_win_from_space(sk, rcvbuf);
+			WRITE_ONCE(tp->window_clamp,
+				   tcp_win_from_space(sk, rcvbuf));
 		}
 	}
 	tp->rcvq_space.space = copied;
@@ -911,7 +919,7 @@ static void tcp_rtt_estimator(struct sock *sk, long mrtt_us)
 			tp->rtt_seq = tp->snd_nxt;
 			tp->mdev_max_us = tcp_rto_min_us(sk);
 
-			tcp_bpf_rtt(sk);
+			tcp_bpf_rtt(sk, mrtt_us, srtt);
 		}
 	} else {
 		/* no previous measure. */
@@ -921,7 +929,7 @@ static void tcp_rtt_estimator(struct sock *sk, long mrtt_us)
 		tp->mdev_max_us = tp->rttvar_us;
 		tp->rtt_seq = tp->snd_nxt;
 
-		tcp_bpf_rtt(sk);
+		tcp_bpf_rtt(sk, mrtt_us, srtt);
 	}
 	tp->srtt_us = max(1U, srtt);
 }
@@ -2126,8 +2134,16 @@ void tcp_clear_retrans(struct tcp_sock *tp)
 static inline void tcp_init_undo(struct tcp_sock *tp)
 {
 	tp->undo_marker = tp->snd_una;
+
 	/* Retransmission still in flight may cause DSACKs later. */
-	tp->undo_retrans = tp->retrans_out ? : -1;
+	/* First, account for regular retransmits in flight: */
+	tp->undo_retrans = tp->retrans_out;
+	/* Next, account for TLP retransmits in flight: */
+	if (tp->tlp_high_seq && tp->tlp_retrans)
+		tp->undo_retrans++;
+	/* Finally, avoid 0, because undo_retrans==0 means "can undo now": */
+	if (!tp->undo_retrans)
+		tp->undo_retrans = -1;
 }
 
 static bool tcp_is_rack(const struct sock *sk)
@@ -2206,6 +2222,7 @@ void tcp_enter_loss(struct sock *sk)
 
 	tcp_set_ca_state(sk, TCP_CA_Loss);
 	tp->high_seq = tp->snd_nxt;
+	tp->tlp_high_seq = 0;
 	tcp_ecn_queue_cwr(tp);
 
 	/* F-RTO RFC5682 sec 3.1 step 1: retransmit SND.UNA if no previous
@@ -2779,13 +2796,37 @@ static void tcp_mtup_probe_success(struct sock *sk)
 	NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMTUPSUCCESS);
 }
 
+/* Sometimes we deduce that packets have been dropped due to reasons other than
+ * congestion, like path MTU reductions or failed client TFO attempts. In these
+ * cases we call this function to retransmit as many packets as cwnd allows,
+ * without reducing cwnd. Given that retransmits will set retrans_stamp to a
+ * non-zero value (and may do so in a later calling context due to TSQ), we
+ * also enter CA_Loss so that we track when all retransmitted packets are ACKed
+ * and clear retrans_stamp when that happens (to ensure later recurring RTOs
+ * are using the correct retrans_stamp and don't declare ETIMEDOUT
+ * prematurely).
+ */
+static void tcp_non_congestion_loss_retransmit(struct sock *sk)
+{
+	const struct inet_connection_sock *icsk = inet_csk(sk);
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	if (icsk->icsk_ca_state != TCP_CA_Loss) {
+		tp->high_seq = tp->snd_nxt;
+		tp->snd_ssthresh = tcp_current_ssthresh(sk);
+		tp->prior_ssthresh = 0;
+		tp->undo_marker = 0;
+		tcp_set_ca_state(sk, TCP_CA_Loss);
+	}
+	tcp_xmit_retransmit_queue(sk);
+}
+
 /* Do a simple retransmit without using the backoff mechanisms in
  * tcp_timer. This is used for path mtu discovery.
  * The socket is already locked here.
  */
 void tcp_simple_retransmit(struct sock *sk)
 {
-	const struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *skb;
 	int mss;
@@ -2825,14 +2866,7 @@ void tcp_simple_retransmit(struct sock *sk)
 	 * in network, but units changed and effective
 	 * cwnd/ssthresh really reduced now.
 	 */
-	if (icsk->icsk_ca_state != TCP_CA_Loss) {
-		tp->high_seq = tp->snd_nxt;
-		tp->snd_ssthresh = tcp_current_ssthresh(sk);
-		tp->prior_ssthresh = 0;
-		tp->undo_marker = 0;
-		tcp_set_ca_state(sk, TCP_CA_Loss);
-	}
-	tcp_xmit_retransmit_queue(sk);
+	tcp_non_congestion_loss_retransmit(sk);
 }
 EXPORT_SYMBOL(tcp_simple_retransmit);
 
@@ -3057,7 +3091,7 @@ static void tcp_fastretrans_alert(struct sock *sk, const u32 prior_snd_una,
 			return;
 
 		if (tcp_try_undo_dsack(sk))
-			tcp_try_keep_open(sk);
+			tcp_try_to_open(sk, flag);
 
 		tcp_identify_packet_loss(sk, ack_flag);
 		if (icsk->icsk_ca_state != TCP_CA_Recovery) {
@@ -3539,7 +3573,7 @@ static void tcp_cong_control(struct sock *sk, u32 ack, u32 acked_sacked,
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 
 	if (icsk->icsk_ca_ops->cong_control) {
-		icsk->icsk_ca_ops->cong_control(sk, rs);
+		icsk->icsk_ca_ops->cong_control(sk, ack, flag, rs);
 		return;
 	}
 
@@ -3575,8 +3609,10 @@ static void tcp_snd_sne_update(struct tcp_sock *tp, u32 ack)
 
 	ao = rcu_dereference_protected(tp->ao_info,
 				       lockdep_sock_is_held((struct sock *)tp));
-	if (ao && ack < tp->snd_una)
+	if (ao && ack < tp->snd_una) {
 		ao->snd_sne++;
+		trace_tcp_ao_snd_sne_update((struct sock *)tp, ao->snd_sne);
+	}
 #endif
 }
 
@@ -3601,8 +3637,10 @@ static void tcp_rcv_sne_update(struct tcp_sock *tp, u32 seq)
 
 	ao = rcu_dereference_protected(tp->ao_info,
 				       lockdep_sock_is_held((struct sock *)tp));
-	if (ao && seq < tp->rcv_nxt)
+	if (ao && seq < tp->rcv_nxt) {
 		ao->rcv_sne++;
+		trace_tcp_ao_rcv_sne_update((struct sock *)tp, ao->rcv_sne);
+	}
 #endif
 }
 
@@ -4204,6 +4242,13 @@ void tcp_parse_options(const struct net *net,
 				 */
 				break;
 #endif
+#ifdef CONFIG_TCP_AO
+			case TCPOPT_AO:
+				/* TCP AO has already been checked
+				 * (see tcp_inbound_ao_hash()).
+				 */
+				break;
+#endif
 			case TCPOPT_FASTOPEN:
 				tcp_parse_fastopen_option(
 					opsize - TCPOLEN_FASTOPEN_BASE,
@@ -4433,9 +4478,26 @@ static enum skb_drop_reason tcp_sequence(const struct tcp_sock *tp,
 	return SKB_NOT_DROPPED_YET;
 }
 
+
+void tcp_done_with_error(struct sock *sk, int err)
+{
+	/* This barrier is coupled with smp_rmb() in tcp_poll() */
+	WRITE_ONCE(sk->sk_err, err);
+	smp_wmb();
+
+	tcp_write_queue_purge(sk);
+	tcp_done(sk);
+
+	if (!sock_flag(sk, SOCK_DEAD))
+		sk_error_report(sk);
+}
+EXPORT_SYMBOL(tcp_done_with_error);
+
 /* When we get a reset we do this. */
 void tcp_reset(struct sock *sk, struct sk_buff *skb)
 {
+	int err;
+
 	trace_tcp_receive_reset(sk);
 
 	/* mptcp can't tell us to ignore reset pkts,
@@ -4447,24 +4509,17 @@ void tcp_reset(struct sock *sk, struct sk_buff *skb)
 	/* We want the right error as BSD sees it (and indeed as we do). */
 	switch (sk->sk_state) {
 	case TCP_SYN_SENT:
-		WRITE_ONCE(sk->sk_err, ECONNREFUSED);
+		err = ECONNREFUSED;
 		break;
 	case TCP_CLOSE_WAIT:
-		WRITE_ONCE(sk->sk_err, EPIPE);
+		err = EPIPE;
 		break;
 	case TCP_CLOSE:
 		return;
 	default:
-		WRITE_ONCE(sk->sk_err, ECONNRESET);
+		err = ECONNRESET;
 	}
-	/* This barrier is coupled with smp_rmb() in tcp_poll() */
-	smp_wmb();
-
-	tcp_write_queue_purge(sk);
-	tcp_done(sk);
-
-	if (!sock_flag(sk, SOCK_DEAD))
-		sk_error_report(sk);
+	tcp_done_with_error(sk, err);
 }
 
 /*
@@ -4800,13 +4855,8 @@ static bool tcp_try_coalesce(struct sock *sk,
 	if (TCP_SKB_CB(from)->seq != TCP_SKB_CB(to)->end_seq)
 		return false;
 
-	if (!mptcp_skb_can_collapse(to, from))
-		return false;
-
-#ifdef CONFIG_TLS_DEVICE
-	if (from->decrypted != to->decrypted)
+	if (!tcp_skb_can_collapse_rx(to, from))
 		return false;
-#endif
 
 	if (!skb_try_coalesce(to, from, fragstolen, &delta))
 		return false;
@@ -4848,7 +4898,7 @@ static void tcp_drop_reason(struct sock *sk, struct sk_buff *skb,
 			    enum skb_drop_reason reason)
 {
 	sk_drops_add(sk, skb);
-	kfree_skb_reason(skb, reason);
+	sk_skb_reason_drop(sk, skb, reason);
 }
 
 /* This one checks to see if we can put data from the
@@ -5174,6 +5224,16 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 	 */
 	if (TCP_SKB_CB(skb)->seq == tp->rcv_nxt) {
 		if (tcp_receive_window(tp) == 0) {
+			/* Some stacks are known to send bare FIN packets
+			 * in a loop even if we send RWIN 0 in our ACK.
+			 * Accepting this FIN does not hurt memory pressure
+			 * because the FIN flag will simply be merged to the
+			 * receive queue tail skb in most cases.
+			 */
+			if (!skb->len &&
+			    (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN))
+				goto queue_and_out;
+
 			reason = SKB_DROP_REASON_TCP_ZEROWINDOW;
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPZEROWINDOWDROP);
 			goto out_of_window;
@@ -5188,7 +5248,7 @@ queue_and_out:
 			inet_csk_schedule_ack(sk);
 			sk->sk_data_ready(sk);
 
-			if (skb_queue_len(&sk->sk_receive_queue)) {
+			if (skb_queue_len(&sk->sk_receive_queue) && skb->len) {
 				reason = SKB_DROP_REASON_PROTO_MEM;
 				NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPRCVQDROP);
 				goto drop;
@@ -5351,7 +5411,7 @@ restart:
 			break;
 		}
 
-		if (n && n != tail && mptcp_skb_can_collapse(skb, n) &&
+		if (n && n != tail && tcp_skb_can_collapse_rx(skb, n) &&
 		    TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(n)->seq) {
 			end_of_skbs = false;
 			break;
@@ -5375,9 +5435,7 @@ restart:
 			break;
 
 		memcpy(nskb->cb, skb->cb, sizeof(skb->cb));
-#ifdef CONFIG_TLS_DEVICE
-		nskb->decrypted = skb->decrypted;
-#endif
+		skb_copy_decrypted(nskb, skb);
 		TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(nskb)->end_seq = start;
 		if (list)
 			__skb_queue_before(list, skb, nskb);
@@ -5404,13 +5462,9 @@ restart:
 				skb = tcp_collapse_one(sk, skb, list, root);
 				if (!skb ||
 				    skb == tail ||
-				    !mptcp_skb_can_collapse(nskb, skb) ||
+				    !tcp_skb_can_collapse_rx(nskb, skb) ||
 				    (TCP_SKB_CB(skb)->tcp_flags & (TCPHDR_SYN | TCPHDR_FIN)))
 					goto end;
-#ifdef CONFIG_TLS_DEVICE
-				if (skb->decrypted != nskb->decrypted)
-					goto end;
-#endif
 			}
 		}
 	}
@@ -5949,6 +6003,11 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
 	 * RFC 5961 4.2 : Send a challenge ack
 	 */
 	if (th->syn) {
+		if (sk->sk_state == TCP_SYN_RECV && sk->sk_socket && th->ack &&
+		    TCP_SKB_CB(skb)->seq + 1 == TCP_SKB_CB(skb)->end_seq &&
+		    TCP_SKB_CB(skb)->seq + 1 == tp->rcv_nxt &&
+		    TCP_SKB_CB(skb)->ack_seq == tp->snd_nxt)
+			goto pass;
 syn_challenge:
 		if (syn_inerr)
 			TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS);
@@ -5958,6 +6017,7 @@ syn_challenge:
 		goto discard;
 	}
 
+pass:
 	bpf_skops_parse_hdr(sk, skb);
 
 	return true;
@@ -6288,7 +6348,7 @@ static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack,
 			tp->fastopen_client_fail = TFO_DATA_NOT_ACKED;
 		skb_rbtree_walk_from(data)
 			 tcp_mark_skb_lost(sk, data);
-		tcp_xmit_retransmit_queue(sk);
+		tcp_non_congestion_loss_retransmit(sk);
 		NET_INC_STATS(sock_net(sk),
 				LINUX_MIB_TCPFASTOPENACTIVEFAIL);
 		return true;
@@ -6426,7 +6486,8 @@ consume:
 
 		if (!tp->rx_opt.wscale_ok) {
 			tp->rx_opt.snd_wscale = tp->rx_opt.rcv_wscale = 0;
-			tp->window_clamp = min(tp->window_clamp, 65535U);
+			WRITE_ONCE(tp->window_clamp,
+				   min(tp->window_clamp, 65535U));
 		}
 
 		if (tp->rx_opt.saw_tstamp) {
@@ -6973,35 +7034,10 @@ static void tcp_openreq_init(struct request_sock *req,
 #endif
 }
 
-struct request_sock *inet_reqsk_alloc(const struct request_sock_ops *ops,
-				      struct sock *sk_listener,
-				      bool attach_listener)
-{
-	struct request_sock *req = reqsk_alloc(ops, sk_listener,
-					       attach_listener);
-
-	if (req) {
-		struct inet_request_sock *ireq = inet_rsk(req);
-
-		ireq->ireq_opt = NULL;
-#if IS_ENABLED(CONFIG_IPV6)
-		ireq->pktopts = NULL;
-#endif
-		atomic64_set(&ireq->ir_cookie, 0);
-		ireq->ireq_state = TCP_NEW_SYN_RECV;
-		write_pnet(&ireq->ireq_net, sock_net(sk_listener));
-		ireq->ireq_family = sk_listener->sk_family;
-		req->timeout = TCP_TIMEOUT_INIT;
-	}
-
-	return req;
-}
-EXPORT_SYMBOL(inet_reqsk_alloc);
-
 /*
  * Return true if a syncookie should be sent
  */
-static bool tcp_syn_flood_action(const struct sock *sk, const char *proto)
+static bool tcp_syn_flood_action(struct sock *sk, const char *proto)
 {
 	struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue;
 	const char *msg = "Dropping request";
@@ -7102,7 +7138,6 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 		     struct sock *sk, struct sk_buff *skb)
 {
 	struct tcp_fastopen_cookie foc = { .len = -1 };
-	__u32 isn = TCP_SKB_CB(skb)->tcp_tw_isn;
 	struct tcp_options_received tmp_opt;
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct net *net = sock_net(sk);
@@ -7112,21 +7147,28 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 	struct dst_entry *dst;
 	struct flowi fl;
 	u8 syncookies;
+	u32 isn;
 
 #ifdef CONFIG_TCP_AO
 	const struct tcp_ao_hdr *aoh;
 #endif
 
-	syncookies = READ_ONCE(net->ipv4.sysctl_tcp_syncookies);
+	isn = __this_cpu_read(tcp_tw_isn);
+	if (isn) {
+		/* TW buckets are converted to open requests without
+		 * limitations, they conserve resources and peer is
+		 * evidently real one.
+		 */
+		__this_cpu_write(tcp_tw_isn, 0);
+	} else {
+		syncookies = READ_ONCE(net->ipv4.sysctl_tcp_syncookies);
 
-	/* TW buckets are converted to open requests without
-	 * limitations, they conserve resources and peer is
-	 * evidently real one.
-	 */
-	if ((syncookies == 2 || inet_csk_reqsk_queue_is_full(sk)) && !isn) {
-		want_cookie = tcp_syn_flood_action(sk, rsk_ops->slab_name);
-		if (!want_cookie)
-			goto drop;
+		if (syncookies == 2 || inet_csk_reqsk_queue_is_full(sk)) {
+			want_cookie = tcp_syn_flood_action(sk,
+							   rsk_ops->slab_name);
+			if (!want_cookie)
+				goto drop;
+		}
 	}
 
 	if (sk_acceptq_is_full(sk)) {
@@ -7165,7 +7207,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 	/* Note: tcp_v6_init_req() might override ir_iif for link locals */
 	inet_rsk(req)->ir_iif = inet_request_bound_dev_if(sk, skb);
 
-	dst = af_ops->route_req(sk, skb, &fl, req);
+	dst = af_ops->route_req(sk, skb, &fl, req, isn);
 	if (!dst)
 		goto drop_and_free;
 
@@ -7242,7 +7284,12 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 		tcp_rsk(req)->tfo_listener = false;
 		if (!want_cookie) {
 			req->timeout = tcp_timeout_init((struct sock *)req);
-			inet_csk_reqsk_queue_hash_add(sk, req, req->timeout);
+			if (unlikely(!inet_csk_reqsk_queue_hash_add(sk, req,
+								    req->timeout))) {
+				reqsk_free(req);
+				return 0;
+			}
+
 		}
 		af_ops->send_synack(sk, dst, &fl, req, &foc,
 				    !want_cookie ? TCP_SYNACK_NORMAL :