Merge branch 'tcp_drop_reason'

Menglong Dong says:

====================
net: add skb drop reasons to TCP packet receive

In the commit c504e5c ("net: skb: introduce kfree_skb_reason()"),
we added the support of reporting the reasons of skb drops to kfree_skb
tracepoint. And in this series patches, reasons for skb drops are added
to TCP layer (both TCPv4 and TCPv6 are considered).
Following functions are processed:

tcp_v4_rcv()
tcp_v6_rcv()
tcp_v4_inbound_md5_hash()
tcp_v6_inbound_md5_hash()
tcp_add_backlog()
tcp_v4_do_rcv()
tcp_v6_do_rcv()
tcp_rcv_established()
tcp_data_queue()
tcp_data_queue_ofo()

The functions we handled are mostly for packet ingress, as skb drops
hardly happens in the egress path of TCP layer. However, it's a little
complex for TCP state processing, as I find that it's hard to report skb
drop reasons to where it is freed. For example, when skb is dropped in
tcp_rcv_state_process(), the reason can be caused by the call of
tcp_v4_conn_request(), and it's hard to return a drop reason from
tcp_v4_conn_request(). So such cases are skipped  for this moment.

Following new drop reasons are introduced (what they mean can be see
in the document for them):

/* SKB_DROP_REASON_TCP_MD5* corresponding to LINUX_MIB_TCPMD5* */
SKB_DROP_REASON_TCP_MD5NOTFOUND
SKB_DROP_REASON_TCP_MD5UNEXPECTED
SKB_DROP_REASON_TCP_MD5FAILURE
SKB_DROP_REASON_SOCKET_BACKLOG
SKB_DROP_REASON_TCP_FLAGS
SKB_DROP_REASON_TCP_ZEROWINDOW
SKB_DROP_REASON_TCP_OLD_DATA
SKB_DROP_REASON_TCP_OVERWINDOW
/* corresponding to LINUX_MIB_TCPOFOMERGE */
SKB_DROP_REASON_TCP_OFOMERGE

Here is a example to get TCP packet drop reasons from ftrace:

$ echo 1 > /sys/kernel/debug/tracing/events/skb/kfree_skb/enable
$ cat /sys/kernel/debug/tracing/trace
$ <idle>-0       [036] ..s1.   647.428165: kfree_skb: skbaddr=000000004d037db6 protocol=2048 location=0000000074cd1243 reason: NO_SOCKET
$ <idle>-0       [020] ..s2.   639.676674: kfree_skb: skbaddr=00000000bcbfa42d protocol=2048 location=00000000bfe89d35 reason: PROTO_MEM

From the reason 'PROTO_MEM' we can know that the skb is dropped because
the memory configured in net.ipv4.tcp_mem is up to the limition.

Changes since v2:
- remove the 'inline' of tcp_drop() in the 1th patch, as Jakub
  suggested

Changes since v1:
- enrich the document for this series patches in the cover letter,
  as Eric suggested
- fix compile warning report by Jakub in the 6th patch
- let NO_SOCKET trump the XFRM failure in the 2th and 3th patches
====================

Signed-off-by: David S. Miller <[email protected]>

Author: davem330

include/linux/skbuff.h

@@ -346,6 +346,40 @@ enum skb_drop_reason {
 					 * udp packet drop out of
 					 * udp_memory_allocated.
 					 */
+	SKB_DROP_REASON_TCP_MD5NOTFOUND,	/* no MD5 hash and one
+						 * expected, corresponding
+						 * to LINUX_MIB_TCPMD5NOTFOUND
+						 */
+	SKB_DROP_REASON_TCP_MD5UNEXPECTED,	/* MD5 hash and we're not
+						 * expecting one, corresponding
+						 * to LINUX_MIB_TCPMD5UNEXPECTED
+						 */
+	SKB_DROP_REASON_TCP_MD5FAILURE,	/* MD5 hash and its wrong,
+					 * corresponding to
+					 * LINUX_MIB_TCPMD5FAILURE
+					 */
+	SKB_DROP_REASON_SOCKET_BACKLOG,	/* failed to add skb to socket
+					 * backlog (see
+					 * LINUX_MIB_TCPBACKLOGDROP)
+					 */
+	SKB_DROP_REASON_TCP_FLAGS,	/* TCP flags invalid */
+	SKB_DROP_REASON_TCP_ZEROWINDOW,	/* TCP receive window size is zero,
+					 * see LINUX_MIB_TCPZEROWINDOWDROP
+					 */
+	SKB_DROP_REASON_TCP_OLD_DATA,	/* the TCP data reveived is already
+					 * received before (spurious retrans
+					 * may happened), see
+					 * LINUX_MIB_DELAYEDACKLOST
+					 */
+	SKB_DROP_REASON_TCP_OVERWINDOW,	/* the TCP data is out of window,
+					 * the seq of the first byte exceed
+					 * the right edges of receive
+					 * window
+					 */
+	SKB_DROP_REASON_TCP_OFOMERGE,	/* the data of skb is already in
+					 * the ofo queue, corresponding to
+					 * LINUX_MIB_TCPOFOMERGE
+					 */
 	SKB_DROP_REASON_MAX,
 };
 

include/net/tcp.h

@@ -1367,7 +1367,8 @@ static inline bool tcp_checksum_complete(struct sk_buff *skb)
 		__skb_checksum_complete(skb);
 }
 
-bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb);
+bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb,
+		     enum skb_drop_reason *reason);
 
 #ifdef CONFIG_INET
 void __sk_defer_free_flush(struct sock *sk);

include/trace/events/skb.h

@@ -27,6 +27,16 @@
 	EM(SKB_DROP_REASON_IP_NOPROTO, IP_NOPROTO)		\
 	EM(SKB_DROP_REASON_SOCKET_RCVBUFF, SOCKET_RCVBUFF)	\
 	EM(SKB_DROP_REASON_PROTO_MEM, PROTO_MEM)		\
+	EM(SKB_DROP_REASON_TCP_MD5NOTFOUND, TCP_MD5NOTFOUND)	\
+	EM(SKB_DROP_REASON_TCP_MD5UNEXPECTED,			\
+	   TCP_MD5UNEXPECTED)					\
+	EM(SKB_DROP_REASON_TCP_MD5FAILURE, TCP_MD5FAILURE)	\
+	EM(SKB_DROP_REASON_SOCKET_BACKLOG, SOCKET_BACKLOG)	\
+	EM(SKB_DROP_REASON_TCP_FLAGS, TCP_FLAGS)		\
+	EM(SKB_DROP_REASON_TCP_ZEROWINDOW, TCP_ZEROWINDOW)	\
+	EM(SKB_DROP_REASON_TCP_OLD_DATA, TCP_OLD_DATA)		\
+	EM(SKB_DROP_REASON_TCP_OVERWINDOW, TCP_OVERWINDOW)	\
+	EM(SKB_DROP_REASON_TCP_OFOMERGE, TCP_OFOMERGE)		\
 	EMe(SKB_DROP_REASON_MAX, MAX)
 
 #undef EM

net/ipv4/tcp_input.c

@@ -4684,10 +4684,16 @@ static bool tcp_ooo_try_coalesce(struct sock *sk,
 	return res;
 }
 
-static void tcp_drop(struct sock *sk, struct sk_buff *skb)
+static void tcp_drop_reason(struct sock *sk, struct sk_buff *skb,
+			    enum skb_drop_reason reason)
 {
 	sk_drops_add(sk, skb);
-	__kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
+}
+
+static void tcp_drop(struct sock *sk, struct sk_buff *skb)
+{
+	tcp_drop_reason(sk, skb, SKB_DROP_REASON_NOT_SPECIFIED);
 }
 
 /* This one checks to see if we can put data from the
@@ -4773,7 +4779,7 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb)
 	if (unlikely(tcp_try_rmem_schedule(sk, skb, skb->truesize))) {
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPOFODROP);
 		sk->sk_data_ready(sk);
-		tcp_drop(sk, skb);
+		tcp_drop_reason(sk, skb, SKB_DROP_REASON_PROTO_MEM);
 		return;
 	}
 
@@ -4836,7 +4842,8 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb)
 				/* All the bits are present. Drop. */
 				NET_INC_STATS(sock_net(sk),
 					      LINUX_MIB_TCPOFOMERGE);
-				tcp_drop(sk, skb);
+				tcp_drop_reason(sk, skb,
+						SKB_DROP_REASON_TCP_OFOMERGE);
 				skb = NULL;
 				tcp_dsack_set(sk, seq, end_seq);
 				goto add_sack;
@@ -4855,7 +4862,8 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb)
 						 TCP_SKB_CB(skb1)->end_seq);
 				NET_INC_STATS(sock_net(sk),
 					      LINUX_MIB_TCPOFOMERGE);
-				tcp_drop(sk, skb1);
+				tcp_drop_reason(sk, skb1,
+						SKB_DROP_REASON_TCP_OFOMERGE);
 				goto merge_right;
 			}
 		} else if (tcp_ooo_try_coalesce(sk, skb1,
@@ -4883,7 +4891,7 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb)
 		tcp_dsack_extend(sk, TCP_SKB_CB(skb1)->seq,
 				 TCP_SKB_CB(skb1)->end_seq);
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPOFOMERGE);
-		tcp_drop(sk, skb1);
+		tcp_drop_reason(sk, skb1, SKB_DROP_REASON_TCP_OFOMERGE);
 	}
 	/* If there is no skb after us, we are the last_skb ! */
 	if (!skb1)
@@ -4982,6 +4990,7 @@ void tcp_data_ready(struct sock *sk)
 static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
+	enum skb_drop_reason reason;
 	bool fragstolen;
 	int eaten;
 
@@ -5000,6 +5009,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 	skb_dst_drop(skb);
 	__skb_pull(skb, tcp_hdr(skb)->doff * 4);
 
+	reason = SKB_DROP_REASON_NOT_SPECIFIED;
 	tp->rx_opt.dsack = 0;
 
 	/*  Queue data for delivery to the user.
@@ -5008,6 +5018,7 @@ 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) {
+			reason = SKB_DROP_REASON_TCP_ZEROWINDOW;
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPZEROWINDOWDROP);
 			goto out_of_window;
 		}
@@ -5017,6 +5028,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 		if (skb_queue_len(&sk->sk_receive_queue) == 0)
 			sk_forced_mem_schedule(sk, skb->truesize);
 		else if (tcp_try_rmem_schedule(sk, skb, skb->truesize)) {
+			reason = SKB_DROP_REASON_PROTO_MEM;
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPRCVQDROP);
 			sk->sk_data_ready(sk);
 			goto drop;
@@ -5053,20 +5065,24 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 	if (!after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt)) {
 		tcp_rcv_spurious_retrans(sk, skb);
 		/* A retransmit, 2nd most common case.  Force an immediate ack. */
+		reason = SKB_DROP_REASON_TCP_OLD_DATA;
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_DELAYEDACKLOST);
 		tcp_dsack_set(sk, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq);
 
 out_of_window:
 		tcp_enter_quickack_mode(sk, TCP_MAX_QUICKACKS);
 		inet_csk_schedule_ack(sk);
 drop:
-		tcp_drop(sk, skb);
+		tcp_drop_reason(sk, skb, reason);
 		return;
 	}
 
 	/* Out of window. F.e. zero window probe. */
-	if (!before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt + tcp_receive_window(tp)))
+	if (!before(TCP_SKB_CB(skb)->seq,
+		    tp->rcv_nxt + tcp_receive_window(tp))) {
+		reason = SKB_DROP_REASON_TCP_OVERWINDOW;
 		goto out_of_window;
+	}
 
 	if (before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
 		/* Partial packet, seq < rcv_next < end_seq */
@@ -5076,6 +5092,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 		 * remembering D-SACK for its head made in previous line.
 		 */
 		if (!tcp_receive_window(tp)) {
+			reason = SKB_DROP_REASON_TCP_ZEROWINDOW;
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPZEROWINDOWDROP);
 			goto out_of_window;
 		}
@@ -5781,6 +5798,7 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
  */
 void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
 {
+	enum skb_drop_reason reason = SKB_DROP_REASON_NOT_SPECIFIED;
 	const struct tcphdr *th = (const struct tcphdr *)skb->data;
 	struct tcp_sock *tp = tcp_sk(sk);
 	unsigned int len = skb->len;
@@ -5869,6 +5887,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
 				tp->rcv_rtt_last_tsecr = tp->rx_opt.rcv_tsecr;
 				return;
 			} else { /* Header too small */
+				reason = SKB_DROP_REASON_PKT_TOO_SMALL;
 				TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS);
 				goto discard;
 			}
@@ -5924,8 +5943,10 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
 	if (len < (th->doff << 2) || tcp_checksum_complete(skb))
 		goto csum_error;
 
-	if (!th->ack && !th->rst && !th->syn)
+	if (!th->ack && !th->rst && !th->syn) {
+		reason = SKB_DROP_REASON_TCP_FLAGS;
 		goto discard;
+	}
 
 	/*
 	 *	Standard slow path.
@@ -5951,12 +5972,13 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
 	return;
 
 csum_error:
+	reason = SKB_DROP_REASON_TCP_CSUM;
 	trace_tcp_bad_csum(skb);
 	TCP_INC_STATS(sock_net(sk), TCP_MIB_CSUMERRORS);
 	TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS);
 
 discard:
-	tcp_drop(sk, skb);
+	tcp_drop_reason(sk, skb, reason);
 }
 EXPORT_SYMBOL(tcp_rcv_established);
 

net/ipv4/tcp_ipv4.c

@@ -1412,7 +1412,8 @@ EXPORT_SYMBOL(tcp_v4_md5_hash_skb);
 /* Called with rcu_read_lock() */
 static bool tcp_v4_inbound_md5_hash(const struct sock *sk,
 				    const struct sk_buff *skb,
-				    int dif, int sdif)
+				    int dif, int sdif,
+				    enum skb_drop_reason *reason)
 {
 #ifdef CONFIG_TCP_MD5SIG
 	/*
@@ -1445,11 +1446,13 @@ static bool tcp_v4_inbound_md5_hash(const struct sock *sk,
 		return false;
 
 	if (hash_expected && !hash_location) {
+		*reason = SKB_DROP_REASON_TCP_MD5NOTFOUND;
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5NOTFOUND);
 		return true;
 	}
 
 	if (!hash_expected && hash_location) {
+		*reason = SKB_DROP_REASON_TCP_MD5UNEXPECTED;
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5UNEXPECTED);
 		return true;
 	}
@@ -1462,6 +1465,7 @@ static bool tcp_v4_inbound_md5_hash(const struct sock *sk,
 				      NULL, skb);
 
 	if (genhash || memcmp(hash_location, newhash, 16) != 0) {
+		*reason = SKB_DROP_REASON_TCP_MD5FAILURE;
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5FAILURE);
 		net_info_ratelimited("MD5 Hash failed for (%pI4, %d)->(%pI4, %d)%s L3 index %d\n",
 				     &iph->saddr, ntohs(th->source),
@@ -1704,6 +1708,7 @@ INDIRECT_CALLABLE_DECLARE(struct dst_entry *ipv4_dst_check(struct dst_entry *,
  */
 int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
 {
+	enum skb_drop_reason reason;
 	struct sock *rsk;
 
 	if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */
@@ -1726,6 +1731,7 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
 		return 0;
 	}
 
+	reason = SKB_DROP_REASON_NOT_SPECIFIED;
 	if (tcp_checksum_complete(skb))
 		goto csum_err;
 
@@ -1753,7 +1759,7 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
 reset:
 	tcp_v4_send_reset(rsk, skb);
 discard:
-	kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
 	/* Be careful here. If this function gets more complicated and
 	 * gcc suffers from register pressure on the x86, sk (in %ebx)
 	 * might be destroyed here. This current version compiles correctly,
@@ -1762,6 +1768,7 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
 	return 0;
 
 csum_err:
+	reason = SKB_DROP_REASON_TCP_CSUM;
 	trace_tcp_bad_csum(skb);
 	TCP_INC_STATS(sock_net(sk), TCP_MIB_CSUMERRORS);
 	TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS);
@@ -1807,7 +1814,8 @@ int tcp_v4_early_demux(struct sk_buff *skb)
 	return 0;
 }
 
-bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb)
+bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb,
+		     enum skb_drop_reason *reason)
 {
 	u32 limit, tail_gso_size, tail_gso_segs;
 	struct skb_shared_info *shinfo;
@@ -1833,6 +1841,7 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb)
 	if (unlikely(tcp_checksum_complete(skb))) {
 		bh_unlock_sock(sk);
 		trace_tcp_bad_csum(skb);
+		*reason = SKB_DROP_REASON_TCP_CSUM;
 		__TCP_INC_STATS(sock_net(sk), TCP_MIB_CSUMERRORS);
 		__TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS);
 		return true;
@@ -1921,6 +1930,7 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb)
 
 	if (unlikely(sk_add_backlog(sk, skb, limit))) {
 		bh_unlock_sock(sk);
+		*reason = SKB_DROP_REASON_SOCKET_BACKLOG;
 		__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPBACKLOGDROP);
 		return true;
 	}
@@ -1971,13 +1981,13 @@ static void tcp_v4_fill_cb(struct sk_buff *skb, const struct iphdr *iph,
 int tcp_v4_rcv(struct sk_buff *skb)
 {
 	struct net *net = dev_net(skb->dev);
+	enum skb_drop_reason drop_reason;
 	int sdif = inet_sdif(skb);
 	int dif = inet_iif(skb);
 	const struct iphdr *iph;
 	const struct tcphdr *th;
 	bool refcounted;
 	struct sock *sk;
-	int drop_reason;
 	int ret;
 
 	drop_reason = SKB_DROP_REASON_NOT_SPECIFIED;
@@ -2025,7 +2035,8 @@ int tcp_v4_rcv(struct sk_buff *skb)
 		struct sock *nsk;
 
 		sk = req->rsk_listener;
-		if (unlikely(tcp_v4_inbound_md5_hash(sk, skb, dif, sdif))) {
+		if (unlikely(tcp_v4_inbound_md5_hash(sk, skb, dif, sdif,
+						     &drop_reason))) {
 			sk_drops_add(sk, skb);
 			reqsk_put(req);
 			goto discard_it;
@@ -2057,6 +2068,8 @@ int tcp_v4_rcv(struct sk_buff *skb)
 			iph = ip_hdr(skb);
 			tcp_v4_fill_cb(skb, iph, th);
 			nsk = tcp_check_req(sk, skb, req, false, &req_stolen);
+		} else {
+			drop_reason = SKB_DROP_REASON_SOCKET_FILTER;
 		}
 		if (!nsk) {
 			reqsk_put(req);
@@ -2092,10 +2105,12 @@ int tcp_v4_rcv(struct sk_buff *skb)
 		}
 	}
 
-	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
+	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) {
+		drop_reason = SKB_DROP_REASON_XFRM_POLICY;
 		goto discard_and_relse;
+	}
 
-	if (tcp_v4_inbound_md5_hash(sk, skb, dif, sdif))
+	if (tcp_v4_inbound_md5_hash(sk, skb, dif, sdif, &drop_reason))
 		goto discard_and_relse;
 
 	nf_reset_ct(skb);
@@ -2124,7 +2139,7 @@ int tcp_v4_rcv(struct sk_buff *skb)
 	if (!sock_owned_by_user(sk)) {
 		ret = tcp_v4_do_rcv(sk, skb);
 	} else {
-		if (tcp_add_backlog(sk, skb))
+		if (tcp_add_backlog(sk, skb, &drop_reason))
 			goto discard_and_relse;
 	}
 	bh_unlock_sock(sk);
@@ -2166,6 +2181,7 @@ int tcp_v4_rcv(struct sk_buff *skb)
 
 do_time_wait:
 	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) {
+		drop_reason = SKB_DROP_REASON_XFRM_POLICY;
 		inet_twsk_put(inet_twsk(sk));
 		goto discard_it;
 	}