Skip to contentSkip to conversation

tcp: make trace of reset logic complete

PATCH Applied Jason Xing 2 participants 4 messages Apr 1, 2024 – Apr 4, 2024

Subsystems (5)
lore
Share & tools
lore

Jason Xing

author
#0 lore raw

v4 cover tcp: make trace of reset logic complete

From: Jason Xing <[email protected]>

Before this, we miss some cases where the TCP layer could send RST but we cannot trace it. So I decided to complete it :)

v4

1. rebased against latest net-next
2. remove {} and add skb test statement (Eric)
3. drop v3 patch [3/3] temporarily because 1) location is not that useful since we can use perf or something else to trace, 2) Eric said we could use drop_reason to show why we have to RST, which is good, but this seems not work well for those ->send_reset() logic. I need more time to investigate this part.

v3
1. fix a format problem in patch [3/3]

v2
1. fix spelling mistakes

Jason Xing (2):
trace: adjust TP_STORE_ADDR_PORTS_SKB() parameters
trace: tcp: fully support trace_tcp_send_reset

 include/trace/events/net_probe_common.h | 20 ++++++------
 include/trace/events/tcp.h              | 42 +++++++++++++++++++++++--
 include/trace/events/udp.h              |  2 +-
 net/ipv4/tcp_ipv4.c                     |  7 ++---
 net/ipv6/tcp_ipv6.c                     |  3 +-
 5 files changed, 56 insertions(+), 18 deletions(-)
Signature · 1 line
2.37.3
author
#1 lore raw

PATCH v4 1/2 trace: adjust TP_STORE_ADDR_PORTS_SKB() parameters

From: Jason Xing <[email protected]>

Introducing entry_saddr and entry_daddr parameters in this macro for later use can help us record the reverse 4-tuple by analyzing the 4-tuple of the incoming skb when receiving.

---
 include/trace/events/net_probe_common.h | 20 +++++++++++---------
 include/trace/events/tcp.h              |  2 +-
 include/trace/events/udp.h              |  2 +-
 3 files changed, 13 insertions(+), 11 deletions(-)
include/trace/events/net_probe_common.h +11 -9
index 5e33f91bdea3..976a58364bff 100644
@@ -70,14 +70,14 @@
 	TP_STORE_V4MAPPED(__entry, saddr, daddr)
 #endif
 
-#define TP_STORE_ADDR_PORTS_SKB_V4(__entry, skb, protoh)		\
+#define TP_STORE_ADDR_PORTS_SKB_V4(skb, protoh, entry_saddr, entry_daddr) \
 	do {								\
-		struct sockaddr_in *v4 = (void *)__entry->saddr;	\
+		struct sockaddr_in *v4 = (void *)entry_saddr;		\
 									\
 		v4->sin_family = AF_INET;				\
 		v4->sin_port = protoh->source;				\
 		v4->sin_addr.s_addr = ip_hdr(skb)->saddr;		\
-		v4 = (void *)__entry->daddr;				\
+		v4 = (void *)entry_daddr;				\
 		v4->sin_family = AF_INET;				\
 		v4->sin_port = protoh->dest;				\
 		v4->sin_addr.s_addr = ip_hdr(skb)->daddr;		\
@@ -85,28 +85,30 @@
 
 #if IS_ENABLED(CONFIG_IPV6)
 
-#define TP_STORE_ADDR_PORTS_SKB(__entry, skb, protoh)			\
+#define TP_STORE_ADDR_PORTS_SKB(skb, protoh, entry_saddr, entry_daddr)	\
 	do {								\
 		const struct iphdr *iph = ip_hdr(skb);			\
 									\
 		if (iph->version == 6) {				\
-			struct sockaddr_in6 *v6 = (void *)__entry->saddr; \
+			struct sockaddr_in6 *v6 = (void *)entry_saddr;	\
 									\
 			v6->sin6_family = AF_INET6;			\
 			v6->sin6_port = protoh->source;			\
 			v6->sin6_addr = ipv6_hdr(skb)->saddr;		\
-			v6 = (void *)__entry->daddr;			\
+			v6 = (void *)entry_daddr;			\
 			v6->sin6_family = AF_INET6;			\
 			v6->sin6_port = protoh->dest;			\
 			v6->sin6_addr = ipv6_hdr(skb)->daddr;		\
 		} else							\
-			TP_STORE_ADDR_PORTS_SKB_V4(__entry, skb, protoh); \
+			TP_STORE_ADDR_PORTS_SKB_V4(skb, protoh,		\
+						   entry_saddr,		\
+						   entry_daddr);	\
 	} while (0)
 
 #else
 
-#define TP_STORE_ADDR_PORTS_SKB(__entry, skb, protoh)		\
-	TP_STORE_ADDR_PORTS_SKB_V4(__entry, skb, protoh)
+#define TP_STORE_ADDR_PORTS_SKB(skb, protoh, entry_saddr, entry_daddr)	\
+	TP_STORE_ADDR_PORTS_SKB_V4(skb, protoh, entry_saddr, entry_daddr)
 
 #endif
 
include/trace/events/tcp.h +1 -1
index 1db95175c1e5..cf14b6fcbeed 100644
@@ -295,7 +295,7 @@ DECLARE_EVENT_CLASS(tcp_event_skb,
 		memset(__entry->saddr, 0, sizeof(struct sockaddr_in6));
 		memset(__entry->daddr, 0, sizeof(struct sockaddr_in6));
 
-		TP_STORE_ADDR_PORTS_SKB(__entry, skb, th);
+		TP_STORE_ADDR_PORTS_SKB(skb, th, __entry->saddr, __entry->daddr);
 	),
 
 	TP_printk("skbaddr=%p src=%pISpc dest=%pISpc",
include/trace/events/udp.h +1 -1
index 62bebe2a6ece..6142be4068e2 100644
@@ -38,7 +38,7 @@ TRACE_EVENT(udp_fail_queue_rcv_skb,
 		memset(__entry->saddr, 0, sizeof(struct sockaddr_in6));
 		memset(__entry->daddr, 0, sizeof(struct sockaddr_in6));
 
-		TP_STORE_ADDR_PORTS_SKB(__entry, skb, uh);
+		TP_STORE_ADDR_PORTS_SKB(skb, uh, __entry->saddr, __entry->daddr);
 	),
 
 	TP_printk("rc=%d family=%s src=%pISpc dest=%pISpc", __entry->rc,
Signature · 1 line
2.37.3
author
#2 lore raw

PATCH v4 2/2 trace: tcp: fully support trace_tcp_send_reset

In reply to Jason Xing

From: Jason Xing <[email protected]>

Prior to this patch, what we can see by enabling trace_tcp_send is only happening under two circumstances:
1) active rst mode
2) non-active rst mode and based on the full socket

That means the inconsistency occurs if we use tcpdump and trace simultaneously to see how rst happens.

It's necessary that we should take into other cases into considerations, say:
1) time-wait socket
2) no socket
...

By parsing the incoming skb and reversing its 4-tuple can we know the exact 'flow' which might not exist.

Samples after applied this patch:
1. tcp_send_reset: skbaddr=XXX skaddr=XXX src=ip:port dest=ip:port state=TCP_ESTABLISHED
2. tcp_send_reset: skbaddr=000...000 skaddr=XXX src=ip:port dest=ip:port state=UNKNOWN
Note:
1) UNKNOWN means we cannot extract the right information from skb.
2) skbaddr/skaddr could be 0

---
 include/trace/events/tcp.h | 40 ++++++++++++++++++++++++++++++++++++--
 net/ipv4/tcp_ipv4.c        |  7 +++----
 net/ipv6/tcp_ipv6.c        |  3 ++-
 3 files changed, 43 insertions(+), 7 deletions(-)
include/trace/events/tcp.h +38 -2
index cf14b6fcbeed..5c04a61a11c2 100644
@@ -78,11 +78,47 @@ DEFINE_EVENT(tcp_event_sk_skb, tcp_retransmit_skb,
  * skb of trace_tcp_send_reset is the skb that caused RST. In case of
  * active reset, skb should be NULL
  */
-DEFINE_EVENT(tcp_event_sk_skb, tcp_send_reset,
+TRACE_EVENT(tcp_send_reset,
 
 	TP_PROTO(const struct sock *sk, const struct sk_buff *skb),
 
-	TP_ARGS(sk, skb)
+	TP_ARGS(sk, skb),
+
+	TP_STRUCT__entry(
+		__field(const void *, skbaddr)
+		__field(const void *, skaddr)
+		__field(int, state)
+		__array(__u8, saddr, sizeof(struct sockaddr_in6))
+		__array(__u8, daddr, sizeof(struct sockaddr_in6))
+	),
+
+	TP_fast_assign(
+		__entry->skbaddr = skb;
+		__entry->skaddr = sk;
+		/* Zero means unknown state. */
+		__entry->state = sk ? sk->sk_state : 0;
+
+		memset(__entry->saddr, 0, sizeof(struct sockaddr_in6));
+		memset(__entry->daddr, 0, sizeof(struct sockaddr_in6));
+
+		if (sk && sk_fullsock(sk)) {
+			const struct inet_sock *inet = inet_sk(sk);
+
+			TP_STORE_ADDR_PORTS(__entry, inet, sk);
+		} else if (skb) {
+			const struct tcphdr *th = (const struct tcphdr *)skb->data;
+			/*
+			 * We should reverse the 4-tuple of skb, so later
+			 * it can print the right flow direction of rst.
+			 */
+			TP_STORE_ADDR_PORTS_SKB(skb, th, entry->daddr, entry->saddr);
+		}
+	),
+
+	TP_printk("skbaddr=%p skaddr=%p src=%pISpc dest=%pISpc state=%s",
+		  __entry->skbaddr, __entry->skaddr,
+		  __entry->saddr, __entry->daddr,
+		  __entry->state ? show_tcp_state_name(__entry->state) : "UNKNOWN")
 );
 
 /*
net/ipv4/tcp_ipv4.c +3 -4
index a22ee5838751..0d47b48f8cfd 100644
@@ -866,11 +866,10 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 	 * routing might fail in this case. No choice here, if we choose to force
 	 * input interface, we will misroute in case of asymmetric route.
 	 */
-	if (sk) {
+	if (sk)
 		arg.bound_dev_if = sk->sk_bound_dev_if;
-		if (sk_fullsock(sk))
-			trace_tcp_send_reset(sk, skb);
-	}
+
+	trace_tcp_send_reset(sk, skb);
 
 	BUILD_BUG_ON(offsetof(struct sock, sk_bound_dev_if) !=
 		     offsetof(struct inet_timewait_sock, tw_bound_dev_if));
net/ipv6/tcp_ipv6.c +2 -1
index 3f4cba49e9ee..8e9c59b6c00c 100644
@@ -1113,7 +1113,6 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 	if (sk) {
 		oif = sk->sk_bound_dev_if;
 		if (sk_fullsock(sk)) {
-			trace_tcp_send_reset(sk, skb);
 			if (inet6_test_bit(REPFLOW, sk))
 				label = ip6_flowlabel(ipv6h);
 			priority = READ_ONCE(sk->sk_priority);
@@ -1129,6 +1128,8 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 			label = ip6_flowlabel(ipv6h);
 	}
 
+	trace_tcp_send_reset(sk, skb);
+
 	tcp_v6_send_response(sk, skb, seq, ack_seq, 0, 0, 0, oif, 1,
 			     ipv6_get_dsfield(ipv6h), label, priority, txhash,
 			     &key);
Signature · 1 line
2.37.3
#3 lore raw
In reply to Jason Xing
This series was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Hello:

This series was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Quoted 15 lines · On Mon, 1 Apr 2024 15:36:03 +0800 you wrote:

From: Jason Xing <[email protected]>

Before this, we miss some cases where the TCP layer could send RST but we cannot trace it. So I decided to complete it :)

v4

1. rebased against latest net-next
2. remove {} and add skb test statement (Eric)
3. drop v3 patch [3/3] temporarily because 1) location is not that useful since we can use perf or something else to trace, 2) Eric said we could use drop_reason to show why we have to RST, which is good, but this seems not work well for those ->send_reset() logic. I need more time to investigate this part.

Quoted 1 line

[...]

Here is the summary with links:
- [net-next,v4,1/2] trace: adjust TP_STORE_ADDR_PORTS_SKB() parameters

    https://git.kernel.org/netdev/net-next/c/9807080e2170
  - [net-next,v4,2/2] trace: tcp: fully support trace_tcp_send_reset
    https://git.kernel.org/netdev/net-next/c/19822a980e19

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html