[Patch net-next 03/13] tcp: introduce tracepoint trace_tcp_transmit_skb()
From: Cong Wang <hidden>
Date: 2021-08-05 18:58:02
Subsystem:
networking [general], networking [tcp], the rest, tracing · Maintainers:
"David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Neal Cardwell, Linus Torvalds, Steven Rostedt, Masami Hiramatsu
From: Qitao Xu <redacted> Tracepoint trace_tcp_transmit_skb() is introduced to trace skb at the exit of TCP layer on TX side. Reviewed-by: Cong Wang <redacted> Signed-off-by: Qitao Xu <redacted> --- include/trace/events/tcp.h | 11 +++++++++-- net/ipv4/tcp_output.c | 8 +++++++- 2 files changed, 16 insertions(+), 3 deletions(-)
diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index 521059d8dc0a..6147528d93ba 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h@@ -88,11 +88,18 @@ DECLARE_EVENT_CLASS(tcp_event_sk_skb, sk->sk_v6_rcv_saddr, sk->sk_v6_daddr); ), - TP_printk("family=%s sport=%hu dport=%hu saddr=%pI4 daddr=%pI4 saddrv6=%pI6c daddrv6=%pI6c state=%s", + TP_printk("family=%s sport=%hu dport=%hu saddr=%pI4 daddr=%pI4 saddrv6=%pI6c daddrv6=%pI6c state=%s skbaddr=%px", show_family_name(__entry->family), __entry->sport, __entry->dport, __entry->saddr, __entry->daddr, __entry->saddr_v6, __entry->daddr_v6, - show_tcp_state_name(__entry->state)) + show_tcp_state_name(__entry->state), __entry->skbaddr) +); + +DEFINE_EVENT(tcp_event_sk_skb, tcp_transmit_skb, + + TP_PROTO(const struct sock *sk, const struct sk_buff *skb), + + TP_ARGS(sk, skb) ); DEFINE_EVENT(tcp_event_sk_skb, tcp_retransmit_skb,
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 29553fce8502..628b28332851 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c@@ -1420,8 +1420,14 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, gfp_t gfp_mask) { - return __tcp_transmit_skb(sk, skb, clone_it, gfp_mask, + int ret; + + ret = __tcp_transmit_skb(sk, skb, clone_it, gfp_mask, tcp_sk(sk)->rcv_nxt); + if (!ret) + trace_tcp_transmit_skb(sk, skb); + return ret; + } /* This routine just queues the buffer for sending.
--
2.27.0