Re: [PATCH net-next v2 2/2] trace: tcp: Add tracepoint for tcp_sendmsg_locked()

[Kuniyuki Iwashima]

From: Breno Leitao <leitao@xxxxxxxxxx>
Date: Mon, 07 Apr 2025 06:40:44 -0700
> Add a tracepoint to monitor TCP send operations, enabling detailed
> visibility into TCP message transmission.
> 
> Create a new tracepoint within the tcp_sendmsg_locked function,
> capturing traditional fields along with size_goal, which indicates the
> optimal data size for a single TCP segment. Additionally, a reference to
> the struct sock sk is passed, allowing direct access for BPF programs.
> The implementation is largely based on David's patch and suggestions.
> 
> The implementation is largely based on David's patch[1] and suggestions.

nit: duplicate sentences.


> 
> Link: https://lore.kernel.org/all/70168c8f-bf52-4279-b4c4-be64527aa1ac@xxxxxxxxxx/ [1]
> Signed-off-by: Breno Leitao <leitao@xxxxxxxxxx>
> ---
>  include/trace/events/tcp.h | 24 ++++++++++++++++++++++++
>  net/ipv4/tcp.c             |  2 ++
>  2 files changed, 26 insertions(+)
> 
> diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
> index 1a40c41ff8c30..cab25504c4f9d 100644
> --- a/include/trace/events/tcp.h
> +++ b/include/trace/events/tcp.h
> @@ -259,6 +259,30 @@ TRACE_EVENT(tcp_retransmit_synack,
>  		  __entry->saddr_v6, __entry->daddr_v6)
>  );
>  
> +TRACE_EVENT(tcp_sendmsg_locked,
> +	TP_PROTO(const struct sock *sk, const struct msghdr *msg,
> +		 const struct sk_buff *skb, int size_goal),
> +
> +	TP_ARGS(sk, msg, skb, size_goal),
> +
> +	TP_STRUCT__entry(
> +		__field(const void *, skb_addr)
> +		__field(int, skb_len)
> +		__field(int, msg_left)
> +		__field(int, size_goal)
> +	),
> +
> +	TP_fast_assign(
> +		__entry->skb_addr = skb;
> +		__entry->skb_len = skb ? skb->len : 0;
> +		__entry->msg_left = msg_data_left(msg);
> +		__entry->size_goal = size_goal;
> +	),
> +
> +	TP_printk("skb_addr %p skb_len %d msg_left %d size_goal %d",
> +		__entry->skb_addr, __entry->skb_len, __entry->msg_left,
> +		__entry->size_goal));
> +
>  DECLARE_TRACE(tcp_cwnd_reduction_tp,
>  	TP_PROTO(const struct sock *sk, int newly_acked_sacked,
>  		 int newly_lost, int flag),
> diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
> index ea8de00f669d0..270ce2c8c2d54 100644
> --- a/net/ipv4/tcp.c
> +++ b/net/ipv4/tcp.c
> @@ -1160,6 +1160,8 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size)
>  		if (skb)
>  			copy = size_goal - skb->len;
>  
> +		trace_tcp_sendmsg_locked(sk, msg, skb, size_goal);

skb could be NULL, so I think raw_tp_null_args[] needs to be updated.

Maybe try attaching a bpf prog that dereferences skb unconditionally
and see if the bpf verifier rejects it.

See this commit for the similar issue:

commit 5da7e15fb5a12e78de974d8908f348e279922ce9
Author: Kuniyuki Iwashima <kuniyu@xxxxxxxxxx>
Date:   Fri Jan 31 19:01:42 2025 -0800

    net: Add rx_skb of kfree_skb to raw_tp_null_args[].


> +
>  		if (copy <= 0 || !tcp_skb_can_collapse_to(skb)) {
>  			bool first_skb;
>  
> 
> -- 
> 2.47.1