Re: [PATCH net-next 0/3] introduce two new tracepoints for udp

[David Miller]

From: Tony Lu <tonylu@xxxxxxxxxxxxxxxxx>
Date: Wed, 29 May 2019 21:06:54 +0800

> This series introduces two new tracepoints trace_udp_send and
> trace_udp_queue_rcv, and removes redundant new line from
> tcp_event_sk_skb.

Why?

Is it faster than using kprobes?

Is it more reliable?

Are the events _so_ useful that they warrant a tracepoint and thus
creating a semi-stable interface for tracing and introspection via
ebpf and similar technologies?

Again, you have to say why in your log message(s).