Re: [PATCH 09/12] trace-cmd analyze: Add wake up latency timings

From: Steven Rostedt <rostedt@goodmis.org>
To: John Keeping <john@metanate.com>
Cc: linux-trace-devel@vger.kernel.org, williskung@google.com,
	kaleshsingh@google.com, linux-rt-users@vger.kernel.org
Subject: Re: [PATCH 09/12] trace-cmd analyze: Add wake up latency timings
Date: Fri, 25 Mar 2022 16:31:48 -0400	[thread overview]
Message-ID: <20220325163148.0bf4146b@gandalf.local.home> (raw)
In-Reply-To: <20220325161333.65ffb789@gandalf.local.home>

On Fri, 25 Mar 2022 16:13:33 -0400
Steven Rostedt <rostedt@goodmis.org> wrote:

> I'll fix it in v2.

BTW, I plan on rewriting a lot of this to move it into libtracecmd. That
way other tools can easily extract data from it.

For example, instead of hard coding "page faults" and "wake ups" I plan on
adding a way to register events to trigger on. Something like:

 tracecmd_analyze_register_event_timing(handle, name, start_event,
		 start_field, start_record, end_event, end_field,
		 end_record);

Where "start_record" and "end_record" is if the field is a task to record.

enum {
	TRACECMD_ANALYZE_FL_NONE,
	TRACECMD_ANALYZE_FL_PID,
	TRACECMD_ANALYZE_FL_PID_CPU,
};

Thus, we could have:

	tracecmd_analyze_register_event_timing(handle,
			"Wakeup", sched_waking,
			pid_field, TRACECMD_ANALYZE_FL_PID,
			sched_switch, next_pid,
			TRACECMD_ANALYZE_FL_PID_CPU);

Or something like the above, that will end up turning into the
process_wakeup() function (and properly record the waking pid globally).

One tool I want to use this with, is my ktrace shell (development has
stalled, but I do plan on starting it again). 

  https://git.kernel.org/pub/scm/linux/kernel/git/rostedt/ktrace.git/

And then I could have commands like:

ktrace> open trace.dat
ktrace> analyze register timing sched.sched_switch pid global \
  sched.sched_switch next_pid cpu
ktrace> analyze
ktrace> analyze show cpu 1

And it only shows cpu 1 data.

ktrace> analyze show pid 512

to see just pid 512

etc.

-- Steve