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: Sat, 26 Mar 2022 11:32:32 -0400	[thread overview]
Message-ID: <20220326113232.0dbfd503@rorschach.local.home> (raw)
In-Reply-To: <Yj71frLcA+W9hkMw@donbot>

On Sat, 26 Mar 2022 11:14:06 +0000
John Keeping <john@metanate.com> wrote:

> I started having a look at whether it's possible to add an overview of
> interrupts from irq:irq_handler_entry & irq:irq_handler_exit and I guess
> that would fit in TRACECMD_ANALYZE_FL_NONE above, but it would be good
> to have per-cpu stats for irqs as well so does that mean an additional
> record type is needed?

I actually created that enum for exactly that ;-) To be able to add
different metrics in the future. Which is why it states "PID". I
actually did think about adding things like

  TRACECMD_ANALYZE_FL_IRQ

To say this is to add irq interruptions. Perhaps even add:

  TRACECMD_ANALYZE_FL_STOLEN

for guest stolen times.

I haven't decided on that API yet. Perhaps we could add new custom
timings too, and include them somehow. I have to think about a good API
to use.

-- Steve