From mboxrd@z Thu Jan  1 00:00:00 1970
From: Peter Zijlstra <peterz@infradead.org>
Subject: Re: [PATCH v2] cpuidle: Add 'above' and 'below' idle state metrics
Date: Mon, 10 Dec 2018 23:51:34 +0100
Message-ID: <20181210225134.GB24811@worktop.programming.kicks-ass.net>
References: <3514439.dzOWKx1Cjx@aspire.rjw.lan>
 <20181210122104.GL5289@hirez.programming.kicks-ass.net>
 <CAJZ5v0j-a6cHayRtcsUPbUTBT0iBSUvmcUy6pfSSvUVcTMBxbw@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Return-path: <linux-kernel-owner@vger.kernel.org>
Content-Disposition: inline
In-Reply-To: <CAJZ5v0j-a6cHayRtcsUPbUTBT0iBSUvmcUy6pfSSvUVcTMBxbw@mail.gmail.com>
Sender: linux-kernel-owner@vger.kernel.org
To: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>, Linux PM <linux-pm@vger.kernel.org>, Doug Smythies <dsmythies@telus.net>, Linux Kernel Mailing List <linux-kernel@vger.kernel.org>, "open list:DOCUMENTATION" <linux-doc@vger.kernel.org>, Daniel Lezcano <daniel.lezcano@linaro.org>, Giovanni Gherdovich <ggherdovich@suse.cz>, Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
List-Id: linux-pm@vger.kernel.org

On Mon, Dec 10, 2018 at 10:36:40PM +0100, Rafael J. Wysocki wrote:
> On Mon, Dec 10, 2018 at 1:21 PM Peter Zijlstra <peterz@infradead.org> wrote:

> > One question on this; why is this tracked unconditionally?
> 
> Because I didn't quite see how to make that conditional in a sensible way.

Something like:

	if (static_branch_unlikely(__tracepoint_idle_above) ||
	    static_branch_unlikely(__tracepoint_idle_below)) {

		// do stuff that calls trace_idle_above() /
		// trace_idle_below().

	}

> These things are counters and counting with the help of tracepoints
> isn't particularly convenient (and one needs debugfs to be there to
> use tracepoints and they require root access etc).

Root only should not be a problem for a developer; and aren't these
numbers only really interesting if you're prodding at the idle governor?

> > Would not a tracepoint be better?; then there is no overhead in the
> > normal case where nobody gives a crap about these here numbers.
> 
> There is an existing tracepoint that in principle could be used to
> produce this information, but it is such a major PITA in practice that
> nobody does that.  Guess why. :-)

Sounds like you need to ship a convenient script or something :-)

> Also, the "usage" and "time" counters are there in sysfs, so why not these two?
> 
> And is the overhead really that horrible?

Dunno; it could be cold cachelines, at which point it can be fairly
expensive. Also, being stuck with API is fairly horrible if you want to
'fix' it.