From mboxrd@z Thu Jan  1 00:00:00 1970
From: alexandre.belloni@free-electrons.com (Alexandre Belloni)
Date: Fri, 16 Jan 2015 12:07:42 +0100
Subject: [RFC 2/2] clocksource: don't suspend/resume when unused
In-Reply-To: <54B8EF15.4000602@linaro.org>
References: <1421399874-29119-1-git-send-email-alexandre.belloni@free-electrons.com>
 <1421399874-29119-3-git-send-email-alexandre.belloni@free-electrons.com>
 <alpine.DEB.2.11.1501161122320.5526@nanos>
 <20150116103530.GB3843@piout.net> <54B8EA54.5020002@linaro.org>
 <20150116104804.GE3843@piout.net> <54B8EF15.4000602@linaro.org>
Message-ID: <20150116110742.GJ3843@piout.net>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On 16/01/2015 at 11:59:33 +0100, Daniel Lezcano wrote :
> On 01/16/2015 11:48 AM, Alexandre Belloni wrote:
> >Hi,
> >
> >On 16/01/2015 at 11:39:16 +0100, Daniel Lezcano wrote :
> >>>Isn't that already the case?
> >>>Right now, if you call clocksource_suspend, it doesn't matter whether
> >>>the clocksource has an enable or not, it will be suspended. Maybe I'm
> >>>mistaken but my patch doesn't seem to change that behaviour.
> >>
> >>Actually, if there is no enable/disable callback, then CLOCK_SOURCE_USED
> >>will be never set, hence the condition will always fail and the suspend
> >>callback won't be called.
> >>
> >
> >It is set in clocksource_enable/disable, even if there is no
> >enable/disable callback.
> 
> Ah, right. But shouldn't we set the flag only if the callback is present and
> succeed as Boris mentioned it ?
> 

What Boris was suggesting was that if the enable exist, set it only if
it succeed. Which gives something like that:

int clocksource_enable(struct clocksource *cs)
{
	int ret = 0;

	if (cs->enable)
		ret = cs->enable(cs);

	if (!ret)
		cs->flags |= CLOCK_SOURCE_USED;

	return 0;
}

I will use that version in v2.


-- 
Alexandre Belloni, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com