[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <7hpowpr0da.fsf@baylibre.com>
Date: Mon, 25 Jan 2016 10:16:01 -0800
From: Kevin Hilman <khilman@...libre.com>
To: Tony Lindgren <tony@...mide.com>
Cc: Pali Rohár <pali.rohar@...il.com>,
Sebastian Reichel <sre@...nel.org>,
Ivaylo Dimitrov <ivo.g.dimitrov.75@...il.com>,
Pavel Machek <pavel@....cz>,
Aaro Koskinen <aaro.koskinen@....fi>,
Nishanth Menon <nm@...com>, linux-omap@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: Missing OMAP PM layer
Tony Lindgren <tony@...mide.com> writes:
> * Pali Rohár <pali.rohar@...il.com> [160124 12:24]:
>> Hello,
>>
>> make menuconfig allows me to choose "OMAP PM layer selection" and the
>> only one option is CONFIG_OMAP_PM_NOOP "No-op/debug PM layer".
>>
>> What does it mean? Power manager is noop?
>>
>> I see that it has only two corresponding files in mainline kernel:
>> arch/arm/mach-omap2/omap-pm.h
>> arch/arm/mach-omap2/omap-pm-noop.c
>>
>> Nokia's kernels (for N900 and N950) had also:
>> arch/arm/mach-omap2/omap-pm-srf.c
>>
>> Can somebody explain it what happened with omap power management?
>>
>> Looks like that omap-pm.h provides some API, but the only implementation
>> is noop which do nothing.
>
> I believe none of that is needed any longer in mainline.
>
> Kevin, care to descrbibe what should be done here?
We had created the OMAP PM layer as a pluggable layer where we could
experiment with different approaches for adding constratints, etc. I
can't remember anymore what all was in there, but we gutted most of it
after switching to runtime PM.
The SRF was a "shared resource framework" that came out of a TI kernel
that was never accepted upstream either, but was one of the
implementation of the OMAP PM layer.
Kevin
Powered by blists - more mailing lists