[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <200907300127.19358.rjw@sisk.pl>
Date: Thu, 30 Jul 2009 01:27:18 +0200
From: "Rafael J. Wysocki" <rjw@...k.pl>
To: Dmitry Torokhov <dmitry.torokhov@...il.com>
Cc: Arnaud Faucher <arnaud.faucher@...il.com>,
Carlos Corbacho <carlos@...angeworlds.co.uk>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Frans Pop <elendil@...net.nl>,
Manuel Lauss <manuel.lauss@...il.com>,
Erik Ekman <erik@...o.se>,
Mark Brown <broonie@...nsource.wolfsonmicro.com>
Subject: Re: [PATCH 1/1] acer-wmi: switch driver to dev_pm_ops
On Wednesday 29 July 2009, Dmitry Torokhov wrote:
> On Wed, Jul 29, 2009 at 10:49:46PM +0200, Rafael J. Wysocki wrote:
> > On Wednesday 29 July 2009, Arnaud Faucher wrote:
> >
> > > As per the pm.h documentation .thaw is called after RAM image has been
> > > created, in order to restore hardware state in case RAM image failed and
> > > the system cannot power off.
> >
> > That's not correct (please see above). .thaw() is called after creating the
> > image in case .freeze() has changed the state of the device. This often is not
> > necessary, though, so .thaw() may be skipped in many cases. Of course, you
> > should know exactly what you're doing.
> >
>
> Umm, but thaw() _is_ called in case of hibernate failure:
>
> case PM_EVENT_THAW:
> case PM_EVENT_RECOVER:
> if (ops->thaw) {
> error = ops->thaw(dev);
> suspend_report_result(ops->thaw, error);
> }
> break;
>
> so I don't believe you can easily skip thaw() if you have freeze() that
> stops/resets device.
_RECOVER means "image creation failed or unable to run the image kernel", so
from the devices' point of view it's the same as _THAW.
Still, if your .freeze() modifies the device's registers, then most likely
.thaw() _is_ necessary.
Best,
Rafael
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists