Re: [linux-pm] [RFC] input: syfs switches for SKE keypad

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Sunday, October 10, 2010, Alan Stern wrote:
> On Sat, 9 Oct 2010, Rafael J. Wysocki wrote:
> 
> > On Wednesday, October 06, 2010, Alan Stern wrote:
> > > On Wed, 6 Oct 2010, Rafael J. Wysocki wrote:
> > > 
> > > > > Mobile folks wish to power down some devices (most often input -
> > > > > touchscreen, keypad) under certain circumstances to save power.
> > > > > So far they were doing that by adding "disable" hook to individual
> > > > > drivers and while I did allow that in for some devices I feel that we
> > > > > need more standardised solution, preferably one that could re-use
> > > > > existing PM hooks in drivers.
> > > > 
> > > > There's no interface for that at the PM core level, but I think we should
> > > > add it, perhaps in analogy to the autosuspend one.  Namely, we can add a
> > > > flag for drivers who want to make their suspend/resume callbacks to be
> > > > reachable directly from user space.  Setting that flag would enable a sysfs
> > > > attribute in /sys/devices/.../power/ allowing user space to invoke
> > > > pm_runtime_suspend() and pm_runtime_resume() for the given device.
> > > 
> > > We already have power/control.  If the subsystem sets it to "on" by 
> > > default and the driver suspends the device whenever it is idle, then 
> > > userspace can control the power level by writing "auto" or "on" to 
> > > power/control.
> > 
> > The problem is that with the 'auto' setting the driver decides when to suspend
> > and the driver need not know it's the right time.
> > 
> > Suspending a graphics adapter when the user presses a "turn screen off"
> > button it a good example of this.  The graphics driver may not know the button
> > was pressed and it has to be told about that.  OTOH the button driver need
> > not know what exactly should be done when it is pressed.  There may be a user
> > space component in between that processes the button event and should be able
> > to tell the graphics driver to suspend.
> 
> You are clearly right; userspace has to tell the driver when it is okay
> to suspend.  My point was that we already have a mechanism in the PM
> core for doing this -- assuming the driver is written appropriately (to 
> assume that it should try to suspend whenever possible).  We shouldn't 
> need to add another mechanism.

OK, so how is a graphics driver going to figure out it should suspend when the
button is pressed in the example above?

Rafael