On Mon, Oct 11, 2010 at 12:06:32PM -0400, Alan Stern wrote: > On Sun, 10 Oct 2010, Dmitry Torokhov wrote: > > > > Will the driver use autosuspend for the 30-second delay? Then all you > > > have to do is this: When the button is pressed, write 0 to > > > power/autosuspend_delay_ms, causing an immediate runtime suspend. > > > Then before turning the display back on, write 30000 to set the delay > > > to 30 seconds again. You can leave power/control set to "auto" the > > > whole time. > > > > > > > This means messing up with the currently set timeout. If userspace could > > have an option of "accelerating" expiring of the current timeout that > > would be better. > > Accelerating the expiration of the timeout _is_ a form of messing with > the currently-set timeout. It can't be "better". I was thinking about the case where timeout is user-supplied setting. Instead of having application store the original value, write 0, and later on restore it, I thought it would be better (as in simpler) to have a separate sysfs entry to "expire" the original timeout immediately. > > > > I don't like the idea of having an "off" setting in power/control. > > > What happens if the user turns a disk drive controller "off" and the > > > system needs to read or write something on that disk drive? > > > > > > > I see. Well, for some devices "sicky off" > > "sticky off"? The device that stays off and doe snot power itself up even if it is needed. > > > maybe the one that makes most > > sense. For others we may need an option that would put them into low > > power mode while allowing bringing them back if they are needed. > > Can you give any examples where "sticky off" is really useful (other > than just for debugging)? > A mobile device might have a set of keys that, once device is put into lower power mode, should not bring the device out of that mode. This I would call a "sticky off" case. > > In case of touchscreens/keypads they are most likely configured as > > wakeup sources so when user actually tried to use the device they'll be > > brought online. > > I agree, but it's important to distinguish between wakeup sources in > particular and interrupt sources in general. As long as the system > as a whole isn't suspended (it might be running or it might be in > cpuidle), any interrupt source can cause a runtime resume. By > contrast, when the system is suspended then only wakeup sources will > cause anything to happen. Right. -- Dmitry _______________________________________________ linux-pm mailing list linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/linux-pm
