Re: klists and struct device semaphores

[Patrick Mochel]

On Thu, 31 Mar 2005, David Brownell wrote:

> On Thursday 31 March 2005 10:26 am, Patrick Mochel wrote:

> > Conversely, we only want to automatically suspend the device, or allow the
> > device to be explicitly put to sleep, if the device is not being used.
>
> And be able to suspend itself, even if it's open, if it's idle enough and
> can wake itself up automatically.

Certainly that's dependent on the type of device. It'll be relatively easy
to begin with devices at an open()/close() level, and consider any device
that's open() to be active and ineligible for suspend. This will allow us
to setup the infrastructure and get some decent savings from definitely
inactive devices.

We then move to a more fine-grained activity-detection and reaction
mechanism, like being idle after N seconds of inactivity. But, that will
be completely dependent on the type of device and the driver.

In fact with a usage counter, a driver that wants to be more intelligent
about how much it's being used can simply use that (decrement it and
automatically suspend) when it's inactive for a while.

Right?


	Pat
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/