On 7/29/06, Mark Underwood <basicmark@xxxxxxxxx> wrote:
> The lazy polling approach I described in my last post to Vojtech
> ("block until there's a new readout or N milliseconds have passed,
> whichever is later") looks like a more general, accurate and efficient
> interface.
This sounds like a good idea. You could do a similar thing using sysfs by
providing a entry in sysfs which tells userland when the next update is going
to happen, the userland app can then decide to use this as it's next poll time
or not.
The driver may not know when the next update will happen, e.g., if its
data source is event-based. Moreover, the driver may be able to
*decide* when the next update will happen (i.e., when to query
poll-based hardware) if the clients say what they need in advance. The
"lazy poll" / "reverse select" handles both of these.
Shem
-
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
