[coyotos-dev] Sleep, wakeup, and persistence

[Valerio Bellizzomi]

On 14/09/2007, at 23.56, Sam Mason wrote:

>On Fri, Sep 14, 2007 at 10:53:15PM +0200, Valerio Bellizzomi wrote:
>> >That way, our embedded system can have the time punched into its
>> >panel/ask some network server and eventually get an answer.  In the
>> >meantime, it can be relying on the kernel for short timeouts.  A PC
>> >would be able to get a good estimate of the real time from its clock
>> >quickly so will be able to start servicing "real time" alarms (for
lack
>> >of a better word) quickly.
>> 
>> There is a problem with that, I have a P4 mainboard that gains seconds
>> faster than NTP clock.
>> The mainboard clock counts faster than real time, so when NTP time is
>> updated, the real time actually goes backwards.
>
>That's kind of why I was was saying leave it to user land.  You could
>tell your system to bootstrap the clock with the PC's realtime clock
>(does it even want to do that?) and only to start answering alarms (this
>still seems like a bad term) when it's corrected the local clock.

Yes. The application needs to be warned when the clock has been corrected,
this can be implemented in user-space.

>
>Still got no idea how this sort of design would be implemented though.

Charlie's idea is interesting:

"One option is to implement these forms of sleep outside the kernel, 
in a server that is aware of restarts, but does not expose that 
awareness to its clients."


>
>
>  Sam