Re: Rép. : Re: [Xenomai-core] rtdm, rtc and printk driver.

[Jeff Webb <jeff.webb@domain.hid>]

Jan Kiszka wrote:

> The long-term question for me is if there is a need to synchronise the
> system clock (or some skin timebase in future Xenomai versions) on an
> external sources like the RTC here. Not that I'm desperately locking for
> work, it's just good to know user requirements whenever further design
> decisions [on the time subsystem] have to be made.

Our real-time simulation requires synchronization with an external time base.  The external timing source generates a one-pulse-per-second signal that is connected to an interrupt line on the real-time PC.  We use this  one-PPS interrupt to calibrate the task execution period and remove any accumulated clock skew to keep our simulation outputs in-sync with other parts of the simulation.

I am satisfied with the current implementation, but thought I would respond, since you were asking for input.  I hadn't thought about the possibility of some sort of built-in timebase synchronization functionality, but I imagine this sort of thing is fairly common, and it may prove useful to include it in Xenomai.

-Jeff