Ticker interrupt priority

[Martin Galvan]

On Mon, Dec 28, 2015 at 3:26 PM, Joel Sherrill <joel at rtems.org> wrote:
> A couple of odd guesses. If there are non-RTEMS interrupts, they must
> be the highest priority.

Precisely, I'd like to know why the ticker interrupt must always have
a lower priority.

> My other guess would be that the interrupt vectoring code isn't handling
> the nesting correctly. Perhaps a dispatch is missed or it doesn't know
> it is nested.

We're still looking into it, and will send a patch after we fix it.