[meta-xilinx] Interrupt forwarding in AMP configuration

[Eric Wong]

On Fri, Dec 23, 2016 at 10:18 AM, Eric Wong <ewong3 at gmail.com> wrote:
> But my interrupt is not active low, it is rising-edge.  The signal
> just idles high and I want the interrupt to trigger on the rising
> trailing edge of a negative pulse in the signal.
>
> Supporting only active-high and rising-edge interrupts does sort of
> imply the signal should idle low.  Though it doesn't strictly rule out
> a signal idling high and interrupting on the rising trailing edge of a
> negative pulse, but something apparently can't handle the interrupt
> signal idling high even if it's a rising edge trigger.

The other confounding thing is that this kind of interrupt handling
(idle high, trigger on rising trailing edge) works without Linux/CPU0
in the picture.  If I run only the CPU1 firmware via JTAG, it works.