On Fri, Jan 31, 2014 at 06:09:20PM +0000, Peter Maydell wrote:
> On 28 January 2014 20:32, Christoffer Dall <christoffer.d...@linaro.org>
> wrote:
[...]
>
> This looks broadly right; a couple of comments below.
>
[...]
> > +
> > +static void gic_set_irq_generic(GICState *s, int irq, int level,
> > + int cm, int target)
> > +{
> > + if (level) {
> > + GIC_SET_LEVEL(irq, cm);
> > + DPRINTF("Set %d pending mask %x\n", irq, target);
> > + if (GIC_TEST_EDGE_TRIGGER(irq)) {
> > + GIC_SET_PENDING(irq, target);
> > + }
> > + } else {
> > + if (GIC_TEST_EDGE_TRIGGER(irq)) {
> > + GIC_CLEAR_PENDING(irq, target);
> > + }
>
> This doesn't look right. A falling edge for an edge triggered
> interrupt should just CLEAR_LEVEL and leave the state of the
> pending latch alone.
>
duh, yeah, thanks for spotting.
[...]
Thanks,
-Christoffer
