On 10.08.2016 13:30, Dmitry Osipenko wrote: [snip]
> I suggest to provide "reset" function, otherwise it's likely that you would > get > unexpected result or crash on QEMU reset. This also applies to the "interrupt > controller" patch. > Correction: I didn't notice that "interrupt controller" state hasn't any resettable variables, so it doesn't need reset. -- Dmitry