On Sun, Jan 12, 2025 at 08:51:25PM -0800, Nicolin Chen wrote: > We will still have the fault queue: > if (error is handled by PRI) > report via fault queue; // need response This is an important point, we can't generate a response for the PRI unless the fault queue is used, so even though the vEVENTQ could carry the PRI event, it must not be done. If there is no fault handle available the kernel must NAK the PRI event internally, never send it to the vEVENTQ. Jason