2014-03-23 09:44+0100, Paolo Bonzini: > Il 21/03/2014 19:58, Radim Krčmář ha scritto: > >>> + /* > >>> + * Return 0 for coalesced interrupts; for edge-triggered interrupts, > >>> + * this only happens if a previous edge has not been delivered due > >>> + * do masking. For level interrupts, the remote_irr field tells > > (^ to) > >>> + * us if the interrupt is waiting for an EOI. > > > >How do we know we haven't delivered an edge-triggered interrupt when > >ioapic_service() always clears this IRR bit? > > We know we have at least delivered it to the local APIC(s). If it > is masked in the ioredirtbl, ioapic_service doesn't clear the IRR. I see, LAPIC can't "refuse" an interrupt, so kvm_irq_delivery_to_apic() won't (mustn't) fail. (The code looks fragile if it returned -1.) Thanks. -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
