Re: [PATCH 4/5] KVM: Kick resamplefd for split kernel irqchip

[Peter Xu <peterx@redhat.com>]

On Fri, Feb 28, 2020 at 11:34:46AM +0100, Paolo Bonzini wrote:
> On 26/02/20 23:55, Peter Xu wrote:
> > +
> > +            /* Kick resamplefd if KVM is bypassed */
> > +            kvm_resample_fd_notify(n);
> 
> This is only needed for level-triggered interrupts, so it can be placed
> below the test.

Yes I was thinking it the same way, however...  I noticed that VFIO
will even mask edge INTx after getting IRQ (vfio_intx_handler).  And
if look into current KVM kernel implementation, it has done the same
logic to ack edge triggerred irqs (kvm_ioapic_update_eoi_one), we
should logically ack it here too even for edge-triggered?

Thanks,

> 
> Paolo
> 
> > +            if (((entry >> IOAPIC_LVT_TRIGGER_MODE_SHIFT) & 1) !=
> > +                IOAPIC_TRIGGER_LEVEL) {
> >                  continue;
> 

-- 
Peter Xu