On 31/01/19 09:52, Luwei Kang wrote: > Some Posted-Interrupts from passthrough devices may be lost or > overwritten when the vCPU is in runnable state. > > The SN (Suppress Notification) of PID (Posted Interrupt Descriptor) will > be set when the vCPU is preempted (vCPU in KVM_MP_STATE_RUNNABLE state > but not running on physical CPU). If a posted interrupt coming at this > time, the irq remmaping facility will set the bit of PIR (Posted > Interrupt Requests) without ON (Outstanding Notification). > So this interrupt can't be sync to APIC virtualization register and > will not be handled by Guest because ON is zero. > > Signed-off-by: Luwei Kang <luwei.kang@xxxxxxxxx> > --- > arch/x86/kvm/vmx/vmx.c | 5 +++++ > arch/x86/kvm/x86.c | 2 +- > 2 files changed, 6 insertions(+), 1 deletion(-) > > diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c > index 4341175..8ed9634 100644 > --- a/arch/x86/kvm/vmx/vmx.c > +++ b/arch/x86/kvm/vmx/vmx.c > @@ -1221,6 +1221,11 @@ static void vmx_vcpu_pi_load(struct kvm_vcpu *vcpu, int cpu) > new.sn = 0; > } while (cmpxchg64(&pi_desc->control, old.control, > new.control) != old.control); > + /* * Clear SN before reading the bitmap. The VT-d firmware * writes the bitmap and reads SN atomically (5.2.3 in the * spec), so it doesn't really have a memory barrier that * pairs with this, but we cannot do that and we need one. */ > + smp_mb__after_atomic(); > + > + if (!bitmap_empty((unsigned long *)pi_desc->pir, NR_VECTORS)) > + pi_test_and_set_on(pi_desc); You can add pi_set_on for use here. The fast path with pi_clear_sn should also be removed. > } > /* > diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c > index 3d27206..5bcf2c4 100644 > --- a/arch/x86/kvm/x86.c > +++ b/arch/x86/kvm/x86.c > @@ -7794,7 +7794,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) > * 1) We should set ->mode before checking ->requests. Please see > * the comment in kvm_vcpu_exiting_guest_mode(). > * > - * 2) For APICv, we should set ->mode before checking PIR.ON. This > + * 2) For APICv, we should set ->mode before checking PID.PIR. This This should be PID.ON. Paolo > * pairs with the memory barrier implicit in pi_test_and_set_on > * (see vmx_deliver_posted_interrupt). > * >
