[PATCH] KVM: only abort guest entry if timer count goes from 0->1

[Avi Kivity]

From: Marcelo Tosatti <[EMAIL PROTECTED]>

Only abort guest entry if the timer count went from 0->1, since for 1->2
or larger the bit will either be set already or a timer irq will have
been injected.

Using atomic_inc_and_test() for it also introduces an SMP barrier
to the LAPIC version (thought it was unecessary because of timer
migration, but guest can be scheduled to a different pCPU between exit
and kvm_vcpu_block(), so there is the possibility for a race).

Noticed by Avi.

Signed-off-by: Marcelo Tosatti <[EMAIL PROTECTED]>
Signed-off-by: Avi Kivity <[EMAIL PROTECTED]>

diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c
index 9e3391e..c0f7872 100644
--- a/arch/x86/kvm/i8254.c
+++ b/arch/x86/kvm/i8254.c
@@ -198,14 +198,11 @@ static int __pit_timer_fn(struct kvm_kpit_state *ps)
        struct kvm_vcpu *vcpu0 = ps->pit->kvm->vcpus[0];
        struct kvm_kpit_timer *pt = &ps->pit_timer;
 
-       atomic_inc(&pt->pending);
-       smp_mb__after_atomic_inc();
-       if (vcpu0) {
+       if (!atomic_inc_and_test(&pt->pending))
                set_bit(KVM_REQ_PENDING_TIMER, &vcpu0->requests);
-               if (waitqueue_active(&vcpu0->wq)) {
-                       vcpu0->arch.mp_state = KVM_MP_STATE_RUNNABLE;
-                       wake_up_interruptible(&vcpu0->wq);
-               }
+       if (vcpu0 && waitqueue_active(&vcpu0->wq)) {
+               vcpu0->arch.mp_state = KVM_MP_STATE_RUNNABLE;
+               wake_up_interruptible(&vcpu0->wq);
        }
 
        pt->timer.expires = ktime_add_ns(pt->timer.expires, pt->period);
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 180ba73..73f43de 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -945,8 +945,8 @@ static int __apic_timer_fn(struct kvm_lapic *apic)
        int result = 0;
        wait_queue_head_t *q = &apic->vcpu->wq;
 
-       atomic_inc(&apic->timer.pending);
-       set_bit(KVM_REQ_PENDING_TIMER, &apic->vcpu->requests);
+       if(!atomic_inc_and_test(&apic->timer.pending))
+               set_bit(KVM_REQ_PENDING_TIMER, &apic->vcpu->requests);
        if (waitqueue_active(q)) {
                apic->vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE;
                wake_up_interruptible(q);
--
To unsubscribe from this list: send the line "unsubscribe kvm-commits" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html