Apparently the development of update_affinity() overlapped with the
promotion of irq_lock to be _irqsave, so the patch didn't convert this
lock over. This will make lockdep complain.
Fix this by disabling IRQs around the lock.
Cc: stable@vger.kernel.org
Fixes: 08c9fd042117 ("KVM: arm/arm64: vITS: Add a helper to update the affinity of an LPI")
Reported-by: Jan Glauber <jan.glauber@caviumnetworks.com>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 static int update_affinity(struct vgic_irq *irq, struct kvm_vcpu *vcpu)
 {
        int ret = 0;
+       unsigned long flags;
 
-       spin_lock(&irq->irq_lock);
+       spin_lock_irqsave(&irq->irq_lock, flags);
        irq->target_vcpu = vcpu;
-       spin_unlock(&irq->irq_lock);
+       spin_unlock_irqrestore(&irq->irq_lock, flags);
 
        if (irq->hw) {
                struct its_vlpi_map map;