On Thu, Jan 07, 2021, Paolo Bonzini wrote: > On 07/01/21 18:00, Sean Christopherson wrote: > > Ugh, I assume this is due to one of the "premature" nested_ops->check_events() > > calls that are necessitated by the event mess? I'm guessing kvm_vcpu_running() > > is the culprit? > > > > If my assumption is correct, this bug affects nVMX as well. > > Yes, though it may be latent. For SVM it was until we started allocating > svm->nested on demand. > > > Rather than clear the request blindly on any nested VM-Exit, what > > about something like the following? > > I think your patch is overkill, KVM_REQ_GET_NESTED_STATE_PAGES is only set > from KVM_SET_NESTED_STATE so it cannot happen while the VM runs. Yeah, which is why I was hoping we could avoid clearing the request on every nested exit. > Something like this is small enough and works well. I've no argument against it working, rather that I dislike clearing the request on every exit. Except for the ->check_events() case, hitting the scenario where there's a pending request at the time of nested VM-Exit would ideally be treated as a KVM bug. On the other hand, clearing nested-specific request on nested VM-Exit is logically sound, so I guess I'm ok with the minimal patch.
