| Message ID | 20230921203331.3746712-5-seanjc@google.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | KVM: guest_memfd fixes | expand |
On 9/22/2023 4:33 AM, Sean Christopherson wrote: > Add an assertion that there are no in-progress MMU invalidations when a > VM is being destroyed, with the exception of the scenario where KVM > unregisters its MMU notifier between an .invalidate_range_start() call and > the corresponding .invalidate_range_end(). > > KVM can't detect unpaired calls from the mmu_notifier due to the above > exception waiver, but the assertion can detect KVM bugs, e.g. such as the > bug that *almost* escaped initial guest_memfd development. > > Link: https://lore.kernel.org/all/e397d30c-c6af-e68f-d18e-b4e3739c5389@linux.intel.com > Signed-off-by: Sean Christopherson <seanjc@google.com> > --- > virt/kvm/kvm_main.c | 9 ++++++++- > 1 file changed, 8 insertions(+), 1 deletion(-) > > diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c > index 54480655bcce..277afeedd670 100644 > --- a/virt/kvm/kvm_main.c > +++ b/virt/kvm/kvm_main.c > @@ -1381,9 +1381,16 @@ static void kvm_destroy_vm(struct kvm *kvm) > * No threads can be waiting in kvm_swap_active_memslots() as the > * last reference on KVM has been dropped, but freeing > * memslots would deadlock without this manual intervention. > + * > + * If the count isn't unbalanced, i.e. KVM did NOT unregister between Nit: Readers can get it according to the code context, but is it better to add "MMU notifier" to tell what to "unregister" to make the comment easier to understand? > + * a start() and end(), then there shouldn't be any in-progress > + * invalidations. > */ > WARN_ON(rcuwait_active(&kvm->mn_memslots_update_rcuwait)); > - kvm->mn_active_invalidate_count = 0; > + if (kvm->mn_active_invalidate_count) > + kvm->mn_active_invalidate_count = 0; > + else > + WARN_ON(kvm->mmu_invalidate_in_progress); > #else > kvm_flush_shadow_all(kvm); > #endif
On Wed, Sep 27, 2023, Binbin Wu wrote: > > > On 9/22/2023 4:33 AM, Sean Christopherson wrote: > > Add an assertion that there are no in-progress MMU invalidations when a > > VM is being destroyed, with the exception of the scenario where KVM > > unregisters its MMU notifier between an .invalidate_range_start() call and > > the corresponding .invalidate_range_end(). > > > > KVM can't detect unpaired calls from the mmu_notifier due to the above > > exception waiver, but the assertion can detect KVM bugs, e.g. such as the > > bug that *almost* escaped initial guest_memfd development. > > > > Link: https://lore.kernel.org/all/e397d30c-c6af-e68f-d18e-b4e3739c5389@linux.intel.com > > Signed-off-by: Sean Christopherson <seanjc@google.com> > > --- > > virt/kvm/kvm_main.c | 9 ++++++++- > > 1 file changed, 8 insertions(+), 1 deletion(-) > > > > diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c > > index 54480655bcce..277afeedd670 100644 > > --- a/virt/kvm/kvm_main.c > > +++ b/virt/kvm/kvm_main.c > > @@ -1381,9 +1381,16 @@ static void kvm_destroy_vm(struct kvm *kvm) > > * No threads can be waiting in kvm_swap_active_memslots() as the > > * last reference on KVM has been dropped, but freeing > > * memslots would deadlock without this manual intervention. > > + * > > + * If the count isn't unbalanced, i.e. KVM did NOT unregister between > Nit: Readers can get it according to the code context, but is it better to > add "MMU notifier" to tell what to "unregister" to make the comment easier > to understand? Agreed, I'll add that when applying.
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 54480655bcce..277afeedd670 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1381,9 +1381,16 @@ static void kvm_destroy_vm(struct kvm *kvm) * No threads can be waiting in kvm_swap_active_memslots() as the * last reference on KVM has been dropped, but freeing * memslots would deadlock without this manual intervention. + * + * If the count isn't unbalanced, i.e. KVM did NOT unregister between + * a start() and end(), then there shouldn't be any in-progress + * invalidations. */ WARN_ON(rcuwait_active(&kvm->mn_memslots_update_rcuwait)); - kvm->mn_active_invalidate_count = 0; + if (kvm->mn_active_invalidate_count) + kvm->mn_active_invalidate_count = 0; + else + WARN_ON(kvm->mmu_invalidate_in_progress); #else kvm_flush_shadow_all(kvm); #endif
Add an assertion that there are no in-progress MMU invalidations when a VM is being destroyed, with the exception of the scenario where KVM unregisters its MMU notifier between an .invalidate_range_start() call and the corresponding .invalidate_range_end(). KVM can't detect unpaired calls from the mmu_notifier due to the above exception waiver, but the assertion can detect KVM bugs, e.g. such as the bug that *almost* escaped initial guest_memfd development. Link: https://lore.kernel.org/all/e397d30c-c6af-e68f-d18e-b4e3739c5389@linux.intel.com Signed-off-by: Sean Christopherson <seanjc@google.com> --- virt/kvm/kvm_main.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)