On Mon, 2024-06-24 at 16:29 +0800, Yan Zhao wrote: > > @@ -1057,7 +1057,7 @@ void kvm_tdp_mmu_zap_all(struct kvm *kvm) > > * KVM_RUN is unreachable, i.e. no vCPUs will ever service the > > request. > > */ > > lockdep_assert_held_write(&kvm->mmu_lock); > > - for_each_tdp_mmu_root_yield_safe(kvm, root) > > + __for_each_tdp_mmu_root_yield_safe(kvm, root, -1, KVM_DIRECT_ROOTS) > nit: update the comment of kvm_tdp_mmu_zap_all() Yea, good idea. It's definitely needs some explanation, considering the function is called "zap_all". A bit unfortunate actually. > and explain why it's > KVM_DIRECT_ROOTS, not KVM_DIRECT_ROOTS | KVM_INVALID_ROOTS. Explain why not to zap invalid mirrored roots?
