On 23/05/2018 16:44, Sean Christopherson wrote: > Unloading the MMU on nested entry/exit doesn't seem deliberate, > e.g. why bother with VPID handling in prepare_vmcs02() if KVM > intends to unconditionally flush? I think figuring out how to > avoid unloading the MMU in those cases will resolve the issue > of the TLB being flushed on every switch between L1 and L2, > though I get the feeling that that will mean doing a holistic > analysis of the (nested) MMU handling. My plan there was just to add a third kvm_mmu struct, which (as a start) requires getting rid of all references to vcpu->arch.mmu in arch/x86/kvm/. Paolo
