Hi, This series aims to fix various issues in nested VPID emulation. The 1st patch fixes a bug that INVEPT was executed on L1<->L2 transitions to flush TLB but with the wrong EPTP because it hasn't been loaded yet with the destination EPTP. The patch handles this issue easily by just delaying vmx_flush_tlb() to be called after EPTP has been set to destination by using KVM_REQ_TLB_FLUSH. The 2nd patch fixes a bug in L0 emulation of INVVPID which can use a wrong VPID02 in case KVM failed to allocate vmx->nested.vpid02. The 3rd patch fixes a critical issue of invalidating wrong mappings from TLB when performing VPID02 related invalidations. The issue is that these invalidations mistakenly invalidates combined and guest-physical mappings tagged by EPTP01 instead of invalidating linear and combined mappings tagged by VPID02. The 4rd patch optimizes L1<->L2 transitions by avoiding unnecessary TLB flush in case L1 use both VPID and EPT and thus L0 can separate TLB entries of L1 and L2 by EPTP tag. Regards, -Liran
