> On 16 Jul 2019, at 22:41, Sean Christopherson <sean.j.christopherson@xxxxxxxxx> wrote: > > On Tue, Jul 16, 2019 at 10:34:08PM +0300, Liran Alon wrote: >> If we really want to be pedantic, we can parse guest page-tables to see if PTE >> have U/S bit set to 1. What do you think? > > Performance aside, walking the guest page tables would fall apart if a > different vCPU modified the guest's page tables. True. :) So let’s just stick with checking only CR4.SMAP=1 then. -Liran
