Re: [PATCH 3/5] mmu: don't set the present bit unconditionally

[Xiao Guangrong]

On 06/29/2016 04:18 PM, Paolo Bonzini wrote:
> On 29/06/2016 05:17, Xiao Guangrong wrote:
> > > +++ b/arch/x86/kvm/mmu.c
> > > @@ -2516,13 +2516,17 @@ static int set_spte(struct kvm_vcpu *vcpu, u64
> > > *sptep,
> > >                gfn_t gfn, kvm_pfn_t pfn, bool speculative,
> > >                bool can_unsync, bool host_writable)
> > >    {
> > > -    u64 spte;
> > > +    u64 spte = 0;
> > >        int ret = 0;
> > > +    struct kvm_mmu *context = &vcpu->arch.mmu;
> > > +    bool execonly = !(context->guest_rsvd_check.bad_mt_xwr &
> > > +              (1ull << VMX_EPT_EXECUTABLE_MASK));
> >
> > Could we introduce a new field, say execonly, to "struct kvm_mmu"?
> > That would make the code be more clearer.
>
> Given how execonly is used, let's add shadow_present_mask instead.

Yup, it is better.