Re: [PATCH] KVM: x86: handle wrap around 32-bit address space

[Sean Christopherson]

On Mon, Apr 27, 2020 at 05:28:54PM -0700, Jim Mattson wrote:
> On Mon, Apr 27, 2020 at 9:59 AM Paolo Bonzini <pbonzini@xxxxxxxxxx> wrote:
> > @@ -1568,8 +1568,17 @@ static int skip_emulated_instruction(struct kvm_vcpu *vcpu)
> >          */
> >         if (!static_cpu_has(X86_FEATURE_HYPERVISOR) ||
> >             to_vmx(vcpu)->exit_reason != EXIT_REASON_EPT_MISCONFIG) {
> > -               rip = kvm_rip_read(vcpu);
> > -               rip += vmcs_read32(VM_EXIT_INSTRUCTION_LEN);
> > +               orig_rip = kvm_rip_read(vcpu);
> > +               rip = orig_rip + vmcs_read32(VM_EXIT_INSTRUCTION_LEN);
> > +#ifdef CONFIG_X86_64
> > +               /*
> > +                * We need to mask out the high 32 bits of RIP if not in 64-bit
> > +                * mode, but just finding out that we are in 64-bit mode is
> > +                * quite expensive.  Only do it if there was a carry.
> > +                */
> > +               if (unlikely(((rip ^ orig_rip) >> 31) == 3) && !is_64_bit_mode(vcpu))
> 
> Is it actually possible to wrap around 0 without getting a segment
> limit violation, or is it only possible to wrap *to* 0 (i.e. rip==1ull
> << 32)?

Arbitrary wrap is possible.  Limit checks are disabled for flat segs, it's
a legacy bug^W feature.