MIME-Version: 1.0
References: <43980946-7bbf-dcef-7e40-af904c456250@linux.microsoft.com>
 <Y+p1j7tYT+16MX6B@google.com> <35ff8f48-2677-78ea-b5f3-329c75ce65c9@redhat.com>
 <Y+qLe42h9ZPRINrG@google.com>
In-Reply-To: <Y+qLe42h9ZPRINrG@google.com>
From:   Paolo Bonzini <pbonzini@redhat.com>
Date:   Mon, 13 Feb 2023 20:56:08 +0100
Message-ID: <CABgObfaZQOvt6v0yGz3MR7FBU7DcrTTGmS6M8RWCX0uy6WML1Q@mail.gmail.com>
Subject: Re: "KVM: x86/mmu: Overhaul TDP MMU zapping and flushing" breaks SVM
 on Hyper-V
To:     Sean Christopherson <seanjc@google.com>
Cc:     Jeremi Piotrowski <jpiotrowski@linux.microsoft.com>,
        kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
        Tianyu Lan <ltykernel@gmail.com>,
        "Michael Kelley (LINUX)" <mikelley@microsoft.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Mon, Feb 13, 2023 at 8:12 PM Sean Christopherson <seanjc@google.com> wrote:
> > My reading of the spec[1] is that HV_X64_NESTED_ENLIGHTENED_TLB will cause
> > svm_flush_tlb_current to behave (in Intel parlance) as an INVVPID rather
> > than an INVEPT.
>
> Oh!  Good catch!  Yeah, that'll be a problem.
>
> > So svm_flush_tlb_current has to be changed to also add a
> > call to HvCallFlushGuestPhysicalAddressSpace.  I'm not sure if that's a good
> > idea though.
>
> That's not strictly necessary, e.g. flushes from kvm_invalidate_pcid() and
> kvm_post_set_cr4() don't need to effect a full flush.  I believe the virtual
> address flush is also sufficient for avic_activate_vmcb().  Nested (from KVM's
> perspective, i.e. running L3) can just be mutually exclusive with
> HV_X64_NESTED_ENLIGHTENED_TLB.
>
> That just leaves kvm_mmu_new_pgd()'s force_flush_and_sync_on_reuse and the
> aforementioned kvm_mmu_load().
>
> That said, the above cases where a virtual address flush is sufficient are
> rare operations when using NPT, so adding a new KVM_REQ_TLB_FLUSH_ROOT or
> whatever probably isn't worth doing.
>
> > First, that's a TLB shootdown rather than just a local thing;
> > flush_tlb_current is supposed to be relatively cheap, and there would be a
> > lot of them because of the unconditional calls to
> > nested_svm_transition_tlb_flush on vmentry/vmexit.
>
> This isn't a nested scenario for KVM though.

Yes, but svm_flush_tlb_current() *is* also used in nested scenarios so
it's like you said below---you would have to disable enlightened TLB
when EFER.SVME=1 or something like that.

> > Depending on the performance results of adding the hypercall to
> > svm_flush_tlb_current, the fix could indeed be to just disable usage of
> > HV_X64_NESTED_ENLIGHTENED_TLB.
>
> Minus making nested SVM (L3) mutually exclusive, I believe this will do the trick:
>
> +       /* blah blah blah */
> +       hv_flush_tlb_current(vcpu);
> +

Yes, it's either this or disabling the feature.

Paolo