Received-SPF: pass (google.com: domain of linux-kernel+bounces-85426-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Precedence: bulk
MIME-Version: 1.0
References: <20240227232100.478238-1-pbonzini@redhat.com> <Zd6LK7RpZZ8t-5CY@google.com>
 <CABgObfYpRJnDdQrxp=OgjhbT9A+LHK36MHjMvzcQJsHAmfX++w@mail.gmail.com> <Zd9hrfJ5xRI6HeZp@google.com>
In-Reply-To: <Zd9hrfJ5xRI6HeZp@google.com>
From: Paolo Bonzini <pbonzini@redhat.com>
Date: Wed, 28 Feb 2024 18:20:27 +0100
Message-ID: <CABgObfaPRpEnndHwkZpOZ=JOZjJPyh2KXYLh5ZGMFMSboZqj9w@mail.gmail.com>
Subject: Re: [PATCH 00/21] TDX/SNP part 1 of n, for 6.9
To: Sean Christopherson <seanjc@google.com>
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org, michael.roth@amd.com, 
	isaku.yamahata@intel.com, thomas.lendacky@amd.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Wed, Feb 28, 2024 at 5:39=E2=80=AFPM Sean Christopherson <seanjc@google.=
com> wrote:
> > > This doesn't work.  The ENC flag gets set on any SNP *capable* CPU, w=
hich results
> > > in false positives for SEV and SEV-ES guests[*].
> >
> > You didn't look at the patch did you? :)
>
> Guilty, sort of.  I looked (and tested) the patch from the TDX series, bu=
t I didn't
> look at what you postd.  But it's a moot point, because now I did look at=
 what you
> posted, and it's still broken :-)
>
> > It does check for has_private_mem (alternatively I could have dropped t=
he bit
> > in SVM code for SEV and SEV-ES guests).
>
> The problem isn't with *KVM* setting the bit, it's with *hardware* settin=
g the
> bit for SEV and SEV-ES guests.  That results in this:
>
>   .is_private =3D vcpu->kvm->arch.has_private_mem && (err & PFERR_GUEST_E=
NC_MASK),
>
> marking the fault as private.  Which, in a vacuum, isn't technically wron=
g, since
> from hardware's perspective the vCPU access was "private".  But from KVM'=
s
> perspective, SEV and SEV-ES guests don't have private memory

vcpu->kvm->arch.has_private_mem is the flag from the SEV VM types
series. It's false on SEV and SEV-ES VMs, therefore fault->is_private
is going to be false as well. Is it ENOCOFFEE for you or ENODINNER for
me? :)

Paolo

> And because the flag only gets set on SNP capable hardware (in my limited=
 testing
> of a whole two systems), running the same VM on different hardware would =
result
> in faults being marked private on one system, but not the other.  Which m=
eans that
> KVM can't rely on the flag being set for SEV or SEV-ES guests, i.e. we ca=
n't
> retroactively enforce anything (not to mention that that might break exis=
ting VMs).
>