Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Content-Type: text/plain;
        charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 11.1 \(3445.4.7\))
Subject: Re: [PATCH RFC] KVM: x86: tell guests if the exposed SMT topology is
 trustworthy
From:   Liran Alon <liran.alon@oracle.com>
In-Reply-To: <E392C60C-A596-49DD-B604-8B3C473ACAA2@dinechin.org>
Date:   Fri, 8 Nov 2019 17:52:19 +0200
Cc:     Sean Christopherson <sean.j.christopherson@intel.com>,
        Vitaly Kuznetsov <vkuznets@redhat.com>,
        KVM list <kvm@vger.kernel.org>, x86@kernel.org,
        Paolo Bonzini <pbonzini@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        Jim Mattson <jmattson@google.com>,
        open list <linux-kernel@vger.kernel.org>,
        "H. Peter Anvin" <hpa@zytor.com>,
        "Peter Zijlstra (Intel)" <peterz@infradead.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <5B82FF2C-0309-4D67-85E2-646AFB77B2FD@oracle.com>
References: <20191105161737.21395-1-vkuznets@redhat.com>
 <20191105193749.GA20225@linux.intel.com>
 <20191105232500.GA25887@linux.intel.com>
 <943488A8-2DD7-4471-B3C7-9F21A0B0BCF9@dinechin.org>
 <713ECF67-6A6C-4956-8AC6-7F4C05961328@oracle.com>
 <E392C60C-A596-49DD-B604-8B3C473ACAA2@dinechin.org>
To:     Christophe de Dinechin <christophe.de.dinechin@gmail.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On 8 Nov 2019, at 17:35, Christophe de Dinechin =
<christophe.de.dinechin@gmail.com> wrote:
>=20
>=20
>=20
>> On 7 Nov 2019, at 16:02, Liran Alon <liran.alon@oracle.com> wrote:
>>=20
>>=20
>>=20
>>> On 7 Nov 2019, at 16:00, Christophe de Dinechin =
<christophe.de.dinechin@gmail.com> wrote:
>>>=20
>>=20
>>>=20
>>> I share that concern about the naming, although I do see some
>>> value in exposing the cpu_smt_possible() result. I think it=E2=80=99s =
easier
>>> to state that something does not work than to state something does
>>> work.
>>>=20
>>> Also, with respect to mitigation, we may want to split the two cases
>>> that Paolo outlined, i.e. have KVM_HINTS_REALTIME,
>>> KVM_HINTS_CORES_CROSSTALK and
>>> KVM_HINTS_CORES_LEAKING,
>>> where CORES_CROSSTALKS indicates there may be some
>>> cross-talk between what the guest thinks are isolated cores,
>>> and CORES_LEAKING indicates that cores may leak data
>>> to some other guest.
>>>=20
>>> The problem with my approach is that it is shouting =E2=80=9Cdon=E2=80=
=99t trust me=E2=80=9D
>>> a bit too loudly.
>>=20
>> I don=E2=80=99t see a value in exposing CORES_LEAKING to guest. As =
guest have nothing to do with it.
>=20
> The guest could display / expose the information to guest sysadmins
> and admin tools (e.g. through /proc).
>=20
> While the kernel cannot mitigate, a higher-level product could for =
example
> have a policy about which workloads can be deployed on a system which
> may leak data to other VMs.
>=20
> Christophe

Honestly, I don=E2=80=99t think any sane cloud provider will schedule =
vCPUs of different guests on same physical CPU core and report this to =
guest.
Therefore, I think this is only relevant for use-cases where the guest =
owner is also the host/hypervisor owner. And therefore, doesn=E2=80=99t =
need this
information exposed in a CPUID bit.
I see your point regarding how in theory it could be used, but I think =
we should wait and see if such use-case exists before defining this =
interface.

-Liran