Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <YQuZdVuaGG/Cr62y@kroah.com> <YQuaJ78y8j1UmBoz@kroah.com>
 <fdf8b6b6-58c3-8392-2fc6-1908a314e991@linux.intel.com> <YQwlHrJBw79xhTSI@kroah.com>
 <21db8884-5aa1-3971-79ef-f173a0a95bef@linux.intel.com> <YQwpa+LAYt7YZ5dh@kroah.com>
 <1e0967ee-c41e-fd5d-f553-e4d7ab88838c@linux.intel.com> <CAPcyv4h26QmPhJKjtUB-VPc4TP95rBAer78OzkgXncg77tFWVw@mail.gmail.com>
 <9b2956f5-3acf-e798-ff0f-002d2d5254db@linux.intel.com> <CAPcyv4ju0kzUuamiuE=xC_DoEArk1KxD+n+-TUw0LLTWoSj2VA@mail.gmail.com>
 <YQzF/utgrJfbZuHh@kroah.com>
In-Reply-To: <YQzF/utgrJfbZuHh@kroah.com>
From:   Dan Williams <dan.j.williams@intel.com>
Date:   Fri, 6 Aug 2021 07:36:25 -0700
Message-ID: <CAPcyv4jXoj9_xJ9p3rfoHgZc6xsjnvbeqHwxw8Y2h+kx4-qs+Q@mail.gmail.com>
Subject: Re: [PATCH v1] driver: base: Add driver filter support
To:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc:     Andi Kleen <ak@linux.intel.com>,
        Kuppuswamy Sathyanarayanan 
        <sathyanarayanan.kuppuswamy@linux.intel.com>,
        "Rafael J . Wysocki" <rafael@kernel.org>,
        Jonathan Corbet <corbet@lwn.net>,
        Kuppuswamy Sathyanarayanan <knsathya@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Linux Doc Mailing List <linux-doc@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, Aug 5, 2021 at 10:17 PM Greg Kroah-Hartman
<gregkh@linuxfoundation.org> wrote:
>
> On Thu, Aug 05, 2021 at 06:00:25PM -0700, Dan Williams wrote:
> > That said, per-device authorization is a little bit different than
> > per-driver trust. Driver trust is easy to reason about for a built-in
> > policy, while per-device authorization is easy for userspace to reason
> > about for "why is this device not talking to its driver?".
>
> See my other email about how the "per driver" trust is the wrong model,
> you need to stick to "per device" trust.  Especially given that you are
> giving control of your kernel drivers over to third parties, you already
> trust them to do the right thing.

Andi, if the number of TDX devices is small could they grow an SPDM
over virtio channel? Then you can measure trust from the VM to the VMM
to the attestation server.