Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Tue, 2 Aug 2022 21:23:08 -0700
From:   Eric Biggers <ebiggers@kernel.org>
To:     Daniil Lunev <dlunev@chromium.org>
Cc:     Zdenek Kabelac <zdenek.kabelac@gmail.com>,
        Brian Geffon <bgeffon@google.com>,
        Mike Snitzer <snitzer@kernel.org>,
        linux-kernel@vger.kernel.org, dm-devel@redhat.com,
        Mikulas Patocka <mpatocka@redhat.com>,
        Alasdair Kergon <agk@redhat.com>
Subject: Re: [dm-devel] [PATCH 1/1] dm: add message command to disallow
 device open
Message-ID: <Yun4LH+StcuBXRtO@sol.localdomain>
References: <20220704000225.345536-1-dlunev@chromium.org>
 <20220704100221.1.I15b3f7a84ba5a97fde9276648e391b54957103ff@changeid>
 <YtB45Lte5UhlEE6y@redhat.com>
 <CAONX=-dEG121RQ6L-4fPMXrLXb3JeYNVNiPzHXNaRLbwRzb3bw@mail.gmail.com>
 <alpine.LRH.2.02.2207150528170.5197@file01.intranet.prod.int.rdu2.redhat.com>
 <cca5b463-a860-de8d-b7e4-a8d30aef2ff2@gmail.com>
 <CAONX=-fJHgfGkwR5A1MT+8FHckueehOsUS_LyHkjrgp4Y+vOgw@mail.gmail.com>
 <CAONX=-ft=ewFDui4jmd2fvcNr2EJc90=ZNOueDdp6HaPZmvObQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAONX=-ft=ewFDui4jmd2fvcNr2EJc90=ZNOueDdp6HaPZmvObQ@mail.gmail.com>
Precedence: bulk

On Wed, Aug 03, 2022 at 02:12:26PM +1000, Daniil Lunev wrote:
> Hello all
> To signal boost here. What can we do to advance the discussion on this
> topic? Can we move forward with the approach or are there any
> alternative suggestions how the desired behaviour can be achieved?
> Thanks,
> --Daniil
> 
> On Tue, Jul 19, 2022 at 9:42 AM Daniil Lunev <dlunev@chromium.org> wrote:
> >
> > We understand that if someone acquires root it is a game over. The intent of
> > this mechanism is to reduce the attack surface. The exposure might be a
> > certain system daemon that is exploited into accessing a wrong node in
> > the filesystem. And exposing modifiable system memory is a pathway for
> > further escalation and leaks of secrets. This is a defense in depth mechanism,
> > that is intended to make attackers' lives harder even if they find an
> > exploitable
> > vulnerability.
> > We understand that in regular situations people may not want the behaviour,
> > that is why the mechanism is controlled via a side channel - if a message is
> > never sent - the behaviour is not altered.
> > --Daniil

This seems like an access control policy, which the Linux kernel already has a
lot of mechanisms for.  Chrome OS already uses SELinux.  Couldn't this be solved
by giving the device node an SELinux label that no one has permission to open?

- Eric