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;
MIME-Version: 1.0
References: <CAHk-=whcA0kAWGqet9oJdyttjhP7J4PcU2Ly=n_Qyo6rMJNxRg@mail.gmail.com>
 <CAKOZuevv_QgvzmusA+yey76vo4hn9bEVwud2RumuOr0K4++15A@mail.gmail.com>
 <CAHk-=wj0o8BiUrYiM5ZcTS6JxBh1M60O36Ms8-EqLQ139=1L4g@mail.gmail.com>
 <20190330171215.3yrfxwodstmgzmxy@brauner.io> <CAHk-=whF6gnRVbJaArZna4e=tejfrzmNQtRbWjnuSSKpBn+jQg@mail.gmail.com>
 <132107F4-F56B-4D6E-9E00-A6F7C092E6BD@amacapital.net> <CAHk-=wiaLtH41Mc5qAjOeCWavPqV0DhS581KYa0QBt8uraTK1Q@mail.gmail.com>
 <20190331211041.vht7dnqg4e4bilr2@brauner.io> <CAHk-=wi3AE1-iRQ_7LOeSMNAMrGxRdC=gTjD30duVw4XRchcNQ@mail.gmail.com>
 <18C7FCB9-2CBA-4237-94BB-9C4395A2106B@amacapital.net> <20190401120450.e4k2m434qyqj4yrn@brauner.io>
In-Reply-To: <20190401120450.e4k2m434qyqj4yrn@brauner.io>
From:   Jann Horn <jannh@google.com>
Date:   Mon, 1 Apr 2019 15:43:13 +0200
Message-ID: <CAG48ez0KeQU32HV8OGY4-LNPF0Q+PtaH7H47GmKAd1Hhho=jXw@mail.gmail.com>
Subject: Re: [PATCH v2 0/5] pid: add pidfd_open()
To:     Christian Brauner <christian@brauner.io>,
        Al Viro <viro@zeniv.linux.org.uk>,
        Andy Lutomirski <luto@amacapital.net>
Cc:     Linus Torvalds <torvalds@linux-foundation.org>,
        Daniel Colascione <dancol@google.com>,
        Andrew Lutomirski <luto@kernel.org>,
        David Howells <dhowells@redhat.com>,
        "Serge E. Hallyn" <serge@hallyn.com>,
        Linux API <linux-api@vger.kernel.org>,
        Linux List Kernel Mailing <linux-kernel@vger.kernel.org>,
        Arnd Bergmann <arnd@arndb.de>,
        "Eric W. Biederman" <ebiederm@xmission.com>,
        Konstantin Khlebnikov <khlebnikov@yandex-team.ru>,
        Kees Cook <keescook@chromium.org>,
        Alexey Dobriyan <adobriyan@gmail.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Michael Kerrisk-manpages <mtk.manpages@gmail.com>,
        Jonathan Kowalski <bl0pbl33p@gmail.com>,
        "Dmitry V. Levin" <ldv@altlinux.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Oleg Nesterov <oleg@redhat.com>,
        Nagarathnam Muthusamy <nagarathnam.muthusamy@oracle.com>,
        Aleksa Sarai <cyphar@cyphar.com>,
        Joel Fernandes <joel@joelfernandes.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Apr 1, 2019 at 2:04 PM Christian Brauner <christian@brauner.io> wro=
te:
> On Sun, Mar 31, 2019 at 08:13:38PM -0600, Andy Lutomirski wrote:
> > > On Mar 31, 2019, at 3:17 PM, Linus Torvalds <torvalds@linux-foundatio=
n.org> wrote:
> > >> On Sun, Mar 31, 2019 at 2:10 PM Christian Brauner <christian@brauner=
.io> wrote:
> > >>
> > >> I don't think that we want or can make them equivalent since that wo=
uld
> > >> mean we depend on procfs.
> > >
> > > Sure we can.
> > >
> > > If /proc is enabled, then you always do that dance YOU ALREADY WROTE
> > > THE CODE FOR to do the stupid ioctl.
> > >
> > > And if /procfs isn't enabled, then you don't do that.
> > >
> > > Ta-daa. Done. No stupid ioctl, and now /proc and pidfd_open() return
> > > the same damn thing.
> > >
> > > And guess what? If /proc isn't enabled, then obviously pidfd_open()
> > > gives you the /proc-less thing, but at least there is no crazy "two
> > > different file descriptors for the same thing" situation, because the=
n
> > > the /proc one doesn't exist.
> > >
> >
> > I wish we could do this, and, in a clean design, it would be a no-brain=
er.  But /proc has too much baggage.  Just to mention two such things, ther=
e=E2=80=99s =E2=80=9Cnet=E2=80=9D and =E2=80=9C../sys=E2=80=9D.  This crud =
is why we have all kinds of crazy rules that prevent programs in sandboxes =
from making a new mounts and mounting /proc in it.  If we make it possible =
to clone a new process and this access /proc without having /proc mounted, =
we=E2=80=99ll open up a big can of worms.
> >
> > Maybe we could have a sanitized view of /proc and make a pidfd be a dir=
ectory fd pointing at that.
>
> We can also just create something like an internal bind-mount without a
> parent, i.e. similar to
>
> open_tree(<internal-procfs-mount>, "<pid>", OPEN_TREE_CLONE);
>
> on a clone(CLONE_PIDFD);
>
> that would block any openat(fd, "..");

Or we add a check to follow_dotdot()/follow_dotdot_rcu() that throws
an error if nd->path.mnt->mnt_flags has some new flag for "no dotdot
traversal on this mountpoint", and then set that on the internal procfs
mount... if Al Viro doesn't think that that's too hideous.