linux-kernel - Re: [PATCH 1/2] fs/exec: Explicitly unshare fs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAG48ez3hM+-V39QpFaNfRJxVrQVBu2Dm-B-xFN2GEt9p81Vd2Q@mail.gmail.com>
Date:   Fri, 14 Oct 2022 17:35:26 +0200
From:   Jann Horn <jannh@...gle.com>
To:     Andy Lutomirski <luto@...nel.org>
Cc:     Christian Brauner <brauner@...nel.org>,
        Kees Cook <keescook@...omium.org>,
        "Eric W. Biederman" <ebiederm@...ssion.com>,
        Jorge Merlino <jorge.merlino@...onical.com>,
        Al Viro <viro@...iv.linux.org.uk>,
        Thomas Gleixner <tglx@...utronix.de>,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org,
        linux-fsdevel@...r.kernel.org,
        John Johansen <john.johansen@...onical.com>,
        Paul Moore <paul@...l-moore.com>,
        James Morris <jmorris@...ei.org>,
        "Serge E. Hallyn" <serge@...lyn.com>,
        Stephen Smalley <stephen.smalley.work@...il.com>,
        Eric Paris <eparis@...isplace.org>,
        Richard Haines <richard_c_haines@...nternet.com>,
        Casey Schaufler <casey@...aufler-ca.com>,
        Xin Long <lucien.xin@...il.com>,
        "David S. Miller" <davem@...emloft.net>,
        Todd Kjos <tkjos@...gle.com>,
        Ondrej Mosnacek <omosnace@...hat.com>,
        Prashanth Prahlad <pprahlad@...hat.com>,
        Micah Morton <mortonm@...omium.org>,
        Fenghua Yu <fenghua.yu@...el.com>,
        Andrei Vagin <avagin@...il.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        apparmor@...ts.ubuntu.com, linux-security-module@...r.kernel.org,
        selinux@...r.kernel.org, linux-hardening@...r.kernel.org
Subject: Re: [PATCH 1/2] fs/exec: Explicitly unshare fs_struct on exec

On Fri, Oct 14, 2022 at 5:18 AM Andy Lutomirski <luto@...nel.org> wrote:
> On Thu, Oct 6, 2022, at 7:13 AM, Jann Horn wrote:
> > On Thu, Oct 6, 2022 at 11:05 AM Christian Brauner <brauner@...nel.org> wrote:
> >> On Thu, Oct 06, 2022 at 01:27:34AM -0700, Kees Cook wrote:
> >> > The check_unsafe_exec() counting of n_fs would not add up under a heavily
> >> > threaded process trying to perform a suid exec, causing the suid portion
> >> > to fail. This counting error appears to be unneeded, but to catch any
> >> > possible conditions, explicitly unshare fs_struct on exec, if it ends up
> >>
> >> Isn't this a potential uapi break? Afaict, before this change a call to
> >> clone{3}(CLONE_FS) followed by an exec in the child would have the
> >> parent and child share fs information. So if the child e.g., changes the
> >> working directory post exec it would also affect the parent. But after
> >> this change here this would no longer be true. So a child changing a
> >> workding directoro would not affect the parent anymore. IOW, an exec is
> >> accompanied by an unshare(CLONE_FS). Might still be worth trying ofc but
> >> it seems like a non-trivial uapi change but there might be few users
> >> that do clone{3}(CLONE_FS) followed by an exec.
> >
> > I believe the following code in Chromium explicitly relies on this
> > behavior, but I'm not sure whether this code is in active use anymore:
> >
> > https://source.chromium.org/chromium/chromium/src/+/main:sandbox/linux/suid/sandbox.c;l=101?q=CLONE_FS&sq=&ss=chromium
>
> Wait, this is absolutely nucking futs.  On a very quick inspection, the sharable things like this are fs, files, sighand, and io.    files and sighand get unshared, which makes sense.  fs supposedly checks for extra refs and prevents gaining privilege.  io is... ignored!  At least it's not immediately obvious that io is a problem.
>
> But seriously, this makes no sense at all.  It should not be possible to exec a program and then, without ptrace, change its cwd out from under it.  Do we really need to preserve this behavior?

I agree that this is pretty wild.

The single user I'm aware of is Chrome, and as far as I know, they use
it for establishing their sandbox on systems where unprivileged user
namespaces are disabled - see
<https://chromium.googlesource.com/chromium/src/+/main/docs/linux/suid_sandbox.md>.
They also have seccomp-based sandboxing, but IIRC there are some small
holes that mean it's still useful for them to be able to set up
namespaces, like how sendmsg() on a unix domain socket can specify a
file path as the destination address.

(By the way, I think maybe Chrome wouldn't need this wacky trick with
the shared fs_struct if the "NO_NEW_PRIVS permits chroot()" thing had
ever landed that you
(https://lore.kernel.org/lkml/0e2f0f54e19bff53a3739ecfddb4ffa9a6dbde4d.1327858005.git.luto@amacapital.net/)
and Mickaël Salaün proposed in the past... or alternatively, if there
was a way to properly filter all the syscalls that Chrome has to
permit for renderers.)

(But also, to be clear, I don't speak for Chrome, this is just my
understanding of how their stuff works.)