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;
Date:   Sat, 30 May 2020 09:09:29 -0700
From:   Kees Cook <keescook@chromium.org>
To:     Christian Brauner <christian.brauner@ubuntu.com>
Cc:     Jann Horn <jannh@google.com>, Sargun Dhillon <sargun@sargun.me>,
        Linux Containers <containers@lists.linux-foundation.org>,
        Aleksa Sarai <cyphar@cyphar.com>,
        Jeffrey Vander Stoep <jeffv@google.com>,
        Linux API <linux-api@vger.kernel.org>,
        kernel list <linux-kernel@vger.kernel.org>,
        Chris Palmer <palmer@google.com>,
        Robert Sesek <rsesek@google.com>,
        Tycho Andersen <tycho@tycho.ws>,
        Matt Denton <mpdenton@google.com>,
        Al Viro <viro@zeniv.linux.org.uk>
Subject: Re: [PATCH v2 2/3] seccomp: Introduce addfd ioctl to seccomp user
 notifier
Message-ID: <202005300908.303040EF9@keescook>
References: <20200528110858.3265-1-sargun@sargun.me>
 <20200528110858.3265-3-sargun@sargun.me>
 <202005282345.573B917@keescook>
 <20200530011054.GA14852@ircssh-2.c.rugged-nimbus-611.internal>
 <202005291926.E9004B4@keescook>
 <CAG48ez0+BvbLoSc+zcZwnwfOSCFt2LHnUkzzt-d4LQFJYXZC9w@mail.gmail.com>
 <20200530135827.cxltfmiqara4yaki@wittgenstein>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200530135827.cxltfmiqara4yaki@wittgenstein>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Sat, May 30, 2020 at 03:58:27PM +0200, Christian Brauner wrote:
> On Sat, May 30, 2020 at 05:17:24AM +0200, Jann Horn wrote:
> > On Sat, May 30, 2020 at 4:43 AM Kees Cook <keescook@chromium.org> wrote:
> > > I mean, yes, that's certainly better, but it just seems a shame that
> > > everyone has to do the get_unused/put_unused dance just because of how
> > > SCM_RIGHTS does this weird put_user() in the middle.
> > >
> > > Can anyone clarify the expected failure mode from SCM_RIGHTS? Can we
> > > move the put_user() after instead?
> > 
> > Honestly, I think trying to remove file descriptors and such after
> > -EFAULT is a waste of time. If userspace runs into -EFAULT, userspace
> [...]
> 
> There's really no point in trying to save a broken scm message imho.

Right -- my concern is about stuffing a fd into a process without it
knowing (this is likely an overly paranoid concern, given that if the
process is getting EFAULT at the end of a list of fds, all the prior
ones will be installed too..)

-- 
Kees Cook