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:   Tue, 16 Jun 2020 11:35:46 -0700
From:   Kees Cook <keescook@chromium.org>
To:     Andy Lutomirski <luto@kernel.org>
Cc:     LKML <linux-kernel@vger.kernel.org>,
        Christian Brauner <christian@brauner.io>,
        Sargun Dhillon <sargun@sargun.me>,
        Tycho Andersen <tycho@tycho.ws>, Jann Horn <jannh@google.com>,
        "zhujianwei (C)" <zhujianwei7@huawei.com>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        Matthew Wilcox <willy@infradead.org>,
        Will Drewry <wad@chromium.org>, Shuah Khan <shuah@kernel.org>,
        Matt Denton <mpdenton@google.com>,
        Chris Palmer <palmer@google.com>,
        Jeffrey Vander Stoep <jeffv@google.com>,
        Aleksa Sarai <cyphar@cyphar.com>,
        Hehuazhen <hehuazhen@huawei.com>, X86 ML <x86@kernel.org>,
        Linux Containers <containers@lists.linux-foundation.org>,
        LSM List <linux-security-module@vger.kernel.org>,
        Linux API <linux-api@vger.kernel.org>
Subject: Re: [RFC][PATCH 0/8] seccomp: Implement constant action bitmaps
Message-ID: <202006161131.5A21C01@keescook>
References: <20200616074934.1600036-1-keescook@chromium.org>
 <CALCETrVLA22khom-iSu5iTJGKfUykOnwyz8j7Tm9g6hWW_e0Aw@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CALCETrVLA22khom-iSu5iTJGKfUykOnwyz8j7Tm9g6hWW_e0Aw@mail.gmail.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, Jun 16, 2020 at 10:01:43AM -0700, Andy Lutomirski wrote:
> On Tue, Jun 16, 2020 at 12:49 AM Kees Cook <keescook@chromium.org> wrote:
> >
> > Hi,
> >
> 
> > In order to build this mapping at filter attach time, each filter is
> > executed for every syscall (under each possible architecture), and
> > checked for any accesses of struct seccomp_data that are not the "arch"
> > nor "nr" (syscall) members. If only "arch" and "nr" are examined, then
> > there is a constant mapping for that syscall, and bitmaps can be updated
> > accordingly. If any accesses happen outside of those struct members,
> > seccomp must not bypass filter execution for that syscall, since program
> > state will be used to determine filter action result.
> 
> >
> > During syscall action probing, in order to determine whether other members
> > of struct seccomp_data are being accessed during a filter execution,
> > the struct is placed across a page boundary with the "arch" and "nr"
> > members in the first page, and everything else in the second page. The
> > "page accessed" flag is cleared in the second page's PTE, and the filter
> > is run. If the "page accessed" flag appears as set after running the
> > filter, we can determine that the filter looked beyond the "arch" and
> > "nr" members, and exclude that syscall from the constant action bitmaps.
> 
> This is... evil.  I don't know how I feel about it.  It's also

Thank you! ;)

> potentially quite slow.

I got the impression that (worst-case: a "full" filter for every
arch/syscall combo) ~900 _local_ TLB flushes per filter attach wouldn't be
very slow at all. (And the code is optimized to avoid needless flushes.)

> I don't suppose you could, instead, instrument the BPF code to get at
> this without TLB hackery?  Or maybe try to do some real symbolic
> execution of the BPF code?

I think the "simple emulator" path[1] might get us a realistically large
coverage. I'm going to try it out, and see what it looks like.

-Kees

[1] https://lore.kernel.org/lkml/202006160757.99FD9B785@keescook/

-- 
Kees Cook