Re: [PATCH v2 seccomp 4/6] seccomp/cache: Lookup syscall allowlist for fast path

From: YiFei Zhu <zhuyifei1999@gmail.com>
To: Kees Cook <keescook@chromium.org>
Cc: Linux Containers <containers@lists.linux-foundation.org>,
	YiFei Zhu <yifeifz2@illinois.edu>, bpf <bpf@vger.kernel.org>,
	kernel list <linux-kernel@vger.kernel.org>,
	Aleksa Sarai <cyphar@cyphar.com>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Andy Lutomirski <luto@amacapital.net>,
	Dimitrios Skarlatos <dskarlat@cs.cmu.edu>,
	Giuseppe Scrivano <gscrivan@redhat.com>,
	Hubertus Franke <frankeh@us.ibm.com>,
	Jack Chen <jianyan2@illinois.edu>, Jann Horn <jannh@google.com>,
	Josep Torrellas <torrella@illinois.edu>,
	Tianyin Xu <tyxu@illinois.edu>,
	Tobin Feldman-Fitzthum <tobin@ibm.com>,
	Tycho Andersen <tycho@tycho.pizza>,
	Valentin Rothberg <vrothber@redhat.com>,
	Will Drewry <wad@chromium.org>
Subject: Re: [PATCH v2 seccomp 4/6] seccomp/cache: Lookup syscall allowlist for fast path
Date: Thu, 24 Sep 2020 20:55:03 -0500	[thread overview]
Message-ID: <CABqSeAS1wmCL8gRW+atNO0ZBe0JTzUcbQAR6n461AwhCotNVZg@mail.gmail.com> (raw)
In-Reply-To: <202009241640.7E3C54CF@keescook>

On Thu, Sep 24, 2020 at 6:46 PM Kees Cook <keescook@chromium.org> wrote:
> This protects us from x32 (i.e. syscall_nr will have 0x40000000 bit
> set), but given the effort needed to support compat, I think supporting
> x32 isn't much more. (Though again, I note that NR_syscalls differs in
> size, so this test needs to be per-arch and obviously after
> arch-discovery.)
>
> That said, if it really does turn out that x32 is literally the only
> architecture doing these shenanigans (and I suspect not, given the MIPS
> case), okay, fine, I'll give in. :) You and Jann both seem to think this
> isn't worth it.

MIPS has the sparse syscall shenanigans... idek how that works. Maybe
someone can clarify?

> I think this linear search for the matching arch can be made O(1) (this
> is what I was trying to do in v1: we can map all possible combos to a
> distinct bitmap, so there is just math and lookup rather than a linear
> compare search. In the one-arch case, it can also be easily collapsed
> into a no-op (though my v1 didn't do this correctly).

I remember yours was:

static inline u8 seccomp_get_arch(u32 syscall_arch, u32 syscall_nr)
{
[...]
        switch (syscall_arch) {
        case SECCOMP_ARCH:
                seccomp_arch = SECCOMP_ARCH_IS_NATIVE;
                break;
#ifdef CONFIG_COMPAT
        case SECCOMP_ARCH_COMPAT:
                seccomp_arch = SECCOMP_ARCH_IS_COMPAT;
                break;
#endif
        default:
                seccomp_arch = SECCOMP_ARCH_IS_UNKNOWN;
        }

What I'm relying on here is that the compiler will unroll the loop.
How does the compiler perform switch statements? I was imagining it
would be similar, with "case" corresponding to a compare on the
immediate, and the assign as a move to a register, and break
corresponding to a jump. this would also be O(n) to the number of
arches. Yes, compilers can also do an O(1) table lookup, but that is
nonsensical here -- the arch numbers occupy the MSBs.

That said, does O(1) or O(n) matter here? Given that n is at most 3
you might as well consider it a constant.

Also, does "collapse in one arch case" actually worth it? Given that
there's a likely(), and the other side is a WARN_ON_ONCE(), the
compiler will layout the likely path in the fast path and branch
prediction will be in our favor, right?

YiFei Zhu