Received-SPF: pass (google.com: domain of linux-kernel+bounces-214123-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Message-ID: <5a8a3c85760c19be66965630418e09a820f79277.camel@linux.ibm.com>
Subject: Re: [PATCH v4 12/35] kmsan: Support SLAB_POISON
From: Ilya Leoshkevich <iii@linux.ibm.com>
To: SeongJae Park <sj@kernel.org>, Alexander Potapenko <glider@google.com>
Cc: Alexander Gordeev <agordeev@linux.ibm.com>,
        Andrew Morton
 <akpm@linux-foundation.org>,
        Christoph Lameter <cl@linux.com>,
        David
 Rientjes <rientjes@google.com>,
        Heiko Carstens <hca@linux.ibm.com>,
        Joonsoo
 Kim <iamjoonsoo.kim@lge.com>, Marco Elver <elver@google.com>,
        Masami
 Hiramatsu <mhiramat@kernel.org>,
        Pekka Enberg <penberg@kernel.org>,
        Steven
 Rostedt <rostedt@goodmis.org>,
        Vasily Gorbik <gor@linux.ibm.com>, Vlastimil
 Babka <vbabka@suse.cz>,
        Christian Borntraeger <borntraeger@linux.ibm.com>,
        Dmitry Vyukov <dvyukov@google.com>,
        Hyeonggon Yoo <42.hyeyoo@gmail.com>, kasan-dev@googlegroups.com,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org,
        linux-s390@vger.kernel.org, linux-trace-kernel@vger.kernel.org,
        Mark Rutland <mark.rutland@arm.com>,
        Roman Gushchin <roman.gushchin@linux.dev>,
        Sven Schnelle
 <svens@linux.ibm.com>
Date: Fri, 14 Jun 2024 01:44:39 +0200
In-Reply-To: <20240613233044.117000-1-sj@kernel.org>
References: <20240613233044.117000-1-sj@kernel.org>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.50.4 (3.50.4-1.fc39) 
Content-Transfer-Encoding: quoted-printable
Precedence: bulk
MIME-Version: 1.0

On Thu, 2024-06-13 at 16:30 -0700, SeongJae Park wrote:
> Hi Ilya,
>=20
> On Thu, 13 Jun 2024 17:34:14 +0200 Ilya Leoshkevich
> <iii@linux.ibm.com> wrote:
>=20
> > Avoid false KMSAN negatives with SLUB_DEBUG by allowing
> > kmsan_slab_free() to poison the freed memory, and by preventing
> > init_object() from unpoisoning new allocations by using __memset().
> >=20
> > There are two alternatives to this approach. First, init_object()
> > can be marked with __no_sanitize_memory. This annotation should be
> > used
> > with great care, because it drops all instrumentation from the
> > function, and any shadow writes will be lost. Even though this is
> > not a
> > concern with the current init_object() implementation, this may
> > change
> > in the future.
> >=20
> > Second, kmsan_poison_memory() calls may be added after memset()
> > calls.
> > The downside is that init_object() is called from
> > free_debug_processing(), in which case poisoning will erase the
> > distinction between simply uninitialized memory and UAF.
> >=20
> > Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
> > ---
> > =C2=A0mm/kmsan/hooks.c |=C2=A0 2 +-
> > =C2=A0mm/slub.c=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 | 13 ++++++++=
+----
> > =C2=A02 files changed, 10 insertions(+), 5 deletions(-)
> >=20
> [...]
> > --- a/mm/slub.c
> > +++ b/mm/slub.c
> > @@ -1139,7 +1139,12 @@ static void init_object(struct kmem_cache
> > *s, void *object, u8 val)
> > =C2=A0	unsigned int poison_size =3D s->object_size;
> > =C2=A0
> > =C2=A0	if (s->flags & SLAB_RED_ZONE) {
> > -		memset(p - s->red_left_pad, val, s->red_left_pad);
> > +		/*
> > +		 * Use __memset() here and below in order to avoid
> > overwriting
> > +		 * the KMSAN shadow. Keeping the shadow makes it
> > possible to
> > +		 * distinguish uninit-value from use-after-free.
> > +		 */
> > +		__memset(p - s->red_left_pad, val, s-
> > >red_left_pad);
>=20
> I found my build test[1] fails with below error on latest mm-unstable
> branch.
> 'git bisect' points me this patch.
>=20
> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 CC=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 mm/slub.o
> =C2=A0=C2=A0=C2=A0 /mm/slub.c: In function 'init_object':
> =C2=A0=C2=A0=C2=A0 /mm/slub.c:1147:17: error: implicit declaration of fun=
ction
> '__memset'; did you mean 'memset'? [-Werror=3Dimplicit-function-
> declaration]
> =C2=A0=C2=A0=C2=A0=C2=A0 1147 |=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 __memset(p - s->red_=
left_pad, val, s-
> >red_left_pad);
> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0 ^~~~~~~~
> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0 memset
> =C2=A0=C2=A0=C2=A0 cc1: some warnings being treated as errors
>=20
> I haven't looked in deep, but reporting first.=C2=A0 Do you have any idea?
>=20
> [1]
> https://github.com/awslabs/damon-tests/blob/next/corr/tests/build_m68k.sh
>=20
>=20
> Thanks,
> SJ
>=20
> [...]

Thanks for the report.

Apparently not all architectures have=C2=A0__memset(). We should probably go
back to memset_no_sanitize_memory() [1], but this time mark it with
noinline __maybe_unused __no_sanitize_memory, like it's done in, e.g.,
32/35.

Alexander, what do you think?

[1]
https://lore.kernel.org/lkml/20231121220155.1217090-14-iii@linux.ibm.com/