Message-ID: <1448389547.27159.8.camel@decadent.org.uk>
Subject: Re: [PATCH] RDS: fix race condition when sending a message on
 unbound socket.
From: Ben Hutchings <ben@decadent.org.uk>
To: Quentin Casasnovas <quentin.casasnovas@oracle.com>,
        santosh shilimkar <santosh.shilimkar@oracle.com>
Cc: lkml <linux-kernel@vger.kernel.org>,
        Vegard Nossum <vegard.nossum@oracle.com>,
        Sasha Levin <sasha.levin@oracle.com>, Chien Yen <chien.yen@oracle.com>,
        "David S. Miller" <davem@davemloft.net>
Date: Tue, 24 Nov 2015 18:25:47 +0000
In-Reply-To: <20151103112545.GD921@chrystal.uk.oracle.com>
References: <1445008302-14169-1-git-send-email-quentin.casasnovas@oracle.com>
	 <56213845.6020900@oracle.com> <20151103112545.GD921@chrystal.uk.oracle.com>
Content-Type: multipart/signed; micalg="pgp-sha512";
	protocol="application/pgp-signature"; boundary="=-a4CKAiLxbcSunNHKoquy"
Mime-Version: 1.0
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2943
Lines: 78


--=-a4CKAiLxbcSunNHKoquy
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, 2015-11-03 at 12:25 +0100, Quentin Casasnovas wrote:
> On Fri, Oct 16, 2015 at 10:47:49AM -0700, santosh shilimkar wrote:
> > On 10/16/2015 8:11 AM, Quentin Casasnovas wrote:
> > > Sasha's found a NULL pointer dereference in the RDS connection code w=
hen
> > > sending a message to an apparently unbound socket.=C2=A0=C2=A0The pro=
blem is caused
> > > by the code checking if the socket is bound in rds_sendmsg(), which c=
hecks
> > > the rs_bound_addr field without taking a lock on the socket.=C2=A0=C2=
=A0This opens a
> > > race where rs_bound_addr is temporarily set but where the transport i=
s not
> > > in rds_bind(), leading to a NULL pointer dereference when trying to
> > > dereference 'trans' in __rds_conn_create().
> > >=20
> > > Vegard wrote a reproducer for this issue, so kindly ask him to share =
if
> > > you're interested.
> > >=20
> > > I cannot reproduce the NULL pointer dereference using Vegard's reprod=
ucer
> > > with this patch, whereas I could without.
> > >=20
> > > Complete earlier incomplete fix to CVE-2015-6937:
> > >=20
> > > =C2=A0=C2=A0=C2=A074e98eb08588 ("RDS: verify the underlying transport=
 exists before creating a connection")
> > >=20
>=20
> For reference, this has been assigned CVE-2015-7990 on the oss-sec thread=
:
>=20
> =C2=A0 http://seclists.org/oss-sec/2015/q4/179

But the patch doesn't seem to have gone anywhere. =C2=A0Santosh, can you
apply this in your tree and ask David to pull, or should Quentin re-
send it to netdev?

Ben.

--=20
Ben Hutchings
Unix is many things to many people,
but it's never been everything to anybody.
--=-a4CKAiLxbcSunNHKoquy
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
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=bTZP
-----END PGP SIGNATURE-----

--=-a4CKAiLxbcSunNHKoquy--
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/