Subject: Re: emul vs. symlinks
To: Rhialto <rhialto@azenomei.knuffel.net>
From: Bill Studenmund <wrstuden@netbsd.org>
List: tech-kern
Date: 07/26/2004 09:09:56
--/NkBOFFp2J2Af1nK
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Jul 21, 2004 at 03:44:24PM +0200, Rhialto wrote:
> On Tue 13 Jul 2004 at 23:25:21 +0200, Quentin Garnier wrote:
> > Le Mon, 12 Jul 2004 13:13:40 -0700
> > Bill Studenmund a ecrit :
> > > I think if we do this right, we can get rid of emul_find().
> >=20
> > Well, I'm not so sure about this, since it's syscall-dependent.  Arguab=
ly
> > we could flag some way the calling lwp.
>=20
> Well, if we do it "right", we'll have introduced an algorithm with
> exponential complexity in the name-lookup mechanism (exponential in the
> number of traversed symlinks). I smell a Denial Of Service attack, or at
> least bad system performance. Do we want that?

I don't think we face such an attack for two reasons. 1) The limit on the=
=20
number of symlinks traversed would still apply. 2) You have to be root to=
=20
add /emul/foo directories, and if you were root, there would be a lot of=20
other nasty ways to kill the system.

Take care,

Bill

--/NkBOFFp2J2Af1nK
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)

iD8DBQFBBSzUWz+3JHUci9cRApJRAJoCKClXZbR1LoWngA4CpznbR1On8gCeNhvh
wqBrU8EvUUVAjTZPevcuDJM=
=qmfk
-----END PGP SIGNATURE-----

--/NkBOFFp2J2Af1nK--