Subject: Re: namei() for binary emulations
To: Christos Zoulas <christos@astron.com>
From: Quentin Garnier <cube@cubidou.net>
List: tech-kern
Date: 07/06/2005 01:09:21
--DKU6Jbt7q3WqK7+M
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Jul 06, 2005 at 01:16:14AM +0300, Christos Zoulas wrote:
> In article <20050705211915.GA1749@gallia.cubidou.net>,
> Quentin Garnier  <cube@cubidou.net> wrote:
> >-=3D-=3D-=3D-=3D-=3D-
> >
> >On Tue, Jul 05, 2005 at 01:30:29PM -0700, Jason Thorpe wrote:
> >>=20
> >> On Jul 5, 2005, at 12:12 PM, matthew green wrote:
> >>=20
> >> >an easier fix would be to use relative links in {/emul/netbsd32,}/=20
> >> >usr/lib*.
> >>=20
> >> Hm, maybe we need a @emul magic link? :-)
> >
> >And how does namei() guess what to expand it to?  That's the whole
> >point of my hack, making namei() aware it is being used under binary
> >emulation.
>=20
> There is /../ that by-passes /emul lookup, and install -l r that helps
> with relative symlinks.

But /../ works only for paths passed to the syscall, it doesn't work
during resolution (i.e., a symlink to /../dir is not magic).

--=20
Quentin Garnier - cube@cubidou.net - cube@NetBSD.org
"When I find the controls, I'll go where I like, I'll know where I want
to be, but maybe for now I'll stay right here on a silent sea."
KT Tunstall, Silent Sea, Eye to the Telescope, 2004.

--DKU6Jbt7q3WqK7+M
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (NetBSD)

iQEVAwUBQssTIdgoQloHrPnoAQK1pggAsChzqbSYmlWEC5ko72p/806xTLXk7xI+
4+4AX5vB4dk9K9OIP1NwIckEPslkknaxxY+KWw0vXZqBnoo1WCv50isOOKpjpdYl
nnk85kl0b+2W5G8p61OCDz1Rg6DFRN7jrRn9ULHAXfAFdy9oqg95XGhKXR37NuLR
qcjGy0nJKGqbXkJVtYpssQT+BwMFeR7lrSx3YjZKhptEhZiTWE5oqYpgEjKf1q8R
hqxVMf79tRBIcmESdk4nEX97NkzUUL0TsSEMQSMKDyavP1HftEijv5DhEDVI5yid
9YUqqfW/uMCasxTqBS4jez2snxHKKPfQHfJfq9MKaC8Nalxnutxmdw==
=h2MT
-----END PGP SIGNATURE-----

--DKU6Jbt7q3WqK7+M--