--envbJBWh7q8WU6mo
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sat, Jun 24, 2006 at 03:26:06PM +0200, Juergen Hannken-Illjes wrote:
> On Fri, Jun 23, 2006 at 10:26:00AM -0700, Bill Studenmund wrote:
> [snip]
>=20
> Correct me if I'm wrong... Could this simply replace vn_lock/VOP_UNLOCK
> outside of file systems?  It looks like we would get always things like

No.

For two reasons. 1) the locking scopes are vastly different. 2) We don't=20
know what locking will be needed by the transaction. Merging the two=20
operations may be a huge mess.

>     vn_hold(vp, V_WAIT);
>     vn_lock(vp, LK_...);
>     ...
>     VOP_UNLOCK(vp);
>     vn_release(vp);
>=20
> What would we need to support layered file systems?

Layered file systems other than union would just lock the underlying file=
=20
system.

Take care,

Bill

--envbJBWh7q8WU6mo
Content-Type: application/pgp-signature
Content-Disposition: inline

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

iD8DBQFEoANUWz+3JHUci9cRAjoWAKCVNN4bHHjQN0/VqwRqM44Zpbn51ACdHFsL
V9yzaZ7okHSKAghJLJiE1Qg=
=c5Pd
-----END PGP SIGNATURE-----

--envbJBWh7q8WU6mo--