Subject: Re: new kpi proposal, sysdisk(9)
To: YAMAMOTO Takashi <yamt@mwd.biglobe.ne.jp>
From: Bill Studenmund <wrstuden@netbsd.org>
List: tech-kern
Date: 12/30/2006 10:35:31
--ew6BAiZeqk4r7MaW
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sun, Dec 31, 2006 at 02:24:46AM +0900, YAMAMOTO Takashi wrote:
> > I agree we have a problem with wedges in that they act as independent=
=20
> > devices as opposed to magic, free-form partitions. Thus detecting overl=
ap=20
> > is harder; we don't have a central repository like we do with a partiti=
on=20
> > table. But that's arguably a bug of our wedge implementation. ;-)
>=20
> dkwedge_add() has code to detect overlapped wedges.
> isn't it what you are talking about?

Kinda. Jason and I have a difference of opinion on this. :-)

I personally think we should permit overlapping wedges and enforce the=20
no-overlap on open. Jason, obviously, believes we should only have=20
non-overlapping ones and thus no overlap problem.

I want conceptual support for overlapping wedges mainly for how MBR disks=
=20
get partitioned. I'd like a wedge for an extended partition, and a wedge=20
for the file system there-in and a wedge for the extended partition in the=
=20
extended partition and a wedge for the file system in that and so on. I'd=
=20
like the non-file-system ones as they are what the partitioning tools work=
=20
with.

How exactly we maintain a name space of these, I'm not sure. :-|

Take care,

Bill

--ew6BAiZeqk4r7MaW
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (NetBSD)

iD8DBQFFlrFzWz+3JHUci9cRAp4VAJwL9SYA6mQFk6vMIyfw9FTgETWwDQCbBllz
2LKl7sa3nxGhtTuCtfTQeIM=
=uldM
-----END PGP SIGNATURE-----

--ew6BAiZeqk4r7MaW--