Re: devfs persistence

From: Stephen C. Tweedie (
Date: Wed May 03 2000 - 05:15:49 EST


On Wed, May 03, 2000 at 12:07:17PM +0200, Andries Brouwer wrote:
> On Tue, May 02, 2000 at 10:09:22PM +0100, Stephen C. Tweedie wrote:
> > > I wouldnt mind if only Linux-type partition tables had room for
> > > volume labels and the like. People that really want to have both
> > > DOS type partition tables and disk/partition volume labels and uuids
> > > might use a Linux-type "index" subpartition of a DOS type partition table.
> >
> > A logical device superblock in a partition already defines a
> > subpartition in this way, and is completely portable. I guess I
> > just don't see the need for yet another mechanism!
> You lost me.

The question is, what is a ``Linux-type "index" subpartition of a DOS
type partition table''?

It sounds to me like some complex structure that you fit into a single
DOS-type partition. And we already have that --- the raid superblock
lets you subdivide a DOS-type partition into the control information
(the raid superblock) and the content.

The only advantage you'd get by adding extra linux-only partitioning
would be to let you subdivide the DOS-type partitions even further, but
I don't see the advantage in that, since it doesn't actually gain us
any functionality: you can already do that with LVM.

So the way I read the discussion is:

Andries: "Why not divide a DOS partition into more bits, adding some
          control information and letting us create different linux-
          only partitions inside?"

Stephen: "Because the raid and LVM code already let you do that. Why
          do you want yet another mechanism to do the same thing?"

I don't see the advantage in making linux-only partitions. Partitions
are not the core of the problem --- block devices are. And we can fix
all the uuid problems by using logical block devices other than simple


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

This archive was generated by hypermail 2b29 : Sun May 07 2000 - 21:00:11 EST