> Before anyone does anything with new-format disklabels, I'd like to
> point out Charles' old "wedge" proposal. I think it's a very good
> proposal, and should be implemented.

I like it too; it's a lot like some other stuff that's been kicked
around on the lists.

The biggest problem I don't quite see is how it gets bootstrapped.
This is probably doable, especially given whom the proposal came from
(mycroft is far from stupid) but it's not clear from what I read there
how it would happen.  I can't see any way to avoid having *some*
disklabel knowledge in the kernel (or the bootblocks, on ports where
the bootblocks are NetBSD code - for compatability, I'd prefer to keep
it in the kernel).

As for the dump-partition idea, it occurs to me that there's no
conceptual reason the dump partition can't be changed at runtime, so
that if it turns out to be more convenient, you can have no dump
partition during early boot and then set it early in the boot scripts,
after wedgeconfig has run.

					der Mouse

			       mouse@rodents.montreal.qc.ca
		     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B