> I had some trouble with the Sparc's need for partitions being whole
> multiples of cylinders, but it works now.

In my experience they don't need to be.  The restriction is that
partitions must _begin_ on cylinder boundaries (or at least what the
disklabel says are cylinder boundaries), but I've never had any trouble
using _sizes_ that are not multiples of the "cylinder" size.

> I then built a ccd with the thing.  Interesting to note that the
> disktab entry for the ccd creates a label where [ns=2048/nt=1]
> resulting in the same net effect you suggested with [ns=64/nt=32]

> Any pros or cons to either approach or is this all just arbitrary
> fiction?

There probably are, but I don't know what they are.  Someone who knows
the ins and outs of FFS in more detail than I could probably say.

					der Mouse

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