[current-users: this started on port-i386.  Read on to see the
cross-port relevance that's prompting me to send to current-users.]

[all: when replying, remember to check which list you're replying to!]

> Even if we have 16* partitions allowed, [...]

> both ufs & ffs will support upto 8 logical partitions within the
> physical partition.

The i386 port is odd in that it has two levels of partitioning; there
is FDISK-style partitioning, and within a NetBSD FDISK-style partition,
there is traditional-UNIX (sub)partitioning.  ufs/ffs runs within a
single block device; it doesn't give a damn what's behind that, as long
as the driver supports I/O to it.  Thus, it really doesn't make sense
to speak of ufs/ffs as supporting any particular number of partitions.

> * - the increase to 16 logical partitions is only a recent
> addition(?) to NetBSD - and I don't know for sure - unless I'm
> confused with FreeBSD.

Long ago, I increased the partitions-per-drive count from 8 to 16 on
the sparc and sun3 ports; the changes are still private to me (though
some version of the sparc changes is in the PR database - I can look up
the PR number if anyone's interested), but I'll be glad to give a copy
of them to anyone interested.  (The biggest problem involved in
switching the number of partitions per drive is the brief period during
which the kernel disagrees with /dev about how many partitions there
are per drive - you have to either use only drive 0 during that period
or else use the "wrong" names so as to get the right minor number.)

I've been trying to think of some way to handle things so that the
number of partitions per pack can change without causing upgrade hell
for people who have trouble with handling that overlap interval.  I
haven't come up with a scheme I'm really satisfied with, yet; I'd love
to discuss this with anyone who has any ideas on the matter.

					der Mouse

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