Subject: Disk partitioning (again, sigh)
To: None <port-i386@NetBSD.ORG>
From: Ken Hornstein <kenh@cmf.nrl.navy.mil>
List: port-i386
Date: 07/31/1998 02:08:43
Alright, even though I've done this a billion times, I guess I'm simply
an idiot.

Remember that friend's Sony Vaio laptop I was trying to get NetBSD
installed on?  Well, I've got it on there, except the dang thing
won't boot NetBSD.

Well, that's not quite true ... I can boot of a floppy and tell the boot
blocks to use wd0a:netbsd, and it works just fine.  But when I set
the NetBSD partition active, or a try to use OS-BS, I get "No operating
system".

Now, I blew away and re-created the partition a couple of times (don't
ask), so I _think_ what happened is that I lost track of the original
translated geometry, and I partitioned the disk with the wrong translated
geometry, and the C/H/S partition labels don't point to the right
spot (because I'm using the wrong translated geometry).

Now, the question(s) I have are:

- Does my analysis make sense?  (One of these days I'm going to understand
  this whole mess completely!)
- Is there a way to determine the "correct" translated geometry
  associated with an IDE drive?  BTW, the BIOS on this machine
  _SAYS_ it's using the "correct" geometry (4000+ cylinders), so
  I'm assuming that gets translated _somewhere_ ... or am I completely
  off-base?
- Do MBR partitions have to start on a cylinder boundry to work
  correctly?
- As I understand it, the crucial limitation is that the BIOS only
  has enough bits for 1023 cylinders, so the other parameters have
  to be made larger so the cylinders can be made smaller.  Is that
  right?

--Ken