So, I've finally been pulled into the evil world of IDE.  My EDSI controller
(or disk??) in my old 486 died, and since the disk was tiny and all the data
was on tape I thought I'd replace it with an IDE disk (since they're cheap).

I started the install, got the disk labelled and put down filesystems on it
and finally copied a kernel from my 1.2 boot disk onto it.  However, when I 
reboot, the bootblock refuses to load the kernel... I hit return after the 
prompt and the bootblocks just sit there rather stupidly.

The drive (and controller) seem to boot OK from the small DOS partition I
created on the disk to test it out, so I would venture that the problem
ins't totally controller-specific (although I'd bet that the controller is
doing something dumb that DOS, being DOS, just doesn't care about but the
NetBSD bootblocks do).

This is an old ISA machine, the controller is a SIIG CI-1050 (this is what
the card and the BIOS say, the box says 'Model CN2419').  The card does have
an on-board BIOS which claims to do LBA, but when the controller attaches the
drive the drive is attached as 780cyl/64hd/63sec rather than the native
geometry of 3121cyl/16hd/63sec that the drive claims it is.  Don't know if
this is important, though.

I've tried disklabelling the drive both ways (with the native and the 
controller-specified geometries) and still no dice.  I do have a DOS
partition on the first part of the disk, although currently it's there
just as a test and I'll probably re-do the disk as all-NetBSD when/if
I can get the controller and the bootblocks to agree, and as I said
before, I *can* boot DOS.

I recall people talking about dumb BIOS-es that don't put the boot drive
in the right register, which then confuses the bootblocks... Could this be
the problem?? If so, could someone send me a patch so I can build a new-and-
improved install floppy and give that a try??  If not, any other ideas?

Thanks for any help/ideas.
--rafal

----
Rafal Boni                                                rkboni@concentric.net