> True, but we still have to deal with that old software.  And aren't
> there still some machines out there that do use C/H/S addressing?
> We still should support them, if they exist.

And what does this have to do with the price of crack in Tibet?  We
*do* support them.  That has nothing to do with what geometry NetBSD
or the BIOS uses.

> I'd rather not have to use fdisk to find the "geometry".

Then how in the Hell do you expect it to calculate the CHS address of
the partitions correctly?  It *has* to know the geometry of the disk.
(And indeed the DOS FDISK gets this information from the BIOS.)  There
are a finite number of places for it to get this information from, and
we already have one (disklabel(8)) that's flexible and works.

> I assumed we were trying to make it easier for "joe average user" to
> install on a machine they have.  This situation sounds much more like
> what a developer might do.  At least the user would have more
> knowledge about the system than "joe average user".  And yes, I agree,
> if a disk was moved from one SCSI controller to another one with
> different BIOS C/H/S parameters, you would loose if the partition
> didn't start at sector 0 or such.  But that is a problem I suspect is
> not unique to NetBSD.  That just shows how poor the MBR design is.

No, it shows that we need to give the user a way to change what NetBSD
thinks the geometry is -- just like the user can change the BIOS's
view of the geometry (at least for IDE/ATA disks) using the setup
program.  The mechanism for doing this is disklabel(8).  I simply fo
not comprehend why you want to introduce a new mechanism that people
have to learn.

> Ok, please tell me how to do the dedicated install.

That's not particularly relevant until the other known problems are
fixed.

> Do you want the kernel to present these to the user at boot?

Of course not; that would be completely obnoxious.  I want them to be
presented as options in the installation program's partitioning tool
-- if needed.

> So how does the install tools get all the BIOS geometries and how does
> it know which disks have been successfully matched and which one have
> not?

I didn't specify that.  If I have to work out all the details, then I
might as well do it myself.  (Indeed, it looks like I may have to
anyway...)

> I would think a simple message telling the user that the NetBSD a
> partition must be located entirely within the first X Mb of the disk
> would be sufficient.  You could even put in a reference to the
> description of why this is needed.  X will need to be computed 
> by the install program depending on the C/H/S of the BIOS.

Great.  So we confuse the user with a number that's seemingly pulled
out of the program's arse, and then say `to become unconfused, refer
to this section in the back of the manual, which hopefully you printed
out before booting the installation program'.

That's not quite my idea of user-friendly.

> We also need
> to know if we have to restrict the a partition to the first 
> 1024 cylinders.

If (and only if) the boot program is unable to use `large disk'
support to get around the geometry issue.