A bit rearranged and mailed to you.  Maybe the rest of
the crowd is getting tired of all this .

Wolfgang Solfrank <ws@tools.de> writes:

I did get this.  Indented below is my previous mail.  I think its
subtle and ok.  But I agree its what dos does.
> In addition, you still seem to have missed the part where I wrote "...
> during the first time it writes the mbr to disk".  Or maybe, I haven't made
> that clear enough?  What I meant is, the boot code and partition table are
> initialized on disk, if you instruct fdisk to write the mbr anyway.  I.e.,
> if you call fdisk with any of the options -i, -u or -a.

	> Actually it will initialize its incore data structures even if it
	> doesn't write to disk, so it will tell sysinst that the partition table
	> is empty.  With sysinst, you then add a NetBSD partition, and when
	> sysinst calls fdisk again to write this partition (using the fdisk -u...
	> below the fdisk -i you removed), fdisk will initialize the mbr on disk.
	It is an extension that I had not appreciated earlier, that fdisk would
	install boot code (i.e. fdisk -i), when it was installing mbr
	partition records, if the mbr was invalid.


> But that's not really different from what we have already.  The old fdisk
> code would do mostly the same (yes, even if _not_ called with the -i option).
> The only difference is that the old code does initialize the partition entry
> 3 as a partition covering the whole disk except the first track, while the
> new code initializes all four partitions to zero.

Your plan does correctly, what the old code tried to do w/o ever messing up 
the mbr partition records.  I'm arguing for something slightly better?
A menu to tell the user that there is no mbr and ask what to do.