Path: utzoo!attcan!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!math.lsa.umich.edu!math.lsa.umich.edu!hyc
From: hyc@math.lsa.umich.edu (Howard Chu)
Newsgroups: comp.os.minix
Subject: Re: minix ST 1.5 patch 3
Message-ID: <1990Oct13.120111.21252@math.lsa.umich.edu>
Date: 13 Oct 90 12:01:11 GMT
References: <meulenbr.655388337@cst> <1320@carol.fwi.uva.nl>
Sender: usenet@math.lsa.umich.edu
Organization: University of Michigan Math Dept., Ann Arbor
Lines: 44

In article <1320@carol.fwi.uva.nl> croes@fwi.uva.nl (Felix A. Croes) writes:
>meulenbr@cst.philips.nl (Frans Meulenbroeks) writes:
>>Hi. Here is patch 3 for minix ST 1.5

>[...]

>>The floppy driver can now deal with various disk formats. 
>>This is done through the boot block.

>[...]

>>Small note: the fs changes in this posting are not yet accepted by ast.
>>If he decides not to incorporate them (or something functionally
>>equivalent) both the support for removable hard disks and the new floppy
>>driver may not be usable in future versions. 
>>Of course I'll try to persuade him to incorporate these changes.

>  Why not use a separate (minor) device type for each disk format? That is the
>tidy way of doing things, no patches to the fs are required and the cdiff to
>stfloppy.c is so small it's almost trivial.

This implies that you know in advance all the different formats you will
encounter. In reality, ST disk drives are not all like (obviously!), some
will successfully format out to 85 tracks most only out to 82 or so. Some
can reliably use 11 sectors per track, most won't. This change allows you
to use properly formatted disks (by some clever TOS program, I guess) and
not have to worry what format it is and which device name/minor dev number
to reference.

It also allows using high density floppy disks, again, without having to
expand table entries in the fs for new device number assignments. I don't
think having to recompile the FS every time I want to read a different
disk is very tidy at all.

Just as a point of reference, the standard format is a 720K disk of 80 tracks
and 9 sectors per track. I routinely use 10 sectors per track and 82 tracks.
That's standard for me, but not so standard that I'm going to hardcode those
numbers into the device driver. The hardware is flexible, why shouldn't the
software be as well?
--
  -- Howard Chu @ University of Michigan
  one million data bits stored on a chip, one million bits per chip
	if one of those data bits happens to flip,
		one million data bits stored on the chip...