> > 	$FSBIN/$FSTYPE/$FSCMD $ARGS
> > ...
> > 	for i in *
> > 	do
> > 		# for each FS for which an "fstyp" exists..
> > 		if test -x $i/fstyp
> >		then
> >			[stuff]
> >...
> 
> So what you're saying is that the above is significantly better than
> 
>     $FSCMD_$FSTYPE $ARGS
> 
> and
> 
>     for i in *_fstyp; do [stuff] ...
> 
> Or am I missing something here?

You're missing the fact that the file system support may have arrived
as the rsult of a union mount, and that the organization to allow
this requires that the per-FS components be logically and simply
seperable from the other components.

For instance, I may wish to have a directory overlay from /usr for
the additional fs's other than FFS, since only FFS is needed for boot.
This is only an example of something I might want to do.  I can
provide others, if necessary, and if you want to spend the time on
them.

In reality, given the discussion Doug Rabson, Mike Smith, and I had,
it should be obious that an ELF executable could be considered a
"bootfs" image, and all other FS types could result from a directed
overlay from that into a real FS name space.  That is, i support
variosu FS types as single modules which may be agregated with boot
and kernel images.

All the better to boot from VFATSFS or EXT2FS as root and have a kernel
that doesn't support FFS at all until a module is loaded from one
of the FS types it does support.


					Regards,
					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.