On Tue, 29 Aug 1995 08:37:06 -0500 (CDT) 
 rhealey@kas.helios.mn.org (Rob Healey) wrote:

 > 	Noone has answered my question as to why creating a dkn variable
 > 	is the correct solution and using the kernel variable dk_ndrive (iostat
 > 	uses to determine the number of drives in the system) is the
 > 	wrong solution.

Did you use send-pr?  I must've missed it...

 > 	It seems to me that dkn was a loose end hanging around in the x86
 > 	port.

...and the hp300, and the sparc, and ... It seems like it's an historic 
BSD (for lack of a better term) thing.

 > 	So, can someone answer me as to why dk_ndrive is the WRONG variable
 > 	to use in ccd.c? It would appear to be how the kernel keeps track
 > 	of the number of "disks" there are in the system.

It's not readily clear to me _what_ to use at this point ... that's why I 
asked for input :-)  Indeed, cgd and I discussed the dk foo quite a bit 
while I was working on the ccd.  This is not really a new issue :-)

--------------------------------------------------------------------------
Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939