Subject: Re: xmcd with ATAPI?
To: Jason Downs <downsj@teeny.org>
From: Jason Thorpe <thorpej@nas.nasa.gov>
List: port-i386
Date: 12/14/1996 09:51:51
On Sat, 14 Dec 1996 09:48:17 -0800 
 Jason Downs <downsj@teeny.org> wrote:

 > It already does, just hack the "FreeBSD ioctl interface" to compile under
 > NetBSD.  i.e., add it to the ifdefs, and then configure xmcd to use that
 > instead of raw SCSI.

Ah, cool...

 > It's arguable that the raw SCSI interface in xmcd is bogus, since raw SCSI
 > is so insecure.

Err, why?  "Raw SCSI" is just fine ... as long as you have write access
to the device, who cares?  Alternatively, the scsi_ioctl() function could
check for a certain set of commands that are known to not alter the
device's state (MODE SENSE, READ, etc.) and allow only those is the device
is open for reading, and allow all if the device is open for writing.

 > >	(2) the ATAPI code needs to become a `scsibus' attached to
 > >	    an IDE controller (which is how BSD/OS does it).
 > 
 > This is a gruesome hack in BSD/OS, too, and I've never seen it work nearly
 > as well as they'd like to pretend it does.

...considering what ATAPI is, the `scsibus' approach is exactly the right
one.

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