Re: scsictl identifies ATAPI CD drive as related to scsibusN (but it isn't)

[Martin Husemann <martin%duskware.de@localhost>]

On Sun, Jun 14, 2009 at 05:25:11PM +0200, Magnus Eriksson wrote:
> So what is going on anyway, does scsictl simply print the wrong bus type 
> in this case?

At the lower kernel levels, scsi and atapibuses share common code.

> # scsictl cd0 identify
> /dev/rcd0d: scsibus0 target 0 lun 0 <_NEC, DVD_RW ND-3550A, 1.05>

So in case of atapi, lun will always be 0, target will be 0 for master
and 1 for slave, and the bus name just is plain wrong, and the count is
local to the controller the bus hangs off.

We could make scsictl detect the parent bus type and do pretty output,
or just s/scsibus/bus/ and live with the vagueness.

Martin