umass/cam headaches

[M. Warner Losh]

In message: <20070705184309.GD2749 at roadrunner.q.local>
            Ulrich Spoerlein <uspoerlein at gmail.com> writes:
: I'm still fighting with getting my ext. Plextor drive to read retail
: DVDs and noticed these new printfs. I don't know *what* has been done to
: CAM or cd(4) or umass(4) during the last week. But my drive is working
: way better than before.

I'm unsure what I've done, but you are welcome.

: Still lots of oddities:
: Jul  4 21:10:26 roadrunner kernel: umass0: Unsupported ATAPI command 0x46 - trying anyway
: Jul  4 21:10:26 roadrunner kernel: umass0: Unsupported ATAPI command 0x4a - trying anyway

These are status commands for the media, as you've discovered.

: Or these here with a different media
: Jul  5 20:20:12 roadrunner kernel: (cd0:umass-sim0:0:0:0): READ(10). CDB: 28 0 0 23 4 60 0 0 1 0 
: Jul  5 20:20:12 roadrunner kernel: (cd0:umass-sim0:0:0:0): CAM Status: SCSI Status Error
: Jul  5 20:20:12 roadrunner kernel: (cd0:umass-sim0:0:0:0): SCSI Status: Check Condition
: Jul  5 20:20:12 roadrunner kernel: (cd0:umass-sim0:0:0:0): ILLEGAL REQUEST asc:6f,3
: Jul  5 20:20:12 roadrunner kernel: (cd0:umass-sim0:0:0:0): Reserved ASC/ASCQ pair

I have no clue what this means.  But it isn't 'unsupported command' so
I don't think quirking would help.

: Googleing around, it looks like Linux is defining these codes as:
: #define GET_CONFIGURATION 0x46
: #define GET_EVENT_STATUS_NOTIFICATION 0x4a
: 
: libcdio:
: CDIO_MMC_GPCMD_GET_EVENT_STATUS = 0x4a
: 
: Looking at /sys/cam/scsi/scsi_cd.h it looks like it could handle at
: least 0xa3 and 0xa4. So should umass.c simply pass them through?

Chances are those commands caused problems for other devices.

: Then what about 0x46 and 0x4a? Could they be the source of my problem?
: What's with the SCSI status errors? Is READ(10) telling me it's using 10
: byte commands? The CDB is 10 bytes.

I'm afraid I don't know your problem.  What kind of retail DVDs are
you reading?  Movies?

Warner