Request Sense DESC bit, minor conflict between MMC-6 and SPC-4

Ralph Weber roweber at IEEE.org
Tue Apr 21 16:39:21 PDT 2009


* From the T10 Reflector (t10 at t10.org), posted by:
* Ralph Weber <roweber at ieee.org>
*
Mike,
To my mind, the SPC-3 wording represents the typical T10 approach
to handling Reserved bits that are subsequently given a new meaning
(i.e., the sender of the new bit setting must be prepared to
deal with old-device behaviors).
In SPC-4, CAP apparently felt that the normal Reserved bit rules
no longer applied when it approved 06-264r2:
http://www.t10.org/cgi-bin/ac.pl?t=d&f=06-264r2.pdf
SPC-4 REQUEST SENSE parameter data clarifications [Elliott]
Now that you have exposed the question, I guess CAP will have
to reconsider how best to represent backwards compatibility
in this regard (or even whether backwards compatibility matters).
All the best,
.Ralph
Mike Berhan wrote:
> * From the T10 Reflector (t10 at t10.org), posted by:
> * "Mike Berhan" <mikeb at bustrace.com>
> *
> The MMC specification refers one to SPC-3 for the Request Sense CDB
> definition.  It states this regarding the DESC bit in the CDB:
>
> "Since MM Drives support only a 32-bit LBA format, MM Drives ignore the
> setting of the Desc bit in the
> REQUEST SENSE command CDB and return only fixed format sense data.
>
> SPC-3 does not seem to be specific (unless I've missed it) on what the
drive
> should do if it doesn't support Descriptor format sense data but receives a
> DESC=1 request.  However, SPC-4 is very clear on this in Table 251 (DESC
> bit).  It states:
>
> "The device server shall return no parameter data and terminate the REQUEST
> SENSE command with CHECK CONDITION status with the sense key set to ILLEGAL
> REQUEST and the additional sense code set to INVALID FIELD IN CDB."
>
> A quick test of the optical drives I have attached show that they do ignore
> the DESC bit thereby meeting the MMC requirement but contradicting the SPC
> requirement.	This is a minor issue as I suspect no optical software (other
> than test software) issues a Request Sense with the DESC bit set.  It is a
> conflict between MMC-6 and SPC-4 that I don't believe has been mentioned
> before.
>
> It seems to me that MMC-6 should adopt what SPC-4 has defined but I'll
leave
> that to the two committees if they're interested in addressing this.
>
> Regards,
>
> Mike
>
>
> *
> * For T10 Reflector information, send a message with
> * 'info t10' (no quotes) in the message body to majordomo at t10.org
>
>
>
>   
*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org



More information about the T10 mailing list