Apparent mode sub-page conflict in SPC-3 and SPC-4

Ralph Weber roweber at IEEE.org
Sun Nov 14 20:18:53 PST 2010


Formatted message: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1011141_f.htm">HTML-formatted message</a>

Regarding:
> That disagreement is extended in SPC-4 Revision 01 where table 244 in the
mode parameters section was modified to add:
>	0Ah F0 - FEh (See specific device type)
>
> I wasn't able to find any documentation on why that change was made except
maybe due to the comment "Several purely editorial changes made by the ANSI
Editor in the published version of SPC-3 were included in SPC-4 r01."
The cited change is a clear extension of the first change proposed
(top of page 2) in approved document 04-397r3 
<http://www.t10.org/cgi-bin/ac.pl?t=d&f=04-397r3.pdf&gt;. Inclusion of these
changes in SPC-4 was approved (implicitly) by the March 2005 SAT
working group (minutes in 05-095r0 
<http://www.t10.org/cgi-bin/ac.pl?t=d&f=05-095r0.pdf&gt;).
As to the question of how Mode Subpage codes E0h to FEh are
intended to be used ... This is clearly a mess.
The conflict that Curtis describes has existed for all of
the life of SPC-4.
    * In the MODE SENSE(6) command, the E0h to FEh subpage codes
      are defined as vendor specific, but ...
    * In the Mode Parameters subclause, the *only* vendor specific
      code is page/subpage code combination is 00h/??h
There is no need to dredge up ancient history. SPC-4 r27 
<http://www.t10.org/cgi-bin/ac.pl?t=f&f=spc4r27.pdf&gt;
exhibits the conundrum perfectly well.
Comments on how best to proceed are hereby solicited.
All the best,
.Ralph
On 11/11/2010 9:10 AM, Ballard, Curtis C (StorageWorks) wrote:
> * From the T10 Reflector (t10 at t10.org), posted by:
> * "Ballard, Curtis C (StorageWorks)"<curtis.ballard at hp.com>
> *
> In SPC-3 Revision 23 and in SPC-4 Revision 00 there are two tables that
show mode sub-pages:
>
> Table 99 in the MODE SENSE(6) text; and
> Table 244 in section 7.4.5, Parameters for all device types, Mode
parameters, page/sub-page formats and codes
>
> In those documents the MODE SENSE(6) text specifies the following:
>	Mode pages 01h - 1Fh sub-pages E0h - FEh is vendor specific (sub-page
format)
>
> And the mode parameters section where ranges of sub-pages are listed
specifies the following:
>	18h 01h - FEh (See specific SCSI transport protocol)
>	19h 01h - FEh (See specific SCSI transport protocol)
>	01h 00h - FEh (See specific device type)
>	03h - 08h 00h - FEh (See specific device type)
>	0Bh - 14h 00h - FEh (See specific device type)
>	1Bh 00h - FEh (See specific device type)
>	1Dh - 1Fh 00h - FEh (See specific device type)
>	20h - 3Eh 00h - FEh (See specific device type)
>
> Is these two versions there is disagreement between the tables for several
mode page/sub-page combinations because the sub-page range for E0h - FEh is
vendor specific in table 99 and says to see specific device type in table
244.  The reference to the specific device type implies that the specific
device type can assign one of those sub-pages.
>
> That disagreement is extended in SPC-4 Revision 01 where table 244 in the
mode parameters section was modified to add:
>	0Ah F0 - FEh (See specific device type)
>
> I wasn't able to find any documentation on why that change was made except
maybe due to the comment "Several purely editorial changes made by the ANSI
Editor in the published version of SPC-3 were included in SPC-4 r01."
>
> At least one mode sub-page specified as "vendor specific" in the MODE
SENSE(6) text has been assigned in the current draft of SPC-4, Mode Page 0Ah,
Sub-page F0h which is the "Control Data Protection Mode Page".
>
> Is the E0h - FEh range vendor specific or is it defined in the applicable
command standard?  If it is supposed to be vendor specific SSC-4 needs to
select a different sub-page value and the SPC mode parameters lists need
synchronized.
>
> Curtis Ballard
> Hewlett Packard
>
> *
> * 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