Sivan, George, Yes, I believe we agreed to effectively make the descriptor byte count even by adding a reserved byte. I am still pondering whether to add the byte inside or outside the descriptor ... with outside meaning in the CDB definition and in the Capability definition. I am concerned that the Capability format does not include a Format Type field. Instead, this is in the Credential (which means it is not a part of the Extended CDB Descriptor definition). The presence of a reserved byte in the Capability tempts me to use it (or a nibble of it) for the Capability format. I need to wrestle with this before solidifying my inside/outside position. All the best, .Ralph
Penokie, George wrote:

Ralph,
 
I seem to remember saying in the meeting last week that we decided we
were going to make the MAM thing an even byte count. Or am I thinking of
a different odd byte thing.

Bye for now, 
George Penokie
 
LSI Corporation 
3033 41st St. NW 
Suite 100 
Rochester, MN 55901
 
507-328-9017 
george.penokie@lsi.com


-----Original Message-----
From: owner-t10@t10.org [mailto:owner-t10@t10.org] On Behalf Of Sivan
Tal
Sent: Tuesday, March 18, 2008 1:47 PM
To: Ralph Weber
Cc: owner-t10@t10.org; 't10@t10.org'
Subject: Re: Comments on: CbCS 'correction' proposals

* From the T10 Reflector (t10@t10.org), posted by:
* Sivan Tal <SIVANT@il.ibm.com>
*
Ralph,

That makes the RECEIVE CREDENTIAL CDB a potentially odd length CDB.

If you append a single reserved byte at the end of 'Table x6 - CbCS
logical unit and volume credential request descriptor' (in 08-145r0) it
makes the CDB even length, and also aligns the Capability descriptor to
be even length.
This is just one way to deal with it, assuming odd length CDB is not
welcome.

- Sivan.



 

             Ralph Weber

             <roweber@IEEE.org

             >
To 
             Sent by:                  "'t10@t10.org'" <t10@t10.org>

             owner-t10@t10.org
cc 
 

 
Subject 
             03/10/2008 06:49          Re: Comments on: CbCS
'correction'  
             AM                        proposals

 

 

 

 

 

 





* From the T10 Reflector (t10@t10.org), posted by:
* Ralph Weber <roweber@ieee.org>
*
Sivan Tal wrote:

I confirm the new revision addresses those comments, thanks.

A couple of typo/editorial:
1) In 6.19.1.2 and in 6.19.1.3 where you say "as show in" it should be

"as

shown in"

These corrections appear in:

http://www.t10.org/ftp/t10/document.08/08-128r2.pdf

2) Table x3 - the last byte shoud be at offset 55 instead of 56 (we 
start counting at 0).

The problem here is that the specified MAM attribute (0401h MEDIUM
SERIAL NUMBER) has a length of 37 bytes, specifically 32 bytes of serial
number plus 5 bytes of header.

N.B. This means a 36-byte CbCS designation descriptor field is not large
enough to hold it. (I just realized this myself, and have not yet
decided how to fix it.)

All the best,

.Ralph

*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo@t10.org

*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo@t10.org