RE>RE- SCSI-3 Reseved Field

Charles Monia, SHR3-2/W3, 237-6757 29-Apr-1994 1453 monia at starch.enet.dec.com
Fri Apr 29 11:54:18 PDT 1994


Jim McGrath wrote:
============================================
I do not see how the suggestions address Bob's original point - the sender 
should be instructed not to set reserved bits, but the target (or logical unit - 
in my devices they are one in the same) should not be required to check them.  
Checking for reserved bits is a time consuming process (or gate consuming if 
done in hardware) - I agree with Bob that there is little reason to perform this 
cheack in the real world.

On parameter data, I do not think SCSI-2 requires any checking of parameter data 
today.  I would be opposed to an extension that would break de fecto SCSI-3 
software (i.e. exiting SCSI-2 software).
===========================================
I thought Bob's issue pertained to checking FCP data structures that would not 
be visible to the logical unit. The purpose of the proposal was to define the 
data structures for which testing was required in a way that clearly excludes
these.


With regard to SCSI-2 requirements, the following is from rev 10h , pp 72 of
the SCSI-2 spec. I've included the major section headings to establish the 
context in which the quoted paragraph appears.



"6. SCSI Commands and Status
.....
6.1 Command Implementation Requirements
...
6.1.1 Reserved

Reserved bits fields bytes and code values are set aside for future 
standardization. Their use and interpretation may be specified by future 
extensions to this standard. A reserved bit field or byte shall be set to zero 
or in accrodance with a future extension to this standard. A target that 
receives a reserved bit, field or byte that is not zero or receives a reserved 
code value shall terminate the command with a CHECK CONDITION status. and the 
sense key shall be set to ILLEGAL REQUEST. It shall also be acceptable for a 
target to interpret a bit field, byte or code value in accordance with a future 
extension to this standard."


As I read the above, the enforcement requirements seem to apply both to the 
command descriptor block and parameter data. At least the proposed SCSI-3 
wording was based on that assumption. At any rate, I believe some direction from 
the committee is required here.

BTW: SAM goes out of its way to distinguish between "target" and "logical unit".
In a number of places within section 6 of the SCSI-2 spec, , the word "target" 
is incorrectly used  in place of "logical unit". I had assumed the use of target 
in the quoted paragraph was another one of those instances.

Thanks,
Charles






More information about the T10 mailing list