98-115r0 - Sundry Minor Enhancements for SPC-2

Gerry_Houlder at notes.seagate.com Gerry_Houlder at notes.seagate.com
Mon Feb 2 13:59:13 PST 1998


* From the T10 (formerly SCSI) Reflector (t10 at symbios.com), posted by:
* Gerry_Houlder at notes.seagate.com
*
This is text from Ralph's original message:

>The following should be added at the end of the third paragraph after
>Table 36 in clause 7.20.1 to embody the committee's position in SPC-2:
>
>   "If a field whose length exceeds one byte is reserved, it shall be
>   treated as a multiple-byte field.  If several consecutive bytes are
>   individually reserved, each shall be treated as a single-byte field."
I think the concept of a multiple byte reserved field is abhorent. A
multiple byte reserved field can only be a multiple byte defined field in
the future, right? Can we really have the foresight to know whether those
reserved bytes will be bits, one byte fields, or multiple byte fields in
the future? The concept of multiple byte reserved fields makes error
reported more complicated, not less complicated. What if one rev. of the
draft has a multiple byte reserved field and another rev. has consecutive
single byte reserved fields? Must we continue to redesign device firmware
every time an editor tries to switch to a more efficient editorial style?

I don't like the statement "The contents of the sense-key specific data for
the ILLEGAL REQUEST sense key is not specified by SCSI-2 when the error
occurs in a reserved field." True, the wording doesn't specifically address
reserved fields but it does address the "command descriptor block or
parameter data that was in error". It is clear about how to report multiple
byte fields and fields of one byte or less. The only thing that is unclear
is is whether a reserved field can be multiple bytes or not. This used to
be very clear because old versions of the standard put the word "reserved"
in each byte of a reserved area and more recent editing style lets one
"reserved" handle multiple bytes in some cases. Don't let editorial style
dictate a technical change!

Change the proposed wording to read as follows:

"If several consecutive bytes are reserved, each byte shall be treated as a
single byte field."


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




More information about the T10 mailing list