Approved SSA-TL2 comment resolution
scheible at vnet.ibm.com
scheible at vnet.ibm.com
Wed Feb 26 22:07:35 PST 1997
* From the SCSI Reflector (scsi at symbios.com), posted by:
* scheible at VNET.IBM.COM
Accredited Standards Committee
NCITS, National Committee for Information Technology Standards
Date: February 26, 1997
Ref. Doc.: SSA-TL2 rev 4
Reply to: John Scheible
To: T10 Membership
From: John Scheible
Subject: Letter ballot comment resolution on SSA-TL2 rev 4
This proposal documents the responses to the T10 letter ballot comments for SSA-TL2 rev 4 as approved by
T10.1 during its February 26, 1997 plenary meeting.
1 Organization(s) did not vote
46 Total voting organizations
3 Ballot(s) included comments
This 2/3rds majority ballot passed.
Comments attached to YesC ballot from John P. Scheible of IBM Corp.:
RESPONSE: Accept all as written with modifications to IBM-19) and IBM-20) as shown.
I consider all comments editorial (E) except for 9, 18, 19,and 20.
IBM-1) Global: Change "(( " to "(" and ")|" to" ))". Also "(s).)" to "(s)|."
IBM-2) Change "<any letter>( "to "<any letter> (" to add a space.
IBM-3) Table of tables: Change heading from "Table" to "Tables" (plural).
IBM-4) Table of tables: Table numbers for 41-60 should be bold.
IBM-5) Introduction: Change "Clause 4 contains" to "Clause 4 defines" for consistency.
IBM-6) 3.2: Add abbreviation OUI (Organizationally Unique Identifier).
IBM-7) 4, last paragraph: Change "Figures and tables (highest) take precedence over text (lowest)." to "In
case of conflict, figures take precedence over tables and both figures and tables take precedence
IBM-8) end of 6.2: Change "(see 0)" to "(see 9.5)". Also applies to end of 7.2.7 and 7.2.8.
IBM-9) 9.3, paragraph after psuedocode: Change "Channels 0- 127 are addressed one byte. Channels 129-
16 383 ." to "Channels 0 - 127 are addressed one byte. Channels 128 - 16 383." (128 was left
IBM-10) 10.2.3, fifth para, end first sentence: change ".)." to ")."
IBM-11) Table 20, first column: Change small caps to all caps (values, not field names).
IBM-12) 10.3, add semicolon to d) first list, add period to e) of second list.
IBM-13) 10.4.1, c): Change "Ready State" to "Ready state" (lower case "S").
IBM-14) 10.6: Change TL2 to SSA-TL2 (global).
IBM-15) 11.1.3, element h): Change "mod" to "modulo" (two places) to match I).
IBM-16) Table 29, sixth row from end: Should not be bold.
IBM-17) Table 32, bottom of byte 12: should be solid line, not dotted.
IBM-18) Table 37, version 00h: Change "SSA-TL1 implementations before standardization" to
"Implementations prior to standardization." Since version 00h is not SSA-TL1.
IBM-19) Table 37, version 04h: Change "SSA-IA/96PH" to either "reserved" or "SSA-IA/97PH" depending on
whether the SSA-IA plans to document this version. Confusion will occur since someone may try to
obtain the non-existent document.
RESPONSE: Change SSA-IA96 (04) to reserved.
IBM-20) 12.2.7, SSA-TL paragraph: Correct name, add "if possible" and add examples, by changing:
"The SSA-TL field defined in Table 37 identifies the version of SSA-TL being used by the sender. If
multiple levels of SSA-TL are supported, then highest value shall be reported that is equal to or less
than the SSA-TL field in the associated QUERY NODE SMS."
"The SSA-TL VERSION field defined in Table 37 identifies the version of SSA-TL being used by the
sender. If multiple levels of SSA-TL are supported and any are numerically less that the SSA-TL
VERSION field of the QUERY NODE SMS, then the highest value shall be reported in the QUERY
NODE REPLY SMS that is equal to or less than the SSA-TL VERSION field in the associated QUERY
NODE SMS. The following are examples:
a) An SSA-TL1/SSA-TL2 Configutor sends a QUERY NODE SMS indicating an SSA-TL version of
SSA-TL2. An SSA-TL1 only node responds with a QUERY NODE REPLY SMS indicating SSA-TL1.
The Configutor now knows to use SSA-TL1 to communicate with the node.
b) An SSA-TL1/SSA-TL2 Configutor sends a QUERY NODE SMS indicating an SSA-TL version of
SSA-TL2. An SSA-TL1/SSA-TL2 or SSA-TL2 only node responds with a QUERY NODE REPLY SMS
indicating SSA-TL2. The Configutor now knows to use SSA-TL2 to communicate with the node.
c) An SSA-TL1 only Configutor sends a QUERY NODE SMS indicating an SSA-TL version of SSA-
TL1. An SSA-TL1/SSA-TL2 node responds with a QUERY NODE REPLY SMS indicating SSA-TL1.
The Responder now knows to use SSA-TL1 to communicate with the node.
d) An SSA-TL1 only Configutor sends a QUERY NODE SMS indicating an SSA-TL version of SSA-
TL1. An SSA-TL2 only node responds with a QUERY NODE REPLY SMS indicating SSA-TL2. The
Configutor now knows it cannot communicate with the node, and does not register with it.
RESPONSE: Make two minor changes as shown above.
IBM-21) 12.2.7, LONG bit paragraph: Remove the extraneous " (".
IBM-22) Table 53, blank rows between byte 3 and 4 should be removed.
Comments attached to No ballot from Edward A. Gardner of Ophidian Designs:
OPH-1) I am uncomfortable with VLSI's negative vote remaining unresolved. I would like to see X3T10.1
respond to that vote and comment before forwarding this for review. If X3T10.1 has already
responded to VLSI's vote, please refer me to the relevant document and I will amend my vote.
RESPONSE: Add and implementer's note after 8, paragraph 6 (Brad Kitson (VLSI) would change his vote to
Yes based on this response).
"Note n: Hardware implementers should allow at least 16 bytes (excluding CRC) of dedicated buffer space for
future extensions to the Link Reset and Extended Link Reset frames".
Comments attached to No ballot from Gene Milligan of Seagate Technology:
RESOLUTION: (see each comment)
SEA-1) The ballot does not show what the response was to the X3T10.1 "NO" ballot.
RESPONSE: See OPH-1 above.
SEA-2) The ballot does not indicate why Rev 4 is being forwarded when X3T10.1 ballotted to forward Rev 3.
RESPONSE: T10.1 voted to forward Rev 3 as modified, and instructed the editor to create rev 4 with
SEA-3) The draft should have an editorial review checking at least for the appropriate use of key words (e.g.
this drafts contains at least three musts and fifteen cans.
RESPONSE: Replace "must" with "shall" in 9.4.1 para 2, 13.1.1 second bullets a) and e). Replace
"can" with "may" in 10.1.1 para 2, 10.1.7 para 1, 10.1.8 para 1, Table 19 note, 10.2.4 para 6, 10.4.2
bullets d) f) h), 12.2.12 last para, 13.1.1 para 1, B.4.2 para 1, B.4.3 para 1 and 3. Replace "can"
with "shall" in 12.2.16 bullet a). Replace "it can" with "the REGION IDENTIFIER field may" in 12.2.17
para 4. Replace "could" with "may" in 13.1.1 para 1 and second bullet f), 10.2.4 para 2, and 13.3.2
SEA-4) If someone impliments SSA-TL2 and an appropriate selection of the other SSA drafts will they be
interoperable with the defacto installed base of SSA subsystems? (This is a question and not the
basis for the NO - see comments 1-3.)
RESPONSE: Yes, the different levels of PH1 / PH2 , TL1 / TL2, and S2P / S3P may interoperate and
co-exist on the same loop simultaneously. The physical layers may connect to each other and
speed matching is done on a link by link basis. The transport layers identify themselves in the
QUERY NODE and QUERY NODE REPLY SMSs. Likewise the S2P and S3P layers identify
themselves in the QUERY NODE REPLY SMS and whichever protocol layer is needed may be used
between different source and destination pairs.
Web page access
X3T10.1 is planning web sites to disseminate information regarding the implementation of the standard. The
following changes will be made:
1) Add the following to the end of clause 2.2
Further information regarding the implementation of this standard may be available at
http://www.ncits.org/t10 or http://www.ssaia.org. Some tables within this standard may have extensions
defined at these sites.
2) Add the following note to Tables 6, 29, 37, and B.5:
Add a table note: NOTE - This table is accurate as of the time of publication. Further information may be
available at the web sites (see 2.2).
Voice: (512) 823-8208
FAX: (512) 838-3822
Email: Scheible at vnet.ibm.com
* For SCSI Reflector information, send a message with
* 'info scsi' (no quotes) in the message body to majordomo at symbios.com
More information about the T10