Draft minutes of SAS Protocol WG -- Mar 10-11, 2003

John Lohmeyer lohmeyer at t10.org
Wed Mar 12 19:29:18 PST 2003


* From the T10 Reflector (t10 at t10.org), posted by:
* John Lohmeyer <lohmeyer at t10.org>
*

Accredited Standards Committee*
InterNational Committee for Information Technology Standards (INCITS)

                                                     Doc. No.: T10/03-122r0
                                                         Date: March 12, 2003
                                                     Reply to: John Lohmeyer
To:       T10 Membership
From:     Ralph Weber Subject:  SAS Protocol WG -- Mar 10-11, 2003
          Dallas, TX

Agenda
1.     Opening Remarks
2.     Approval of Agenda
3.     Attendance and Membership
4.     Old Business
4.1      SAS Working Draft Review (SAS) [Elliott]
4.1.1      4.6.7.5 Expander route index order wording (reflector messages)
4.1.2      6.7.4.8.1 SP15 state SP to SL_IR handoff
4.1.3      6.8 SP_DWS wording (reflector messages and 03-114r0) (Pak Seto, Intel
and George Penokie, IBM)
4.1.4      7.12.5.2 Subtractive port NO DESTINATION vs BAD DESTINATION (Rob
Elliott, HP)
4.1.5      7.14.2 SL transmitter and receiver BREAK should be ignored inside
OPEN address frames (03-120r0) (George Penokie, IBM)
4.1.6      7.16.1 SSP link layer - Livelock paragraph for port layer (note for
Mark Evans, Maxtor)
4.1.7      7.16.7.5 Concern on reflector about DONE Timeout vs. Credit Timeout
(Brian Day, LSI Logic)
4.1.8      7.16.7.10 SSP_TC should be a single state state machine (Rob
Elliott, HP)
4.1.9      7.16.7.11 SSP_TAN should be single state state machine (Rob
Elliott, HP)
4.1.10     8.1 Port layer rewrite (Mark Evans, Maxtor)
4.1.11     10.2.9 SCSI vital product data - device name format waiting on CAP
approval of 02-419r2 (Rob Elliott, HP)
4.1.12     J SAS logo - need to resolve with SCSI Trade Association (John
Lohmeyer, LSI Logic)
4.2      SAS Disconnect-Reconnect Mode Parameters (02-324r0) [Gardner]
4.3      CRC on Frame Header [Roberts]
4.4      Transport Layer Retries (02-487r0) [Jones]
5.     New Business
5.1      SAS Revision 3 T10 letter ballot comment resolution (03-055r4)
[Elliott]
5.2      SAS Requests and Confirmations (03-023r0) [Evans]
5.3      SAS Port Layer Description Modifications (03-024r1) [Evans]
5.4      SAS logo or icon? [Penokie]
5.5      SAS: SP_DWS Wording Changes (03-114r0) [Penokie]
5.6      SAS: Terminating Discovery (03-115r0) [Nixon]
5.7      SAS Invalid Initiator Connection Tag (03-119r0) [Day]
5.8      SAS: Section 7 SL changes (03-120r0) [Penokie]
5.9      SAS: SPC: Media Access Commands During SCSI Power Conditions (03-121r0)
[Penokie]
5.10     SAS: Scrambling Text [Coomes]
5.11     SAS Timeline [Elliott]
5.12     Transition SSP_TF2 to SSP_TF4 wording changes (03-126r0) [Penokie]
5.13     UML Diagrams [Penokie]
6.     Review of Recommendations
7.     Meeting Schedule
8.     Adjournment


Results of Meeting

1.     Opening Remarks

John Lohmeyer called the meeting to order at 9:00 a.m. Monday, March 10, 2003.
He thanked Paul Aloisi and Bob Kando of Texas Instruments for hosting the
meeting. As usual, the people present introduced themselves.


2.     Approval of Agenda

The draft agenda was approved with the following changes:

4.1.1      4.6.7.5 Expander route index order wording (reflector messages)
4.1.2      6.7.4.8.1 SP15 state
4.1.3      6.8 SP_DWS wording (reflector messages and 03-114r0) (Pak Seto, Intel
and George Penokie, IBM)
4.1.4      7.12.5.2 Subtractive port NO DESTINATION vs BAD DESTINATION (Rob
Elliott, HP)
4.1.5      7.14.2 SL transmitter and receiver BREAK should be ignored inside
OPEN address frames (03-120r0) (George Penokie, IBM)
4.1.6      7.16.1 SSP link layer - Livelock paragraph for port layer (note for
Mark Evans, Maxtor)
4.1.7      7.16.7.5 Concern on reflector about DONE Timeout vs. Credit Timeout
(Brian Day, LSI Logic)
4.1.8      7.16.7.10 SSP_TC should be a single state state machine (Rob
Elliott, HP)
4.1.9      7.16.7.11 SSP_TAN should be single state state machine (Rob
Elliott, HP)
4.1.10     8.1 Port layer rewrite (Mark Evans, Maxtor)
4.1.11     10.2.9 SCSI vital product data - device name format waiting on CAP
approval of 02-419r2 (Rob Elliott, HP)
4.1.12     J SAS logo - need to resolve with SCSI Trade Association (John
Lohmeyer, LSI Logic)
5.10     SAS: Scrambling Text [Coomes]
5.11     SAS Timeline [Elliott]

The following items was added or revised during the course of the meeting:

5.12     Transition SSP_TF2 to SSP_TF4 wording changes (03-126r0) [Penokie]
5.13     UML Diagrams [Penokie]


3.     Attendance and Membership

Attendance at working group meetings does not count toward minimum attendance
requirements for T10 membership. Working group meetings are open to any person
or organization directly and materially affected by T10's scope of work. The
following people attended the meeting:

         Name          S        Organization         Electronic Mail Address
---------------------- -- ------------------------- -------------------------
Mr. Ron Roberts        P  Adaptec, Inc.             rkroberts at aol.com
Mr. Robert H. Nixon    P  Emulex                    bob.nixon at emulex.com
Mr. Ralph O. Weber     P  ENDL Texas                roweber at acm.org
Mr. Mike Fitzpatrick   P  Fujitsu                   mfitzpatrick at fcpa.
                                                    fujitsu.com
Mr. Rob Elliott        P  Hewlett Packard Co.       Elliott at hp.com
Mr. Werner Glinka      AV Hitachi America           werner.glinka at attglobal.
                                                    net
Mr. Dan Colegrove      AV Hitachi Global Storage    dcolegro at us.ibm.com
                          Tech.
Mr. George O. Penokie  P  IBM / Tivoli Systems      gop at us.ibm.com
Mr. Cris Simpson       P  Intel Corp.               cris.simpson at intel.com
Mr. Robert Sheffield   A  Intel Corp.               robert.l.sheffield at intel.
                                                    com
Mr. John Lohmeyer      P  LSI Logic Corp.           lohmeyer at t10.org
Mr. Brian Day          V  LSI Logic Corp.           brian.day at lsil.com
Mr. Tim Hoglund        V  LSI Logic Corp.           tim.hoglund at lsil.com
Mr. Mark Evans         P  Maxtor Corp.              mark_evans at maxtor.com
Mr. Martin Czekalski   V  Maxtor Corp.              marty_czekalski at maxtor.
                                                    com
Mr. Nathan Obr         A# Microsoft Corp.           natobr at microsoft.com
Mr. Edward A. Gardner  AV Ophidian Designs          eag at ophidian.com
Mr. Curtis Stevens     V  Pacific Digital Corp.     cstevens at pacificdigital.
                                                    com
Mr. Bill Galloway      P  Pivot3, Inc.              billg at pivot3.com
Mr. Larrie Carr        V  PMC-Sierra, Inc.
Mr. Arie Krantz        V  QLogic Corp.              l_krantz at qlc.com
Mr. Jim Jones          A  Quantum Corp.             jim.jones at quantum.com
Mr. Gerald Houlder     P  Seagate Technology        gerry.houlder at seagate.com
Mr. Jim Coomes         V  Seagate Technology        jim.coomes at seagate.com
Mr. John Worden        V  Seagate Technology        john.worden at seagate.com
Mr. Vit Novak          P  Sun Microsystems, Inc.    vit.novak at sun.com
Mr. Robert Kando       A. Texas Instruments         Robert.Kando at ti.com
Mr. Phil Shelton       A  UNISYS Corporation        phil.shelton at unisys.com
Mr. Haluk Aytac        V  Velio                     haluk at velio.com
Mr. Kenneth Hirata     P  Vixel Corp.               Ken.Hirata at Vixel.com

30 People Present

Status Key:  P    -  Principal
             A,A# -  Alternate
             AV   -  Advisory Member
             L    -  Liaison
             V    -  Visitor


4.     Old Business

4.1    SAS Working Draft Review (SAS) [Elliott]

4.1.1  4.6.7.5 Expander route index order wording (reflector messages)

After discussing the reflector messages, the group concluded that in the
absence of any specific proposed changes the SAS document will remain as
written.

4.1.2  6.7.4.8.1 SP15 state SP to SL_IR handoff

Mark Evans described a problem where one end of a connection can miss an
IDENTIFY due to clock tolerance and COMSAS detection tolerance. He proposed
adding a 2 ms delay prior to transmitting the IDENTIFY.

Rob Elliott agreed to announce the problem and proposed solutions on the
reflector, and the group agreed to consider a solution for the problem on the
second day of the meeting.

George Penokie moved that IDENTIFY be accepted before the SNTT timer expires.
Bill Galloway seconded the motion. In the absence of any objections, the
motion passed unanimously. The SAS editor was instructed to incorporate the
results of the motion as part of resolving the SAS letter ballot.

4.1.3  6.8 SP_DWS wording (reflector messages and 03-114r0) (Pak Seto, Intel and
George Penokie, IBM)

This topic was covered under item 5.5.

4.1.4  7.12.5.2 Subtractive port NO DESTINATION vs BAD DESTINATION (Rob
Elliott, HP)

The group agreed to change BAD DESTINATION to NO DESTINATION when a request
comes in a subtractive routing port that cannot be forwarded to anywhere
except back to the subtractive routing port.

4.1.5  7.14.2 SL transmitter and receiver BREAK should be ignored inside OPEN
address frames (03-120r0) (George Penokie, IBM)

This topic was discussed under item 5.8.

4.1.6  7.16.1 SSP link layer - Livelock paragraph for port layer (note for
Mark Evans, Maxtor)

Mark agreed to include this issue in his proposal (03-024r1), which was
discussed later in the meeting (see 5.3).

4.1.7  7.16.7.5 Concern on reflector about DONE Timeout vs. Credit Timeout
(Brian Day, LSI Logic)

Brian Day described a race condition that occurs when one end of a connection
sends DONE but doesn't give the other end credits. George Penokie stated that
this race condition is no different from several other conditions and
therefore no changes are required. The group eventually agreed to make a
change such that the DONE timer is not started unless the device knows that
credit is available at the other device or a CREDIT BLOCKED has been sent to
the other device.

George agreed to incorporate the changes that are needed in the next revision
of 03-114r0, which was discussed later in the meeting.

4.1.8  7.16.7.10 SSP_TC should be a single-state state machine (Rob Elliott,
HP)

George Penokie agreed to incorporate the resolution for this issue in 03-120r0.

4.1.9  7.16.7.11 SSP_TAN should be single-state state machine (Rob Elliott,
HP)

George Penokie agreed to incorporate the resolution for this issue in 03-120r0,
which was discussed later in the meeting.

4.1.10 8.1 Port layer rewrite (Mark Evans, Maxtor)

This topic was covered under item 5.3.

4.1.11 10.2.9 SCSI vital product data - device name format waiting on CAP
approval of 02-419r2 (Rob Elliott, HP)

Rob Elliott noted that if CAP approves 02-419r2, SAS would be changed
accordingly.

Rob asked that discussion of this topic be deleted from future agendas.

4.1.12 J SAS logo - need to resolve with SCSI Trade Association (John
Lohmeyer, LSI Logic)

This topic was discussed under item 5.4.

4.2    SAS Disconnect-Reconnect Mode Parameters (02-324r0) [Gardner]

Ed Gardner asked that discussion of this topic be removed from future agendas.

4.3    CRC on Frame Header [Roberts]

Ron Roberts asked that this topic be deferred to SAS-2.

4.4    Transport Layer Retries (02-487r0) [Jones]

Jim Jones proposed that this topic be deferred to SAS-2, provided that it can
be considered a high-priority item in the SAS-2 effort. George Penokie noted
that it is a high-priority item for his company.


5.     New Business

5.1    SAS Revision 3 T10 letter ballot comment resolution (03-055r4) [Elliott]

Rob Elliott noted that the following documents are relevant:

03-055r4.pdf - comments sorted by author
03-055r4.zip - contains a zip file with sas-r03 with comments (PDF database)
sas-r03e.pdf - latest version of SAS
sas-r03e.zip - Frame source for SAS rev 3e

Rob Elliott led the group in a discussion of the letter ballot comments on
SAS. Rob summarized the status of the comment resolution (as of the beginning
of the meeting) as follows:

Total comments: 2727
 Accept - Done: 2046
 Accept - Last: 5
 Accept - Todo: 3
 Reject (only): 569
     Port todo: 90
         Refer: 11
   Todo (only): 1
          Left: 2
  Total worked: 2725 (99.93%)

Rob reviewed several comment resolutions where a need for general agreement
was indicated.

George Penokie moved that all references to RBC and the RBC amendment be
removed from SAS. Cris Simpson seconded the motion. The motion passed on a
vote of 6:3:4.

The group reviewed a preliminary proposal (no number assigned) from Bob
Sheffield for new text describing the use of the STP Affiliation feature and
requested several modifications.

5.2    SAS Requests and Confirmations (03-023r0) [Evans]

Mark Evans ask that discussion of this topic be deferred to the next meeting.

5.3    SAS Port Layer Description Modifications (03-024r1) [Evans]

Mark Evans presented a proposed changed to the SAS port layer description
(03-024r1). The group advised Mark regarding the resolution of three issues
listed in the introduction to the proposal. Then the group reviewed the
proposal in detail, recommending several wording changes. Mark agreed to
prepare a new revision of the proposal.

5.4    SAS logo or icon? [Penokie]

John Lohmeyer reported that George sent a e-mail to STA detailing the SAS
issues with using the trademarked STA logo. John Lohmeyer and Marty Czekalski
discussed the conversion of the STA logo to a SAS icon. Rob Elliott noted that
we need a resolution this week. However, a final resolution for the issue was
not available before the meeting adjourned.

Chair's note:  Shortly after the meeting, the STA Board agreed to allow usage
of a SAS icon, which is essentially the same as the SAS logo minus all
lowercase test. See the STA Liaison Report in the T10 minutes (03-124).

5.5    SAS: SP_DWS Wording Changes (03-114r0) [Penokie]

George Penokie presented proposed changes to the dword synchronization wording
(03-114r2).

George Penokie moved that 03-114r2 be incorporated into SAS as part of the
letter ballot resolution. Bill Galloway seconded the motion. In the absence of
any objections, the motion passed unanimously.

5.6    SAS: Terminating Discovery (03-115r0) [Nixon]

Bob Nixon presented a proposal to allow a discovery operation to be terminated
in the event that the discovery data appears to be incorrect (03-115r0). The
group requested minor changes and Bob agreed to prepare a revised proposal.
The group recommended that 03-115r1 (r0 as revised) be accepted as resolving a
letter ballot comment.

5.7    SAS Invalid Initiator Connection Tag (03-119r0) [Day]

Brian Day presented a proposal concerning the handling of invalid initiator
connection tags (03-119r0). The group instructed the editor to incorporate
03-119r0 in SAS as part of the letter ballot resolution.

5.8    SAS: Section 7 SL changes (03-120r0) [Penokie]

George Penokie presented revised SL state diagram changes (03-120r0). The
proposal included an explicit statement that BREAKs be ignored during a
receipt of an address frame. Jim Coomes argued that the proposal requires
transmitters to be more complicated.

The group requested several changes and George agreed to prepare a revised
proposal.

George Penokie moved that 03-120r1 (r0 as revised) be accepted for inclusion
in SAS. Bill Galloway seconded the motion. The motion passed on a vote of
8:1:1.

Later in the meeting, George Penokie moved that 03-120r2 be accepted for
inclusion in SAS, replacing r1, which was accepted previously. Bill Galloway
seconded the motion. The motion passed on a vote of 13:0:0.

5.9    SAS: SPC: Media Access Commands During SCSI Power Conditions (03-121r0)
[Penokie]

George Penokie presented a proposal to clarify that media access commands are
not functional in the Active_Wait power condition (03-121r0). Mark Evans
agreed to cut the material from George's proposal and paste it in his proposal
02-464r1. Rob Elliott noted that the SAS aspects of the proposal have already
been incorporated in rev 03e.

5.10   SAS: Scrambling Text [Coomes]

Jim Coomes presented a proposal to revise and correct the SAS text describing
the scrambling algorithm (a draft of 03-129r0). The group discussed the
proposal and requested minor changes. The group instructed the editor to
incorporate 03-129r0 in SAS to resolve a letter ballot comment.

5.11   SAS Timeline [Elliott]

Rob Elliott presented an e-mail describing the anticipated course of events in
the completion of development of SAS.

Concerns were raised about allowing time for rebuttals concerning comments
that have been rejected. Rob acknowledged that time must be left for rebutting
rejected comments.

The process for getting SAS published as a ISO standard was also discussed.

5.12   Transition SSP_TF2 to SSP_TF4 wording changes (03-126r0) [Penokie]

George Penokie presented a proposal to clear up the wording for the TF2 to TF4
transition (02-126r0), specifically when the transition takes place. The group
requested changes and George agreed to prepare a new revision.

George Penokie moved that 03-126r1 (r0 as revised) be recommended for
inclusion in the SAS letter ballot resolution. Mark Evans seconded the motion.
In the absence of any objections, the motion passed unanimously.

5.13   UML Diagrams [Penokie]

George Penokie requested that this topic be removed from future agendas.


6.     Review of Recommendations

The following documents were accepted for inclusion in SAS during this
meeting:

03-114r2, SP_DWS Wording Changes
03-115r1, Terminating Discovery
03-119r0, SAS Invalid Initiator Connection Tag
03-120r2, Section 7 SL changes
03-129r0, Scrambling Text
03-126r1, Transition SSP_TF2 to SSP_TF4 wording changes

The following recommendations were made during this meeting:

that IDENTIFY be accepted before the SNTT timer expires
that BAD DESTINATION be chanted to NO DESTINATION when a request comes in a
subtractive routing port that cannot be forwarded to anywhere except back to
the subtractive routing port
that all references to RBC and the RBC amendment be removed from SAS

The following documents were accepted for inclusion in SAS during the February
24-25, 2003 meeting:

03-034r4, SAS Expander internal devices
03-089r0, SAS PHY CHANGE COUNT
03-091r1, SAS Target Port Transfer Tag
03-109r1, SAS STP affiliations
03-112r0, SAS SL_IR state receiving IDENTIFY frames

The following motions were passed during the February 24-25, 2003 meeting:

that in section 7.8.2, `the application client within an initiator device
shall perform a discovery process' be changed to `the application client
within an initiator device should perform a discovery process' (i.e., `shall'
becomes `should').
that a new confirmation, Done Received, be added to the port layer and that
the same confirmation be added from the port layer to the transport layer and
|from the transport layer to the application layer.
that the statement in 7.12.4.1 that says "This state shall not respond to
incoming BREAKs, OPEN_REJECTs, and OPEN_ACCEPTs until after the OPEN address
frame has been transmitted." be changed to "This state shall not respond to
incoming OPEN_REJECTs or OPEN_ACCEPTs until after the OPEN address frame has
been transmitted."
that the SATA references be changed to the appropriate document references in
T13.
that the T10 Vice Chair contact STA and request usage of the STA logo for SAS
with no restrictions.

The following recommendations were made during the January 13-14, 2003
meeting:

Accept the comment in 7.1.4.4 that increases the number of broadcast
primitives from 4 to 8. [14:0:4]
Make one of the new broadcast-reserved primitives a broadcast-reserved-changed
primitive. [16:2:3]
Accept the comment regarding the 20 dword rule in 7.1.6.3. [11:0:6]
Remove the 64 entry limit of the number of indices in fanout
expanders.[10:1:6]
Remove near end analog loop back. [8:2:10]
Remove note 12. [5:1:10]
SAS Definitions (03-022r1) [Evans] {r0 as revised, unanimous}
SAS SMP error codes (03-058r0) [Fairchild] {unanimous}


7.     Meeting Schedule

SAS Protocol Working Group meetings are scheduled for:

Monday-Tuesday, May 5-6, 2003 commencing at 9 a.m. in Nashua, NH at the Crowne
Plaza Nashua (603-886-1200), hosted by Hitachi Cable Manchester.

Additional teleconference calls may be announced on the T10 reflector as
needed.


8.     Adjournment

The meeting was adjourned at 6:55 p.m. on Tuesday March 11, 2003.

 
*
* 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