Minutes of Fast-xx SG and SPI-2 WG - 11/3/97

John Lohmeyer John.Lohmeyer at symbios.com
Tue Nov 11 12:33:45 PST 1997


* From the T10 (formerly SCSI) Reflector (t10 at symbios.com), posted by:
* John Lohmeyer <John.Lohmeyer at Symbios.com>
*
Accredited Standards Committee*
National Committee for Information Technology Standards (NCITS)

                                                    Doc. No.: T10/97-272r0
                                                    Date: November 11, 1997
                                                    Project:  1142-D
                                                    Ref. Doc.:
                                                    Reply to: John Lohmeyer

 To:      Membership of T10

From:     Ralph Weber, Secretary T10
          Larry Lamers, Vice-chair T10
          John Lohmeyer, Chair T10

Subject:  Minutes of Fast-xx Study Group and SPI-2 Working Group
          November 3, 1997 -- Palm Springs, CA


  Agenda
1.  Opening Remarks
2.  Approval of Agenda
3.  Attendance and Membership
4.  Fast-xx Study Group {Monday Morning}
   4.1 Low-Attenuation Cables (97-213, 97-228) [Rogers, Bellino]
   4.2 Cable Attenuation & Margin Budget Analysis at 80 MHz (97-269) [Gintz]
   4.3 Bias Reduction Proposal (97-214) [Bridgewater]
   4.4 Dual Clocking Proposal (97-208) [Bastiani]
   4.5 Project Proposal for SPI-3 (97-262) [Lamers]
   4.6 Fast 80 MHz SCSI Testing (97-276) [Bastiani]
5.  SPI-2 Working Group {Monday Afternoon}
   5.1 SPI-2 specifications on interconnect (97-235) [Ham]
   5.2 0.5 pF balance capacitance for LVD SCSI (Table 29 SPI-2R13) (97-236)
   [Aloisi/Kapraun]
   5.3 Capacitance Measurement at 2 Volts [Wallace]
   5.4 Universal Backplane Annex (97-121r0) [Wallace]
   5.5 Latching and Counting (97-216) [Guss]
   5.6 SPI-2 Revision 16 Review [Penokie]
   5.7 SPI-2 Timing Tolerance (97-268) [Asami]
   5.8 Load Compensation (97-281) [Novak]
   5.9 Proposal to remove clause 10.5 from SPI-2 Rev 16 [Lohmeyer]
6.  Meeting Schedule
7.  Adjournment



                            Results of Meeting

1. Opening Remarks

John Lohmeyer, the T10 Chair, called the meeting to order at 9:00 a.m., Monday
November 3, 1997.  He thanked Norm Harris of Adaptec for hosting the meeting.

As is customary, the people attending introduced themselves and a copy of the
attendance list was circulated.

2. Approval of Agenda

The agenda was approved with the addition of Greg Kapraun to item 5.2:

5.2 0.5 pF balance capacitance for LVD SCSI (Table 29 SPI-2R13) (97-236)
[Aloisi/Kapraun]

The following agenda items were added during the meeting:

4.6 Fast 80 MHz SCSI Testing (97-276) [Bastiani]

3. Attendance and Membership

Attendance at working group meetings does not count toward minimum attendance
requirements for T10 member-ship.  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. Lawrence J. Lamers P  Adaptec, Inc.             ljlamers at ix.netcom.com
Mr. Vincent Bastiani   A# Adaptec, Inc.             bastiani at corp.adaptec.com
Mr. Tak Asami          V  Adaptec, Inc.             asami at itc.adaptec.com
Mr. Wally Bridgewater  V  Adaptec, Inc.             wally at eng.adaptec.com
Mr. Bill Gintz         V  Adaptec, Inc.             bgintz at corp.adaptec.com
Mr. Tom Schneider      V  Adaptec, Inc.             schneid at itc.adaptec.com
Mr. Charles Brill      P  AMP, Inc.                 cebrill at amp.com
Mr. John Laskowsky     V  Berg Electronics          laskowj at bergelect.com
Mr. Domininc K.V.      V  Berg Electronics          zarreldv at bergelect.com
Zarrella
Mr. Richard Wagner     P  Cable Design Technologies rwagner at montrose-cdt.com
Mr. Edward Haske       O  CMD Technology            haske at cmd.com
Mr. Charles Tashbook   P  Dallas Semiconductor      charles.tashbook at dalsemi.
                                                    com
Mr. Greg McSorley      P  Data General / Clariion   greg_mcsorley at dgc.ceo.dg.
                                                    com
Dr. William Ham        A# Digital Equipment Corp.   bill.ham at digital.com
Mr. George Penokie     P  IBM Corp.                 gop at us.ibm.com
Mr. Tim Bradshaw       P  Iomega Corp.              bradshat at iomega.com
Mr. Dean Wallace       P  Linfinity Micro           75671.3443 at compuserve.com
Mr. Louis Grantham     A  Linfinity Micro           lgrantham at linfinity.com
Mr. Alan Littlewood    P  LSI Logic Corp.           alanl at lsil.com
Mr. Makesh             V  Lucent Technologies       makesh at lucent.com
Kothandaraman
Dr. Bernard Morris     V  Lucent Technologies       blmorris at lucent.com
Mr. Brian Davis        V  Mylex Corp.               briand at mylex.com
Mr. Skip Jones         P  QLogic Corp.              sk_jones at qlc.com
Mr. Ting Li Chan       A  QLogic Corp.              t_chan at qlc.com
Mr. Patrick McGarrah   P  Quantum Corp.             pmcgarra at tdh.qntm.com
Mr. Mark Evans         A  Quantum Corp.             mevans at qntm.com
Mr. Bruce Leshay       V  Quantum Corp.             bleshay at tdh.qntm.com
Mr. Richard Uber       V  Quantum Corp.             duber at tdh.qntm.com
Mr. John A. Fobel      O  Rancho Technology, Inc.   johnf at rancho.com
Mr. Gene Milligan      P  Seagate Technology        Gene_Milligan at notes.
                                                    seagate.com
Mr. Gerald Houlder     A  Seagate Technology        Gerry_Houlder at notes.
                                                    seagate.com
Mr. Daniel (Dan) F.    O  Seagate Technology        daniel_f_smith at notes.
Smith                                               seagate.com
Mr. Vit Novak          A  Sun Microsystems, Inc.    vit.novak at sun.com
Mr. John Lohmeyer      P  Symbios Logic Inc.        john.lohmeyer at symbios.com
Mr. Ralph O. Weber     A  Symbios Logic Inc.        roweber at acm.org
Mr. Paul D. Aloisi     P  Unitrode Corporation      aloisi at unitrode.com
Mr. Gregory Kapraun    V  Western Digital Corp.     Gregory.D.Kapraun at wdc.com

37 People Present

Status Key:  P    -  Principal
             A,A# -  Alternate
             O    -  Observer
             L    -  Liaison
             V    -  Visitor

For the benefit of those present at this meeting who will not be attending the
T10 plenary meeeting, John described the pilot membership program developed
for T10 by the T10 leadership and the NCITS secretariat.  Details of John's
presentation can be found in the minutes of the T10 plenary meeting (97-279).

4. Fast-xx Study Group {Monday Morning}

4.1 Low-Attenuation Cables (97-213, 97-228) [Rogers, Bellino]

Neither Kirk Rogers nor Bob Bellino were present.  John Lohmeyer mentioned a
presentation from Bob Bellino on Lower Attenuation Cables (97-228r0) that was
in the July mailing, but not discussed at any meeting.  He expressed the
belief that the data was beneficial to those concerned with cable
characteristics.

4.2 Cable Attenuation & Margin Budget Analysis at 80 MHz (97-269) [Gintz]

Bill Gintz presented a revised timing budget analysis (97-269, revised from
97-207).  Bill reported his conclusions first.  He described the Fast-80
amplitude and jitter budgets results to be shown as covering: analytical
projections, worst case margins 2.5 ns/6.25 ns, multi-drop effect, reflections
and ISI, and a hot plug simulation.  He described the key features of his
empirical results as:

   Reflections dominate jitter,
   Multi-drop dominates amplitude,
   Skew and ISI are unwieldy, and
   Hot plug is problematical.

For about an hour, the group discussed Bill's analysis: results, methodology,
and assumptions.  Bill's summary was as follows:

   1.  Dual-edge clock still seems to be the best solution - The group
   generally agreed with this conclusion.
   2.  Cable skew and differential capacitance dominate offsets - There was no
   argument on this point either.
   3.  Cable induced ISI still enigmatic - more data is being taken - Bill
   Gintz and Bill Ham agreed that testing is important and that the data
   gathered thus far are insufficient.  Bill Gintz expressed the hope that one
   of the cable companies will do the needed testing.  Bill Ham expressed a
   preference for looking at silicon solutions while keeping the current cable
   definition.
   4.  The next step is ISI dual-edge test silicon to verify margin analysis
   5.  Jitter margin with nominal driver and skew control expected to be 2-6
   ns
   6.  Hot-plugging is problematical - more test data is needed

To this list, Bill Ham proposed the addition of a seventh summary point
involving duty cycle distortion (DCD) issues.  Bill Gintz argued that the DCD
issues are not as serious as Bill Ham was suggesting.  After a brief
discussion, it appeared that Bill Gintz accepted DCD issues as potentially
important.

Other topics discussed included:

   1.  Skew compensation or CRC for data signals. Bill Gintz and Vince
   Bastiani extolled the virtues of HIPPI electronic skew compensation.  Then,
   Bill Gintz noted that CRC could solve the problems too.  Bill Ham noted
   that skew compensation fixes the problem, but CRC only detects it.  The
   consensus appeared to be that electronic skew compensation would be
   preferable, but it will most likely not be needed until we attempt much
   higher data rates than Fast-80.
   2.  Case 4 hot plugging.  Vince asked what had happened to the proposal to
   provided staged contact resistance to a hot plugging connector.  Bill Ham
   said AMP was still investigating the design.  The group also discussed
   whether LVD is robust enough to do case 4 hot plugging without special
   connectors; whether today's parity and REQ/ACK counting were sufficient, or
   whether CRC should be employed.  (See 5.9 for more on this topic.)

4.3 Bias Reduction Proposal (97-214) [Bridgewater]

Wally Bridgewater asked that discussion of this item be deferred to the next
meeting.

4.4 Dual Clocking Proposal (97-208) [Bastiani]

Vince Bastiani presented his discussion of dual-edge clocking as a review of
time interval testing for 80 MHz (see 4.6).  He asked that this topic or one
very similar to it be included on the next agenda.

4.5 Project Proposal for SPI-3 (97-262) [Lamers]

The first topic for argument was the name and the group agreed to change
"SCSI-4 Parallel Interconnect - 3" to "SCSI Parallel Interconnect - 3".  The
group continued wordsmithing the proposal at length.

Dan Smith proposed creation of a group to create and review the electrical
design data needed to reach the kind of transfer rates proposed by the group.
The needs for and procedures of such a group were discussed.

In the absence of any objections, the group unanimously recommended that the
T10 plenary approve the revised SPI-3 project proposal (97-262r2).

4.6 Fast 80 MHz SCSI Testing (97-276) [Bastiani]

Vince Bastiani presented measured data analyzing setup and hold timings (see
97-276) based on full 80 MHz signals (as opposed to dual edge clocking).  A
much-discussed component of the data measurement equipment was a HP 3310A
Timing Interval Analyzer.  Bill Ham noted a very large jitter in the base-line
case.  After some discussion, Gene Milligan led the group to the conclusion
that the jitter could have been induced by the pseudo-random data that was
being driven on the other data lines.

After a lengthy review of the data and discussion, Vince concluded that there
would be sufficient margin to detect bits if dual-edge clocking is used.

5. SPI-2 Working Group {Monday Afternoon}

5.1 SPI-2 specifications on interconnect (97-235) [Ham]

Bill's presentation was deferred to item 5.6, since George Penokie has already
incorporated the changes in SPI-2 revision 16.

5.2 0.5 pF balance capacitance for LVD SCSI (Table 29 SPI-2R13) (97-236)
[Aloisi/Kapraun]

After a brief introduction from Paul Aloisi, Greg Kapraun presented new data
(97-277) with capacitance imbal-ances introduced in each of the thirteen
drives in the test configuration.  The group was reminded that Greg's previous
data (where only one drive was modified to have a capacitance imbalance) could
be found in 97-236.

Greg's conclusion was that the 0.5 pF requirement should not be relaxed.
However, questions were raised about the degree to which the data presented
supported the conclusion.  Greg said that he will conduct more detailed tests,
including measuring the actual capacitance after the altering hardware has
been installed.

Wally Bridgewater asked at what frequency the capacitance tests were to be
performed.  Paul Aloisi said that the testing method was specified in one of
the source documents for SPI-2.  After reading SPI-2, Wally suggested that a
change in the frequency might be in order, especially for Fast-xx.

5.3 Capacitance Measurement at 2 Volts [Wallace]

Dean Wallace asked that this item be dropped from this and future agendas.

5.4 Universal Backplane Annex (97-121r0) [Wallace]

Dean Wallace asked that this topic be deferred to the next working group
meeting.

5.5 Latching and Counting (97-216) [Guss]

Dave Guss was not present.  Bill Ham expressed concern that the latching and
counting topic doesn't have a home.  There was a brief argument between SPI-2
and SPI-3 as potential homes.  The consensus appeared to be that the material
should be integrated in SPI-3.

Additional discussion, based on new document revisions if any, was deferred to
the next working group meeting.

5.6 SPI-2 Revision 16 Review [Penokie]

George Penokie reviewed the SPI-2 changes between revisions 15 and 16.  George
noted several technical changes, for which he will request plenary votes to
approve making the changes in SPI-2 revision 17.  In the absence of any
objections, the working group recommended that the plenary approve the
technical changes made during its review.

5.7 SPI-2 Timing Tolerance (97-268) [Asami]

After a lengthy discussion of the tolerance issues raised by Tak Asami, the
group almost reached an agreement that the Receive Period Tolerance row be
removed from table 42 in SPI-2 revision 16.  But, that agreement was
short-lived when the group discussed additional statements that should be in
the standard.  Another lengthy discussion followed.  Much of the debate
centered on a concern that relaxing the specifications on the time interval
significantly would impinge on setup and hold times.  Also, the transfer
period tolerance table entries were added in response to a problem that
perhaps is better solved with the other clarifications that were made to the
SDTR negotiation.

Bruce Leshay noted that the transfer-to-transfer jitter does not scale
linearly with transfer rate; 0,25% may be okay for Fast-10, but it is
certainly too small for Fast-40.

Eventually, the group agreed to recommend to the plenary that:
1)  every entry in the Transmit Period Tolerance row in Table 42 of SPI-2
revision 17 be changed to 1 ns; and
2)  every entry in the Receive Period Tolerance row in Table 42 of SPI-2
revision 17 be changed to 1.1 ns.

5.8 Load Compensation (97-281) [Novak]

Vit Novak expressed concern that his presentation concerned single-ended SCSI.
He wondered whether the group would want to spend time on this topic, in view
of the late hour.  George Penokie asked that topics not intended for SPI-2
should be deferred.  Vit presented an executive summary of his work.

Basically, Vit's presentation contained a review of experiments he performed
to compensate for device capacitance presented to the bus by inserting an
inductor and resistor on both sides of each signal for each device.  The
circuit reduces the disturbance to the bus signals caused by the device.

John Lohmeyer stated that this topic would be placed under SPI-3 in agenda for
the next meeting.

5.9 Proposal to remove clause 10.5 from SPI-2 Rev 16 [Lohmeyer]

John Lohmeyer proposed removing clause 10.5 (Hot Plugging) from SPI-2.  Gene
Milligan asked what really goes away if clause 10.5 is removed.  George
Penokie said that the restriction against case 4 hot plugging goes away for
LVD and SE/LVD multi-mode.  John said that prohibiting case 4 hot plugging for
LVD and multi-mode is overly restrictive.  All testing and analysis done to
date has shown that case 4 hot plugging with real drivers and receivers does
indeed work.  While this does not guarantee that all LVD implementations will
work, the standard should not prohibit those that do work.

After a little research, Gene Milligan reported that removing clause 10.5 has
no effect, since equivalent words are present in clause 10.4.  At this point,
John proposed amending clause 10.4.  The group proceeded to discuss the
quality of the hardware implementation that is necessary to produce a system
capable of case 4 hot plugging.

Gerry Houlder raised concerns about drive vendors incorrectly being blamed for
not conforming to the "hot plugging rule".  John Lohmeyer responded that the
market demands are forcing case 4 hot plugging with LVD products and customers
will go to non-SCSI products if the demands are not met.

The group then considered the case 4 hot plugging conditions that might
produce undetected errors.  Bill Ham described problems with latching and
counting that can result in undetected errors (data corruption).

Gene Milligan moved that clause 10.5 be removed; that the second sentence from
bullet a) be removed from clause 10.4; and that a warning regarding undetected
errors be added to clause 10.4.  Before a second was made, the group debated
the "undetected" wording in Gene's motion.  John Lohmeyer seconded the motion.
The real focus of the debate that followed was the exact wording of the note
to be added to 10.4.

After four attempts, the group agreed on the following wording:  "LVD may
require more stringent system design to tolerate transients that occur during
insertion and removal events."

The motion passed 13:2.

6. Meeting Schedule

The next meeting of SPI-2 and SPI-3 Working Group associated with a T10
meeting week will be March 16, 1998, in San Diego, CA at the Hyatt Islandia
Hotel (619-224-1234), hosted by QLogic.  Several other meetings were
discussed, but the T10 plenary later authorized the following meetings:

  SPI-2 Editors Meeting           Dec 3, 1997     Irvine, CA / QLogic
  SPI-3 Protocol Meeting          Dec 4, 1997     Irvine, CA / QLogic
  SPI-3 Physical Meeting          Jan 21, 1998    Irvine, CA / QLogic
  SPI-2 LB Comments Resolution    Jan 22, 1998    Irvine, CA / QLogic

(STA has also planned SCSI Harbor meetings for the afternoon of Dec 2nd and
Jan 20th.)

7. Adjournment

The meeting was adjourned at 6:47 p.m. on Monday November 3, 1997.



--
John Lohmeyer                 E-Mail: john.lohmeyer at symbios.com
Symbios Logic Inc.             Voice: 719-533-7560
4420 ArrowsWest Dr.              Fax: 719-533-7036
Colo Spgs, CO 80907-3444    SCSI BBS: 719-533-7950 300--28800 baud

*
* 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