draft minutes of Parallel SCSI WG - March 27-28, 2000
John Lohmeyer
lohmeyer at t10.org
Wed Mar 29 14:06:38 PST 2000
* From the T10 Reflector (t10 at t10.org), posted by:
* John Lohmeyer <lohmeyer at t10.org>
*
A PDF version of these minutes is available at:
ftp://ftp.t10.org/t10/document.00/00-184r0.pdf
Accredited Standards Committee*
National Committee for Information Technology Standards (NCITS)
Doc. No.: T10/00-184r0
Date: March 29, 2000
Reply to: John Lohmeyer
To: T10 Membership
From: John Lohmeyer
Subject: Parallel SCSI Working Group Meeting -- March 27-28, 2000
Milpitas, CA
Agenda
1. Opening Remarks
2. Approval of Agenda
3. Attendance and Membership
4. SPI-4 Physical Topics
4.1 Free Running Clock
4.1.1 SPI-4 clocking proposal (99-262) [Petty]
4.1.2 ISI Compensation Selection [Penokie]
4.2 Training Patterns
4.2.1 Ultra320 SCSI Calibration Protocol (00-133) [Leshay]
4.2.2 Proposal for training pattern to be included in SPI-4 (00-132)
[Evans]
4.2.3 Proposed Training for Skew Compensation (00-174) [Bastiani]
4.2.4 Training Patterns Modeling Results (00-193) [Penokie]
4.3 Timings
4.3.1 Proposal for Fast-160 to be included in SPI-4 (99-295) [Milligan]
4.3.2 Clock Frequency Restrictions for Fast-160 [Evans]
4.3.3 Fast-160 Driver and Receiver Masks (00-198) [Aloisi]
4.4 Test Configurations
4.4.1 Use of eye measurements (00-126) [Bastiani]
4.4.2 ISI Measurements (99-337) [Bridgewater/Bastiani]
4.4.3 Ultra320 into fully populated 10-slot backplane (00-195) [Brown]
4.4.4 Requirements for Measuring Receive Signals in SPI-4 and beyond
(00-149) [Ham]
4.4.5 Receiver Input Voltage Budget for Eye Patterns (00-158)
[Bridgewater]
4.5 Other SPI-4 Physical Topics
4.5.1 Precompensation and test measurements (00-194) [Maniloi]
4.5.2 320 data transfer rates on 25 meter cables (00-153) [McGarrah]
4.5.3 640 data transfer rates, a first look (00-154) [McGarrah]
4.5.4 Comments on adaptive filtering for Ultra320 SCSI (00-196) [Brown]
4.5.5 Pre-Emphasis Experimental Data (99-167) [Gasparik]
4.5.6 Expander Topics [Ham]
4.5.7 Eye window amplitude (00-200) [Bridgewater]
4.5.8 The zero offset problem for receiver equalization (00-201)
[Bridgewater]
4.5.9 Next Meeting Testing Data [Penokie]
5. SPI-4 Protocol Topics
5.1 Margin Control (99-264) [Lamers]
5.2 Flow Control & Read Streaming (00-142) [Lamers]
5.3 PPR Message Enhancements (99-283) [Lamers]
6. Expanders and Domain Validation Topics
6.1 SCSI out of band communications method (99-213) [Petty]
7. New Business
8. Meeting Schedule
9. Adjournment
Results of Meeting
1. Opening Remarks
John Lohmeyer, the T10 Chair, called the meeting to order at 1:00 p.m.,
Monday, March 27, 2000. He thanked Larry Lamers 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 draft agenda was approved with the following changes:
4.2.4 Training Patterns Modeling Results (00-193) [Penokie] <added>
4.5.1 Driver Precompensation with Receiver Filtering for Ultra4 (00-175)
[Smith] <changed to:>
4.5.1 Precompensation and test measurements (00-194) [Maniloi]
4.5.2 320 data transfer rates on 25 meter cables (00-153) [McGarrah]
<deleted>
4.5.3 640 data transfer rates, a first look (00-154) [McGarrah] <deleted>
4.5.5 Pre-Emphasis Experimental Data (99-167) [Gasparik] <deleted>
During the course of the meeting, the following agenda items were
added/revised:
4.3.2 Clock Frequency Restrictions for Fast-160 [Evans]
4.3.3 Fast-160 Driver and Receiver Masks (00-198) [Aloisi]
4.4.3 Ultra320 SCSI vs. Ultra160 SCSI Eye Diagram Data (00-169) [Brown]
<changed to:>
4.4.3 Ultra320 into fully populated 10-slot backplane (00-195) [Brown]
4.5.4 Receiver Equalization (00-168) [Brown] <changed to:>
4.5.4 Comments on adaptive filtering for Ultra320 SCSI (00-196) [Brown]
4.5.7 Eye window amplitude (00-200) [Bridgewater]
4.5.8 The zero offset problem for receiver equalization (00-201)
[Bridgewater]
4.5.9 Next Meeting Testing Data [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. Lawrence J. Lamers P Adaptec, Inc. ljlamers at ieee.org
Mr. Vincent Bastiani A# Adaptec, Inc. bastiani at corp.adaptec.com
Mr. Wally Bridgewater A# Adaptec, Inc. wally at eng.adaptec.com
Mr. Tariq Abou-Jeyab O Adaptec, Inc. tajeyab at corp.adaptec.com
Mr. Lee Hearn V Adaptec, Inc. lee_hearn at corp.adaptec.
com
Mr. Bart Raudebaugh P Ancot Corp. bart at ancot.com
Ms. Christine V Berg Electronics cmulroney at fciconnect.com
Mulrooney
Mr. Bill Galloway P BREA Technologies, Inc. billg at breatech.com
Mr. Robert C. Elliott P Compaq Computer Corp. Rob_Elliott at compuserve.
com
Dr. William Ham A Compaq Computer Corp. bill.ham at digital.com
Mr. Keith Childs V Compaq Computer Corp.
Mr. Charles Tashbook P Dallas Semiconductor charles.tashbook at dalsemi.
com
Mr. Eugene Lew P Fujitsu elew at fcpa.fujitsu.com
Mr. Sam Ray V IBM str at us.ibm.com
Mr. George O. Penokie P IBM Corp. gop at us.ibm.com
Mr. Brian Schuh V IBM Corp. bschuh at us.ibm.com
Mr. Dennis Moore P KnowledgeTek, Inc. dmoore at ix.netcom.com
Mr. John Lohmeyer P LSI Logic Corp. lohmeyer at t10.org
Mr. Frank Gasparik V LSI Logic Corp. frank.gasparik at lsil.com
Mr. Alan Littlewood V LSI Logic Corp. alanl at lsil.com
Mr. William Petty V LSI Logic Corp. william.petty at lsil.com
Mr. Brian Day V LSI Logic Corp. brian.day at lsil.com
Mr. Keith Maloney V LSI Logic Corp. keith.maloney at lsil.com
Mr. Dennis Rehm V LSI Logic Corp. dennis.rehm at lsil.com
Mr. Mark Strauss V Lucent Technologies msstrauss at lucent.com
Mr. Makesh V Lucent Technologies makesh at lucent.com
KoMaudarauan
Mr. Richard Moore A# QLogic Corp. r_moore at qlc.com
Mr. Ting Li Chan V QLogic Corp. t_chan at qlc.com
Mr. Mark Evans P Quantum Corp. mark.evans at quantum.com
Mr. Russ Brown V Quantum Corp. russ.brown at quantum.com
Mr. Bruce Leshay V Quantum Corp. bleshay at tdh.qntm.com
Mr. Ken Erickson V Quantum Corp. ken.erickson at quantum.com
Mr. Duncan Penman V Quantum Corp. duncan_penman at quantum.com
Mr. Hollis Poche V Quantum Corp. hollis.poche at quantum.com
Mr. Gene Milligan P Seagate Technology Gene_Milligan at notes.
seagate.com
Mr. Daniel (Dan) F. O Seagate Technology daniel_f_smith at notes.
Smith seagate.com
Mr. Mayank R. Patel V Seagate Technology mayank_r_patel at notes.
seagate.com
Mr. Umesh Chandra V Seagate Technology umesh_chandra at notes.
seagate.com
Mr. Bruce Maniloi V Seagate Technology bruce_maniloi at notes.
seagate.com
Mr. John Masiewicz V Seagate Technology john_masiewicz at notes.
seagate.com
Mr. Ron Roberts V Sierra-Pac Technology rkroberts at aol.com
Mr. Paul D. Aloisi P Texas Instruments Paul_Aloisi at ti.com
Mr. Donald R. Getty A# Texas Instruments donald_getty at ti.com
Mr. Mike Kosco V Texas Instruments mike at mvbuilders.com
Mr. John T. Wilson V Texas Instruments jwilson at dal.asp.ti.com
45 People Present
Status Key: P - Principal
A,A# - Alternate
O - Observer
L - Liaison
V - Visitor
4. SPI-4 Physical Topics
4.1 Free Running Clock
4.1.1 SPI-4 clocking proposal (99-262) [Petty]
Bill Petty asked that this item be deferred to a subsequent meeting.
4.1.2 ISI Compensation Selection [Penokie]
George Penokie presented a hand-drawn foil with a matrix of possible
requirements on equalization and precompensation. He suggested that the
results might shed some light on possible ways to resolve this issue. He
asked for a per-company vote with each company allowed to vote Yes, No, or
Abstain on each row.
Bill Ham argued that these features should not be optional. It is hard to
design systems for optional features at the physical layer.
The voting was:
Driver Receiver
Pre-Comp Equalizer Yes No Abstain
-------- --------- --- -- -------
Silent Silent 0 12 1
Mandatory Silent 6 5 2
Silent Mandatory 4 6 3
Mandatory Optional 6 5 1
Mandatory Mandatory 4 6 1
Optional Mandatory 4 7 1
It was suggested that the voting might be more reveling if each company was
allowed only one Yes vote. The results of this `vote' were:
Driver Receiver
Pre-Comp Equalizer Yes
-------- --------- ---
Silent Silent 0
Mandatory Silent 4
Silent Mandatory 1
Mandatory Optional 4
Mandatory Mandatory 3
Optional Mandatory 1
There was a short discussion of the significance of these two votes and it was
agreed that we should wait until the next meeting before doing any more
voting. Several people observed that more data is needed.
4.2 Training Patterns
4.2.1 Ultra320 SCSI Calibration Protocol (00-133) [Leshay]
Bruce Leshay presented 00-133r2, Ultra320 SCSI Calibration Protocol. Bruce
noted an error on slide 8 where `100ns' should have been `200ns'.
There was some discussion regarding the startup of DATA OUT phase (slide 9)
beginning before the assertion of the REQ signal by the target. Instead the
initiator begins transmitting ACK pulses approximately 90 ns after it detects
that the C/D, I/O, and MSG signals are in the correct state for DATA OUT
phase.
The question of whether the free-running clock should be paused during the 50
ns data-line pause was left open until the next meeting. Bill Petty was given
an action item to investigate this issue and give the response to Bruce Leshay
as soon an possible.
4.2.2 Proposal for training pattern to be included in SPI-4 (00-132) [Evans]
Mark Evans asked that this item be deferred to a subsequent meeting.
4.2.3 Proposed Training for Skew Compensation (00-174) [Bastiani]
Wally Bridgewater presented 00-174, Proposed Training for Skew Compensation.
Wally asked for an additional training pattern that would be specific to his
implementation of equalization. Some concern was expressed over how many
vendor-specific training patterns might be needed.
4.2.4 Training Patterns Modeling Results (00-193) [Penokie]
George Penokie presented 00-193, Training Patterns Modeling Results. His
conclusion was that the impact of training on system throughput is low. He
recommended that SPI-4 require training once per direction per connection, to
be remembered throughout that connection only.
George projected his spreadsheet on the screen and showed the effect of
changing various parameters as requested by several people who were present.
George Penokie moved that we recommend the summary section 00-193r0 be
incorporated into SPI-4. Bruce Leshay seconded the motion. The motion
passed
15:0:0.
4.3 Timings
4.3.1 Proposal for Fast-160 to be included in SPI-4 (99-295) [Milligan]
Gene Milligan presented portions of 99-295r3, Proposal for Fast-160 to be
included in SPI-4. Several updates were made and Gene agreed to prepare a rev
4 document.
4.3.2 Clock Frequency Restrictions for Fast-160 [Evans]
George Penokie suggested that Fast-160 have a restricted frequency range on
the clock period, rather than just being a maximum frequency. This would
simplify both skew compensation and filter design. There was general
agreement that such a restriction is a good idea. Gene Milligan noted that
deleting some occurrences of `or greater' from his proposal (99-295) should
accomplish this goal and agreed to do so in the next revision of it.
4.3.3 Fast-160 Driver and Receiver Masks (00-198) [Aloisi]
Paul Aloisi presented 00-198r0, Fast-160 Driver and Receiver Masks. Paul
suggested that the various Fast-160 proposals should be written in a similar
format. This proposal also included masks for Fast-320, which some people
judged to be premature.
4.4 Test Configurations
4.4.1 Use of eye measurements (00-126) [Bastiani]
Vince Bastiani asked that this item be deferred to a subsequent meeting.
4.4.2 ISI Measurements (99-337) [Bridgewater/Bastiani]
Vince Bastiani asked that this be deleted from the agenda.
4.4.3 Ultra320 into fully populated 10-slot backplane (00-195) [Brown]
Russ Brown presented 00-195, Ultra320 into fully populated 10-slot backplane.
This presentation showed the benefits of equalization with an existing
backplane.
4.4.4 Requirements for Measuring Receive Signals in SPI-4 and beyond (00-149)
[Ham]
Bill Ham asked that this item be deferred to a subsequent meeting.
4.4.5 Receiver Input Voltage Budget for Eye Patterns (00-158) [Bridgewater]
This item was not covered separately from item 4.5.7.
4.5 Other SPI-4 Physical Topics
4.5.1 Precompensation and test measurements (00-194) [Maniloi]
Bruce Maniloi presented 00-194r0, Presentation on precompensation and test
measurements. The data presented supports the case that precompensation alone
is adequate for Fast-160.
There was considerable discussion about the relative merits of transmit
precompensation vs. receiver equalization with no immediate closure. It was
suggested that this issue might be more easily resolved if we could agree on
some worst-case configurations.
George Penokie observed that if SPI-4 were to specify a minimum transmit
precompensation level and it were to also include a mandatory training pattern
for receiver equalization, then the decision to include (or not include)
receiver equalization could be optional in SPI-4 devices.
4.5.2 320 data transfer rates on 25 meter cables (00-153) [McGarrah]
Mark Evans asked that this item be deleted from the agenda.
4.5.3 640 data transfer rates, a first look (00-154) [McGarrah]
Mark Evans asked that this item be deleted from the agenda.
4.5.4 Comments on adaptive filtering for Ultra320 SCSI (00-196) [Brown]
Russ Brown presented 00-196, Comments on adaptive filtering for Ultra320 SCSI.
4.5.5 Pre-Emphasis Experimental Data (99-167) [Gasparik]
Frank Gasparik asked that this item be deleted from the agenda.
4.5.6 Expander Topics [Ham]
Bill Ham presented 00-199, Proposal for expander requirements, domain
validation basic integrity check, and related target requirements. He edited
the document on screen with input from the group, producing revision 0. He
stated an intention to prepare a revision 1, which would include a few more
changes too extensive to do online.
4.5.7 Eye window amplitude (00-200) [Bridgewater]
Wally Bridgewater presented 00-200r0, Eye window amplitude. Wally said that
precompensation mitigates the overdrive requirement. Thus we can reduce the
overdrive requirement from 100 millivolts to 80 millivolts. Or if we leave it
at 100 millivolts, we will have increased margin for the chip.
4.5.8 The zero offset problem for receiver equalization (00-201)
[Bridgewater]
Wally Bridgewater presented 00-201, the zero offset problem for receiver
equalization. He described an implementation of an equalizer circuit that he
has been investigating. He is concerned that large amounts of equalization
will introduce significant noise.
4.5.9 Next Meeting Testing Data [Penokie]
George Penokie asked that those with concerns about equalization or
precompensation state their concerns so that the proponents can have an
opportunity to address those concerns. The equalizer concerns were:
1. Effects of offset in the equalizer circuit which are included in the eye
mask (inject the offset before the equalizer circuit).
2. What was the procedure that Russ used to determine "worst-case" crosstalk?
3. Effects of crosstalk to/from a signal going the opposite direction from the
data (ACK or REQ).
4. Can the testers of equalizers run with longer patterns?
The precompensation concerns were:
1. Understand the difference of test setup vs. real silicon
2. Effects of crosstalk to/from a signal going the opposite direction from the
data (ACK or REQ)
3. Does pre-comp increase common mode signal?
4. Does the power of the driver increase?
5. Ensure testing done with worst-case crosstalk.
6. Does pre-comp increase driver capacitance?
5. SPI-4 Protocol Topics
5.1 Margin Control (99-264) [Lamers]
Larry Lamers asked that this item be moved to the Commands, Architecture, and
Protocol working group agenda in May. He also asked for reflector discussion
on this topic.
5.2 Flow Control & Read Streaming (00-142) [Lamers]
Larry Lamers asked that this item be moved to the Commands, Architecture, and
Protocol working group agenda in May. He also asked for reflector discussion
on this topic.
5.3 PPR Message Enhancements (99-283) [Lamers]
Larry Lamers asked that this item be moved to the Commands, Architecture, and
Protocol working group agenda in May. He also asked for reflector discussion
on this topic.
6. Expanders and Domain Validation Topics
6.1 SCSI out of band communications method (99-213) [Petty]
Bill Petty asked that this item be deferred to a subsequent meeting.
7. New Business
There was no new business.
8. Meeting Schedule
The next meeting of the Parallel SCSI Working Group will be Wednesday, April
26, 2000 commencing at 1:00 p.m. until 6 p.m. and continuing Thursday, April
27, 2000 at 9:00 a.m. until 5:00 p.m. at the Embassy Suites (719-599-9100) in
Colorado Springs, CO hosted by LSI Logic.
The subsequent meetings of this group are May 16 starting at 9 a.m. in Nashua,
NH and Thursday, June 15, 2000 commencing at 9:00 a.m. and continuing Friday,
June 16, 2000 in Chicago, IL.
9. Adjournment
The meeting was adjourned at 4:29 p.m. on Tuesday March 28, 2000.
--
John Lohmeyer Email: lohmeyer at t10.org
LSI Logic Corp. Voice: +1-719-533-7560
4420 ArrowsWest Dr. Fax: +1-719-533-7183
Colo Spgs, CO 80907
*
* 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