Draft Minutes of SPI Physical Working Group - Oct 13-14, 1999
John Lohmeyer
lohmeyer at t10.org
Fri Oct 15 10:13:23 PDT 1999
* From the T10 Reflector (t10 at t10.org), posted by:
* John Lohmeyer <lohmeyer at t10.org>
*
These minutes are available in PDF format at:
ftp://ftp.t10.org/t10/document.99/99-290r0.pdf
Accredited Standards Committee*
National Committee for Information Technology Standards (NCITS)
Doc. No.: T10/99-290r0
Date: October 15, 1999
Reply to: John Lohmeyer
To: T10 Membership
From: John Lohmeyer
Subject: SCSI Physical Working Group Meeting -- October 13-14, 1999
Colorado Springs, CO
Agenda
1. Opening Remarks
2. Approval of Agenda
3. Attendance and Membership
4. SPI-3 Topics
4.1 Cable Media Performance Testing (98-219, 99-111) [Ham]
5. SPI-4 Topics
5.1 Load Compensation [Novak]
5.2 Cable/System Simulation Issues (99-204) [Wallace]
5.3 Next Generation Cable Performance [Ham]
5.4 SCSI out of band communications method (99-213) [Petty]
5.5 Presentation on Ultra320 (99-302) [Bastiani]
5.6 Extended Addressing for SPI-4 (99-249, 99-250) [Monia]
5.7 SPI-4 ISI reduction via transmit pre-compensation (99-260) [Petty]
5.8 SPI-4 timing budget utilizing receiver signal deskew method (99-261)
[Petty]
5.9 SPI-4 clocking proposal (99-262) [Petty]
5.10 Margin Control (99-264) [Lamers]
5.11 Setting Extended Addresses (99-273) [Milligan]
5.12 Flow Control & Continue I/O Process Flag (99-142) [Lamers]
5.13 Proposal for Fast-160 to be included in SPI-4 (99-295) [Milligan]
5.14 Fast-160 Simplified Timing presentation (99-296) [Milligan]
5.15 Proposal for a training pattern for SPI-4 (99-297) [Evans]
5.16 Proposal for turn-on/turn-off of a free-running clock (99-298)
[Evans]
5.17 Fast-160 Presentation (99-300) [Patel]
5.18 Compaq inputs on SPI-4 (99-301) [Ham]
6. Domain Validation Technical Report Topics
7. New Business
7.1 How to verify goodness (99-305) [Bastiani]
7.2 Where do we go from here? [Penokie]
8. Meeting Schedule
9. Adjournment
Results of Meeting
1. Opening Remarks
John Lohmeyer, the T10 Chair, called the meeting to order at 1:20 p.m.,
Wednesday October 13, 1999. He thanked LSI Logic 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 additions and changes:
5.18 Compaq inputs on SPI-4 (99-301) [Ham]
The following agenda items were added or revised during the course of the
meeting:
7.1 How to verify goodness (99-305) [Bastiani]
7.2 Where do we go from here? [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. Bill Galloway P BREA Technologies, Inc. billg at breatech.com
Mr. Robert C. Elliott P Compaq Computer Corp. Robert.Elliott at compaq.com
Dr. William Ham A Compaq Computer Corp. bill.ham at digital.com
Mr. Roger Cummings P Distributed Processing cummings_roger at dpt.com
Tech.
Mr. George O. Penokie P IBM Corp. gop at us.ibm.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. 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. Bruce Leshay V Quantum Corp. bleshay at tdh.qntm.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. Bill Gintz V Seus, Inc. wcgintz at ix.netcom.com
Mr. Paul D. Aloisi P Unitrode Corporation Paul_Aloisi at ti.com
21 People Present
Status Key: P - Principal
A,A# - Alternate
O - Observer
L - Liaison
V - Visitor
4. SPI-3 Topics
4.1 Cable Media Performance Testing (98-219, 99-111) [Ham]
This topic was deferred to a subsequent meeting.
5. SPI-4 Topics
5.1 Load Compensation [Novak]
This topic was deferred to a subsequent meeting.
5.2 Cable/System Simulation Issues (99-204) [Wallace]
This topic was deferred to a subsequent meeting.
5.3 Next Generation Cable Performance [Ham]
This topic was deferred to a subsequent meeting.
5.4 SCSI out of band communications method (99-213) [Petty]
Bill Petty said he had not revised his proposal since he had first presented
it. The topic was deferred to a subsequent meeting.
5.5 Presentation on Ultra320 (99-302) [Bastiani]
Vince Bastiani presented 99-302. He proposed that the minimum driver rise
time be increased as compared to SPI-3. It was noted that driver rise time
only affects short cables; on longer cables the cable attenuation controls the
rise time. Bill Ham argued that fast rise times on short cables are not a
problem because the signal is stronger (due to less attenuation) so the
glitches do not reach the receiver threshold. No agreement was reached.
5.6 Extended Addressing for SPI-4 (99-249, 99-250) [Monia]
This topic was deferred to a subsequent meeting.
5.7 SPI-4 ISI reduction via transmit pre-compensation (99-260) [Petty]
Bill Petty said he had made no changes to make to this proposal. The rev. 0
proposal was discussed again. There was some discussion of whether it is
better to use timing compensation or amplitude compensation. It was suggested
that either method be allowed in SPI-4.
5.8 SPI-4 timing budget utilizing receiver signal deskew method (99-261)
[Petty]
Bill Petty presented 99-261r1, SPI-4 timing budget utilizing receiver signal
deskew method. Bill noted that the REQ and ACK edges are sent at the same
time as the data. This is necessary during training and simplifies the normal
data phases as it allows the receiving device to determine its own setup and
hold timing.
The issue of how often to do training was discussed. Bill noted that the LSI
Logic performance analysis group planned to study the impact of using a
training pattern on all data phases for various workloads-- if the impact is
not too great, then he would propose doing training on each data phase.
Bill said he would revise his proposal based on the input received.
Late in the meeting, Gene Milligan presented a hand-drawn foil with some
differences in the timing budget. Bruce Leshay suggested that rather than
creating another document dealing with this subject that Gene should work with
Bill Petty to capture Gene's differences. Gene agreed provided that Bill's
document accommodate the timing budget for a non-free running clock. Bill
agreed to work with Gene.
5.9 SPI-4 clocking proposal (99-262) [Petty]
Bill Petty presented 99-262r1, SPI-4 REQ/ACK ISI removal via qualified
clocking method. He had doubled the setup and hold budget for the P1 signal
by only sampling the P1 signal on the assertion edge of the clock signal.
This proposal would also simplify the internal data path design as
compared to
using both edges.
Bill noted that his proposal would require that all REQ/ACK offsets be even
and that all transfers be done in pairs.
5.10 Margin Control (99-264) [Lamers]
This topic was deferred to a subsequent meeting.
5.11 Setting Extended Addresses (99-273) [Milligan]
This topic was deferred to a subsequent meeting.
5.12 Flow Control & Continue I/O Process Flag (99-142) [Lamers]
This topic was deferred to a subsequent meeting.
5.13 Proposal for Fast-160 to be included in SPI-4 (99-295) [Milligan]
Gene Milligan presented 99-296r0, Proposal for Fast-160 to be included in
SPI-4. Gene reviewed the timing assumptions and the various wording changes
that would be necessary to incorporate his Fast-160 proposal (without a
free-running clock) into SPI-4, assuming we start with the SPI-3 document.
Dan Smith presented 99-303r0, Skew Measurements and 99-304r0, ISI
Measurements.
5.14 Fast-160 Simplified Timing presentation (99-296) [Milligan]
Gene Milligan presented 99-296r0, Fast-160 Simplified Timing presentation.
The more-detailed proposal is contained in 99-295r0 and was covered (later)
under agenda item 5.13.
5.15 Proposal for a training pattern for SPI-4 (99-297) [Evans]
Mark Evans briefly described 99-297r0, proposal for a training pattern for
SPI-4. There was a discussion over whether the pattern should be 010101... or
00110011... or 00110011...010101... The key difference is whether there can
more than 6.25 ns of signal skew. If so, then the lower frequency pattern
will be necessary. If not, then only the high frequency pattern is needed.
We need more test data to make this decision.
5.16 Proposal for turn-on/turn-off of a free-running clock (99-298) [Evans]
Bruce Leshay presented 99-298r0, proposal for turn-on/turn-off of a free
running clock. Bruce noted that this proposal does not address ISI training
and some changes would be necessary to deal with the ISI training sequence.
There was a discussion about the 400 ns. settling time following the changes
to the phase signals (C/D, I/O, and MSG). Bruce assured the group that his
proposal did not make any changes to this time, which has been in SCSI
forever. The question of why 400 ns. kept coming up throughout the meeting.
John noted that 400 ns. was considered a very small time in the early days of
SCSI and was most likely selected because it allowed signals to settle without
depending on incident wave switching.
It was noted that for a DT DATA OUT phase there are some problems with the P0
signal timing, which goes the opposite direction of the rest of the data
signals. Since P0 does not transition very often, the easiest solution is to
add extra setup and hold time around the P0 transitions. Bruce agreed to
revise his proposal per the group's input.
5.17 Fast-160 Presentation (99-300) [Patel]
Mayank Patel presented 99-300r0, Fast-160 Presentation. A major difference in
this proposal from the other Fast-160 proposals is that it does not include a
free running clock. The contention is that there is adequate timing budget to
make Fast-160 work without requiring a free-running clock.
5.18 Compaq inputs on SPI-4 (99-301) [Ham]
Bill Ham presented 99-301, Compaq's inputs on SPI-4. Some items were
controversial. John reminded the group that the controversial items may need
to be resolved at the T10 plenary level and the working group should not bog
down on these items.
6. Domain Validation Technical Report Topics
John Lohmeyer said that the SCSI Domain Validation (SDV) Technical Report
project had been authorized by NCITS and project number 1378-DT had been
assigned.
7. New Business
7.1 How to verify goodness (99-305) [Bastiani]
Vince Bastiani presented 99-305r0, how to verify goodness. His main point was
that it would be useful to have specific test configurations to use to
evaluate the `goodness' of the various SPI-4 proposals.
7.2 Where do we go from here? [Penokie]
George Penokie said that he believes we now have proposals for most of the
areas of SPI-4. We now need to get serious about the detailed numbers. He
proposed that we target the December meeting for making decisions on the SPI-4
direction. He wants convergence on the technical direction so he can begin
work on putting together the SPI-4 draft document.
8. Meeting Schedule
The next meeting of the SCSI Physical Working Group will be Monday, November
1, 1999 commencing at 1 p.m. recessing on Monday evening and resuming at 9
a.m. until 6 p.m. on Tuesday, November 2, 1999 at the Embassy Suites on
Monterey Bay (831-393-1115) in Monterey, CA hosted by Adaptec, Inc.
John Lohmeyer presented a foil showing the planned SCSI Physical Working Group
meetings:
Meeting Dates Location / Contact
SCSI PHY WG 11/1/99 1:00p Mon Monterey, CA / Larry Lamers
11/2/99 Tue Monterey, CA / Larry Lamers
SCSI PHY WG 12/2/99 1:00p Thur Rochester, MN / Fasig
12/3/99 Fri Rochester, MN / Fasig
SCSI PHY WG 1/10/00 1:00p Mon Brisbane, Aus. / Dal Allan
1/11/00 Tue. Brisbane, Aus. / Dal Allan
SCSI PHY WG 2/2/00 1:00p Wed Huntington Beach, CA / Skip Jones ***
2/3/00 Thur Huntington Beach, CA / Skip Jones
*** Proposed Meeting -- Not authorized yet
Gene Milligan and Mark Evans noted that the 12/2/99 and 12/3/99 meeting is in
conflict with the 12/3/99 T13 meeting. John said that since T10 had already
approved this meeting, the issue should be addressed at the 11/4/99 T10
meeting.
It was noted that the proposed meeting on 2/2/00 and 2/3/00 is in conflict
with a proposed Cable Performance and SSM meeting proposed for 2/1/00 and
2/2/00. John took an action item to contact Skip Jones regarding this issue.
9. Adjournment
The meeting was adjourned at 12:20 p.m. on Thursday October 14, 1999.
--
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