Date: January 12, 1989 To: X3T9.2 Membership From: Larry Lamers, X3T9.2 Recording Secretary John Lohmeyer, X3T9.2 Chairman Subject: January 1989 SCSI-2 Working Group Meeting Minutes Jeff Stai of Western Digital hosted a working group meeting January 9-11, 1989 in Costa Mesa, CA at the Red Lion Inn. The working group thanks Jeff and Western Digital for hosting this productive meeting and for making copies of SCSI-2 Rev 6a for the attendees. The working group also thanks the Red Lion Inn for thoughtfully providing us with incentive to get exercise on Wednesday with the false fire alarms. The final agenda was: SCSI-2 Items 1. Review of SCSI-2 revision 6a. 2. Review of LOG SELECT/SENSE changes and clarifications. (Jim Semenak) 3. Preliminary report on cable/connector testing. (Kurt Chan) 4. Rotational Position Locking (Steve Cornaby) 5. Helical Scan Devices (Pete Bramhall) 6. Single-Ended Termination (Paul Boulay) 7. Terminate Immediate [89-007] (Steve Goldman, Paul Nitza) 8. Request Sense Polling (Kurt Chan) 9. Asynchronous Event Notification Procedure [88-168] (Kurt Chan) 10. Response time after reset [89-005] (Kurt Chan) 11. Tape Read Density Reporting [89-006] (Kurt Chan) ESDI Items 31. ESDI Document [87-105R3, 89-002] (Larry Lamers, Kiyotake Kumazawa) SCSI-3 Items [not covered due to lack of time] 50. 87-186 SEND DIAGNOSTIC Pages (Bill Spence) 51. 87-203 & 87-217 LOAD SKIP MASK command proposal (Greg Floryance, Dave McIntyre) 52. 87-206 More that 8 devices on wide SCSI (David Harms) 53. 88-002 Search Command modifications (Jeff Stai) 54. 88-007 Expanded RelAdr Bit Definition (Paul Boulay) 55. 88-69R1, 88-92, & 88-100 Autoconfiguration SSWG (Jerry Marazas, Paul Nitza, Jim McGrath, Paul Boulay) 56. 88-127 Error Handling Action Codes (George Penokie) 57. 88-158 TERMINATE IMMEDIATE message (Steve Goldman) 58. 89-017 Optional PARTITION command (Pete Bramhill) The following people attended the meeting: Name Status Organization ------------------------------ ------ ------------------------------ Mr. Robert N. Snively P Adaptec, Inc. Mr. Charles Brill P AMP, Inc. Mr. Jerry Katzung O Apple Computer Mr. Gus Andrade O Apple Computer Mr. Ken Post P Archive Corp. Mr. James J. Semenak P AT&T Mr. Joe Lawlor O AT&T Mr. Dan Davies P Cipher Data Products, Inc. Mr. Chris Nieves O Computer Consoles Mr. Steve Goldman P DPT Mr. Tom Treadway A DPT Mr. Gerald Maurer P Emulex Corp. Mr. I. Dal Allan P ENDL Mr. Bill Duran O Exabyte Corp. Mr. Robert Liu P Fujitsu America, Inc. Mr. Bob Pentecost P Hewlett Packard Co. Mr. Kurt Chan A Hewlett Packard Co. Mr. Pete Bramhall O Hewlett Packard Co. Mr. Kiyotake Kumazawa O Hitachi America, Ltd. Mr. George Penokie P IBM Corp. Mr. Gerald Marazas A IBM Corp. Mr. Gary R. Stephens A IBM Corp. Mr. David A. Buesing O IBM Corp. Mr. Craig Scott A Interphase Corp. Mr. Erik Walberg O Konica Technology Corp. Mr. Paul Boulay A LMS-OSD Mr. Wayne Ihde V LMS-OSD Mr. William Homans P LMS-TSD Mr. Skip Jones A Maxtor Corp. Mr. Bob Masterson P Methode Electronics, Inc. Mr. Stephen Cornaby P Micropolis Corp. Mr. John Lohmeyer P NCR Corp. Mr. Paul Nitza P OTL Engineering Mr. James McGrath P Quantum Corp. Mr. Lawrence J. Lamers P Sony Corp. of America Mr. Curtis A. Ridgeway P Sun Microsystems, Inc. Mr. D. W. Spence P Texas Instruments Mr. Lou Edwards O Verbatim Mr. Jeff Stai P Western Digital 39 attendees + 9 personal computers Status Key: P - Principal A - Alternate O - Observer V - Visitor The following documents were distributed at the meeting: Document Doc Date Author Description of Document ------------- -------- --------------- --------------------------------------- X3T9.2/88-165 12/16/88 K. Chan Input current and resistor tolerances X3T9.2/88-167 12/14/88 S. Cornaby Rotational Position Locking X3T9.2/88-168 12/22/88 B. Pentecost AEN Initialization Procedure X3T9.2/89-2 12/22/88 K. Kumazawa ESDI "Index to Sector" Synchronization X3T9.2/89-4 1/6/89 P. Boulay Proposed Single-Ended SCSI Bus Terminator X3T9.2/89-5 1/6/89 K. Chan Recommended Response Time after RST X3T9.2/89-6 K. Chan Tape Read Density Reporting X3T9.2/89-7 1/7/89 P. Nitza Terminate Immediate with Auto Sense X3T9.2/89-8 Stephens/ "Brief" Description of Density Code Davies operation X3T9.2/89-9 1/9/89 P. Bramhill COPY and Partitioned Tape X3T9.2/89-10 1/9/89 P. Bramhill REQUEST SENSE EOM data on Partitioned Tape X3T9.2/89-11 1/9/89 P. Bramhill MODE SELECT Device Configuration Parameters X3T9.2/89-12 1/9/89 P. Bramhill Separation of BOM and BOP0 Rev 1 X3T9.2/89-13 1/9/89 P. Bramhill 9.2.13 SPACE Command, Rev 6a, page 9-30 Rev 1 X3T9.2/89-14 1/6/89 B. Pentecost READ POSITION Command X3T9.2/89-15 1/6/89 B. Pentecost Write with Length = 0 X3T9.2/89-16 1/10/89 P. Bramhill Active Format field in Device Configuration Parameters Page X3T9.2/89-17 1/10/89 P. Bramhill Introduce a new optional command X3T9.2/89-18 K. Chan Presentation on 25 mil centerline cable testing Agenda Items for SCSI-2 1. Review of SCSI-2 revision 6a. Most of the meeting time was spent reviewing SCSI-2 Rev 6a, which was distributed at the meeting. This is an interim document that only contains sections 1 - 9 plus 13. The editors prepared this document to shown the changes made at Austin and San Diego. There are no plans to distribute Rev 6a outside the working group since Rev 7 will be in the next mailing. Models - when will they get into the document? There was some favorable sentiment was expressed for Greg Fry's disk model, although some people thought it was too simple. There was some discussion of having a logical layer and a physical layer. Gary Stephens will provide a copy of his proposal from yesteryear on this subject. John & Larry were assigned to develop the section 7 model and Jeff was assigned to develop the section 8 model. If a target truncates the REQUEST SENSE data, the initiator should assume that all data beyond the truncation point logically has a value of zero. Gus Andrade requested that the wording for target response to the ATN signal during a data phase be improved to suggest that targets should respond at logical block boundaries. The proposed change was accepted. A long discussion interrupted by a fire drill took place on the interactive effects of PREVENT/ALLOW, RESERVE/RELEASE, and multiple initiators. The resolution was that PREVENT/ALLOW are independent of device reservations and the prevention condition remains in effect until all initiators that previously prevented medium removal have allowed medium removal. John Lohmeyer drafted the revised wording which will be included in S2R7. Third-party reserve will be mandatory for all device types, including sequential-access. There were MANY other corrections made which were noted in the editor's copies of the document. These will be included in S2R7. 2. Review of LOG SELECT/SENSE changes and clarifications. (Jim Semenak) Jim, as editor of section seven, will heal himself, as any good physician should do. 3. Preliminary report on cable/connector testing. (Kurt Chan) Kurt presented his preliminary results (see document 89-018). 4. Rotational Position Locking [88-167] (Steve Cornaby) Steve's proposal met with enough opposition that he will re-draft his proposal for the next plenary meeting. Jim Semenak accepted an action item to revise the existing wording on RPL to cover some of the topics of the discussion. 5. Helical Scan Devices (Pete Bramhall) There were several proposals from Hewlett Packard related to using helical scan devices as sequential-access devices. Most of these proposals were simple corrections that should have been made when partitions were added to the sequential-access device model. The resolution of these proposals is as follows: Doc Author Description Resolution ------- ------------ -------------------------- ------------------------------ 89-9 P. Bramhill COPY and Partitioned Tape end-of-medium changed to end-of-partition 89-10 P. Bramhill REQUEST SENSE EOM data Will fix medium to be on Partitioned Tape partition 89-11 P. Bramhill MODE SELECT Device Accepted Configuration Parameters 89-12R1 P. Bramhill Separation of BOM and BOP0 Accepted 89-13R1 P. Bramhill 9.2.13 SPACE Command, Rejected Rev 6a, page 9-30 89-14 B. Pentecost READ POSITION Command Partially accepted 89-15 B. Pentecost Write with Length = 0 Rejected 89-16 P. Bramhill Active Format field in Some of the information was Device Configuration added to the Error Recovery Parameters Page Page 89-17 P. Bramhill Introduce a new optional Put in SCSI-3 queue command 6. Single-Ended Termination [88-165, 89-004] (Paul Boulay, Kurt Chan) Paul made a presentation of alternatives for terminators on single-ended systems using an off-the-shelf voltage regulator and several different resistor values. This proposal allows the impedance of the termination to be lowered so that it better matches the typical cable impedance. Paul still hasn't resolved whether it is better to specify Vout as 2.6v or 2.8v. The lower voltage allows 100 ohm impedance while the higher voltage improves noise margins at the expense of 110 ohms. Potential problems are size reduction and thermal considerations (worst case dissipation is 1 watt). Paul will generate an appendix for S2R7. There was a request made for Kurt Chan to accept an action item to test the configurations in the proposal. Kurt Chan reported on problems related to noise margins and powered-down devices plus the need to specify resistor tolerances for the single-ended terminator. His document was accepted. A note will be included to recommend that powered-down devices not load the bus. 7. Terminate Immediate [89-007] (Steve Goldman, Paul Nitza) Paul Nitza accepted an action item to re-draft the proposal for the TERMINATE IMMEDIATE message and submit to the next plenary meeting. The proposal to include auto sense will be a SCSI-3 item, unless Paul Nitza can come up with a better justification for including it in SCSI-2. 8. REQUEST SENSE Polling (Kurt Chan) What started out as an innocent question from Kurt generated a lengthy discussion on the BBS and even more (sometimes vehement) discussion in the meeting. John Lohmeyer accepted an action item to draft new wording for the REQUEST SENSE description on page 7-44 of S2R6a. This wording was reviewed later in the meeting and accepted for S2R7. 9. Asynchronous Event Notification Procedure [88-168] (Kurt Chan) Accepted for inclusion in Appendix H. 10. Response time after reset [89-005] (Kurt Chan) There was some discussion regarding what a host should do if it cannot respond in 250 milliseconds. Bottom line is that the host will not be able to act as a target in these circumstances. The proposal was not accepted. 11. Tape Read Density Reporting [89-006] (Kurt Chan) Gary Stephens accepted an action item to review the tape model and draft some wording for clarification. This wording was assigned document number 89-008. 31. ESDI Document [87-105R3, 89-002] (Larry Lamers, Kiyotake Kumazawa) A meeting of persons interested in ESDI was held at 8:15 AM Tuesday, January 10, 1989. The modifications to the ESDI document for the public review comments were made to the ISO-style version of the ESDI document, however a document was not submitted by Dal Allan in a format acceptable for distribution. Larry Lamers prepared a version of the document in ANSI format and used 'CompareRite' to indicate the changes made so that the working group could review the document. Dal Allan objected to this document being used, claiming that he was still the editor of the document and he would provide a good quality, ANSI format, camera-ready copy for the next document distribution. Kumazawa-san's proposal was favorably received. It was suggested that the command code be changed and a revised proposal be brought to the next plenary meeting for a vote. The SCSI-3 items were not discussed due to limited time.