Draft Agenda -- X3T9.2 Meeting #83 August 21-22, 1989 -- Colorado Springs, CO 1. Opening Remarks 2. Approval of Agenda 3. Attendance and Membership 3.1 Roll Call of Members in Jeopardy 3.2 1990 - 1991 Service Fees (X3T9.2/89-87) 4. Approval of Minutes - June Meeting San Jose, CA (X3T9.2/89-75) 5. Document Distribution 6. Liaison Reports 6.1 ISO 6.2 IT8 6.3 SCSI-2 Common Access Method Committee Report 6.4 Fiber Channel 6.5 X3B7 Report 7. Review of Old Action Items 8. Working Group and/or Project Status Reports 8.1 ESDI (Tuesday p.m.) 8.1.1 Editorial Error in ESDI Rev 3A (X3T9.2/89-102) 8.1.2 Maxtor Issue on Synchronized Spindles 8.2 SCSI General Working Group (X3T9.2/89-090) 9. Old Business 9.1 Cable Testing Results (Tuesday a.m.) 9.2 Connector Vendor Statements as to Patent Intentions (Tuesday a.m.) 9.3 Forwarding Motion on SCSI-2 10. New Business 10.1 Single-cable 16-bit SCSI (X3T9.2/89-94 Rev 1) 10.2 CD-ROM Read Subchannel Issue (X3T9.2/89-95, -98) 10.3 Editing the SCSI-3 document 10.4 Review of new documents 10.5 Agenda for the Oklahoma City Working Group 10.6 SCSI Requirements for Fiber Channel 11. Review of Action Items 12. Meeting Schedule 12.1 General Working Group Schedule for 1990 13. Adjournment Minutes -- X3T9.2 Meeting #83 August 21-22, 1989 -- Colorado Springs, CO 1. Opening Remarks John Lohmeyer, the chairman, called the meeting to order at 9:00 a.m., Monday August 21, 1989. He thanked John Morse and Del Shoemaker of Digital Equipment Corp. for hosting the meeting and he also thanked Classic Conferences for arranging the meeting. The chairman announced that X3T9 had forwarded SCSI-2 Revision 10 to X3 at its last meeting. After the SPARC compliance review had been completed, Del Shoemaker realized that the X3T9 forwarding motion had actually failed due to the large number of absent members. Rather than issue a letter ballot, which would close after our August meetings, Del elected to try again at the August meeting. This created an opportunity to revise the document in an attempt to resolve the Sony negative ballot. (See agenda item 9.3.) As is customary, the attendees introduced themselves. A copy of the X3T9.2 membership list was circulated for attendance and corrections. Copies of a revised draft agenda and the recent document register were made available to all attendees. Information on X3T9.2 and Mailing Subscription forms were made available for new attendees. 2. Approval of Agenda The revised agenda was approved. 3. Attendance and Membership The membership requirements were reviewed. X3 rules permit one vote per organization. An individual from a new organization must attend one plenary meeting as an observer before he/she may apply for voting membership via a letter to the chairman of X3T9.2. The individual may vote at the second plenary meeting. Working group meetings do not count toward attendance requirements. The chairman stated that since the last meeting the following changes had occurred in the voting membership: Mr. Robert B. Mason and Mr. Nick Seroff are the new principal and alternate for Flexstar, Inc. Mr. Orrin Char has replaced Mr. D. Bethune as alternate from Intergraph Corp. During the meeting the following membership changes were received: Laura Shellhamer replaced John Mangan as the Miniscribe principal member. No replacement alternate has been selected. Mark O'Dell replaced Ed Carmona as the alternate for Distributed Logic Corp. Rudy Stalzer and Douglas Hagerman became additional alternates for Digital Equipment Corp. The net effect of these changes plus the two terminations for non- attendance, below, is that X3T9.2 now has 77 voting organizations down from 78 at the last meeting. Enclosure (1) is the list of attendees at the meeting. Enclosure (2) is the X3T9.2 membership changes since the last minutes and enclosure (3) is the current X3T9.2 membership list. 3.1 Roll Call of Members in Jeopardy The chairman stated that he had sent jeopardy letters to all representatives of the following organizations: Intergraph Corp., Laserdrive Ltd., and Optotech, Inc. The letters said that these organizations had failed to attend two of the last three plenary meetings and that their membership would be terminated unless they attended the June plenary meeting. Optotech was present, but Intergraph Corp. and Laserdrive Ltd. were not present. The principals and alternates from these later organizations were changed to observer status. 3.2 1990 - 1991 Service Fees (X3T9.2/89-87) The 1990 and 1991 Service Fees have been announced. The fees for participating in X3T9.2 will be $250 per year for a principal (including one alternate). Additional alternates and observers are billed $250 per year. The Subscription Fee to receive mailings is in addition to these fees and will be announced for 1990 later this year. The Service Fees are the same ($250) for each Technical Committee and Task Group. The Subscription Fees vary depending on the expected size of the subgroup's mailings and whether the subscriber is overseas. 4. Approval of Minutes - June Meeting San Jose, CA (X3T9.2/89-75) The June minutes were approved with the following corrections: Item 1: The June meeting actually start on June 19 and not on April 24. Item 10.2.6: The words "...to avoid an incorrect initiator connection..." will be inserted in the second paragraph after "...MESSAGE OUT phase...". Attendance List: James McGrath of Quantum attended the meeting. The June minutes will be updated on the SCSI Bulletin Board, but not redistributed. 5. Document Distribution The last mailing included both SCSI-2 Revision 10 and ESDI Revision 3A. The chairman had not received any complaints about the quality of the mailing. Members with mailing subscription problems should deal directly with Katrina Gray of the X3 Secretariat. She can be reached at (202) 626-5741. John Lohmeyer requested that all documents for the September mailing be sent to him by September 11, 1989. The following new or revised documents were distributed and/or discussed at the meeting: Document Doc Date Author Description of Document ------------- -------- --------------- --------------------------------------- X3T9.2/86-109 8/18/89 Too Many Draft of extended Small Computer System Rev 10a Interface, "SCSI-2" X3T9.2/89-4 8/21/89 P. Boulay Proposed Single-Ended SCSI Bus Rev 2 Terminator X3T9.2/89-101 8/4/89 G. Penokie Maximum Burst Size Rounding Definition X3T9.2/89-102 8/7/89 D. Allan Correction to ESDI document X3T9.2/89-103 8/7/89 T. Wicklund Comments on SCSI-2 Rev 10 X3T9.2/89-104 8/9/89 A. Hospodor Ltr. to J. Lohmeyer re X3B7.1 Activities X3T9.2/89-105 8/15/89 J. Lohmeyer Ltr to A. Hospodor re his August 9 letter (see 89-104) X3T9.2/89-106 8/9/89 B. Mattox Request to re-consider ribbon-style contacts for high-density connector X3T9.2/89-107 8/10/89 K. Kumazawa ESDI Synchronous Spindle Configuration Parameter X3T9.2/89-108 8/10/89 R. Lin Sony Proposal for CD-ROM Device Rev 1 X3T9.2/89-109 8/17/89 G. Houlder New COPY Segment Descriptors for Wide SCSI-3 X3T9.2/89-110 8/17/89 G. Houlder Target Initiated SDTR Message Control X3T9.2/89-112 8/17/89 R. Snively Comments on 16-bit Single-Connector Proposal X3T9.2/89-113 8/18/89 L. Lamers ESDI Rev. 3A Clarification X3T9.2/89-114 8/17/89 P. Boulay SCSI-2 CD-ROm Command Set Changes X3T9.2/89-115 8/17/89 Chan/ Matheson Alternative 2 Termination: Worst-Case Analysis and Recommendations X3T9.2/89-116 8/19/89 Chan/ Matheson Shielded Cable Evaluation X3T9.2/89-117 8/21/89 B. Homans Standard INQUIRY Data Format X3T9.2/89-118 8/18/89 R. Cummings An Introduction to/update on the Fiber Channel The current document register is enclosure (4). 6. Liaison Reports 6.1 ISO Gene Milligan, the X3T9 International Representative, gave the following report: The SCSI-2 Revision 10 has been officially submitted to ISO. Gene Milligan, the X3T9 IR, has been granted permission to host an SC 13 working group meeting to exclusively address SMD-E comments. An invitation to such a meeting would only be extended if we are unable to resolve the SMD issues prior to the end of the SC 13 meeting. No U.S. responses have been received relative to the letter ballot comments which could potentially impact the SMD-E ANS. The FDDI PMD DIS 9314-3 ISO letter ballot passed with comments. A 19 May letter from Ross proposed resolution of the PMD DIS letter ballot comments. However Germany has disagreed with some of the proposed resolutions. An ISO letter ballot has been issued on the ESDI DIS 10222 which will close 12/89. A recommendation is due from X3T9 prior to 10/06/89. The IPI-2 Tape editor George Reichenberg has submitted a camera ready copy to ISO. ISO style versions of IPI-3 Tape will be submitted to Sc 13 upon approval by X3T9.3. An unchanged and a version with technical changes are both available. There was an ISO letter ballot questionnaire on the issue of whether IS 6548 Process I/O should be reaffirmed. The IR suggests that T9 should be prepared to respond to any questions as to why the U.S.A. failed to submit a vote. With the matrix of questions and the small number of votes it is somewhat difficult to judge the results. A camera ready copy is still required for the following: Document: Description: Editor: IS 9317 Tape Streaming Cart/Cas Drive Domshy IS 9318-2 IPI-2 Disk Wire IS 9318-3 IPI-3 Disk Truestedt An ISO style copy is still required for the following: Document: Description: Editor: DIS 9318-6 IPI Enhanced Physical Sanderson For those interested in additional references, Mr. Milligan has provided a document list of all current documents relating to SC 13 or known to be relating to X3T9's international responsibilities. He has also provided a cryptic summary of SC 13 projects. An SC 25 TAG meeting will be held after the X3T9 meeting p73 Friday 8/25/89. Delegates are requested to attend. The TAG plans to review all the proposed agenda items. 6.2 IT8 At the last meeting John Lohmeyer reported that he had received a request from ASC IT8 for two device type codes. There was some debate over why two codes were needed and John accepted an action item to call the IT8 secretary to learn more about their request. He had spoken to the IT8 officers: William K. Smythe, Robert Strum, and Chris Goldsmith. Presently, IT8 has a draft standard which is just about to enter its first public review period (see X3T9.2/89-77). It references SCSI-1, but includes some SCSI-2 features. They used the SCSI-2 scanner command set as a model for their command set. They probably only need one device type code at present, but they expect to define further graphic art/prepress devices in the near future. They may need a new device type code for color input devices or they may wish to have a separate identity for their host devices to avoid confusion with more conventional host adapters. John proposed that device type codes 0Ah and 0Bh be assigned as "Reserved for use by ASC IT8". If IT8's standard is approved before SCSI-2, these code values could be further identified during the final editing. This proposal was accepted as part of SCSI-2 Revision 10b. 6.3 SCSI-2 Common Access Method Committee Report Dal Allan reported. There have been two CAM committee meetings and one ATA Working Group held since the last plenary meeting. There has been a notable lack of success in getting a CAM working document on the table. The committee has changed direction twice; after the last SCSI working group meeting a decision was made to split into a specific operating system and platform orientation for each CAM module, which was reversed at Costa Mesa to return to a common structure for all CAM modules. At Costa Mesa there were five proposals presented which were very close in philosophy. An agreement was reached on a data structure to be passed from the operating system to the SIM layer. The content of the data structure is specific to each operating system. An optional host bus adapter interface structure which is less dependent on the specific operating system was agreed to in principle. There is now a need to develop a basic document to coalesce into a proposed standard. A description to form the basis of a working document will be prepared in September. At the Oklahoma City meeting, Microsoft and IBM are planning to make a joint presentation on OS/2, and IBM will be making an "architectural exploration" of a DOS BIOS for SCSI-attachment. The ATA (AT-Bus interface) document has been revised and is in the latest CAM committee mailing. The working group has made significant progress on clarifications but three areas still need clarification: the DASP and PDIAG signals, a definition of command acceptance, and timeouts. The EATA portion of the document had no changes. A proposal is expected from the quarter-inch cartridge people on using EATA to support tape drives. 6.4 Fiber Channel Dal Allan reported using a presentation developed by Roger Cummings of StorageTek (see document 89-118R0). He commented that there still is a lack of participation by companies in the SCSI committee on the Fiber Channel project and as a result the preponderance of High Speed Channel attendees may skew the services and functions away from some of the needs that SCSI users have. The justification for going to a Fiber Channel interface is that there is a market demand to place I/O devices further apart with faster data transfer rates. These requirements precipitate a change in architecture. Fiber Channel is intended to be a new physical interface, able to transport SCSI, IPI, and HSC commands. Two layers are used for framing, protocol, encoding, and routing. See also item 10.6 of these minutes for further information on Fiber Channel. 6.5 X3B7 Report John Lohmeyer reported that Dal Allan had arranged with some X3B7 members to provide X3T9.2 with a draft proposal on SCSI device testing in time for the September mailing. Dal informed these people that X3T9.2 would schedule time to review this proposal at the October working group meeting. John received a letter (X3T9.2/89-104) from Andy Hospodor, technical editor for the X3B7.1 Diagnostic Command Set document. In this letter, Andy welcomed interested parties to attend the X3B7.1 meeting August 21- 23 in Phoenix. He also mentioned that there are future X3B7.1 meetings scheduled for October 1989 and February 1989 [sic] in the Phoenix area. Andy complained about the June X3T9.2 minutes and the July working group minutes. He did not like the statements that "Disk companies have not been strong participants in ... X3B7" and that "X3B7 is basing its proposal on SCSI-1, not SCSI-2." He listed several disk drive companies that have representatives on X3B7.1 and he stated that their document references X3T9.2/86-109, SCSI-2, Revision 10. John said that he had responded to Andy in a letter (X3T9.2/89-105). The letter invited X3B7.1 members to attend upcoming X3T9.2 meetings. It also stated that the minutes accurately reflected the statements made in X3T9.2 meetings. John read Andy's letter and said he would distribute the letter in the next X3T9.2 mailing. Dal Allan read a statement regarding this issue, defending previous statements he made at the SCSI meetings. Dal said he receives the minutes of X3B7.1 and that the May minutes showed the only drive manufacturers present were DEC and HP. The HP representative is principally involved with test equipment engineering and not drive manufacturing. The intent of these statements is not to tear down the X3B7.1 activity, but to point out that they may lack the necessary participation to develop a standard that would actually be followed. 7. Review of Old Action Items 1. The X3T9.2 connector vendors will prepare statements concerning their business intentions on SCSI-2 connector patents and patent applications. This action item was previously completed by AMP and Fujitsu. The other connector vendors have not submitted statements. While the committee was still interested in receiving such statements, the action item will be dropped. 2. John Lohmeyer will get clarification on the device type request from IT8. Complete. 3. Larry Lamers and John Lohmeyer will prepare SCSI-2 Revision 10. Complete. 4. Dal Allan will prepare ESDI Rev 3a and send it to John Lohmeyer. Complete. 5. John Lohmeyer will request X3T9 to substitute ESDI Revision 3a for ESDI Revision 3 to resolve the Maxtor public review comment without substantive changes. Complete. 6. John Lohmeyer will transmit SCSI-2 Revision 10 to X3T9 for further processing. Complete. 8. Working Group and/or Project Status Reports 8.1 ESDI (Tuesday p.m.) ESDI Revision 3A is currently in an X3 letter ballot for final approval. The letter ballot closes September 20. 8.1.1 Editorial Error in ESDI Rev 3A (X3T9.2/89-102) There was an editorial error in a field in ESDI that specifies whether the device is a master, master control, slave or unsynchronized. This is a two-bit field. When the unsynchronized value was added to this field, it was supposed to have the high-order bit set only. Unfortunately, the editor reverted to hexadecimal and made this value 80h. It should have been 02h. A few organizations at the meeting indicated that they had begun to implement the error, but they were willing to switch. (See the next agenda item.) 8.1.2 Maxtor Issue on Synchronized Spindles Maxtor submitted a letter (89-113) requesting that the synchronized spindle portion of ESDI be clarified. In a few cases, there was apparent confusion over the terms "master" and "master control". Dal Allan led a discussion over how these two options are different. Part of the confusion also stemmed from ambiguity over whether some terms referred to the device or the signal. During the course of the discussion, it was also noted that the SCSI and ESDI documents differ in that ESDI only reports the transition from synchronized to unsynchronized. SCSI reports both transitions. The attendees preferred the ESDI method and may make a proposal to change SCSI-2 at a later date. [Chairman's note: At this point in the meeting, SCSI-2 Rev 10b had already been forwarded.] Gene Milligan moved and Sam Karunanithi seconded that the ESDI technical editor be directed to make changes to the ESDI document during the final edit to reflect the following clarifications: 1) The master control signal is not necessarily tied to index. It is a signal generated by the drive designated as the master control and sent to the controller which may use it in synchronizing drives. 2) A drive designated as a master control shall synchronize to the slave sync signal. A drive designated as a master control or a slave shall accept rotational position offsets. 3) A drive designated as a master generates a slave sync signal at its index. It does not synchronize to the slave sync signal (a output only). A drive designated as a master does not accept a rotational position offset. 4) Add the following clarification to page 36: Attention shall be asserted whenever there is a change from the synchronized condition to the unsynchronized condition (from bit 11 = 1 to bit 11 = 0). 5) Change the value of unsynchronized from 80h to 02h. The motion passed unanimously. 8.2 SCSI General Working Group (X3T9.2/89-090) John Lohmeyer reported that a working group meeting had been held July 10-11, 1989 in Chicago with 31 people present. Joe Lawlor of AT&T hosted the meeting. The meeting principally focused on the long deferred SCSI-3 issues. Several proposals were so dated, that the authors withdrew them. Some of these will be updated and re-submitted; others will probably be forgotten. There was considerable interest in the single-cable 16-bit SCSI proposal that would used a single 68-position connector (X3T9.2/89-94). There was also a discussion of the Sony CD-ROM issue. Paul Boulay gave a presentation explaining Sony's issue for the non-CD-ROM experts and suggested other alternatives to Sony's preferred solution. (For more information on this issue, see agenda item 10.2, below.) Robert Kellert disagreed that there was a clear yes on the question of whether a 16-bit alternative implementation can be done in a single chip. (See document 89-90, item 11, paragraph 3.) 9. Old Business 9.1 Cable Testing Results (Tuesday a.m.) Kurt Chan presented results of testing shielded cables (see document 89- 116). There still is a need for more data from the cable manufacturers to distill meaningful information from the test results. Kurt requested that committee members review his analysis of terminator power (see document 89-115). Discussion on this will occur at the next working group meeting. Bob Bellino of Madison Cable asked if the SCSI cable requirement of 90 ohms minimum and 132 ohms maximum is realistic. Their test on a Time Domain Reflectometer for 127 +/- 5 ohm cable indicated that on a single- ended system, the resulting impedance is 81 ohms. He felt the problem is a result of a lack of clarity in specifying how impedance is to be tested. He asked the committee to improve this specification. The committee responded in its usual manner. Bob Bellino will organize a special working group to study this issue for round cables. It will meet on Monday October 16 starting at 1:00 p.m. in parallel with the X3T9.2 plenary meeting in Raleigh, NC. Those interested in participating may contact Bob at (508) 752-7320. 9.2 Connector Vendor Statements as to Patent Intentions (Tuesday a.m.) The committee has received such statements from AMP and Fujitsu. This item will be dropped due to the lack of intentions on the part of the other connector vendors. 9.3 Forwarding Motion on SCSI-2 The chairman said that although X3T9.2 and X3T9 had both forwarded SCSI-2 Rev 10 at their June meetings, the X3T9 vote had actually failed mostly due to the large number of absent members. When Del Shoemaker realized the error, he requested that the X3 Secretariat hold up the public review notice until X3T9 could re-ballot the forwarding motion. X3T9 is planning to do so at its August 25 meeting. The X3T9 vote failed by one vote, 12 Yes, 1 No, 0 Abstain, and 12 Absent. Sony cast the No vote, saying that they had some issues with the CD-ROM section, but did not have a proposal ready. Since SCSI-2 had incurred a delay, John Lohmeyer and Paul Boulay prepared a revision 10a document in hopes of solving the Sony issues and incorporating several minor changes that have been identified since the revision 10 document was prepared. Changes had been identified from several sources: 1. Informal comments received at the July working group meeting. 2. Tom Wicklund's review of Rev 10 (89-103). 3. Tom Wicklund's further comments (FAX to J. Lohmeyer Aug. 16) 4. Sony CD-ROM Proposal (89-108) 5. Four messages from Greg Leonhardt received on the SCSI BBS. 6. Requests from several vendors for Vendor IDs. 7. A phone call from Bill Homans re the Device-Type Qualifier field. 8. A phone message from Bill Spence on unit attention. 9. A phone call from Gerry Houlder on the ABORT message. 10. An email series of messages with Kurt Chan on the IDENTIFY fields. 11. A phone call from Gene Milligan on clearing contingent allegiance. 12. George Penokie's proposal on maximum burst size rounding (89-101). 13. Ken Post (and another person) pointed out an error in appendix B. John and Paul had copies of the changes available at the meeting. The committee reviewed these changes and made several more changes, designating the final document Revision 10b. Paul Boulay moved and Rita Lin seconded that SCSI-2 Revision 10b, which consists of the Revision 10a document plus the changes made during the meeting be forwarded to X3T9 for further processing. A role call ballot was taken with the following results: Organization Name Yes No Abstain Absent ------------------------- ------------------------ --- -- ------- ------ 3M Company Mr. Wayne E. Roen 1 ACT Technology 1 AMP, Inc. Mr. Bob Whiteman 1 AT&T 1 Adaptec, Inc. Mr. Al Wilhelm 1 Advanced Micro Devices Mr. Nicos Syrimis 1 Amphenol Corp. Mr. Michael Wingard 1 Ancot Corp. Mr. Jan V. Dedek 1 Apple Computer 1 Archive Corp. 1 Burndy Corp. 1 Cinch Connector Mr. John Guennewig 1 Cipher Data Products, Inc Mr. Bharat Shah 1 Cipher/Optimem 1 Ciprico Inc. Mr. Tom Wicklund 1 Cirrus Logic Inc. Mr. Robert Kellert 1 Congruent Software, Inc. 1 Cygnet Systems 1 DDC Pertec 1 DPT 1 Data General Corp. 1 Datacopy Corp. 1 Digital Equipment Corp. Mr. Rudy Stalzer 1 Distributed Logic Corp. Mr. Duncan McDonald 1 ENDL Mr. I. Dal Allan 1 Eastman Kodak Co. 1 Emulex Corp. Mr. Chuck Micalizzi 1 Exabyte Corp. 1 Flexstar, Inc. Mr. Robert B. Mason 1 Fujitsu America, Inc. Mr. Robert Liu 1 Future Domain Mr. Kenneth Post 1 Hewlett Packard Co. Mr. Kurt Chan 1 Hirose Electric U.S.A. 1 Hitachi Cable Manchester Mr. Fielding S. Tabb 1 Honda Connector Mr. David McFadden 1 IBM Corp. Mr. George Penokie 1 ICL 1 IMPRIMIS/CDC Mr. Gene Milligan 1 ITT Cannon 1 Interphase Corp. 1 Iomega Corp. Mr. Geoff Barton 1 JAE 1 LMS-OSD Mr. Paul Boulay 1 Madison Cable Corp. Mr. Robert Bellino 1 Maxtor Corp. 1 Methode Electronics, Inc. Mr. Bob Masterson 1 Micropolis Corp. Mr. Stephen Cornaby 1 Organization Name Yes No Abstain Absent ------------------------- ------------------------ --- -- ------- ------ Miniscribe Corp. Ms. Laura Shellhamer 1 Mitsubishi Electronics Am Mr. John Spongr 1 Molex Inc. Mr. Craig Bixler 1 NCR Corp. Mr. John Lohmeyer 1 NEC I.S. 1 Nat. Inst. of Stnds. & Te 1 National Semiconductor Mr. James Schuessler 1 Northern Telecom, Inc. Mr. David Wehrman 1 OTL Engineering 1 Olivetti 1 Optotech, Inc. Mr. Jim Harper 1 Panasonic Industrial Co. Mr. Charles Yang 1 Panduit Corp. Mr. Ed Cieniawa 1 Priam Corp. 1 Quantum Corp. 1 Ricoh Corporation 1 Scientific Micro Systems 1 Seagate Technology Mr. Mike Eneboe 1 Siemens Info. Systems, In 1 Sony Corp. of America Ms. Rita Lin 1 Stewart Connector Systems 1 Storage Technology Corp. Mr. Dennis Appleyard 1 Sun Microsystems, Inc. 1 Tandem Computer Inc. 1 Technology Forums, Ltd. 1 Texas Instruments Mr. D. W. Spence 1 Thomas & Betts Mr. Harvey Waltersdorf 1 UNISYS Mr. Peter Dougherty 1 Western Digital Mr. Jeff Stai 1 Zenith Data Systems Mr. Greg Leonhardt 1 Totals: 44 0 0 33 The motion passed: 44 Yes, 0 No, and 33 Absent. Gene Milligan suggested that we mail the changed pages to ISO. X3T9.2 requested that X3T9 attempt to get the ISO document synchronized with Revision 10b. 10. New Business 10.1 Single-cable 16-bit SCSI (X3T9.2/89-94 Rev 1) There was a discussion on Gerry Houlder's proposed fixes for the COPY command (89-109) to allow increased addressability. We may wish to make a further change in anticipation of Fiber Channel. Jeff Stai commented that we should make conscious decision on whether 8- bit devices should be permitted on 16-bit cables. If so, we should insure that signal skew issues are addressed. George Penokie took an action item to develop the alternative pin assignments for presentation at the Oklahoma City working group meeting. A special working group was appointed to study the issue of defining a transceiver for SCSI-3 that provides both high performance and can be integrated into protocol chips. This could result in a high-current single-ended transceiver or a low-power differential transceiver. Gary Murdock of National Semiconductor (408-721-7269) will organize the group and they will meet Wednesday evening during the Okalahoma City SCSI working group meeting. 10.2 CD-ROM Read Subchannel Issue (X3T9.2/89-95, -98, -108) At the Chicago working group meeting, Rita Lin submitted 89-98 explaining why Sony voted No on the SCSI-2 forwarding motions at X3T9.2 and X3T9. At the same meeting, Paul Boulay submitted 89-95 further explaining the Sony issue for those people who were not experts on CD-ROMs. He also outlined alternative solutions. On August 7, Mike Yokoyama uploaded 89-108 to the SCSI Bulletin Board containing an re-edited Section 13 with the wording changes that Sony felt would resolve their issue. Rita Lin sent a revised version of this proposal to John Lohmeyer several days later via facsimile (89-108R1). The differences appeared to be confined to the explanation of the proposal. Paul Boulay, the Section 13 editor, agreed to edit the Sony proposal into the appropriate style and to review it for consistency with other portions of the document and acceptability with LMS. In doing this task he found a few exceptions. He sent a facsimile of these to Sony and prepared Section 13 of the Revision 10a proposal. The revised proposal was acceptable to Sony and was included in the SCSI- 2 Revision 10b document. 10.3 Editing the SCSI-3 document At the Chicago working group meeting, the question of whether X3T9.2 could find a way to hire a professional editor (either for SCSI-2 or SCSI-3) was raised. Several issues were raised concerning this suggestion included who would judge the quality of the editor's work and how could X3T9.2 collect the funds and distribute them. John Lohmeyer agreed to discuss the financial issue with Del Shoemaker. Del's initial reaction was that the X3 Secretariat could probably arrange an escrow account for this purpose much like the escrow account that was used to host SC-13 meetings. In that case, X3T9 members voluntarily contributed to the account and the money was collected prior to authorizing X3T9 to host the SC-13 meeting. During the X3T9.2 meeting discussion continued on this topic. It was "guesstimated" that hiring an editor to convert SCSI-2 to ISO style might cost $30,000. This editor would not need to be a technical expert on SCSI-2 and would not need to attend committee meetings. If we were to hired an editor for SCSI-3, he/she would have to be technically competent on SCSI and would have to attend the meetings. Hiring such a person and paying for travel expenses might cost $150,000 per year. 10.4 Review of new documents Target Initiated SDTR Message Control (89-110R0) [Gerry Houlder] There was considerable debate over this proposal. It would add a bit to the control mode page to specify whether the target is permitted to initiate an SDTR exchange. It became apparent that there was enough opposition that the proposal could not be included in SCSI-2 Rev 10b. The proposal was remanded to the working group for further discussion. The other new documents were discussed under other agenda items. 10.5 Agenda for the Oklahoma City Working Group The following draft agenda was defined: 1. Additional SEND DIAGNOSTIC Pages (87-186R0) [Spence] 2. More Than 8 Devices On Wide SCSI (87-207) [Harms] 3. Search Command Modifications (88-002) [Stai] 4. Autoconfiguration 5. Autosense 6. Request For Data DAT Device Type [Andrews] 7. Documentation Layering [Stephens] 8. Single-Cable 16-bit Wide SCSI [Penokie, Lohmeyer, Lamers] 9. 16-bit Everything Proposal (89-97) [Houlder] 10. Alternate Physical Layers (e.g., fiber optics) 11. LOGICAL UNIT RESET Message Proposal (89-058) [Lohmeyer] 12. Arbitration Fairness (89-61) [Penokie] 13. Arbitration Unfairness (89-65) [Buesing] 14. Terminator power analysis (89-115) [Chan] 15. Target Initiated SDTR Message Control (89-110) [Gerry Houlder] 10.6 SCSI Requirements for Fiber Channel This item was added to the agenda after the mailing went out. The July 17-18 Fiber Channel working group of X3T9.3 requested X3T9.2 to provide a list of Fiber Channel requirements for SCSI. John Lohmeyer submitted such a list prepared by NCR as a starting point for X3T9.2 to prepare list for the Fiber Channel working group. The NCR list was as follows: NCR Requirements for using SCSI over Fiber Channel Maximum number of devices: 256 Transfer rate: 400 - 1000 Mbit/sec Cost/node: $100 - 250 Maximum distance between nodes: 1 km Connector Physical Size: Smaller than SCSI-2 High-Density Protocol: Datagram with SCSI CDB Frame Size: 40% < 64 bytes 60% 1 - 4 kbytes Error Detection: Required Detected Error Rate: < 1 error in 10E12 bits transferred Undetected Error Rate: < 1 error in 10E18 bits transferred Error Correction: Not required (will use re-transmission) Features: Live insertion of devices. Fiber Optic link to a box containing more than one device -- electrical links within the box (possibly parallel). This list was revised to reduce the low-end transfer rate to 200 Mbit/sec. Dal Allan moved and Jeff Stai seconded that the revised list of requirements above be sent from X3T9.2 to the Fiber Channel working group. The motion passed unanimously. John Lohmeyer accepted an action item to revise the list of requirements and submit them to the Fiber Channel working group. [Chairman's note: During the Fiber Channel meeting, one other feature was identified. Autoconfiguration was added to the list as it is associated with the live insertion of devices.] 11. Review of Action Items 1. Larry Lamers and John Lohmeyer will prepare SCSI-2 Revision 10b. 2. John Lohmeyer will revise the list of fiber channel requirements from X3T9.2 and submit them to the fiber channel working group. 3. Dal Allan will clarify the ESDI document and send an electronic copy to John Lohmeyer. 4. John Lohmeyer will transmit SCSI-2 Revision 10b to X3T9 for further processing. 5. George Penokie will develop the two alternatives on 16-bit SCSI pin- outs for presentation at the Oklahoma City working group. 12. Meeting Schedule The next meeting of X3T9.2 will be October 16-17, 1989 at the Howard Johnson Hotel and Conference Center (919) 782-4433 in Raleigh, North Carolina hosted by IBM. Please mention X3T9/IBM when making reservations to receive the special room rate of $66.28 single and $71.68 double, tax inclusive. The cut-off date for reservations is September 18, 1989. The plenary meeting schedule is: Date Location Host -------------------- ------------------------ --------------------- December 4-5, 1989 San Diego, CA NCR February 19-20, 1990 Austin, TX Motorola April 23-24, 1990 St. Petersburg Beach, FL AMP (*) June 18-19, 1990 Wichita, KS NCR August 20-21, 1990 Seattle, WA Boeing October 15-16, 1990 Valley Forge, PA UNISYS December 3-4, 1990 Southern CA ?? February 18-19, 1991 Dallas, TX EDS April 22-23, 1991 St. Petersburg, FL AMP June 17-18, 1991 Minneapolis, MN Stanatek August 19-20, 1991 Vancouver, BC tbd or San Diego, CA Western Digital October 14-15, 1991 Atlanta, GA AT&T December 9-10, 1991 Phoenix, AZ DEC (*) The location or date may change. The usual hotel is not available. 12.1 General Working Group Schedule for 1989 The next SCSI working group meeting was scheduled for September 6-7, 1989 in Oklahoma City, OK by Gene Milligan of Imprimis. The hotel is the Sheraton Century Center Hotel, One North Broadway, Oklahoma City, OK 73102 (405 235-2780). The room rates are $55 single and $60 double, plus tax. See X3T9.2/89-083 for more information. A SCSI CAM meeting will be held on September 8 in the same hotel. The SCSI Working Group meeting schedule for the remainder of 1989 is: Date Location Host Note -------------------- ----------------- ------------- ------------------ Oct 30-31, 1989 Irvine Ken Post See X3T9.2/89-100 The following weeks were selected. There was some debate over which days of these weeks should be assigned to the working group meetings. Other days in the same weeks will likely be assigned for Fiber Channel working group meetings and CAM meetings. Dal proposed that Monday be assigned for the CAM group, Tuesday and Wednesday be assigned for SCSI, and Thursday and Friday be assigned to Fiber Channel. Bill Spence indicated that he preferred that SCSI remain on Monday and Tuesday. The decision on this issue was deferred until the October meeting. John Lohmeyer request that potential hosts contact him. Date ---------------------- Jan 8-12, 1990 Mar 5-9 or 12-16, 1990 May 7-11, 1990 (*) Jul 9-13, 1990 Sep 5-9, 1990 Oct 29-Nov 2, 1990 (*) This date may be impacted by the April meeting problem, above. 13. Adjournment The meeting was adjourned at 3:00 p.m. on Tuesday August 22.