Final Agenda -- X3T9.2 Meeting #80 February 20-21, 1989 -- Austin, TX 1. Opening Remarks 2. Approval of Agenda 3. Attendance and Membership 3.1 Roll Call of Members in Jeopardy 4. Approval of Minutes - December Meeting San Diego, CA (X3T9.2/88-164) 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 7. Review of Old Action Items 8. Working Group and/or Project Status Reports 8.1 ESDI 8.1.1 ESDI Synchronized Sector Offset (X3T9.2/89-023) 8.2 Flexible Disk 8.3 SCSI General Working Group (X3T9.2/89-019) 9. Old Business 9.1 Proposed alternative single-ended terminator (X3T9.2/89-4) 9.2 Review of SCSI-2 Draft Document (X3T9.2/86-109 Rev 7) 9.2.1 Vendor ID Registration 9.2.2 Required sense data: 8 or 18 bytes? 9.2.3 Combine ASC and ASCQ into single 16-bit field? 9.2.4 Should field names be first letter caps? 9.2.5 Should REQUEST SENSE command failure clear sense data? (89-024) 9.2.6 SEARCH DATA command transfer length inconsistency (89-025) 9.2.7 READ DEFECT DATA command [Gerry Houlder] 9.2.8 Comments on SCSI-2 (89-30, 31, 32, 33, 34, 35, 37) 9.3 Cable Testing Results (X3T9.2/89-18) 9.4 TERMINATE IMMEDIATE message re-visited (X3T9.2/88-158 Rev 1, 89-7) 10. New Business 10.1 Tape READ underlength detection (X3T9.2/89-021) [Penokie] 10.2 SCSI-2 Forwarding motion (X3T9.2/86-109 Rev 7) 10.3 ESDI Forwarding motion (X3T9.2/87-105 Rev 3) 10.4 Review of new documents 10.5 Agenda for the Milpitas Working Group 11. SCSI-3 Activities 11.1 Autoconfiguration SSWG 12. Review of Action Items 13. Meeting Schedule 13.1 General Working Group Schedule for 1989 13.2 Methods for Support of Future Plenary Meetings 14. Adjournment X3T9.2 Meeting #80 February 20-21, 1989 -- Austin, TX 1. Opening Remarks John Lohmeyer, the chairman, called the meeting to order at 9:00 a.m., Monday February 20, 1989. He thanked Rhonda Alexis-Dirvin of Motorola for hosting the meeting and he also thanked Classic Conferences for arranging the meeting. As is customary, the attendees introduced themselves. A copy of the X3T9.2 membership list was circulated for attendance and corrections. Copies of the draft agenda and the recent document register were made available to all attendees. Information on X3T9.2 and Pilot Subscription forms were made available for new attendees. 2. Approval of Agenda The draft agenda was approved with the following additions: 9.2.7 READ DEFECT DATA command [Gerry Houlder] 9.2.8 Comments on SCSI-2 (89-30, 31, 32, 33, 34, 35, 37) 13.2 Methods for Support of Future Plenary Meetings 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. Jim Erickson replaced Mr. Ray Calvin as the Hirose principal. Mr. Oscar Kornblum replaced Mr. Michael Kosmala as the Hirose alternate. Mr. Glen Nance replaced Mr. Tony Lo as the Quantum alternate. Ms. Rita Lin replaced Mr. Lawrence Lamers as the Sony Corporation principal. Mr. Simon Kustedjo replaced Mr. Masayuki Yokoyama as the Sony alternate. The following voting membership changes occurred during the meeting: Apple Computer has been reinstated as a voting member of X3T9.2. Mr. Harlan Andrews will be the principal and Mr. Jerry Katzung will be the alternate. Mr. Lawrence Lamers replaced Bob Mortensen as the Maxtor principal. ACT Technology was reinstated with Mr. Nathan Lee designated as the principal and Mr. Jerry L. Amstutz as the alternate. Exabyte Corporation became voting members with Mr. Michael Braitberg named the principal and Mr. Bill Duran named the alternate. DDC Pertec Corporation became voting members with Mr. Rick Ehrecke named the principal and Mr. Mike Racelo named the alternate. At the end of the meeting, the following changes were made to the AT&T representation: Mr. James Semenak was replaced by Mr. Joe Lawlor as principal and Mr. David Harms was replaced by Mr. Christopher Phea as alternate. Jim Semenak has taken early retirement from his SCSI editing job; X3T9.2 thanks Jim for his dedication in editing Section 9 and later in editing Sections 7 and 8. 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: Datacopy Corp., DEST Corp., ELCO Corp., Hirose Electric U.S.A., Honeywell Bull, Inc., ICL, JAE, Maxon Systems, Inc., Northern Telecom, Inc., Panduit Corp., Priam Corp., Ricoh Corporation, Storage Technology Corp., Viking Connectors, and Winchester Electronics. 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 February plenary meeting. Subsequent to sending out the letters, Ed Cieniawa of Panduit called to inform the chairman that he had attended the October 1988 meeting. It was also pointed out to the chairman that Honeywell Bull and Bull Systems are the same organization and therefore are not in jeopardy. The chairman has corrected the errors in his database. Representatives were present from all of the organizations except: DEST Corp., ELCO Corp., Maxon Systems, Inc., and Winchester Electronics. These organizations were terminated, that is changed to Observer status. In the case of DEST, where all participants have left the company, the organization was dropped from the committee lists. Winchester Electronics has elected to not participate in 1989 and was dropped from the committee lists. 4. Approval of Minutes - December Meeting San Diego, CA (X3T9.2/88-164) The minutes were approved with no changes. 5. Document Distribution The last mailing was 858 pages. The X3 Secretariat asked the copying service to put the 1989 Pilot Subscription Form in front of the mailing. They did, causing all the front and back pages to be reversed. The chairman expressed his regrets for the inconvenience. The task of maintaining the subscription database for the X3T9 groups has now been given to Katrina Gray, Standards Secretary, of the X3 Secretariat. Members with subscription problems should deal directly with Ms. Gray who can be reached at (202) 626-5741. John Lohmeyer requested that all documents for the March mailing be sent to him before March 15, 1989. Starting now it is a new year and subscriptions must be renewed to continue getting the documents. The document subscription fee for 1989 is $110 U.S.A. and $220 overseas. The following new or revised documents were distributed and/or discussed at the meeting: Document Doc Date Author Description of Document ------------- -------- --------------- --------------------------------------- X3T9.2/89-4 1/30/89 P. Boulay Proposed Single-Ended SCSI Bus Rev 1 Terminator X3T9.2/89-24 1/24/89 J. Lohmeyer When not to clear sense data X3T9.2/89-25 1/26/89 J. Lohmeyer SEARCH DATA Transfer Length Inconsistency X3T9.2/89-26 2/6/89 J. McGrath March 6-8, 1989 SCSI Working Group and CAM Meeting Announcement X3T9.2/89-30 2/18/89 P. Nitza Review of SCSI-2 Revision 7 X3T9.2/89-31 2/10/89 B. Pentecost DA FORMAT UNIT command, IP Modifier X3T9.2/89-32 2/16/89 P. Bramhall Comments on SCSI-2 Working Draft Rev 7 X3T9.2/89-33 2/17/89 D. Appleyard Attention Condition Timing Delays X3T9.2/89-34 2/17/89 D. Appleyard No Operation Message X3T9.2/89-35 2/20/89 D. Davies Tape Read Overlength Detection w/SILI Bit X3T9.2/89-37 2/20/89 Boulay/ Medium Changer Element Status Pentecost Descriptors X3T9.2/89-38 2/19/89 V. Novak Preliminary Results of SCSI-2 Cable Testing X3T9.2/89-39 2/20/89 Rikkonen et. SCSI-2 Cable Testing Summary Report al. X3T9.2/89-40 2/17/89 J. Morse SCSI Cable Test Configurations Enclosure (4) is the current 1989 document register. 6. Liaison Reports 6.1 ISO John Lohmeyer said that Bill Burr had recently sent the final camera- ready original of the ISO SCSI standard (IS 9316:1989) to Keith Brannon, the ISO Editor. Bill had spent hundreds of hours converting the ANSI SCSI standard into the ISO style and format. John read a letter (89-029) that he had sent to Bill on behalf of X3T9.2, thanking Bill for his efforts. Mr. Milligan was not present to give his report, but he left the following report which the chairman read: "The IR is providing X3T9 with an updated International Document Register. It lists 24 documents which require some follow-up or reports of closure. The ISO/IEC JTC 1/SC 13 Secretariat is actively working with the X3T9 project editors and the ISO editorial staff to complete the publication of numerous X3T9 standards. X3T9 will be discussing the vote on the new ISO work item proposal for FDDI profiles at its meeting Friday. They will also address the Danish request to revise ANSI procedures. X3T9.5 is planning to address the Japanese comments on 50 micron fiber at its February meeting. Mr. Milligan intends to ask Mr. Groenke for information on the status of the SC 83 merger and any possible impact upon the upcoming meeting schedule. A subsequent set of comments on the SCSI DIS has been received from Mr. Mori of UNISYS Japan in 13N491 dated 12/28/88. A request for comments on an IEC Interconnection for Telesoftware and Teleprinting has been received in 13N492 dated 1/4/89. It does not appear to impinge upon X3T9 activity. X3 has devised a new classification of documents that can be withheld from committee participants other than international delegates. The X3T9 IR has sent a letter to X3T9 officers including sub- committees requesting they obtain volunteer candidates for the SC 13 or SC 83 meeting later this year." SC 25 has been formed to replace ISO SC 13 and IEC SC83. The German Secretariat of SC 13 will be the SC 25 Secretariat. A primary task for SC 25 at the next meeting in Copenhagen, Denmark is to define the scope of the new committee. This meeting is scheduled for September 25-29, 1989. 6.2 IT8 The IT8 committee had requested that a reference to their activities be included in SCSI-2. A reference to their work, as well as to Fiber Channel and SCSI-3, was placed in Appendix H, Other SCSI Standardization Activities. 6.3 SCSI-2 Common Access Method Committee Report Dal Allan reported on the January CAM meeting. At that meeting Jerry Armstrong of NCR described the agreed-upon structure. The structure presented is: CPU & Operating System - - - - - - - - - - - - - - - - - - - - - - - - - - - - +-----------------------+ | OSD Driver | Operating System Dependent Control Block +-----------------------+ | Transport Module | Transport Control Block +-----------------------+ | SCSI Interface Module | Common Access Method Control Block +-----------------------+ | Protocol Hardware | Command Descriptor Block +-----------------------+ - The OSD is Operating System Dependent and will contain the references that are necessary to cross the border in and out of the CAM layer. - The Transport Module contains a truncated form of the CAM Control Block for functions which do not need a SCSI CDB (e.g., CAM configuration). - The Interface Module manages the CAM Control Block and contains all the information necessary to dispatch a CDB on behalf of the requester. - The Protocol Hardware is the SCSI host adapter. There are issues on how pointers should be handled and appropriate addresses (virtual or physical, size, pointer type, etc.). Dal is working on a suitable format for the standardization of the CAM. There are many editing and style rules to publish standards. It is expected that the proposals will be integrated into a working document that is in a standards format after the February meeting. 6.4 Fiber Channel Dal Allan reported. The most recent working group meeting was held Jan 30, 31. Canstar proposed an implementation based on active hubs and "knock-on-door" dispatching of packets and Grumman proposed a token-ring with features that reduce overhead and ring latency. Data General described a point-to-point link developed in fiber using 200 Mbaud transmitters and receivers which can attain almost 16 MBs with large blocks. IBM described the list of features needed to run "fast" in an HSC-like extension using single mode fiber for up to 10 km. Technology presentations were made by Gazelle on the "Hot Rod" parallel to serial (40:1) conversion components. Ancor covered the technology and use of 1 Gbs circuit switches. These look attractive for providing parallel transfer paths for multi-switched environments. Action items taken to be worked on included the criteria for connectors, (IEC is supposed to work closely with the committee on this); and on security, which will be a big issue. Of concern to all SCSI committee members should be that there is no representation of SCSI needs in the Fiber Channel working group, HSC is well-represented and so too is IPI. Some have even questioned whether SCSI should be supported by the Fiber Channel. The next Fiber Channel working group meeting is in San Rafael March 20- 21. The following meeting will be in Minneapolis sometime in May. 7. Review of Old Action Items 1. Jim Semenak will correct page 7-31 paragraph 3 in SCSI-2 R5 which references a section that does not exist and should contain a description of time ordering of log entries. Complete. 2. John Lohmeyer and Larry Lamers will revise the ISO ESDI document to correct the cover page, to include the figures, and to correct the statement concerning transfer rate reporting requirements. Completed by Dal Allan. 3. John Lohmeyer will correct the wording concerning byte ordering in section 4, delete the [] note in section 5 and include Jeff Stai's figure. Complete. 4. Paul Boulay will develop an appendix on single-ended terminators. Carried over. 5. Paul Boulay will add the READ LONG command to the CD-ROM command set. Complete. 6. Paul Boulay will write an implementors note on restoring mode parameters following a third-party release. Complete. 7. Jeff Stai will develop cabling recommendations. Complete. 8. Jeff Stai will add the WRITE SAME command to section 15. Complete. 8. Working Group and/or Project Status Reports 8.1 ESDI Dal Allan provided John Lohmeyer with ESDI Version 3 for the mailing. This document incorporates the formal and informal public review comments into the ISO-style version of ESDI. A letter from X3T9 to NCR summarizes the changes made as a result of the NCR comment. NCR has informed the X3 Secretariat that ESDI Version 3 satisfactorily addresses their comment (X3T9.2/89-027). 8.1.1 ESDI Synchronized Sector Offset (X3T9.2/89-002, -023) A document from Kiyotake Kumazawa of Hitachi (89-002) was briefly reviewed during the SCSI general working group meeting in January. It was favorably received and as a result, Dal Allan has submitted a document (89-023) detailing the changes he believes are desirable to support the sector offset in the ESDI document. This change was incorporated in ESDI version 3. 8.2 Flexible Disk John Lohmeyer reported that the revised Flexible Disk standard has been published and is available from ANSI for $14. It is known as X3.80-1988. This agenda item will be dropped from future agendas. 8.3 SCSI General Working Group (X3T9.2/89-019 Rev 1) Jeff Stai hosted a working group meeting January 9-11 in Costa Mesa, CA. There were 39 attendees. Most of the meeting was spent reviewing an interim revision of SCSI-2 designated Rev 6a. This document was not distributed outside the working group because it was a partial document and has been superseded by Revision 7 which was in the mailing. This was a good, productive meeting. It is the first time that some sections of the document have been reviewed page-by-page. Many helpful comments were made and the document has been improved considerably as a result. Pete Bramhall of Hewlett Packard submitted approximately 9 proposals all related to sequential-access devices and most related to helical scan implementations of sequential-access devices. Most of these were at least partially accepted. One, introducing a new command, was placed in the queue for SCSI-3. The TERMINATE IMMEDIATE message was again brought up for discussion along with the autosense proposal which had previously been placed in SCSI-2 and later deleted. The working group felt that some form of the TERMINATE IMMEDIATE message (later re-named TERMINATE I/O PROCESS) may be desirable, but they were generally opposed to adding autosense to SCSI-2 (see agenda item 9.4). Revised wording concerning REQUEST SENSE polling was accepted and included in SCSI-2 Rev 7. 9. Old Business 9.1 Proposed alternative single-ended terminator (X3T9.2/89-4) Paul Boulay made a presentation for termination based on a Linear Technology voltage regulator. The proposal maintains positive noise margins at 2.0 volts, which the older 220/330 termination does not. The noise margins are even less if impedances are based on the exhibited reflection properties in cables. Dennis O'Niel of Linear Technology (408-432-1900 ext 322) provide the LT1086 voltage regulator information. A fixed voltage low-dropout regulator would work marvelously, however such a part is not available. Paul has developed an outline for a proposed appendix on termination. Amphenol agreed to provide six prototypes of external terminators for the persons doing the cable evaluations. 9.2 Review of SCSI-2 Draft Document (X3T9.2/86-109 Rev 7) Larry Lamers and John Lohmeyer met in Wichita the week after the January working group meeting to assemble revision 7. The other editors contributed their sections via the SCSI bulletin board. Larry and John spent 150 man-hours working on the document that week. They had not finished with all the known editorial problems when the deadline came for the mailing. In particular, the section and table references in section 9 are known to have some errors. John had since corrected these references and he had deleted the extra RELEASE UNIT command (the first one should be deleted). He also made a number of other editorial corrections to this section (such as changing notes containing requirements into text). It had been pointed out that the table of contents has some errors in that page numbers don't always match the document. This was due to anomalies in the comparison program. Larry has been working with the vendor to get these anomalies fixed. John suggested that a marked-up version of revision 7 be made during the meeting by the general editor and that it be designated revision 8. He was hopeful that the marked-up document could be forwarded to X3T9 and that a clean copy of revision 8 could be prepared for the X3T9 letter ballot. 9.2.1 Vendor ID Registration John Lohmeyer stated that more than a year after X3T9.2 had requested guidance from X3T9 and X3 on vendor registration, we still did not have a satisfactory procedure for vendor ID registration. There has been a great deal of activity in the area of defining international registration authority procedures, but we still did not have a clear procedure to follow. He did not relish becoming as a test case for registration authority. He proposed that we avoid the problem by defining the vendor ID field in the INQUIRY data as vendor specific when non-blank values are returned. An all-blank value should be defined to mean that vendor ID field is not implemented. The committee could maintain a list of vendor IDs that are known to it, but it shouldn't attempt to guarantee uniqueness. Peter Johansson submitted proposed wording for an implementors note to describe the intention of this field. John Morse moved and Jim McGrath seconded that the Vendor ID List be included as an appendix and Peter Johansson's wording, as modified, be included in the document. John Lohmeyer moved and Steve Cornaby seconded that the motion be amended so that the Vendor ID List be deleted entirely from the document. The proposed amendment failed: 18 in favor, 20 opposed. The main motion passed: 35 in favor, 3 opposed. 9.2.2 Required sense data: 8 or 18 bytes? Jim Semenak stated that with the incorporation of ASC and ASCQs in the body of the document it now seems that it should be mandatory to return those bytes in the sense data. Bill Spence moved and Jim McGrath seconded that SCSI-2 targets be required to support the first 18 bytes of sense data. The motion passed: 25 in favor, 1 opposed. 9.2.3 Combine ASC and ASCQ into single 16-bit field? John Lohmeyer suggested that we should consider combining the ASC and ASCQ fields into one 16-bit field. The notion of these being separate fields is missing in most of the documentation. We usually say "... and set the additional sense code to XYZ", where XYZ is the name of both an ASC and an ASCQ. The whole 16-bit field should be called the additional sense code. If necessary, we could refer to the first byte as the primary ASC and the second byte as the secondary ASC. There were several objections to this approach. In the interest of forwarding the suggestion was withdrawn. 9.2.4 Should field names be first letter caps? Many people have complained that the policy of using all lower-case for field names in the text descriptions is confusing. The descriptions often are difficult to read because the field names may be interrupted as normal English rather than a field name. John Lohmeyer spoke to the ANSI editor, Harvey Rosenfeld, the previous week about this policy. ANSI has had a change of heart and now allows initial capital letters for such purposes. They would also be willing to consider using a different font. Mr. Rosenfeld agreed to send the current style manual and the laser printer guidelines to both John Lohmeyer and Larry Lamers. He also indicated that the ANSI style manual probably will be changed to be more compatible with the ISO style. This is not definite, yet. If we start out with the ANSI rules, he did not believe we would be required to switch to the ISO style. He unofficially advised us to use the ISO style if we plan to submit SCSI-2 to ISO. 9.2.5 Should REQUEST SENSE command failure clear sense data? (89-024) John Lohmeyer had received a call from a person who had induced parity errors while testing a vendor's product. A parity error was introduced during the DATA IN phase of a REQUEST SENSE command. The host adapter issued an INITIATOR DETECTED ERROR message which the target responded to with MESSAGE REJECT message. After the command completed, the initiator issued a second REQUEST SENSE command. This command completed with NO SENSE sense key being returned. The target had cleared the sense data from the original error condition and gave the initiator no opportunity to get the sense data after the induced parity error. John submitted 89-024 to revise the REQUEST SENSE wording to make it clear that the target should preserve the sense data until it is successfully transferred to the initiator. After some discussion it was felt that the proposed change would cause some other problems (e.g., stacked sense data) and that the real problem was the target's inability to deal with initiator detected errors. Changes to the wording of the document were not made. 9.2.6 SEARCH DATA command transfer length inconsistency (89-025) John Lohmeyer stated that he had received a call from Mark Koch of Pentax Teknologies saying that the SEARCH DATA commands are ambiguous as to whether the transfer length field is the number of bytes of parameter data to be passed or the maximum number of blocks to be searched. John submitted 89-025, which would define the transfer length as the number of bytes of parameter data. Jeff Stai saw John's proposal on the SCSI BBS and responded with the following message to John: The transfer length is the number of blocks, not the number of bytes in the parameters list. Transfer length is defined exactly as for READ. The length of the parameter list is implicit in the command, just like READ CAPACITY and REASSIGN BLOCKS: there is a 14-byte header which includes a field defining the length of the following search argument(s). What is missing is a paragraph that says: "The transfer length field specifies the number of contiguous logical blocks that shall be searched. A transfer length of zero indicates that no logical blocks shall be searched. This condition shall not be considered an error. Any other value indicates the number of logical blocks that shall be searched." "Records" are sub-fields within a block (like a FAT entry) and may span multiple blocks. The "transfer length" specifies the number of blocks limit, while "number of records" specifies the number of records limit. For instance, if you had four records per block, and wanted to search 7 records, you would specify 2 blocks and the search would end after the 7 records, not the whole two blocks. The target needs both data points to understand the record structure. I think you will find that all controllers that currently do search (including ours) implement "transfer length" as a number of blocks. John Lohmeyer said he didn't really care which definition was accepted provided that the ambiguity was removed. He had a preference for the architectural purity of specifying the length of the parameter list in the CDB, but was willing to accept the product precedent since this command has been around for a long time. Peter Johansson moved and Larry Lamers seconded that the Transfer Length field be changed to a Parameter List Length field and the wording from 89-25 be accepted. Dal Allan moved and Jim McGrath seconded that the motion be amended so that bytes 7 and 8 be labeled vendor-specific. Paul Nitza moved and Bob Snively seconded that the amendment be revised so that bytes 7 and 8 be labeled Number of Blocks to Search. The amendment passed: 11 in favor, 8 opposed. The revised amendment stated that bytes 7 and 8 be labeled Number of Blocks to Search. The amendment passed: 18 in favor, 4 opposed. The revised motion stated that the Transfer Length field be changed to Number of Blocks to Search and the wording from Jeff Stai's message be accepted. The motion passed: 18 in favor, 5 opposed. Jim McGrath moved and Dal Allan seconded that the Number of Blocks to Search field be changed to vendor specific along with an implementors note stating that industry practice is to use this field as a number of blocks field. The motion failed: 7 in favor, 17 opposed. 9.2.7 READ DEFECT DATA command [Gerry Houlder] The paragraphs in the READ DEFECT DATA command describing the defect data format field are obscure because they reference the FORMAT UNIT command and then modify its meaning for this field. We should completely define the meaning of this field within the READ DEFECT DATA command description. The values for this field should be: 0 Vendor Specific 1-3 Reserved 4 Bytes From Index 5 Physical Sector Format 6 Vendor Specific 7 Reserved Plus something should be done to the paragraph following the implementors note. It is confusing and seems to imply a conflict. Peter Johansson moved and Paul Nitza seconded that the corrections in document 89-30, page 12 be accepted to resolve this issue and that it be modified to state that when physical sector format is used that it may or may not include the complete list. The motion passed unanimously. 9.2.8 Comments on SCSI-2 (89-30, 31, 32, 33, 34, 35, 37) [P. Nitza, B. Pentecost, P. Bramhall, D. Appleyard, D. Davies, P. Boulay] A poll was taken on what the correct method for handling a command received without a queue tag. The proposal is to change from handling it as an untagged queue to handling it as an ordered queue tag command. This change was accepted: 21 in favor, 12 opposed. Peter Johansson moved and Bob Snively seconded that the logical block address returned in the sense data in REASSIGN BLOCKS be changed to the last successfully reassigned block. The motion failed: 2 in favor, 12 opposed. Most of Paul Nitza's comments (document 89-030) were accepted. The comment on Duplicate Nexus was substantially re-written during the meeting. Bob Pentecost's implementors note on initialization pattern (document 89-031) was accepted. Pete Bramhall's comments (document 89-32) were accepted. Dennis Appleyard's wording for the addition of attention condition timing delays (document 89-033) was accepted with a modification to change the length of delay to "two deskew delays" from "a deskew delay plus a cable transfer delay". His comment on the NO OPERATION message (document 89-034) was accepted for inclusion as an example. Dan Davies revised wording for the sequential-access READ command (document 89-035) was accepted. Paul Boulay's revised wording for Medium Element Status Descriptors (document 89-037) was accepted. In addition several hand-written comments were received that noted misspellings and incorrect references. The general editor has an action item to incorporate these changes in a draft for review at the next working group. Successful completion of this task may allow the document to complete the 30-day letter ballot at X3T9 prior to the next plenary meeting. Following the letter ballot the document will go into its public review period. 9.3 Cable Testing Results (X3T9.2/89-18) John Morse presented an interim report on the testing. No particular cable/connector configuration fails absolutely. All cable/connector configurations potentially have some problems. See X3T9.2/89-40. A complete report is slated for April. Lowering term power (but within specification) causes failures. Trapezoidal drivers are better with less undershoot and smaller glitches. Alex Pappas presented his summary of testing from Sun. See 89-038 and 89-039. Maintaining terminator power is critical to safe operation. Targets should provide terminator power since there is significant drop through long cables. More detailed analysis will be available in April. Kurt Chan presented an update on HP's testing. See 88-165 and 89-019. Spice simulations are being undertaken to correlate the bench testing and examine different terminations. Meeting the standard does not guarantee that the implementation works. 9.4 TERMINATE IMMEDIATE message re-visited (X3T9.2/88-158 Rev 1, 89-7) The TERMINATE IMMEDIATE message was assigned to a SCSI-3 Specific-Subject Working Group (SSWG) at the last meeting by a vote of 19 to 13. Because of the close vote, the January working group reviewed this topic and concluded that a properly-edited, linguistically-correct proposal for a TERMINATE IMMEDIATE message without the actual block count message might be acceptable. The strongest objections to the proposal were on the grounds that it might delay the forwarding schedule and that the proposed mechanisms for returning the residue information were ugly. Paul Nitza had submitted a revised proposal, which DPT has mailed to the principals, detailing the changes needed to incorporate a TERMINATE I/O PROCESS message (X3T9.2/88-158 Rev 1). Jim McGrath moved and Peter Johansson seconded that the TERMINATE I/O PROCESS message become a SCSI-3 item. The motion failed with 13 in favor and 20 opposed. Steve Goldman moved and Paul Nitza seconded that the TERMINATE I/O PROCESS message (88-158 R1) be incorporated into SCSI-2. Jim McGrath moved and Bob Snively seconded that the motion be amended to make it an appendix. The amendment failed: 8 in favor, 23 opposed. The main motion passed with 23 in favor and 10 opposed. 10. New Business 10.1 Tape READ underlength detection (X3T9.2/89-021) [Penokie] This proposal was withdrawn in favor of 89-35, which was accepted. 10.2 SCSI-2 Forwarding motion (X3T9.2/86-109 Rev 7) Larry Lamers moved and Bill Burr seconded that SCSI-2 (X3T9.2/86-109 Revision 8) be forwarded to X3T9 for further processing. A roll call vote was taken with the following results: X3T9.2 Roll Call Vote Results on Forwarding SCSI-2 Organization Voter Vote Yes No Abstain Absent -------------------------- ----------------------- ---- --- -- ------- ------ 3M Company Mr. Robert C. Herron N 0 1 0 0 ACT Technology Mr. Nathan Lee Y 1 0 0 0 Adaptec, Inc. Mr. Robert N. Snively Y 1 0 0 0 Advanced Micro Devices Mr. Scott Smyers Y 1 0 0 0 AMP, Inc. Mr. Charles Brill Y 1 0 0 0 Amphenol Corp. Mr. Michael Wingard Y 1 0 0 0 Ancot Corp. Mr. Jan V. Dedek Y 1 0 0 0 Apple Computer Mr. Harlan Andrews Y 1 0 0 0 Archive Corp. Mr. Sam Washburn Y 1 0 0 0 Aspen Peripherals (none) A 0 0 0 1 AT&T Mr. James J. Semenak Y 1 0 0 0 Ballard Synergy Corp. (none) A 0 0 0 1 Burndy Corp. Mr. David Barnum N 0 1 0 0 Cinch Connector Mr. John Guennewig Y 1 0 0 0 Cipher Data Products, Inc. Mr. Dan Davies Y 1 0 0 0 Cipher/Optimem (none) A 0 0 0 1 Ciprico Inc. (none) A 0 0 0 1 Cirrus Logic Inc. Mr. Robert Kellert Y 1 0 0 0 Congruent Software, Inc. Mr. Peter Johansson N 0 1 0 0 Cygnet Systems Mr. Bart Raudebaugh Y 1 0 0 0 Data General Corp. Mr. Sam Pendleton Y 1 0 0 0 Datacopy Corp. Mr. Mark P. Pearson Y 1 0 0 0 DDC Pertec Mr. Rick Ehrecke Y 1 0 0 0 Digital Equipment Corp. Mr. John Morse Y 1 0 0 0 Distributed Logic Corp. Mr. Duncan McDonald Y 1 0 0 0 DPT Mr. Tom Treadway Y 1 0 0 0 Eastman Kodak Co. Mr. Lou Edwards Y 1 0 0 0 Emulex Corp. Mr. Gerald Maurer Y 1 0 0 0 ENDL Mr. I. Dal Allan Y 1 0 0 0 Exabyte Corp. Mr. Bill Duran Y 1 0 0 0 Fujitsu Mr. Robert Liu Y 1 0 0 0 Hewlett Packard Co. Mr. Bob Pentecost Y 1 0 0 0 Hirose Electric U.S.A. Mr. Oscar Kornblum Y 1 0 0 0 Hitachi Mr. Fielding S. Tabb Y 1 0 0 0 Honda Connector Mr. David McFadden Y 1 0 0 0 Honeywell Bull, Inc. (none) A 0 0 0 1 IBM Corp. Mr. George Penokie Y 1 0 0 0 ICL Mr. Tony Salthouse Y 1 0 0 0 IMPRIMIS/CDC Mr. Gene Milligan Y 1 0 0 0 Intergraph Corp. (none) A 0 0 0 1 Interphase Corp. Mr. Craig Scott Y 1 0 0 0 Iomega Corp. Mr. Geoff Barton Y 1 0 0 0 ITT Cannon (none) A 0 0 0 1 JAE Mr. W. Dean Wilson Y 1 0 0 0 Laserdrive Ltd. (none) A 0 0 0 1 LMS Mr. Paul Boulay Y 1 0 0 0 Maxtor Corp. Mr. Lawrence J. Lamers Y 1 0 0 0 Methode Electronics, Inc. Mr. Bob Masterson Y 1 0 0 0 Micropolis Corp. Mr. Stephen Cornaby Y 1 0 0 0 Miniscribe Corp. (none) A 0 0 0 1 X3T9.2 Roll Call Vote Results on Forwarding SCSI-2 Organization Voter Vote Yes No Abstain Absent -------------------------- ----------------------- ---- --- -- ------- ------ Mitsubishi Electronics Mr. Daniel E. Moczarny Y 1 0 0 0 Amer Molex Inc. Mr. Craig Bixler Y 1 0 0 0 Nat. Inst. of Stnds. & Mr. William E. Burr Y 1 0 0 0 Tech. National Semiconductor Mr. James Schuessler Y 1 0 0 0 NCR Corp. Mr. John Lohmeyer Y 1 0 0 0 NEC Mr. Bruce Anderson Y 1 0 0 0 Northern Telecom, Inc. Mr. David Filpus Y 1 0 0 0 Olivetti Mr. Dennis P. Trupski Y 1 0 0 0 Optotech, Inc. (none) A 0 0 0 1 OTL Engineering Mr. Paul Nitza Y 1 0 0 0 Panasonic (Matsushita) (none) A 0 0 0 1 Panduit Corp. Mr. Ed Cieniawa Y 1 0 0 0 Peripheral Technology Inc. (none) A 0 0 0 1 Priam Corp. Mr. Mike Buckert Y 1 0 0 0 Prime Computer, Inc. (none) A 0 0 0 1 Quantum Corp. Mr. James McGrath Y 1 0 0 0 Ricoh Corporation Mr. Toshi Iizuka Y 1 0 0 0 Scientific Micro Systems Mr. Bob Reid Y 1 0 0 0 Seagate Technology Mr. D. Michael Robinson Y 1 0 0 0 Siemens Mr. Dean Konell Y 1 0 0 0 Sony Corp. of America Ms. Rita Lin Y 1 0 0 0 Stewart Connector Systems Mr. David A. Hatch N 0 1 0 0 Storage Technology Corp. Mr. Dennis Appleyard Y 1 0 0 0 Sun Microsystems, Inc. Mr. Curtis A. Ridgeway Y 1 0 0 0 Technology Forums, Ltd. (none) A 0 0 0 1 Texas Instruments Mr. D. W. Spence Y 1 0 0 0 Thomas & Betts Mr. Harvey Waltersdorf N 0 1 0 0 UNISYS Mr. Peter Dougherty Y 1 0 0 0 Viking Connectors (none) A 0 0 0 1 Western Digital (none) A 0 0 0 1 ========================== ======================= ==== === == ======= ====== Total: 59 5 0 16 The motion carried. Beth Stephens and Larry Lamers were appointed the Technical Editors for SCSI-2. 10.3 ESDI Forwarding motion (X3T9.2/87-105 Rev 3) Dal Allan moved and Chuck Brill seconded that ESDI (X3T9.2/87-105 Revision 4) be forwarded to X3T9 for further processing. A roll call vote was taken with the following results: X3T9.2 Roll Call Vote Results on Forwarding ESDI Organization Voter Vote Yes No Abstain Absent -------------------------- ----------------------- ---- --- -- ------- ------ 3M Company (none) A 0 0 0 1 ACT Technology Mr. Nathan Lee Y 1 0 0 0 Adaptec, Inc. Mr. Robert N. Snively Y 1 0 0 0 Advanced Micro Devices Mr. Scott Smyers Y 1 0 0 0 AMP, Inc. Mr. Charles Brill Y 1 0 0 0 Amphenol Corp. Mr. Michael Wingard Y 1 0 0 0 Ancot Corp. Mr. Jan V. Dedek Y 1 0 0 0 Apple Computer Mr. Harlan Andrews Y 1 0 0 0 Archive Corp. Mr. Sam Washburn Y 1 0 0 0 Aspen Peripherals Mr. John Herron Y 1 0 0 0 AT&T Mr. James J. Semenak Y 1 0 0 0 Ballard Synergy Corp. (none) A 0 0 0 1 Burndy Corp. Mr. David Barnum Y 1 0 0 0 Cinch Connector Mr. John Guennewig Y 1 0 0 0 Cipher Data Products, Inc. Mr. Dan Davies Y 1 0 0 0 Cipher/Optimem (none) A 0 0 0 1 Ciprico Inc. (none) A 0 0 0 1 Cirrus Logic Inc. Mr. Robert Kellert Y 1 0 0 0 Congruent Software, Inc. Mr. Peter Johansson Y 1 0 0 0 Cygnet Systems Mr. Bart Raudebaugh Y 1 0 0 0 Data General Corp. Mr. Sam Pendleton Y 1 0 0 0 Datacopy Corp. Mr. Mark P. Pearson Y 1 0 0 0 DDC Pertec (none) A 0 0 0 1 Digital Equipment Corp. Mr. John Morse Y 1 0 0 0 Distributed Logic Corp. Mr. Duncan McDonald Y 1 0 0 0 DPT Mr. Tom Treadway Y 1 0 0 0 Eastman Kodak Co. Mr. Lou Edwards Y 1 0 0 0 Emulex Corp. Mr. Gerald Maurer Y 1 0 0 0 ENDL Mr. I. Dal Allan Y 1 0 0 0 Exabyte Corp. Mr. Bill Duran Y 1 0 0 0 Fujitsu Mr. Robert Liu Y 1 0 0 0 Hewlett Packard Co. Mr. Bob Pentecost Y 1 0 0 0 Hirose Electric U.S.A. Mr. Oscar Kornblum Y 1 0 0 0 Hitachi Mr. Fielding S. Tabb Y 1 0 0 0 Honda Connector Mr. David McFadden Y 1 0 0 0 Honeywell Bull, Inc. (none) A 0 0 0 1 IBM Corp. Mr. George Penokie Y 1 0 0 0 ICL (none) A 0 0 0 1 IMPRIMIS/CDC Mr. Gene Milligan Y 1 0 0 0 Intergraph Corp. (none) A 0 0 0 1 Interphase Corp. Mr. Craig Scott Y 1 0 0 0 Iomega Corp. Mr. Geoff Barton Y 1 0 0 0 ITT Cannon (none) A 0 0 0 1 JAE Mr. W. Dean Wilson Y 1 0 0 0 Laserdrive Ltd. (none) A 0 0 0 1 LMS Mr. Paul Boulay Y 1 0 0 0 Maxtor Corp. Mr. Lawrence J. Lamers Y 1 0 0 0 Methode Electronics, Inc. Mr. Bob Masterson Y 1 0 0 0 Micropolis Corp. Mr. Stephen Cornaby Y 1 0 0 0 Miniscribe Corp. (none) A 0 0 0 1 X3T9.2 Roll Call Vote Results on Forwarding ESDI Organization Voter Vote Yes No Abstain Absent -------------------------- ----------------------- ---- --- -- ------- ------ Mitsubishi Electronics Mr. Daniel E. Moczarny Y 1 0 0 0 Amer Molex Inc. Mr. Craig Bixler Y 1 0 0 0 Nat. Inst. of Stnds. & Mr. William E. Burr Y 1 0 0 0 Tech. National Semiconductor Mr. James Schuessler Y 1 0 0 0 NCR Corp. Mr. John Lohmeyer Y 1 0 0 0 NEC Mr. Bruce Anderson Y 1 0 0 0 Northern Telecom, Inc. Mr. David Filpus Y 1 0 0 0 Olivetti Mr. Dennis P. Trupski Y 1 0 0 0 Optotech, Inc. (none) A 0 0 0 1 OTL Engineering Mr. Paul Nitza Y 1 0 0 0 Panasonic (Matsushita) (none) A 0 0 0 1 Panduit Corp. Mr. Ed Cieniawa Y 1 0 0 0 Peripheral Technology Inc. (none) A 0 0 0 1 Priam Corp. Mr. Mike Buckert Y 1 0 0 0 Prime Computer, Inc. (none) A 0 0 0 1 Quantum Corp. Mr. James McGrath Y 1 0 0 0 Ricoh Corporation Mr. Toshi Iizuka Y 1 0 0 0 Scientific Micro Systems Mr. Bob Reid Y 1 0 0 0 Seagate Technology Mr. D. Michael Robinson Y 1 0 0 0 Siemens Mr. Dean Konell Y 1 0 0 0 Sony Corp. of America Ms. Rita Lin Y 1 0 0 0 Stewart Connector Systems (none) A 0 0 0 1 Storage Technology Corp. Mr. Dennis Appleyard Y 1 0 0 0 Sun Microsystems, Inc. Mr. Curtis A. Ridgeway Y 1 0 0 0 Technology Forums, Ltd. (none) A 0 0 0 1 Texas Instruments Mr. D. W. Spence Y 1 0 0 0 Thomas & Betts Mr. Harvey Waltersdorf Y 1 0 0 0 UNISYS Mr. Peter Dougherty Y 1 0 0 0 Viking Connectors (none) A 0 0 0 1 Western Digital (none) A 0 0 0 1 ========================== ======================= ==== === == ======= ====== Total: 61 0 0 19 The motion carried. 10.4 Review of new documents See 9.2.8 above. 10.5 Agenda for the Milpitas Working Group The following draft agenda was established for the Milpitas working group meeting: 1. Review of the SCSI-2 Revision 8 changes. SCSI-3 Items 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] 11. SCSI-3 Activities The project proposal for SCSI-3 has been approved by X3. There were no reports from the SSWGs. 12. Review of Action Items 1. Paul Boulay will develop an appendix on single-ended terminators. 2. Larry Lamers will prepare revision 8 of SCSI-2. 3. John Lohmeyer will request X3T9 letter ballots on forwarding SCSI-2 and ESDI. 13. Meeting Schedule The next meeting of X3T9.2 will be April 24-25, 1989 at the Tradewinds Hotel (813-367-6461) in St. Petersburg Beach, Florida hosted by AMP. Please mention X3T9/AMP when making reservations to receive the special room rate of $98.00 Single or Double. The cut-off date for reservations is April 3, 1989. The plenary meeting schedule for the remainder of 1989 is: Date Location Host -------------------- ------------------------ --------------------- June 19-20, 1989 San Jose, Ca AMD August 21-22, 1989 Denver DEC October 16-20, 1989 Research Triangle Park, NC IBM December 4-5, 1989 San Diego, CA NCR 13.1 General Working Group Schedule for 1989 The SCSI-2/SCSI-3 Working Group meeting schedule for 1989 is: Date Location Host Note -------------------- ----------------- ------------- ------------------- March 6-7, 1989 Milpitas, CA Jim McGrath {see 89-026R0.TXT} May 8-9, 1989 Wichita, KS John Lohmeyer July 10-12, 1989 Chicago, IL Joe Lawlor {tentative} September 6-8, 1989 ? ? {tentative date} Oct 30 - Nov 1, 1989 ? ? {tentative date} The March and May meetings have been shortened to two days. 13.2 Methods for Support of Future Plenary Meetings Del Shoemaker reported that X3T9 is having some difficulty in getting hosts for meetings with sufficient lead time to insure being able to find a hotel with adequate meeting facilities. In the current environment a lead time of 18 to 24 months is needed to get adequate facilities. Most companies are not willing to undertake a commitment that far into the future. Del presented several proposals for dealing with this: 1) establish an escrow account so that X3T9 could make the future commitments by adding a surcharge to hotel room rates; 2) charge an up-front meeting fee of $100; or 3) reduce meeting costs by eliminating refreshments. Option 3 may not be a sufficient measure to deal with the problem. At this point several members volunteered as hosts. At least two of these members commented that they had offered to host meetings in the past but had been rejected because their companies were not located in resort locations. Del Shoemaker conducted several straw polls concerning these options and planned to discuss the results at the X3T9 meeting on Friday. 14. Adjournment The meeting was adjourned at 2:05 p.m. on Tuesday, February 21, 1989.