X3T9.2/92-126R0 To: Membership of X3T9.2 From: Lawrence J. Lamers Date: May 21, 1992 Subject: Minutes of X3T9.2 May General Working Group Agenda -- X3T9.2 May General Working Group Harrisburg, PA -- May 18-20, 1992 1. Opening Remarks 2. Attendance and Membership 3. Revised SCSI-3 project proposals 4. SCSI-3 Architecture Model (92-116R1) [Monia] 5. Direct Attach Device Interface [Smyers] 6. Generic Packetized Protocol [Stephens] 7. Serial Bus Protocol [Gardner] 8. Fiber Channel Protocol [Snively] 9. SGS-Thomson Serial Technology Presentation [Crowell] 10. Enhancement of Sequential Access for Medium Changers (92-6) [Jones] 11. Alternate P Cable Connector for SCSI-3 11.1 SCSI-3 Connector Framework (Ham) 11.2 What is not being requested (Jay Near) 11.3 Blind Mating 12. ATA Extensions Issues 12.1 Logical Block Addressing (92-74) [Western Digital] 12.2 Read/Write Multiple Issues (92-___) [Novell] 12.3 Command Code 0 (92-121R0) [Newman] 12.4 IDE (ATA) Document Changes (92-110) [Intel] 12.5 DMA Timing Issues [Emulex] 12.6 Power Management [Zenith] 12.7 68-pin Assignments on a PCMCIA Physical with ATA electrical [Integral, Cirrus] 12.8 READ LONG Command [Seagate] 12.9 3-Pin Power Connector [Seagate] 12.10Cable Select Signal [IBM] 12.11Error bit in Status Register [Adaptec] 13. Device Type 0 Support for Media Interchange (92-56) [Wilhelm] 14. SCSI-3 Queuing Model (91-098 R8) [Penokie] 15. CHANGE INQUIRY Command (92-85) [Penokie] 16. Tolerance Analysis of P-cable Connection (92-91) [McGrath] 17. Data Compression Sense Codes (92-111) [Dougherty] 18. SCSI-2 TIB on Queuing [Gallant] 19. Patent Issues [Allan] 20. Working Group Schedule 21. Adjournment Results of Meeting 1. Opening Remarks Dal Allan, the Vice-Chair, called the meeting to order at 2:00 p.m., Monday May 18, 1992. He thanked Chuck Brill of AMP for hosting the meeting. As is customary, the people attending introduced themselves. A copy of the attendance list was circulated for attendance and corrections. It was stated that this is an authorized X3T9.2 General Working Group meeting. The meeting was conducted under Roberts Rules of Order. Since this is a working group meeting no final actions are taken. The working group prepares a set of recommendations for approval by the X3T9.2 Plenary Committee on Lower Level Interfaces. The minutes of this meeting will be posted to the SCSI BBS and the SCSI Reflector within 14 calendar days. 2. Attendance and Membership Attendance at working group meetings does not count toward meeting the attendance requirements for X3T9.2 membership. The general working group meetings are open to any person or company to attend and express their opinion on the subjects being discussed. The voting rules for the meeting are those of the parent committee, X3T9.2. These rules are: one vote per company; and any participating company member may vote. The following people attended the meeting: Name Status Organization Phone Number ------------------------- ------ ------------------------------ -------------- Mr. Robert C. Herron A 3M Company (512) 984-6807 Mr. Ken Wolfswinkel V 3M Company (512) 984-6719 Mr. Norm Harris P Adaptec, Inc. (408) 945-8600 Mr. Thomas Newman A Adaptec, Inc. (408) 945-8600 Mr. Charles Brill P AMP, Inc. (717) 561-6198 Mr. Bob Whiteman A AMP, Inc. (717) 780-7481 Mr. Bill Garver O AMP, Inc. (717) 986-3223 Mr. J. Michael Evans O AMP, Inc. (717) 780-4248 Mr. Bill Hurdle O AMP, Inc. (919) 727-5623 Mr. C. Edward Reynolds S AMP, Inc. (717) 780-6558 Mr. David McAdoo V AMP, Inc. (919) 727-5781 Mr. Ken Talentino O Amphenol (203) 265-8901 Mr. Jeff Rosa P Amphenol Interconnect (607) 786-4222 Mr. Michael Wingard A Amphenol Interconnect (607) 786-4241 Mr. Richard M. Ross O Amphenol/Spectra-strip (203) 281-3200 Mr. Scott Smyers P Apple Computer (408) 974-7057 Mr. Robert Brown O Areal Technology (408) 436-6871 Mr. Robert Brown V Areal Technology Mr. Thomas Debiec R Berk-Tek (717) 354-6200 Mr. Rich Mizia O C&M Corp. (203) 774-4812 Mr. John Geldman A Cirrus Logic Inc. (510) 226-2368 Mr. Michael Lazar P Connector Ind. Consulting Gr (914) 949-5450 Mr. Stephen R. Cornaby P Conner Peripherals (303) 651-2881 Mr. Greg McSorley O Data General Corp. (508) 870-8829 Mr. Charles Monia P Digital Equipment Corp. (508) 841-6757 Dr. William Ham A Digital Equipment Corp. (508) 841-2629 Mr. John A. Gallant A Digital Equipment Corp. (603) 881-2472 Mr. Ralph Weber A Digital Equipment Corp. (603) 881-1497 Mr. James A. Somerville O DuPont (302) 992-5869 Mr. Douglas Wagner V Dupont (805) 498-0325 Mr. Mehran Raitezani Emulex Corp. (714) 668-5331 Mr. Skip Jones A Emulex Corp. (714) 668-5058 Mr. I. Dal Allan P ENDL (408) 867-6630 Mr. D. W. Spence A ENDL Associates (512) 255-0339 Mr. Joel Urban O Fujitsu Microelectronics Inc (408) 922-8928 Mr. Jeffrey L. Williams P Hewlett Packard Co. (208) 323-5030 Mr. Kurt Chan A Hewlett Packard Co. (916) 785-5621 Mr. George Drennan Hewlett-Packard (208) 323-2513 Mr. Levi Lebo Hewlett-Packard (208) 323-4652 Mr. Tom Kulesza O Honda Connector (708) 913-9566 Mr. George Penokie P IBM Corp. (507) 253-5208 Mr. Gerald Marazas A IBM Corp. (407) 982-4423 Mr. Thomas H. Newsom S IBM Corp. (407) 982-6447 Mr. Giles Frazier S IBM Corp. (512) 838-1802 Mr. Jay H. Neer S IBM Corp. (407) 982-5651 Ms. Gricell Co V IBM Corp. (512) 838-3664 Mr. Peter Thompson Inmos Limited 44-454-616616 Mr. John Blagaila V Integral Peripherals (303) 449-8009 Mr. Bruce Bonner V Intel (916) 351-4411 Mr. Kevin R. Pokorney O Intellistor, Inc. (303) 682-6649 Mr. Lawrence J. Lamers P Maxtor Corp. (408) 432-3889 Mr. Reginald J. Murray V Ministor Peripherals Corp. (408) 943-0165 Mr. Joe Dambach P Molex Inc. (708) 969-4550 Mr. Jim McGrath A Molex Inc. (708) 969-4550 Mr. Rod DeKoning NCR Corp. 316-636-8842 Mr. John Blagaila R Prairie Tek Mr. Gene Milligan A Seagate Technology (405) 324-3070 Mr. Gerald Houlder A Seagate Technology (612) 844-5869 Mr. Hale Landis A Seagate Technology (408) 439-2443 Mr. Forrest Crowell P SGS-Thomson Microelectronics (714) 957-6018 Mr. Robert L. Simpson P Sony Corp. of America (408) 944-4348 Mr. Vit Novak A Sun Microsystems, Inc. (415) 336-2455 Mr. Stephen Gross SunDisk Corp (408) 562-0500 Mr. Mark Granahan Texas Instruements (214) 997-5955 Mr. Harvey Waltersdorf P Thomas & Betts (803) 676-2905 Mr. Peter Dougherty P UNISYS (714) 380-6270 Mr. Tom Hanan A Western Digital (714) 932-7472 Mr. Shishir Shah A Western Digital (714) 932-7235 Mr. Saied Zangenehpour P Zenith Data Systems (616) 982-3509 3. Revised SCSI-3 project proposals Larry Lamers gave a status of the project proposals being prepared for SCSI-3. He noted that only the protocol projects were included in the mailing for a plenary vote at the June meeting. The command set project proposals still need some work, but should be on the agenda for the August meeting. 4. SCSI-3 Architecture Model (92-116R1) [Monia] Charles Monia presented the Packetized SCSI Systems architecture. This effort is still in its early development stage and Charles is requesting comment and guidance. SAM will not attempt to defince a common payload packet for packetized interconnects. The SAM document will describe functional requirements of SCSI. 5. Direct Attach Device Interface [Smyers] Scott Smyers gave a brief introduction to the concepts of DADI. He is planning a draft document. He contrasted a pure memory model of access to disk drives with a channel model. The channel model allows a device to improve its performance by allowing it to manipulate the order of data transfer to take advantage of its physical characteristics. A pure memory model does not since it expects an immediate response and does not queue requests. Commands are not used per se, but rather a memory access request. The DMA model uses a pointer and length to read and write data. The device is responsible for manipulating the pointers instead of the host. The DMA model uses a control structure that may contain multiple requests, so that optimization of data transfer can occur. The control structure is a linked list of requests. 6. Generic Packetized Protocol [Stephens] Gary Stephens gave his presentation on Encapsulating SCSI - Generic Packetized Protocol. He explained how it could be used in a heterogeneous interface environment to transport SCSI over a mixture of Ethernet, FDDI, Fiber Channel etc. Scott Smyers questioned how the routing could be accomplished as there are no tools to manage data movement over such a complex structure. 7. Serial Bus Protocol [Gardner] Ed Gardner was not present and there was no discussion of this topic. 8. Fiber Channel Protocol [Snively] Bob Snively was not present and there was no discussion of this topic. 9. SGS-Thomson Serial Technology Presentation [Crowell] Forrest Crowell presented further information on the DS-Link. The DS-Link is currently used for transputers, but SGS-Thomson is willing to license just the DS-Link for use as a serial interface. 200Mbit point to point full duplex, actual payload uni-directional 18.6 Mb/sec, bi-directional 32.8 Mb/sec. Peter Thompson covered the communications concepts, architecture and protocol used by DS-Links in the T9000 transputer. 10. Enhancement of Sequential Access for Medium Changers (92-6) [Jones] Kevin Jones was not present. Dal Allan will meet with Kevin Jones next week in England. He will report back at the next meeting. 11. Alternate P Cable Connector for SCSI-3 11.1 SCSI-3 Connector Framework (Ham) Bill Ham presented is connector framework, including the list of requirements from a system viewpoint. He identified seven levels of basic applications features. 11.2 What is not being requested (Jay Near) Jay Near presented the IBM usability testing done in Boca Raton. Folks within IBM were used as test subjects. Out of a sample size of 59, 3 damaged connectors occurred. 11.3 Blind Mating There was some concern that the issue of blind mating may need some further consideration based on the weight of evidence presented. 12. ATA Extensions Issues 12.1 Logical Block Addressing (92-74) [Western Digital] Shishir Shah reviewed the latest proposal for logical block addressing. This proposal follows a natural order. Hal pointed that Word 60-61 should follow the same convention as Word 57-58 and reflect the total number of logical blocks available. Accepted. The ATA task file is only valid at end of transfer or when error occurs, and not after every interrupt. He also proposed that the drive will maintain the emulation geometry through soft resets. Hal Landis beat him back on this one since use of 66h in SET FEATURES solves this problem. Shishir has an action item to draft a public review response. 12.2 Read/Write Multiple Issues (92-___) [Novell] Dal Allan read the Novell request for an additional bit for multiple operations. Their request is to redefine a vendor specific bit, which is not allowed. Dal will inform Novell of this and indicate that they need to re- submit a request with a reserved bit. Dal read the request for a method of determining the current mode, i.e. whether it is in multiple mode or not. The request is for the lower order byte of word 59 to return the multiple size. Drives that do not support read/write sector when a multiple is other than zero is a violation of ATA. Steve Gross pointed out the need for a valid bit on the multiple size. Bit 8 of word 59 could will be used as a valid bit and also double as the honesty bit. 12.3 Command Code 0 (92-121R0) [Newman] Tom Newman proposed having Command Code 0h be a no operation command that did not modify the task file registers so that 16-bit memory accesses can be made. Tom will submit this as a public review comment. 12.4 IDE (ATA) Document Changes (92-110) [Intel] This will be an official public review comment, but no one has had time to review it yet. 12.5 DMA Timing Issues [Emulex] Mehran Ramezani has action item to develop wording for the ATA document regarding the DMA timing. The stopping early recommendation, is not elegant, may have some patent issues, requires hardware changes and so was rejected in favor of a 40ns timing requirement. 12.6 Power Management [Zenith] A set of 3 commands to power down system. REST (E7h), READ DRIVE STATE (E9h), RESTORE DRIVE STATE (EAh) , with a feature register of ACh. REST is get prepared to shutdown within 30 seconds. The vary next command must be READ DRIVE STATE. READ DRIVE STATE returns 512 bytes of data, the last bit of the last byte is not used by the vendor, all the other bits are vendor specific. If this bit is a one on a RESTORE DRIVE STATE then the drive generates an interrupt. 12.7 68-pin Assignments on a PCMCIA Physical with ATA electrical [Integral, Cirrus] Cable Select has a power usage problem, John inverted the signal to minimize power usage. The 10k pull-up could be on the adapter. Could also convert it to active signal by strapping it to Vcc or GND, could also reverse the meaning of CSEL in ATA. Remanded to the special interest group at Maxtor on June 1, 1992. 12.8 READ LONG Command [Seagate] The drive does not check ECC on READ LONG commands. The error status is incorrect, the drive should not set either the Corrected or Uncorrected Error bits when performing a READ LONG. Both of these entries should be blank. Word 22, stating the number of ECC bytes passed is confusing, and should state the this is the number of available ECC bytes. 12.9 3-Pin Power Connector [Seagate] Several mfg's have their 3-pin power wired differently than the ATA standard. Hal Landis cited three variations. Saied Zangenehpour agreed to make a public review comment on this item. The ATA document has it as 12 Vdc/GND/5 Vdc; Seagate has it as 5/12/GND, Maxtor is GND/12/5, and Miniscribe was 5/12/GND. 12.10Cable Select Signal [IBM] George Penokie questioned Section 6.3.15.2 stating that it is not implementable as defined. Why is CSEL specifiec as an open-colletor output? It seems that it was carried over from the definition of DASP. CSEL should state that the signal should have a 10k ohm pull up resistor. 12.11Error bit in Status Register [Adaptec] What should the Error Bit mean when a command is aborted? If the drive aborts an operation should it set the Error Bit to indicate that the command is over? It seems that the answer should be yes. 13. Device Type 0 Support for Media Interchange (92-56) [Wilhelm] At Norm Harris's request this item was tabled until the next meeting. 14. SCSI-3 Queuing Model (91-098 R8) [Penokie] George Penokie presented his latest revision of the SCSI-3 Queuing Model. There was still debate on a single vs multiple queue and the meaning of ordered. Behavior of the ORDERED QUEUE TAG with regard to the ordering boundaries (per initiator or per target-lun basis). The question could not be decided, and a vote will be requested at the next plenary. Does ordering apply to the set of I/O processes within the LUN or the set of I/O processes for a specific initiator? There was significant disagreement with the addition of a new message for handling the continuation of ACA. The methods of clearing ACA were also debated. 15. CHANGE INQUIRY Command (92-85) [Penokie] George Penokie's effort to retrict his proposed new command was not successful as several members questioned why there was no ability to revert to factory default information in Inquiry. The possibility of errors during the execution of the change means Check Reset has to be supported and if there are multiple initiators then Unit Attention has to be provided. The longer the discussion, the more it looked as if this capability should remain Vendor Unique, or be implemented as a form of Mode Select. George agreed to consider withdrawing the proposal. 16. Tolerance Analysis of P-cable Connection (92-91) [McGrath] This topic was discussed by the connector folks offline. The discussion reached a consensus of the issues. Bob Whiteman will prepare a study of the dimensions and a tolerance analysis for the next meeting. 17. Data Compression Sense Codes (92-111) [Dougherty] George Penokie and Peter Dougherty worked out the specific codes. Peter will revise the proposal. 18. SCSI-2 TIB on Queuing [Gallant] John Gallant explained his objective in putting together a TIB for error recovery under SCSI-2 Command Queueing as a means to reduce the number of implementation variations. There was general concern on whether the techniques will be SCSI-3 compatible, or represent another point of difference. 19. Patent Issues [Allan] Dal Allan spoke on the patent issues currently affecting the disk drive industry. 20. Working Group Schedule The General Working Group meeting schedule is: Date Location Host ------------------ ----------------- ------------------------ Jul 20-24, 1992 Rochester, MN IBM Sep 21-25, 1992 Toronto, Canada Canstar Nov 9-13, 1992 Sunnyvale, CA Amdahl Jan 11-15, 1993 Sunnyvale, CA Tandem Mar 15-19, 1993 Orange County, CA Emulex May 17-21, 1993 Sante Fe, NM Los Alamos Jul 19-23, 1993 Boston, MA DEC Sep 13-17, 1993 Poughkipsee, NY IBM Nov 8-12, 1993 Colo Sprg, CO NCR The tentative schedule for the X3T9.2 July working group week is: Day Meeting Times --------- ---------------------------------- -------------- Monday Small Form Factor Committee 9:00a - 1:00p ATA Extensions 2:00p - 5:00p Tuesday X3T9.2 General Working Group 9:00a - 1:00p SPI Working Group 2:00p - 8:00p Wednesday X3T9.2 General Working Group 9:00a - 5:00p Thursday SCSI Editors 9:00a - 5:00p 21. Adjournment The meeting was adjourned at 2:00 p.m. on Wednesday May 20, 1992.