Accredited Standards Committee X3, Information Processing Systems Doc. No.: X3T9.2/93-15 Date: January 29,1993 To: Membership of X3T9.2 From:William Dallas, Editor of CAM-2 Subject:Minutes of CAM-2 Study Group Meeting (01/11/93) ___________________________________________________________________________ Agenda 1.0Opening Remarks 2.0Attendance and Introductions 3.0Call for items to be placed on the current agenda 4.0Status of Action Items 5.0Backwards compatibility discussion - Dallas 6.0Operating System defined CCB sizes discussion - Dallas 7.0SCSI-3 Bus/target/lun addressing - Dallas 8.0Proposal for the development of a standard SIM port interface architecture for CAM - Monia 9.0Dual pathing to a device discussion - Dallas 10.0EDT tables and configuration driver (SCSI-3 transports) - Dallas 11.0Review of Action Items 12.0Next Meeting Items 13.0Randell Jesup's Public Review Comments 1.0Opening Remarks Bill Dallas, Editor of the CAM-2 Study Group, called the meeting to order at 9:30 a.m., Monday January 11, 1993. It was noted that the meeting was poorly attended and a copy of the attendance list was circulated for attendance and corrections. As is customary, the people attending introduced themselves. Feb 2 11:24 1993 X3T9.2/93-15 Page 2 2.0Attendance The following people attended the meeting: NameOrganizationTelephone --------------------------------------------------------------------------- Mr. Norm HarrisAdaptec, Inc.(408) 945-8600 ext. 2230 Mr. Scott SmyersApple Computer(408) 974-7057 Mr. Charles MoniaDigital Equipment Corp.(508) 841-6757 Mr. Ralph WeberDigital Equipment Corp.(603) 881-1497 Mr. William DallasDigital Equipment Corp.(603) 881-2508 Mr. Richard WhalenDigital Equipment Corp. (508) 841-2684 Mr. James Doucette Digital Equipment Corp.(508) 841-6564 Mr. Todd SprenkeTandem Computers(408) 285-6264 8 People Present 3.0Call for items to be placed on the current agenda Due to public review comments on CAM-1 from Mr. Randell Jesup of Commodore Business Machines it was decided to place the public review comments on the agenda. 4.0Status of Action Items No new status. 13.0 Randell Jesup's Public Review Comments [Dallas] Most of the morning was dedicated to the public review comments made by Randell Jesup. Bill Dallas presented a draft of responses to the comments for the study group members. The group went over the draft and made some text/editorial changes to the responses. Approximately 50 percent of the public review comments dealt with minor oversights with the CAM-1 document. The remainder of the comments dealt with section 10 (Target mode). It was noted that the target mode section of the specification has implementation holes. Bill Dallas mentioned that he had spoken to Mr. John Gallant and Mr. Richard Napolatano in reference to them and other volunteers reworking section 10 of the CAM-1 document. It was noted that both John and Richard were agreeable to the idea. The group also consulted with Mr. John B. Lohmeyer, Chair X3T9.2, for direction in determining whether the CAM-1 specification needed to be pulled and corrections made to the document. Due to the substantive changes needed to the document it was decided to pull the document for rework. Feb 2 11:24 1993 X3T9.2/93-15 Page 3 Bill Dallas will write a response letter for review and approval of the X3T9.2 membership. 5.0Backwards compatibility discussion [Dallas] Bill Dallas presented an idea to preserve backwards compatibility for CAM-1 CCB's. This involved having the CAM-2 function codes mimic the CAM-1 function codes, with the only difference being bit 6 of the function code being set for CAM-2. This was discussed and Scott Smyers proposed that the function code be a single value signifying a CAM-2 CCB and that the function/opcode for CAM-2 be the the second item within the CCB header. Scott Smyers idea was discussed and agreed upon. A CAM-2 CAM Control Block (CCB) header was devised. The new header has both backwards compatibility with CAM-1 CCBs and guaranteed future compatibility (due to a CCB version field in the first two bytes). The following diagram shows the CAM-1 and CAM-2 headers, side-by-side: CAM-1 CAM-2 +------------------+ +------------------+ | | | CCB | +-- --+ +-- version --+ | CCB | | | +-- address --+ +------------------+ | | | CAM-2 | +-- --+ +-- Function --+ | | | Code | +------------------+ +------------------+ | | | | +-- CCB length --+ +-- CCB length --+ | | | | +------------------+ +------------------+ | Function Code | | CAM-2 CCB ID | +------------------+ +------------------+ | . | | . | | . | | . | | . | | . | Feb 2 11:24 1993 X3T9.2/93-15 Page 4 6.0Operating System defined CCB sizes discussion [Dallas] Bill Dallas lead a discussion on why CCB's must be a fixed size. One of the primary reasons that CAM was developed is to allow Hardware Vendors a consistent method of interfacing their hardware to a platform. He stated that if the CAM work group allowed an Operating System to define the CCB size/member size we will destroy one of the rationals for developing CAM. This lead to a discussion on the the merits and drawbacks of allowing the OS to define the size of the CCB's. It was decided to have the CAM-2 document ordered where a separate section of the document would define the size of and member offset of the CCB's. The group also decided to cast the CAM-2 document in terms of boolean, integer, and pointer fields etc. 7.0 SCSI-3 Bus/target/lun addressing [Dallas] Bill Dallas raised a concern with the new transports (FCP/GPP etc) being developed. There should be a consistent method of addressing across these models. A small discussion took place on how to get the different transport working groups to develop a consistent model across all transports. There was no decision on this issue. 8.0 Proposal for the development of a standard SIM port interface architecture for CAM [Monia] Jim Doucette of Digital Equipment Corporation presented a CAM hardware SIMport proposal/overview. Complete SIMport specification documents were given to all present. It was also mentioned that copies of the documents can be obtained over the internet by contacting Richard Whalen ( whalen@starch.enet.dec.com or voice 508-841-2684 ). Jim's presentation consisted of why a CAM SIM port architecture is needed, and the construction of the SIMport architecture. Jim requested that the CAM Study Group accept as an action item to consider the SIMport architecture as the standard intelligent hardware CAM CCB port It was decided by the group to make a decision on the request at the next CAM-2 study group meeting (March 1993). 9.0Dual pathing to a device discussion [Dallas] Bill Dallas was asked by Gary Stephens (GPP) to see if CAM-2 could start addressing the dual paths to a single device that the GPP model presents. This item did not get addressed due to lack of information. More information is needed on the dual pathing GPP model, and if other transports need this functionality. Feb 2 11:24 1993 X3T9.2/93-15 Page 5 10.0 EDT tables and configuration driver (SCSI-3 transports) [Dallas] Bill Dallas raised a concern about the sparse addressing space with some of the new transports, and the memory usage within the EDT. It was discussed if any specific actions were needed by the configuration driver to deal with the memory utilization problem. The overall consensus of the group felt that the memory utilization was an implementation specific problem and should not be address within CAM-2. 11.0Review of Action Items Action items from this meeting are: Bill Dallas to draft a formal response to Mr. Jesup for review by X3T9.2. Bill Dallas to draft changes to CAM-1 document in response to Randell Jesup's public review comments. Draft will not include Section 10 at this time. Individuals to review the SIMport architecture to be able to make an informed decision at the next CAM-2 meeting. 12.0 Next Meeting Items (March 15, 1993). Review CAM-1 changes (except Section 10). Decide if this study group wants to work on a standard CCB SIM port specification for the CAM-2 document. If any of person would like to place an item on the agenda for the next CAM-2 study group meeting, please contact the editor. Bill Dallas internet dallas@wasted.enet.dec.com voice (603) 881-2508 fax(603) 881-2257