T10/05-135r0 Voting Results on T10 Letter Ballot 05-134r0 on Forwarding FCP-3 to First Public Review Ballot closed: 2005/04/28 12:00 noon MDT Organization Name S Vote Add'l Info --------------------------------- -------------------- - ---- ---------- Adaptec, Inc. Tim Symons P Yes Agilent Technologies Pat Thaler P Yes Amphenol Interconnect Michael Wingard P Yes Broadcom Corp. Paul Griffith P Yes Brocade Robert Snively P Yes Cmnts Cisco Systems, Inc. Claudio DeSanti P Yes CNT David Peterson P Yes Crossroads Systems, Inc. DNV Dallas Semiconductor James A. Lott, Jr. P Yes Dell, Inc. Kevin Marks P Yes EMC Corp. David Black A Yes Emulex Robert H. Nixon P Yes Cmnts ENDL Ralph O. Weber P Yes FCI Douglas Wagner P Yes Foxconn Electronics Elwood Parsons P Yes Fujitsu Mike Fitzpatrick P Yes General Dynamics Nathan Hastad P Yes Hewlett Packard Co. Rob Elliott P Yes Cmnts Hitachi Cable Manchester Zane Daggett P Yes Hitachi Global Storage Tech. Dan Colegrove P Yes IBM Corp. George O. Penokie P No Cmnts Intel Corp. Robert Sheffield P Yes Iomega Corp. David Hawks P Yes Lexar Media, Inc. Martin Furuhjelm A Yes LSI Logic Corp. John Lohmeyer P Yes Maxtor Corp. Mark Evans P Yes Microsoft Corp. Jeff Mastro A Yes Molex Inc. Jay Neer P Yes Nvidia Corp. Mark Overby P Yes Panasonic Technologies, Inc Terence J. Nelson P Yes Philips Electronics William P. McFerrin P Yes Pivot3, Inc. Bill Galloway P Abs Cmnts PMC-Sierra Rachelle Trent P Yes QLogic Corp. Craig W. Carlson A Yes Quantum Corp. Paul Suhler A Yes Seagate Technology Gerald Houlder P No Cmnts Sierra Logic, Inc. William Martin P Yes Storage Technology Corp. Erich Oetting P Yes Cmnts Sun Microsystems, Inc. Vit Novak P Yes Texas Instruments Paul D. Aloisi P Yes Toshiba Yutaka Arakawa P Yes TycoElectronics Ashlie Fan P Yes Veritas Software Roger Cummings P Yes Vitesse Semiconductor Gregory Tabor P Yes Western Digital Curtis Stevens P Yes Xiotech Corp. Jeff Williams P Yes Ballot totals: (42:2:1:1=46) 42 Yes 2 No 1 Abstain 1 Organization(s) did not vote 46 Total voting organizations 7 Ballot(s) included comments This 2/3rds majority ballot passed. 42 Yes are more than half the membership eligible to vote minus abstentions [greater than 22] AND 42 Yes are at least 30 (2/3rds of those voting, excluding abstentions [44]) AND 42 Yes are equal to or exceed a quorum [15] Key: P Voter is principal member A Voter is alternate member Abs Abstain vote DNV Organization did not vote Cmnts Comments were included with ballot NoCmnts No comments were included with a vote that requires comments ************************************************************** Comments attached to Yes ballot from Robert Snively of Brocade: Brocade-01 (E) Page: ii Location: Abstract Problem Description: The abstract should add be rewritten to: "This standard describes the frame format and protocol definitions required to transfer commands and data between a SCSI (Small Computer System Interface) initiator and target using the Fibre Channel family of standards. The second version added optional retransmission, task ordering, and confirmation capabilities. This third version incorporates bi-directional commands, removes information that is now contained in other standards, and describes additional error recovery capabilities for the Fibre Channel Protocol." Suggested Solution: Make requested changes Brocade-02 (E) Page: xv Location: Foreword Problem Description: This document is not BSR INCITS 350. This text should have a nice big TBD with indicators about filling this in afterwards. Suggested Solution: Make requested changes Brocade-03 (E) Page: 5 Location: 3.2 Problem Description: The paragraph format for FC-FS-2 should be adjusted for the proper paragraph indent. Suggested Solution: Make requested changes Brocade-04 (E) Page: 43 Location: 9.2.2 Problem Description: The text of the second paragraph uses the phrase "(i.e., each FCP_DATA IU shall begin on a word boundary)." There is no definition of a word or of a word boundary in the document. I believe that the phrase should either be stricken or replaced with the words "(i.e., the two low-order bits of FCP_DATA_RO shall be zero.)" Suggested Solution: Make requested changes Brocade-05 (E) Page: 22 Location: 3.1 Problem Description: The word "word" is used several times throughout the document, but there is no glossary definition for the word. Suggested Solution: Add a glossary entry for the word "word" as follows: "word: a grouping of 4 bytes (32 bits) with a location beginning on a 4-byte boundary with respect to the beginning of an IU and treated as a unit." Brocade-06 (E) Page: many Location: many Problem Description: "can" should in all cases be replaced with "may". Suggested Solution: Make requested changes Brocade-07 (E) Page: 26 Location: 6.3.1 Problem Description: The fourth paragraph should be corrected to read: "An accept response code indicating other than REQUEST EXECUTED (see 6.3.5 and FC-LS) shall be provided if the PRLI Service Parameter page is incorrect or if the requested image pair is not established. Suggested Solution: Make requested changes Brocade-08 (E) Page: 93 Location: C.1 Problem Description: In C.4, C.11, and C.12, the word "cannot" should be replaced with "shall not". While I recognize that annexes do not show normative behavior, the words in these cases are used to describe behavior that is absolutely prohibited by this and other standards. Suggested Solution: Make requested changes Brocade-09 (E) Page: many Location: many Problem Description: The word "which" should be examined for proper usage in each of the places it is used and corrected to be removed, replaced with "that", to have the sentence rewritten, or to be unchanged depending on the correct meaning and according to the proper writing style guides. Suggested Solution: Make requested changes ************************************************************** Comments attached to Yes ballot from Robert H. Nixon of Emulex: Emulex comments on FCP-3 Revision 3f (March 23, 2005) Emulex-001 Page 2 Subclause 3.1.1 Access controls are not referenced in FCP-3. Remove the definitions related to access controls in 3.1.1, 3.1.2, and 3.1.3. Emulex-002 Page 11 subclause 4.2 It appears that between 6.3.4 and 10.2.10, there isn't really any way to suppress first burst transfer that is outside the scope of FCP-3. Change the first full sentence at the top of page 11 to "If the initiator and target have negotiated to disable the initial transfer ready (see 6.3.4), the initiator shall send an initial FCP_DATA IU after sending the FCP_CMND IU without transferring sequence initiative to the target (see 10.2.10), and the target shall process the initial FCP_DATA IU without having first sent an FCP_XFER_RDY IU." Emulex-003 subclause 4.2 It appears that between 6.3.4 and 10.2.10, there isn't really any way to suppress first burst transfer that is outside the scope of FCP-3. Change the last sentence of the third paragraph on page 11 to "If the initiator and target have negotiated to disable the initial transfer ready (see 6.3.4), the initiator shall send an initial FCP_DATA IU after sending the FCP_CMND IU without transferring sequence initiative to the target (see 10.2.10), and the target shall process the initial FCP_DATA IU without having first sent an FCP_XFER_RDY IU." Emulex-004 Page 19 Subclause 4.14 Since this subclause allows implicit PRLI, it should say "The Process Login (PRLI) ELS may be used to establish the FCP operating relationships..." Emulex-005 Page 19 Subclause 4.15 This subclause describes Link Management. Its last sentence is "Implicit login functions are allowed", which looks like it belongs at the end of the first paragraph of subclause 4.13 on Port Login/Logout. ************************************************************** Comments attached to Yes ballot from Rob Elliott of Hewlett Packard Co.: HPQ comment number 1 Page=16 Subtype=StrikeOut Author=relliott Comment= Introduction Delete "at data rates from 265 Mbits up to 10 Gbits per second" since that will become obsolete. HPQ comment number 2 Page=16 Subtype=Highlight Author=relliott Comment= Introduction Change "The Fibre Channel Protocol for SCSI, Third revision (FCP-3) standard" to "This standard" HPQ comment number 3 Page=16 Subtype=Highlight Author=relliott Comment= Introduction Change "Fibre Channel Protocol for SCSI, Third Version (FCP-3) standard" to "This standard" HPQ comment number 4 Page=18 Subtype=Highlight Author=relliott Comment= 1 Scope Change "the SCSI-3 Architecture Model - 3 (SAM-3)." to "SCSI Architecture Model - 3 (SAM-3)" HPQ comment number 5 Page=19 Subtype=StrikeOut Author=relliott Comment= 3.1.1 access controls Delete unused term: 3.1.1 access controls: Mechanisms allowing a managing application client to control the set of initiators that have access to a target. The access control is enforced by the target (see SPC-3). HPQ comment number 6 Page=19 Subtype=StrikeOut Author=relliott Comment= 3.1.2 access controls data: Delete unused term 3.1.2 access controls data: Information sent to the target by the managing application client that is used by the target to control the set of initiators that have access to the target (see SPC-3). HPQ comment number 7 Page=19 Subtype=StrikeOut Author=relliott Comment= 3.1.3 access controls enrollment state Delete unused term 3.1.3 access controls enrollment state: A state established in the target by the managing application client. The state governs the behavior of the target in controlling the set of initiators that have access to the target (see SPC-3) HPQ comment number 8 Page=19 Subtype=Highlight Author=relliott Comment= 3.1.4 acknowledged class Change "Acknowledged classes of service include Class 1, Class 2, and Class 4 service" to "(e.g., Class 1, Class 2, and Class 4)" HPQ comment number 9 Page=19 Subtype=Text Author=relliott Comment= 3.1.4 acknowledged class Add a definition: 3.1.xx unacknowledged class: Any class of service that does not acknowledge transfers (e.g., Class 3)(see FC-FS-2). HPQ comment number 10 Page=19 Subtype=Highlight Author=relliott Comment= 3.1.5 address identifier "used to identify source" s/b "used to identify the source" HPQ comment number 11 Page=19 Subtype=Text Author=relliott Comment= 3.1 Definitions If a cross reference like "(see SAM-3)" applies to the whole term being defined, I recommend using ". See SAM-3." rather than "(see SAM-3)". Use the parenthesis style only when the cross-reference applies to the object in front of the parenthesis. For example, this would mean to see SAM-3 for more info on "command": command: A request describing a unit of work to be performed by a device server. See SAM-3. while this would mean to see it for more info on "device server": command: A request describing a unit of work to be performed by a device server (see SAM-3). HPQ comment number 12 Page=20 Subtype=Highlight Author=relliott Comment= 3.1.15 Destination_Identifier Change "Destination_Identifier" to "Destination_Identifier (D_ID)" HPQ comment number 13 Page=22 Subtype=Highlight Author=relliott Comment= 3.1.54 Source_Identifier Change "Source_Identifier" to "Source_Identifier (S_ID)" HPQ comment number 14 Page=22 Subtype=Highlight Author=relliott Comment= 3.2 Abbreviations D_ID row "(see FC-FS-2)" s/b "(see 3.1.15)" HPQ comment number 15 Page=23 Subtype=Highlight Author=relliott Comment= 3.2 Abbreviations S_ID row "(see FC-FS-2)" s/b "(see 3.1.54)" HPQ comment number 16 Page=23 Subtype=Highlight Author=relliott Comment= 3.2 Abbreviations IU row "(see FC-FS-2)" s/b "(see 3.1.27)" HPQ comment number 17 Page=23 Subtype=Text Author=relliott Comment= 3.2 Abbreviations Add: ABTS Abort Sequence (see FC-FS-2) FCP_RJT FCP FC-4 Link Service Reject link service (see 8.3) PRLI Process Login ELS (see 6.3 and FC-LS) PRLO Process Logout ELS (see 6.4 and FC-LS) REC Read Exchange Concise ELS (see 6.5 and FC-LS) SRR Sequence Retransmission Request link service (see 8.2) HPQ comment number 18 Page=23 Subtype=Highlight Author=relliott Comment= 3.2 Abbreviations in PLOGI, change "Extended Link Service" to "ELS" HPQ comment number 19 Page=23 Subtype=Highlight Author=relliott Comment= 3.2 Abbreviations In FLOGI, change "Extended Link Service" to "ELS" HPQ comment number 20 Page=23 Subtype=Highlight Author=relliott Comment= 3.2 Abbreviations In LOGO, change "Extended Link Service" to "ELS" HPQ comment number 21 Page=27 Subtype=Highlight Author=relliott Comment= 4.2 Device management "Execute Command remote procedure call. (See SAM-3)." s/b "Execute Command procedure call (see SAM-3)." HPQ comment number 22 Page=27 Subtype=Underline Author=relliott Comment= 4.2 Device management This is not quite right: "An application client begins an FCP I/O operation when it invokes an Execute Command remote procedure call. (See SAM-3). The Execute Command call conveys a single request or a list of linked requests from the application client to the FCP service delivery subsystem." Problems: 1. Execute Command is now called a "procedure call," not a "remote procedure call" 2. Execute Command is a model for the collective operation of multiple SCSI transport protocol services. The application client doesn't invoke it, per se. 3. Task management functions also fall into the category of I/O operations - Execute Command only models commands. Each tmf has its own procedure call: ABORT TASK (), ABORT TASK SET(), etc. (see SAM-3 section 7.1) 4. In terms of protocol services, the application client invokes Send SCSI Command () or Send Task Management Request (). HPQ comment number 23 Page=27 Subtype=Underline Author=relliott Comment= 4.2 Device management After fixing the first sentences to cover task management functions too, change "one SCSI command" to "one SCSI command or task management function" HPQ comment number 24 Page=27 Subtype=Underline Author=relliott Comment= 4.2 Device management After fixing the first sentences to cover task management functions too, change "SCSI command" to "SCSI command or task management function" HPQ comment number 25 Page=27 Subtype=Underline Author=relliott Comment= 4.2 Device management "The FCP_CMND IU payload is the Send SCSI Command protocol service request (see SAM-3)" The initiator port sends the FCP_CMND IU payload to implement the Send SCSI Command protocol service request. HPQ comment number 26 Page=27 Subtype=Text Author=relliott Comment= 4.2 Device management Near "One FCP_DATA IU shall follow each FCP_XFER_RDY IU" clarify that each FCP_DATA IU contains one or more solicited data frames. HPQ comment number 27 Page=27 Subtype=Highlight Author=relliott Comment= 4.2 Device management Change "Device management" to "FCP I/O operations" HPQ comment number 28 Page=28 Subtype=Highlight Author=relliott Comment= 4.2 Device management "INTERMEDIATE CONDITION MET" s/b "INTERMEDIATE-CONDITION MET" HPQ comment number 29 Page=29 Subtype=Highlight Author=relliott Comment= 4.4 Precise delivery Change: "An application client may determine if a device server supports the precise delivery function by using the MODE SENSE and MODE SELECT commands to examine and set the enable precise delivery checking (EPDC) bit in the Fibre Channel Logical Unit Control page. See 10.3." to something like: "The ENABLE PRECISE DELIVERY CHECKING (EPDC) bit in the Fibre Channel Logical Unit Control mode page (see 10.3) indicates if precise delivery is enabled or disabled and may allow the application client to change the setting." HPQ comment number 30 Page=30 Subtype=Highlight Author=relliott Comment= 4.4 Precise delivery d) "by receipt of ... an FCP_DATA IU" implies that the entire FCP_DATA IU must be received. Is that the case, or is receipt of the first frame in the FCP_DATA IU sufficient? HPQ comment number 31 Page=30 Subtype=Highlight Author=relliott Comment= "the application client places a one byte unsigned integer in the COMMAND REFERENCE NUMBER field of each command..." The application client provides the Command Reference Number argument to the Send SCSI Command protocol service; the initiator port puts it in the FCP_CMND frame. HPQ comment number 32 Page=30 Subtype=Highlight Author=relliott Comment= 4.4 Precise delivery item e) "in the order of increasing CRN, highest CRN last." needs to account for wrapping (if you receive commands with CRN 254, 255, 1, and 2, you don't assume that they were received in order 1, 2, 254, 255, which this statement says) HPQ comment number 33 Page=30 Subtype=Text Author=relliott Comment= 4.4 Precise delivery a)b)c) list Use T10 style with ; endings HPQ comment number 34 Page=30 Subtype=Highlight Author=relliott Comment= 4.5 Confirmed completion Change "PRLI parameters are used to determine that confirmed completion is accepted by an initiator and may be requested by a target communicating with that initiator." to something like: The CONFIRMED COMPLETION ALLOWED field in the FCP Service Parameter page for PRLI request (see 6.3.4) and accept (6.3.5) is used to negotiate use of confirmed completion. HPQ comment number 35 Page=30 Subtype=Highlight Author=relliott Comment= 4.5 Confirmed completion Change "is provided by the confirmed completion function, optionally implemented by FCP-2 devices." to "may be provided by the optional confirmed completion function." HPQ comment number 36 Page=30 Subtype=Highlight Author=relliott Comment= 4.5 Confirmed completion FCP_CONF_REQ should be smallcaps HPQ comment number 37 Page=30 Subtype=Highlight Author=relliott Comment= 4.5 Confirmed completion Change "A target may invoke" to "If the CONFIRMED COMPLETION ALLOWED field is set to one in the PRLI accept FCP Service Parameter page, the target may request" HPQ comment number 38 Page=31 Subtype=Highlight Author=relliott Comment= 4.5 Confirmed completion After "INTERMEDIATE" add "or INTERMEDIATE-CONDITION MET" HPQ comment number 39 Page=31 Subtype=Highlight Author=relliott Comment= 4.5 Confirmed completion above first a)b) list After "completion" add ":" HPQ comment number 40 Page=31 Subtype=Text Author=relliott Comment= 4.5 Confirmed completion a)b) and a)b)c) list Use T10 style with ; endings HPQ comment number 41 Page=31 Subtype=Highlight Author=relliott Comment= 4.5 Confirmed completion Change "Confirmed completion shall not be requested for" to "Targets shall not request confirmed completion for" HPQ comment number 42 Page=31 Subtype=Highlight Author=relliott Comment= 4.6 Retransmission "as indicated by the PRLI bits," Specifically name the fields/bits HPQ comment number 43 Page=31 Subtype=Highlight Author=relliott Comment= 4.6 Retransmission "unsuccessfully transmitted data" is in the title, and the text discusses data retransmission. However, clause 12 also discusses retransmission of FCP_CMND, FCP_XFER_RDY, etc. Consider removing "data" HPQ comment number 44 Page=32 Subtype=Highlight Author=relliott Comment=Change "(REC and SRR)" to "(i.e., REC and SRR)" HPQ comment number 45 Page=32 Subtype=Square Author=relliott Comment= 4.8 Discovery of FCP caps Table 2 Change all the references to "6.3.4" to "6.3". 6.3 points to the PRLI request only; really some of them apply to the PRLI accept. HPQ comment number 46 Page=32 Subtype=Highlight Author=relliott Comment= 4.8 Discovery of FCP caps Table 2 For target overlay, change "MODE SENSE command" to "Disconnect-Reconnect mode page EMDP bit" HPQ comment number 47 Page=32 Subtype=Highlight Author=relliott Comment= 4.8 Discovery of FCP caps Table 2 For initiator overlay, change "MODE SENSE command" to "Fibre Channel Logical Unit Control page EPDC bit" and the reference to "4.4 and 10.3" HPQ comment number 48 Page=32 Subtype=Highlight Author=relliott Comment= 4.8 Discovery of FCP caps Table 2 For target overlay, change "MODE SENSE command" to "Fibre Channel Logical Unit Control page EPDC bit" and the reference to "4.4 and 10.3" HPQ comment number 49 Page=32 Subtype=Highlight Author=relliott Comment= 4.9 TMF Change: should end with an FCP_RSP IU completion status of Task Management function incorrect logical unit number (i.e., 09h) and may end with an FCP_RSP IU completion status of Task Management function complete (i.e., 00h) to: a) should end with an FCP_RSP IU with the RSP_CODE field set to 09h (i.e., task management function incorrect logical unit number); and b) may end with an FCP_RSP IU with the RSP_CODE field set to 00h (i.e., task management function complete); HPQ comment number 50 Page=33 Subtype=Text Author=relliott Comment= 4.9 Task mgmt Table 3 Add double lines below header row and above notes row HPQ comment number 51 Page=34 Subtype=Highlight Author=relliott Comment= 4.10 Clearing effects Table 4 and table 5 Change "N_Port or L_Port" to "FCP_Port" in the "PRLI parameters cleared" row. PRLI only applies to Nx_Ports, so the generic L_Port term (which includes NL_ and FL_ Ports) does not apply. HPQ comment number 52 Page=34 Subtype=Highlight Author=relliott Comment= 4.10 Clearing effects Table 4 Clearing effects In "CRN (Command Reference Number) (set to one)" should be "CRN set to one" HPQ comment number 53 Page=35 Subtype=Highlight Author=relliott Comment= 4.10 Clearing effects Table 5 Clearing effects "CRN (Command Reference Number) (set to one)" s/b "CRN set to one" HPQ comment number 54 Page=37 Subtype=StrikeOut Author=relliott Comment= 5.2 Use of WWN Delete "vital" HPQ comment number 55 Page=37 Subtype=Text Author=relliott Comment= 5 FC protocol overview Consider merging chapter 4 and 5. Chapter 5 seems to just continue describing general topics. HPQ comment number 56 Page=37 Subtype=Highlight Author=relliott Comment= 5.3 FCP Information Units (IUs) Move 5.3 into clause 9 FCP Information Units formats HPQ comment number 57 Page=38 Subtype=Text Author=relliott Comment= 5.3 IUs Table 6 Use double-line above notes HPQ comment number 58 Page=39 Subtype=Text Author=relliott Comment= 5.3 IUs Table 7 Use double-line above notes HPQ comment number 59 Page=40 Subtype=Text Author=relliott Comment= 5.4.1 FC-FS-2 frame header Right justify Bits Left justify Word HPQ comment number 60 Page=40 Subtype=Highlight Author=relliott Comment= 5.4 FC-FS-2 mappings to SCSI-3 functionality This is an awkward name for the section defining the frame header. Rename this to "FC-FS-2 frame header" and 5.4.1 to "FC-FS-2 frame header overview" HPQ comment number 61 Page=40 Subtype=Text Author=relliott Comment= 5.4 Frame header Consider moving 5.4 into clause 9 by the rest of the frame definitions (the IU contents) HPQ comment number 62 Page=40 Subtype=Text Author=relliott Comment= 5.4.2.x Add "field" after each title (already in 5.4.2.12 parameter field) HPQ comment number 63 Page=41 Subtype=Highlight Author=relliott Comment= 5.4.2.12 PARAMETER field After "task retry identification" add "(see 4.7)" HPQ comment number 64 Page=42 Subtype=Highlight Author=relliott Comment= 6.1 Link service requirements Change: "Process Login and Process Logout ELSs defined by FC-LS, the Process Login FCP Service Parameter pages defined in this standard, and the Read Exchange Concise ELS" to "PRLI and PRLO ELSs defined by FC-LS, the PRLI FCP Service Parameter pages defined by 6.3, and the REC ELS" HPQ comment number 65 Page=42 Subtype=Highlight Author=relliott Comment= 6.2 Overview of PRLI/PRLO Change "the following subclauses" to "6.3.4 and 6.3.5" HPQ comment number 66 Page=42 Subtype=Text Author=relliott Comment= 6 FCP link service definitions 8 FC-4 Link Service definitions Consider combining clause 6 and clause 8 since they both discuss link services HPQ comment number 67 Page=44 Subtype=Highlight Author=relliott Comment= 6.3.4 Process Login request page format Change "Process Login" to "PRLI" HPQ comment number 68 Page=44 Subtype=Text Author=relliott Comment= 6.3.4/5 Are these pages also used by PRLO in 6.4? If so rename them to not include PRLI/"Login" HPQ comment number 69 Page=45 Subtype=Highlight Author=relliott Comment= 6.3.4 FCP service parameter page word 3, bit 9 Change "the task retry identification function" to "task retry identification" globally. Add "(see 4.7)" after the first one in each section. HPQ comment number 70 Page=47 Subtype=Highlight Author=relliott Comment= 6.3.4 Process Login accept page format Change "Process Login" to "PRLI" HPQ comment number 71 Page=48 Subtype=Highlight Author=relliott Comment= 6.4 PRLO "The ACC shall present a response FCP Service Parameter page for the request FCP Service Parameter page." Where are these page formats defined? In they mean 6.3.4/6.3.5 then the names need to be changed to reflect PRLO also uses them. Add cross references and make sure naming is consistent. HPQ comment number 72 Page=50 Subtype=Text Author=relliott Comment= 8.1 FC-4 Link Services Expand the paragraph into one paragraph per field HPQ comment number 73 Page=51 Subtype=Highlight Author=relliott Comment= 8.2 SRR Table 13 Words s/b Word HPQ comment number 74 Page=51 Subtype=Highlight Author=relliott Comment= 8.2 SRR Table 14 Words s/b Word HPQ comment number 75 Page=52 Subtype=Highlight Author=relliott Comment= 8.3 FCP_RJT Table 15 Words s/b Word HPQ comment number 76 Page=54 Subtype=Text Author=relliott Comment= 9 IUs Since an FCP_DATA IU can be spread out over multiple frames, does that mean an FCP_CMND IU, FCP_XFER_RDY IU, or FCP_RSP IU can also be delivered in multiple frames? If not, there should be a statement in each section saying so. HPQ comment number 77 Page=54 Subtype=Highlight Author=relliott Comment= 9.1.1 FCM_CMND IU format Table 18 After "ADDITIONAL FCP_CDB" add "(if any") HPQ comment number 78 Page=54 Subtype=StrikeOut Author=relliott Comment= Table 18 Remove (MSB) and (LSB) from FCP_CDB and ADDITIONAL FCP_CDB fields. They have substructures. HPQ comment number 79 Page=54 Subtype=Highlight Author=relliott Comment= 9.1.1 FCP_CMND IU Change: an FCP_RSP IU containing a RSP_CODE field set to "FCP_CMND Fields Invalid". to: an FCP_RSP IU with the RSP_CODE field set to 02h (i.e., FCP_CMND fields invalid) HPQ comment number 80 Page=54 Subtype=Highlight Author=relliott Comment= 9.1.2.1 FCP_LUN field Change: "address of the destination logical unit in the attached subsystem. See SAM-3." to: "address of the logical unit (i.e., the logical unit number)(see SAM-3)." HPQ comment number 81 Page=55 Subtype=Highlight Author=relliott Comment= 9.1.2.1 FCP_LUN field Change "If the addressed logical unit does not exist, the target shall report that the logical unit number is incorrect or that the logical unit is not installed (see SAM-3 and SPC-3)." to: "If the addressed logical unit does not exist, the task manager shall follow the rules for selection of incorrect logical units defined in SAM-3" HPQ comment number 82 Page=56 Subtype=Text Author=relliott Comment= 9.1.2.5 TM FLAGS field Table 20 - TASK MGMT FLAGS Since more than one function at a time is prohibited, change this from a bit-by-bit table to an encoded value table: Code Task management function 40h CLEAR ACA 20h Obsolete 10h LOGICAL UNIT RESET 04h CLEAR TASK SET 02h ABORT TASK SET All others Reserved and adjust the wording above the table to reflect the change (e.g. say "field is set to a nonzero value" rather than "any bit is set to one") HPQ comment number 83 Page=56 Subtype=Highlight Author=relliott Comment= 9.1.2.5 TM Flags field CLEAR ACA description Change: the normal Task Management function complete RSP_CODE shall be contained in the returned FCP_RSP IU." to: The FCP_RSP IU shall contain a RSP_CODE field set to 00h (i.e., task management function complete). HPQ comment number 84 Page=56 Subtype=Highlight Author=relliott Comment= 9.1.2.5 TM FLAGS field Change: the FCP_RSP IU that indicates completion of the task management function shall contain a RSP_CODE field set to "FCP_CMND fields invalid". to: the FCP_RSP IU shall contain the RSP_CODE field set to 02h (i.e., FCP_CMND fields invalid). HPQ comment number 85 Page=56 Subtype=StrikeOut Author=relliott Comment= 9.1.2.4 TASK ATTRIBUTE field Delete "SIMPLE requests that the task be managed according to the rules for a SIMPLE task attribute. HEAD OF QUEUE requests that the task be managed according to the rules for a HEAD OF QUEUE task attribute. ORDERED requests that the task be managed according to the rules for an ORDERED task attribute. Mechanisms to assure delivery of commands to a device server in the correct order are described in 4.4. ACA requests that the task be managed according to the rules foran automatic contingent allegiance (ACA) task attribute." since table 19 already says that. HPQ comment number 86 Page=56 Subtype=Highlight Author=relliott Comment= 9.1.2.5 TASK MANAGEMENT FLAGS field "TASK ATTRIBUTES field" s/b "TASK ATTRIBUTE field" HPQ comment number 87 Page=57 Subtype=Highlight Author=relliott Comment= 9.1.2.5 TM Flags field "TASK ABORTED completion status" s/b "the TASK ABORTED status" HPQ comment number 88 Page=57 Subtype=Highlight Author=relliott Comment= 9.1.2.5 TASK MANAGEMENT FLAGS field LU RESET description "A task for an initiator other than the initiator that sent the LOGICAL UNIT RESET may be ended in the target. The initiator for that task shall determine by a timeout that the task did not finish. Subsequent retries fail because the task resources have been cleared in the target, so the initiator shall clear the Exchange resources with a recovery abort sequence. See 12.3." The "may" is incorrect - the target _must_ end tasks for other initiators according to SAM-3. This text is discussing details that are best left to SAM-3. This text might be trying to describe the ramifications of the SAM-3 rules; reword more as a note if that is the case. HPQ comment number 89 Page=57 Subtype=Highlight Author=relliott Comment= 9.1.2.5 TASK MANAGEMENT FLAGS field LU RESET description "A task for an initiator other than the initiator that sent the LOGICAL UNIT RESET may be completed by returning CHECK CONDITION status with the sense key set to UNIT ATTENTION and the additional sense code set to POWER ON, RESET, OR BUS DEVICE RESET OCCURRED. The initiator shall then clear all other tasks for that target and logical unit using the ABORT TASK task management function. See 9.1.3." This is all material covered by SAM-3 and shouldn't be mentioned here. Some problems with the above text: * SAM-3 allows more additional sense code options (e.g. POWER ON OCCURRED or BUS DEVICE RESET FUNCTION OCCURRED) * the logical unit _must_ create a unit attention condition - there's no "may" about it * it's not necessary for the initiator to use ABORT TASK (i.e. ABTS) on old tasks if it receives the unit attention condition from the logical unit - that itself proves the tasks are gone. HPQ comment number 90 Page=57 Subtype=Highlight Author=relliott Comment= 9.1.2.5 TASK MANAGEMENT FLAGS field CLEAR TASK SET description "A task for an initiator other than the initiator that sent the CLEAR TASK SET may be ended in the target. The initiator for that task shall determine by a timeout that the task did not finish. Subsequent retries fail because the task resources have been cleared in the target, so the initiator shall clear the Exchange r esources with a recovery abort sequence. See 12.3." The "may be ended" is too vague - SAM-3 requires they be ended. This is discussing details that are best left to SAM-3. This text might be trying to describe the ramifications of the SAM-3 rules; reword more as a note if that is the case. HPQ comment number 91 Page=57 Subtype=Highlight Author=relliott Comment= 9.1.2.5 TASK MANAGEMENT FUNCTION flags CLEAR ACA description There is no such thing as "contingent allegiance" in SAM-3. This is restating rules best left to SAM-3. HPQ comment number 92 Page=58 Subtype=Highlight Author=relliott Comment= 9.1.2.5 TM Flags field NOTE 4 "TASK ABORTED completion status" s/b "the TASK ABORTED status" HPQ comment number 93 Page=58 Subtype=Highlight Author=relliott Comment= 9.1.2.7 RDDATA and WRDATA bits NOTE 5 Change: with the RSP_CODE field to "FCP_CMND fields invalid" to with the RSP_CODE field set to 02h (i.e., FCP_CMND fields invalid) HPQ comment number 94 Page=58 Subtype=Highlight Author=relliott Comment= 9.1.2.5 TASK MANAGEMENT FUNCTION flags ABORT TASK SET description "all tasks in the task set from the initiator requesting the ABORT TASK SET to be aborted" Actually, all tasks from the I_T nexus requesting the ABORT TASK SET are aborted, not all tasks from the initiator. This is restating SAM-3 rules that are best left to SAM-3. HPQ comment number 95 Page=58 Subtype=Highlight Author=relliott Comment= 9.1.2.6 ADDITIONAL FCP_CDB LENGTH field Change: "The value of the ADDITIONAL FCP_CDB LENGTH field shall be zero" to "The ADDITIONAL FCP_CDB LENGTH field shall be set to zero" HPQ comment number 96 Page=58 Subtype=Text Author=relliott Comment= Global "command" and "operation" are used inconsistently for unidirectional and bidirectional commands. I suggest this terminology be used: Type of command -> operations used by the command read command -> read operation(s) write command -> write operation(s) bidirectional command -> read operation(s) and write operation(s) There are several places where a rule is described for "read operations" that applies to the read portion of a bidirectional command or "write operations" that also applies to the write portion of a bidirectional command, but the rule is then restated for bidirectional commands. This could lead to inconsistency. If there is concern that "read command" might be interpreted as only the READ opcode and "write command" might be interpreted as only the WRITE opcode, introduce/define them as "read command (e.g., INQUIRY, REPORT LUNS, MODE SENSE, READ, and RECEIVE DIAGNOSTIC RESULTS)" and "write command (e.g., MODE SELECT, SEND DIAGNOSTIC, and WRITE)". HPQ comment number 97 Page=59 Subtype=Text Author=relliott Comment= 9.1.3 ABORT TASK This does not belong in the FCP_CMND IU subclause. Move the details to somewhere else (e.g. 4.9) and just have the FCP_CMND subclause point there from the TASK MANAGEMENT FLAGS field table 20 as "Note: the ABORT TASK task management function is described in 4.9") HPQ comment number 98 Page=60 Subtype=Highlight Author=relliott Comment= 9.2.1 FCP_XFER_RDY overview "FCP_XFER_RDY IUs shall be transmitted preceding each write FCP_DATA IU" mixes plural and singular, and s/b "An FCP_XFER_RDY IU shall be transmitted preceding each write FCP_DATA IU". HPQ comment number 99 Page=60 Subtype=Text Author=relliott Comment= 9.2.2 FCP_DATA_RO In this subclause add "This is the "Offset of the data being transferred" field defined in FC-FS-2. And add "This is the "Length of the data being transferred" field defined in FC-FS-2. HPQ comment number 100 Page=60 Subtype=Highlight Author=relliott Comment= 9.2.3 FCP_BURST_LEN field Change "disconnect-reconnect page of MODE SELECT and MODE SENSE commands. See 10.2.7." to "Disconnect-Reconnect mode page (see 10.2.7)" HPQ comment number 101 Page=60 Subtype=Highlight Author=relliott Comment= 9.2.2 FCP_DATA_RO field Change "disconnect-reconnect page of the MODE SELECT and MODE SENSE commands (see 10.2)." to "Disconnect-Reconnect mode page (see 10.2.7)" HPQ comment number 102 Page=60 Subtype=Highlight Author=relliott Comment= 9.2.2 FCP_DATA_RO field Change "0 modulo 4" with "a multiple of 4". As written, it could either mean: (fcp_data_ro mod 4) = 0 (the intended meaning) fcp_data_ro = (0 mod 4) = 0 HPQ comment number 103 Page=61 Subtype=Text Author=relliott Comment= 9.3.1 FCP_DATA IU overview "If more than one FCP_DATA IU is used... the relative offset ... is used" This implies the relative offset is only used in the first FCP_DATA frame in a FCP_DATA IU. Is that the case, or must it be set correctly in each data frame? If the latter, then more rules are needed saying so. HPQ comment number 104 Page=61 Subtype=Text Author=relliott Comment= 9.3.1 FCP_DATA IU overview Add some text like this to explain that FCP DATA IU means the entire sequence of solicited data frames, not an individual frame: An FCP_DATA IU is a sequence (see 5.3) of one or more solicited data frames. The last frame of an FCP_DATA IU for a write operation transfers Sequence Initiative. The last frame of an FCP_DATA IU for a read operation holds Sequence Initiative. HPQ comment number 105 Page=61 Subtype=Highlight Author=relliott Comment= 9.3.1 FCP_DATA IU "Class of Service" s/b "class of service" HPQ comment number 106 Page=61 Subtype=Highlight Author=relliott Comment= 9.3.1 FCP_DATA IU Change: the target shall post the error code "FCP_DATA Parameter mismatch with FCP_DATA_RO" in the FCP_RSP_INFO field of the FCP_RSP IU. to: the target shall return an FCP_RSP IU with the RSP_CODE field set to 03h (i.e., FCP_DATA parameter mismatch with FCP_DATA RO). HPQ comment number 107 Page=61 Subtype=Highlight Author=relliott Comment= 9.3.1 FCP_DATA IU overview Change "parameters of the disconnect-reconnect page of the MODE SENSE and MODE SELECT commands" to "Disconnect-Reconnect mode page" globally. Add "(see 10.2)" after the first use in each section. HPQ comment number 108 Page=61 Subtype=Highlight Author=relliott Comment= the PRLI FCP Service Parameters specify WRITE FCP_XFER_RDY DISABLED" to "If the WRITE FCP_XFER_RDY DISABLED bit is set to one in the PLRI FCP Service Parameter page (see 6.3)" HPQ comment number 109 Page=61 Subtype=Highlight Author=relliott Comment= 9.3.1 FCP_DATA IU Change: "PRLI service parameter DATA OVERLAY ALLOWED for the initiator is zero," to: "DATA OVERLAY ALLOWED bit is set to one in the PLRI FCP Service Parameter page (see 6.3)" HPQ comment number 110 Page=61 Subtype=Highlight Author=relliott Comment= EMDP bit in the disconnect-reconnect page (see 10.2) of the MODE SELECT and MODE SENSE commands." to "the EMDP bit in the Disconnect-Reconnect mode page (see 10.2)." HPQ comment number 111 Page=62 Subtype=Highlight Author=relliott Comment= 9.3.2 FCP_DATA IUs Add paragraph break between "target. The initiator" since the rest of the paragraph is not only applicable during first bursts. HPQ comment number 112 Page=62 Subtype=Text Author=relliott Comment= 9.3.2 FCP_DATA IUs Clarify that commands that fail with protocol-level errors ("the amount of data requested or transferred does not match the number of bytes calculated from FCP_DL and FCP_RESID...") need to result in CHECK CONDITION status rather than GOOD status if the recovery procedures fail. HPQ comment number 113 Page=63 Subtype=Highlight Author=relliott Comment= 9.4.1 FCP_RSP IU "The FCP_RSP IU shall return the completion status of all task management functions using the FCP_RSP_INFO field. "all task management functions" is too broad. An ABORT TASK task management function does not involve an FCP_RSP IU. Change to: The target shall send a FCP_RSP IU for each task management function delivered with an FCP_CMND IU, indicating the completion status of the task management function in the RSP_CODE field. HPQ comment number 114 Page=64 Subtype=Highlight Author=relliott Comment= 9.4.1 FCP_RSP IU overview Table 22 After "FCP_RSP_INFO (m bytes long)" add "(if any)(see table 23 in 9.4.16)" HPQ comment number 115 Page=64 Subtype=StrikeOut Author=relliott Comment= 9.4.1 FCP_RSP IU overview Table 22 Delete (MSB) and (LSB) from FCP_SNS_INFO, since it has substructures HPQ comment number 116 Page=64 Subtype=StrikeOut Author=relliott Comment= 9.4.1 FCP_RSP IU overview Table 22 Delete (MSB) and (LSB) from FCP_RSP_INFO, since it has substructures HPQ comment number 117 Page=64 Subtype=Highlight Author=relliott Comment= 9.4.1 FCP_RSP IU overview Table 22 After "FCP_SNS_INFO (n bytes long)" add "(if any)" HPQ comment number 118 Page=64 Subtype=Highlight Author=relliott Comment= 9.4.1 FCP_RSP IU overview Table 22 After "FCP_BIDIRECTIONAL_READ_RESID" add "(if any)" HPQ comment number 119 Page=64 Subtype=Highlight Author=relliott Comment= Global including Table 22 In tables, change "RESERVED" from smallcaps to "Reserved" in mixed case HPQ comment number 120 Page=64 Subtype=Highlight Author=relliott Comment= 9.4.3 FCP_BIDI_RSP "are present." is accurate when referring to the FCP_BIDIRECTIONAL_READ_RESID field, which may or may not be in the IU (the IU is truncated if not present). The FCP_BIDIR_READ_RESID_UNDER and _OVER bits are always present, though. They're just "set to zero" if FCP_BIDI_RSP is set to zero. HPQ comment number 121 Page=65 Subtype=Highlight Author=relliott Comment= 9.4.8 FCP_RESID_OVER "The application client should examine the FCP_RESID field" This permissive wording might be part of the reason some HBAs don't always notice overflows/underflows. Upgrade to "shall" HPQ comment number 122 Page=65 Subtype=Highlight Author=relliott Comment= 9.4.7 FCP_RESID_UNDER "The application client should examine the FCP_RESID field" This permissive wording might be part of the reason some HBAs don't always notice overflows/underflows. Upgrade to "shall" HPQ comment number 123 Page=66 Subtype=Text Author=relliott Comment= 9.4.12 FCP_RESID field Reorder the text to put all the underflow sentences together and all the overflow sentences together. HPQ comment number 124 Page=67 Subtype=Highlight Author=relliott Comment= The number shall be 4, or 8." to "The FCP_RSP_LEN field shall be set to either 00000004h or 00000008h. HPQ comment number 125 Page=67 Subtype=StrikeOut Author=relliott Comment= 9.4.15 FCP_RSP_LEN Delete "Other values of length are reserved for future standardization." which is true of every field which has undefined values. (matching a comment received in SAS letter ballot) HPQ comment number 126 Page=68 Subtype=Highlight Author=relliott Comment= Change: indicates "Task Management function failed" to: "is set to 05h (i.e., task management function failed)" HPQ comment number 127 Page=68 Subtype=Highlight Author=relliott Comment= Values 04h and 05h are not valid responses to SCSI commands. Replace this sentence with a footnote in the table for 04h, 05h, and 09h that says: "Only valid when responding to a task management function" HPQ comment number 128 Page=68 Subtype=Highlight Author=relliott Comment= 9.4.16 FCP_RSP_INFO field Table 24 "Task Management" s/b "Task management" 4 times HPQ comment number 129 Page=68 Subtype=Highlight Author=relliott Comment= 9.4.16 FCP_RSP_INFO field Table 24 Parameter s/b parameter HPQ comment number 130 Page=68 Subtype=Highlight Author=relliott Comment= Table 23 "RESERVED" in bytes 4-7 Add "(if any)", since the length is allowed to be 4 rather than 8 in 9.4.15. HPQ comment number 131 Page=70 Subtype=Highlight Author=relliott Comment= 10.1 Mode page code overview Table 25 Capitalize R in "reconnect" HPQ comment number 132 Page=70 Subtype=Highlight Author=relliott Comment= 10.1 Mode page code overview Table 25 Change "page" to "mode page" in each row in this table HPQ comment number 133 Page=70 Subtype=Highlight Author=relliott Comment= 10.2.1 Disconnect-Reconnect mode page "disconnect-reconnect page" s/b "Disconnect-Reconnect mode page" HPQ comment number 134 Page=70 Subtype=StrikeOut Author=relliott Comment= 10.1 Overview of mode page codes Delete "block descriptors and the" HPQ comment number 135 Page=70 Subtype=Highlight Author=relliott Comment= 10.1 Overview of mode pages Change "pages" to "mode pages" HPQ comment number 136 Page=70 Subtype=Highlight Author=relliott Comment= 10.1 Overview of mode pages Change "mode page codes" to "mode pages" HPQ comment number 137 Page=71 Subtype=Highlight Author=relliott Comment= 10.2.1 Disconnect-reconnect mode page Change "FC-AL-2 loops" to "arbitrated loops (see FC-AL-2)" HPQ comment number 138 Page=71 Subtype=Highlight Author=relliott Comment= 10.2.1 Disconnect-Reconnect Table 26 "Disconnect-reconnect page" s/b "Disconnect-Reconnect mode page" HPQ comment number 139 Page=72 Subtype=Highlight Author=relliott Comment= 10.2.7 MAXIMUM BURST SIZE field Change "transfer to the initiator or request from the initiator." to "transfer to the initiator in a single Data-In FCP_DATA IU or request from the initiator in an FCP_XFER_RDY IU." HPQ comment number 140 Page=73 Subtype=Text Author=relliott Comment= 10.2.8 EMDP bit Clarify that for bidirectional commands, EMDP applies independently to the read data and the write data. With EMDP=0, the read sequences must be in order with relation to themselves and the write sequences must be in order with relation to the write sequences, but there is no read-to-write sequence ordering requirement affected by this bit. (a given command may have certain requirements for interleaving or not, but EMDP doesn't override those) HPQ comment number 141 Page=73 Subtype=Highlight Author=relliott Comment= 10.2.9 FAA, FAB, FAC bits Change "in a loop configuration" to "attached to an arbitrated loop (see FC-AL-2)" HPQ comment number 142 Page=73 Subtype=Highlight Author=relliott Comment= 10.2.9 FAA, FAB, FAC bits "The FAB bit controls arbitration when the initiator wishes to send one or more FCP_XFER_RDY IU frames to a target." is wrong. The initiator does not send FCP_XFER_RDY frames. This should probably be "when the target wishes to send...to an initiator." IBM will probably complain about "wishes" too. HPQ comment number 143 Page=73 Subtype=Highlight Author=relliott Comment= 10.2.9 FAA, FAB, FAC bits "or when the initiator wishes to send an FCP_CMND IU frames to target." How does a mode page field, by definition in a target device, place a requirement on an initiator? Perhaps this means if the target port is really a target/initiator port, it controls the functionality of the initiator role? HPQ comment number 144 Page=73 Subtype=Highlight Author=relliott Comment= 10.2.10 FIRST BURST SIZE field Change "write transfer ready is disabled" to "WRITE FCP_XFER_RDY DISABLED (smallcaps) is negotiated as being set to one in the PRLI FCP Service Parameter page (see 6.3.4)". Or in 6.3.4 define 'write transfer ready" and just add "(see 6.3.4)" here. Also change similar wording multiple times in this section HPQ comment number 145 Page=74 Subtype=StrikeOut Author=relliott Comment= 10.4.1 FC Port Control mode page Remove "The page shall not be implemented by logical units other than LUN 0." Any logical unit should be allowed to implement this if it wants. HPQ comment number 146 Page=74 Subtype=Highlight Author=relliott Comment= 10.3 FC LU Control mode page "CRN field" Change CRN to small caps. Add "in the FCP_CMND IU (see 9.1.2.2)" HPQ comment number 147 Page=74 Subtype=Highlight Author=relliott Comment= 10.4.1 FC Port Control mode page Change "The page" to "This mode page" throughout the paragraph HPQ comment number 148 Page=74 Subtype=Highlight Author=relliott Comment= 10.4.1 FC Port Contol mode page Change "page" to "mode page" throughout the section HPQ comment number 149 Page=75 Subtype=Highlight Author=relliott Comment= 10.4.4 ALWLI bit Change "Single Connector Attach - 2 (SCA-2) SFF-8067 connector" to "SCA-2 connector (see SFF-8067)" HPQ comment number 150 Page=75 Subtype=Highlight Author=relliott Comment= 10.4.4 ALWLI bit Change "FC-AL-2 loop" to "arbitrated loop (see FC-AL-2)" HPQ comment number 151 Page=75 Subtype=Highlight Author=relliott Comment= 10.4.2 DTOLI bit Change "by an arbitrated loop" to "to an arbitrated loop (see FC-AL-2)" HPQ comment number 152 Page=75 Subtype=Highlight Author=relliott Comment= 10.4.3 DTIPE bit Change "arbitrated loop" to "arbitrated loop (see FC-AL-2)" HPQ comment number 153 Page=75 Subtype=Highlight Author=relliott Comment= 10.4.5 RHA bit Change "arbitrated loop" to "arbitrated loop (see FC-AL-2)" HPQ comment number 154 Page=76 Subtype=Highlight Author=relliott Comment= 10.4.6 DLM bit Change "FC-AL-2 loop" to "arbitrated loop (see FC-AL-2)" HPQ comment number 155 Page=76 Subtype=Highlight Author=relliott Comment= 10.4.8 PLPB bit Change "FC-AL-2 loop" to "arbitrated loop (see FC-AL-2)" HPQ comment number 156 Page=76 Subtype=Highlight Author=relliott Comment= 10.4.7 DDIS bit Change "arbitrated loop" to "arbitrated loop (see FC-AL-2)" HPQ comment number 157 Page=76 Subtype=Highlight Author=relliott Comment= 10.4.9 DTFD bit Change "by an arbitrated loop" to "to an arbitrated loop (see FC-AL-2)" HPQ comment number 158 Page=77 Subtype=Highlight Author=relliott Comment= 10.4.10 RR_TOVseq_init Change RR_TOV SEQ_INIT to small caps and smallcaps/subscript (twice above table 29) HPQ comment number 159 Page=78 Subtype=Text Author=relliott Comment= 11.1 Timers Table 30 Use double-line above notes HPQ comment number 160 Page=79 Subtype=Highlight Author=relliott Comment= 11.3 R_A_TOV "following receipt of the BA_ACC to ABTS" is unclear. Maybe "following receipt of the BAA_ACC response to ABTS"? HPQ comment number 161 Page=81 Subtype=Text Author=relliott Comment= 12.1.2 Sequence level error recovery There should be a cross-reference to 12.4 somewhere in 12.1.2 since that's where the details are provided. There should also be a reference to 12.5, since that is at the same level and it is apparently used by the recovery described in 12.4. HPQ comment number 162 Page=82 Subtype=Highlight Author=relliott Comment= 12.2.3 Error detection mechanism Twice on the page: "b) ... no ACK has been received for FCP_DATA IU(s)" implies that each FCP_DATA IU is a frame (since each frame is ACKed). Change to "ACKs have not been received for all the frames in an FCP_DATA IU". HPQ comment number 163 Page=82 Subtype=Highlight Author=relliott Comment= 12.2.3 Error detection mechanisms "classes of Service" s/b "classes of service" HPQ comment number 164 Page=82 Subtype=Highlight Author=relliott Comment= 12.2.2 FCP-3 error detection for all classes "read-type command" s/b "read command" HPQ comment number 165 Page=83 Subtype=Highlight Author=relliott Comment= 12.3.1 Recovery abort requirements Change "Sequence level error recovery." to "sequence level error recovery." HPQ comment number 166 Page=83 Subtype=Highlight Author=relliott Comment= 12.3 Change "Exchange level recovery" to "exchange level error recovery" HPQ comment number 167 Page=83 Subtype=Highlight Author=relliott Comment= 12.3.1 Recovery abort requirements Change requirements to overview HPQ comment number 168 Page=84 Subtype=Highlight Author=relliott Comment= 12.3.4 Additional error recover by initiator Change "Sequence level recovery" to "sequence level error recovery" HPQ comment number 169 Page=84 Subtype=Highlight Author=relliott Comment= 12.3.5 Additional error recovery by target Change "Sequence level recovery" to "sequence level error recovery" HPQ comment number 170 Page=85 Subtype=Highlight Author=relliott Comment= 12.4.1.3 FCP_CMND IU recovery "see figure C.1 and figure C.2" can probably also reference C.3 (acknowledged classes), so change to "see figure C.1, figure C.2, and figure C.3." HPQ comment number 171 Page=85 Subtype=Highlight Author=relliott Comment= 12.4.1Using information from REC Change "Sequence level recovery" to "sequence level error recovery" in header HPQ comment number 172 Page=85 Subtype=Highlight Author=relliott Comment= 12.4.1.3 FCP_CMND IU recovery Change "Sequence level recovery" to "sequence level error recovery" HPQ comment number 173 Page=85 Subtype=Highlight Author=relliott Comment= 12.4.1.4 FCP_XFER_RDY IU recovery Change "Sequence level recovery" to "sequence level error recovery" HPQ comment number 174 Page=85 Subtype=StrikeOut Author=relliott Comment= 12.4.1.3 FCP_CMND IU recovery Delete "using information from REC" which is not in the other 12.4.1.x titles HPQ comment number 175 Page=86 Subtype=Highlight Author=relliott Comment= 12.4.1.5 FCP_RSP IU recovery Change "Sequence level recovery" to "sequence level error recovery" HPQ comment number 176 Page=87 Subtype=Highlight Author=relliott Comment= 12.4.1.6 FCP_DATA IU recovery - write Change "Sequence level recovery" to "sequence level error recovery" HPQ comment number 177 Page=87 Subtype=Highlight Author=relliott Comment= 12.4.1.8 FCP_CONF IU recovery Change "Sequence level recovery" to "sequence level error recovery" HPQ comment number 178 Page=87 Subtype=Highlight Author=relliott Comment= 12.4.1.7 FCP_DATA IU recovery - read Change "Sequence level recovery" to "sequence level error recovery" HPQ comment number 179 Page=89 Subtype=Text Author=relliott Comment= 12.5 Second-level error recovery In each of the 12.5.x titles, add the word "recovery" or "error recovery" e.g. 12.5.1 ABTS error recovery HPQ comment number 180 Page=90 Subtype=Text Author=relliott Comment= Annexes When creating the .pdf file, include the annex titles in the bookmarks HPQ comment number 181 Page=92 Subtype=Text Author=relliott Comment= A.4 Send SCSI Command Send SCSI command is 1 of the 4 steps, it is not a four-step service itself. HPQ comment number 182 Page=93 Subtype=Highlight Author=relliott Comment= A.6 title "services" s/b "function SCSI transport protocol services" HPQ comment number 183 Page=93 Subtype=Text Author=relliott Comment= A6. Task management services Describe the SCSI transport protocol services (Send Task Management Request, Task Management Request Received, etc.) defined in SAM-4 and how they are implemented by FCP-3 HPQ comment number 184 Page=96 Subtype=Text Author=relliott Comment= B.1.2 and B.1.5 Remove B.1.2 and move B.1.5 into its place. HPQ comment number 185 Page=97 Subtype=Text Author=relliott Comment= Tables B.6, B.7, B8 Show multiple FCP_XFER_RDY frames (and corresponding write FCP_DATA frames) as well. Show multiple read FCP_DATA frames. HPQ comment number 186 Page=97 Subtype=Text Author=relliott Comment= B.1.7 Table B.6 End table with doubleline. HPQ comment number 187 Page=97 Subtype=Text Author=relliott Comment= B.1.8 Table B.7 End table with doubleline. HPQ comment number 188 Page=98 Subtype=Text Author=relliott Comment= B.1.9 Table B.8 End table with doubleline. HPQ comment number 189 Page=99 Subtype=Text Author=relliott Comment= B.1.10 Table B.9 End table with doubleline. HPQ comment number 190 Page=101 Subtype=Text Author=relliott Comment= B.1.12 Table B.11 End table with doubleline. HPQ comment number 191 Page=102 Subtype=Text Author=relliott Comment= B.2 write example Figure B.1 Line up each "ACK" with its arrow There is room to make this figure wider. HPQ comment number 192 Page=102 Subtype=Highlight Author=relliott Comment= B.2 FCP write, frame level Figure B.1 Change "write I/O operation" to "write operation" HPQ comment number 193 Page=104 Subtype=Highlight Author=relliott Comment= B.3 FCP read example Figure B.3 Change "read I/O operation" to "read operation" HPQ comment number 194 Page=104 Subtype=Text Author=relliott Comment= B.3 FCP read example Figure B.3 Add [ and ] like in figure B.1 identifying the scope of the sequences. Line up the arrows better. HPQ comment number 195 Page=110 Subtype=Text Author=relliott Comment= C.1 Introduction Figure C.4 Does anything prevent the target from sending an FCP_XFER_RDY after the ACK? The target doesn't know the ACK was lost... HPQ comment number 196 Page=114 Subtype=Highlight Author=relliott Comment= C.1 Introduction Figure C.8 can s/b may HPQ comment number 197 Page=114 Subtype=Highlight Author=relliott Comment= C.1 Introduction Figure C.8 Change "Sequence level recovery" to "sequence level error recovery" HPQ comment number 198 Page=114 Subtype=Highlight Author=relliott Comment= C.1 Introduction Figure C.8 Change "Exchange level recovery" to "exchange level error recovery" HPQ comment number 199 Page=129 Subtype=Highlight Author=relliott Comment= C.1 Introduction Figure C.23 can s/b may HPQ comment number 200 Page=140 Subtype=Highlight Author=relliott Comment= D.1.1 item 8) Change "device" to "peripheral device" HPQ comment number 201 Page=142 Subtype=Text Author=relliott Comment= E.2.1 ABTS Table E.1 Add another horizontal line between Bit 0=0 and Bit 0 = 1 HPQ comment number 202 Page=143 Subtype=Highlight Author=relliott Comment= Table E.2 Change validity to Validity HPQ comment number 203 Page=143 Subtype=Highlight Author=relliott Comment= byte I think FC-FS-2 just uses "SEQ_ID" when referring to this ************************************************************** Comments attached to No ballot from George O. Penokie of IBM Corp.: IBM-001 PDF pg 4, pg iv, Change History The change history needs to be deleted before this document goes to public review. IBM-002 PDF pg 16, pg xvi, Introduction This << The Fibre Channel Protocol for SCSI, Third revision (FCP-3) standard has six annexes. >> should be changed to << The Fibre Channel Protocol for SCSI, Third revision (FCP-3) standard has the following annexes. >> as there are only five. A number here is always wrong and there is no need for it. IBM-003 PDF pg 18, pg 1, 2.2 Published standard and technical report references I find it hard to believe you have to reference FCP for anything norminative is this standard. This << ANSI X3.269-1996, Fibre Channel Protocol for SCSI (FCP) >> should be deleted. IBM-004 PDF pg 19, pg 2, 3.1.1 access controls: This << the set of initiators that have access to a target.>> should be << the set of initiator ports that have access to a SCSI target device.>> IBM-005 PDF pg 19, pg 2, 3.1.1 access controls: This <> should be << The access control is enforced by the SCSI target device>> IBM-006 PDF pg 19, pg 2, 3.1.1 access controls: This << sent to the target by the managing application client that is used by >> should be << sent to the SCSI target device by the managing application client that is used by >> IBM-007 PDF pg 19, pg 2, 3.1.2 access controls data: This << the target to control the set of initiators that have access to the target >> should be << the SCSI target device to control the set of initiators that have access to the SCSI target device >> IBM-008 PDF pg 19, pg 2, 3.1.3 access controls enrollment state: This << A state established in the target by the managing application client. The state governs the behavior of the target in controlling the set of initiators that have access to the target (see SPC-3)>> should be << A state established in the SCSI target device by the managing application client. The state governs the behavior of the SCSI target device in controlling the set of initiator ports that have access to the SCSI target device (see SPC-3)>> IBM-009 PDF pg 20, pg 3, 3.1.23 fully qualified exchange identifier: This << identify an FCP I/O operation. See 5.1. >> should be << identify an FCP I/O operation (see 5.1). >> IBM-010 PDF pg 20, pg 3, 3.1.28 initiator: This << the word 'initiator' also refers to an FCP_Port using the Fibre Channel protocol to perform >> should be << the word initiator also refers to an FCP_Port using the Fibre Channel protocol to perform >> IBM-011 PDF pg 20, pg 3, 3.1.29 initiator port identifier: This << Address a target uses to identify the initiator device (see SAM-3). >> should be <> IBM-012 PDF pg 21, pg 4, 3.1.31 logical unit: This <> should be << A SCSI target device object, containing a device server and task manager, that implements a device model and manages tasks to process commands sent by an application client (see SAM-3). >> IBM-013 PDF pg 22, pg 5, 3.1.58 tag: This << The initiator-specified component of a task identifier that uniquely identifies one task among the several tasks coming from an initiator to a logical unit. >> should be << The application client specified component of a task identifier that uniquely identifies one task among the several tasks coming from an application client to a logical unit. >> IBM-014 PDF pg 22, pg 5, 3.1.59 target: The definition titled << target >> should be << changed to << SCSI target port >> IBM-015 PDF pg 22, pg 5, 3.1.59 target: This << A SCSI device that receives SCSI commands and directs such commands to one or more logical units for execution. In this standard, the word 'target' also refers to an FCP_Port using the Fibre Channel protocol to perform the SCSI target functions defined by SAM-3 >> should be << A SCSI target device object that contains a task router and acts as the connection between device servers and task managers and the service delivery subsystem through which indications and responses are routed (see SAM-3). In this standard, the term SCSI target port also refers to an FCP_Port using the Fibre Channel protocol to perform the SCSI target port functions defined by SAM-3 >> IBM-016 PDF pg 22, pg 5, 3.1.62 task attribute: This should be << The queuing specification for a task (SIMPLE, ORDERED, HEAD OF QUEUE, ACA) (see SAM-3). >> should be << This should be << The queuing specification for a task (e.g., SIMPLE, ORDERED, HEAD OF QUEUE, ACA) (see SAM-3). >> IBM-017 PDF pg 23, pg 6, 3.2 Abbreviations This << FCP X3.269-1996, Fibre Channel Protocol for SCSI (see 2.2). Also: referring both to FCP and to this standard. >> should be << FCP Refers to this standard. >> IBM-018 PDF pg 23, pg 6, 3.2 Abbreviations Add SCS-2 to the list. IBM-019 PDF pg 23, pg 6, 3.2 Abbreviations This << SCSI-3 Small Computer System Interface-3, the SCSI architecture specified by SAM-3 and extended by the companion standards referenced in SAM-3. >> is completely incorrect and false statement and should be deleted. IBM-020 PDF pg 25, pg 8, 3.4 Editorial conventions, last paragraph This << precedence to resolve the conflicts is text; then tables; and finally figures. Exceptions >> should be << precedence to resolve the conflicts is text, then tables, and finally figures. Exceptions >> IBM-021 PDF pg 26, pg 9, 4.1 Structure and concepts, 3rd paragraph This << Fibre Channel Arbitrated Loop (FC-AL) is an alternative multiple port topology that allows communication between two ports on the loop or between a port on >> should be << Fibre Channel Arbitrated Loop-2 (FC-AL-2) is an alternative multiple port topology that allows communication between two ports on the loop or between a port on >> IBM-022 PDF pg 26, pg 9, 4.1 Structure and concepts Global - a.b.c lists and 1,2,3 lists should not have line spaces between the items in the list. This needs to be fixed. IBM-023 PDF pg 26, pg 9, 4.1 Structure and concepts Having all this space between the start of a sentence and the end of the sentence is not a good idea. Move the table anchor to it's own paragraph and this will not be a problem. IBM-024 PDF pg 27, pg 10, 4.2 Device management, 1st paragraph This << for the execution of one SCSI command, including the local storage address and characteristics of data to be transferred by the command. >> should be << for the processing of one SCSI command, including the local storage address and characteristics of data to be transferred by the command. >> IBM-025 PDF pg 27, pg 10, 4.2 Device management, 1st paragraph This << The execution of the individual steps of the protocol is consistent with the SCSI architectural model as defined by SAM-3. >> should be << The processing of the individual steps of the protocol is consistent with the SCSI architectural model as defined by SAM-3. >> IBM-026 PDF pg 28, pg 11, 4.2 Device management, 4th paragraph This << When the device server for the command has completed the interpretation of the command and has determined that read data transfer is required, the FCP_Port that is the target transmits a solicited data IU to the initiator containing the FCP_DATA IU payload. >> should be << When the device server for the command has completed the interpretation of the command and has determined that read data transfer is required, the target FCP_Port transmits a solicited data IU to the initiator FCP_Port. The solicited data IU shall contain the FCP_DATA IU payload. >> IBM-027 PDF pg 28, pg 11, 4.2 Device management, 5th paragraph This << The FCP_Port that is the initiator then transmits the solicited data IU to the target containing the FCP_DATA IU payload >> should be << The initiator FCP_Port then transmits the solicited data IU to the target FCP_Port. The solicited data IU shall contain the FCP_DATA IU payload >> IBM-028 PDF pg 28, pg 11, 4.2 Device management, 5th paragraph This << the FCP_Port that is the target transmits a solicited data IU to the initiator containing the FCP_DATA IU payload. >> should be << the target FCP_Port transmits a solicited data IU to the initiator FCP_Port. The solicited data IU shall contain the FCP_DATA IU payload. >> IBM-029 PDF pg 28, pg 11, 4.2 Device management, 8th paragraph This << The target shall present the FCP_RSP using the IU that allows command linking, I5 (see 5.3). The initiator shall continue the same Exchange with an FCP_CMND IU, beginning the next SCSI command. >> should be << The target FCP_Port shall present the FCP_RSP using the IU that allows command linking, I5 (see 5.3). The initiator FCP_Port shall continue the same Exchange with an FCP_CMND IU, beginning the next SCSI command. >> IBM-030 PDF pg 29, pg 12, 4.2 Device management, 8th paragraph This << linked in the FCP I/O operation except the last are executed in the manner described above. SAM-3 defines the >> should be << linked in the FCP I/O operation except the last are processed in the manner described above. SAM-3 defines the >> IBM-031 PDF pg 29, pg 12, 4.2 Device management, 9th paragraph This<< If command queueing resources are unavailable in the target when a command is received, the >> should be << If command queueing resources are unavailable in the logical unit when a command is received, the >> IBM-032 PDF pg 29, pg 12, 4.2 Device management, last paragraph This << SCSI allows the initiator function in any FCP_Port and the target function in any FCP_Port. For FCP I/O operations between a host and a peripheral subsystem, the host typically takes on the initiator role and the peripheral subsystem typically takes on the target role. For host to host communications, either one of the communicating pair may take on the initiator role. For device to device communications, typically used to implement extended copy and other third-party operations, the initiator role is adopted by the managing FCP device. >> should be << SCSI allows the SCSI initiator port function in any FCP_Port and the SCSI target port function in any FCP_Port. For FCP I/O operations between a host and a peripheral subsystem, the host typically takes on the SCSI initiator port role and the peripheral subsystem typically takes on the target role. For host to host communications, either one of the communicating pair may take on the SCSI initiator port role. For device to device communications, typically used to implement extended copy and other third-party operations, the SCSI initiator port role is adopted by the managing FCP device. >> IBM-033 PDF pg 29, pg 12, 4.4 Precise delivery of SCSI commands, 1st paragraph This << delivery and execution of SCSI commands is often not critical. Any changes in execution sequence caused by link failures or switch latencies are not important and the recovery and retry mechanisms may be executed while other activities are continued by the application client and the device server. >> should be << delivery and processing of SCSI commands is often not critical. Any changes in processing sequence caused by link failures or switch latencies are not important and the recovery and retry mechanisms may be processed while other activities are continued by the application client and the device server. >> IBM-034 PDF pg 29, pg 12, 4.4 Precise delivery of SCSI commands, 2nd paragraph This << the commands are guaranteed to be executed in order. >> should be << the commands are guaranteed to be processed in order. >> IBM-035 PDF pg 29, pg 12, 4.4 Precise delivery of SCSI commands, 3rd paragraph The term << EPDC>> should be in small caps. IBM-036 PDF pg 30, pg 13, 4.4 Precise delivery of SCSI commands, 5th paragraph The term << EPDC>> should be in small caps. IBM-037 PDF pg 30, pg 13, 4.4 Precise delivery of SCSI commands, item a) This << See tables 4 and 5 for the actions that cause the CRN to be transmitted by the initiator to be set to one and the CRN expected by the device server to be set to one. >> should be << See table 4 and table 5 for the actions that cause the CRN to be transmitted by the initiator FCP_Port to be set to one and the CRN expected by the device server to be set to one. >>. Note there are two changes in this sentence. IBM-038 PDF pg 30, pg 13, 4.4 Precise delivery of SCSI commands, last paragraph This << required for that command. For example, commands such as INQUIRY, TEST UNIT READY, REPORT LUNS and MODE SENSE/SELECT used for booting and initialization may use a CRN of zero. >> should be << required for that command (e.g., commands such as INQUIRY, TEST UNIT READY, REPORT LUNS and MODE SENSE/SELECT used for booting and initialization may use a CRN of zero). >> IBM-039 PDF pg 30, pg 13, 4.5 Confirmed completion of FCP I/O Operations, 1st paragraph This << PRLI parameters are used to determine that confirmed completion is accepted by an initiator and may be requested by a target communicating with that initiator. >> should be << PRLI parameters are used to determine that confirmed completion is accepted by an initiator FCP_Port and may be requested by a target FCP_Port communicating with that initiator FCP_Port. >> IBM-040 PDF pg 30, pg 13, 4.5 Confirmed completion of FCP I/O Operations, 2nd paragraph This << A target may invoke the confirmed completion function by setting the FCP_CONF_REQ bit to one in the FCP_RSP IU. Upon receiving the request in the FCP_RSP IU, the initiator shall transmit an FCP_CONF IU to the target, indicating to the target that the FCP_RSP IU has been received by the initiator. >> should be << A target FCP_Port may invoke the confirmed completion function by setting the FCP_CONF_REQ bit to one in the FCP_RSP IU. Upon receiving the request in the FCP_RSP IU, the initiator FCP_Port shall transmit an FCP_CONF IU to the target FCP_Port , indicating to the target FCP_Port that the FCP_RSP IU has been received by the initiator FCP_Port. >> IBM-041 PDF pg 31, pg 14, 4.5 Confirmed completion of FCP I/O Operations, 3rd paragraph This << The confirmed completion function allows the retry of unsuccessful notifications of errors and confirms that the initiator and the target both agree upon the state of a state dependent device. >> should be << The confirmed completion function allows the retry of unsuccessful notifications of errors and confirms that the initiator FCP_Port and the target FCP_Port both agree upon the state of a state dependent device. >> IBM-042 PDF pg 31, pg 14, 4.5 Confirmed completion of FCP I/O Operations, 6th paragraph This << If command linking is being performed, the target shall not request confirmed completion for an FCP_RSP IU containing INTERMEDIATE status. The target may request confirmed completion >> should be << If command linking is being performed, the target FCP_Port shall not request confirmed completion for an FCP_RSP IU containing INTERMEDIATE status. The target FCP_Port may request confirmed completion >> IBM-043 PDF pg 31, pg 14, 4.5 Confirmed completion of FCP I/O Operations, 1st a,b,c list item a) This << linked commands, or >> should be << linked commands; or >> IBM-044 PDF pg 31, pg 14, 4.5 Confirmed completion of FCP I/O Operations, 2nd a,b,c, list item a) This << may be used to confirm that an initiator has received an FCP_RSP IU reporting a SCSI CHECK CONDITION status, together with accompanying autosense data. Upon receiving the FCP_CONF IU, the target may discard its copy of the autosense data. >> should be << may be used to confirm that an initiator FCP_Port has received an FCP_RSP IU reporting a SCSI CHECK CONDITION status, together with accompanying autosense data. Upon receiving the FCP_CONF IU, the target FCP_Port may discard its copy of the autosense data. >> IBM-045 PDF pg 31, pg 14, 4.5 Confirmed completion of FCP I/O Operations, 2nd a,b,c list This list is not formed correctly. It should be a)...; b)...; and c).... IBM-046 PDF pg 31, pg 14, 4.5 Confirmed completion of FCP I/O Operations, 2nd a,b,c, list item b) This <> should be << has been successfully transferred to the initiator FCP_Port. That allows subsequent queued state dependent operations to be performed, since the FCP_CONF IU confirms that the FCP_RSP IU has been received by the initiator FCP_Port.>> IBM-047 PDF pg 31, pg 14, 4.5 Confirmed completion of FCP I/O Operations, 2nd a,b,c, list item b) This << be used to confirm that an initiator has received the FCP_RSP IU for targets that require state dependent synchronization with initiators. >> should be << be used to confirm that an initiator FCP_Port has received the FCP_RSP IU for target FCP_Ports that require state dependent synchronization with initiator FCP_Ports. >> IBM-048 PDF pg 31, pg 14, 4.6 Retransmission of unsuccessfully transmitted data The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-049 PDF pg 31, pg 14, 4.7 Task retry identification The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-050 PDF pg 32, pg 15, 4.7 Task retry identification, last paragraph This << Devices that agree to perform recovery shall support task retry identification. If both devices agree to support task retry identification, a task >> should be << FCP_Ports that agree to perform recovery shall support task retry identification. If both the initiator FCP_Port and target FCP_Port agree to support task retry identification, a task >> IBM-051 PDF pg 32, pg 15, 4.7 Task retry identification, last paragraph This << If the devices do not agree to support task retry identification, the PARAMETER field is zero >> should be << If the FCP_Ports do not agree to support task retry identification, the PARAMETER field is zero >> IBM-052 PDF pg 32, pg 15, 4.8 Discovery of FCP capabilities The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause including those it table 2. IBM-053 PDF pg 33, pg 16, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, 1st paragraph This are << FCP target objects >>? Do you mean << target FCP_Ports. >>? or something else. If so how does that relate to SAM-3 objects? This needs to be fixed. IBM-054 PDF pg 33, pg 16, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, 1st paragraph This << A 'Y' in the corresponding column of either table indicates the object is cleared to its default, saved, or initial value within the >> does not make sense. What is << initial value within the device >> supposed to mean? IBM-055 PDF pg 33, pg 16, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, 1st paragraph second sentence The term << upon >> is the wrong font. IBM-056 PDF pg 33, pg 16, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, 1st paragraph This << Rows indicating an effect for all initiator ports have the specified effect on all ports, regardless of the link that attaches the initiator port to the target. >> should be << Rows indicating an effect for all initiator FCP_Ports have the specified effect on all initiator FCP_Ports and all target FCP_Ports, regardless of the link that attaches the initiator FCP_Port to the target FCP_Port. >> IBM-057 PDF pg 34, pg 17, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 4 - in two places in header I have no idea what a << Target object >> is. I will assume it is the << Target FCP_Port >>. If so change it, if not then what is it? IBM-058 PDF pg 34, pg 17, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 4 The term <> should be changed to << initiator FCP_Port >> in all cases in this table. IBM-059 PDF pg 34, pg 17, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 4 Why are there some cells with only one << N >> for two rows? What is that supposed to mean? Every row entry needs to have a N, Y, or -. This needs to be fixed. IBM-060 PDF pg 34, pg 17, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 4 This << The Target shall clear the object only if ESTABLISH IMAGE PAIR is set >> should be << The target FCP_Port shall clear the object only if ESTABLISH IMAGE PAIR is set >> IBM-061 PDF pg 34, pg 17, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 4 This<< A target port should send PRLO to all logged-in initiator ports >> should be << A target FCP_Port should send PRLO to all logged-in initiator ports >> IBM-062 PDF pg 35, pg 18, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 5 (in two places in header) I have no idea what a << Target object >> is. I will assume it is the << Target FCP_Port >>. If so change it, if not then what is it? IBM-063 PDF pg 35, pg 18, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 5 The term <> should be changed to << initiator FCP_Port >> in all cases in this table. IBM-064 PDF pg 35, pg 18, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 5 Why are there some cells with only one << N >> for two rows? What is that supposed to mean? Every row entry needs to have a N, Y, or -. This needs to be fixed. IBM-065 PDF pg 35, pg 18, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 5 This << shall be individually aborted by the initiator via the recovery >> should be << shall be individually aborted by the initiator FCP_Port via the recovery >> IBM-066 PDF pg 35, pg 18, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 5 This << For multiple-LUN targets, CLEAR TASK SET, ABORT >> should be << For multiple-logical unit SCSI target devices, CLEAR TASK SET, ABORT >> IBM-067 PDF pg 35, pg 18, 4.10 Clearing effects of task management, FCP, FC-FS-2, FC-LS, and FC-AL-2 actions, Table 5 This << affect only the addressed LUN. >> should be << affect only the addressed logical unit. >> IBM-068 PDF pg 36, pg 19, 4.13 Port Login/Logout The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-069 PDF pg 37, pg 20, 5.1 FCP addressing and Exchange identification, 2nd paragraph This << Addressability of logical units uses the logical unit number provided in the FCP_CMND IU. >> should be << Addressability of logical units uses the LUN provided in the FCP_CMND IU. >> IBM-070 PDF pg 37, pg 20, 5.1 FCP addressing and Exchange identification, Last paragraph This << The target uses the OX_ID, and, if it has >> should be << The target FCP_Port uses the OX_ID, and, if it has >> IBM-071 PDF pg 37, pg 20, 5.2 Use of World Wide Names, 1st paragraph The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this paragraph IBM-072 PDF pg 37, pg 20, 5.2 Use of World Wide Names, Last paragraph This << unit having a LUN of 0 may be the same as the Node_Name of the target. The Worldwide_Name for the port shall be different from the Worldwide_Name for the node. >> should be << unit having a LUN of 0 may be the same as the Node_Name of the SCSI target device. The Worldwide_Name for the target FCP_Port shall be different from the Worldwide_Name for the node. >> IBM-073 PDF pg 37, pg 20, 5.3 FCP Information Units (IUs), 1st paragraph This << for IUs sent to targets, and in table 7 for IUs sent to initiators. Each >> should be << for IUs sent to target FCP_Ports, and in table 7 for IUs sent to initiator FCP_Ports. Each >> IBM-074 PDF pg 37, pg 20, 5.3 FCP Information Units (IUs) Having all this space between the start of a sentence and the end of the sentence is not a good idea. Move the table anchor to it's own paragraph and this will not be a problem. IBM-075 PDF pg 38, pg 21, 5.3 FCP Information Units (IUs), Table 6 title This <> should be << sent to target FCP_Ports >> IBM-076 PDF pg 38, pg 21, 5.3 FCP Information Units (IUs), Table 6 The term << SCSI-3 >> should be << SCSI >> as there is no such thing as SCSI-3. IBM-077 PDF pg 39, pg 22, 5.3 FCP Information Units (IUs), Table 7 title This <> should be << sent to initiator FCP_Ports >> IBM-078 PDF pg 39, pg 22, 5.3 FCP Information Units (IUs), Table 7 The term << SCSI-3 >> should be << SCSI >> as there is no such thing as SCSI-3. IBM-079 PDF pg 40, pg 23, 5.4.1 FC-FS-2 frame header Global - None of the field names are in small caps except parameter. All field names have to be changed to small caps in the tables and everywhere those values are used in text. IBM-080 PDF pg 41, pg 24, 5.4.2.12 PARAMETER field, 3rd paragraph The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this paragraph IBM-081 PDF pg 42, pg 25, 6.2 Overview of Process Login/Logout The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-082 PDF pg 42, pg 25, 6.2 Overview of Process Login/Logout, last paragraph This << Process Login has two actions that can be performed, selected by the ESTABLISH IMAGE PAIR bit (see 6.3.4): >> should be << Process Login has two actions that may be performed, selected by the ESTABLISH IMAGE PAIR bit (see 6.3.4): >> IBM-083 PDF pg 43, pg 26, 6.3.1 Use of Process Login by the Fibre Channel protocol, 3rd paragraph The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this paragraph IBM-084 PDF pg 43, pg 26, 6.3.1 Use of Process Login by the Fibre Channel protocol, 3dr paragraph This << Some capabilities require support by both the Originator and Responder before they can be used (see 6.3.4). >> should be << Some capabilities require support by both the Originator and Responder before they are able to be used (see 6.3.4). >> IBM-085 PDF pg 45, pg 28, Word 3, Bit 9: TASK RETRY IDENTIFICATION REQUESTED: This << then it shall be used between the initiator and all logical units for that port. The >> should be << then it shall be used between the initiator FCP_Port and all logical units address through that initiator FCP_Port. The >> IBM-086 PDF pg 45, pg 28, Word 3, Bit 8: RETRY: The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-087 PDF pg 45, pg 28, Word 3, Bit 7: CONFIRMED COMPLETION ALLOWED: The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> and the term << initiator function >> should be << initiator FCP_Port function >> and the term << target function >> should be << target FCP_Port function>>in all cases in this subclause. IBM-088 PDF pg 46, pg 29, Word 3, Bit 6: DATA OVERLAY ALLOWED: The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> and the term << initiator function >> should be << initiator FCP_Port function >> and the term << target function >> should be << target FCP_Port function>>in all cases in this subclause. IBM-089 PDF pg 46, pg 29, Word 3, Bit 6: DATA OVERLAY ALLOWED: This << in the application client buffer more than once during execution of a command. >> should be << in the application client buffer more than once during processing of a command. >> IBM-090 PDF pg 46, pg 29, Word 3, Bit 5: INITIATOR FUNCTION: The term << initiator >> should be << initiator FCP_Port >> and the term << initiator function >> should be << initiator FCP_Port function >> in all cases in this subclause. IBM-091 PDF pg 46, pg 29, Word 3, Bit 4: TARGET FUNCTION: The term << target >> should be << target FCP_Port >> and the term << target function >> should be << target FCP_Port function>>in all cases in this subclause. IBM-092 PDF pg 46, pg 29, Word 3, Bit 1: READ FCP_XFER_RDY DISABLED: This << Targets shall not send FCP_XFER_RDY on read operations. >> should be << Target FCP_Ports shall not send FCP_XFER_RDY on read operations. >> IBM-093 PDF pg 46, pg 29, Word 3, Bit 0: WRITE FCP_XFER_RDY DISABLED: The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-094 PDF pg 48, pg 31, 6.5 Read Exchange Concise (REC), item e) This << number of bytes transmitted by the target for a read. >> should be << number of bytes transmitted by the target FCP_Port for a read. >> IBM-095 PDF pg 49, pg 32, 7.1 Overview of FC-4 specific objects for the Fibre Channel protocol This << of the operations which can be performed to register objects with a Name >> should be << of the operations which are performed to register objects with a Name >> IBM-096 PDF pg 49, pg 32, 7.2 FC-4 Features object, 3dr paragraph This << Request CT_IU, which requests the FC-4 Features object for a specified >> should be << Request CT_IU, that requests the FC-4 Features object for a specified >> IBM-097 PDF pg 49, pg 32, 7.2 FC-4 Features object, 1st paragraph This << unit with logical unit number 0 of the FCP_Port, as specified by SPC-3. For >> should be << unit with LUN 0 of the FCP_Port, as specified by SPC-3. For >> IBM-098 PDF pg 50, pg 33, 8.2 Sequence Retransmission Request (SRR) The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-099 PDF pg 50, pg 33, Addressing:, 1st paragraph This << The S_ID field designates the initiator requesting the information retransmission. The D_ID field designates the target that is to receive the request. In the event that the target responds to the SRR with an FCP_RJT, the target shall return CHECK CONDITION status with the sense key set to HARDWARE ERROR and the additional sense code set to INITIATOR DETECTED ERROR MESSAGE RECEIVED. A target that has agreed during PRLI to support retransmission should not reject requests for retransmission of the requested frames >> should be << The S_ID field designates the initiator FCP_Port requesting the information retransmission. The D_ID field designates the target FCP_Port that is to receive the request. In the event that the target FCP_Port responds to the SRR with an FCP_RJT, the device server shall return CHECK CONDITION status with the sense key set to HARDWARE ERROR and the additional sense code set to INITIATOR DETECTED ERROR MESSAGE RECEIVED. A target FCP_Port that has agreed during PRLI to support retransmission should not reject requests for retransmission of the requested frames >> IBM-100 PDF pg 51, pg 34, Addressing:, 1st paragraph This << units that do not support retransmission on a target that supports retransmission for other logical units shall be >> should be << units that do not support retransmission on a target FCP_Port that supports retransmission for other logical units shall be >> IBM-101 PDF pg 51, pg 34, Payload for SRR FCP FC-4 Link Service request: Global - None of the field names are in small caps except parameter. All field names have to be changed to small caps in the tables and everywhere those values are used in text. IBM-102 PDF pg 51, pg 34, Payload for SRR FCP FC-4 Link Service request:, 2nd paragraph under table 13 This << Relative Offset of the lowest byte the initiator has identified as requiring retransmission. >> should be << Relative Offset of the lowest byte the initiator FCP_Port has identified as requiring retransmission. >> IBM-103 PDF pg 54, pg 37, 9.1.1 FCP_CMND IU format, 1st paragraph This << of bits is set in the FCP_CMND IU, the target shall respond with an FCP_RSP IU containing >> should be << of bits is set in the FCP_CMND IU, the target FCP_Port shall respond with an FCP_RSP IU containing >> IBM-104 PDF pg 54, pg 37, 9.1.2.1 FCP_LUN field, 1st paragraph This << logical unit in the attached subsystem. >> should be <> s IBM-105 PDF pg 55, pg 38, 9.1.2.1 FCP_LUN field, 2nd paragraph This << Each target shall accept an INQUIRY command addressed to logical unit with logical unit number 0. If logical unit numbers other than zero are supported by the target, logical unit number 0 shall implement >> should be << Each target FCP_Port shall accept an INQUIRY command addressed to LUN 0. If LUNs other than zero are supported by the SCSI target device, LUN 0 shall implement >> IBM-106 PDF pg 55, pg 38, 9.1.2.1 FCP_LUN field, Last paragraph This << the target shall report that the logical unit number is incorrect or that >> should be << the SCSI target device shall report that the logical unit number is incorrect or that >> IBM-107 PDF pg 55, pg 38, 9.1.2.2 COMMAND REFERENCE NUMBER field The term << initiator >> should be << initiator FCP_Port >> in all cases in this subclause. IBM-108 PDF pg 56, pg 39, 9.1.2.5 TASK MANAGEMENT FLAGS field, 1st paragraph This << shall be requested by the initiator (Exchange Originator) using >> should be << shall be requested by the initiator FCP_Port (Exchange Originator) using >> IBM-109 PDF pg 56, pg 39, 9.1.2.5 TASK MANAGEMENT FLAGS field, 1st paragraph This << function shall not be executed and the FCP_RSP IU that indicates completion of the task management >> should be << function shall not be processed and the FCP_RSP IU that indicates completion of the task management >> IBM-110 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, 5th paragraph in CLEAR ACA This << The CLEAR ACA is transmitted by the initiator (Exchange Originator) using a new Exchange. >> should be << The CLEAR ACA is transmitted by the initiator FCP_Port (Exchange Originator) using a new Exchange. >> IBM-111 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, 5th paragraph in CLEAR ACA This << It shall not be sent to a target with a NORMACA bit equal to zero in the INQUIRY data. >> should be << It shall not be sent to a logical unit with a NORMACA bit equal to zero in the INQUIRY data. >> IBM-112 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, 2nd paragraph LOGICAL UNIT RESET This << The LOGICAL UNIT RESET is transmitted by the initiator (Exchange Originator) using a new Exchange. LOGICAL UNIT RESET resets the internal states of the target and logical unit as shown in 4.10. >> should be << The LOGICAL UNIT RESET is transmitted by the initiator FCP_Port (Exchange Originator) using a new Exchange. LOGICAL UNIT RESET resets the internal states of the target FCP_Port and logical unit as shown in 4.10. >>. IBM-113 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, 2nd paragraph LOGICAL UNIT RESET This << cleared by the following mechanisms. >> should be << cleared by the following mechanisms: >> IBM-114 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, item a) logical unit reset This << A recovery abort sequence (see 12.3) may be generated by the initiator that sent the LOGICAL UNIT RESET for each task in the logical unit known to that initiator. >> should be << A recovery abort sequence (see 12.3) may be generated by the initiator FCP_Port that sent the LOGICAL UNIT RESET for each task in the logical unit known to that initiator FCP_Port;>> IBM-115 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, item b) LOGICAL UNIT RESET This << A task for an initiator other than the initiator that sent the LOGICAL UNIT RESET may be ended in the target. The initiator for that task shall determine by a timeout that the task did not finish. Subsequent retries fail because the task resources have been cleared in the target, so the initiator shall clear the Exchange resources with a recovery abort sequence. See 12.3. >> should be << A task for an initiator FCP_Port other than the initiator FCP_Port that sent the LOGICAL UNIT RESET may be ended in the logical unit. The initiator FCP_Port for that task shall determine by a timeout that the task did not finish. Subsequent retries fail as a result of the task resources have been cleared in the logical unit, so the initiator FCP_Port shall clear the Exchange resources with a recovery abort sequence. See 12.3; or >> IBM-116 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, item b) LOGICAL UNIT RESET This << A task for an initiator other than the initiator that sent the LOGICAL UNIT RESET may be completed by returning CHECK CONDITION status with the sense key set to UNIT ATTENTION and the additional sense code set to POWER ON, RESET, OR BUS DEVICE RESET OCCURRED. The initiator shall then clear all other tasks for that target and logical unit using the ABORT TASK task management function. See 9.1.3. >> should be << A task for an initiator FCP_Port other than the initiator FCP_Port that sent the LOGICAL UNIT RESET may be completed by returning CHECK CONDITION status with the sense key set to UNIT ATTENTION and the additional sense code set to POWER ON, RESET, OR BUS DEVICE RESET OCCURRED. The initiator FCP_Port shall then clear all other tasks for that target FCP_Port and logical unit using the ABORT TASK task management function. See 9.1.3. >> IBM-117 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, 1st paragraph CLEAR TASK SET This << tasks from all initiators in the specified task set to be aborted as defined >> should be << tasks from all initiator FCP_Ports in the specified task set to be aborted as defined >> IBM-118 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, 2nd paragraph CLEAR TASK SET This << is transmitted by the initiator (Exchange Originator) using a new Exchange. CLEAR TASK SET resets internal states of the target as shown in 4.10. >> should be << is transmitted by the initiator FCP_Port (Exchange Originator) using a new Exchange. CLEAR TASK SET resets internal states of the target FCP_Port as shown in 4.10. >> IBM-119 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, 2nd paragraph CLEAR TASK SET This << or more of the following mechanisms. >> should be << or more of the following mechanisms: >> IBM-120 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, item a) CLEAR TASK SET This <> should be << A recovery abort sequence (see 12.3) may be generated by the initiator FCP_Port that sent the CLEAR TASK SET for each task known to that initiator FCP_Port;>> IBM-121 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, item b) CLEAR TASK SET This << A task for an initiator other than the initiator that sent the CLEAR TASK SET may be ended in the target. The initiator for that task shall determine by a timeout that the task did not finish. Subsequent retries fail because the task resources have been cleared in the target, so the initiator shall clear the Exchange resources with a recovery abort sequence. See 12.3. >> should be << A task for an initiator FCP_Port other than the initiator FCP_Port that sent the CLEAR TASK SET may be ended in the logical unit. The initiator FCP_Port for that task shall determine by a timeout that the task did not finish. Subsequent retries fail because the task resources have been cleared in the logical unit, so the initiator FCP_Port shall clear the Exchange resources with a recovery abort sequence. See 12.3; or >> IBM-122 PDF pg 57, pg 40, 9.1.2.5 TASK MANAGEMENT FLAGS field, item b) CLEAR TASK SET This << A task for an initiator other than the initiator that sent the CLEAR TASK SET may be completed by returning CHECK CONDITION status with the sense key set to UNIT ATTENTION and the additional sense code set to POWER ON, RESET, OR BUS DEVICE RESET OCCURRED. The initiator shall then clear all other tasks for that target using the ABORT TASK task management function. See 9.1.3. >> should be << A task for an initiator FCP_Port other than the initiator FCP_Port that sent the CLEAR TASK SET may be completed by returning CHECK CONDITION status with the sense key set to UNIT ATTENTION and the additional sense code set to POWER ON, RESET, OR BUS DEVICE RESET OCCURRED. The initiator FCP_Port shall then clear all other tasks for that target FCP_Port using the ABORT TASK task management function. See 9.1.3. >> IBM-123 PDF pg 58, pg 41, 9.1.2.5 TASK MANAGEMENT FLAGS field, ABORT TASK SET The term << initiator >> should be << initiator FCP_Port >> in all cases in this subclause. IBM-124 PDF pg 58, pg 41, 9.1.2.7 RDDATA and WRDATA bits, 1st, 2nd, and 3rd paragraphs The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >>. IBM-125 PDF pg 58, pg 41, 9.1.2.7 RDDATA and WRDATA bits, 5th paragraph This << The target shall return CHECK CONDITION status with the sense key set to ILLEGAL REQUEST and the >> should be << The device server shall return CHECK CONDITION status with the sense key set to ILLEGAL REQUEST and the >> IBM-126 PDF pg 58, pg 41, 9.1.2.7 RDDATA and WRDATA bits, NOTE 5 This << Targets compliant to previous versions of this standard may terminate the command >> should be << Device servers compliant to previous versions of this standard may terminate the command >> IBM-127 PDF pg 59, pg 42, 9.1.3 Additional mechanisms for performing task management functions - ABORT TASK, 1st paragraph This << The ABORT TASK task management function causes the target to abort the specified task using the recovery abort protocol, if the task exists. The action is defined in SAM-3. The ABORT TASK is performed by the initiator (Exchange Originator) using the recovery abort (see 12.3). The specified Exchange shall be terminated by the initiator using the recovery abort. >> should be << The ABORT TASK task management function causes the device server to abort the specified task using the recovery abort protocol, if the task exists. The action is defined in SAM-3. The ABORT TASK is performed by the initiator FCP_Port (Exchange Originator) using the recovery abort (see 12.3). The specified Exchange shall be terminated by the initiator FCP_Port using the recovery abort. >> IBM-128 PDF pg 59, pg 42, 9.2.1 Overview and format of FCP_XFER_RDY IU The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-129 PDF pg 60, pg 43, 9.2.1 Overview and format of FCP_XFER_RDY IU, 3dr paragraph This << given FCP names for use in this document. >> should be << given FCP names for use in this standard>> IBM-130 PDF pg 60, pg 43, 9.2.2 FCP_DATA_RO field, 1st paragraph This<< This may be used by the target to request data out of order >> should be << This may be used by the target FCP_Port to request data out of order >> IBM-131 PDF pg 60, pg 43, 9.2.3 FCP_BURST_LEN field, 1st paragraph This << requests the transfer from the initiator of an IU of that length. The >> should be << requests the transfer from the initiator FCP_Port of an IU of that length. The >> IBM-132 PDF pg 61, pg 44, 9.3.1 FCP_DATA IU overview The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-133 PDF pg 61, pg 44, 9.3.2 FCP_DATA IUs for SCSI read and SCSI write operations Except as noted in the other comment on this subclause the term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-134 PDF pg 61, pg 44, 9.3.2 FCP_DATA IUs for SCSI read and SCSI write operations, 2dr paragraph This << data beyond FCP_DL be transferred, the target shall set the FCP_RESID_OVER bit (see 9.4.8) to one in the FCP_RSP IU and >> should be << data beyond FCP_DL be transferred, the device server shall set the FCP_RESID_OVER bit (see 9.4.8) to one in the FCP_RSP IU and >> IBM-135 PDF pg 62, pg 45, 9.3.3 FCP_DATA IUs for bidirectional SCSI commands Except as noted in the other 2 comments on this subclause the term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-136 PDF pg 62, pg 45, 9.3.3 FCP_DATA IUs for bidirectional SCSI commands, 3rd paragraph This << If a bidirectional command requested that data beyond FCP_DL be transferred, the target shall set the FCP_RESID_OVER bit (see 9.4.8) to one in the FCP_RSP IU and shall: >> should be << If a bidirectional command requested that data beyond FCP_DL be transferred, the device server shall set the FCP_RESID_OVER bit (see 9.4.8) to one in the FCP_RSP IU and shall: >> IBM-137 PDF pg 62, pg 45, 9.3.3 FCP_DATA IUs for bidirectional SCSI commands, Paragraph above 2nd a,b,c, list This << If a bidirectional command requests that data beyond the value specified in the FCP_BIDIRECTIONAL_READ_DL field be transferred, the target shall set the FCP_BIDI_READ_RESID_OVER bit (see 9.4.5) to one in the FCP_RSP IU and shall: >> should be << If a bidirectional command requests that data beyond the value specified in the FCP_BIDIRECTIONAL_READ_DL field be transferred, the device server shall set the FCP_BIDI_READ_RESID_OVER bit (see 9.4.5) to one in the FCP_RSP IU and shall: >> IBM-138 PDF pg 63, pg 46, 9.4.1 Overview and format of FCP_RSP IU, 5th paragraph This << If a SCSI device error is detected by a target while the target has Sequence Initiative for the Exchange associated with the error, the target should complete any Sequence that has already been started, keep Sequence Initiative and transmit an FCP_RSP IU with CHECK CONDITION status and the sense data that describes the error. If a SCSI device error is detected by a target while the target does not have Sequence Initiative for the Exchange associated with the error, it shall wait until Sequence Initiative has been returned and then transmit an FCP_RSP IU with CHECK CONDITION status and the sense data that describes the error. >> should be << If a SCSI device error is detected by a target FCP_Port while the target FCP_Port has Sequence Initiative for the Exchange associated with the error, the target FCP_Port should complete any Sequence that has already been started, keep Sequence Initiative and an FCP_RSP IU with CHECK CONDITION status and the sense data that describes the error transmitted. If a SCSI device error is detected by a device server while the target FCP_Port does not have Sequence Initiative for the Exchange associated with the error, the target FCP_Port shall wait until Sequence Initiative has been returned and then an FCP_RSP IU with CHECK CONDITION status and the sense data that describes the error shall be transmitted. >> IBM-139 PDF pg 63, pg 46, 9.4.1 Overview and format of FCP_RSP IU, 6th paragraph This << the target may implicitly terminate the affected Exchange. >> should be << the target FCP_Port may implicitly terminate the affected Exchange. >> IBM-140 PDF pg 65, pg 48, 9.4.6 FCP_CONF_REQ bit This << An FCP_CONF_REQ bit of one indicates that the initiator shall transmit an FCP_CONF IU to confirm receipt of the CP_RSP Sequence. An FCP_CONF_REQ bit of zero indicates that the initiator shall not transmit an FCP_CONF IU. >> should be << An FCP_CONF_REQ bit of one indicates that the initiator FCP_Port shall transmit an FCP_CONF IU to confirm receipt of the CP_RSP Sequence. An FCP_CONF_REQ bit of zero indicates that the initiator FCP_Port shall not transmit an FCP_CONF IU. >> IBM-141 PDF pg 65, pg 48, 9.4.10 FCP_RSP_LEN_VALID bit, 1st paragraph This << When the FCP_RSP_LEN_VALID bit is set to one, the content of the SCSI STATUS CODE field is not reliable and shall be ignored by the initiator. >> should be << When the FCP_RSP_LEN_VALID bit is set to one, the content of the SCSI STATUS CODE field is not reliable and shall be ignored by the application client. >> IBM-142 PDF pg 65, pg 48, 9.4.10 FCP_RSP_LEN_VALID bit, 2nd paragraph This << For task management functions transmitted to the target using an FCP_CMND IU, the FCP_RSP_LEN_VALID bit shall be set to one, the >> should be << For task management functions transmitted to the logical unit using an FCP_CMND IU, the FCP_RSP_LEN_VALID bit shall be set to one, the >> IBM-143 PDF pg 66, pg 49, 9.4.12 FCP_RESID field, 6th paragraph This << Targets are not required to verify that the data length implied by the contents of the CDB cause an overrun or underrun before beginning execution of a SCSI command. >> should be << Device servers are not required to verify that the data length implied by the contents of the CDB cause an overrun or underrun before beginning processing of a SCSI command. >> IBM-144 PDF pg 66, pg 49, 9.4.12 FCP_RESID field, NOTE 6 This << Some early target implementations presented the FCP_RSP IU without the FCP_RESID, FCP_SNS_LEN, and FCP_RSP_LEN fields if the FCP_RESID_UNDER, FCP_RESID_OVER, FCP_SNS_LEN_VALID, and FCP_RSP_LEN_VALID bits were all set to zero. Initiators should be tolerant of this non-standard behavior. >> should be << Some early device server implementations presented the FCP_RSP IU without the FCP_RESID, FCP_SNS_LEN, and FCP_RSP_LEN fields if the FCP_RESID_UNDER, FCP_RESID_OVER, FCP_SNS_LEN_VALID, and FCP_RSP_LEN_VALID bits were all set to zero. Application clients should be tolerant of this non-standard behavior. >> IBM-145 PDF pg 67, pg 50, 9.4.13 FCP_BIDIRECTIONAL_READ_RESID field, 4th paragraph This << Targets are not required to verify that the data length implied by the contents of the CDB cause an overrun or underrun before beginning execution of a SCSI command. >> should be << Device servers are not required to verify that the data length implied by the contents of the CDB cause an overrun or underrun before beginning processing of a SCSI command. >> IBM-146 PDF pg 67, pg 50, 9.4.16 FCP_RSP_INFO field, 1st paragraph This << failures detected during the execution of an FCP I/O operation. If none >> should be << failures detected during the processing of an FCP I/O operation. If none >> IBM-147 PDF pg 67, pg 50, 9.4.16 FCP_RSP_INFO field, 1st paragraph This << The FCP_RSP_INFO field shall contain valid information if the target detects any of the conditions indicated by an FCP RSP_CODE. >> should be << The FCP_RSP_INFO field shall contain valid information if the target FCP_Port detects any of the conditions indicated by an FCP RSP_CODE. >> IBM-148 PDF pg 68, pg 51, 9.4.16 FCP_RSP_INFO field, 2nd to last paragraph This<< If the RSP_CODE indicates 'Task Management function failed', the state of the target is unknown. >> should be << If the RSP_CODE indicates 'Task Management function failed', the state of the logical unit is unknown. >> IBM-149 PDF pg 69, pg 52, 9.5 FCP_CONF IU The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-150 PDF pg 70, pg 53, 10.1 Overview of mode page codes for the Fibre Channel protocol This << Clause 10 describes the block descriptors and the pages >> should be << This clause describes the block descriptors and the pages >> IBM-151 PDF pg 70, pg 53, 10.2.1 Overview and format of Disconnect-Reconnect mode page for FCP, 1st paragraph This << The initiator communicates with the device server to determine what values are most appropriate for a device server. The device server communicates the parameter values in this mode page to the target port, normally the Fibre Channel interface circuitry. This communication is internal to the target and FCP device and is outside the scope of this standard. >> should be << The application client communicates with the device server to determine what values are most appropriate for a device server. The device server communicates the parameter values in this mode page to the target FCP_Port, normally the Fibre Channel interface circuitry. This communication is internal to the SCSI target device and FCP device and is outside the scope of this standard. >> IBM-152 PDF pg 72, pg 55, 10.2.3 BUFFER EMPTY RATIO field This << should be prior to transmitting an FCP_XFER_RDY IU that requests the initiator to send data. >> should be << should be prior to transmitting an FCP_XFER_RDY IU that requests the initiator FCP_Port to send data. >> IBM-153 PDF pg 72, pg 55, 10.2.4 BUS INACTIVITY LIMIT field, 1st paragraph This << The BUS INACTIVITY LIMIT field indicates the maximum time that the target is permitted to maintain an interconnect >> should be << The BUS INACTIVITY LIMIT field indicates the maximum time that the target FCP_Port is permitted to maintain an interconnect >> IBM-154 PDF pg 72, pg 55, 10.2.4 BUS INACTIVITY LIMIT field, 1stpparagraph This << and the target holding the bus detects that the limit is going to be exceeded, >> should be << and the target FCP_Port holding the bus detects that the limit is going to be exceeded, >> IBM-155 PDF pg 72, pg 55, 10.2.5 DISCONNECT TIME LIMIT field, 1st paragraph This << Targets in configurations having the concept of interconnect tenancy >> should be << Targets FCP_Ports in configurations having the concept of interconnect tenancy >> IBM-156 PDF pg 72, pg 55, 10.2.7 MAXIMUM BURST SIZE field, 1st paragraph This << the maximum size of all bytes in an FCP_DATA IU that the device server shall transfer to the initiator or request from the initiator. >> should be << the maximum size of all bytes in an FCP_DATA IU that the device server shall transfer to the application client or request from the application client.>> IBM-157 PDF pg 73, pg 56, 10.2.8 EMDP bit This << The enable modify data pointers (EMDP) bit indicates whether or not the target may use the random buffer >> should be << The enable modify data pointers (EMDP) bit indicates whether or not the target FCP_Port may use the random buffer >> IBM-158 PDF pg 73, pg 56, 10.2.8 EMDP bit This << If the EMDP bit is set to zero, the target shall generate continuously increasing relative >> should be << If the EMDP bit is set to zero, the target FCP_Port shall generate continuously increasing relative >> IBM-159 PDF pg 73, pg 56, 10.2.8 EMDP bit This << If the EMDP bit is set to one, the target may transfer the FCP_DATA IUs for a single SCSI command >> should be << If the EMDP bit is set to one, the target FCP_Port may transfer the FCP_DATA IUs for a single SCSI command >> IBM-160 PDF pg 73, pg 56, 10.2.9 FAA, FAB, FAC bits The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-161 PDF pg 73, pg 56, 10.2.9 FAA, FAB, FAC bits This << The FAA bit controls arbitration when the target wishes to send one or more FCP_DATA IU frames to an initiator. The FAB bit controls arbitration when the initiator wishes to send one or more FCP_XFER_RDY IU frames to a target. The FAC bit controls arbitration when the target wishes to send an FCP_RSP IU frame to an initiator or when the initiator wishes to send an FCP_CMND IU frames to target. >> should be << The FAA bit controls arbitration when the target FCP_Port has one or more FCP_DATA IU frames to send to an initiator FCP_Port. The FAB bit controls arbitration when the initiator FCP_Port has one or more FCP_XFER_RDY IU frames to send to a target FCP_Port. The FAC bit controls arbitration when the target FCP_Port has an FCP_RSP IU frame to send to an initiator FCP_Port or when the initiator FCP_Port has an FCP_CMND IU frames to send to a target FCP_Port. >> IBM-162 PDF pg 73, pg 56, 10.2.10 FIRST BURST SIZE field The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-163 PDF pg 75, pg 58, 10.4.2 DISABLE TARGET ORIGINATED LOOP INITIALIZATION (DTOLI) bit The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-164 PDF pg 75, pg 58, 10.4.3 DISABLE TARGET INITIATED PORT ENABLE (DTIPE) bit The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-165 PDF pg 75, pg 58, 10.4.4 ALLOW LOGIN WITHOUT LOOP INITIALIZATION (ALWLI) This <> should be << shall use the hard address available in the SCA-2 connector (see SFF-8067) or in device >> IBM-166 PDF pg 75, pg 58, 10.4.4 ALLOW LOGIN WITHOUT LOOP INITIALIZATION (ALWLI) The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-167 PDF pg 75, pg 58, 10.4.5 REQUIRE HARD ADDRESS (RHA) bit This << its hard address available in the SCA-2 SFF-8067 connector or device address jumpers during loop >> should be << its hard address available in the SCA-2 connector (see SFF-8067) or device address jumpers during loop >> IBM-168 PDF pg 75, pg 58, 10.4.5 REQUIRE HARD ADDRESS (RHA) bit The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-169 PDF pg 76, pg 59, 10.4.6 DISABLE LOOP MASTER (DLM) bit The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-170 PDF pg 76, pg 59, 10.4.7 DISABLE DISCOVERY (DDIS) bit This << shall not require receipt of Address or Port Discovery (ADISC or PDISC ELSs) following loop initialization as >> should be << shall not require receipt of Address or Port Discovery (i.e., ADISC or PDISC ELSs) following loop initialization as >> IBM-171 PDF pg 76, pg 59, 10.4.7 DISABLE DISCOVERY (DDIS) bit The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-172 PDF pg 76, pg 59, 10.4.8 PREVENT LOOP PORT BYPASS (PLPB) bit The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-173 PDF pg 76, pg 59, 10.4.9 DISABLE TARGET FABRIC DISCOVERY (DTFD) bit The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-174 PDF pg 77, pg 60, 10.4.10 SEQUENCE INITIATIVE RESOURCE RECOVERY TIMEOUT VALUE (RR_TOVSEQ_INIT) field, 1st paragraph This << The RR_TOVSEQ_INIT (see 11.4) field is defined by bytes 6 and 7 in the following manner. >> should be << The RR_TOVSEQ_INIT timer (see 11.4) operation is defined by the RR_TOVSEQ_INIT field and the RR_TOV INITS field. >> IBM-175 PDF pg 78, pg 61, 11.1 Summary of timers for the Fibre Channel protocol Global - The correct abbreviation for seconds is << s >> not << sec >> or << sec. >> or << seconds >>. This needs to be fixed throughout this standard starting with table 30. IBM-176 PDF pg 78, pg 61, 11.1 Summary of timers for the Fibre Channel protocol, table 30 note 2 and note 4 The term << target >> should be << target FCP_Port >> IBM-177 PDF pg 79, pg 62, 11.2 Error_Detect Timeout (E_D_TOV), 2nd to last paragraph This << Target devices that support Class 2 shall implement this timer for the purpose of timing out missing ACKs. >> should be << Target FCP_Ports that support Class 2 shall implement this timer for the purpose of timing out missing ACKs. >> IBM-178 PDF pg 79, pg 62, 11.4 Resource Recovery Timeout (RR_TOV) The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-179 PDF pg 80, pg 63, 11.5 Read Exchange Concise Timeout Value (REC_TOV) The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-180 PDF pg 80, pg 63, 11.6 Upper Level Protocol Timeout (ULP_TOV), Last paragraph This << caused by command queuing and multi-initiator congestion. >> should be << caused by command queuing and multi-initiator FCP_Port congestion. >> IBM-181 PDF pg 81, pg 64, 12.1.1 Exchange level error recovery, 2nd paragraph This << To recover from these errors, all FCP compliant initiators shall be capable of invoking the recovery abort function to terminate a failing exchange and to recover the associated resources as described in 12.3. All FCP compliant targets shall be capable of executing the requested recover >> should be << To recover from these errors, all FCP compliant initiator FCP_Ports shall be capable of invoking the recovery abort function to terminate a failing exchange and to recover the associated resources as described in 12.3. All FCP compliant target FCP_Ports shall be capable of processing the requested recover >> IBM-182 PDF pg 81, pg 64, 12.1.2 Sequence level error recovery The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-183 PDF pg 81, pg 64, 12.2.2 FCP-3 error detection using protocol errors for all classes of service The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-184 PDF pg 83, pg 66, 12.3.1 Recovery abort requirements, last paragraph This << All FCP initiators shall be capable of invoking the recovery abort protocol to terminate failing commands for later retry (see 9.1.2.5). All FCP targets shall >> should be << All initiator FCP_Ports shall be capable of invoking the recovery abort protocol to terminate failing commands for later retry (see 9.1.2.5). All target FCP_Ports shall >> IBM-185 PDF pg 83, pg 66, 12.3.2 Initiator invocation of recovery abort The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-186 PDF pg 83, pg 66, 12.3.2 Initiator invocation of recovery abort, 2nd paragraph This << effect immediately. For example, if ABTS is sent following transmission of a READ command, the initiator may receive some or all of the requested read data before receiving the BA_ACC to the ABTS. >> should be << effect immediately (e.g., if ABTS is sent following transmission of a READ command, the initiator may receive some or all of the requested read data before receiving the BA_ACC to the ABTS). >> IBM-187 PDF pg 83, pg 66, 12.3.3 Target response to recovery abort The term << target >> should be << target FCP_Port >> in all cases in this subclause including in the subclause title. IBM-188 PDF pg 84, pg 67, 12.3.4 Additional error recovery by initiator The term << initiator >> should be << initiator FCP_Port >> in all cases in this subclause including the title of the subclause. IBM-189 PDF pg 84, pg 67, 12.3.5 Additional error recovery by target The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause including the subcaluse title. IBM-190 PDF pg 85, pg 68, 12.4.1.1 Polling Exchange state with REC The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-191 PDF pg 85, pg 68, 12.4.1.1 Polling Exchange state with REC This << Exchange. Subclauses 12.4.1.2 through 12.4.1.8 define optional error detection and recovery procedures for acknowledged and unacknowledged classes of service. >> should be << Exchange. Optional error detection and recovery procedures for acknowledged and unacknowledged classes of service are defined in 12.4.1.2, 12.4.1.3, 12.4.1.4, 12.4.1.5, 12.4.1.6, 12.4.1.7, and 12.4.1.8. >> IBM-192 PDF pg 85, pg 68, 12.4.1.2 Detection of errors while polling with REC The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-193 PDF pg 85, pg 68, 12.4.1.3 FCP_CMND IU recovery using information from REC The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-194 PDF pg 85, pg 68, 12.4.1.4 FCP_XFER_RDY IU recovery The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-195 PDF pg 85, pg 68, 12.4.1.4 FCP_XFER_RDY IU recovery, 2nd paragraph This<< but not received by the initiator, issue an SRR in a new Exchange to request retransmission of the FCP_XFER_RDY IU.>> does not make any sense. Perhaps there is supposed to be a << then >>. If so it would become << but not received by the initiator, then the target FCP_Port issues an SRR in a new Exchange to request retransmission of the FCP_XFER_RDY IU.>> IBM-196 PDF pg 86, pg 69, 12.4.1.5 FCP_RSP IU recovery The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-197 PDF pg 87, pg 70, 12.4.1.6 FCP_DATA IU recovery - write operations The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-198 PDF pg 87, pg 70, 12.4.1.7 FCP_DATA IU recovery - read operations The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause except as noted in the other comment in this subclause. IBM-199 PDF pg 87, pg 70, 12.4.1.7 FCP_DATA IU recovery - read operations, 2nd paragraph This << the target shall send an FCP_RSP IU with CHECK CONDITION status and sense data containing a sense key of HARDWARE ERROR and an additional sense code of INITIATOR DETECTED ERROR MESSAGE RECEIVED. >> should be << the device server shall send an FCP_RSP IU with CHECK CONDITION status and sense data containing a sense key of HARDWARE ERROR and an additional sense code of INITIATOR DETECTED ERROR MESSAGE RECEIVED. >> IBM-200 PDF pg 87, pg 70, 12.4.1.8 FCP_CONF IU recovery The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause except as noted in the other comment in this subclause. IBM-201 PDF pg 88, pg 71, 12.4.2.2 Missing ACK The term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-202 PDF pg 88, pg 71, 12.4.2.3 Distinguishing Exchange to be aborted The term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-203 PDF pg 89, pg 72, 12.5.1 ABTS The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause. IBM-204 PDF pg 89, pg 72, 12.5.2 REC The term << initiator >> should be << initiator FCP_Port >> in all cases in this subclause. IBM-205 PDF pg 89, pg 72, 12.5.3 SRR The term << initiator >> should be << initiator FCP_Port >> in all cases in this subclause. IBM-206 PDF pg 89, pg 72, 12.6 Responses to FCP type frames before PLOGI or PRLI The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this subclause IBM-207 PDF pg 89, pg 72, 12.6 Responses to FCP type frames before PLOGI or PRLI, Last paragraph This << If an FCP device receives a frame of category 0001b or 0011b (solicited data or solicited control) and the FCP >> should be << If an FCP device receives a frame of category 0001b or 0011b (i.e., solicited data or solicited control) and the FCP >> IBM-208 PDF pg 90, pg 73, A.1 Definition of procedure terms, 1st paragraph This << FCP-3 services are provided to the application client by the initiator to request and manage tasks as described by the SAM-3 standard. SAM-3 further defines how the target enables the device server to receive and process the tasks addressed to a logical unit. The Fibre Channel protocol is described in terms of the services provided by the initiator and target. >> should be << FCP-3 services are provided to the application client by the initiator FCP_Port to request and manage tasks as described by the SAM-3 standard. SAM-3 further defines how the target FCP_Port enables the device server to receive and process the tasks addressed to a logical unit. The Fibre Channel protocol is described in terms of the services provided by the initiator FCP_Port and target FCP_Port . >> IBM-209 PDF pg 90, pg 73, A.1 Definition of procedure terms, table A.1 This << initiator port >> should be << initiator FCP_Port >>. IBM-210 PDF pg 90, pg 73, A.1 Definition of procedure terms, table A.1 This << target port >> should be << target FCP_Port >> IBM-211 PDF pg 94, pg 77, B.1.2 SCSI FCP read operation, Annex B All the tables that have column titles of << Initiator function >> and <> should be << Initiator FCP_Port function >> and << Target FCP_Port function >>. IBM-212 PDF pg 107, pg 90, C.1 Introduction, Figure c.1 This << REC can be optionally used at any time to ascertain status of an Exchange. It can also be used in conjunction with ABTS(Sequence) to obtain additional information useful in the Error Recovery process. >> should be << REC may be used at any time to ascertain status of an Exchange. It may also be used in conjunction with ABTS(Sequence) to obtain additional information useful in the Error Recovery process. >> IBM-213 PDF pg 109, pg 92, C.1 Introduction, Figure C.3 This << Both the initiator and target establish Recovery Qualifiers. >> should be << Both the initiator FCP_Port and target FCP_Port establish Recovery Qualifiers. >> IBM-214 PDF pg 110, pg 93, C.1 Introduction, Figure C.4 This << CNT of FCP_CMND. Note that the issuance of RRQ is not necessary in this case, since the >> should be << CNT of FCP_CMND. The issuance of RRQ is not necessary in this case, since the >> IBM-215 PDF pg 110, pg 93, C.1 Introduction, Figure C.4 This << target has not established a Recovery Qualifier. However, the initiator cannot reclaim the >> should be << target FCP_Port has not established a Recovery Qualifier. However, the initiator FCP_Port is not able to reclaim the >> IBM-216 PDF pg 111, pg 94, C.1 Introduction, Figure C-5 This << indicates the initiator holds Sequence Initiative and the Exchange is open. The initiator sends an SRR requesting the FCP_XFER_RDY be resent. The target resends the >> should be << indicates the initiator FCP_Port holds Sequence Initiative and the Exchange is open. The initiator FCP_Port sends an SRR requesting the FCP_XFER_RDY be resent. The target FCP_Port resends the >> IBM-217 PDF pg 112, pg 95, C.1 Introduction, Figure C.6 This << by the initiator. The BA_ACC payload is SEQ_ID invalid, low SEQ_CNT = 0, high SEQ_CNT = SEQ_CNT in ABTS frame = 1. Both target and initiator establish Recovery Qualifiers. >> should be << by the initiator FCP_Port. The BA_ACC payload is SEQ_ID invalid, low SEQ_CNT = 0, high SEQ_CNT = SEQ_CNT in ABTS frame = 1. Both target FCP_Port and initiator FCP_Port establish Recovery Qualifiers. >> IBM-218 PDF pg 113, pg 96, C.1 Introduction, Figure C.7 This << FCP_XFER_RDY was received by the initiator. >> should be << FCP_XFER_RDY was received by the initiator FCP_Port. >> IBM-219 PDF pg 113, pg 96, C.1 Introduction, Figure C.7 This <> should be << There is no need for the target FCP_Port to issue >> IBM-220 PDF pg 114, pg 97, C.1 Introduction, Figure C.8 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-221 PDF pg 115, pg 98, C.1 Introduction, Figure C.9 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-222 PDF pg 116, pg 99, C.1 Introduction, Figure C.10 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-223 PDF pg 117, pg 100, C.1 Introduction, Figure C.11 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-224 PDF pg 117, pg 100, C.1 Introduction, Figure C.11 This << The associated resources cannot be reused for a period of R_A_TOV. For in-order delivery, >> should be << The associated resources is not able to be reused for a period of R_A_TOV. For in-order delivery, >> IBM-225 PDF pg 118, pg 101, C.1 Introduction, Figure C.12 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-226 PDF pg 118, pg 101, C.1 Introduction, Figure C.13 This << he associated resources cannot be reused for a period of R_A_TOV.>> should be << he associated resources is not able to be reused for a period of R_A_TOV.>> IBM-227 PDF pg 119, pg 102, C.1 Introduction, Figure C.14 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-228 PDF pg 121, pg 104, C.1 Introduction, Figure C.16 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-229 PDF pg 123, pg 106, C.1 Introduction, Figure C.17 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-230 PDF pg 125, pg 108, C.1 Introduction, Figure C.19 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-231 PDF pg 127, pg 110, C.1 Introduction, Figure C.21 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-232 PDF pg 127, pg 110, C.1 Introduction, Figure C.21 This << FCP_RSP can be received anytime after the transmission of FCP_CMND due to out of order delivery. >> should be << FCP_RSP may be received anytime after the transmission of FCP_CMND due to out of order delivery. >> IBM-233 PDF pg 128, pg 111, C.1 Introduction, Figure C.22 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-234 PDF pg 128, pg 111, C.1 Introduction, Figure C.22 This << FCP_RSP can be received at any time after the last FCP_DATA frame has been transmitted.>> should be << FCP_RSP may be received at any time after the last FCP_DATA frame has been transmitted.>> IBM-235 PDF pg 129, pg 112, C.1 Introduction, Figure C.23 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-236 PDF pg 129, pg 112, C.1 Introduction, Figure C.23 This << discarded and the context for the Exchange can be purged.>> should be << discarded and the context for the Exchange is able to be be purged.>> IBM-237 PDF pg 131, pg 114, C.1 Introduction, Figure C.25 The term << initiator >> should be << initiator FCP_Port >> and the term << target >> should be << target FCP_Port >> in all cases in this figure. IBM-238 PDF pg 131, pg 114, C.1 Introduction, Figure C.25 This << The resources associated with the Recovery Qualifier can be reclaimed when R_A_TOV expires. >> should be << The resources associated with the Recovery Qualifier are able to be reclaimed when R_A_TOV expires. >> IBM-239 PDF pg 132, pg 115, C.1 Introduction, Figure C.26 This << received, the target would view the ABTS as having been issued on a new Exchange. >> should be << received, the target FCP_Port views the ABTS as having been issued on a new Exchange. >> IBM-240 PDF pg 133, pg 116, C.1 Introduction, Figure C.28 This << indicates that the REC was never received by the target. >> should be << indicates that the REC was never received by the target FCP_Port. >> IBM-241 PDF pg 134, pg 117, C.1 Introduction, Figure C.28 The term << initiator >> should be << initiator FCP_Port >> in all cases in this figure. IBM-242 PDF pg 137, pg 120, C.1 Introduction, Figure C.31 This << The Recovery Qualifier is established on the initiator side and is timed out >> should be << The Recovery Qualifier is established on the initiator FCP_Port side and is timed out >> IBM-243 PDF pg 140, pg 123, D.1.1 Initiator discovery of Fabric-attached targets, 1st paragraph This << The following procedure may be used by initiators for discovering and authenticating >> should be << The following procedure may be used by initiator FCP_Ports for discovering and authenticating >> IBM-244 PDF pg 140, pg 123, D.1.1 Initiator discovery of Fabric-attached targets The title if this subclause should be changed to << D.1.1 Discovery of Fabric-attached target FCP_Ports IBM-245 PDF pg 140, pg 123, D.1.1 Initiator discovery of Fabric-attached targets This 1,2,3 list is not in the correct form. It should be 1) ...; 2)...: x-1) ...; and x) ..... IBM-246 PDF pg 140, pg 123, D.1.1 Initiator discovery of Fabric-attached targets, Item 7 This << If the INQUIRY succeeds, issue a REPORT LUNS command to LUN 0 to obtain a list of the logical units accessible through the target >> should be << If the INQUIRY succeeds, issue a REPORT LUNS command to LUN 0 to obtain a list of the logical units accessible through the target FCP_Port >> IBM-247 PDF pg 140, pg 123, D.1.2 Initiator discovery of loop-attached targets, 1st paragraph This << The following procedure may be used by initiators for discovering and >> should be << The following procedure may be used by initiator FCP_Ports for discovering and >> IBM-248 PDF pg 140, pg 123, D.1.2 Initiator discovery of loop-attached targets This title of this section should be << D.1.2 Discovery of loop-attached target FCP_Ports IBM-249 PDF pg 141, pg 124, D.1.2 Initiator discovery of loop-attached targets, Item 1) This << to identify those devices that are present on the loop >> should be << to identify SCSI devices that are present on the loop >> IBM-250 PDF pg 141, pg 124, D.1.2 Initiator discovery of loop-attached targets, Item 2) This << if the device is determined to be an >> should be << if the SCSI device is determined to be an >> IBM-251 PDF pg 141, pg 124, D.1.2 Initiator discovery of loop-attached targets This 1,2,3 list is not in the correct form. It should be 1) ...; 2)...: x-1) ...; and x) ..... IBM-252 PDF pg 141, pg 124, D.1.2 Initiator discovery of loop-attached targets, item 3) This << units supported by the target >> should be << units access through the target FCP_Port >> IBM-253 PDF pg 141, pg 124, D.1.2 Initiator discovery of loop-attached targets, Item 4) The term << EVPD >> needs to be in small caps. IBM-254 PDF pg 141, pg 124, D.2 Fabric and Device Authentication This 1,2,3 list is not in the correct form. It should be 1) ...; 2)...: x-1) ...; and x) ..... IBM-255 PDF pg 141, pg 124, D.2 Fabric and Device Authentication, Item 2) This << All N_Ports and NL_Ports, including initiators and targets, validate the current >> should be << All N_Ports and NL_Ports, including initiator FCP_Ports and target FCP_Ports, validate the current >> IBM-256 PDF pg 141, pg 124, D.2 Fabric and Device Authentication, item 3) This << Address Identifier of that port. >> should be << Address Identifier of that FCP_Port. >> IBM-257 PDF pg 141, pg 124, D.2 Fabric and Device Authentication, Item 4) This << Initiators and targets validate N_Port and NL_Port logins following >> should be << Initiator FCP_Ports and target FCP_Ports validate N_Port and NL_Port logins following >> IBM-258 PDF pg 141, pg 124, D.2 Fabric and Device Authentication, iten 4) This << all open Exchanges with that initiator or target are terminated ( >> should be << all open Exchanges with that initiator FCP_Port or target FCP_Port are terminated ( >> IBM-259 PDF pg 141, pg 124, D.3 Logical unit authentication, 1st paragraph The term << EVPD >> needs to be in small caps. IBM-260 PDF pg 141, pg 124, D.3 Logical unit authentication, 1st paragraph This term << LUN >> needs to be in small caps. IBM-261 PDF pg 142, pg 125, E.2.1 Abort Sequence (ABTS) Request fields, 1st paragraph This << The initiator or target may transmit an ABTS Frame. >> should be << The initiator FCP_Port or target FCP_Port may transmit an ABTS Frame. >> IBM-262 PDF pg 142, pg 125, E.2.1 Abort Sequence (ABTS) Request fields, Table E.1 1st row This << the ABTS Initiator may not have Sequence Initiative for the Sequence being aborted). >> should be << the ABTS Initiator FCP_Port may not have Sequence Initiative for the Sequence being aborted). >> IBM-263 PDF pg 142, pg 125, E.2.1 Abort Sequence (ABTS) Request fields, Table E.1 3rd row This << any Exchange) between that pair of ports. >> should be << any Exchange) between that pair of FCP_Ports. >> IBM-264 PDF pg 143, pg 126, E.2.2 Basic Accept (BA_ACC) Frame to ABTS, 1st paragraph This << An initiator or target may accept ABTS with BA_ACC. >> should be << An initiator FCP_Port or target FCP_Port may accept ABTS with BA_ACC. >> IBM-265 PDF pg 143, pg 126, E.2.2 Basic Accept (BA_ACC) Frame to ABTS, Table E.2 This << ABTS Initiator for Abort Sequence >> should be << ABTS Initiator FCP_Port for Abort Sequence >> IBM-266 PDF pg 143, pg 126, E.2.3 Basic Reject (BA_RJT) Frame to ABTS, 1st paragraph This << A target may reject ABTS with BA_RJT. When it does, >> should be << A target FCP_Port may reject ABTS with BA_RJT. When it does, >> IBM-267 PDF pg 144, pg 127, E.3.1 RRQ request format, table E.4 This << Source_ID of the initiator >> should be << Source_ID of the initiator FCP_Port>> IBM-268 PDF pg 144, pg 127, E.3.1 RRQ request format, Last paragraph This << of the RRQ, the target responds with ACC. >> should be << of the RRQ, the target FCP_Port responds with ACC. >> ************************************************************** Comments attached to Abs ballot from Bill Galloway of Pivot3, Inc.: Not materially affected by this proposal. ************************************************************** Comments attached to No ballot from Gerald Houlder of Seagate Technology: Comments are also provided in document 05-168r0. Both PDF and XLS versions of this document are provided. Seagate-001 (E) Page: Location: Global Problem Description: Globally, The word protocol in: Fibre Channel Protocol is sometimes capitalized and sometimes not. The use of a capital Protocol better differentiates that the protocol defined by this standard is being referenced Suggested Solution: Change all Fibre Channel protocol to Fibre Channel Protocol Seagate-002 (E) Page: 9 Location: 4.10 Problem Description: If FC-FS-2 is being used, this reference should be FC-AL-2. Note, FC-AL-2 is used in other places in the document. Suggested Solution: Change references from FC-AL to FC-AL-2. A global change would be good. Seagate-003 (E) Page: 17 Location: 4.10 Problem Description: In Table 4, the text in the table notes is running into the border on the right side. Suggested Solution: change format? Seagate-004 (E) Page: 18 Location: 4.12 Problem Description: A reference after Reset LIP(y,x) would be beneficial Suggested Solution: Add a reference to FC-AL-2 Seagate-005 (E) Page: 18 Location: 4.12 Problem Description: In Table 5, the text in the table notes is running into the border on the right side. Suggested Solution: change format? Seagate-006 (E) Page: 19 Location: 4.15 Problem Description: The last sentence is not really related to the other text in this clause. 4.13 needs a comment about implicit login. Suggested Solution: Delete this sentence in 4.15 and add in 4.13 Seagate-007 (E) Page: 20 Location: 5.2 Problem Description: In the second sentence . . . by FC-FS-2 and its extensions. What are extensions? Suggested Solution: Suggest deleting "and its extensions" Seagate-008 (E) Page: 20 Location: 5.3 Problem Description: This does not convey that FC_DATA IUs may be multiple frame sequences. If FC_DATA IUs were required to be single frame sequences, they would satisfy this statement. Suggested Solution: Add text to 9.3.1 FCP_DATA IU. See Seagate-009 Seagate-009 (E) Page: 44 Location: 9.3.1 Problem Description: Add text to indicate the FCP_DATA IU may be a multiple frame Sequence Suggested Solution: Add between the second and third paragraphs:FCP_DATA IUs carry the SCSI data transfers for a command. An FCP_DATA IU is a single FC Sequence consisting of one or more FC data frames Seagate-010 (E) Page: 53 Location: 10.1 Problem Description: The first sentence does not need to say "Clause 10 describes . . . Other clauses do not have this intro. Just describe the function. Suggested Solution: Change the first sentence to: The block descriptors and pages used with MODE SELECT and MODE SENSE commands control and report the behavior of the Fibre Channel Protocol. ************************************************************** Comments attached to Yes ballot from Erich Oetting of Storage Technology Corp.: 01 E, 4.1, pg 9, par 3, first s, remove "to execute the steps required". 02 E, 4.1, pg 9, par 4, replace "is not defined, although" with "is not defined by this standard.", (break into two sentences). 03 E, 4.1, pg 9, par 6, replace with "This standard defines four kinds of functional management:" 04 E, 4.2, pg 10, par 3, s 1, replace "device server for the command has completed the interpretation of" with "device server has interpreted". 05 E, 4.2, pg 10, par 3, s 1, replace "to the initiator to indicate" with "to the initiator indicating". 06 E, 4.2, pg 10, par 3, s 2, replace "FCP-Port that is the initiator" with "initiator FCP-Port". 07 E. 4.2, pg 11, par 2 of pg, replace "device server for the command has completed the interpretation of" with "device server has interpreted". 08 E. 4.2, pg 11, par 3 of pg, replace "device server for the command has completed the interpretation of" with "device server has interpreted". 09 T, 4.2, pg 12, par 1 of pg, modify last sentence to permit asking for confirmation on last FCP_RESP_IU in a series of linked commands as permitted by clause 4.5. 10 E, 4.2, pg 12, par 1 of pg, last s, add ",see clause 5.3" after I4. 11 E, 4.6, pg 14, par 2, replace "any initiator" with "the initiator". 12 E, 5.2, pg 20, par 2, first s, remove the word "vital". 13 E, 5.2, pg 20, par 2, first s, replace "as defined by SPC-3." with ", see SPC-3." 14 E, 8.3, pg 35, move the "FCP_RJT Reason Code Discriptions" text into Table 16. 15 E, 9.1.2.5, pg 41, last para in clause, replace "resources to be cleared may" with "resources may". 16 E, 9.2.1, pg 42, first s, replace "write command" with "write operation". 17 E, 10.2.1, pg 53, replace first s. with "The disconnect-reconnect page (see table 26) allows the application client to modify the behavior of the service delivery subsystem.". 18 E, 12.2.2, pg 63, second item d on page, replace the word "reas" with "read". ******************** End of Ballot Report ********************