FC-SB-4 Revision 2.00 Letter Ballot Comments Database (09-283v4)
Company Tech/Edit Page Sec/table/fig locator
Comment Proposed Solution Resolution Key Status Notes+I56
ENDL-1 T 81 The format of the SB-4 transport-data Starting with figure 18 and continuing Figure 18 is only showing the fomat AinP C
IU fails to account for the fact that this throughout the document, a method must be of a Data IU. It does not show how
IU frequently will be broken up into devised to show that the transport-data IU can long or how short or how few or
smaller units of transfer as per the automatically be subdivided to meet the needs how many are used in a data
requirements of first burst processing, of the underlying transport hardware. Every transfer operation. The smallest unit
transfer-ready IUs, and suchlike. This instance of transport-data IU must be of transfer is an IU. A transfer is
produces many problems. One inspected to see if it is the currently defined broken up into multiple IU's if
example is that a strict reading of the larger entity or one of the subdivisions thereof. multiple transfer readies (implicit or
working draft as written requires not) are used. Section 220.127.116.11
transfer of the entire transport-data IU figure 50 shows the format of the
(up to 4 GB) as a first burst if first- entire data transfer (refered to as
transfer-ready is disabled. There is no the transport data field).
smaller unit/IU of data to transfer.
Section 18.104.22.168 deleted - figure 50
is redundant with figure 18; section
8.14.1 states that one or more
transport data IUs shall be used for
a write operation. Sections 8.14.2
and 9.3.2 describe additional detail
regarding transport-mode data
transfer for both read and write
operations including how much data
can be sent in each transport-mode
data IU and the use of multiple data
IUs for write operations - added a
reference to those sections in
8.14.1. Added wording to the
definition of the TCOB command
that the COB is sent in the first transp
ENDL-2 T 142 8.14.1, p 2, s 2 The description of first burst At least the following two facts need to be The first burst size limit is already AinP C there is confusion caused by the
processing: When first-transfer-ready mentioned. First, the first burst transfer ends explained in 8.14.2. Change this location of section x.x that
is disabled, the channel shall send a when Sequence Initiative is transferred from sentence to cover seq. initiative: describes the transport data format.
transport-data IU immediately the channel to the control unit. Second, the When first-transfer-ready is It is in the middle of the section
following the transport-command IU. " number of bytes transferred in a first burst disabled, the channel shall send a describing the rules and use of data
is incomplete. shall not exceed the First Burst Size returned transport-data IU IU's. Consider putting the data
in the PRLI ACC (see 22.214.171.124). immediately following the transport- format in a separate section.
command IU. This first transport-
data IU shall transfer sequence Handling of Sequence Initiative for
initiative to the control unit. all IU's is described in tables 7 and
Alternatively, add rule to 8.2.3: For 8 - sequence initiative is always
a write operations with first transfer transferred for transport-data IUs.
ready is disabled, the first transport-
data IU sent by a channel shall not
contain a number of bytes
exceeding the first burst size
obtained from the control unit during
process login and sequence
initiative for the transport exchange
shall be transferred to the recipient.
Changes: Sections 8.14.2 and 9.3.2
provide details for transport-mode
data transfer for both read and write
operations including how much data
can be sent in each transport-mode
data IU - added a reference to
those sections in 8.14.1; added
statements to the first paragraph of 8
ENDL-3 T 144 8.14.3, 1st p on pg, s 1 first byte of data in the Data IU [is not first byte after the COB CRC field, if any, in the Fix all occurances of Data IU that AinP C
correct on two counts. Data IU is not transport-data IU mean transport-data IU. Clarify that
the correct term. If a COB is present the first byte of data is the first byte
in the transport-data IU, then the after the COB CRC field, if any, in
offset is from the first byte after the the FIRST transport-data IU of the
COB.] transport data field.
Changed occurences of Data IU to
transport-data IU where
appropriate; added statement to the
definition of the COB that "the first
byte of data for the write operation
is the first byte following the COB
IBM-1 T 81 8.3 figure 18 Final CRC size should be 0/4 Change to 0/4 Changed per proposed solution. A C
because intermediate data Ius don't
have final CRC.
IBM-2 T 143 126.96.36.199, last paragraph All pad and CRC byte counts are All pad, CRC and COB byte counts are Section 188.8.131.52 deleted (see ENDL- A C
included in the DL. - need to include included in the DL 1). The definition of the DL covers
COB bytes inclusion of pad, CRC and COB,
COB pad and COB CRC.
IBM-3 T 143 8.14.3 Need to document that a COB is Editors choice Definition of the TCOB command AinP C
present in either the transport- already describes how the COB
command IU or the transport-data IU may be provided. Added statement
when the TCOB DCW is in the TCA. to 8.14.3 (COB definition) to
Otherwise, it is not present. indicate COB is only provided when
the first command in TCA is the
TCOB command and added
reference to the TCOB command
IBM-4 E 143 8.14.3 p2 s3 CDB should be TCA change to TCA Changed per proposed solution. A C
IBM-5 T 160 184.108.40.206 Table 36 Once the a first entry is determined to Eliminate word 1 Changed per proposed solution. A C
be not appropriate, the transfer
processing will stop. We don't need
IBM-6 T 164 8.16.1 p3 s3 Should also state that the data IU Otherwise, it Added the following sentences to A C
transfers sequence initiative shall be sent in a response to a transport- - the first paragraph of to 8.14.1,
data IU that did not exhaust the write data byte Transport Data IU Overview:"When
count for the operation and transfers then transport-data IU is sent by the
sequence initiative to the control unit and only channel, sequence initiative for the
when an abnormal condition does not exist for exchange shall be transferred to the
the operation. recipient. When then transport-data
IU is sent by the control unit,
sequence initiative for the exchange
shall be held by the control unit.".
IBM-7 T 182 220.127.116.11 A TCOB command does not have the A TCOB command does not have the chaining Changed per proposed solution. A C
chaining flag not set - is the second flag not set
"not" supposed to be here?
IBM-8 T 141 18.104.22.168.1 The presence of the TCOB does not Add the following Note:The presence of a Note added A C
affect chaining rules at the device. TCOB as the first DCW in the TCA does not
The command following the TCOB affect DCW chaining requirements of a device.
DCW is still considered to be the first For the purpose of enforcing DCW chaining
command in chain. rules, the DCW following the TCOB shall be
considered the first in chain.
IBM-9 E 6 3.2.13 Need definition for CRC Offset Block Insert this definition at 3.2.13: CRC The following definition inserted at A C
Offset Block - the CRC Offset Block, when 3.2.13: CRC Offset Block - the
present contains a list of 1-word values CRC Offset Block contains a list of
identifying the offset of each intermediate CRC 1-word values identifying the byte
in a write transport data IU. offset of each intermediate CRC
word in the write transport data.
IBM-10 E 10 3.4.6 Need acronym for TCOB Insert at 3.4.6: TCOB Changed per proposed solution. A C
= Transfer CRC Offset Block
IBM-11 E 140 22.214.171.124.1 reference to table 12 should be table Change to reference table 25. Changed per proposed solution. A C
IBM-12 E 5 3.2.10 The definition should not describe the Delete all text starting with " A Connection… Need a location and text to cover AinP C
use or the rules this definition in the document.
Changed per proposed solution.
This is covered by the definition of
IBM-13 E 7 3.2.31 The definition should not describe the Delete "from 1 to 30" Changed per proposed solution. A C
use or the rules
IBM-14 E 7 3.2.36 The definition should not describe the Delete "SB-4 link-control functions are not Changed per proposed solution. A C
use or the rules performed in transport mode"
IBM-15 T 23 5.3 Note 1 does not take transport mode Modify text to either make it generic to Made generic by changing: AinP C
into acount command and transport mode, or add text for "...by not initiating new exchange
transport mode pairs and allowing all channel
programs to complete and
exchange pairs to end." to
"...by not initiating new exchanges
and allowing all channel programs
to complete and exchanges to end."
IBM-16 T 31 6.3.6 Second Paragraph: If the CU Delete the last sentence of the paragraph Changed per proposed solution. A C
performs a RSCN when SB-4 process "The control unit may optionally perform
login service parameters change, the the RSCN ELS when other SB-4 process login
channel will not see the new service parameter capabilities change at the
parameters control unit. "
IBM-17 E 32 126.96.36.199 Last sentence implies that a CU has Change "Logical Paths" to "one or more Changed per proposed solution. A C
to have 2 or more Logical Paths Logical Paths"
IBM-18 T 46 188.8.131.52 This section needs to specify when Add text to describe the CUE rules, handling Added statement "When the control AinP C
the CU sends CUE the case where PRLI will be issued by the unit is able to accept initiation IUs
channel and the case where it won't containing a device-level function,
control-unit-end status shall be sent
to each destination that accepted a
status byte with the control-unit-
busy indication (see 9.4.10)."
IBM-19 T 47 184.108.40.206 The Init Delay Time says that it is the Byte 0 of word 3 contains a binary integer that Changed per proposed solution. AinP C
time before the chan can initiate an specifies in units of seconds the minimum
exchange. But it is really the time amount of time that the recipient of the PRLO
before the chan sends an RNID, ELS request shall delay before sending an
which it must do (see 220.127.116.11) before ELS or initiating an exchange or exhange pair
it opens any command or transport with the sender of the PRLO ELS request
IBM-20 T 67 18.104.22.168 Some channels send an RNID prior to Change text to allow either ordering of Added Statement to 7.3.3 "Channel AinP C
PLOGI. PLOGI/RNID node-idenitifier acquisition may be
be performed prior to channel
IBM-21 T 68 22.214.171.124 The channel does not enforce this Delete last sentence of paragraph "a)" Deleted sentence per proposed AinP C
rule because we have found that solution and added statements to
there are some Control Units that do indicate that unsuccessful node ID
not support RNID that would not acquisition does not prevent the
come on line if we enforced it channel from attempting to
establish logical paths to the control
IBM-22 E 78 8.2.3 para "c)" could be misinterpreted change it to read "…..same device and logical Changed per proposed solution. A C
path if the new transport-command….."
IBM-23 E 78 8.2.3 para "d) 2)" Should the 'or' at the end Change the paragraph to "….the channel shall The channel does not do something AinP C
of the paragraph be there? send a transport-data IU to the control unit. If 'extra" in the case of first transfer
first transfer ready is not disabled or if the ready NOT disabled, so it does not
This paragraph does not define the transport command IU specified a read really need to be stated. The
action if first transfer ready is not operation, the channel shall wait for an IU from sentence that begins, At this point..
disabled. the CU…." Covers both cases. The "or"
delete the "or" at the end of the paragraph should be changed to "and"
Item d.2 split into items d.2 and d.3.
Item d.3 covers all cases.
IBM-24 T 82 8.4.2 In a transport-response IU, the model Change the last sentence to "…and bits 16-31 May need more words in more AinP C
dependent field is supposed to be of word 1 shall be set to bits 16-31 of word one places for this.
echoed back from the command IU of the command IU for which the response IU
is being sent" Added the following line to the end
of the paragraph: "In transport-
command IUs, bits 16-31 of word 1
shall be set to a model dependent
value; in transport-response IUs,
bits 16-31 of word 1 shall be set to
bits 16-31 of word 1 of the
command IU for which the response
IU is being sent."
IBM-25 T 82 8.4.5 Para 1: How is the channel supposed Separate the discussion on commands changed to: "...otherwise, an A C
to tell if the device address received by a CU from responces received by address exception condition shall
corresponds to a device configured to a channel be recognized by the control unit ."
IBM-26 E 135 126.96.36.199 Other than the glossary, there is no delete "is obtained from the TCW and" from Changed per proposed solution. A C
description of what a TCW is. the first sentence.
IBM-27 E 135 188.8.131.52 Other than the glossary, there is no Delete the first sentence of the first paragraph Deleted "The channel shall set bit AinP C
description of what a TCW is. 30 equal to the R bit in the TCW"
and indicate that the R bit is bit 30
It is stated in section 8.13.1 that other in the next sentence.
than the LRC, all fields in the TCCB
are provided by the host system
program. Therefore, the channel
does not set the R bit from the TCW
(what ever that is)
IBM-28 E 135 184.108.40.206 Other than the glossary, there is no Delete the first sentence of the first paragraph Deleted "The channel shall set bit AinP C
description of what a TCW is. 31 equal to the W bit in the TCW"
and indicate that the W bit is bit 31
It is stated in section 8.13.1 that other in the next sentence.
than the LRC, all fields in the TCCB
are provided by the host system
program. Therefore, the channel
does not set the W bit from the TCW
(what ever that is)
IBM-29 T 140 220.127.116.11.1 READ: The bytes provided by some Delete the last sentence of the paragraph This only applies to READ R C
control units were not set up by a commands, not control or sense
previous write, so the order can not commands and therefore, by
use the write command as a definition, refers to data previously
reference written.. Some byte ordering should
IBM-30 T 142 18.104.22.168 Section 8.14.1 clearly states that Make the first paragraph: "For a write 22.214.171.124 is the Transport Data AinP C
there can be multiple Data IUs for a operation, the transport data field is a variable Format. It represents the whole
write operation, one for each transfer length field, made up of one or more Data-IUs, transfer, not a single data IU. See
ready. In 126.96.36.199, implies that each which contains data, data and pad bytes, data ENDL-1 for clarifying text.
Data IU can contain a COB, and must and CRC fields, data and pad bytes and CRC
contain a final CRC. Fields, or just a CRC field, and if a CRC offset Section 188.8.131.52 deleted. See ENDL-
block (COB) is included in the Data field, it 1.
also includes the COB, any COB pad bytes
and the COB CRC."
Add a similar "one or more Data IUs to the
IBM-31 T 146 8.15.3 The intent of the model dependent Change the text to indicatet that the entire SB- Changed per proposed solution. AinP C
bytes was to take the place of the 4 header are echo'd back
command mode "token", so the CU
would echo it back
IBM-32 E 153 184.108.40.206.1 Last sentence of the second to last Remove () Changed per proposed solution. A C
paragraph should not have parenthsis
IBM-33 E 154 220.127.116.11.2 Last sentence of the second to last Remove () Changed per proposed solution. A C
paragraph should not have parenthsis
IBM-34 E 155 18.104.22.168.3 "Meaningless queueing time" is not Change note to "…and present zero in the Changed per proposed solution. A C
an option in the preceeding text. queue-time field"
Either queueing is disabled or zero
must be presented. In either case,
zero must be returned in the queue-
IBM-35 E 141 22.214.171.124.1 The paragraph with the heading Suggest Moving the "bit" definitions under Descriptions of the basic operations AinP C
"Transfer CRC Offset Block" should table 24 and leaving this here. This is really moved to under table 24. Modified
be removed. Reason: The previous the only place the TCOB DCW Command the description of the transport
descriptions of "Read", "Write", code is defined. basic operation to indicate that it
"Control", "Sense", and "Transport" may perform a write operation, as
are descriptions of entries in Table well as a read operation
24. This new paragraph is a
description of an entry in table 25 and
its presence is confusing.
Additionally, the command is
described sufficiently elsewhere.
IBM-36 T 159 126.96.36.199 I think RCQ value 5 should be Remove RCQ value 5. Shift others up. This RCQ does provide some AinP C
removed. Reason: If more than one additional information, so keep RCQ
TCOB command is specified, then 5. Changed definition of RCQ
the additional TCOB commands will, 4 from "A TCOB command was
by definition, not be the first DCW in encountered that was not the first
the TCA and this is covered by RCQ DCW in the TCA" to "The first
value 4. If you keep both values, then TCOB command encountered was
I think you need some sort of not the first DCW in the TCA".
IBM-37 T 182 188.8.131.52 The third bullet should be removed. Remove "A second TCOB command is Changed "A transfer-CRC-offset- AinP C
Reason: Same reason as stated for encountered in the TCA." block (TCOB) command is
section 184.108.40.206 above. encountered that is not the first
DCW in the TCA" to "The first
command encountered is not the
first DCW in the TCA."
IBM-38 (late) T 161,163 Figure 61, 220.127.116.11.6, The state-dependent information field In Figure 61 show the state-dependent Changed per proposed solution. A C
18.104.22.168.7 in the Interrogate extended status information field as words 5-7 and move the
needs to be expanded from 2 words device-level identifier field to word 8. In section
to 3 words. 22.214.171.124.6 change "Words 5-6…" to "Words 5-
7...". In section 126.96.36.199.7 change "Word 7..."
to "Word 8..."
IBM-39 (late) E 10 3.4.6 Definition of TCA should be "transport Correct the wording. Changed per proposed solution. A C
command area" not "transport control
area"; TCAH should be "transport
command area header" not "transport
control area header".
IBM-40 (late) T 143 8.14.3 No PAD bytes needed for COB in Specify that pad bytes are only provided when Changed per proposed solution A C
DCW - DCWs start on a word the COB is provided in the transport data in
boundary. 8.14.3. Remove sentence "The control-data
count does not include any pad bytes." in
TCOB definition (188.8.131.52.1).
IBM-41 (late) T 143 8.14.3 Definition of COB is the transport Move the definition of the COB to the Changed per proposed solution A C
data section but may be sent in the definition of the TCOB in section 184.108.40.206.1.
EMC-1 T 140 220.127.116.11.1 TCOB command should be Require TCOB command. Split table 25 into 2 tables - AinP C Propose that this is the resolution
mandatory. removed transport mode specific to all comments regarding TCOB
commands into a new table. In the mandatory requirements, pending
While the current specification new table explained that review by affected parties that
provides a preferred mechanism for requirements of the transport made the comments.
explicitly locating Intermediate CRC commands (Interrogate is required
offsets, it is optional. There are no and TCOB is recommended) and
guarantees that this will ever actually what happens if they are received in
be implemented, so as to address command mode.
concerns whereby identifying CRC In the description of the TCOB
locations without a proprietary device command under the (new) table
specification becomes possible. explain: "The TCOB command is
recommended in TCW I/O
Operations whenever intermediate
CRCs are present in write data. The
device indicates support for the
TCOB using model dependent
configuration commands. A TCOB
command should not be sent to a
device that does not indicate
support for this command."
With this additional text, removed
note at the bottom of 8.14.3 about
EMC-2 T 183 18.104.22.168 R/W DL validation ("when the last Expanded the 2 bullets on DL AinP C
DCW becomes current…"). checking to 4 bullets:
<bullet> For a read operation, when
It seems as though execution of a a DCW in the TCA becomes
DCW should be stopped if an current, the sum of the data counts
aggregate data count ever exceeds for all previous DCWs and the
the DL value - regardless if the DCW current DCW plus all pad and CRC
is last. bytes exceeds the DL field in the
<new bullet> when the last DCW in
the TCA becomes current, the sum
of the data counts for all DCWs plus
pad bytes plus 4 bytes for CRC all
pad and CRC bytes is not equal to
the DL field in the TCCB.
<bullet> For a write operation, when
a DCW in the TCA becomes
EMC-3 (Late) E 164 8.16.3 Reference is made to a maximum- Either the field in figure 12 in 22.214.171.124 needs to Changed per proposed solution. A C
burst size field that is provided during change ("max-burst-size"), or to change the
process login in the PRLI_ACC but text in 8.16.3 (ie, remove the text limiting the
such a field is not defined, only the burst-length to at most the 'first-burst-size').
'first-burst-size' field is specified.
Change section 8.16.3 to state that the value
of the burst length "shall not exceed the
amount of data remaining to be transferred for
the write operation."
SUN-1 T 143 8.14.3 The conditions where the COB is sent Specify that the COB shall be required when See EMC-1. AinP C Transport Mode was developed to
to the Control Unit are not specified. Intermediate CRC is inlcuded. A second improve application performance
The COB is specified as optional, not solution is, specify how a Control Unit can through reduction of I/O protocol
required. request a COB block from the Channel Image. overhead. Existing implementations
are acheivable without the
presence of and additional
overhead that would be the product
of a required TCOB. The
requriement of a TCOB would
create interoperability issues with
exsisting implemetnations. The
use of a TCOB is provided for by
this specification and the
requirements for use is up to the
device and is specified using
SUN-2 T 144 126.96.36.199 It is not specified in the draft that Indicate the location of all intermediate CRC See EMC-1. AinP C The Data Length is only used to
there is any indication that an fields when Intermediate CRC is included in indicate the number of bytes of data
intermediate CRC is included in the the Data IU. expected to be transfered between
data IU. There is a data count that channel and control unit for a TCW
includes an intermediate CRC count, I/O operation. The TCOB,
but not its location in the data IU. An when used, indicates the presence
error in that count cannot be detected of CRCs and location within the
or identified as an error. Identifying data. When TCOB is not used,
multiple intermediate CRC fields with given the DL and the DCW counts,
padding become ambiguous from the one can calculate whether or not
data count alone. intermediate CRCs are present.
Propose reject of comment.
SUN-3 T 144 8.14.6 It is specified that the final CRC shall Specify that the COB shall be required when See EMC-1. AinP C You must either support TCOB or
be checked by the Control Unit. It is Intermediate CRC is inlcuded. device dependant means for
ambiguous or impossible to generate knowing the location of CRCs
and check the final CRC, when the within the data. A miscalculation of
data counts included in the final CRC the presence or location of
are not completely specified. It is not intermdiate CRCs will result in a
specified how to specify when the final CRC error.
final CRC should not be checked.
SUN-4 T 141 188.8.131.52.2 No explanation of how to handle a CC The CC bit may be 0 only if there is a single Add: "The CC bit shall be set to 0 in A C The CC bit is also 0 in the last
bit of 0. DCW in the TCW (see 184.108.40.206). the last DCW in the TCA. (see DCW in the TCA. Section 220.127.116.11
18.104.22.168). indicates how to handle CC=0
In section 22.214.171.124.2, added cases.
statement to the CC definition that
the CC bit shall be set to zero in the
last DCW in the TCA (see 126.96.36.199).
SUN-5 E 143 188.8.131.52 4th paragraph, last sentence, "The remove the word, "pad" 184.108.40.206 has been deleted. See A C
method used for inserting CRC bytes ENDL-1
pad and …"
SUN-6 T 135 220.127.116.11 Sentence starting with, "Bytes 0-2 Add to the sentence, "and is ignored by the under figure 45, change text to: A C
shall be set to hex '00 00 00' …" control unit." "Bytes 0-2 shall be set to hex ‘00 00
Does not describe how the control 00’ by the channel and shall be
unit handles the field. ignored by the control unit." "
SUN-7 E 182 18.104.22.168 encountered is misspelled as correct the spelling corrected the spelling. A C
'encounted' in several of the TCCB
content error discriptions.
SUN-8 (late) E 150 Fig. 57 "N less than of equal …" "N less than or equal …" Changed per proposed solution. A C
SUN-9 (late) T 150 22.214.171.124 Last sentence is out of place in FC- Remove the sentence. Changed per proposed solution A C
SB-4, "Extended status is stored
directly into system storage at a
SUN-10 (late) E 183 126.96.36.199, 2nd paragraph "When the an interrogate …" "When an interrogate …" Changed per proposed solution A C
SUN-11 (late) E 206 10.2.2 "The following to apply transport …" "The following apply to transport …" Changed per proposed solution A C
SUN-12 (late) T 213 10.11 "… or if a response other than Include LS_RJT as a valid response. Change Changed per proposed solution A C
LS_ACC is received for the REC text to, "… or if a response other than LS_ACC
ELS." or LS_RJT is received …"
Emulex-001 T 3 2.1 ANSI no longer distributes standards See FC-FS-3 subclause 2.1 for the language Implement boiler plate provided by AinP C
from certain organizations. Although the ANSI has requested (per email, Snively to Bob Nixon via email to Roger
FC-SB-4 doesn't reference any of T11 editors, 26 Oct 2006). Hathorn
them, it claims ANSI can source
them. Changed according to FC-FS-3
Emulex-002 E 6 3.2.25 in the middle of the second line of Change "The mode of operation and is either Change "The mode of operation A C
3.2.25 is a spurious "and" command mode or transport mode" to "The and is either command mode or
mode of operation is either command mode or transport mode" to "The mode of
transport mode". operation is either command mode
or transport mode".
Emulex-003 T 6 3.2.27 The definition of N_Port_ID Change "Nx_Port" to "N_Port". Changed "Nx_Port" to "N_Port". A C
references Nx_Port, but Nx_Port is
Emulex-004 E 8 3.4.1 Numbered or listed items do not FC-SB-4 does not have a huge number of Commentor is incorrect about the AinP C
follow ANSI/ISO convention: Most of lists. Why not give the ANSI and ISO editors source of the conventions. The
the numbered lists do not require some help: incorporate the list conventions requirement is that what we do is
ordering. Some lists are embedded in from other standards and make the lists specificed in the conventions
paragraphs. conform? (The risk of not doing it is the ANSI section of the document, but the
and/or ISO editors may try. Who knows document must follow what is
whether they will always guess right?) stated in the conventions section.
Review to ensure that the
convention documents what we do
or that we do what the conventions
say we do.
In 3.4.1, Change "Numbered or
listed items in this standard do not
represent any priority unless
explicitly indicated" to "Items listed
in numbered or lettered lists in this
standard do not represent any
priority unless explicitly indicated"
and review ordered lists to ensure
they state specifically when order is
Changed per proposed solution.The
following sections have ordering
7.1 N_Port Link Initialization
- channel initialization procedure
- control unit initialization procedure
The statement n the document "The
initialization process may be
Emulex-005 E 12 4.4.2 In the first sentence of 4.4.2, there Would you believe "A channel provides the Changed to "A channel provides the A C
seem to be an excess "and" and a functions specified by FC-FS-2, FC-LS, and functions specified by FC-FS-2, FC-
shortage of commas. SB-4, and performs their prescribed LS, and SB-4, and performs their
protocols."? prescribed protocols."
Emulex-006 E 20 4.7.4 Beware of references using In this case, I'd recommend changing "(See Removed current reference and A C
(sub)clause numbers. If the reference FC-FS-2, clause 9.4)" to "(See N_Port_ID in added (see FC-FS-3) just after the
document is updated (e.g., from FC- FC-FS-3)" word N_Port.
FS-2 to FC-FS-3), you need to verify
the (sub)clause number.
Emulex-007 E 31 6.3.6 In the third line of the third paragraph Delete either "to" or "with". Either one works Changed "to all channel nodes to A C
from the top of page 31 appears "to for me. with which the control unit currently
with which". That's too many has at least one logical channel
prepositions for the need. pathestablished. To "to all channel
nodes with which the control unit
currently has at least one logical
channel path established.
Emulex-008 T 35 188.8.131.52 The paragraph after the bit list in the I have no idea. Changed "A fabric which supports A C
description of byte 1 of the Node the fabric requirements for FC-SB-4
Parameters of an RNID Specific Node- shall use code 1 or 3. See clause
Identification Data page is "A fabric 184.108.40.206." to "A node which
which supports the fabric supports the requirements for FC-
requirements for FC-SB-4 shall use SB-4 shall use code 1 or 3."
code 1 or 3. See clause 220.127.116.11.". It removed the reference as it is
doesn't make sense…neither party is incorrect and unnecessary.
a fabric, and the crossreference is to
a specification of the Specific Link-
Emulex-009 T 45 6.3.15 I didn't see anything specific about It should be ok to have the channel and CU SB-4 chair to investigate. R C
behavior in event of PRLO collisions respond to the PRLO as normal, even though No change.
between a channel and a control unit. each has sent a PRLO and is waiting for a
Is no specific treatment needed? Do PRLO response.
both just proceed to completion?
Emulex-010 T 71 18.104.22.168 What is the meaning of the struck-out These will go away in the final R C
note? Please resolve specifically revision. This revision shows all
whether NPIV is allowed or changes from FC-SB-3.
prohibited. (there is a similar struck-
out note on page 17)
Emulex-011 T 76 Table 7 Link Control (request) and Link Add footnote 1 references to Link Control Command mode refers to device- R C
Control (response) do not reference (request) and Link Control (response) level operations so footnote 1 is not
footnote 1, leaving it open that they necessary.
might not require command mode.
Emulex-012 T global global References in the document body globally change FC-FS-2 references to FC-FS- Will implement if containable. A C
should be to the development 3. Convention in the document has
version of the reference (if there is globally change FC-LS to FC-LS-2. been to reference the latest
one) globally change FC-SP to FC-SP-2. published standards. However,
on page 17, change FC-SW-4 to FC-SW-5. there is one new function in FC-LS-
In 2.3, add "INCITS Project 2118-D, Fibre 2 that is required for the PRLO
Channel - Physical Interface - 5 (FC-PI-5)". function. At least make sure that
On page 14, change "FC-PI-2" to "FC-PI-3, FC- reference is correct.
PI-4, FC-PI-5," Changed per proposed solution.
Emulex-013 E global global "which", which is very widely used, is Reject this comment, which would require a Accept the proposal and reject the R C
used in many cases that require which-hunt that is unjustifiably onerous comment.
"that". because the meaning is not impaired by which
word is chosen.
Red - editor to research or working group needs to discuss. Includes late comments.
Yellow - working group action item
Pink - editor to incorporate
Purple - complete
Pumpkin - Comments covered in 7/7/09 Interim call - and 7/4 adHoc - to be changed to Pink or Yellow
Green - internal discussion required.
Summary 72 All Keys:
0 All Open O Open: An action has been identified
and is not complete
41 All Accepted A Accepted: The issue has been
resolved and the resolution
indicates any necessary changes
5 All Rejected R Rejected: The issue has been
rejected, and the resolution
indicates the reason. The resolution
may also indicate changes found
useful to improve the readability of
0 All Withdrawn W Withdrawn: The commenter has
withdrawn the comment.
26 All Accepted in Principle AinP Accepted in Principle: The comment
issue has been accepted in
principle and the resolution
indicates any necessary changes
0 All Not Processed not considered yet
42 All Technical
#REF! All Open Technical
#REF! All Accepted Technical
#REF! All Rejected Technical
#REF! All Withdrawn Technical
#REF! All Not Processed Technical
30 All Editorial
#REF! All Open Editorial
#REF! All Accepted Editorial
#REF! All Rejected Editorial
#REF! All Withdrawn Editorial
#REF! All Not Processed Editorial