IEEE P802.3ae D2.1 Comment Resolution

Brad Booth

Informations

• Publié par: Vaep
• Nombre de lectures: 35
• Langue: English

Extrait

P802.3ae Draft 2.1 Comments
Cl SC P L Cl SC P L00 # 33 00 00 # 664
Stoltz, Mario ChipIng.de, an Intel co William G. Lane CSU, Chico
Comment Type E Comment Status R Comment Type E Comment Status R
Mixed usage of "baud" (20 occurrences) and "b/s" (many, many more) in the draft does not There is a general inconsistency in the way clock frequencies and transmission rates are
seem to follow any plan. specified in the various clauses, including for example:
SuggestedRemedy
Clause 46
Replace all "bauds" with "b/s" unless there actually WAS a plan that simply escaped the 9.58 Gb/s STS-192 payload rate (see 46.1.3),
commentor's limited viewfield. 156.25 MHz+/-0.01% (46.3.1.1 -- TX_CLK at the XGMII),
nominally 156.25 MHz (46.3.2.1 - RX_CLK at the XGMII), Proposed Response Response Status C
REJECT.
Clause 47
nominal rate of 3.125 Gbaud (47.1.3)
There are cases where baud and b/s do imply different things. The editor respectfully requests 3.125 Gbaud +/-100ppm (47.3.3)
that the commenter re-submit a comment in the next ballot period that indicates where this
wording should be changed. Clause 48Cl 00 SC P L 48 # 507
312.5 MHz +/-100 ppm (48.3.1.1)
Law, David J 3Com nominal rate of 312.5 MHz ( 100 ppm, as governed by frequency and tolerance of XGMII
TX_CLK (48.3.2.1.2)Comment Type E Comment Status A
The address for the IEEE on the front sheet is incorrect Clause 49
644.53 Mtransfers/s and 599.04 Mtransfers/s (49.1.5)SuggestedRemedy
Please correct the address to read:-
Clause 50
payload capacity of STS-192c / VC-4-64c, i.e., 9.58464 Gb/s. (50.1)
3 Park Avenue,
9.95328 Gb/s effective data rate (50.1.2)New York,
nominal rate of 599.04 MHz corresponding to the STS-192c payload rate of 9.58464 Gb/s
NY 10016-5997,
(50.2.1..2)USA
Proposed Response Response Status C Clause 51
nominal clock rate of 644.53125 MHz and 622.08 MHz in 10GBASE-R and 10GBASE-W ACCEPT.
operations, respectively. (51.4)
9.95328Gb/s (10GBASE-W family) or 10.3125Gb/s (10GBASE-R family) bit clock (51.4)Change incorporated in D2.2.
Clause 52
10.3125 +/- 100 ppm (table 52-6)
9.95328 +/- 100 ppm (table 52-6)
Clause 53
2.48832 GHz bit clock to generate the 622.08 MHz data-group clock (53.1.3)
rate of 2.48832 GHz (53.3.2)
Clause 54
3.125 GBd +/-100ppm (table 54-7)
2.48832 Gbd +/-100ppm (table 54-11)
The problem is more than just terminology. The number of significant digits used in defining
transfer rates varies from 4 to 8 for values that are all supposedly derived from the same
TX_CLK which has a specified tolerance of +/-0.01%.
TYPE: TR/technical required T/technical E/editorial COMMENT STATUS: D/dispatched A/accepted R/rejected SORT ORDER: Clause, Page, Line, Subclause Page 1 of 144
RESPONSE STATUS: O/open W/written C/closed U/unsatisfied Z/withdrawn Cl 00 SC 00P802.3ae Draft 2.1 Comments
SuggestedRemedy
Cl SC P L00 00 00 # 289
The editors need to agree as a group on:
Christensen, Benny Intel / GIGA
A consistent terminology, when and when not to use nominal, whether to use Hz, b/s, or Baud Comment Type T Comment Status R
How to specify tolerance values for transfer rate - % or ppm.
X-referencing between/across major clauses may not be correct due to renumbering in the The number of significant digits that should be used in representing these rates.
individual clauses.
Proposed Response Response Status C
SuggestedRemedy
REJECT.
Fix it.
The Editor-in-Chief rejected this comment because the comment doesn't provide direction and Proposed Response Response Status C
relies on the editors to determine the terminology and significant figures. The Editor-in-Chief will REJECT.
re-submit this comment in the next ballot cycle, on behalf of the commenter, to provide a specific
direction that the editors should take. Editor respectfully requests that commenter re-submit this comment in the next ballot cycle, and
that the commenter indicate the incorrect cross-references.Cl 00 SC 00 P L # 638
William G. Lane CSU, Chico Cl 00 SC 53 P 456 L 1 # 587
Rich Taborek nSerial CorporationComment Type E Comment Status R
The representation of logic values zero and one are sometimes represented in lower case and Comment Type T Comment Status A
other times in upper case. No electrical interface and associated jitter specifications is specified for the LW4-PMA as is the
SuggestedRemedy case for the LX4-PMA in clause 47. Clause 47 specs are not applicable to clause 53 as they
apply to 8B/10B transmission code only.Chose a consistent way for all editors to use.
SuggestedRemedyProposed Response Response Status C
The LW4-PMA is incomplete and should be deleted.
REJECT.
Proposed Response Response Status C
ONE and ZERO are used as a variable expression; whereas, zero and one are used as a logic ACCEPT.
value. The editor respectfully requests that the commenter review the use of these in the next
draft and submit comments in the next ballot cycle to indicate which are used incorrectly. Move to accept response and remove all technical content that exclusively supports 10GBASE-
LW4, and make all editorial changes necessary to remove references to 10GBASE-LW4.Cl 00 SC 00 P 00 L # 301
Christensen, Benny Intel / GIGA Moved: B. Grow
Seconded: T. DineenComment Type T Comment Status A
Technical (75%)
Several clause headlines (like clause 47) and in-line text use the spelled 'TEN' Gigabit ..... for
XSBI or XAUI where 'X' is the acronym for '10'Definition in clause 1.4.xxx use '10' Gigabit 802.3 Voting Members
Attachment Unit (XAUI)not the spelled version. Y: 54 N: 6 A: 17
SuggestedRemedy PASSES
change all 'Ten' to '10' as per definition
P802.3ae Task Force Members
Proposed Response Response Status C Y: 73 N: 6 A: 37
PASSESACCEPT.
Change incorporated in D2.2.
TYPE: TR/technical required T/technical E/editorial COMMENT STATUS: D/dispatched A/accepted R/rejected SORT ORDER: Clause, Page, Line, Subclause Page 2 of 144
RESPONSE STATUS: O/open W/written C/closed U/unsatisfied Z/withdrawn Cl 00 SC 53P802.3ae Draft 2.1 Comments
Cl SC P L Cl SC P L01 1.4 5 35 # 535 04 4.2.8 27 17 # 64
Law, David J 3Com Brown, Benjamin AMCC
Comment Type E Comment Status A Comment Type T Comment Status R
The numbering of the side stream scrambling definition does not match the Y2K 802.3 edition. The MAC should only use the provided FCS if it is present and the MAC supports that mode.
SuggestedRemedy SuggestedRemedy
Correct the definiotn numbering to match the Y2K edition. Define new Transmit State Variable: supportsFCSParam: Boolean; {Indicates that the MAC
supports using a provided fcsParamValue from the MAC Client}Then replace "if
Proposed Response Response Status C fcsParamPresent then" with "if fcsParamPresent and supportsFCSParam then"
ACCEPT. Alternative solution: Modify definition of fcsParamPresent to only be true when the MAC
supports this mode, but this presupposes the MAC Client knows the modes the MAC operates
Cl 01 SC 1.4 P 5 L 35 # 536 in.
Law, David J 3Com Proposed Response Response Status C
REJECT. Comment Type E Comment Status A
I do not believe the additional variable is necessary. The proposed "alternativeThe term 'Lane' needs to be added to the definitions.
solution" has already been incorporated into Clause 4, in principle.
SuggestedRemedy The behavior of the MAC in the presence and absence of fcsParamPresent in
both modes has been adequately described in Clause 4, with the appropriateAdd the term 'Lane' to the definitions.
"shall" statements, and there is no need to complicate the Pascal more than
Proposed Response Response Status C necessary.
See 4.1.2.1.1 (Pg. 14, Ln. 28-31), 4.2.3.1 (Pg. 19, Ln. 8-10),ACCEPT.
4.3.2 (Pg. 39, Ln. 20-21).
Cl 01 SC 1.5 P 7 L 10 # 5
Cl SC P L04 4.4.2 41 29 # 614
Stoltz, Mario ChipIng.de, an Intel co
Bottorff, Paul Nortel Networks
Comment Type E Comment Status A
Comment Type T Comment Status R
"OIF" is wrongly explained as "optical interface forum".
The ifsStretchRatio inserts more idles than necessary.
SuggestedRemedy
SuggestedRemedy
Replace with correct reading "optical internetworking forum"
Change the value from 104 to 105.
Proposed Response Response Status C
Proposed Response Response Status C
ACCEPT.
REJECT.
I believe that the value for ifsStretchRatio is correct. It has been defined suchCl 02 SC 2.3.1.2 P 10 L 43 # 176
that for the specified number of bits, one octet of IPG will be generated by the
Edwards, Gareth D. Xilinx Pascal code. The editor humbly requests that the commenter re-submit this
comment in the next ballot cycle, with a detailed explanation why this valueComment Type E Comment Status A
should be any different.
Missing "the" in sentence "There is sufficient information associated with
mac_service_data_unit for the MAC sublayer entity to determine the length of the data unit."
SuggestedRemedy
Change sentence to "There is sufficient information associated with the mac_service_data_unit
for the MAC sublayer..."
Propo