Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: Clause 48 comments




Ben,

I realize that a response to this note is long overdue, but I'm finally
getting back to it. Here are my responses to your clause 48 comments.

Ben Brown wrote:
> 
> Rich,
> 
> Just a few comments regarding clause 48. Sorry if these were
> already brought up last week.
> 
> 48.1.4 - page 128
>   I thought clause 47 was for XAUI and this clause was for
>   LAN WWDM. I'm confused by the statement on line 48 talking
>   about this being a chip-to-chip interface.

You are correct in which clause does what. However, the LAN WWDM PCS and
PMA constitute a chip-to-chip interconnect as does XAUI. It's up to the
LAN WWDM PMD to turn the electrons into photons. 
 
> 48.1.6 - page 129, line 11
>   Change "map onto to the" to "map onto the"

Done.
 
> 48.2.2 - page 131, line 54
>   Change "PCS Receive process" to "PCS Deskew process"

Done.
 
> 48.2.4.3 - page 135, figure 48-5
>   Under Code_Sel decodings, 0 should be used to indicate /K/

Done.
 
> 48.2.4.3.2 - page 135, line 46
>   Change "initiator" to "transmitter"

Done.
 
> 48.2.4.3.3 - page 21, line 21
>   Change "many" to "may"

Done.
 
> General
>   Change "single-code groups" to "single-code code-groups"
 
Done. 2 occurrences.

> 48.2.5.1.3
>   - rx_even=TRUE for both cgbad and cggood makes me think
>     this hasn't been updated yet from clause 36

Correct. Revised in rev 1.01.

>   - rx_code-group - line 22 - Change "all three lanes" to
>     "all four lanes"

Done.
 
> 48.2.5.2.1
>   Page 141, figure 48-6 - If /T/ is lost, the next packet
>     is also lost. Should there be some detection of /I/ that
>     returns the machine to XGMII_IPG?

Yes. The State Diagrams are being revised to reflect earlier /I/
detection. 

>   Page 142, figure 48-7 - Should this state machine describe
>     how to pad octets after /T/ with /K/?

I'd rather do this with a function rather than a state diagram.

>   Page 143, line 3 - Reference to Figure 48-5 should be Figure
>     48-8

Done.
  
> 48.2.5.2.2 - page 143, line 25
>   Reference to Figure 48-8 should be Figure 48-9

Done.
  
> 48.2.5.2.3 - page 143, lines 45-49
>   This, also, looks like Clause 36 stuff.

Correct. Revised in rev 1.01.

> 48.3.2 - page 149, line 32
>   The note references clauses 38 & 39. This should be
>   clause 54 (I think).

Done.
  
> 48.3.2.3 - page 150
>   I would have expected some discussion of comma alignment
>   in this section. Is this done somewhere else?

Correct. Revised in rev 1.01.

> 48.3.4 - page 150, lines 42
>   Do you expect this reference to Annex 36A for test patterns
>   to change or will Annex 36A change to reflect that these
>   test patterns also support 10GBASE-X (and XAUI?)?

I expect the reference to change. Editor's note inserted.
 
> Ben
> 
> --
> -----------------------------------------
> Benjamin Brown
> AMCC
> 603-798-4115 - Home
> bbrown@xxxxxxxx
> -----------------------------------------

-- 

Best Regards,
Rich
                                      
------------------------------------------------------- 
Richard Taborek Sr.                 Phone: 408-845-6102       
Chief Technology Officer             Cell: 408-832-3957
nSerial Corporation                   Fax: 408-845-6114
2500-5 Augustine Dr.        mailto:rtaborek@xxxxxxxxxxx
Santa Clara, CA 95054            http://www.nSerial.com