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

Re: Link status inconsistancies


I apologize for some of the inconsistencies, specifically associated
with Signal/Sequence in taborek_2_1100. Too much
"hip-pocket-architecting" was going on. I just sent a note to the
reflector clearing up these issues and proposing the Link Fault
encodings for all 10GE interfaces.

Best Regards,

"THALER,PAT (A-Roseville,ex1)" wrote:
> Rich,
> I'm trying to add link status siganlling to clause 49 and, looking at
> taborek_2_1100 and walker_1_0700, there seem to be some inconsistancies both
> withing toborek_2_1100 and between that and walker_1_0700 page 19 both of
> which we adopted by motions.
> Walker says that XGMII 0x9c,1 is to be used for 10 Gig Ethernet Link
> Signalling. 0x5c, 1 is for Fibre Channel ordered sets. 0x5c,1 is K28.2 in
> XAUI coding while 0x9c is K28.4.
> When I look at taborek_2_1100, page 13 says that the status signalling uses
> the "signal ordered set". However page 14 says that it uses K28.2 and says
> that is the same as the FC sequence.
> When I look at your presentation to Fibre Channel, T11/00-488v0, it says
> (page 26) that K28.2 is for primative signals and K28.4 is for primative
> sequences.
> I think that the fault signalling should use the FC sequence rather than the
> FC signal set because the
> FC signal set may include signals that one wouldn't want to delay to the
> next A. The FC sequences are also used to signal various link states and
> therefore share reaction time and duration needs with our Link status.
> Which is it suppose to be?
> Regards,
> Pat
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