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

Re: Draft 2.0 (figure 48-6)




Brian,

A_CNT_0 is not the counter, it is a variable set to 1 when the counter
gets down to zero. However, I see your point. This variable should not
be reset until ||A|| is sent, thereby guaranteeing sufficient ||A||s in
the PCS stream. Look's like another comment. Thanks.

Best Regards,
Rich
   
--

Brian Cruikshank wrote:
> 
> Rich,
> 
> I have another question on the same figure.  It is on the
> conditions for going from SEND_DATA to SEND_A or SEND_K.
> 
> If the path to SEND_A requires (code_sel=1*A_CNT_0=1) this
> path may be very rare.  The proposed PRBS counter (I believe)
> increments every clock cycle whether it is data or idle.  If
> the IPG is consistently low, there may not be many /A characters
> sent.
> 
> Maybe it would be better to leave the A_CNT_0=X out of
> these conditions.  Another alternative is to make the PRBS counter
> only count during idles; that would make /A characters appear
> every 12 packets on average (24 average divided by 2 possible min
> IPG locations).
> 
> I believe there was some discussion this already, but I cannot
> remember if it was resolved.
> 
> /Brian Cruikshank
                                   
------------------------------------------------------- 
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