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

RE: Clause 48: Errors after T.

  D2.3 modified the definition of the check_end function to look at the
current ||T|| column as well as the column after ||T|| and push back any
running-disparity error past /T/ to the previous column.  This was done to
address a problem raised by Pat Thaler when an error is detected in lanes
following the /T/ in the ||T|| column.

  In general however, errors between /S/ and /T/ should not be pushed back,
as this might result in an /E/ code destroying the /S/ or being moved before
the /S/ delimiter.

  To be clear, consider the following cases and how the current definition
of check_end would handle them:
 {D="data code" I=idle  X="a detected running disparity error" E=the error

case 1:  (RD error in column after ||T||)
lane0   DEX
lane1   DTI
lane2   DII
lane3   DII

case 2:  (RD error after /T/ in ||T||)
lane0   DDI
lane1   DTI
lane2   DII
lane3   EXI

case 3: (RD error before /T/ in ||T||)
lane0   DXI  (note, the detected "X" should be replaced with an "E" - no
push back)
lane1   DTI
lane2   DII
lane3   DII

clearly case3 could be handled such that the "E" is pushed back to the
previous column, but that's not how check_end is currently worded.   Since
an ||S|| should not immediately precede a ||T|| there should be no
significant problem with pushing any RD error anywhere in ||T|| back one
column, if you feel its a big enough issue then by all means fire off a
comment against D3.0 and we can address it in May.

  Bob Noseworthy
  (603) 862-4342
  Clause 48 quasi-co-editor
  UNH InterOperability Lab

> -----Original Message-----
> From: owner-stds-802-3-hssg@xxxxxxxx
> [mailto:owner-stds-802-3-hssg@xxxxxxxx]On Behalf Of Justin Gaither
> Sent: Monday, March 26, 2001 5:59 PM
> To: 802.3ae
> Subject: Clause 48: Errors after T.
> Everyone,
> During the Plenary, it was determined that we need to push an error that
> occurs after the /T/ back 1 column so that it is included inside the /S/
> and /T/ delimeters.  My question is, can we just push all errors back on
> column, or must we determine if it is in the column following a //T//?
> I could not find this in D2.3, where did it get put?
> Thanks
> --
> Justin Gaither                       Phone: 512-306-7292  x529
> RocketChips a Division of Xilinx     Fax:   512-306-7293
> 500 N. Capital of TX Hwy.
> Bldg 3                         email: jgaither@xxxxxxxxxxxxxxx
> Austin, TX 78746               WWW: