RE: [EFM] RE: OAM Transport Proposal
Comments are in line...
At 04:27 PM 05/02/2002 -0700, Kevin Daines wrote:
>I'm just commenting on one paragraph of your e-mail, included below. (Rest of message deleted for everyone but Pat who enjoys Ethernet in the First Mile :)
>--- from Sanjeev e-mail ---
>Since 802.3ah is tasked to define new PHYs so I do not see why a PHY can not be defined
>to guarantee to pass a complete preamble to RS and don't see why specifying so for a PHY
>would break 802.3 spec. Correct me if I am wrong, it is not expected for 802.3ah 1000Mbps
>PHY to be compatible with 1000Base-X PHY or 100Mbps copper PHY to be compatible
>with 100Base-T PHY. So, if that is not the case what is the big issue here?
>KQD> My reading of and interpretation of the 802.3ah objectives leads me to believe you are wrong, at least in part. Consider the following objective:
>Provide a family of physical layer specifications:
>- 1000BASE-LX extended temperature range optics
>- 1000BASE-X >= 10km over single SM fiber
>- 100BASE-X >= 10km over single SM fiber
>- PHY for PON, >= 10km, 1000Mbps, single SM fiber, >=1:16
>- PHY for PON, >= 20km, 1000Mbps, single SM fiber, >=1:16
>- PHY for single pair non-loaded voice grade copper distance >=750m and speed >= 10Mbps full-duplex
>Note: This is taken from http://www.ieee802.org/3/efm/public/mar02/objectives_2_0302.pdf
SM:-Thank you for the info, appreciate it.
>Yes, EFM is defining new PHYs for PON and for copper.
SM:-Here you by "PHY" means the "PCS+PMA+PMD" all are subject to change, correct?
If yes then there will be complete new definition of the PCS also.
>However, EFM is defining new PMDs - not PHYs - compatible with Clause 36 1000BASE-X PCS and Clause 24 100BASE-X PCS.
SM:-100Base-X PCS does not do any shrinkage to preamble so not an issue. As is well known the 1000Base-X
PCS at TX makes use of this feature. If the PMD part of the 1000Base-X PHY is being re-written perhaps
the 1000Base-X PCS clause could also be updated with that "optonal OAMiP capable PCS would not shrink
preamble but instead would vary IPG to meet even-odd alignment". This is just a suggestion. Could be solved
>So, I believe it IS expected for 802.3ah to be compatible with 1000BASE-X.
SM:-Making above modification would not break anything and would be
compatible with 1000Base-X as it is optional. What it would mean though
that PCS designers/suppliers need to open up their designs and update it
if they wish to incorprorate this feature and add value to their designs/devices.
Those who dont see any value would just walk away and dont implement it.
>Can this compatibility be achieved through the creation of a new RS? Can this compatibility be achieved by using a lesser number of preamble bytes? Can we resolve the issue of MDC/MDIO connectivity? I believe the answer to each of these is yes. We just need to decide the path the Task Force is going to take (and give us editors some direction!)
SM:-Not that it is not good to solve this with smaller number of bytes, the problem is that as is
there are only 8 bytes out which 1 byte is gone for CRC, 2 are gone for PHY ID, 1 is
gone for SOP then only left 4. I am not sure shrinking these header bytes is really a good
idea for future enhancements. If you look at other protocols/standards those are trying to
solve the problem with many more bytes. So, I think more is better.
>Editor, IEEE P802.3ah EFM OAM