Re: Break Link and Remote Fault
Seto,
As I see it, a transport for LSS which does not use Idle (in the absence of
packets) or IPG (in the presence of packets) becomes an alternative link
signaling protocol. Any such alternative link signaling protocol must still be
robust and support the required functions of a link protocol. Depending on
whether the link protocol must operate over a serial or parallel link, the
following functions may be applicable: synchronization, deskew and clock
tolerance compensation. LSS is defined atop the XAUI protocol, applicable to
parallel links as well as 64B/66B, applicable to serial links.
I have yet to see an alternative link signaling protocol proposed which meets
all link protocol requirements as well as LSS does. Do you have anything
specific in mind when you refer to an LSS protocol that doesn't use the IPG?
A question back to you and the Task Force at large: If LSS information is
terminated in the PHY and the MAC never knows about it, is anyone concerned
about the fact that LSS, like XAUI and SUPI, alter the PHY representation of the
IPG?
Best Regards,
Rich
--
"Seto, Koichiro" wrote:
>
> [Date: 07/17/2000 From Seto]
>
> Shawn,
>
> I think using MAC frame means a big change to existing MAC, thus not a
> good idea.
>
> LSS is a very solid link signaling proposal. I think we should start
> >from LSS. The issue is whether we should use IPG or not.
>
> - 1000BASE-X uses Configuration order set (code set) to communicate Link
> Failure, but not during IPG.
> - 100BASE-FX uses special code sequence to communicate Remote Fault, but
> not during IPG.
>
> I'd like to take a straw poll how many would support LSS if it does not
> mandate to use IPG. ;-)
>
> Sincerely,
> Seto
-------------------------------------------------------
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@nSerial.com
Santa Clara, CA 95054 http://www.nSerial.com