Re: Link Status thoughts
Tripathi,
Thanks for the explanation. I've researched SRP and have a good idea of
what you're suggesting now. My sense is that ring protection protocols
and 10GE are mutually exclusive. 10GE is a point-to-point only MAC and
Physical Layer protocol. As such, link status information need not be
passed to the MAC. Note that this does not mean that Ethernet Station
Management is required to be slow (i.e. not real time). It also does not
preclude the construction of large mission critical Ethernet MANs and
WANs which employ switched point-to-point links. I do believe that the
latter is outside the scope of P802.3ae.
Best Regards,
Rich
--
Devendra Tripathi wrote:
>
> Hi Rich,
>
> >Tripathi,
> >
> >Please help me with a definition of and requirements for "ring
> >protection mechanisms". My sense is that robust handling of
> >signal_detect at each link element receiver addresses these
> >requirements.
>
> Yes, it does. Only thing is that since ring protection protocols (like SRP)
> would be running at MAC layer, the information about link status must reach
> the MAC.
>
> >Please also explain what you mean by a "MDC/MDIO multi master interface"
> >and how this ties into the processing Link Status information.
>
> If the link status is going to reach only upto PHY (through the status
> register 1.2) , the only way it can reach the MAC (mgmt entity) is through
> MDC/MDIO interface. Since MAC mgmt entity is the master of this interface,
> there is no way PHY can communicate the change in link status in real time
> to MAC. If the protocol is updated such that PHY could become master, it
> could do so. That is what I meant by multi-master interface.
>
> Regards,
> Tripathi.
>
> >Best Regards,
> >Rich
> >
> >--
> >
> >Devendra Tripathi wrote:
> > >
> > > Hi Rich,
> > >
> > > One related point to this is the ring protection mechanisms which are being
> > > handled separately. It is important that Link status information reaches
> > > the MAC layer in real time
> > > which as you have pointed here is not possible via MDC/MDIO (unless it is
> > > made multi master
> > > interface).
> > >
> > > Best Regards,
> > >
> > > Devendra Tripathi
> > > Vitesse Semoconductor Corporation
> > > 3100 De La Cruz Boulevard
> > > Santa Clara, CA 95054
> > > Phone: (408) 986-4380 Ext 103
> > > Fax: (408) 986-6050
-------------------------------------------------------
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