[RPRWG] Side selection
All,
My data path model has assumed that each MAC interface
consists of two sets of signals: one for each of the
ring0 and ring1 attachments.
This would be consistent with allowing the client to
specify the path of the packet, in a spatial fashion,
as this would be based on which link the packet were
to be sent.
If one were to allow the MAC to make this decision,
then there would seem to be difficulties in passing
the frame between attach-point interfaces. This would
not seem to be desirable.
Thus, from a practical point of view, it would seem
that the MAC should make all ring-selection decisions.
Comments?
DVJ
David V. James, PhD
Chief Architect
Network Processing Solutions
Data Communications Division
Cypress Semiconductor
110 Nortech Parkway
San Jose, CA 95134
Work: +1.408.942.2010
Cell: +1.650.954.6906
Fax: +1.408.942.2099
Work: djz@xxxxxxxxxxx
Base: dvj@xxxxxxxxxxxx
-----Original Message-----
From: David James [mailto:djz@xxxxxxxxxxx]
Sent: Monday, November 19, 2001 10:33 AM
To: stds-802-17@xxxxxxxx
Subject: Outline ordering
Editors,
I would like to proposse changing (slightly) the ordering of clauses
proposed last week:
From (last weeks):
Annex I: Code examples
Annex J: Stratum clock distribution
Annex K: Implementation guidelines
To:
Annex I: Stratum clock distribution
Annex J: Implementation guidelines
Annex K: Code examples
The reason is rather pragmatic; the C-code section has a
distinct format (landscape, rather than portrait). Thus,
it is more difficult to switch section formats twice if
this is surrounded by normal text. Better for the editor
and the reader is the "special" stuff is at the end.
DVJ
David V. James, PhD
Chief Architect
Network Processing Solutions
Data Communications Division
Cypress Semiconductor
110 Nortech Parkway
San Jose, CA 95134
Work: +1.408.942.2010
Cell: +1.650.954.6906
Fax: +1.408.942.2099
Work: djz@xxxxxxxxxxx
Base: dvj@xxxxxxxxxxxx