XAUI electrical status
Work on XAUI electrical issues has been proceeding in several quarters, but
with little public discussion of late. Following is a summary of XAUI
electrical issues to spur both proposals on D3.0 and further work in
preparation for the Interim next month. Jitter specific issues are being
tracked separately by Anthony and are not listed here.
Compliance channel phase response
(See
http://grouper.ieee.org/groups/802/3/ae/public/adhoc/serial_pmd/xaui_documen
ts/XAUI_mar01_agenda.pdf page 8.)
- Please consider whether a phase specification is really needed for driver
compliance.
Return loss
(See
http://grouper.ieee.org/groups/802/3/ae/public/adhoc/serial_pmd/xaui_documen
ts/XAUI_mar01_agenda.pdf page 9.)
- Measurements in progress by Ishwar.
- Additional chip measurements would be useful.
- Definitive analysis presentation needed to determine acceptable limits
with respect to system performance
Vmax
(See
http://grouper.ieee.org/groups/802/3/ae/public/adhoc/serial_pmd/xaui_documen
ts/XAUI_mar01_agenda.pdf page 10.)
- Solution proposals needed.
Crosstalk
(See
http://grouper.ieee.org/groups/802/3/ae/public/adhoc/serial_pmd/xaui_documen
ts/XAUI_mar01_agenda.pdf page 11.)
- Study being done by John D.
- Additional issue: The far end eye of Fig. 47-4 is used to define both
driver compliance (47.3.3.5) and the valid receive signal (47.3.4.1). There
is no difference in vertical eye opening to account for crosstalk in the
receive case.
Compliance channel low frequency spec limit
- Technical opinions welcome concerning how low in frequency the compliance
channel spec should cover.
Driver compliance eyes
- Data pattern is not defined.
- D3.0 allows using either a system clock or a golden PLL for eye trigger,
but eye results will differ with some data patterns.
Please review and resubmit relevant XAUI comments on D2.1 which were
deferred to D3.0:
http://grouper.ieee.org/groups/802/3/ae/comments/d2.1/index.html
Thank you all.
-Dawson Kesling
Editor, Clause 47