SAS PHY teleconference reminder - 9/23/02 2pm - 4pm CDT

Alvin.Cox at seagate.com Alvin.Cox at seagate.com
Mon Sep 23 07:54:25 PDT 2002


* From the T10 Reflector (t10 at t10.org), posted by:
* Alvin.Cox at seagate.com
*
SAS PHY Teleconference Monday, September 23, 2:00 pm - 4:00 pm CDT

PARTICIPANT INFORMATION:
-----------------------
All Participants should use the following information to reach the
conference call:

Toll Free Dial in Number:  (866)279-4742
International Access/Caller Paid Dial In Number:  (309)229-0118

PARTICIPANT CODE:  3243413

Webex information:

Meeting Summary
Meeting Name:  SAS PHY WG
Scheduled Time: 9/23/2002 at 1:45PM (GMT -05:00) Central Time, USA & Canada
(DayLight Time).
Meeting Number: 281981011
Any meeting attendee can use this number to join the meeting, at
          https://seagate.webex.com/join/

Password: LED101


Agenda:

1.  LED driver requirements

Barry Olawsky presentation concerning system level impact of 15mA current,
3.6V max.

2.  Test Patterns

It was determined that since there is a fixed 24 byte header, that the
scrambling offset can be factored into the data pattern such that the
CJTPAT can be accomplished by a fixed pattern, independent of the address.
Aligns sent during the transmission could be negative on the test pattern
effectiveness. A note should be added concerning aligns during this test
pattern. Bernhard to revise pattern and work with Alvin to document. Review
Status.

A description of the CRPAT will be included but no instructions on how to
achieve it on the wire.

All participating on the call are to check with their safety agency experts
to determine if a documented means of disabling scrambling requires the
system to be tested in that mode.


3.  Common mode specification

Russ is concerned that long cables may not allow the skew aspect to be
covered by the common mode specification. He will perform additional tests
and share results on the call.

4.  Test loads

Posting due 9/20 (Cox) Not posted yet. Review figures during call.

5.  OOB transmission levels

Targets should transmit at SAS levels as they are keyed to not connect to
SATA devices. Initiators and expanders should begin at SATA voltage levels
since a SATA device may be attached to it. Voltage level may increase to
SAS level if OOB is not completed.

During the meeting last week, it was noted that some areas of the OOB text
do not properly describe the transition to idle time.

Cox to provide text for review during call.

6.  Define conditions for "loss of signal"
Based on the impact to the system, what defines a loss of signal? One bit?
Voltage level loss? Etc.

All to review specification use of "loss of signal" and determine if
necessary. Suspected carry-over from FCAL concerning optical connection
rather than copper. Discuss on call.

7. Review what PHY group wants to count on the SMP REPORT PHY ERROR LOG.

List below made intially by PHY working group. Is this what we want? What
interactive errors stop counts on others?

PHY group suggested error report items in order of significance:

* Loss of signal
* OOB signal detected but can't negotiate OOB (expander log)
* Speed negotiation at each speed (count of successes for G1, G2, ?)
* Disparity error count
* Illegal character count
* CRC error count
* Loss of synchronization

PHY group to reconsider list as is currently proposed. Juzer has provided
the following link that should be considered when evaluating the need for
this log page.

www.reed.com/Papers/EndtoEnd.html

The paper presents a persuasive but heuristic argument of the end-to-end
principle.  The principle suggests that functions (like loss of signal and
other error check functions) placed at low levels of a system may be
redundant or of little value when compared with the cost of providing them
at that level.

8.  New items.

9.  Meeting schedule.

*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org




More information about the T10 mailing list