Questions/comments regarding 11-036r6 proposal (Transmitter training)

Karthikeyan, Kishore K kishore.k.karthikeyan at intel.com
Wed Jun 15 14:01:55 PDT 2011


Formatted message: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1106154_f.htm">HTML-formatted message</a>
Attachment #1: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1106154_image001.png">image001.png</a>

Hi George
Thanks for the response.
Regarding #1
Table 9 says that the pattern marker has to be present before the TTIU and
not after the 58 scrambled dwords. My problem with this statement is that
when will I generate this invalid pattern marker message? The valid/invalid
pattern marker messages go to the PTT_PL state machine. So are we saying that
the phy's receiver should generate an invalid pattern marker message even
though no invalid pattern marker was detected (according to the definition of
invalid pattern marker that we defined in 5.7.4.2.3.4.3). And how many
messages will we generate and for how long? The statement makes it vague and
open to different interpretations.
Regarding #8:
I made the comment about "the phy shall" because you seemed to have changed
the wording from "the phy shall" to "this state shall" in the same section
5.9.4.4.1 --> see the first line of this section marked 1). I was trying to
make it consistent because paragraph 1 of this state's description says "this
state shall" and paragraph 2 says "the phy shall"
Thanks
Kishore
From: Penokie, George [mailto:George.Penokie at lsi.com]
Sent: Wednesday, June 15, 2011 12:46 PM
To: Karthikeyan, Kishore K; t10 at t10.org
Subject: RE: Questions/comments regarding 11-036r6 proposal (Transmitter
training)
Kishore,
Thanks for your comments. See my responses below.
Bye for now,
George Penokie
LSI Corporation
3033 41st St. NW
Suite 100
Rochester, MN 55901
507-328-9017
george.penokie at lsi.com
From: owner-t10 at t10.org [mailto:owner-t10 at t10.org] On Behalf Of Karthikeyan,
Kishore K
Sent: Wednesday, June 15, 2011 12:32 PM
To: t10 at t10.org
Subject: Questions/comments regarding 11-036r6 proposal (Transmitter
training)
Hi all
I had some comments on the 11-036r6 proposal that I was hoping to get covered
in the call today but the email I sent last night bounced from t10 reflector
and unfortunately, I missed today's conference call too.
Given below are the review comments/questions from my review of pages 1- 88
of 11-036r6.
Thanks
Kishore
#1:
Pg 33, last line
"If the phy's receiver does not detect a pattern marker after a Train_Tx
pattern then the pattern marker shall be invalid.
What is the above statement trying to say? As there is no time bound for this
pattern detection, what scenario is this statement trying to cover?
GP - The Train-Tx pattern is very specific and clearly defined in table 9. If
there is no patter marker immediately following the 58 dwords of zeros that
is considered an invalid pattern marker.
#2:
Pg 35
The control/status TTIU is used by a phy to:
a) ....
b) ......
Statement b) does not match the contents of the fields. Training status word
contents are not used to receive status of the attached phy's transmitter. It
is used to convey the current status of the transmitter training to the
attached phy.
GP - It depends on what side of the but you are on. Your statement is correct
but so is the one in the proposal. For now I am not making any change here.
#3:
Pg 44
[cid:image001.png at 01CC2B5C.F3B4F4D0]
Shouldn't this read as follows --> then a new Train_Tx-SNW shall be performed
(if possible) or Train_Rx-SNW shall be performed (if Train_Tx-SNW not
possible) based on the next highest, untried, commonly supported settings.
GP - Yes - Fixed
#4:
Pg 37
In the paragraph about phy reset problem
A phy reset problem occurs:
a) ----
b) ----
c) ----
Shouldn't we also add one more condition to this to cover Train_Tx-SNW?
d) after a Train_Tx-SNW, if the Train_Tx-SNW is invalid and there are no
additional, untried, commonly supported settings.
See pg 81 --> 5.9.4.5.1
GP - Yes - Fixed
#5:
Pg 54
Just below the figure
Spelling mistake -->  Transmitter
GP - Fixed
#6:
Page 53, last line
Grammatical error --> after "of"
GP - Fixed
#7:
pg 55
Typo in figure --> TTIU instead of ITTU
GP - Fixed
#8:
Pg 81
Instead of "the phy shall", it should be "this state shall" in 2)
GP - The 'the phy shall' wording is the way these things are stated in the
SPxx state machines.  Just look in SLP-2 and you see that statement all over
the place in the SP state machine sections.
#9:
Pg 88
typo (use 1,2 and 3 instead of 1,1,1 in these paragraphs for the "transmitter
Adjustment complete" and the "current coefficient")
GP - Fixed
#10:
Just for my understanding, what was the reason for disabling OOB detection
during Train-Tx window? We don't disable it for Train-Rx then why disable it
for Train-Tx?
GP - I am not an expert on this but what I understand is that Manchester
encoded period is close to OOB timing and that could cause problems.



More information about the T10 mailing list