Issues with diagrams in 11-036r9

Stephen Finch Stephen.Finch at wdc.com
Tue Jul 26 06:51:59 PDT 2011


Formatted message: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1107262_f.htm">HTML-formatted message</a>

Back in SAS-2 we had definitions for ALT (Actual lock time) and ATT (Actual
Training Time).  Something like these definitions would be useful.  These
acronyms were removed in SPL-1, but the definition was kept.  In the draft
of SPL-2, before the addition of 11-036, these terms were still in
existence.  11-036 did not add an equivalent “actual” time for the TX
Train
window and, a point I had missed, deleted the “Actual training time” term
for the Train_Rx-SNW.	 What is needed in figure 17 is the usage of the
term “Actual Transmitter Training time”.   If the “<” is the proposed
solution, I think text needs to be added to clarify that “<” means actual
transmitter training time.  I really don’t know if I’m comfortable with
the
use of “<”, but I’ll not raise any objection at this time. 
You are right, I mis-typed.  Figure 16 shows Train_Rx-SNW as consisting of
RCDT and MTT times.
Figure 17 calls the window after the Train_Tx-SNW by the name Train_Rx-SNW,
but that window in this diagram does not contain an RCDT.  For that reason,
technically, the receiver training taking place after the Train_Tx-SNW is
not a Train_Rx-SNW.  The only thing I can think of that fixes this problem
is add a name that defines the training portion of the Train_Rx-SNW window,
then extend the definition of the Train_Tx-SNW to have three parts:  RCDT,
Tx Training and Rx Training, and use the name for the Train_Rx window second
half in the Train_Tx-SNW definition.
Regards,
Stephen Finch
720 864-4751
720 883-1414 (cell)
Western Digital, Longmont
This communication is for the exclusive use of the addressee and may contain
Western Digital confidential and/or non-disclosable information. Any
distribution, reading, copying or use of this communication by anyone other
than the addressee is strictly prohibited and may be unlawful. If you have
received this message by mistake, please delete it immediately and advise
the sender you have done so by sending an E-Mail to Stephen Finch
<stephen.finch at wdc.com>
From: Penokie, George [mailto:George.Penokie at lsi.com] 
Sent: Tuesday, July 26, 2011 7:27 AM
To: Stephen Finch; t10 at t10.org
Subject: RE: Issues with diagrams in 11-036r9
Steve,
Figure 16 does shows Train_Rx-SNW not Train_Tx-SNW so I’m not sure what
your
point is in 1.
I do understand your point about MRTT and MTTT. That is also a problem in
the current (and past) standards. I have resolved it by placing a < in front
of the MTTT and MRTT when it they are intended to not represent the maximum
time being reached. I have made this change in all the relevant figures in
SPL-2.
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 Stephen
Finch
Sent: Monday, July 25, 2011 11:42 AM
To: t10 at t10.org
Subject: RE: Issues with diagrams in 11-036r9
Resent with the WD proprietary message removed.
Stephen Finch
720 864-4751
720 883-1414 (cell)
Western Digital, Longmont
From: owner-t10 at t10.org [mailto:owner-t10 at t10.org] On Behalf Of Stephen
Finch
Sent: Monday, July 25, 2011 9:00 AM
To: t10 at t10.org
Subject: Issues with diagrams in 11-036r9
Figure 17, on page 46, shows the Train_Tx-SNW as being composed of RCDT and
MTTT before exiting to Train_RX-SNW.  There are two issues with this:
1.	  Figure 16 shows the Train_Tx-SNW.  It consists of RCDT followed by
MRTT.  In the case of Train_Tx-SNW exiting to Train_Rx-SNW, the RCDT does
not exist.  Therefore, technically, what the Train_Tx-SNW is going to is
only a part of the Train_Rx-SNW.
2.	 Figure 17 shows MTTT as occurring before the transition to
Train_Rx-SNW.  In fact, if MTTT is met, the transition will not be to
Train_Rx-SNW.
3.	 Figure 16 has the same issue for MRTT as in item 2.
I think the text in the state machines is what is intended.  It just that
one could misinterpret the diagrams.
Stephen Finch
720 864-4751
720 883-1414 (cell)
Western Digital, Longmont



More information about the T10 mailing list