[T10] Clarification requested for the detection of scrambled DWORDS in Train-TX Speed Negotiation window

Lana Chan lana at cadence.com
Tue Jan 17 11:30:10 PST 2017


Hello,

We have a query regarding Scrambling in Train-Tx Speed Negotiation Window in Physical Layer in which the  Train-Tx comprises of exchange of Transmitter Training Pattern between two SAS devices.

Per the Working Draft SAS Protocol Layer - 4 (SPL-4), T10/BSR INCITS 538 Revision 08a, Section - 5.11.4.2.3.4.1 Phy's transmitter initial condition,  the transmitter training pattern comprises of the following

[cid:image001.png at 01D219A9.4CFF68A0]


Can you please provide clarification for the following scenario:

1.       Receiver does not detect 1st pattern marker. So the 1st TTIU will be missed along with 1st set of 58 scrambled dwords.

2.       Receiver detects 2nd pattern marker and also detects 2nd TTIU pattern.

We see two possible understandings:

Understanding 1 : Receiver tries to detect 58 dwords and tries to descramble them.
Here it will not be able to detect those 58 scrambled dwords because receiver will try to descramble data dwords using
initial seed(As it has missed initial Transmitter Training Pattern) whereas transmitter would send scramble dwords using
a new seed. Based on this understanding receiver will not be able to detect any of the of Transmitter Training Pattern.
And Train Tx will be unsuccessful.

Understanding 2 : Receiver shall ignore the detection of 58 scrambled dwords.
Here if the receiver ignore those 58 dwords then there is no question of descrambling
Dwords. Hence Train-Tx SNW will be successfully detected.

The specification also indicates the following :
The scrambler is the same as that defined for the link layer (see 6.8) and shall be initialized at the end of
RateChangeDelayTime. The scrambler shall not be reinitialized for the remainder of the Train_Tx-SNW. (Reference -Working Draft SAS Protocol Layer - 4 (SPL-4),
T10/BSR INCITS 538 Revision 08a, Section - 5.11.4.2.3.4.1 Phy's transmitter initial condition)

RateChangeDelayTime(RCDT) is the time for which transmitter drives idle at the beginning of Train-Tx SNW before
Transmitting any data.

Can you please provide us clarification as to whether or not we should be following scenario two or otherwise.

Kind Regards
Lana




[http://www.cadence.com/mail/footer_logocdns2.jpg]



Lana May Chan   |   Senior Product Marketing Manager, Verification IP

P: 4084285749   M: 4086236108   www.cadence.com<http://www.cadence.com>








-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.t10.org/pipermail/t10/attachments/20170117/2801cc13/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 73791 bytes
Desc: image001.png
URL: <http://www.t10.org/pipermail/t10/attachments/20170117/2801cc13/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.gif
Type: image/gif
Size: 64 bytes
Desc: image002.gif
URL: <http://www.t10.org/pipermail/t10/attachments/20170117/2801cc13/attachment-0001.gif>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image003.jpg
Type: image/jpeg
Size: 2688 bytes
Desc: image003.jpg
URL: <http://www.t10.org/pipermail/t10/attachments/20170117/2801cc13/attachment-0001.jpg>


More information about the T10 mailing list