OOB signal detect

Elliott, Robert (Server Storage) elliott at hp.com
Thu May 20 16:05:02 PDT 2004


* From the T10 Reflector (t10 at t10.org), posted by:
* "Elliott, Robert (Server Storage)" <elliott at hp.com>
*
> Amit,
> After going through the standard this is what I understood. 
> 
> The OOB signals start with an IDLE followed by ALIGN bursts. 
> At least in fig 58 of SAS 1.1 standard there seems to be an 
> assumption of "any transition" being any signal other than 
> an IDLE. As an idle burst is followed by an ALIGN burst it 
> can be counted as the 1st ALIGN burst. Hence after 4th ALIGN
> the COMINIT should be detected.  Also as a COMINIT OOB signal 
> is more than one instance of 6 IDLEs + 6 ALIGNS. There is a 
> great chance that an IDLE after COMINIT negation period could 
> easily be detected hence the COMINIT being detected after the 
> 4th ALIGN in that instance.
> I guess it doesn't matter if a COMINIT is detected after 4th 
> ALIGN or 5th ALIGN. I think it can be detected at the 6th 
> ALIGN as long as the receiver has detected 4 successive 
> IDLE+ALIGN bursts.
> 
> Guru

Correct.  Although the transmitter is only supposed to send 6
of them, the receiver is required to accept any number 4 and
higher.  If you get 3000 of them in a row, you still consider
it the same OOB signal and keep waiting for an IDLE period that
doesn't have the same time.

The SP state machine differentiates between COMnnn Detected
and COMnnn Completed whenever it matters.

--
Rob Elliott, elliott at hp.com
Hewlett-Packard Industry Standard Server Storage Advanced Technology
https://ecardfile.com/id/RobElliott


*
* 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