Clarification from SAS 1.1 Standard (T10/1601-D)

Elliott, Robert (Server Storage) Elliott at hp.com
Wed Oct 19 14:24:17 PDT 2005


* From the T10 Reflector (t10 at t10.org), posted by:
* "Elliott, Robert (Server Storage)" <Elliott at hp.com>
*
> * From the T10 Reflector (t10 at t10.org), posted by:
> * David Bergman <dbergman at dalsemi.com>
> *
> I have 3 spec clarification questions:
> 
> 1) Section 4.4.2.3.  I read this section in 2 possible ways.  Could I 
> please get an answer to which reading is correct?
> 
>     * Reading 1 - The expander will perform a hard reset on all of the
>       other phys within the port, but not on any of the phys 
> external to
>       the port.  I think this is the intended reading.
>     * Reading 2 - The expander will not perform a hard reset on any
>       other phys, even in the same port.

You have to carefully differentiate between these terms in this section:
* link reset sequence: OOB + speed negotiation + IDENTIFY
* hard reset sequence: OOB + speed negotiation + HARD_RESET + OOB +
speed 

An expander device never generates a _hard_ reset sequence on any of its
phys because of an incoming hard reset sequence.  It does, however,
generate a _link_ reset sequence for all phys contained in the same port
as the phy that received the hard reset sequence (see the rule in
4.4.2.1, which applies all ports).

> 2) Section 7.9.5.3.2.2 refers to a Tx Identify Address Frame request 
> being a gating item to this state transition.  I am unclear 
> where this request originates, and under what circumstances.  Table
K.5 
> seems to indicate it comes from the Management Application layer, but
the 
> conditions for this request are unclear to me.  Could I please get an 
> explanation of the conditions for this Tx Identify Address 
> Frame request?

That represents the ability for a device to let its software/firmware
control when to send the IDENTIFY address frame, implying there could be
a delay before it chooses to do so.

> 3) I have found no change bars in revision 10 of this document.  Does 
> that mean it is identical to revision 9E?

Revision 10 was the one forwarded, so it has no revision history or
change bars.  It's very close to 9e; you can sort the comments by date
in 05-169r6 and see what changed since July 24th.

> Thanks for the clarifications.
> 
> Dave Bergman
> Dallas Semiconductor
> dbergman at dalsemi.com
> (972) 371-3112

--
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