How many SAS addresses should an expander has?

Elliott, Robert (Server Storage) Elliott at hp.com
Wed Oct 16 09:31:27 PDT 2002


* From the T10 Reflector (t10 at t10.org), posted by:
* "Elliott, Robert (Server Storage)" <Elliott at hp.com>
*

> -----Original Message-----
> From: Seto, Pak-lung [mailto:pak-lung.seto at intel.com] 
> Sent: Wednesday, October 16, 2002 9:37 AM
> To: 't10 at t10.org'
> Subject: How many SAS addresses should an expander has?
> 
> 
> * From the T10 Reflector (t10 at t10.org), posted by:
> * "Seto, Pak-lung" <pak-lung.seto at intel.com>
> *
> First, ignoring the STP/SATA attachment first.
> 
> For all links within an expander, during the Identify address 
> frame phase, according to the SAS address field definition on 
> page 127 v2.0a
> 
> "An SAS ADDRESS field indicates the SAS address of the device 
> transmitting the IDENTIFY address frame."

That wording is a bit dated.  To be more precise it should say:
a) indicates the SAS address of the port for end devices; or
b) indicates the SAS address of the expander device for expander
devices.


> From the above definition, it means if any number of links 
> from an expander are connected to an end devices, it will 
> automatically (wants to) become a group of wide links and 
> there is no choice of keeping it as individual separate 
> links?  

Correct.

>        Also, what happen if the SAS Addresses received from 
> the end device(s) are not the same for all those links?

Each set of matching input addresses causes a separate expander port to
be created.

> In addition, in current SAS spec. individual links are 
> performing link initializaiton/speed negotiation 
> independently.  And it is not guaranteed that all links 
> within a wide links group to be negotiatied to have the same 
> negotiated link rate.  It may have problem if this wide links 
> group happens to have the lowest negotiated link rate on the 
> pathway between a particually initiator/target pair.  I think 
> the standard should either require all links within a wide 
> links group to negotiated to have the same link rate (I 
> believe it is a little bit to restrictive), or if any phy of 
> a wide links group when it is reporting it's link rate during 
> discovery to a initiator, it shall report the lowest 
> negotiated link rate of the wide links group that the phy is 
> belonging to (since the expander can select any link within a 
> wide links group to make a connection), because the initiator 
> has no concept of multiple minimum link rates for a pathway 
> connecting to a particular target.

The expander never reports a link rate for the port; it reports
everything on a phy by phy basis.  If a wide link consists of 
physical links running at different rates, the initiator will know.

Assume there is a wide link with both 1.5 Gbps and 3.0 Gbps physical
links.  If an initiator sends an OPEN (3.0) that needs to use that
wide link, that connection can only use a 3.0 Gbps physical link 
of that wide link.  If the initiator sends an OPEN (1.5), the 
connection may use a 1.5 Gbps link or a 3.0 Gbps link (with rate
matching by the expanders).

--
Rob Elliott, elliott at hp.com
Industry Standard Server Storage Advanced Technology
Hewlett-Packard

*
* 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