SAS Expander Routing Resources
gil.romo at qlogic.com
Wed Dec 10 13:59:48 PST 2003
* From the T10 Reflector (t10 at t10.org), posted by:
* Gil Romo <gil.romo at qlogic.com>
Indeed, I believe the standard should define how the expander indicates it has
no more routing resources. I agree with your approach on the expander returning
AIP (WAITING ON CONNECTION).
Here are my suggested changes:
4.6.6, Table 11 - "ECM to expander phy confirmations", Arbitrating (Waiting On
(a AND b AND c AND d)
Either 1) (a AND b AND c AND d); OR 2) (a AND b AND "there are no more resources
to route the connection request"). Not sure yet on exact wording.
7.2.5, Table 57, AIP (WAITING ON CONNECTION)
"... but the destination phys are all being used for connections.",
"but the destination phys are all being used for connections or there are no
more expander resources to route the request."
220.127.116.11, "The ECM shall generate the Arb Won confirmation when ..."
"... and c) there are sufficient resources to route the connection request."
> * 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:
> > * Gil Romo <gil.romo at qlogic.com>
> > *
> > In 4.6.4, the standard states that "The ECR routes messages
> > ... Enough routing
> > resources shall be provided to support at least one connection."
> > What should be the expander's response when it is has no more
> > routing resources
> > (ie, the expander finds a matching destination port that
> > supports the requested
> > connection rate and none of those phys are returning Phy
> > Status (Connection))?
> > I don't believe the standard defines the expected behavior.
> > Thanks,
> > Gil
> > ------------
> > Gilbert Romo
> > Circuits & Integration
> > QLogic Corporation, Aliso Viejo, California
> > Office: 949-389-6266
> > E-mail: gil.romo at qlogic.com
> One approach is to assume the ECM knows and does not offer any
> pathway resolution (no Arb Won, Arb Lost, or Arb Reject) in
> that situation. The request remains in limbo until it can
> be serviced.
> Another is to let the ECM treat it as a Waiting On Connection
> situation (the fact that it is waiting on a connection between
> two unrelated phys, not a connection occupying a destination
> phy, doesn't matter much).
> If you think the standard needs to describe one or both of
> these (or some other approach), please suggest the text and
> where it should go.
> Rob Elliott, elliott at hp.com
> Hewlett-Packard Industry Standard Server Storage Advanced Technology
> * For T10 Reflector information, send a message with
> * 'info t10' (no quotes) in the message body to majordomo at t10.org
Circuits & Integration
QLogic Corporation, Aliso Viejo, California
E-mail: gil.romo at qlogic.com
* 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