[openfabrics-ewg] Changes in SM for the new SRP daemon

[Ishai Rabinovitz]

Of course I'm doing this check.
It remained the same as in the original algorithm, so I ignored it in my
mail.

Ishai 

-----Original Message-----
From: Roland Dreier [mailto:rdreier at cisco.com] 
Sent: Thursday, April 06, 2006 1:24 AM
To: Rimmer, Todd
Cc: Ishai Rabinovitz; openfabrics-ewg at openib.org
Subject: Re: [openfabrics-ewg] Changes in SM for the new SRP daemon

    Todd> Ishai, The algorithm you describe is incomplete.  I assume
    Todd> by:

    Ishai> From the response to this query the initiator can check if
    Ishai> this port is an SRP target. (dm bit capability is set)
 
    Todd> That you will stop at the SA query step and assume all ports
    Todd> supporting a DM are SRP targets.  There can of course be
    Todd> other nodes in a fabric which support a DM but do not
    Todd> support SRP.
 
    Todd> To be complete, DM queries must be made to the identified
    Todd> nodes to check their IOCs and Service entries to identify
    Todd> what protocols they support.  Only those which have
    Todd> available IOCs and have Service entries supporting SRP
    Todd> should be considered SRP targets.

Obviously he has to do the full IO unit and service table query.
Otherwise there would be no way to know what service to connect to when
trying to reach the SRP target.

So I don't think you have to worry about that.

 - R.