[openib-general] mapping between IP address and device name

[David M. Brean]

The Sun implementation uses a Service ID from the "Local OS
Administrative" range when installing an entry in the SA.  So, the first
byte should be 0x2 in the ServiceID component in the SA record.  The
Service Name in the SA record contained the IP address with the NFS port
number appended at the end.  So, this component of the SA record would
appear as <IPaddress>:2049 where IPaddress is provided by the NFS
server.  So, the well-known TCP port number for NFS is used in the
Service Name.  [Note, the information that I have is from a discussion
with the Sun NFS team a year ago.  I haven't used an analyzer to examine
the data crossing the IB links.]

I think that a well-known Service ID has more value when using the CM
private-data mechanism.

[I have some comments about ATS vs CM private-data mechanisms that I'll
send in seperate mail.]

-David

Jay Rosser wrote:

>
>>> On the subject of NFS/RDMA, what is the IB ServiceID space that is 
>>> used? If I recall correctly, I have seen simply the value 2049 (i.e. 
>>> the standard TCP/UDP port number) used in some implementations (i.e. 
>>> 00 00 00 00 00 00 20 49). Is there a mapping onto an IB ServiceID 
>>> defined?
>>>   
>>
>>
>> We aren't currently using the portmapper to discover the serviceid that
>> the NFS/RDMA server is listening on. Brent Callaghan chose serviceid 
>> 2049
>> as a convenience in Sun's first implementation, and so far it has stuck.
>>
>> Theoretically the server can listen on any endpoint it chooses, this is
>> how NFS/TCP and NFS/UDP work. But typically all servers use the well
>> known port. It's probably a good idea to define a better default 
>> mapping.
>>  
>>
> Using a well-known port seems okay to me.
>
> However, if I understand correctly, the 2049 ServiceId is in the 
> IBTA's ServiceID space (i.e. for IBTA-defined well-known ServiceIDs - 
> see IB1.2 Vol1 pg 1181 A3.2.3.1). It would probably make sense to use 
> the IETF ServiceID space (A3.2.3.2) but that would require someone to 
> administer it.
>
> Thanks,
>
> Jay
>
>> Tom.
>>
>>  
>>
>
>