[EMAIL PROTECTED] wrote:
> On Thu, 2006-03-16 at 10:45 -0500, James Lentini wrote:
>> 
>> On Thu, 16 Mar 2006, Steve Wise wrote:
>> 
>> swise> Just to clarify:  Tom is talking about iWARP devices here that
>> swise> support the native stack -and- the rdma stack with the same
>> swise> offload device.  For IB devices, the above "collision" doesn't
>> swise> matter because IB really doesn't use the ip addresses at all
>> swise> after the RC is setup.  AND, the native stack TCP support in
>> IB swise> (IPoIB) uses a different QP and service type (UD).
> So there's no problem with said collision.
>> 
>> If there is a fix for the iWARP collision problem, wouldn't there be
>> an analogous fix for IB? Why wouldn't we make that fix?
> 
> Perhaps.  To be honest, I have followed the SDP/IB side of
> this thread enough...  I just wanted to make a point that
> iwarp devices have a different issue.
> 

But it really isn't a different issue.

If you are representing to the application that you are
listening for TCP connections on IP address X port Y then
there really shouldn't be another listener allowed for X:Y.

Isn't that part of presenting yourself as SOCK_STREAM and
claiming to use an IP address?

_______________________________________________
openib-general mailing list
[email protected]
http://openib.org/mailman/listinfo/openib-general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general

Reply via email to