[openib-general] Re: CM REP-REJ bug

[Sean Hefty]

Libor Michalek wrote:

>>My thinking on this was that the for the sender of the REQ, the remote 
>>ID would be set to 0.  The sender of the REJ would set the local ID to 
>>0, since it wouldn't have one (unless it had previously sent an MRA). 
>>Does this make sense, or am I missing something else here?
> 
>   How does the user set the local ID to zero, just reach into the
> connection identifier and set local_id? This will not cause any
> lookup or other problems? Seems unusual to start reaching into the
> cm_id and start changing values. How about if ib_send_cm_rej()
> made this change based on the CM state in which it was called?

The local/remote IDs shouldn't be modified by the user.

I see the issue now.  The problem is that the sender of the REJ is 
setting the local comm ID to a non-zero value.  It needs to be set to 
zero if rejecting a REQ.  Thanks for finding this; I'll create a patch 
for it.

- Sean