[openib-general] agent_mad_send

Tue Oct 26 15:29:22 PDT 2004

In agent_mad_send, a call is made to create an address handle.  Immediately after calling ib_post_send_mad, the address handle is destroyed.  I think that we want to wait until the send is completed before destroying the address handle, and require this of all callers of ib_post_send_mad.

Also, I don't think that we want to have this code access the port_priv structure, such as the send_list_lock (which ends up being acquired twice).  Queuing of MADs should be done by the ib_post_send_mad call, and not by its caller.  This will be needed to handle QP overflow anyway.  Agent_send_handler has a similar issue.

I will try to submit a patch for this in a couple of days, but I'm currently trying to separate the send_posted_list for the two QPs for better error and completion handling.

- Sean

-- 