[ofa-general] OFED HA related question

[Tang, Changqing]

Roland:

	Suppose I get IBV_EVENT_DEVICE_FATAL async event from the first
HCA on my node,
can I continue to call ibv_poll_cq() to get back all the work-requests I
posted before ?
or do I need to keep track these work-requests? I am afraid
ibv_poll_cq() will return error by itself. Also can I call
ibv_dereg_mr() to free the memory I registered to this HCA ?

	If I continue to use the second HCA, does the failure of first
HCA affect the operation
of second HCA (from driver point of view) ?

	Thanks 

--CQ