[openib-general] [PATCH] Initial CM implementation

[Roland Dreier]

Just a quick read through, more comments later:

	int	local_cm_response_timeout:5;
	int	flow_control:1;

These should be "unsigned," not "int."  A signed 1-bit int doesn't
make much sense, and I think you'll probably run into sign trouble if
someone passes a local_cm_response_timeout of 20 or something for the
5-bit field.

In cm_send_handler(), you have:

	case IB_WC_RESP_TIMEOUT_ERR:
		cm_process_send_timeout(msg);
		break;

but can this ever happen?  I thought that the MAD layer always treated
CM MADs as unsolicited which means that responses are not matched with
requests so timeouts don't happen.  Or am I misunderstanding the MAD
layer semantics?

I see a lot of setting of state to TIMEWAIT, but I don't see where the
TIMEWAIT timeout happens.

TIMEWAIT handling should probably be done using the same cm.wq
workqueue as receive handling goes through -- this eliminates the
problem of connection setup starving timewait reaping that I mentioned
earlier.

 - R.