[openib-general] more CM byte ordering.

[Libor Michalek]

Sean,

  I found a couple more spots in the active connect case of incorrect
byte ordering, this time around starting psn. Here is a tested patch.

-Libor


Index: infiniband/core/cm.c
===================================================================
--- infiniband/core/cm.c	(revision 1725)
+++ infiniband/core/cm.c	(working copy)
@@ -711,7 +711,7 @@
 				       param->remote_cm_response_timeout);
 	cm_req_set_qp_type(req_msg, param->qp->qp_type);
 	cm_req_set_flow_ctrl(req_msg, param->flow_control);
-	cm_req_set_starting_psn(req_msg, cpu_to_be32(param->starting_psn));
+	cm_req_set_starting_psn(req_msg, param->starting_psn);
 	cm_req_set_local_resp_timeout(req_msg,
 				      param->local_cm_response_timeout);
 	cm_req_set_retry_count(req_msg, param->retry_count);
Index: infiniband/core/cm_msgs.h
===================================================================
--- infiniband/core/cm_msgs.h	(revision 1725)
+++ infiniband/core/cm_msgs.h	(working copy)
@@ -529,7 +529,7 @@
 
 static inline u32 cm_rep_get_starting_psn(struct cm_rep_msg *rep_msg)
 {
-	return cpu_to_be32(be32_to_cpu(rep_msg->offset20) >> 8);
+	return be32_to_cpu(rep_msg->offset20) >> 8;
 }
 
 static inline void cm_rep_set_starting_psn(struct cm_rep_msg *rep_msg,