RapidIO II IP Core link-response Timeout Behavior Does Not Match Documentation

According to the RapidIO II MegaCore Function User Guide, after the RapidIO II IP core sends a link-request, if it times out waiting for a link-response, it sends another link-request, for a total of seven attempts. After seven timeouts, the output port error recovery state machine enters the fatal_error state. When the state machine enters the fatal_error state, the IP core sets the PORT_ERR field of the Port 0 Error and Status CSR to the value of 1.

However, the RapidIO II IP core does not implement this behavior. Instead, the IP core tries indefinitely, sending additional link-request packets forever, and the output port error recovery state machine does not enter the fatal_error state as a result of any of these timed out attempts. Instead, the IP core may stall indefinitely.

This issue has no workaround.

This issue is fixed in version 14.1 of the RapidIO II MegaCore function.