This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

NetworkProcessor: receiving "Command Complete" event?

Other Parts Discussed in Thread: CC2541

I am simulating a peripheral that when connected to, pump out notifications every 10-15 ms.  After some time (can be 1000 notifications, can be 3000) , there is no "Command Status" reply to my last GATT_Notification. During the last run, the last event that I received was a HCI_Command_Complete_Event (0x0E) instead of   HCI_Vendor_Specific_Event (0xFF).

1. Is that expected?  If I only send notification commands, should I receive anything other than HCI_Vendor_Specific_Event replies? I thought anything other than HCI_Vendor_Specific_Event is filtered before it reaches the application processor.

2. Is there any way to debug what is the cause of the lock-ups? It is really hindering my prototyping at this point as I cannot implement the longer running tests. I can only send data for up to two minutes, then I have to physically cut the power to the CC2541 . It is not responding to UTIL_Reset, soft nor hard.

  • I captured one of those events that I'm decoding as HCI_Command_Complete_Event:  "04 0e 04 01 00 00 01"

    My "normal traffic" looks like:

       TX: 01 9b fd 19 00 00 00 05 00 8c 19 00 00 cc 00 00 00 55 9b 07 00 00 00 00 00 00 00 00 00
       RX: 04 ff 06 7f 06 00 9b fd 00

    so the 04 0e 04 does not look like bit flips. At the same time, decoding it as an event results in a 4 bytes payload while the HCI_Command_Complete looks to need at least 6: Command complete needs 5 (2 for timeout, 1 for num hci buffers and 2 for opcode) plus 1 for Status.

    Can somebody explain what is going on here?

  • I'm getting the same result. How did you solve the problem?