You are not logged in.
Hi,
I'm using my 802.11 extension with 4 antennas (I added a buffer to store raw data for my proccessing, but it just passively record the raw data from antennas without modifying and Rx flow in the reference design). Usually it run smoothly, but today I got this error:
Stuck waiting for MAC_HW_LASTBYTE_ADDR1: wlan_mac_get_last_byte_index() = 0
MAC HW Status: 0x000C04A2
Rx Hdr Params: 0x0100013D
Rx PHY Status: 0x00000000
I guess you had thought about the problem here when you wrote the code. So can you please suggest me what is the root cause of it?
Thank you so much
Offline
That print occurs when the DCF is polling the Rx MAC waiting for the first 14 bytes of the received packet, but those 14 bytes are never received. This should never happen - the DCF only beings polling the Rx MAC after an RX_START event, which (if things are working normally) is always followed by at least 14 payload bytes.
We added that print a long time ago when debugging bad states between the Rx PHY and MAC core. We have not seen that print occur in our testing of recent releases. Can you clarify - what version of the 802.11 Ref Design are you using? And what specific changes have you made to the hardware and software designs?
Offline
I'm using this version of reference design "Mango_802.11_RefDes_v1.7.7" with this phy_rx "wlan_phy_rx_pmd_axiw_v4_01_x". I have added a buffer to record raw data from 4 antennas for Arrival of Angle estimation as illustrated in the below image
this image
The buffer takes the RFA/B/C/D and RSSI as well as valid signal as input only. But it does activate the AGC when the RSSI is higher than a threshold. In the software, I also enable the use of RSSI pkt detection (by setting the value of RSSI threshold to 200). I've used it for several days intermittently (each session is around 2 to 4 hours), but it was the first time I saw this error. Is there any case that the AGC causes that error?
Last edited by vutran (2018-Sep-03 13:26:25)
Offline
Nothing immediately jumps out to me as to what could be causing this. It would be worth trying to add some additional debugging prints when the error occurs to see if, for example, the error coincides with any particular actions. For example, it would be useful to know if the error only ever occurs during a reception where the RSSI packet detection fires. It's been a long time since we've enabled RSSI packet detection in the design.
At the very least, I think you can recover from the error once it happens by adding the following just after those error prints and then returning from the function:
wlan_mac_reset_rx_started(1); wlan_mac_reset_rx_started(0);
It doesn't avoid the problem, but it should reset the MAC's RX_STARTED latch and let you get back into a working state automatically rather than hanging forever.
Offline
Actually, I've tried to recreate the error (with RSSI detection enabled/disabled), but even I left my WARP overnight, transmitting millions of packets, I couldn't see the error again.
Thanks for your help.
Offline
Thanks for the update. Incredibly rare bugs like this are frustrating to track down. We'll keep pondering what it might be.
Offline