TX2 WiFi issues: dhd_prot_ioctl : bus is down. we have nothing to do

Autonomous Machines Jetson & Embedded Systems Jetson TX2
1

We are seeing a recurring issue with WiFi on the TX2. It’ll work just fine, until the following shows up in dmesg:

[44612.683942] dhd_bus_txctl: ctrl_frame_stat == TRUE txcnt_timeout=1
[44612.690148] dhdcdc_query_ioctl: dhdcdc_msg failed w/status -5
[44612.690159] dhd_dpc_thread is consuming too much time Looped 52 times for 1000 iterations in 100ms timeout
[44612.705553]  wl_cfg80211_get_station : 
[44612.709210] Could not get rssi (-5)
[44612.712705]  wl_cfg80211_get_station : 
[44612.716366] force cfg80211_disconnected: -5
[44617.803938] dhd_bus_txctl: ctrl_frame_stat == TRUE txcnt_timeout=2
[44617.810139] dhdcdc_set_ioctl: dhdcdc_msg failed w/status -110
[44617.810147] dhd_dpc_thread is consuming too much time Looped 50 times for 1000 iterations in 100ms timeout
[44617.825544] dhd_check_hang: Event HANG send up due to  re=0 te=2 e=-110 s=2
[44617.832516] dhd_check_hang: Event HANG send up due to  re=0 te=2 e=-110 s=2
[44617.839492]  wl_notifier_change_state : 
[44617.843242] wlan0:error(-110)
[44617.846225] dhd_prot_ioctl : bus is down. we have nothing to do
[44617.852202] dhd_prot_ioctl : bus is down. we have nothing to do
[44617.858139]  wl_do_escan : 
[44617.860758] error (-1)
[44617.863126]  wl_cfg80211_scan : 
[44617.866178] scan error (-1)
[44617.869215] [12-09 22:44:05.411] wl_cfg80211_disconnect: Reason 3
[44618.131926]  wl_cfg80211_disconnect : 
[44618.135943] Link down event is not received
[44618.162937] dhd_prot_ioctl : bus is down. we have nothing to do
[44618.168879] CFGP2P-ERROR) wl_cfgp2p_bss_isup : 
[44618.173240] 'cfg bss -C 0' failed: -1
[44618.176918] CFGP2P-ERROR) wl_cfgp2p_bss_isup : 
[44618.181278] NOTE: this ioctl error is normal when the BSS has not been created yet.
[44618.188953] dhd_prot_ioctl : bus is down. we have nothing to do
[44618.194929]  wl_notifier_change_state : 
[44618.198685] wlan0:error(-1)
[44618.205472] dhd_prot_ioctl : bus is down. we have nothing to do
[44618.211448] dhd_prot_ioctl : bus is down. we have nothing to do
[44618.217387] CFGP2P-ERROR) wl_cfgp2p_set_management_ie : 
[44618.222529] vndr ie set error : -1
[44618.225964] dhd_prot_ioctl : bus is down. we have nothing to do
[44618.226932]  wl_is_linkdown : 
[44618.226932] Link down Reason : WLC_E_LINK
[44618.239011]  wl_dongle_down : 
[44618.241931] WLC_DOWN error (-1)
[44618.292096] wl_android_wifi_off in
[44618.295712] dhd_prot_ioctl : bus is down. we have nothing to do
[44618.311287] dhd_prot_ioctl : bus is down. we have nothing to do
[44618.317386] dhd_prot_ioctl : bus is down. we have nothing to do
[44618.330868] dhdsdio_isr : bus is down. we have nothing to do
[44618.336590] gpio tegra-gpio-aon wake69 for gpio=59(FF:3)
[44618.336591] Disabling wake69
[44618.337959] sdhci-tegra 3440000.sdhci: Tuning done, restoring the best tap value : 68
[44618.357938] wifi_platform_set_power = 0
[44618.564946]  wl_cfg80211_hang : 
[44618.568072] In : chip crash eventing

From what I’ve been able to trace through the driver, it looks like it times out sending a control message, then the dhd bus goes down and never comes back up.

We’re using the 4.9.140 kernel with the nvidia-provided RT patch.

Any ideas what could be wrong?
Thanks in advance!

3

May I know your Jetpack version?

In the same time pls try to uprade the network-manager and test again.
“sudo apt-get --only-upgrade install network-manager”

Also share the message: sudo nmcli dev wifi before your test.

4

We’re on L4T 32.4.3. We use our own rootfs though (with the firmware files from the sample rootfs v32.4.3 copied over), so we don’t have network-manager or nmcli available - we use wpa_supplicant directly instead.

I know that makes things harder for you to reproduce, but the error that I’m getting to me looks more like a kernel/driver/hardware issue than something due to what’s running in userspace?

Is there a way to restart the WiFi driver / hardware through sysfs or similar maybe?