Wifi crash at startup

[pottendo]

hi,
I've moved my application to the recent step 50.0.1.
Now Wifi won't startup and crashes for both 32 and 64 bit builds on Pi3's (A+ and B+).
I can reproduce with a fresh circle checkout and trying to run the demo: addon/wlan/sample/hello_wlan
I did copy the recent drivers into firmware.
Here's the log (32bit build):

logger: Circle 50.0.1 started on Raspberry Pi 3 Model B+ 1GB
logger: Revision code is a020d3, compiler has been GCC 14.3.1
00:00:00.28 timer: SpeedFactor is 3.57
00:00:01.14 usbdev0-1: Device ven424-2514, dev9-0-2 found (HS)
00:00:01.19 usbdev0-1: Interface int9-0-1 found
00:00:01.19 usbdev0-1: Function is not supported
00:00:01.20 usbdev0-1: Interface int9-0-2 found
00:00:01.20 usbdev0-1: Using device/interface int9-0-2
00:00:01.87 usbdev0-1: Device ven424-2514, dev9-0-2 found (HS)
00:00:01.92 usbdev0-1: Interface int9-0-1 found
00:00:01.93 usbdev0-1: Function is not supported
00:00:01.93 usbdev0-1: Interface int9-0-2 found
00:00:01.94 usbdev0-1: Using device/interface int9-0-2
00:00:02.61 usbdev0-1: Device ven424-7800 found (HS)
00:00:02.66 usbdev0-1: Using device/interface ven424-7800
00:00:02.74 lan7800: MAC address is B8:27:EB:6D:54:74
00:00:02.89 usbhub: Port 1: Device configured
00:00:02.89 usbhub: Port 1: Device configured
00:00:02.90 dwroot: Device configured
00:00:02.92 sdhost: emmc1: sdhost-bcm2835 loaded
00:00:04.43 emmc: Capacity is 30436 MBytes
00:00:04.44 emmc: Found a valid version 3.0x SD card
00:00:04.58 wlan: ether4330: chip 0x4345 rev 6 type 1
00:00:05.18 wlan: ether4330: firmware ready
00:00:05.18 wlan: ether4330: addr B8:27:EB:38:01:21
00:00:05.21 kernel: Compile time: Nov 15 2025 15:23:02
00:00:05.27 wpa: Setting country code to 'AT'
00:00:08.87 wpa: Trying to associate with 04:b4:fe:a6:7d:3c (SSID='pottendo' freq=5290 MHz)
00:00:11.70 wpa: Associated with 04:b4:fe:a6:7d:3c
00:00:12.40 wpa: WPA: Key negotiation completed with 04:b4:fe:a6:7d:3c [PTK=CCMP GTK=CCMP]
00:00:12.40 wpa: CTRL-EVENT-CONNECTED - Connection to 04:b4:fe:a6:7d:3c completed [id=0 id_str=]
00:00:12.49 wpa: WPA: Key negotiation completed with 04:b4:fe:a6:7d:3c [PTK=CCMP GTK=CCMP]
00:00:13.50 wpa: WPA: Key negotiation completed with 04:b4:fe:a6:7d:3c [PTK=CCMP GTK=CCMP]
00:00:14.49 wpa: WPA: Key negotiation completed with 04:b4:fe:a6:7d:3c [PTK=CCMP GTK=CCMP]
00:00:15.49 wpa: CTRL-EVENT-DISCONNECTED bssid=04:b4:fe:a6:7d:3c reason=0
00:00:15.50 wpa: Added BSSID 04:b4:fe:a6:7d:3c into ignore list, ignoring for 10 seconds
00:00:15.51 wpa: BSSID 04:b4:fe:a6:7d:3c ignore list count incremented to 2, ignoring for 10 seconds
00:00:15.61 ../src/drivers/driver_circle.cpp(344): stack[7] is 0x48128
00:00:15.62 ../src/drivers/driver_circle.cpp(344): stack[13] is 0x40D84
00:00:15.63 ../src/drivers/driver_circle.cpp(344): stack[22] is 0x20C58
00:00:15.63 ../src/drivers/driver_circle.cpp(344): stack[23] is 0xC828
00:00:15.64 ../src/drivers/driver_circle.cpp(344): stack[27] is 0x8B0DC
00:00:15.64 ../src/drivers/driver_circle.cpp(344): stack[32] is 0x30D40
00:00:15.65 ../src/drivers/driver_circle.cpp(344): stack[35] is 0x20C58
00:00:15.66 ../src/drivers/driver_circle.cpp(344): stack[37] is 0x8C40
00:00:15.66 ../src/drivers/driver_circle.cpp(344): stack[47] is 0xED80
00:00:15.67 ../src/drivers/driver_circle.cpp(344): stack[51] is 0x9508
00:00:15.68 ../src/drivers/driver_circle.cpp(344): assertion failed: params->freqs == 0

my firmware directory:

 399344 Nov 14 20:03 brcmfmac43430-sdio.bin
   4733 Nov 14 20:03 brcmfmac43430-sdio.clm_blob
   1121 Nov 14 20:03 brcmfmac43430-sdio.txt
 416101 Nov 14 20:03 brcmfmac43436-sdio.bin
  11209 Nov 14 20:03 brcmfmac43436-sdio.clm_blob
   1706 Nov 14 20:03 brcmfmac43436-sdio.txt
 442211 Nov 14 20:03 brcmfmac43436s-sdio.bin
   1185 Nov 14 20:03 brcmfmac43436s-sdio.txt
 548728 Nov 14 20:03 brcmfmac43455-sdio.bin
   2676 Nov 14 20:03 brcmfmac43455-sdio.clm_blob
 548728 Nov 14 20:03 brcmfmac43455-sdio.raspberrypi,5-model-b.bin
   2676 Nov 14 20:03 brcmfmac43455-sdio.raspberrypi,5-model-b.clm_blob
   2074 Nov 14 20:03 brcmfmac43455-sdio.raspberrypi,5-model-b.txt
   2074 Nov 14 20:03 brcmfmac43455-sdio.txt
 495898 Nov 14 20:03 brcmfmac43456-sdio.bin
   7163 Nov 14 20:03 brcmfmac43456-sdio.clm_blob
   2053 Nov 14 20:03 brcmfmac43456-sdio.txt

If needed, let me know if I shall share config.txt, cmdline.txt.
I do have a wpa_supplicant.conf file in my root directory.

This setup used to work with previous steps of circle.

It seems very basic, so I'm quite sure I've overlooked something trivial... Sorry in advance for that!
Ethernet network works.

thanx, pottendo

[rsta2]

This is a duplicate, please see issue #616. The failed assertion has been removed on the develop branch (in the hostap submodule). It's not necessary. It's currently not clear, why the connection is disconnected immediately again. It could help to get more messages from wpa_supplicant. Please see this for info, how the debug level can be changed: #616 (comment)

[pottendo]

hi,
...and I thought that I've searched the discussions and issues; obviously not - Sorry.

Anyway, I produced the debug log, but I'm not fluent in this area...
...does it make sense to publish it here? Is there anything inside the log, that I don't even WANT to share, as it contains secrets (obvious passwords are not there, but maybe something different sensitive...)?

As one suspicion is 'bonded WiFis'... Not sure what it is exactly, but I do have a 'Mesh' network comprising 2.4 and 5Ghz WiFis and all that with a WiFi expander (Fritz-Repeater) connecting to my main FritzBox (7490) - all accessible with one SSID.

Removing the assertion helped and it's working now.

I propose to continue on the other thread and you may close this one here.

Many thanks for the rapid response.
bye, pottendo

[rsta2]

You are welcome. There were multiple problems discussed in this issue, so your problem was probably not that easy to find. Good point. If you don't want to post the log here, you could send it to my email address. You will find it on my personal GitHub page, when you are logged in. I assume Mesh networks are currently not supported. But maybe it works with a single AP inside the Mesh. It seems so, otherwise it would not work now. OK, we should continue this discussion in issue #616.

[rsta2]

Continued in issue #616.