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.

WL12xx Linux association problem

Other Parts Discussed in Thread: WL1271

We have problem association to any access point when using the open source wl12xx driver.

Our system setup is:

Beagleboard XM
Module: WL1273 HDK
Interface: SPI@12MHz

Linux kernel 3.0.17
WL12xx driver: compat-wireless release 2012-02-21

Firmware:
wl1271-nvs.bin (MD5: 170b227c9cc5f3020ae58ce25eb93f21)
wl127x-fw-3.bin (MD5: f0fdc227f7f90f016c4fbb4fde6c6fae)

Driver module loads without problem.
It is possible to bring up and down interface with ifconfig.
We can perform scan on both 2.4 and 5GHz band, but fail in association to an access point. See attached log.

8424.insmod_scan_assoc_debug.txt

Is is possible to get any input if this is a driver problem, a firmware problem or an interface (SPI) problem?
We currently suspect problem with SPI, but SPI is stable during firmware loading and scan (also at higher speed 24MHz). What about the omap2-mcspi driver?

  • Hi Magnus,

    From the log i see that:

    [  418.595275] wl12xx: WL1271_ACX_INTR_DATA
    [  418.606842] wl12xx: ERROR command complete timeout

    That error happens when the Driver (run on Host processor) send command to the firmware (run on WL127x processor) however the Driver does not get confirmation from the firmware that it had received the command, that could happen in 3 cases first is that the firmware got stuck, second is that the interface (in your Application is the SPI) stopped working or that the driver did not receive the interrupt from the firmware.

    Before digging in can you please try to work without supplicant? 
    That is after loading the WLAN driver don’t call “wpa_supplicant -i wlan0 -c /mnt/net/ph/open.conf”  Instead please try to connect to non secured AP with IW commands.
     
    Please run the following:
    ifconfig wlan0 up
    iw wlan0 scan
    iw wlan0 connect AP_SSID (AP_SSID is the SSID of the AP you would like to connect to)
    iw wlan0 link
     
    and please share the logs tht you get.
    Regard,
    Eyal
  • Hi Eyal,

    Thank you for your reply. Please see attached log as requested.

    5127.ti1.txt

    Br,

     Magnus

  • Hi Maguns,

    So the Scan operation succeed however the issue happens in the connection stage, however the issue is not related to the WLAN driver, it is somehow related to the Kernel behavioral and maybe the SPI driver.

    Once the issue happens you can see the message:  BUG: scheduling while atomic:...

    I tried Google it and saw the following:

    Scheduling while atomic happen when the scheduler get confused and therefore unable to work properly and this because the scheduler tried to perform a "schedule()" in a section that contains a schedulable code inside of a non schedulable one . for example using sleeps inside of a section protected by a spinlock. trying to use another lock(semaphores, mutexes..) inside of a spinlock-proteced code may also disturb the scheduler. In addition using spinlocks in user space can drive the scheduler to behave as such.

    Hope it helps, also  I will try to verify a bit more.

    Regards,

    Eyal

  • Hi Eyal,

    Thanks for your reply.

    Our current approach will be to switch host platform. We will post again if we see the same problem on the new host.

    Br,
    Magnus

  •  Hi Magnus,

    Thanks for the reply, i will close the that query, please open it again in case there are issues in the future

    Regards,

    Eyal