Re: [PATCH] brcmfmac: stop watchdog before detach and free everything

[Andy Shevchenko]

On Mon, May 28, 2018 at 12:54 PM, Michael Nazzareno Trimarchi
<michael@xxxxxxxxxxxxxxxxxxxx> wrote:
> Hi Arend
>
> On Mon, May 28, 2018 at 11:51 AM, Arend van Spriel
> <arend.vanspriel@xxxxxxxxxxxx> wrote:
>> On 5/28/2018 9:50 AM, Michael Trimarchi wrote:
>>>
>>> Watchdog need to be stopped in brcmf_sdio_remove to avoid
>>> i
>>> The system is going down NOW!
>>> [ 1348.110759] Unable to handle kernel NULL pointer dereference at virtual
>>> address 000002f8
>>> Sent SIGTERM to all processes
>>> [ 1348.121412] Mem abort info:
>>> [ 1348.126962]   ESR = 0x96000004
>>> [ 1348.130023]   Exception class = DABT (current EL), IL = 32 bits
>>> [ 1348.135948]   SET = 0, FnV = 0
>>> [ 1348.138997]   EA = 0, S1PTW = 0
>>> [ 1348.142154] Data abort info:
>>> [ 1348.145045]   ISV = 0, ISS = 0x00000004
>>> [ 1348.148884]   CM = 0, WnR = 0
>>> [ 1348.151861] user pgtable: 4k pages, 48-bit VAs, pgdp = (____ptrval____)
>>> [ 1348.158475] [00000000000002f8] pgd=0000000000000000
>>> [ 1348.163364] Internal error: Oops: 96000004 [#1] PREEMPT SMP
>>> [ 1348.168927] Modules linked in: ipv6
>>> [ 1348.172421] CPU: 3 PID: 1421 Comm: brcmf_wdog/mmc0 Not tainted
>>> 4.17.0-rc5-next-20180517 #18
>>> [ 1348.180757] Hardware name: Amarula A64-Relic (DT)
>>> [ 1348.185455] pstate: 60000005 (nZCv daif -PAN -UAO)
>>> [ 1348.190251] pc : brcmf_sdiod_freezer_count+0x0/0x20
>>> [ 1348.195124] lr : brcmf_sdio_watchdog_thread+0x64/0x290
>>
>>
>> Hi Michael,
>>
>> Thanks for the patch. In normal scenario the callstack looks like this:
>>
>> brcmf_sdio_remove()
>>         -> brcmf_detach()
>>                 -> brcmf_bus_stop()
>>                         -> brcmf_sdio_bus_stop()
>>
>> In brcmf_sdio_bus_stop() the watchdog is terminated. So in what scenario did
>> you encounter this null pointer deref?
>
> Is this happen even when there is not wifi firmware?
> boot without any firmware in the filesystem and then trigger a reboot

Something like the above I had noticed for a long (couple of kernel
releases?) time, but wasn't a big priority to me.
Though, I can test this on my side.

P.S. I think rmmod or echo > unbind will trigger that as well.

-- 
With Best Regards,
Andy Shevchenko