Re: [PATCH v2 3/5] ath10k: drain tx before restarting hw

[Michal Kazior <michal.kazior@tieto.com>]

On 23 May 2014 12:31, Kalle Valo <kvalo@qca.qualcomm.com> wrote:
> Michal Kazior <michal.kazior@tieto.com> writes:
>
>>> Sure, but I still think it's a bit ugly. The right way to fix this would
>>> be to add it to include/lockdep.h, instead of adding custom checks to a
>>> driver.
>>
>> Good point. I wonder if it's generic enough to justify.
>
> No idea. But no matter what it will take some time to get it accepted,
> so we need to solve this in a faster way so that I can apply this patch.
>
>>> But does this check even make sense? There's nothing preventing
>>> to another thread to take lock just after the WARN_ON(), right?
>>
>> There's nothing wrong with other thread holding it. Actually that's
>> the reason for this very check.
>>
>> The point is to prevent ath10k_drain_tx() being called while caller
>> (current thread) holds conf_mutex. If it were to hold conf_mutex then
>> cancel_work_sync() can deadlock as both workers it tries to stop try
>> to get a hold of the lock too.
>
> Ah, now I understand.
>
> What if we just drop the WARN_ON() from this patch just so that I can
> apply the patch and we add a proper code for checking the mutex in a
> followup patch? Are you ok with that?

I'm okay with this as long as we at least have a comment stating that
conf_mutex must not be held while calling the function to avoid
deadlocks.


Michał