Re: [PATCH] cpufreq, store_scaling_governor requires policy->rwsem to be held for duration of changing governors [v2]

[Prarit Bhargava <prarit@redhat.com>]

On 08/04/2014 06:36 AM, Viresh Kumar wrote:
> Sorry for the delay guys, was away :(
> 
>>
>>   Possible unsafe locking scenario:
>>
>>         CPU0                    CPU1
>>         ----                    ----
>>    lock(&policy->rwsem);
>>                                 lock(s_active#9);
>>                                 lock(&policy->rwsem);
>>    lock(s_active#9);
>>
> 
> Thanks for coming to my rescue Stephen :), I was quite sure I got this
> with ondemand
> as well..

Yeah, I'm confused by it a bit because I wasn't able to reproduce it.  But I've
got some pretty clear instructions on how to do it.

> 
> I will be looking very closely at the code now to see what's going wrong.
> And btw, does anybody here has the exact understanding of why this
> lockdep does happen? I mean what was the real problem for which we

The issue is the collision between the setup & teardown of the policy's governor
sysfs files.

On creation the kernel does:

down_write(&policy->rwsem)
mutex_lock(kernfs_mutex) <- note this is similar to the "old" sysfs_mutex.

The opposite happens on a governor switch, specifically the existing governor's
exit, and then we get a lockdep warning.

I tried to reproduce with the instructions but was unable to ... ut that was on
Friday ;) and I'm going to try again this morning.  I've also ping'd some of the
engineers here in the office who are working on ARM to get access to a system to
do further analysis and testing.

I'll ping back later in the day with some results.

Sorry I don't have a better answer or solution yet, Viresh.

P.