Re: [PATCH 1/2] cpufreq: unlock correct rwsem while updating policy->cpu

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Monday, September 16, 2013 10:38:05 PM Viresh Kumar wrote:
> On 16 September 2013 21:57, Jon Medhurst (Tixy) <tixy@linaro.org> wrote:
> > If I take mainline code and just change the line above to:
> 
> You meant this line by above line?
> 
>          unlock_policy_rwsem_write(cpu);
> 
> >    up_write(&per_cpu(cpu_policy_rwsem, (per_cpu(cpufreq_cpu_data,
> > cpu))->last_cpu));
> > then the big_little cpufreq driver works for me.
> 
> That would be same as: up_write(&per_cpu(cpu_policy_rwsem, policy->last_cpu));
> 
> >> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> >> index 43c24aa..c18bf7b 100644
> >> --- a/drivers/cpufreq/cpufreq.c
> >> +++ b/drivers/cpufreq/cpufreq.c
> >> @@ -952,9 +952,16 @@ static void update_policy_cpu(struct cpufreq_policy *policy, unsigned int cpu)
> >>       if (cpu == policy->cpu)
> >>               return;
> >>
> >> +     /* take direct locks as lock_policy_rwsem_write wouldn't work here */
> >> +     down_write(&per_cpu(cpu_policy_rwsem, policy->cpu));
> >> +     down_write(&per_cpu(cpu_policy_rwsem, cpu));
> >> +
> >>       policy->last_cpu = policy->cpu;
> >>       policy->cpu = cpu;
> >>
> >> +     up_write(&per_cpu(cpu_policy_rwsem, cpu));
> >> +     up_write(&per_cpu(cpu_policy_rwsem, policy->cpu));
> >
> > You've just overwritten policy->cpu with cpu.
> 
> Stupid enough :)
> 
> > I tried using
> > policy->last_cpu to fix that, but it still doesn't work for me (giving
> > the lockdep warning I mentioned.) If I change the code to just lock the
> > original policy->cpu lock only, then all is fine.
> 
> Fixed my patch now.. find attached..

Care to resend with a subject indicating that that's a patch update?

Like [PATCH v2] etc. or similar?

Rafael