Re: cpufreq_resume warning issue

[Viresh Kumar <viresh.kumar@linaro.org>]

On 21-03-16, 03:43, Zengtao (B) wrote:
> Hi all:
> 	I have got the warning in the following code, 
> 	/*
> 	 * schedule call cpufreq_update_policy() for first-online CPU, as that
> 	 * wouldn't be hotplugged-out on suspend. It will verify that the
> 	 * current freq is in sync with what we believe it to be.
> 	 */
> 	policy = cpufreq_cpu_get_raw(cpumask_first(cpu_online_mask));
> 	if (WARN_ON(!policy))
> 		return;
> 	schedule_work(&policy->update);
> 
> in my platform, two cores, core 1 don't have the dvfs feature, but core 2 has.
> So I think that is the warning reason.
> 
> My question:
> 1. Do we need to warning here? 

Yes. But your system just broke an assumption we always had. i.e. all the CPUs
take part in DVFS :)

And its not just about this piece of code, but everything else that cpufreq
does.

For example, cpufreq core must be calling ->init() callback for CPU0 as well..
What are you doing there ?

I am not sure what's the right way of handling this thing is going to be.

Why doesn't you have DVFS for both the cores ?

-- 
viresh