Re: [RFC v3] cpufreq: Make sure frequency transitions are serialized

[Viresh Kumar]

On 20 March 2014 14:02, Srivatsa S. Bhat
<srivatsa.bhat@xxxxxxxxxxxxxxxxxx> wrote:
> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> index 199b52b..5283f10 100644
> --- a/drivers/cpufreq/cpufreq.c
> +++ b/drivers/cpufreq/cpufreq.c
> @@ -349,6 +349,39 @@ void cpufreq_notify_post_transition(struct cpufreq_policy *policy,
>  EXPORT_SYMBOL_GPL(cpufreq_notify_post_transition);
>
>
> +void cpufreq_freq_transition_begin(struct cpufreq_policy *policy,
> +               struct cpufreq_freqs *freqs, unsigned int state)
> +{
> +wait:
> +       wait_event(&policy->transition_wait, !policy->transition_ongoing);
> +
> +       mutex_lock(&policy->transition_lock);
> +
> +       if (policy->transition_ongoing) {
> +               mutex_unlock(&policy->transition_lock);
> +               goto wait;
> +       }
> +
> +       policy->transition_ongoing = true;
> +
> +       mutex_unlock(&policy->transition_lock);
> +
> +       cpufreq_notify_transition(policy, freqs, CPUFREQ_PRECHANGE);
> +}
> +
> +void cpufreq_freq_transition_end(struct cpufreq_policy *policy,
> +               struct cpufreq_freqs *freqs, unsigned int state)
> +{
> +       cpufreq_notify_transition(policy, freqs, CPUFREQ_POSTCHANGE);
> +
> +       mutex_lock(&policy->transition_lock);

Why do we need locking here? You explained that earlier :)

Also, I would like to add this here:

    WARN_ON(policy->transition_ongoing);

> +       policy->transition_ongoing = false;
> +       mutex_unlock(&policy->transition_lock);
> +
> +       wake_up(&policy->transition_wait);
> +}
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/