Re: [RFC PATCH 1/4] cpufreq: governor: support scheduler cpufreq callbacks on remote CPUs

[Rafael J. Wysocki]

On Friday, April 29, 2016 04:08:16 PM Viresh Kumar wrote:
> On 19-04-16, 19:39, Steve Muckle wrote:
> > diff --git a/drivers/cpufreq/cpufreq_governor.c b/drivers/cpufreq/cpufreq_governor.c
> > index 20f0a4e114d1..429d3a5b9866 100644
> > --- a/drivers/cpufreq/cpufreq_governor.c
> > +++ b/drivers/cpufreq/cpufreq_governor.c
> > @@ -248,6 +248,20 @@ static void dbs_irq_work(struct irq_work *irq_work)
> >  	schedule_work_on(smp_processor_id(), &policy_dbs->work);
> >  }
> >  
> > +#ifdef CONFIG_SMP
> > +static inline void dbs_irq_work_queue(struct policy_dbs_info *policy_dbs,
> > +				      int cpu)
> > +{
> > +	irq_work_queue_on(&policy_dbs->irq_work, cpu);
> > +}
> > +#else
> > +static inline void dbs_irq_work_queue(struct policy_dbs_info *policy_dbs,
> > +				      int cpu)
> > +{
> > +	irq_work_queue(&policy_dbs->irq_work);
> > +}
> > +#endif
> 
> Any clue, why we don't have a non-SMP version of irq_work_queue_on(), Which can
> simply call irq_work_queue() ?

Because nobody else needs it?

But I agree that it would be nicer to add the stub to irq_work.h.