Re: [PATCH V3 01/14] cpufreq: Create for_each_{in}active_policy()

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

On 08-05-15, 23:46, Rafael J. Wysocki wrote:
> On Friday, May 08, 2015 11:53:44 AM Viresh Kumar wrote:

> > +#define __for_each_active_policy(__policy, __temp, __active)		\
> > +	for (__temp = NULL, __policy = next_policy(NULL, __active);	\
> > +	     &__policy->policy_list != &cpufreq_policy_list ||		\
> > +	     ((__policy = __temp) && 0);				\
> > +	     __temp = __policy, __policy = next_policy(__policy, __active))
> 
> can be rewritten as:
> 
> #define __for_each_active_policy(__policy, __active)		\
> 	for (__policy = first_policy(__active);	__policy;	\
> 	     __policy = next_policy(__policy, __active))
> 
> and you don't need to explain what it does even.  Of course, next_policy()
> has to return 'false' when it can't find anything suitable, but that shouldn't
> be too difficult to arrange I suppose?
> 
> Or if you need __temp because the object pointed to by __policy can go away,
> you can follow the design of list_for_each_entry_safe() here.

I wasn't using __temp to be safe, but to make sure that the loop
points to a 'active' policy at the end. But that isn't required
anymore as the only call site that required this, is fixed in a better
way:

c75de0ac0756 ("cpufreq: Schedule work for the first-online CPU on resume")