Re: [PATCH v7 12/23] PM: EM: Add em_perf_state_from_pd() to get performance states table

[Lukasz Luba <lukasz.luba@arm.com>]

On 1/29/24 18:13, Dietmar Eggemann wrote:
> On 17/01/2024 10:57, Lukasz Luba wrote:
> 
> [...]
> 
>> diff --git a/include/linux/energy_model.h b/include/linux/energy_model.h
>> index 494df6942cf7..5ebe9dbec8e1 100644
>> --- a/include/linux/energy_model.h
>> +++ b/include/linux/energy_model.h
>> @@ -339,6 +339,23 @@ static inline int em_pd_nr_perf_states(struct em_perf_domain *pd)
>>   	return pd->nr_perf_states;
>>   }
>>   
>> +/**
>> + * em_perf_state_from_pd() - Get the performance states table of perf.
>> + *				domain
>> + * @pd		: performance domain for which this must be done
>> + *
>> + * To use this function the rcu_read_lock() should be hold. After the usage
>> + * of the performance states table is finished, the rcu_read_unlock() should
>> + * be called.
>> + *
>> + * Return: the pointer to performance states table of the performance domain
>> + */
>> +static inline
>> +struct em_perf_state *em_perf_state_from_pd(struct em_perf_domain *pd)
> 
> This is IMHO hard to get since:
> 
>    struct em_perf_table {
>      struct rcu_head rcu;
>      struct kref kref;
>      struct em_perf_state state[];
>    };
> 
> So very often a 'struct em_perf_table' is named 'table' and 'struct
> em_perf_table::state' as well. E.g. in em_adjust_new_capacity().
> 
>    struct em_perf_state *new_table;
> 
>    new_table = em_table->state;
> 
> In older EM code, we used 'struct em_perf_state *ps' to avoid this
> confusion, I guess.
> 
> And what you get from the PD is actually a state vector so maybe:
> 
> struct em_perf_state *em_get_perf_states(struct em_perf_domain *pd)
> 
> The 'from_pd' seems obvious because of the parameter?

Rafael proposed that function name in his review comments, so I
followed. I might address your comment about the:
'struct em_perf_state *ps'
while I will do re-basing and sending v8.