Re: [PATCH v3 1/9] opp: add new helper API dev_pm_opp_set_level()

[Praveen Talari <quic_ptalari@quicinc.com>]

Hi Viresh

Thank you for review.

On 5/2/2025 11:07 AM, Viresh Kumar wrote:
> On 02-05-25, 08:40, Praveen Talari wrote:
>> To configure a device to a specific performance level, consumer drivers
>> currently need to determine the OPP based on the exact level and then
>> set it, resulting in code duplication across drivers.
>>
>> The new helper API, dev_pm_opp_set_level(), addresses this issue by
>> providing a streamlined method for consumer drivers to find and set the
>> OPP based on the desired performance level, thereby eliminating
>> redundancy.
>>
>> Signed-off-by: Praveen Talari <quic_ptalari@quicinc.com>
>>
>> v2 -> v3
>> - moved function defination to pm_opp.h from core.c with inline
>> - updated return value with IS_ERR(opp)
>>
>> v1 -> v2
>> - reorder sequence of tags in commit text
> As Trilok mentioned, this is not the right place for this.
>
>> ---
>>   include/linux/pm_opp.h | 27 +++++++++++++++++++++++++++
>>   1 file changed, 27 insertions(+)
>>
>> diff --git a/include/linux/pm_opp.h b/include/linux/pm_opp.h
>> index e7b5c602c92f..31ed8a7b554e 100644
>> --- a/include/linux/pm_opp.h
>> +++ b/include/linux/pm_opp.h
>> @@ -197,6 +197,28 @@ int dev_pm_opp_get_sharing_cpus(struct device *cpu_dev, struct cpumask *cpumask)
>>   void dev_pm_opp_remove_table(struct device *dev);
>>   void dev_pm_opp_cpumask_remove_table(const struct cpumask *cpumask);
>>   int dev_pm_opp_sync_regulators(struct device *dev);
>> +
>> +/*
>> + * dev_pm_opp_set_level() - Configure device for a level
>> + * @dev: device for which we do this operation
>> + * @level: level to set to
>> + *
>> + * Return: 0 on success, a non-zero value if there is an error otherwise.
>> + */
> No need of these for simple wrappers.
>
>> +static inline int dev_pm_opp_set_level(struct device *dev, unsigned int level)
>> +{
>> +	struct dev_pm_opp *opp = dev_pm_opp_find_level_exact(dev, level);
>> +	int ret;
>> +
>> +	if (IS_ERR(opp))
>> +		return IS_ERR(opp);
> IS_ERR is wrong here, should be PTR_ERR.
>
>> +
>> +	ret = dev_pm_opp_set_opp(dev, opp);
>> +	dev_pm_opp_put(opp);
>> +
>> +	return ret;
>> +}
>> +
>>   #else
>>   static inline struct opp_table *dev_pm_opp_get_opp_table(struct device *dev)
>>   {
>> @@ -461,6 +483,11 @@ static inline int dev_pm_opp_sync_regulators(struct device *dev)
>>   	return -EOPNOTSUPP;
>>   }
>>   
>> +static inline int dev_pm_opp_set_level(struct device *dev, unsigned int level)
>> +{
>> +	return -EOPNOTSUPP;
>> +}
>> +
> No need of these too for such wrappers. And then this isn't rebased
How come? i have synced  linux-next today itself and pushed from it, 
even i didn't face any issue.
Let me know how/where rebased i.e linux-next or linux?
> over latest changes in the OPP core.
>
> I modified it and applied the below version to my tree now.
>