On 2/21/19 9:49 AM, Wolfram Sang wrote:
>
>> +static void rcar_gen3_hwmon_action(void *data)
>> +{
>> + struct thermal_zone_device *zone = data;
>> +
>> + thermal_remove_hwmon_sysfs(zone);
>> +}
>> +
>> static int rcar_gen3_thermal_probe(struct platform_device *pdev)
>> {
>> struct rcar_gen3_thermal_priv *priv;
>> @@ -429,6 +437,17 @@ static int rcar_gen3_thermal_probe(struct
>> platform_device *pdev)
>> if (ret < 0)
>> goto error_unregister;
>>
>> + tsc->zone->tzp->no_hwmon = false;
>> + ret = thermal_add_hwmon_sysfs(tsc->zone);
>> + if (ret)
>> + goto error_unregister;
>> +
>> + ret = devm_add_action(dev, rcar_gen3_hwmon_action, zone);
>> + if (ret) {
>> + rcar_gen3_hwmon_action(zone);
>> + goto error_unregister;
>> + }
>> +
>> dev_info(dev, "TSC%d: Loaded %d trip points\n", i, ret);
>> }
>
> This slipped through the cracks for me, I am sorry about that. I know it
> is applied by now but I still wonder if the above functionality couldn't
> be moved into the core as a generic helper? There is nothing R-Car
> specific in here. From what I see it could be something like:
>
> devm_thermal_activate_hwmon_sysfs(dev, zone);
>
> ?
Do you want to grow function name for each parameter to be changed ?
I don't think that scales. However, I don't have any better idea how to
deal with the hwmon thus far.
--
Best regards,
Marek Vasut