Re: [PATCH v5] hwmon: (pmbus/core): Implement regulator get_status

[Naresh Solanki <naresh.solanki@9elements.com>]

On 24-11-2022 07:48 pm, Guenter Roeck wrote:
> On 11/24/22 00:34, Naresh Solanki wrote:
>> From: Patrick Rudolph <patrick.rudolph@9elements.com>
>>
>> Add get_status for pmbus_regulator_ops.
>>
>> ---
>> Changes:
>> - use lock throughout the function
>> - Avoid line continuation upto 100 column
>> - Optimize use of & and | operator
>> - Check for VOUT, IOUT, TEMPERATURE bit in status word before checking
>>    respective status register for fault.
>> - Report regulator current status.
>> - Utilize get_error_flag to check for regulator errors.
>> - Check for return value of function get_error_flag
>>
>> Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
>> Signed-off-by: Naresh Solanki <Naresh.Solanki@9elements.com>
>> ---
>>   drivers/hwmon/pmbus/pmbus_core.c | 44 ++++++++++++++++++++++++++++++++
>>   1 file changed, 44 insertions(+)
>>
>> diff --git a/drivers/hwmon/pmbus/pmbus_core.c 
>> b/drivers/hwmon/pmbus/pmbus_core.c
>> index 20ca26e19db7..0b13214c662f 100644
>> --- a/drivers/hwmon/pmbus/pmbus_core.c
>> +++ b/drivers/hwmon/pmbus/pmbus_core.c
>> @@ -2855,6 +2855,49 @@ static int 
>> pmbus_regulator_get_error_flags(struct regulator_dev *rdev, unsigned
>>       return 0;
>>   }
>> +static int pmbus_regulator_get_status(struct regulator_dev *rdev)
>> +{
>> +    struct device *dev = rdev_get_dev(rdev);
>> +    struct i2c_client *client = to_i2c_client(dev->parent);
>> +    struct pmbus_data *data = i2c_get_clientdata(client);
>> +    u8 page = rdev_get_id(rdev);
>> +    int status, ret;
>> +
>> +    mutex_lock(&data->update_lock);
>> +    status = pmbus_get_status(client, page, PMBUS_STATUS_WORD);
>> +    if (status < 0) {
>> +        ret = status;
>> +        goto unlock;
>> +    }
>> +
>> +    if (status & PB_STATUS_OFF) {
>> +        ret = REGULATOR_STATUS_OFF;
>> +        goto unlock;
>> +    }
>> +
>> +    /* If regulator is ON & reports power good then return ON */
>> +    if (!(status & PB_STATUS_POWER_GOOD_N)) {
>> +        ret = REGULATOR_STATUS_ON;
>> +        goto unlock;
>> +    }
>> +
>> +    if (rdev->desc->ops->get_error_flags) {
> 
> Looking into this again, why is this check necessary ? Isn't this
> the regulator_ops from below ? Also, why not just call
> pmbus_regulator_get_error_flags() directly ?
Yes. Felt that to be the right way.
Will update to do pmbus_regulator_get)error_flags directly
> 
>> +        ret = rdev->desc->ops->get_error_flags(rdev, &status);
>> +        if (ret)
>> +            goto unlock;
>> +
>> +        if (status & (REGULATOR_ERROR_UNDER_VOLTAGE | 
>> REGULATOR_ERROR_OVER_CURRENT |
>> +           REGULATOR_ERROR_REGULATION_OUT | REGULATOR_ERROR_FAIL |
>> +           REGULATOR_ERROR_OVER_TEMP))
>> +            ret = REGULATOR_STATUS_ERROR;
> 
> If the condition above is false, the return value will be 0, or
> REGULATOR_STATUS_OFF. Is that intentional ?
No. It should be REGULATOR_ERR_UNDEFINED. will fix in next revision.
> 
>> +    } else
>> +        ret = REGULATOR_STATUS_UNDEFINED;
>> +
> 
> CHECK: braces {} should be used on all arms of this statement
> #72: FILE: drivers/hwmon/pmbus/pmbus_core.c:2884:
> +    if (rdev->desc->ops->get_error_flags) {
> [...]
> +    } else
> 
> Guenter
> 
>> +unlock:
>> +    mutex_unlock(&data->update_lock);
>> +    return ret;
>> +}
>> +
>>   static int pmbus_regulator_get_low_margin(struct i2c_client *client, 
>> int page)
>>   {
>>       struct pmbus_data *data = i2c_get_clientdata(client);
>> @@ -2995,6 +3038,7 @@ const struct regulator_ops pmbus_regulator_ops = {
>>       .disable = pmbus_regulator_disable,
>>       .is_enabled = pmbus_regulator_is_enabled,
>>       .get_error_flags = pmbus_regulator_get_error_flags,
>> +    .get_status = pmbus_regulator_get_status,
>>       .get_voltage = pmbus_regulator_get_voltage,
>>       .set_voltage = pmbus_regulator_set_voltage,
>>       .list_voltage = pmbus_regulator_list_voltage,
>>
>> base-commit: 2c71b3246ec3246522e8cb7c8191dc7a5d62cc70
> 

Regards,
Naresh