Hi Rudolf,
> Following patch fixes the w83792d driver. It adds error checking to
> device_create file and also cares to destroy the files upon exit.
>
> I will test it as soon I will get the disk back to my test machine with this
> chip. Possible dates are: Saturday evening, Sunday and Monday for sure ;)
> Until now, compiles without warnings.
Applied, with the following fix:
> @@ -1342,12 +1428,16 @@ static int
> w83792d_detach_client(struct i2c_client *client)
> {
> struct w83792d_data *data = i2c_get_clientdata(client);
> - int err;
> + int err, i;
>
> /* main client */
> if (data)
> hwmon_device_unregister(data->class_dev);
>
> + sysfs_remove_group(&client->dev.kobj, &w83792d_group);
> + for (i = 0; i < ARRAY_SIZE(w83792d_group_fan); i++)
> + sysfs_remove_group(&client->dev.kobj, &w83792d_group_fan[i]);
> +
> if ((err = i2c_detach_client(client)))
> return err;
>
>
Only the main client has the files, so the file removal should be
conditional upon "data" too.
The rest looks OK to me. Testing is still welcome nevertheless.
Thanks,
--
Jean Delvare
