On Thu, 2011-12-08 at 16:10 -0500, Lucas Stach wrote:
> This adds a new hwmon-core interface to centralize sysfs handling
> and enable cooperation with other in-kernel drivers.
>
> v3:
> - checkpatch clean
>
> v4:
> - more flexible interface to allow different capabilities for
> every sensor and holes in sensor numbering
> - corrected a few oversights
> - split out vid and vrm handling as separate feature
> - support 2d attributes for trip point and treshold handling
>
> Signed-off-by: Lucas Stach <dev@xxxxxxxxxx>
> ---
[ ... ]
> +
> + hw_dev_attr->sensor_dev.dev_attr.attr.mode = mode;
> + if (mode && S_IRUGO) {
> + if (type == value)
> + hw_dev_attr->sensor_dev.dev_attr.show = &get_num;
> + else
> + hw_dev_attr->sensor_dev.dev_attr.show = &get_text;
> + }
> + if (mode && S_IWUGO)
This and "mode && SIRUGO" look wrong and don't make much sense. Guess
you mean & ?
[ ... ]
> +
> +struct hwmon_device_instance {
> + struct hwmon_device_caps caps;
> + int (*get_numeric_attr) (struct device *hw_device,
> + enum hwmon_feature feature, u32 subfeature, u8 index,
> + u8 nr, int *value);
> + int (*set_numeric_attr) (struct device *hw_device,
> + enum hwmon_feature feature, u32 subfeature, u8 index,
> + u8 nr, int value);
> + int (*get_text_attr) (struct device *hw_device,
> + enum hwmon_feature feature, u32 subfeature, u8 index,
> + char *buf);
> + struct list_head sysfs_node;
Looks like sysfs_node is used internally by the core code, and not by
the driver. If so, it should not be exported.
Regarding inst_data, I think it would be better to keep the current
approach with a local data structure, and reference hwinst from it. So
you would have
data->hw_inst = hw_inst;
i2c_set_clientdata(client, data);
and
data = i2c_get_clientdata(client);
hw_inst = data->hw_inst;
To avoid the many mallocs in the probe function, maybe it would be
possible to pass the various parameters to hwmon_create_sysfs() and have
it return a pointer to hw_inst. Just an idea.
Thanks,
Guenter
_______________________________________________
lm-sensors mailing list
lm-sensors@xxxxxxxxxxxxxx
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
