Re: [PATCH 1/3] hwmon: pmbus: event notification with alarms

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] PMBus core's pmbus_notify() calls sysfs_notify() on the wrong kobject, silently breaking sysfs poll() and udev events for hardware alarms.
--

--- Patch [1]: [PATCH 1/3] hwmon: pmbus: event notification with alarms ---
commit 2d27260577dbfaf59cc5dfe956f36d8429cbffb8
Author: Vincent Jardin <[email protected]>

hwmon: pmbus: event notification with alarms

This commit introduces pmbus_get_hwmon_device() to allow PMBus chip drivers
to retrieve the hwmon class device for an i2c_client. This enables drivers
to call hwmon_notify_event() on per-sensor alarm attributes during work-item
polling when SMBALERT# is unavailable.

> diff --git a/drivers/hwmon/pmbus/pmbus_core.c b/drivers/hwmon/pmbus/pmbus_core.c
> index 3143b9e0316c4..6ba00ad732973 100644
> --- a/drivers/hwmon/pmbus/pmbus_core.c
> +++ b/drivers/hwmon/pmbus/pmbus_core.c

[ ... ]

> @@ -695,6 +695,14 @@ const struct pmbus_driver_info *pmbus_get_driver_info(struct i2c_client *client)
>  }
>  EXPORT_SYMBOL_NS_GPL(pmbus_get_driver_info, "PMBUS");
>  
> +struct device *pmbus_get_hwmon_device(struct i2c_client *client)
> +{
> +	struct pmbus_data *data = i2c_get_clientdata(client);
> +
> +	return data->hwmon_dev;
> +}
> +EXPORT_SYMBOL_NS_GPL(pmbus_get_hwmon_device, "PMBUS");
> +

[Severity: High]
This is a pre-existing issue, but while looking at how hwmon devices are
exposed for alarm notification, I noticed that pmbus_notify() might be
sending notifications to the wrong kobject:

drivers/hwmon/pmbus/pmbus_core.c:pmbus_notify() {
    ...
    sysfs_notify(&data->dev->kobj, NULL, da->attr.name);
    kobject_uevent(&data->dev->kobj, KOBJ_CHANGE);
    ...
}

Since the hwmon attributes are attached to the hwmon device rather than the
parent I2C device, does sysfs_notify() fail to find the attribute and return
silently?

If so, this could prevent userspace processes polling on hardware alarm sysfs
attributes from being awakened when a hardware fault occurs, leaving alarms
unhandled.

Would it make sense to update pmbus_notify() to use data->hwmon_dev->kobj
instead to ensure poll() and udev events work correctly?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1