Re: [PATCH v2 3/5] hwmon: scmi-hwmon: Log number of sensors

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] The scmi_hwmon_probe function configures the wrong sensors because it passes the array index `i` instead of the actual firmware `sensor->id` to `sensor_ops->config_set`.
--

commit adc025236960e1cfd8886f8043954aa0a9f37bc8
Author: Alex Tran <alex.tran@oss.qualcomm.com>

hwmon: scmi-hwmon: Log number of sensors

This commit logs the number of discovered sensors upon a successful probe in
the SCMI hwmon driver. This aligns the driver with the existing SCMI client
driver logging pattern, making it easier to confirm firmware-exposed resources
during debugging.

> diff --git a/drivers/hwmon/scmi-hwmon.c b/drivers/hwmon/scmi-hwmon.c
> index eec223d174c07..456ce11d050c6 100644
> --- a/drivers/hwmon/scmi-hwmon.c
> +++ b/drivers/hwmon/scmi-hwmon.c
> @@ -335,49 +335,50 @@ static int scmi_hwmon_probe(struct scmi_device *sdev)
[ ... ]
>  	for (i = 0; i < nr_count_temp; i++) {
>  		int ret;
>  
>  		sensor = *(scmi_sensors->info[hwmon_temp] + i);
>  		if (!sensor)
>  			continue;
>  
>  		ret = sensor_ops->config_set(ph, i, sensor_config);

This isn't a regression introduced by this commit, but could passing the
local array index i to sensor_ops->config_set() instead of the firmware
sensor->id result in configuring the wrong sensors?

Since SCMI sensor IDs are global and can map to any sensor type, passing
the local array index might mistakenly configure arbitrary unrelated
hardware components.

Could this leave the actual temperature sensors unconfigured and break
subsequent thermal zone registration? Is it possible that thermal
mitigation would fail to function correctly due to the underlying
temperature sensors remaining disabled?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260513-scmi-client-probe-log-v2-0-36607e9dd540@oss.qualcomm.com?part=3