Re: [PATCH 1/7] hwmon: (pmbus/adm1266) use wall-clock seconds for SET_RTC

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] Silent truncation of a 64-bit real-time timestamp causes Y2038 arithmetic overflow.
- [High] Unconditionally overwriting the hardware RTC during probe destroys retained state and defeats cross-reboot correlation.
--

commit 260e18847f33c3f0022668e0092b4f93049b2ea6
Author: Abdurrahman Hussain <abdurrahman@nexthop.ai>

hwmon: (pmbus/adm1266) use wall-clock seconds for SET_RTC

This patch modifies the ADM1266 driver to seed the device internal RTC
with wall-clock time instead of monotonic time since boot. This change is
intended to allow cross-reboot correlation of blackbox records.

> diff --git a/drivers/hwmon/pmbus/adm1266.c b/drivers/hwmon/pmbus/adm1266.c
> index d90f8f80be8e0..a86666c73a5e9 100644
> --- a/drivers/hwmon/pmbus/adm1266.c
> +++ b/drivers/hwmon/pmbus/adm1266.c
> @@ -432,7 +432,7 @@ static int adm1266_set_rtc(struct adm1266_data *data)
>  	char write_buf[6];
>  	int i;
>  
> -	kt = ktime_get_seconds();
> +	kt = ktime_get_real_seconds();

Can this lead to an arithmetic overflow? ktime_get_real_seconds returns
a 64-bit time64_t, but the subsequent serialization loop only processes
4 bytes:

    for (i = 0; i < 4; i++)
        write_buf[2 + i] = (kt >> (i * 8)) & 0xFF;

This silently truncates the upper 32 bits of the timestamp. Should the value
be clamped or bounds-checked to prevent silent wrap-around?

Also, if adm1266_set_rtc is called unconditionally during early probe, will
ktime_get_real_seconds return an uninitialized time near the 1970 epoch
if the system wall-clock has not yet been synchronized via an external RTC
or userspace NTP?

Overwriting the internal hardware RTC on every probe could destroy any valid
time state retained across a warm reboot, breaking the stated goal of
cross-reboot log correlation. Does this code need to preserve the existing
RTC state rather than seeding it at probe?

>  
>  	memset(write_buf, 0, sizeof(write_buf));
>

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260508-adm1266-v1-0-ec08bf29e0ce@nexthop.ai?part=1