[Bug 204807] Hardware monitoring sensor nct6798d doesn't work unless acpi_enforce_resources=lax is enabled

[bugzilla-daemon@bugzilla.kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=204807

--- Comment #136 from Eugene Shalygin (eugene.shalygin@gmail.com) ---
(In reply to Denis Pauk from comment #135)
> Both drivers have used same entry point and difference as I see that: old
> boards return some list of sensors with names,

There was a special WMI interface that reads all sensors, both from SIO and EC,
kind implementing what you did for the nct driver and what is implemented in
the asus-wmi-ec-sensors, but done fully in the DSDT code. 

> new one always returns zero as count of sensors

There is simply no that high-level WMI interface in the new boards.

> and requires some hardcoded list of sensors.

I now think the EC sensors are at the same registers for the old and new
boards.  

> And list of old and new boards is not intersected. 

Exactly! So half the driver will be a dead code anyway. Would it be better to
load only one of the small drivers? Also, for the old boards the nct6775 driver
will load and asus-wmi-sensors provides duplicate readings.  
> 
> As for me, when we have 30% of similar code better have one driver for both
> cases. Currently driver has 1126 lines.

The only shared code between those two is the HWMON interface functions, which
is more or less the same for many HWMON drivers. 

So, would it be simpler to provide 3 drivers: nct6775, asus-wmi-sensors,
asus-wmi-ec-sensors?

> Do you like to be in MAINTAINERS list?

Yes, please. I still have work to do with that (even not all available sensors
are discovered yet).

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.