Rigo

I have found the problem. Essentially when setting up (enumerating) physical devices. It's not safe to assume that device[0] is the primary GPU. The devices seem to be enumerated in...

Scaleway, please fix. This is stupid!

Those are usually the last instructions that are run on most tests. I will have to do some verification and see what's going on. Unfortunately I cannot view your attached...

The programmer and software is absolutely fine. It's one of the preferred tools for doing in-circuit EEPROM work. (that is without desoldering the flash chip). If you're not comfortable with...

Also apologies for the delay, quite busy with a work project. Life and all that :)

Can you please take a look here and submit the same additional information: https://github.com/rigred/sensor_config/issues/7#issuecomment-475893049

In your case modinfo nct6793 bios screenshots of the sensors sudo find /sys/class/hwmon/hwmon[0-9]/ -type f -printf '%p, ' -exec cat {} \; > hwon_dump.csv

There really isn't much of a guideline. The sensors application relies partly on hwmon (that is `/sys/class/hwmon/hwmonX`) for fan data, as well as kernel modules and the `/etc/sensors.d/` config table...

@mchwalisz Have you made any progress on this? If so you can send me what you have and I'll try put something together.

This is presently not something that can be resolved with a config file. The negative readouts are due to a bug/missing feature in the nct6779 kernel module.