Power meter ACPI000D:00: Ignoring unsafe software power cap

I get the following message on Ubuntu 14.04 during boot:

power meter ACPI000D:00: Ignoring unsafe software power cap

The message comes up shortly after the red/orange Ubuntu background appears and the screen goes black into console for a moment. The system loads fine afterwards and the login screen appears as usual. Nevertheless it's annoying.

Does anybody have an idea how to fix this?

EDIT: While there is no issue using Ubuntu via display manager, the message persists in the TTY consoles which I cannot use anymore.

UPDATE: I upgraded to Ubuntu 16.04 and while the warning still shows up shortly during boot and is afterwards in the first line of the TTY consoles, they work as intended, i.e. I can finally log myself in and use them.

This error occurs because the ACPI _PMC (Power Meter Capabilities) object for that device ACPI000D (the power meter) has bit 0 set that indicates that the device supports power measuring capability, however, the kernel driver has detected that the hardware does not support this feature and hence reports that the power capability is probably dubious and so ignores it. Basically, the firmware indicates that power metering is available, but the kernel driver does not believe it for your hardware.

Currently, the only hardware that seems to allow this is certain IBM hardware with "IBM Active Energy Manager".

For more details about this ACPI functionality, consult section 10.4.1 of the ACPI specification.

One can force this feature on using the kernel parameter force_cap_on=1, however this is not advised unless you know what you are doing.

I think this message can most probably be safely ignored. One could turn the kernel error messages off completely to disable this, however, this will bar you from seeing possibly helpful errors that the kernel reports in the future. This message is being reported as dev_err() level device error message, so one would have to disable dev_err level error messages to silence this error.

For example, setting kernel parameter loglevel=2 will report emergencies, alerts and critical errors and ignore errors of level 3 (error) and below. To do this edit /etc/default/grub (with root privilege) and change GRUB_CMDLINE_LINUX_DEFAULT to include loglevel=2, e.g.

sudo gedit /etc/default/grub

edit and set:

GRUB_CMDLINE_LINUX_DEFAULT="quiet splash loglevel=2"

save the changes and run:

sudo update-grub

The long term fix would be to modify the driver and change the dev_err to a dev_warn