28.3 ACPI

ACPI (advanced configuration and power interface) was designed to enable the operating system to set up and control the individual hardware components. ACPI supersedes both PnP and APM. It delivers information about the battery, AC adapter, temperature, fan, and system events, like close lid or battery low.

The BIOS provides tables containing information about the individual components and hardware access methods. The operating system uses this information for tasks like assigning interrupts or activating and deactivating components. Because the operating system executes commands stored in the BIOS, the functionality depends on the BIOS implementation. The tables ACPI can detect and load are reported in /var/log/boot.msg. See Section 28.3.4, acpi=off for more information about troubleshooting ACPI problems.

28.3.1 ACPI in Action

If the kernel detects an ACPI BIOS when the system is booted, ACPI is activated automatically. The boot parameter acpi=force may be necessary for some older machines. The computer must support ACPI 2.0 or later. Check the kernel boot messages in /var/log/boot.msg to see if ACPI was activated.

Subsequently, a number of modules must be loaded. This is done by the start script of acpid. If any of these modules cause problems, the respective module can be excluded from loading or unloading in /etc/sysconfig/powersave/common. The system log (/var/log/messages) contains the messages of the modules, enabling you to see which components were detected.

/proc/acpi now contains a number of files that provide information about the system state or can be used to change some of the states. Some features do not work yet because they are still under development and the support of some functions largely depends on the implementation of the manufacturer.

All files (except dsdt and fadt) can be read with cat. In some files, settings can be modified with echo, for example, echo X > file to specify suitable values for X. One possibility for easy access to those values is the powersave command, which acts as a front-end for the Powersave daemon. The following describes the most important files:

/proc/acpi/info

General information about ACPI.

/proc/acpi/alarm

Here, specify when the system should wake from a sleep state. Currently, this feature is not fully supported.

/proc/acpi/sleep

Provides information about possible sleep states.

/proc/acpi/event

All events are reported here and processed by the Powersave daemon (powersaved). If no daemon accesses this file, events, such as a brief click on the power button or closing the lid, can be read with cat /proc/acpi/event (terminate with Ctrl+C).

/proc/acpi/dsdt and /proc/acpi/fadt

These files contain the ACPI tables DSDT (differentiated system description table) and FADT (fixed ACPI description table). They can be read with acpidmp, acpidisasm, and dmdecode. These programs and their documentation are located in the package pmtools. For example, acpidmp DSDT | acpidisasm.

/proc/acpi/ac_adapter/AC/state

Shows whether the AC adapter is connected.

/proc/acpi/battery/BAT*/{alarm,info,state}

Detailed information about the battery state. The charge level is read by comparing the last full capacity from info with the remaining capacity from state. A more comfortable way to do this is to use one of the special programs introduced in Section 28.3.3, ACPI Tools. The charge level at which a battery event (such as warning, low and critical) is triggered can be specified in alarm.

/proc/acpi/button

This directory contains information about various switches, like the laptop lid and buttons.

/proc/acpi/fan/FAN/state

Shows if the fan is currently active. Activate or deactivate the fan manually by writing 0 (on) or 3 (off) into this file. However, both the ACPI code in the kernel and the hardware (or the BIOS) overwrite this setting when the system gets too warm.

/proc/acpi/processor/*

A separate subdirectory is kept for each CPU included in your system.

/proc/acpi/processor/*/info

Information about the energy saving options of the processor.

/proc/acpi/processor/*/power

Information about the current processor state. An asterisk next to C2 indicates that the processor is idle. This is the most frequent state, as can be seen from the usage value.

/proc/acpi/processor/*/throttling

Can be used to set the throttling of the processor clock. Usually, throttling is possible in eight levels. This is independent of the frequency control of the CPU.

/proc/acpi/processor/*/limit

If the performance (outdated) and the throttling are automatically controlled by a daemon, the maximum limits can be specified here. Some of the limits are determined by the system. Some can be adjusted by the user.

/proc/acpi/thermal_zone/

A separate subdirectory exists for every thermal zone. A thermal zone is an area with similar thermal properties whose number and names are designated by the hardware manufacturer. However, many of the possibilities offered by ACPI are rarely implemented. Instead, the temperature control is handled conventionally by the BIOS. The operating system is not given much opportunity to intervene, because the life span of the hardware is at stake. Therefore, some of the files only have a theoretical value.

/proc/acpi/thermal_zone/*/temperature

Current temperature of the thermal zone.

/proc/acpi/thermal_zone/*/state

The state indicates if everything is ok or if ACPI applies active or passive cooling. In the case of ACPI-independent fan control, this state is always ok.

/proc/acpi/thermal_zone/*/cooling_mode

Select the cooling method controlled by ACPI. Choose from passive (less performance, economical) or active cooling mode (full performance, fan noise).

/proc/acpi/thermal_zone/*/trip_points

Enables the determination of temperature limits for triggering specific actions, like passive or active cooling, suspension (hot), or a shutdown (critical). The possible actions are defined in the DSDT (device-dependent). The trip points determined in the ACPI specification are critical, hot, passive, active1, and active2. Even if not all of them are implemented, they must always be entered in this file in this order. For example, the entry echo 90:0:70:0:0 > trip_points sets the temperature for critical to 90 and the temperature for passive to 70 (all temperatures measured in degrees Celsius).

/proc/acpi/thermal_zone/*/polling_frequency

If the value in temperature is not updated automatically when the temperature changes, toggle the polling mode here. The command echo X > /proc/acpi/thermal_zone/*/polling_frequency causes the temperature to be queried every X seconds. Set X=0 to disable polling.

None of these settings, information, and events need to be edited manually. This can be done with the Powersave daemon (powersaved) and its various front-ends, like powersave, kpowersave, and wmpowersave. See Section 28.3.3, ACPI Tools.

28.3.2 Controlling the CPU Performance

The CPU can save energy in three ways. Depending on the operating mode of the computer, these methods can be combined. Saving energy also means that the system heats up less and the fans are activated less frequently.

Frequency and Voltage Scaling

PowerNow! and Speedstep are the designations AMD and Intel use for this technology. However, this technology is also applied in processors of other manufacturers. The clock frequency of the CPU and its core voltage are reduced at the same time, resulting in more than linear energy savings. This means that when the frequency is halved (half performance), far less than half of the energy is consumed. This technology is independent from APM or ACPI. There are two main approaches to performing CPU frequency scaling—by the kernel itself or by a userspace application. Therefore, there are different kernel governors that can be set below /sys/devices/system/cpu/cpu*/cpufreq/.

userspace governor

If the userspace governor is set, the kernel gives the control of CPU frequency scaling to a userspace application, usually a daemon. In SUSE Linux Enterprise distributions, this daemon is the powersaved package. When this implementation is used, the CPU frequency is adjusted in regard to the current system load. By default, one of the kernel implementations is used. However, on some hardware or in regard to specific processors or drivers, the userspace implementation is still the only working solution.

ondemand governor

This is the kernel implementation of a dynamic CPU frequency policy and should work on most systems. As soon as there is a high system load, the CPU frequency is immediately increased. It is lowered on a low system load.

conservative governor

This governor is similar to the on demand implementation, except that a more conservative policy is used. The load of the system must be high for a specific amount of time before the CPU frequency is increased.

powersave governor

The cpu frequency is statically set to the lowest possible.

performance governor

The cpu frequency is statically set to the highest possible.

Throttling the Clock Frequency

This technology omits a certain percentage of the clock signal impulses for the CPU. At 25% throttling, every fourth impulse is omitted. At 87.5%, only every eighth impulse reaches the processor. However, the energy savings are a little less than linear. Normally, throttling is only used if frequency scaling is not available or to maximize power savings. This technology, too, must be controlled by a special process. The system interface is /proc/acpi/processor/*/throttling.

Putting the Processor to Sleep

The operating system puts the processor to sleep whenever there is nothing to do. In this case, the operating system sends the CPU a halt command. There are three states: C1, C2, and C3. In the most economic state, C3, even the synchronization of the processor cache with the main memory is halted. Therefore, this state can only be applied if no other device modifies the contents of the main memory via bus master activity. Some drivers prevent the use of C3. The current state is displayed in /proc/acpi/processor/*/power.

Frequency scaling and throttling are only relevant if the processor is busy, because the most economic C state is applied anyway when the processor is idle. If the CPU is busy, frequency scaling is the recommended power saving method. Often the processor only works with a partial load. In this case, it can be run with a lower frequency. Usually, dynamic frequency scaling controlled by the kernel on demand governor or a daemon, such as powersaved, is the best approach. A static setting to a low frequency is useful for battery operation or if you want the computer to be cool or quiet.

Throttling should be used as the last resort, for example, to extend the battery operation time despite a high system load. However, some systems do not run smoothly when they are throttled too much. Moreover, CPU throttling does not make sense if the CPU has little to do.

In SUSE Linux Enterprise these technologies are controlled by the powersave daemon. The configuration is explained in Section 28.5, The powersave Package.

28.3.3 ACPI Tools

The range of more or less comprehensive ACPI utilities includes tools that merely display information, like the battery charge level and the temperature (acpi, klaptopdaemon, wmacpimon, etc.), tools that facilitate the access to the structures in /proc/acpi or that assist in monitoring changes (akpi, acpiw, gtkacpiw), and tools for editing the ACPI tables in the BIOS (package pmtools).

28.3.4 Troubleshooting

There are two different types of problems. On one hand, the ACPI code of the kernel may contain bugs that were not detected in time. In this case, a solution will be made available for download. More often, however, the problems are caused by the BIOS. Sometimes, deviations from the ACPI specification are purposely integrated in the BIOS to circumvent errors in the ACPI implementation in other widespread operating systems. Hardware components that have serious errors in the ACPI implementation are recorded in a blacklist that prevents the Linux kernel from using ACPI for these components.

The first thing to do when problems are encountered is to update the BIOS. If the computer does not boot at all, one of the following boot parameters may be helpful:

pci=noacpi

Do not use ACPI for configuring the PCI devices.

acpi=ht

Only perform a simple resource configuration. Do not use ACPI for other purposes.

acpi=off

Disable ACPI.

WARNING: Problems Booting without ACPI

Some newer machines (especially SMP systems and AMD64 systems) need ACPI for configuring the hardware correctly. On these machines, disabling ACPI can cause problems.

Monitor the boot messages of the system with the command dmesg | grep -2i acpi (or all messages, because the problem may not be caused by ACPI) after booting. If an error occurs while parsing an ACPI table, the most important table—the DSDT—can be replaced with an improved version. In this case, the faulty DSDT of the BIOS is ignored. The procedure is described in Section 28.5.4, Troubleshooting.

In the kernel configuration, there is a switch for activating ACPI debug messages. If a kernel with ACPI debugging is compiled and installed, experts searching for an error can be supported with detailed information.

If you experience BIOS or hardware problems, it is always advisable to contact the manufacturers. Especially if they do not always provide assistance for Linux, they should be confronted with the problems. Manufacturers will only take the issue seriously if they realize that an adequate number of their customers use Linux.

For More Information

Additional documentation and help on ACPI: