All SuperH processors supporting rate rounding through the clock. The following processors for the sparc64 architecture are supported by. This has been merged into the acpi-cpufreq module in later kernels. First of all, a scaling driver has to be registered for CPUFreq to work. When you installed your system there is a very good chance your CPU was detected by default and the module you need for for scaling is already running.

Uploader: Tezil
Date Added: 11 February 2015
File Size: 7.26 Mb
Operating Systems: Windows NT/2000/XP/2003/2003/7/8/10 MacOS 10/X
Downloads: 56320
Price: Free* [*Free Regsitration Required]

Performance governor – CPU runs at max frequency cpufrsq of load. The cpb knob is never present for any processors without the underlying hardware feature e.

Kernel version Make sure and kernel version is at least 2. In turn, if any CPUs are registered after the registration of the scaling driver, the CPUFreq xpufreq will be invoked to take note of them at their registration time. As a rule, it also is implemented differently by different vendors. If you don’t find the option it is probably enabled by default.

Every time the governor type changes all values get reset. CPU frequency scaling does what it sounds like.

CPU Performance Scaling — The Linux Kernel documentation

This one smd the user – or. There are some userspace programs that can be run to scale the processors frequency. By “echoing” a new frequency into this.

If that frequency cannot be determined, this attribute should not be present. The CPUFreq governors tested were conservative, ondemand, performance, powersave, and schedutil. As a last resort if any of these don’t work you can try the generic one for ACPI.

acpi-cpufreq: Add support for modern AMD CPUs []

This attribute is functional only if the userspace scaling governor is attached to the given policy. This is how much the frequency is allowed to change in one go. So when the CPU scales it will only go to a max of and not The actual implementation, however, works on the system-wide basis and setting that knob for one policy causes the same value of it to be set for all of the other policies at the same time.


Conservative governor – CPU frequency is scaled based on current load of the system. The Linux Kernel 4. The policy objects created during CPU initialization and other data structures associated with them are torn down when the scaling driver is unregistered which happens when the kernel module containing it is unloaded, for example or when the last CPU belonging to the given policy in unregistered.

CPU frequency scaling

This has been merged into the acpi-cpufreq module in later kernels. To show c;ufreq available governors you can use. To do this the CPU must have the capability to switch the frequency very quickly. National Semiconductors Geode GX. The difference is that it gracefully increases and decreases the CPU speed rather than jumping to max speed the moment there is any load on the CPU.

Otherwise, the frequency will not be changed.

acpi-cpufreq: Add support for modern AMD CPUs

It does not mean that boosting is actually in use at the moment on any CPUs in the system. Typically, it is set to values of the order of 10 ms.

Some scaling drivers also add driver-specific attributes to the policy directories in sysfs to control policy-specific aspects of driver behavior. This is useful if you are running a CPU intensive calculation zmd your laptop that you do not care how long it takes to complete as you can ‘nice’ it and prevent it from taking part in the deciding process of whether to increase your CPU frequency.


Threshold value in percent, 20 by default used to determine the frequency change direction.

Consequently, the same set of scaling governors should be suitable for every supported platform. Those attributes, referred to as governor tunables, can be either global system-wide or per-policy, depending on cufreq scaling driver in use. That callback is expected to initialize the performance scaling cputreq interface for the given CPU or, more precisely, for the set of CPUs sharing the hardware interface it belongs to, represented by its policy object and, if the policy object it has been called for is cpurreq, to set parameters of the policy, like the minimum and maximum frequencies supported by the hardware, the table of available frequencies if the set of supported P-states is not a continuous rangeand the mask of CPUs that belong to the same policy including both online and offline CPUs.

As a result, the scheduler context overhead from this governor is minimum, but it causes additional CPU context switches to happen relatively often and the CPU P-state updates triggered by it can be relatively irregular.

To check if they are try the following command.