1 CPU frequency and voltage scaling code in the Linux(TM) kernel 2 3 4 L i n u x C P U F r e q 5 6 C P U F r e q G o v e r n o r s 7 8 - information for users and developers - 9 10 11 Dominik Brodowski <linux@brodo.de> 12 some additions and corrections by Nico Golde <nico@ngolde.de> 13 14 15 16 Clock scaling allows you to change the clock speed of the CPUs on the 17 fly. This is a nice method to save battery power, because the lower 18 the clock speed, the less power the CPU consumes. 19 20 21Contents: 22--------- 231. What is a CPUFreq Governor? 24 252. Governors In the Linux Kernel 262.1 Performance 272.2 Powersave 282.3 Userspace 292.4 Ondemand 302.5 Conservative 31 323. The Governor Interface in the CPUfreq Core 33 34 35 361. What Is A CPUFreq Governor? 37============================== 38 39Most cpufreq drivers (in fact, all except one, longrun) or even most 40cpu frequency scaling algorithms only offer the CPU to be set to one 41frequency. In order to offer dynamic frequency scaling, the cpufreq 42core must be able to tell these drivers of a "target frequency". So 43these specific drivers will be transformed to offer a "->target" 44call instead of the existing "->setpolicy" call. For "longrun", all 45stays the same, though. 46 47How to decide what frequency within the CPUfreq policy should be used? 48That's done using "cpufreq governors". Two are already in this patch 49-- they're the already existing "powersave" and "performance" which 50set the frequency statically to the lowest or highest frequency, 51respectively. At least two more such governors will be ready for 52addition in the near future, but likely many more as there are various 53different theories and models about dynamic frequency scaling 54around. Using such a generic interface as cpufreq offers to scaling 55governors, these can be tested extensively, and the best one can be 56selected for each specific use. 57 58Basically, it's the following flow graph: 59 60CPU can be set to switch independently | CPU can only be set 61 within specific "limits" | to specific frequencies 62 63 "CPUfreq policy" 64 consists of frequency limits (policy->{min,max}) 65 and CPUfreq governor to be used 66 / \ 67 / \ 68 / the cpufreq governor decides 69 / (dynamically or statically) 70 / what target_freq to set within 71 / the limits of policy->{min,max} 72 / \ 73 / \ 74 Using the ->setpolicy call, Using the ->target call, 75 the limits and the the frequency closest 76 "policy" is set. to target_freq is set. 77 It is assured that it 78 is within policy->{min,max} 79 80 812. Governors In the Linux Kernel 82================================ 83 842.1 Performance 85--------------- 86 87The CPUfreq governor "performance" sets the CPU statically to the 88highest frequency within the borders of scaling_min_freq and 89scaling_max_freq. 90 91 922.2 Powersave 93------------- 94 95The CPUfreq governor "powersave" sets the CPU statically to the 96lowest frequency within the borders of scaling_min_freq and 97scaling_max_freq. 98 99 1002.3 Userspace 101------------- 102 103The CPUfreq governor "userspace" allows the user, or any userspace 104program running with UID "root", to set the CPU to a specific frequency 105by making a sysfs file "scaling_setspeed" available in the CPU-device 106directory. 107 108 1092.4 Ondemand 110------------ 111 112The CPUfreq governor "ondemand" sets the CPU depending on the 113current usage. To do this the CPU must have the capability to 114switch the frequency very quickly. There are a number of sysfs file 115accessible parameters: 116 117sampling_rate: measured in uS (10^-6 seconds), this is how often you 118want the kernel to look at the CPU usage and to make decisions on 119what to do about the frequency. Typically this is set to values of 120around '10000' or more. 121 122show_sampling_rate_(min|max): the minimum and maximum sampling rates 123available that you may set 'sampling_rate' to. 124 125up_threshold: defines what the average CPU usage between the samplings 126of 'sampling_rate' needs to be for the kernel to make a decision on 127whether it should increase the frequency. For example when it is set 128to its default value of '80' it means that between the checking 129intervals the CPU needs to be on average more than 80% in use to then 130decide that the CPU frequency needs to be increased. 131 132ignore_nice_load: this parameter takes a value of '0' or '1'. When 133set to '0' (its default), all processes are counted towards the 134'cpu utilisation' value. When set to '1', the processes that are 135run with a 'nice' value will not count (and thus be ignored) in the 136overall usage calculation. This is useful if you are running a CPU 137intensive calculation on your laptop that you do not care how long it 138takes to complete as you can 'nice' it and prevent it from taking part 139in the deciding process of whether to increase your CPU frequency. 140 141 1422.5 Conservative 143---------------- 144 145The CPUfreq governor "conservative", much like the "ondemand" 146governor, sets the CPU depending on the current usage. It differs in 147behaviour in that it gracefully increases and decreases the CPU speed 148rather than jumping to max speed the moment there is any load on the 149CPU. This behaviour more suitable in a battery powered environment. 150The governor is tweaked in the same manner as the "ondemand" governor 151through sysfs with the addition of: 152 153freq_step: this describes what percentage steps the cpu freq should be 154increased and decreased smoothly by. By default the cpu frequency will 155increase in 5% chunks of your maximum cpu frequency. You can change this 156value to anywhere between 0 and 100 where '0' will effectively lock your 157CPU at a speed regardless of its load whilst '100' will, in theory, make 158it behave identically to the "ondemand" governor. 159 160down_threshold: same as the 'up_threshold' found for the "ondemand" 161governor but for the opposite direction. For example when set to its 162default value of '20' it means that if the CPU usage needs to be below 16320% between samples to have the frequency decreased. 164 1653. The Governor Interface in the CPUfreq Core 166============================================= 167 168A new governor must register itself with the CPUfreq core using 169"cpufreq_register_governor". The struct cpufreq_governor, which has to 170be passed to that function, must contain the following values: 171 172governor->name - A unique name for this governor 173governor->governor - The governor callback function 174governor->owner - .THIS_MODULE for the governor module (if 175 appropriate) 176 177The governor->governor callback is called with the current (or to-be-set) 178cpufreq_policy struct for that CPU, and an unsigned int event. The 179following events are currently defined: 180 181CPUFREQ_GOV_START: This governor shall start its duty for the CPU 182 policy->cpu 183CPUFREQ_GOV_STOP: This governor shall end its duty for the CPU 184 policy->cpu 185CPUFREQ_GOV_LIMITS: The limits for CPU policy->cpu have changed to 186 policy->min and policy->max. 187 188If you need other "events" externally of your driver, _only_ use the 189cpufreq_governor_l(unsigned int cpu, unsigned int event) call to the 190CPUfreq core to ensure proper locking. 191 192 193The CPUfreq governor may call the CPU processor driver using one of 194these two functions: 195 196int cpufreq_driver_target(struct cpufreq_policy *policy, 197 unsigned int target_freq, 198 unsigned int relation); 199 200int __cpufreq_driver_target(struct cpufreq_policy *policy, 201 unsigned int target_freq, 202 unsigned int relation); 203 204target_freq must be within policy->min and policy->max, of course. 205What's the difference between these two functions? When your governor 206still is in a direct code path of a call to governor->governor, the 207per-CPU cpufreq lock is still held in the cpufreq core, and there's 208no need to lock it again (in fact, this would cause a deadlock). So 209use __cpufreq_driver_target only in these cases. In all other cases 210(for example, when there's a "daemonized" function that wakes up 211every second), use cpufreq_driver_target to lock the cpufreq per-CPU 212lock before the command is passed to the cpufreq processor driver. 213 214