cpu-cooling-api.rst

  1=======================
  2CPU cooling APIs How To
  3=======================
  4
  5Written by Amit Daniel Kachhap <amit.kachhap@linaro.org>
  6
  7Updated: 6 Jan 2015
  8
  9Copyright (c)  2012 Samsung Electronics Co., Ltd(http://www.samsung.com)
 10
 110. Introduction
 12===============
 13
 14The generic cpu cooling(freq clipping) provides registration/unregistration APIs
 15to the caller. The binding of the cooling devices to the trip point is left for
 16the user. The registration APIs returns the cooling device pointer.
 17
 181. cpu cooling APIs
 19===================
 20
 211.1 cpufreq registration/unregistration APIs
 22--------------------------------------------
 23
 24    ::
 25
 26	struct thermal_cooling_device
 27	*cpufreq_cooling_register(struct cpumask *clip_cpus)
 28
 29    This interface function registers the cpufreq cooling device with the name
 30    "thermal-cpufreq-%x". This api can support multiple instances of cpufreq
 31    cooling devices.
 32
 33   clip_cpus:
 34	cpumask of cpus where the frequency constraints will happen.
 35
 36    ::
 37
 38	struct thermal_cooling_device
 39	*of_cpufreq_cooling_register(struct cpufreq_policy *policy)
 40
 41    This interface function registers the cpufreq cooling device with
 42    the name "thermal-cpufreq-%x" linking it with a device tree node, in
 43    order to bind it via the thermal DT code. This api can support multiple
 44    instances of cpufreq cooling devices.
 45
 46    policy:
 47	CPUFreq policy.
 48
 49
 50    ::
 51
 52	void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev)
 53
 54    This interface function unregisters the "thermal-cpufreq-%x" cooling device.
 55
 56    cdev: Cooling device pointer which has to be unregistered.
 57
 582. Power models
 59===============
 60
 61The power API registration functions provide a simple power model for
 62CPUs.  The current power is calculated as dynamic power (static power isn't
 63supported currently).  This power model requires that the operating-points of
 64the CPUs are registered using the kernel's opp library and the
 65`cpufreq_frequency_table` is assigned to the `struct device` of the
 66cpu.  If you are using CONFIG_CPUFREQ_DT then the
 67`cpufreq_frequency_table` should already be assigned to the cpu
 68device.
 69
 70The dynamic power consumption of a processor depends on many factors.
 71For a given processor implementation the primary factors are:
 72
 73- The time the processor spends running, consuming dynamic power, as
 74  compared to the time in idle states where dynamic consumption is
 75  negligible.  Herein we refer to this as 'utilisation'.
 76- The voltage and frequency levels as a result of DVFS.  The DVFS
 77  level is a dominant factor governing power consumption.
 78- In running time the 'execution' behaviour (instruction types, memory
 79  access patterns and so forth) causes, in most cases, a second order
 80  variation.  In pathological cases this variation can be significant,
 81  but typically it is of a much lesser impact than the factors above.
 82
 83A high level dynamic power consumption model may then be represented as::
 84
 85	Pdyn = f(run) * Voltage^2 * Frequency * Utilisation
 86
 87f(run) here represents the described execution behaviour and its
 88result has a units of Watts/Hz/Volt^2 (this often expressed in
 89mW/MHz/uVolt^2)
 90
 91The detailed behaviour for f(run) could be modelled on-line.  However,
 92in practice, such an on-line model has dependencies on a number of
 93implementation specific processor support and characterisation
 94factors.  Therefore, in initial implementation that contribution is
 95represented as a constant coefficient.  This is a simplification
 96consistent with the relative contribution to overall power variation.
 97
 98In this simplified representation our model becomes::
 99
100	Pdyn = Capacitance * Voltage^2 * Frequency * Utilisation
101
102Where `capacitance` is a constant that represents an indicative
103running time dynamic power coefficient in fundamental units of
104mW/MHz/uVolt^2.  Typical values for mobile CPUs might lie in range
105from 100 to 500.  For reference, the approximate values for the SoC in
106ARM's Juno Development Platform are 530 for the Cortex-A57 cluster and
107140 for the Cortex-A53 cluster.