1 .. SPDX-License-Identifier: GPL-2.0
3 =======================
4 Energy Model of devices
5 =======================
10 The Energy Model (EM) framework serves as an interface between drivers knowing
11 the power consumed by devices at various performance levels, and the kernel
12 subsystems willing to use that information to make energy-aware decisions.
14 The source of the information about the power consumed by devices can vary greatly
15 from one platform to another. These power costs can be estimated using
16 devicetree data in some cases. In others, the firmware will know better.
17 Alternatively, userspace might be best positioned. And so on. In order to avoid
18 each and every client subsystem to re-implement support for each and every
19 possible source of information on its own, the EM framework intervenes as an
20 abstraction layer which standardizes the format of power cost tables in the
21 kernel, hence enabling to avoid redundant work.
23 The figure below depicts an example of drivers (Arm-specific here, but the
24 approach is applicable to any architecture) providing power costs to the EM
25 framework, and interested clients reading the data from it::
27 +---------------+ +-----------------+ +---------------+
28 | Thermal (IPA) | | Scheduler (EAS) | | Other |
29 +---------------+ +-----------------+ +---------------+
32 +---------+ | +---------+
35 +---------------------+
38 +---------------------+
40 | | | em_dev_register_perf_domain()
41 +----------+ | +---------+
43 +---------------+ +---------------+ +--------------+
44 | cpufreq-dt | | arm_scmi | | Other |
45 +---------------+ +---------------+ +--------------+
48 +--------------+ +---------------+ +--------------+
49 | Device Tree | | Firmware | | ? |
50 +--------------+ +---------------+ +--------------+
52 In case of CPU devices the EM framework manages power cost tables per
53 'performance domain' in the system. A performance domain is a group of CPUs
54 whose performance is scaled together. Performance domains generally have a
55 1-to-1 mapping with CPUFreq policies. All CPUs in a performance domain are
56 required to have the same micro-architecture. CPUs in different performance
57 domains can have different micro-architectures.
66 CONFIG_ENERGY_MODEL must be enabled to use the EM framework.
69 2.2 Registration of performance domains
70 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
72 Drivers are expected to register performance domains into the EM framework by
73 calling the following API::
75 int em_dev_register_perf_domain(struct device *dev, unsigned int nr_states,
76 struct em_data_callback *cb, cpumask_t *cpus);
78 Drivers must provide a callback function returning <frequency, power> tuples
79 for each performance state. The callback function provided by the driver is free
80 to fetch data from any relevant location (DT, firmware, ...), and by any mean
81 deemed necessary. Only for CPU devices, drivers must specify the CPUs of the
82 performance domains using cpumask. For other devices than CPUs the last
83 argument must be set to NULL.
84 See Section 3. for an example of driver implementing this
85 callback, and kernel/power/energy_model.c for further documentation on this
89 2.3 Accessing performance domains
90 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
92 There are two API functions which provide the access to the energy model:
93 em_cpu_get() which takes CPU id as an argument and em_pd_get() with device
94 pointer as an argument. It depends on the subsystem which interface it is
95 going to use, but in case of CPU devices both functions return the same
98 Subsystems interested in the energy model of a CPU can retrieve it using the
99 em_cpu_get() API. The energy model tables are allocated once upon creation of
100 the performance domains, and kept in memory untouched.
102 The energy consumed by a performance domain can be estimated using the
103 em_cpu_energy() API. The estimation is performed assuming that the schedutil
104 CPUfreq governor is in use in case of CPU device. Currently this calculation is
105 not provided for other type of devices.
107 More details about the above APIs can be found in include/linux/energy_model.h.
113 This section provides a simple example of a CPUFreq driver registering a
114 performance domain in the Energy Model framework using the (fake) 'foo'
115 protocol. The driver implements an est_power() function to be provided to the
118 -> drivers/cpufreq/foo_cpufreq.c
120 01 static int est_power(unsigned long *mW, unsigned long *KHz,
121 02 struct device *dev)
125 06 /* Use the 'foo' protocol to ceil the frequency */
126 07 freq = foo_get_freq_ceil(dev, *KHz);
130 11 /* Estimate the power cost for the dev at the relevant freq. */
131 12 power = foo_estimate_power(dev, freq);
135 16 /* Return the values to the EM framework */
142 23 static int foo_cpufreq_init(struct cpufreq_policy *policy)
144 25 struct em_data_callback em_cb = EM_DATA_CB(est_power);
145 26 struct device *cpu_dev;
148 29 cpu_dev = get_cpu_device(cpumask_first(policy->cpus));
150 31 /* Do the actual CPUFreq init work ... */
151 32 ret = do_foo_cpufreq_init(policy);
155 36 /* Find the number of OPPs for this policy */
156 37 nr_opp = foo_get_nr_opp(policy);
158 39 /* And register the new performance domain */
159 40 em_dev_register_perf_domain(cpu_dev, nr_opp, &em_cb, policy->cpus);