redonkable
/
alistair23-linux
1
0
Fork 0
Code Releases Activity

Documentation: power: update Energy Model description 

The Energy Model framework supports also other devices than CPUs. Update
related information and add description for the new usage.

Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Quentin Perret <qperret@google.com>
Signed-off-by: Lukasz Luba <lukasz.luba@arm.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
zero-sugar-mainline-defconfig
Lukasz Luba 2020-05-27 10:58:53 +01:00 committed by Rafael J. Wysocki
parent f0b5694791
commit 7b7570ad0d
1 changed files with 75 additions and 60 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

135
Documentation/power/energy-model.rst
View File

@ -1,15 +1,17 @@
==================== .. SPDX-License-Identifier: GPL-2.0
Energy Model of CPUs
==================== =======================
Energy Model of devices
=======================
1. Overview 1. Overview
----------- -----------
The Energy Model (EM) framework serves as an interface between drivers knowing The Energy Model (EM) framework serves as an interface between drivers knowing
the power consumed by CPUs at various performance levels, and the kernel the power consumed by devices at various performance levels, and the kernel
subsystems willing to use that information to make energy-aware decisions. subsystems willing to use that information to make energy-aware decisions.
The source of the information about the power consumed by CPUs can vary greatly The source of the information about the power consumed by devices can vary greatly
from one platform to another. These power costs can be estimated using from one platform to another. These power costs can be estimated using
devicetree data in some cases. In others, the firmware will know better. devicetree data in some cases. In others, the firmware will know better.
Alternatively, userspace might be best positioned. And so on. In order to avoid Alternatively, userspace might be best positioned. And so on. In order to avoid
@ -25,7 +27,7 @@ framework, and interested clients reading the data from it::
+---------------+ +-----------------+ +---------------+ +---------------+ +-----------------+ +---------------+
| Thermal (IPA) | | Scheduler (EAS) | | Other | | Thermal (IPA) | | Scheduler (EAS) | | Other |
+---------------+ +-----------------+ +---------------+ +---------------+ +-----------------+ +---------------+
| | em_pd_energy() | | | em_cpu_energy() |
| | em_cpu_get() | | | em_cpu_get() |
+---------+ | +---------+ +---------+ | +---------+
| | | | | |
@ -35,7 +37,7 @@ framework, and interested clients reading the data from it::
| Framework | | Framework |
+---------------------+ +---------------------+
^ ^ ^ ^ ^ ^
| | | em_register_perf_domain() | | | em_dev_register_perf_domain()
+----------+ | +---------+ +----------+ | +---------+
| | | | | |
+---------------+ +---------------+ +--------------+ +---------------+ +---------------+ +--------------+
@ -47,12 +49,12 @@ framework, and interested clients reading the data from it::
| Device Tree | | Firmware | | ? | | Device Tree | | Firmware | | ? |
+--------------+ +---------------+ +--------------+ +--------------+ +---------------+ +--------------+
The EM framework manages power cost tables per 'performance domain' in the In case of CPU devices the EM framework manages power cost tables per
system. A performance domain is a group of CPUs whose performance is scaled 'performance domain' in the system. A performance domain is a group of CPUs
together. Performance domains generally have a 1-to-1 mapping with CPUFreq whose performance is scaled together. Performance domains generally have a
policies. All CPUs in a performance domain are required to have the same 1-to-1 mapping with CPUFreq policies. All CPUs in a performance domain are
micro-architecture. CPUs in different performance domains can have different required to have the same micro-architecture. CPUs in different performance
micro-architectures. domains can have different micro-architectures.
2. Core APIs 2. Core APIs
@ -70,14 +72,16 @@ CONFIG_ENERGY_MODEL must be enabled to use the EM framework.
Drivers are expected to register performance domains into the EM framework by Drivers are expected to register performance domains into the EM framework by
calling the following API:: calling the following API::
int em_register_perf_domain(cpumask_t *span, unsigned int nr_states, int em_dev_register_perf_domain(struct device *dev, unsigned int nr_states,
struct em_data_callback *cb); struct em_data_callback *cb, cpumask_t *cpus);
Drivers must specify the CPUs of the performance domains using the cpumask Drivers must provide a callback function returning <frequency, power> tuples
argument, and provide a callback function returning <frequency, power> tuples for each performance state. The callback function provided by the driver is free
for each capacity state. The callback function provided by the driver is free
to fetch data from any relevant location (DT, firmware, ...), and by any mean to fetch data from any relevant location (DT, firmware, ...), and by any mean
deemed necessary. See Section 3. for an example of driver implementing this deemed necessary. Only for CPU devices, drivers must specify the CPUs of the
performance domains using cpumask. For other devices than CPUs the last
argument must be set to NULL.
See Section 3. for an example of driver implementing this
callback, and kernel/power/energy_model.c for further documentation on this callback, and kernel/power/energy_model.c for further documentation on this
API. API.
@ -85,13 +89,20 @@ API.
2.3 Accessing performance domains 2.3 Accessing performance domains
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
There are two API functions which provide the access to the energy model:
em_cpu_get() which takes CPU id as an argument and em_pd_get() with device
pointer as an argument. It depends on the subsystem which interface it is
going to use, but in case of CPU devices both functions return the same
performance domain.
Subsystems interested in the energy model of a CPU can retrieve it using the Subsystems interested in the energy model of a CPU can retrieve it using the
em_cpu_get() API. The energy model tables are allocated once upon creation of em_cpu_get() API. The energy model tables are allocated once upon creation of
the performance domains, and kept in memory untouched. the performance domains, and kept in memory untouched.
The energy consumed by a performance domain can be estimated using the The energy consumed by a performance domain can be estimated using the
em_pd_energy() API. The estimation is performed assuming that the schedutil em_cpu_energy() API. The estimation is performed assuming that the schedutil
CPUfreq governor is in use. CPUfreq governor is in use in case of CPU device. Currently this calculation is
not provided for other type of devices.
More details about the above APIs can be found in include/linux/energy_model.h. More details about the above APIs can be found in include/linux/energy_model.h.
@ -106,42 +117,46 @@ EM framework::
-> drivers/cpufreq/foo_cpufreq.c -> drivers/cpufreq/foo_cpufreq.c
01 static int est_power(unsigned long *mW, unsigned long *KHz, int cpu) 01 static int est_power(unsigned long *mW, unsigned long *KHz,
02 { 02 struct device *dev)
03 long freq, power; 03 {
04 04 long freq, power;
05 /* Use the 'foo' protocol to ceil the frequency */ 05
06 freq = foo_get_freq_ceil(cpu, *KHz); 06 /* Use the 'foo' protocol to ceil the frequency */
07 if (freq < 0); 07 freq = foo_get_freq_ceil(dev, *KHz);
08 return freq; 08 if (freq < 0);
09 09 return freq;
10 /* Estimate the power cost for the CPU at the relevant freq. */ 10
11 power = foo_estimate_power(cpu, freq); 11 /* Estimate the power cost for the dev at the relevant freq. */
12 if (power < 0); 12 power = foo_estimate_power(dev, freq);
13 return power; 13 if (power < 0);
14 14 return power;
15 /* Return the values to the EM framework */ 15
16 *mW = power; 16 /* Return the values to the EM framework */
17 *KHz = freq; 17 *mW = power;
18 18 *KHz = freq;
19 return 0; 19
20 } 20 return 0;
21 21 }
22 static int foo_cpufreq_init(struct cpufreq_policy *policy) 22
23 { 23 static int foo_cpufreq_init(struct cpufreq_policy *policy)
24 struct em_data_callback em_cb = EM_DATA_CB(est_power); 24 {
25 int nr_opp, ret; 25 struct em_data_callback em_cb = EM_DATA_CB(est_power);
26 26 struct device *cpu_dev;
27 /* Do the actual CPUFreq init work ... */ 27 int nr_opp, ret;
28 ret = do_foo_cpufreq_init(policy); 28
29 if (ret) 29 cpu_dev = get_cpu_device(cpumask_first(policy->cpus));
30 return ret; 30
31 31 /* Do the actual CPUFreq init work ... */
32 /* Find the number of OPPs for this policy */ 32 ret = do_foo_cpufreq_init(policy);
33 nr_opp = foo_get_nr_opp(policy); 33 if (ret)
34 34 return ret;
35 /* And register the new performance domain */ 35
36 em_register_perf_domain(policy->cpus, nr_opp, &em_cb); 36 /* Find the number of OPPs for this policy */
37 37 nr_opp = foo_get_nr_opp(policy);
38 return 0; 38
39 } 39 /* And register the new performance domain */
40 em_dev_register_perf_domain(cpu_dev, nr_opp, &em_cb, policy->cpus);
41
42 return 0;
43 }