xref: /dpdk/doc/guides/prog_guide/power_man.rst (revision 00e57b0e550b7df2047e6d0bde8965c7ae17d203)
1..  SPDX-License-Identifier: BSD-3-Clause
2    Copyright(c) 2010-2014 Intel Corporation.
3
4Power Management
5================
6
7The DPDK Power Management feature allows users space applications to save power
8by dynamically adjusting CPU frequency or entering into different C-States.
9
10*   Adjusting the CPU frequency dynamically according to the utilization of RX queue.
11
12*   Entering into different deeper C-States according to the adaptive algorithms to speculate
13    brief periods of time suspending the application if no packets are received.
14
15The interfaces for adjusting the operating CPU frequency are in the power management library.
16C-State control is implemented in applications according to the different use cases.
17
18CPU Frequency Scaling
19---------------------
20
21The Linux kernel provides a cpufreq module for CPU frequency scaling for each lcore.
22For example, for cpuX, /sys/devices/system/cpu/cpuX/cpufreq/ has the following sys files for frequency scaling:
23
24*   affected_cpus
25
26*   bios_limit
27
28*   cpuinfo_cur_freq
29
30*   cpuinfo_max_freq
31
32*   cpuinfo_min_freq
33
34*   cpuinfo_transition_latency
35
36*   related_cpus
37
38*   scaling_available_frequencies
39
40*   scaling_available_governors
41
42*   scaling_cur_freq
43
44*   scaling_driver
45
46*   scaling_governor
47
48*   scaling_max_freq
49
50*   scaling_min_freq
51
52*   scaling_setspeed
53
54In the DPDK, scaling_governor is configured in user space.
55Then, a user space application can prompt the kernel by writing scaling_setspeed to adjust the CPU frequency
56according to the strategies defined by the user space application.
57
58Core-load Throttling through C-States
59-------------------------------------
60
61Core state can be altered by speculative sleeps whenever the specified lcore has nothing to do.
62In the DPDK, if no packet is received after polling,
63speculative sleeps can be triggered according the strategies defined by the user space application.
64
65Per-core Turbo Boost
66--------------------
67
68Individual cores can be allowed to enter a Turbo Boost state on a per-core
69basis. This is achieved by enabling Turbo Boost Technology in the BIOS, then
70looping through the relevant cores and enabling/disabling Turbo Boost on each
71core.
72
73Use of Power Library in a Hyper-Threaded Environment
74----------------------------------------------------
75
76In the case where the power library is in use on a system with Hyper-Threading enabled,
77the frequency on the physical core is set to the highest frequency of the Hyper-Thread siblings.
78So even though an application may request a scale down, the core frequency will
79remain at the highest frequency until all Hyper-Threads on that core request a scale down.
80
81API Overview of the Power Library
82---------------------------------
83
84The main methods exported by power library are for CPU frequency scaling and include the following:
85
86*   **Freq up**: Prompt the kernel to scale up the frequency of the specific lcore.
87
88*   **Freq down**: Prompt the kernel to scale down the frequency of the specific lcore.
89
90*   **Freq max**: Prompt the kernel to scale up the frequency of the specific lcore to the maximum.
91
92*   **Freq min**: Prompt the kernel to scale down the frequency of the specific lcore to the minimum.
93
94*   **Get available freqs**: Read the available frequencies of the specific lcore from the sys file.
95
96*   **Freq get**: Get the current frequency of the specific lcore.
97
98*   **Freq set**: Prompt the kernel to set the frequency for the specific lcore.
99
100*   **Enable turbo**: Prompt the kernel to enable Turbo Boost for the specific lcore.
101
102*   **Disable turbo**: Prompt the kernel to disable Turbo Boost for the specific lcore.
103
104User Cases
105----------
106
107The power management mechanism is used to save power when performing L3 forwarding.
108
109
110Ethernet PMD Power Management API
111---------------------------------
112
113Abstract
114~~~~~~~~
115
116Existing power management mechanisms require developers to change application
117design or change code to make use of it. The PMD power management API provides a
118convenient alternative by utilizing Ethernet PMD RX callbacks, and triggering
119power saving whenever empty poll count reaches a certain number.
120
121* Monitor
122   This power saving scheme will put the CPU into optimized power state and
123   monitor the Ethernet PMD RX descriptor address, waking the CPU up whenever
124   there's new traffic. Support for this scheme may not be available on all
125   platforms, and further limitations may apply (see below).
126
127* Pause
128   This power saving scheme will avoid busy polling by either entering
129   power-optimized sleep state with ``rte_power_pause()`` function, or, if it's
130   not supported by the underlying platform, use ``rte_pause()``.
131
132* Frequency scaling
133   This power saving scheme will use ``librte_power`` library functionality to
134   scale the core frequency up/down depending on traffic volume.
135   The reaction time of the frequency scaling mode is longer
136   than the pause and monitor mode.
137
138The "monitor" mode is only supported in the following configurations and scenarios:
139
140* On Linux* x86_64, `rte_power_monitor()` requires WAITPKG instruction set being
141  supported by the CPU, while `rte_power_monitor_multi()` requires WAITPKG and
142  RTM instruction sets being supported by the CPU. RTM instruction set may also
143  require booting the Linux with `tsx=on` command line parameter. Please refer
144  to your platform documentation for further information.
145
146* If ``rte_cpu_get_intrinsics_support()`` function indicates that
147  ``rte_power_monitor_multi()`` function is supported by the platform, then
148  monitoring multiple Ethernet Rx queues for traffic will be supported.
149
150* If ``rte_cpu_get_intrinsics_support()`` function indicates that only
151  ``rte_power_monitor()`` is supported by the platform, then monitoring will be
152  limited to a mapping of 1 core 1 queue (thus, each Rx queue will have to be
153  monitored from a different lcore).
154
155* If ``rte_cpu_get_intrinsics_support()`` function indicates that neither of the
156  two monitoring functions are supported, then monitor mode will not be supported.
157
158* Not all Ethernet drivers support monitoring, even if the underlying
159  platform may support the necessary CPU instructions. Please refer to
160  :doc:`../nics/overview` for more information.
161
162
163API Overview for Ethernet PMD Power Management
164~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
165
166* **Queue Enable**: Enable specific power scheme for certain queue/port/core.
167
168* **Queue Disable**: Disable power scheme for certain queue/port/core.
169
170* **Get Emptypoll Max**: Get the configured number of empty polls to wait before
171  entering sleep state.
172
173* **Set Emptypoll Max**: Set the number of empty polls to wait before entering
174  sleep state.
175
176* **Get Pause Duration**: Get the configured duration (microseconds) to be used
177  in the Pause callback.
178
179* **Set Pause Duration**: Set the duration of the pause (microseconds) used in
180  the Pause mode callback.
181
182* **Get Scaling Min Freq**: Get the configured minimum frequency (kHz) to be used
183  in Frequency Scaling mode.
184
185* **Set Scaling Min Freq**: Set the minimum frequency (kHz) to be used in Frequency
186  Scaling mode.
187
188* **Get Scaling Max Freq**: Get the configured maximum frequency (kHz) to be used
189  in Frequency Scaling mode.
190
191* **Set Scaling Max Freq**: Set the maximum frequency (kHz) to be used in Frequency
192  Scaling mode.
193
194Intel Uncore API
195----------------
196
197Abstract
198~~~~~~~~
199
200Uncore is a term used by Intel to describe the functions of a microprocessor
201that are not in the core, but which must be closely connected to the core
202to achieve high performance: L3 cache, on-die memory controller, etc.
203Significant power savings can be achieved by reducing the uncore frequency
204to its lowest value.
205
206The Linux kernel provides the driver "intel-uncore-frequency"
207to control the uncore frequency limits for x86 platform.
208The driver is available from kernel version 5.6 and above.
209Also CONFIG_INTEL_UNCORE_FREQ_CONTROL will need to be enabled in the kernel,
210which was added in 5.6.
211This manipulates the context of MSR 0x620,
212which sets min/max of the uncore for the SKU.
213
214API Overview for Intel Uncore
215~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
216
217Overview of each function in the Intel Uncore API,
218with explanation of what they do.
219Each function should not be called in the fast path.
220
221Uncore Power Init
222  Initialize uncore power, populate frequency array
223  and record original min & max for die on pkg.
224
225Uncore Power Exit
226  Exit uncore power, restoring original min & max for die on pkg.
227
228Get Uncore Power Freq
229  Get current uncore freq index for die on pkg.
230
231Set Uncore Power Freq
232  Set min & max uncore freq index for die on pkg
233  to specified index value (min and max will be the same).
234
235Uncore Power Max
236  Set min & max uncore freq to maximum frequency index for die on pkg
237  (min and max will be the same).
238
239Uncore Power Min
240  Set min & max uncore freq to minimum frequency index for die on pkg
241  (min and max will be the same).
242
243Get Num Freqs
244  Get the number of frequencies in the index array.
245
246Get Num Pkgs
247  Get the number of packages (CPU's) on the system.
248
249Get Num Dies
250  Get the number of die's on a given package.
251
252References
253----------
254
255*   The :doc:`../sample_app_ug/l3_forward_power_man`
256    chapter in the :doc:`../sample_app_ug/index` section.
257
258*   The :doc:`../sample_app_ug/vm_power_management`
259    chapter in the :doc:`../sample_app_ug/index` section.
260
261*   The :doc:`../nics/overview` chapter in the :doc:`../nics/index` section
262