1.. SPDX-License-Identifier: BSD-3-Clause 2 Copyright(c) 2010-2014 Intel Corporation. 3 4L3 Forwarding with Power Management Sample Application 5====================================================== 6 7Introduction 8------------ 9 10The L3 Forwarding with Power Management application is an example of power-aware packet processing using the DPDK. 11The application is based on existing L3 Forwarding sample application, 12with the power management algorithms to control the P-states and 13C-states of the Intel processor via a power management library. 14 15Overview 16-------- 17 18The application demonstrates the use of the Power libraries in the DPDK to implement packet forwarding. 19The initialization and run-time paths are very similar to those of the :doc:`l3_forward`. 20The main difference from the L3 Forwarding sample application is that this application introduces power-aware optimization algorithms 21by leveraging the Power library to control P-state and C-state of processor based on packet load. 22 23The DPDK includes poll-mode drivers to configure Intel NIC devices and their receive (Rx) and transmit (Tx) queues. 24The design principle of this PMD is to access the Rx and Tx descriptors directly without any interrupts to quickly receive, 25process and deliver packets in the user space. 26 27In general, the DPDK executes an endless packet processing loop on dedicated IA cores that include the following steps: 28 29* Retrieve input packets through the PMD to poll Rx queue 30 31* Process each received packet or provide received packets to other processing cores through software queues 32 33* Send pending output packets to Tx queue through the PMD 34 35In this way, the PMD achieves better performance than a traditional interrupt-mode driver, 36at the cost of keeping cores active and running at the highest frequency, 37hence consuming the maximum power all the time. 38However, during the period of processing light network traffic, 39which happens regularly in communication infrastructure systems due to well-known "tidal effect", 40the PMD is still busy waiting for network packets, which wastes a lot of power. 41 42Processor performance states (P-states) are the capability of an Intel processor 43to switch between different supported operating frequencies and voltages. 44If configured correctly, according to system workload, this feature provides power savings. 45CPUFreq is the infrastructure provided by the Linux* kernel to control the processor performance state capability. 46CPUFreq supports a user space governor that enables setting frequency via manipulating the virtual file device from a user space application. 47The Power library in the DPDK provides a set of APIs for manipulating a virtual file device to allow user space application 48to set the CPUFreq governor and set the frequency of specific cores. 49 50This application includes a P-state power management algorithm to generate a frequency hint to be sent to CPUFreq. 51The algorithm uses the number of received and available Rx packets on recent polls to make a heuristic decision to scale frequency up/down. 52Specifically, some thresholds are checked to see whether a specific core running a DPDK polling thread needs to increase frequency 53a step up based on the near to full trend of polled Rx queues. 54Also, it decreases frequency a step if packet processed per loop is far less than the expected threshold 55or the thread's sleeping time exceeds a threshold. 56 57C-States are also known as sleep states. 58They allow software to put an Intel core into a low power idle state from which it is possible to exit via an event, such as an interrupt. 59However, there is a tradeoff between the power consumed in the idle state and the time required to wake up from the idle state (exit latency). 60Therefore, as you go into deeper C-states, the power consumed is lower but the exit latency is increased. Each C-state has a target residency. 61It is essential that when entering into a C-state, the core remains in this C-state for at least as long as the target residency in order 62to fully realize the benefits of entering the C-state. 63CPUIdle is the infrastructure provide by the Linux kernel to control the processor C-state capability. 64Unlike CPUFreq, CPUIdle does not provide a mechanism that allows the application to change C-state. 65It actually has its own heuristic algorithms in kernel space to select target C-state to enter by executing privileged instructions like HLT and MWAIT, 66based on the speculative sleep duration of the core. 67In this application, we introduce a heuristic algorithm that allows packet processing cores to sleep for a short period 68if there is no Rx packet received on recent polls. 69In this way, CPUIdle automatically forces the corresponding cores to enter deeper C-states 70instead of always running to the C0 state waiting for packets. 71 72.. note:: 73 74 To fully demonstrate the power saving capability of using C-states, 75 it is recommended to enable deeper C3 and C6 states in the BIOS during system boot up. 76 77Compiling the Application 78------------------------- 79 80To compile the sample application see :doc:`compiling`. 81 82The application is located in the ``l3fwd-power`` sub-directory. 83 84Running the Application 85----------------------- 86 87The application has a number of command line options: 88 89.. code-block:: console 90 91 ./<build_dir>/examples/dpdk-l3fwd_power [EAL options] -- -p PORTMASK [-P] --config(port,queue,lcore)[,(port,queue,lcore)] [--enable-jumbo [--max-pkt-len PKTLEN]] [--no-numa] 92 93where, 94 95* -p PORTMASK: Hexadecimal bitmask of ports to configure 96 97* -P: Sets all ports to promiscuous mode so that packets are accepted regardless of the packet's Ethernet MAC destination address. 98 Without this option, only packets with the Ethernet MAC destination address set to the Ethernet address of the port are accepted. 99 100* --config (port,queue,lcore)[,(port,queue,lcore)]: determines which queues from which ports are mapped to which cores. 101 102* --enable-jumbo: optional, enables jumbo frames 103 104* --max-pkt-len: optional, maximum packet length in decimal (64-9600) 105 106* --no-numa: optional, disables numa awareness 107 108* --empty-poll: Traffic Aware power management. See below for details 109 110* --telemetry: Telemetry mode. 111 112* --pmd-mgmt: PMD power management mode. 113 114See :doc:`l3_forward` for details. 115The L3fwd-power example reuses the L3fwd command line options. 116 117Explanation 118----------- 119 120The following sections provide some explanation of the sample application code. 121As mentioned in the overview section, 122the initialization and run-time paths are identical to those of the L3 forwarding application. 123The following sections describe aspects that are specific to the L3 Forwarding with Power Management sample application. 124 125Power Library Initialization 126~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 127 128The Power library is initialized in the main routine. 129It changes the P-state governor to userspace for specific cores that are under control. 130The Timer library is also initialized and several timers are created later on, 131responsible for checking if it needs to scale down frequency at run time by checking CPU utilization statistics. 132 133.. note:: 134 135 Only the power management related initialization is shown. 136 137.. literalinclude:: ../../../examples/l3fwd-power/main.c 138 :language: c 139 :start-after: Power library initialized in the main routine. 8< 140 :end-before: >8 End of power library initialization. 141 142Monitoring Loads of Rx Queues 143~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 144 145In general, the polling nature of the DPDK prevents the OS power management subsystem from knowing 146if the network load is actually heavy or light. 147In this sample, sampling network load work is done by monitoring received and 148available descriptors on NIC Rx queues in recent polls. 149Based on the number of returned and available Rx descriptors, 150this example implements algorithms to generate frequency scaling hints and speculative sleep duration, 151and use them to control P-state and C-state of processors via the power management library. 152Frequency (P-state) control and sleep state (C-state) control work individually for each logical core, 153and the combination of them contributes to a power efficient packet processing solution when serving light network loads. 154 155The rte_eth_rx_burst() function and the newly-added rte_eth_rx_queue_count() function are used in the endless packet processing loop 156to return the number of received and available Rx descriptors. 157And those numbers of specific queue are passed to P-state and C-state heuristic algorithms 158to generate hints based on recent network load trends. 159 160.. note:: 161 162 Only power control related code is shown. 163 164.. literalinclude:: ../../../examples/l3fwd-power/main.c 165 :language: c 166 :start-after: Main processing loop. 8< 167 :end-before: >8 End of main processing loop. 168 169P-State Heuristic Algorithm 170~~~~~~~~~~~~~~~~~~~~~~~~~~~ 171 172The power_freq_scaleup_heuristic() function is responsible for generating a frequency hint for the specified logical core 173according to available descriptor number returned from rte_eth_rx_queue_count(). 174On every poll for new packets, the length of available descriptor on an Rx queue is evaluated, 175and the algorithm used for frequency hinting is as follows: 176 177* If the size of available descriptors exceeds 96, the maximum frequency is hinted. 178 179* If the size of available descriptors exceeds 64, a trend counter is incremented by 100. 180 181* If the length of the ring exceeds 32, the trend counter is incremented by 1. 182 183* When the trend counter reached 10000 the frequency hint is changed to the next higher frequency. 184 185.. note:: 186 187 The assumption is that the Rx queue size is 128 and the thresholds specified above 188 must be adjusted accordingly based on actual hardware Rx queue size, 189 which are configured via the rte_eth_rx_queue_setup() function. 190 191In general, a thread needs to poll packets from multiple Rx queues. 192Most likely, different queue have different load, so they would return different frequency hints. 193The algorithm evaluates all the hints and then scales up frequency in an aggressive manner 194by scaling up to highest frequency as long as one Rx queue requires. 195In this way, we can minimize any negative performance impact. 196 197On the other hand, frequency scaling down is controlled in the timer callback function. 198Specifically, if the sleep times of a logical core indicate that it is sleeping more than 25% of the sampling period, 199or if the average packet per iteration is less than expectation, the frequency is decreased by one step. 200 201C-State Heuristic Algorithm 202~~~~~~~~~~~~~~~~~~~~~~~~~~~ 203 204Whenever recent rte_eth_rx_burst() polls return 5 consecutive zero packets, 205an idle counter begins incrementing for each successive zero poll. 206At the same time, the function power_idle_heuristic() is called to generate speculative sleep duration 207in order to force logical to enter deeper sleeping C-state. 208There is no way to control C- state directly, and the CPUIdle subsystem in OS is intelligent enough 209to select C-state to enter based on actual sleep period time of giving logical core. 210The algorithm has the following sleeping behavior depending on the idle counter: 211 212* If idle count less than 100, the counter value is used as a microsecond sleep value through rte_delay_us() 213 which execute pause instructions to avoid costly context switch but saving power at the same time. 214 215* If idle count is between 100 and 999, a fixed sleep interval of 100 μs is used. 216 A 100 μs sleep interval allows the core to enter the C1 state while keeping a fast response time in case new traffic arrives. 217 218* If idle count is greater than 1000, a fixed sleep value of 1 ms is used until the next timer expiration is used. 219 This allows the core to enter the C3/C6 states. 220 221.. note:: 222 223 The thresholds specified above need to be adjusted for different Intel processors and traffic profiles. 224 225If a thread polls multiple Rx queues and different queue returns different sleep duration values, 226the algorithm controls the sleep time in a conservative manner by sleeping for the least possible time 227in order to avoid a potential performance impact. 228 229Empty Poll Mode 230------------------------- 231Additionally, there is a traffic aware mode of operation called "Empty 232Poll" where the number of empty polls can be monitored to keep track 233of how busy the application is. Empty poll mode can be enabled by the 234command line option --empty-poll. 235 236See :doc:`Power Management<../prog_guide/power_man>` chapter in the DPDK Programmer's Guide for empty poll mode details. 237 238.. code-block:: console 239 240 ./<build_dir>/examples/dpdk-l3fwd-power -l xxx -n 4 -a 0000:xx:00.0 -a 0000:xx:00.1 \ 241 -- -p 0x3 -P --config="(0,0,xx),(1,0,xx)" --empty-poll="0,0,0" -l 14 -m 9 -h 1 242 243Where, 244 245--empty-poll: Enable the empty poll mode instead of original algorithm 246 247--empty-poll="training_flag, med_threshold, high_threshold" 248 249* ``training_flag`` : optional, enable/disable training mode. Default value is 0. If the training_flag is set as 1(true), then the application will start in training mode and print out the trained threshold values. If the training_flag is set as 0(false), the application will start in normal mode, and will use either the default thresholds or those supplied on the command line. The trained threshold values are specific to the user’s system, may give a better power profile when compared to the default threshold values. 250 251* ``med_threshold`` : optional, sets the empty poll threshold of a modestly busy system state. If this is not supplied, the application will apply the default value of 350000. 252 253* ``high_threshold`` : optional, sets the empty poll threshold of a busy system state. If this is not supplied, the application will apply the default value of 580000. 254 255* -l : optional, set up the LOW power state frequency index 256 257* -m : optional, set up the MED power state frequency index 258 259* -h : optional, set up the HIGH power state frequency index 260 261Empty Poll Mode Example Usage 262~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 263To initially obtain the ideal thresholds for the system, the training 264mode should be run first. This is achieved by running the l3fwd-power 265app with the training flag set to “1”, and the other parameters set to 2660. 267 268.. code-block:: console 269 270 ./<build_dir>/examples/dpdk-l3fwd-power -l 1-3 -- -p 0x0f --config="(0,0,2),(0,1,3)" --empty-poll "1,0,0" –P 271 272This will run the training algorithm for x seconds on each core (cores 2 273and 3), and then print out the recommended threshold values for those 274cores. The thresholds should be very similar for each core. 275 276.. code-block:: console 277 278 POWER: Bring up the Timer 279 POWER: set the power freq to MED 280 POWER: Low threshold is 230277 281 POWER: MED threshold is 335071 282 POWER: HIGH threshold is 523769 283 POWER: Training is Complete for 2 284 POWER: set the power freq to MED 285 POWER: Low threshold is 236814 286 POWER: MED threshold is 344567 287 POWER: HIGH threshold is 538580 288 POWER: Training is Complete for 3 289 290Once the values have been measured for a particular system, the app can 291then be started without the training mode so traffic can start immediately. 292 293.. code-block:: console 294 295 ./<build_dir>/examples/dpdk-l3fwd-power -l 1-3 -- -p 0x0f --config="(0,0,2),(0,1,3)" --empty-poll "0,340000,540000" –P 296 297Telemetry Mode 298-------------- 299 300The telemetry mode support for ``l3fwd-power`` is a standalone mode, in this mode 301``l3fwd-power`` does simple l3fwding along with calculating empty polls, full polls, 302and busy percentage for each forwarding core. The aggregation of these 303values of all cores is reported as application level telemetry to metric 304library for every 500ms from the main core. 305 306The busy percentage is calculated by recording the poll_count 307and when the count reaches a defined value the total 308cycles it took is measured and compared with minimum and maximum 309reference cycles and accordingly busy rate is set to either 0% or 31050% or 100%. 311 312.. code-block:: console 313 314 ./<build_dir>/examples/dpdk-l3fwd-power --telemetry -l 1-3 -- -p 0x0f --config="(0,0,2),(0,1,3)" --telemetry 315 316The new stats ``empty_poll`` , ``full_poll`` and ``busy_percent`` can be viewed by running the script 317``/usertools/dpdk-telemetry-client.py`` and selecting the menu option ``Send for global Metrics``. 318 319PMD power management Mode 320------------------------- 321 322The PMD power management mode support for ``l3fwd-power`` is a standalone mode. 323In this mode, ``l3fwd-power`` does simple l3fwding 324along with enabling the power saving scheme on specific port/queue/lcore. 325Main purpose for this mode is to demonstrate 326how to use the PMD power management API. 327 328.. code-block:: console 329 330 ./build/examples/dpdk-l3fwd-power -l 1-3 -- --pmd-mgmt -p 0x0f --config="(0,0,2),(0,1,3)" 331 332PMD Power Management Mode 333------------------------- 334 335There is also a traffic-aware operating mode that, 336instead of using explicit power management, 337will use automatic PMD power management. 338This mode is limited to one queue per core, 339and has three available power management schemes: 340 341``monitor`` 342 This will use ``rte_power_monitor()`` function to enter 343 a power-optimized state (subject to platform support). 344 345``pause`` 346 This will use ``rte_power_pause()`` or ``rte_pause()`` 347 to avoid busy looping when there is no traffic. 348 349``scale`` 350 This will use frequency scaling routines 351 available in the ``librte_power`` library. 352 353See :doc:`Power Management<../prog_guide/power_man>` chapter 354in the DPDK Programmer's Guide for more details on PMD power management. 355 356.. code-block:: console 357 358 ./<build_dir>/examples/dpdk-l3fwd-power -l 1-3 -- -p 0x0f --config="(0,0,2),(0,1,3)" --pmd-mgmt=scale 359