1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2010-2014 Intel Corporation.
4 L3 Forwarding with Power Management Sample Application
5 ======================================================
10 The L3 Forwarding with Power Management application is an example of power-aware packet processing using the DPDK.
11 The application is based on existing L3 Forwarding sample application,
12 with the power management algorithms to control the P-states and
13 C-states of the Intel processor via a power management library.
18 The application demonstrates the use of the Power libraries in the DPDK to implement packet forwarding.
19 The initialization and run-time paths are very similar to those of the :doc:`l3_forward`.
20 The main difference from the L3 Forwarding sample application is that this application introduces power-aware optimization algorithms
21 by leveraging the Power library to control P-state and C-state of processor based on packet load.
23 The DPDK includes poll-mode drivers to configure Intel NIC devices and their receive (Rx) and transmit (Tx) queues.
24 The design principle of this PMD is to access the Rx and Tx descriptors directly without any interrupts to quickly receive,
25 process and deliver packets in the user space.
27 In general, the DPDK executes an endless packet processing loop on dedicated IA cores that include the following steps:
29 * Retrieve input packets through the PMD to poll Rx queue
31 * Process each received packet or provide received packets to other processing cores through software queues
33 * Send pending output packets to Tx queue through the PMD
35 In this way, the PMD achieves better performance than a traditional interrupt-mode driver,
36 at the cost of keeping cores active and running at the highest frequency,
37 hence consuming the maximum power all the time.
38 However, during the period of processing light network traffic,
39 which happens regularly in communication infrastructure systems due to well-known "tidal effect",
40 the PMD is still busy waiting for network packets, which wastes a lot of power.
42 Processor performance states (P-states) are the capability of an Intel processor
43 to switch between different supported operating frequencies and voltages.
44 If configured correctly, according to system workload, this feature provides power savings.
45 CPUFreq is the infrastructure provided by the Linux* kernel to control the processor performance state capability.
46 CPUFreq supports a user space governor that enables setting frequency via manipulating the virtual file device from a user space application.
47 The Power library in the DPDK provides a set of APIs for manipulating a virtual file device to allow user space application
48 to set the CPUFreq governor and set the frequency of specific cores.
50 This application includes a P-state power management algorithm to generate a frequency hint to be sent to CPUFreq.
51 The algorithm uses the number of received and available Rx packets on recent polls to make a heuristic decision to scale frequency up/down.
52 Specifically, some thresholds are checked to see whether a specific core running a DPDK polling thread needs to increase frequency
53 a step up based on the near to full trend of polled Rx queues.
54 Also, it decreases frequency a step if packet processed per loop is far less than the expected threshold
55 or the thread's sleeping time exceeds a threshold.
57 C-States are also known as sleep states.
58 They 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.
59 However, 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).
60 Therefore, 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.
61 It 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
62 to fully realize the benefits of entering the C-state.
63 CPUIdle is the infrastructure provide by the Linux kernel to control the processor C-state capability.
64 Unlike CPUFreq, CPUIdle does not provide a mechanism that allows the application to change C-state.
65 It actually has its own heuristic algorithms in kernel space to select target C-state to enter by executing privileged instructions like HLT and MWAIT,
66 based on the speculative sleep duration of the core.
67 In this application, we introduce a heuristic algorithm that allows packet processing cores to sleep for a short period
68 if there is no Rx packet received on recent polls.
69 In this way, CPUIdle automatically forces the corresponding cores to enter deeper C-states
70 instead of always running to the C0 state waiting for packets.
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.
77 Compiling the Application
78 -------------------------
80 To compile the sample application see :doc:`compiling`.
82 The application is located in the ``l3fwd-power`` sub-directory.
84 Running the Application
85 -----------------------
87 The application has a number of command line options:
89 .. code-block:: console
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]
95 * -p PORTMASK: Hexadecimal bitmask of ports to configure
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.
100 * --config (port,queue,lcore)[,(port,queue,lcore)]: determines which queues from which ports are mapped to which cores.
102 * --enable-jumbo: optional, enables jumbo frames
104 * --max-pkt-len: optional, maximum packet length in decimal (64-9600)
106 * --no-numa: optional, disables numa awareness
108 * --empty-poll: Traffic Aware power management. See below for details
110 * --telemetry: Telemetry mode.
112 * --pmd-mgmt: PMD power management mode.
114 See :doc:`l3_forward` for details.
115 The L3fwd-power example reuses the L3fwd command line options.
120 The following sections provide some explanation of the sample application code.
121 As mentioned in the overview section,
122 the initialization and run-time paths are identical to those of the L3 forwarding application.
123 The following sections describe aspects that are specific to the L3 Forwarding with Power Management sample application.
125 Power Library Initialization
126 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
128 The Power library is initialized in the main routine.
129 It changes the P-state governor to userspace for specific cores that are under control.
130 The Timer library is also initialized and several timers are created later on,
131 responsible for checking if it needs to scale down frequency at run time by checking CPU utilization statistics.
135 Only the power management related initialization is shown.
139 int main(int argc, char **argv)
141 struct lcore_conf *qconf;
144 uint16_t queueid, portid;
147 uint32_t n_tx_queue, nb_lcores;
148 uint8_t nb_rx_queue, queue, socketid;
152 /* init RTE timer library to be used to initialize per-core timers */
154 rte_timer_subsystem_init();
159 /* per-core initialization */
161 for (lcore_id = 0; lcore_id < RTE_MAX_LCORE; lcore_id++) {
162 if (rte_lcore_is_enabled(lcore_id) == 0)
165 /* init power management library for a specified core */
167 ret = rte_power_init(lcore_id);
169 rte_exit(EXIT_FAILURE, "Power management library "
170 "initialization failed on core%d\n", lcore_id);
172 /* init timer structures for each enabled lcore */
174 rte_timer_init(&power_timers[lcore_id]);
176 hz = rte_get_hpet_hz();
178 rte_timer_reset(&power_timers[lcore_id], hz/TIMER_NUMBER_PER_SECOND, SINGLE, lcore_id, power_timer_cb, NULL);
186 Monitoring Loads of Rx Queues
187 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
189 In general, the polling nature of the DPDK prevents the OS power management subsystem from knowing
190 if the network load is actually heavy or light.
191 In this sample, sampling network load work is done by monitoring received and
192 available descriptors on NIC Rx queues in recent polls.
193 Based on the number of returned and available Rx descriptors,
194 this example implements algorithms to generate frequency scaling hints and speculative sleep duration,
195 and use them to control P-state and C-state of processors via the power management library.
196 Frequency (P-state) control and sleep state (C-state) control work individually for each logical core,
197 and the combination of them contributes to a power efficient packet processing solution when serving light network loads.
199 The rte_eth_rx_burst() function and the newly-added rte_eth_rx_queue_count() function are used in the endless packet processing loop
200 to return the number of received and available Rx descriptors.
201 And those numbers of specific queue are passed to P-state and C-state heuristic algorithms
202 to generate hints based on recent network load trends.
206 Only power control related code is shown.
211 __rte_noreturn int main_loop(__rte_unused void *dummy)
219 * Read packet from RX queues
222 lcore_scaleup_hint = FREQ_CURRENT;
223 lcore_rx_idle_count = 0;
225 for (i = 0; i < qconf->n_rx_queue; ++i)
227 rx_queue = &(qconf->rx_queue_list[i]);
228 rx_queue->idle_hint = 0;
229 portid = rx_queue->port_id;
230 queueid = rx_queue->queue_id;
232 nb_rx = rte_eth_rx_burst(portid, queueid, pkts_burst, MAX_PKT_BURST);
233 stats[lcore_id].nb_rx_processed += nb_rx;
235 if (unlikely(nb_rx == 0)) {
237 * no packet received from rx queue, try to
238 * sleep for a while forcing CPU enter deeper
242 rx_queue->zero_rx_packet_count++;
244 if (rx_queue->zero_rx_packet_count <= MIN_ZERO_POLL_COUNT)
247 rx_queue->idle_hint = power_idle_heuristic(rx_queue->zero_rx_packet_count);
248 lcore_rx_idle_count++;
250 rx_ring_length = rte_eth_rx_queue_count(portid, queueid);
252 rx_queue->zero_rx_packet_count = 0;
255 * do not scale up frequency immediately as
256 * user to kernel space communication is costly
257 * which might impact packet I/O for received
261 rx_queue->freq_up_hint = power_freq_scaleup_heuristic(lcore_id, rx_ring_length);
264 /* Prefetch and forward packets */
269 if (likely(lcore_rx_idle_count != qconf->n_rx_queue)) {
270 for (i = 1, lcore_scaleup_hint = qconf->rx_queue_list[0].freq_up_hint; i < qconf->n_rx_queue; ++i) {
271 x_queue = &(qconf->rx_queue_list[i]);
273 if (rx_queue->freq_up_hint > lcore_scaleup_hint)
275 lcore_scaleup_hint = rx_queue->freq_up_hint;
278 if (lcore_scaleup_hint == FREQ_HIGHEST)
280 rte_power_freq_max(lcore_id);
282 else if (lcore_scaleup_hint == FREQ_HIGHER)
283 rte_power_freq_up(lcore_id);
286 * All Rx queues empty in recent consecutive polls,
287 * sleep in a conservative manner, meaning sleep as
291 for (i = 1, lcore_idle_hint = qconf->rx_queue_list[0].idle_hint; i < qconf->n_rx_queue; ++i) {
292 rx_queue = &(qconf->rx_queue_list[i]);
293 if (rx_queue->idle_hint < lcore_idle_hint)
294 lcore_idle_hint = rx_queue->idle_hint;
297 if ( lcore_idle_hint < SLEEP_GEAR1_THRESHOLD)
299 * execute "pause" instruction to avoid context
300 * switch for short sleep.
302 rte_delay_us(lcore_idle_hint);
304 /* long sleep force ruining thread to suspend */
305 usleep(lcore_idle_hint);
307 stats[lcore_id].sleep_time += lcore_idle_hint;
312 P-State Heuristic Algorithm
313 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
315 The power_freq_scaleup_heuristic() function is responsible for generating a frequency hint for the specified logical core
316 according to available descriptor number returned from rte_eth_rx_queue_count().
317 On every poll for new packets, the length of available descriptor on an Rx queue is evaluated,
318 and the algorithm used for frequency hinting is as follows:
320 * If the size of available descriptors exceeds 96, the maximum frequency is hinted.
322 * If the size of available descriptors exceeds 64, a trend counter is incremented by 100.
324 * If the length of the ring exceeds 32, the trend counter is incremented by 1.
326 * When the trend counter reached 10000 the frequency hint is changed to the next higher frequency.
330 The assumption is that the Rx queue size is 128 and the thresholds specified above
331 must be adjusted accordingly based on actual hardware Rx queue size,
332 which are configured via the rte_eth_rx_queue_setup() function.
334 In general, a thread needs to poll packets from multiple Rx queues.
335 Most likely, different queue have different load, so they would return different frequency hints.
336 The algorithm evaluates all the hints and then scales up frequency in an aggressive manner
337 by scaling up to highest frequency as long as one Rx queue requires.
338 In this way, we can minimize any negative performance impact.
340 On the other hand, frequency scaling down is controlled in the timer callback function.
341 Specifically, if the sleep times of a logical core indicate that it is sleeping more than 25% of the sampling period,
342 or if the average packet per iteration is less than expectation, the frequency is decreased by one step.
344 C-State Heuristic Algorithm
345 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
347 Whenever recent rte_eth_rx_burst() polls return 5 consecutive zero packets,
348 an idle counter begins incrementing for each successive zero poll.
349 At the same time, the function power_idle_heuristic() is called to generate speculative sleep duration
350 in order to force logical to enter deeper sleeping C-state.
351 There is no way to control C- state directly, and the CPUIdle subsystem in OS is intelligent enough
352 to select C-state to enter based on actual sleep period time of giving logical core.
353 The algorithm has the following sleeping behavior depending on the idle counter:
355 * If idle count less than 100, the counter value is used as a microsecond sleep value through rte_delay_us()
356 which execute pause instructions to avoid costly context switch but saving power at the same time.
358 * If idle count is between 100 and 999, a fixed sleep interval of 100 μs is used.
359 A 100 μs sleep interval allows the core to enter the C1 state while keeping a fast response time in case new traffic arrives.
361 * If idle count is greater than 1000, a fixed sleep value of 1 ms is used until the next timer expiration is used.
362 This allows the core to enter the C3/C6 states.
366 The thresholds specified above need to be adjusted for different Intel processors and traffic profiles.
368 If a thread polls multiple Rx queues and different queue returns different sleep duration values,
369 the algorithm controls the sleep time in a conservative manner by sleeping for the least possible time
370 in order to avoid a potential performance impact.
373 -------------------------
374 Additionally, there is a traffic aware mode of operation called "Empty
375 Poll" where the number of empty polls can be monitored to keep track
376 of how busy the application is. Empty poll mode can be enabled by the
377 command line option --empty-poll.
379 See :doc:`Power Management<../prog_guide/power_man>` chapter in the DPDK Programmer's Guide for empty poll mode details.
381 .. code-block:: console
383 ./<build_dir>/examples/dpdk-l3fwd-power -l xxx -n 4 -a 0000:xx:00.0 -a 0000:xx:00.1 \
384 -- -p 0x3 -P --config="(0,0,xx),(1,0,xx)" --empty-poll="0,0,0" -l 14 -m 9 -h 1
388 --empty-poll: Enable the empty poll mode instead of original algorithm
390 --empty-poll="training_flag, med_threshold, high_threshold"
392 * ``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.
394 * ``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.
396 * ``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.
398 * -l : optional, set up the LOW power state frequency index
400 * -m : optional, set up the MED power state frequency index
402 * -h : optional, set up the HIGH power state frequency index
404 Empty Poll Mode Example Usage
405 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
406 To initially obtain the ideal thresholds for the system, the training
407 mode should be run first. This is achieved by running the l3fwd-power
408 app with the training flag set to “1”, and the other parameters set to
411 .. code-block:: console
413 ./<build_dir>/examples/dpdk-l3fwd-power -l 1-3 -- -p 0x0f --config="(0,0,2),(0,1,3)" --empty-poll "1,0,0" –P
415 This will run the training algorithm for x seconds on each core (cores 2
416 and 3), and then print out the recommended threshold values for those
417 cores. The thresholds should be very similar for each core.
419 .. code-block:: console
421 POWER: Bring up the Timer
422 POWER: set the power freq to MED
423 POWER: Low threshold is 230277
424 POWER: MED threshold is 335071
425 POWER: HIGH threshold is 523769
426 POWER: Training is Complete for 2
427 POWER: set the power freq to MED
428 POWER: Low threshold is 236814
429 POWER: MED threshold is 344567
430 POWER: HIGH threshold is 538580
431 POWER: Training is Complete for 3
433 Once the values have been measured for a particular system, the app can
434 then be started without the training mode so traffic can start immediately.
436 .. code-block:: console
438 ./<build_dir>/examples/dpdk-l3fwd-power -l 1-3 -- -p 0x0f --config="(0,0,2),(0,1,3)" --empty-poll "0,340000,540000" –P
443 The telemetry mode support for ``l3fwd-power`` is a standalone mode, in this mode
444 ``l3fwd-power`` does simple l3fwding along with calculating empty polls, full polls,
445 and busy percentage for each forwarding core. The aggregation of these
446 values of all cores is reported as application level telemetry to metric
447 library for every 500ms from the main core.
449 The busy percentage is calculated by recording the poll_count
450 and when the count reaches a defined value the total
451 cycles it took is measured and compared with minimum and maximum
452 reference cycles and accordingly busy rate is set to either 0% or
455 .. code-block:: console
457 ./<build_dir>/examples/dpdk-l3fwd-power --telemetry -l 1-3 -- -p 0x0f --config="(0,0,2),(0,1,3)" --telemetry
459 The new stats ``empty_poll`` , ``full_poll`` and ``busy_percent`` can be viewed by running the script
460 ``/usertools/dpdk-telemetry-client.py`` and selecting the menu option ``Send for global Metrics``.
462 PMD power management Mode
463 -------------------------
465 The PMD power management mode support for ``l3fwd-power`` is a standalone mode.
466 In this mode, ``l3fwd-power`` does simple l3fwding
467 along with enabling the power saving scheme on specific port/queue/lcore.
468 Main purpose for this mode is to demonstrate
469 how to use the PMD power management API.
471 .. code-block:: console
473 ./build/examples/dpdk-l3fwd-power -l 1-3 -- --pmd-mgmt -p 0x0f --config="(0,0,2),(0,1,3)"
475 PMD Power Management Mode
476 -------------------------
478 There is also a traffic-aware operating mode that,
479 instead of using explicit power management,
480 will use automatic PMD power management.
481 This mode is limited to one queue per core,
482 and has three available power management schemes:
485 This will use ``rte_power_monitor()`` function to enter
486 a power-optimized state (subject to platform support).
489 This will use ``rte_power_pause()`` or ``rte_pause()``
490 to avoid busy looping when there is no traffic.
493 This will use frequency scaling routines
494 available in the ``librte_power`` library.
496 See :doc:`Power Management<../prog_guide/power_man>` chapter
497 in the DPDK Programmer's Guide for more details on PMD power management.
499 .. code-block:: console
501 ./<build_dir>/examples/dpdk-l3fwd-power -l 1-3 -- -p 0x0f --config="(0,0,2),(0,1,3)" --pmd-mgmt=scale