1perf-stat(1)
2============
3
4NAME
5----
6perf-stat - Run a command and gather performance counter statistics
7
8SYNOPSIS
9--------
10[verse]
11'perf stat' [-e <EVENT> | --event=EVENT] [-a] <command>
12'perf stat' [-e <EVENT> | --event=EVENT] [-a] \-- <command> [<options>]
13'perf stat' [-e <EVENT> | --event=EVENT] [-a] record [-o file] \-- <command> [<options>]
14'perf stat' report [-i file]
15
16DESCRIPTION
17-----------
18This command runs a command and gathers performance counter statistics
19from it.
20
21
22OPTIONS
23-------
24<command>...::
25	Any command you can specify in a shell.
26
27record::
28	See STAT RECORD.
29
30report::
31	See STAT REPORT.
32
33-e::
34--event=::
35	Select the PMU event. Selection can be:
36
37	- a symbolic event name (use 'perf list' to list all events)
38
39	- a raw PMU event in the form of rN where N is a hexadecimal value
40	  that represents the raw register encoding with the layout of the
41	  event control registers as described by entries in
42	  /sys/bus/event_source/devices/cpu/format/*.
43
44        - a symbolic or raw PMU event followed by an optional colon
45	  and a list of event modifiers, e.g., cpu-cycles:p.  See the
46	  linkperf:perf-list[1] man page for details on event modifiers.
47
48	- a symbolically formed event like 'pmu/param1=0x3,param2/' where
49	  param1 and param2 are defined as formats for the PMU in
50	  /sys/bus/event_source/devices/<pmu>/format/*
51
52	  'percore' is a event qualifier that sums up the event counts for both
53	  hardware threads in a core. For example:
54	  perf stat -A -a -e cpu/event,percore=1/,otherevent ...
55
56	- a symbolically formed event like 'pmu/config=M,config1=N,config2=K/'
57	  where M, N, K are numbers (in decimal, hex, octal format).
58	  Acceptable values for each of 'config', 'config1' and 'config2'
59	  parameters are defined by corresponding entries in
60	  /sys/bus/event_source/devices/<pmu>/format/*
61
62	Note that the last two syntaxes support prefix and glob matching in
63	the PMU name to simplify creation of events across multiple instances
64	of the same type of PMU in large systems (e.g. memory controller PMUs).
65	Multiple PMU instances are typical for uncore PMUs, so the prefix
66	'uncore_' is also ignored when performing this match.
67
68
69-i::
70--no-inherit::
71        child tasks do not inherit counters
72-p::
73--pid=<pid>::
74        stat events on existing process id (comma separated list)
75
76-t::
77--tid=<tid>::
78        stat events on existing thread id (comma separated list)
79
80-b::
81--bpf-prog::
82        stat events on existing bpf program id (comma separated list),
83        requiring root rights. bpftool-prog could be used to find program
84        id all bpf programs in the system. For example:
85
86  # bpftool prog | head -n 1
87  17247: tracepoint  name sys_enter  tag 192d548b9d754067  gpl
88
89  # perf stat -e cycles,instructions --bpf-prog 17247 --timeout 1000
90
91   Performance counter stats for 'BPF program(s) 17247':
92
93             85,967      cycles
94             28,982      instructions              #    0.34  insn per cycle
95
96        1.102235068 seconds time elapsed
97
98--bpf-counters::
99	Use BPF programs to aggregate readings from perf_events.  This
100	allows multiple perf-stat sessions that are counting the same metric (cycles,
101	instructions, etc.) to share hardware counters.
102	To use BPF programs on common events by default, use
103	"perf config stat.bpf-counter-events=<list_of_events>".
104
105--bpf-attr-map::
106	With option "--bpf-counters", different perf-stat sessions share
107	information about shared BPF programs and maps via a pinned hashmap.
108	Use "--bpf-attr-map" to specify the path of this pinned hashmap.
109	The default path is /sys/fs/bpf/perf_attr_map.
110
111ifdef::HAVE_LIBPFM[]
112--pfm-events events::
113Select a PMU event using libpfm4 syntax (see http://perfmon2.sf.net)
114including support for event filters. For example '--pfm-events
115inst_retired:any_p:u:c=1:i'. More than one event can be passed to the
116option using the comma separator. Hardware events and generic hardware
117events cannot be mixed together. The latter must be used with the -e
118option. The -e option and this one can be mixed and matched.  Events
119can be grouped using the {} notation.
120endif::HAVE_LIBPFM[]
121
122-a::
123--all-cpus::
124        system-wide collection from all CPUs (default if no target is specified)
125
126--no-scale::
127	Don't scale/normalize counter values
128
129-d::
130--detailed::
131	print more detailed statistics, can be specified up to 3 times
132
133	   -d:          detailed events, L1 and LLC data cache
134        -d -d:     more detailed events, dTLB and iTLB events
135     -d -d -d:     very detailed events, adding prefetch events
136
137-r::
138--repeat=<n>::
139	repeat command and print average + stddev (max: 100). 0 means forever.
140
141-B::
142--big-num::
143        print large numbers with thousands' separators according to locale.
144	Enabled by default. Use "--no-big-num" to disable.
145	Default setting can be changed with "perf config stat.big-num=false".
146
147-C::
148--cpu=::
149Count only on the list of CPUs provided. Multiple CPUs can be provided as a
150comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
151In per-thread mode, this option is ignored. The -a option is still necessary
152to activate system-wide monitoring. Default is to count on all CPUs.
153
154-A::
155--no-aggr::
156Do not aggregate counts across all monitored CPUs.
157
158-n::
159--null::
160null run - Don't start any counters.
161
162This can be useful to measure just elapsed wall-clock time - or to assess the
163raw overhead of perf stat itself, without running any counters.
164
165-v::
166--verbose::
167        be more verbose (show counter open errors, etc)
168
169-x SEP::
170--field-separator SEP::
171print counts using a CSV-style output to make it easy to import directly into
172spreadsheets. Columns are separated by the string specified in SEP.
173
174--table:: Display time for each run (-r option), in a table format, e.g.:
175
176  $ perf stat --null -r 5 --table perf bench sched pipe
177
178   Performance counter stats for 'perf bench sched pipe' (5 runs):
179
180             # Table of individual measurements:
181             5.189 (-0.293) #
182             5.189 (-0.294) #
183             5.186 (-0.296) #
184             5.663 (+0.181) ##
185             6.186 (+0.703) ####
186
187             # Final result:
188             5.483 +- 0.198 seconds time elapsed  ( +-  3.62% )
189
190-G name::
191--cgroup name::
192monitor only in the container (cgroup) called "name". This option is available only
193in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
194container "name" are monitored when they run on the monitored CPUs. Multiple cgroups
195can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup
196to first event, second cgroup to second event and so on. It is possible to provide
197an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
198corresponding events, i.e., they always refer to events defined earlier on the command
199line. If the user wants to track multiple events for a specific cgroup, the user can
200use '-e e1 -e e2 -G foo,foo' or just use '-e e1 -e e2 -G foo'.
201
202If wanting to monitor, say, 'cycles' for a cgroup and also for system wide, this
203command line can be used: 'perf stat -e cycles -G cgroup_name -a -e cycles'.
204
205--for-each-cgroup name::
206Expand event list for each cgroup in "name" (allow multiple cgroups separated
207by comma).  It also support regex patterns to match multiple groups.  This has same
208effect that repeating -e option and -G option for each event x name.  This option
209cannot be used with -G/--cgroup option.
210
211-o file::
212--output file::
213Print the output into the designated file.
214
215--append::
216Append to the output file designated with the -o option. Ignored if -o is not specified.
217
218--log-fd::
219
220Log output to fd, instead of stderr.  Complementary to --output, and mutually exclusive
221with it.  --append may be used here.  Examples:
222     3>results  perf stat --log-fd 3          \-- $cmd
223     3>>results perf stat --log-fd 3 --append \-- $cmd
224
225--control=fifo:ctl-fifo[,ack-fifo]::
226--control=fd:ctl-fd[,ack-fd]::
227ctl-fifo / ack-fifo are opened and used as ctl-fd / ack-fd as follows.
228Listen on ctl-fd descriptor for command to control measurement ('enable': enable events,
229'disable': disable events). Measurements can be started with events disabled using
230--delay=-1 option. Optionally send control command completion ('ack\n') to ack-fd descriptor
231to synchronize with the controlling process. Example of bash shell script to enable and
232disable events during measurements:
233
234 #!/bin/bash
235
236 ctl_dir=/tmp/
237
238 ctl_fifo=${ctl_dir}perf_ctl.fifo
239 test -p ${ctl_fifo} && unlink ${ctl_fifo}
240 mkfifo ${ctl_fifo}
241 exec {ctl_fd}<>${ctl_fifo}
242
243 ctl_ack_fifo=${ctl_dir}perf_ctl_ack.fifo
244 test -p ${ctl_ack_fifo} && unlink ${ctl_ack_fifo}
245 mkfifo ${ctl_ack_fifo}
246 exec {ctl_fd_ack}<>${ctl_ack_fifo}
247
248 perf stat -D -1 -e cpu-cycles -a -I 1000       \
249           --control fd:${ctl_fd},${ctl_fd_ack} \
250           \-- sleep 30 &
251 perf_pid=$!
252
253 sleep 5  && echo 'enable' >&${ctl_fd} && read -u ${ctl_fd_ack} e1 && echo "enabled(${e1})"
254 sleep 10 && echo 'disable' >&${ctl_fd} && read -u ${ctl_fd_ack} d1 && echo "disabled(${d1})"
255
256 exec {ctl_fd_ack}>&-
257 unlink ${ctl_ack_fifo}
258
259 exec {ctl_fd}>&-
260 unlink ${ctl_fifo}
261
262 wait -n ${perf_pid}
263 exit $?
264
265
266--pre::
267--post::
268	Pre and post measurement hooks, e.g.:
269
270perf stat --repeat 10 --null --sync --pre 'make -s O=defconfig-build/clean' \-- make -s -j64 O=defconfig-build/ bzImage
271
272-I msecs::
273--interval-print msecs::
274Print count deltas every N milliseconds (minimum: 1ms)
275The overhead percentage could be high in some cases, for instance with small, sub 100ms intervals.  Use with caution.
276	example: 'perf stat -I 1000 -e cycles -a sleep 5'
277
278If the metric exists, it is calculated by the counts generated in this interval and the metric is printed after #.
279
280--interval-count times::
281Print count deltas for fixed number of times.
282This option should be used together with "-I" option.
283	example: 'perf stat -I 1000 --interval-count 2 -e cycles -a'
284
285--interval-clear::
286Clear the screen before next interval.
287
288--timeout msecs::
289Stop the 'perf stat' session and print count deltas after N milliseconds (minimum: 10 ms).
290This option is not supported with the "-I" option.
291	example: 'perf stat --time 2000 -e cycles -a'
292
293--metric-only::
294Only print computed metrics. Print them in a single line.
295Don't show any raw values. Not supported with --per-thread.
296
297--per-socket::
298Aggregate counts per processor socket for system-wide mode measurements.  This
299is a useful mode to detect imbalance between sockets.  To enable this mode,
300use --per-socket in addition to -a. (system-wide).  The output includes the
301socket number and the number of online processors on that socket. This is
302useful to gauge the amount of aggregation.
303
304--per-die::
305Aggregate counts per processor die for system-wide mode measurements.  This
306is a useful mode to detect imbalance between dies.  To enable this mode,
307use --per-die in addition to -a. (system-wide).  The output includes the
308die number and the number of online processors on that die. This is
309useful to gauge the amount of aggregation.
310
311--per-core::
312Aggregate counts per physical processor for system-wide mode measurements.  This
313is a useful mode to detect imbalance between physical cores.  To enable this mode,
314use --per-core in addition to -a. (system-wide).  The output includes the
315core number and the number of online logical processors on that physical processor.
316
317--per-thread::
318Aggregate counts per monitored threads, when monitoring threads (-t option)
319or processes (-p option).
320
321--per-node::
322Aggregate counts per NUMA nodes for system-wide mode measurements. This
323is a useful mode to detect imbalance between NUMA nodes. To enable this
324mode, use --per-node in addition to -a. (system-wide).
325
326-D msecs::
327--delay msecs::
328After starting the program, wait msecs before measuring (-1: start with events
329disabled). This is useful to filter out the startup phase of the program,
330which is often very different.
331
332-T::
333--transaction::
334
335Print statistics of transactional execution if supported.
336
337--metric-no-group::
338By default, events to compute a metric are placed in weak groups. The
339group tries to enforce scheduling all or none of the events. The
340--metric-no-group option places events outside of groups and may
341increase the chance of the event being scheduled - leading to more
342accuracy. However, as events may not be scheduled together accuracy
343for metrics like instructions per cycle can be lower - as both metrics
344may no longer be being measured at the same time.
345
346--metric-no-merge::
347By default metric events in different weak groups can be shared if one
348group contains all the events needed by another. In such cases one
349group will be eliminated reducing event multiplexing and making it so
350that certain groups of metrics sum to 100%. A downside to sharing a
351group is that the group may require multiplexing and so accuracy for a
352small group that need not have multiplexing is lowered. This option
353forbids the event merging logic from sharing events between groups and
354may be used to increase accuracy in this case.
355
356--quiet::
357Don't print output. This is useful with perf stat record below to only
358write data to the perf.data file.
359
360STAT RECORD
361-----------
362Stores stat data into perf data file.
363
364-o file::
365--output file::
366Output file name.
367
368STAT REPORT
369-----------
370Reads and reports stat data from perf data file.
371
372-i file::
373--input file::
374Input file name.
375
376--per-socket::
377Aggregate counts per processor socket for system-wide mode measurements.
378
379--per-die::
380Aggregate counts per processor die for system-wide mode measurements.
381
382--per-core::
383Aggregate counts per physical processor for system-wide mode measurements.
384
385-M::
386--metrics::
387Print metrics or metricgroups specified in a comma separated list.
388For a group all metrics from the group are added.
389The events from the metrics are automatically measured.
390See perf list output for the possible metrics and metricgroups.
391
392-A::
393--no-aggr::
394Do not aggregate counts across all monitored CPUs.
395
396--topdown::
397Print complete top-down metrics supported by the CPU. This allows to
398determine bottle necks in the CPU pipeline for CPU bound workloads,
399by breaking the cycles consumed down into frontend bound, backend bound,
400bad speculation and retiring.
401
402Frontend bound means that the CPU cannot fetch and decode instructions fast
403enough. Backend bound means that computation or memory access is the bottle
404neck. Bad Speculation means that the CPU wasted cycles due to branch
405mispredictions and similar issues. Retiring means that the CPU computed without
406an apparently bottleneck. The bottleneck is only the real bottleneck
407if the workload is actually bound by the CPU and not by something else.
408
409For best results it is usually a good idea to use it with interval
410mode like -I 1000, as the bottleneck of workloads can change often.
411
412This enables --metric-only, unless overridden with --no-metric-only.
413
414The following restrictions only apply to older Intel CPUs and Atom,
415on newer CPUs (IceLake and later) TopDown can be collected for any thread:
416
417The top down metrics are collected per core instead of per
418CPU thread. Per core mode is automatically enabled
419and -a (global monitoring) is needed, requiring root rights or
420perf.perf_event_paranoid=-1.
421
422Topdown uses the full Performance Monitoring Unit, and needs
423disabling of the NMI watchdog (as root):
424echo 0 > /proc/sys/kernel/nmi_watchdog
425for best results. Otherwise the bottlenecks may be inconsistent
426on workload with changing phases.
427
428To interpret the results it is usually needed to know on which
429CPUs the workload runs on. If needed the CPUs can be forced using
430taskset.
431
432--td-level::
433Print the top-down statistics that equal to or lower than the input level.
434It allows users to print the interested top-down metrics level instead of
435the complete top-down metrics.
436
437The availability of the top-down metrics level depends on the hardware. For
438example, Ice Lake only supports L1 top-down metrics. The Sapphire Rapids
439supports both L1 and L2 top-down metrics.
440
441Default: 0 means the max level that the current hardware support.
442Error out if the input is higher than the supported max level.
443
444--no-merge::
445Do not merge results from same PMUs.
446
447When multiple events are created from a single event specification,
448stat will, by default, aggregate the event counts and show the result
449in a single row. This option disables that behavior and shows
450the individual events and counts.
451
452Multiple events are created from a single event specification when:
4531. Prefix or glob matching is used for the PMU name.
4542. Aliases, which are listed immediately after the Kernel PMU events
455   by perf list, are used.
456
457--hybrid-merge::
458Merge the hybrid event counts from all PMUs.
459
460For hybrid events, by default, the stat aggregates and reports the event
461counts per PMU. But sometimes, it's also useful to aggregate event counts
462from all PMUs. This option enables that behavior and reports the counts
463without PMUs.
464
465For non-hybrid events, it should be no effect.
466
467--smi-cost::
468Measure SMI cost if msr/aperf/ and msr/smi/ events are supported.
469
470During the measurement, the /sys/device/cpu/freeze_on_smi will be set to
471freeze core counters on SMI.
472The aperf counter will not be effected by the setting.
473The cost of SMI can be measured by (aperf - unhalted core cycles).
474
475In practice, the percentages of SMI cycles is very useful for performance
476oriented analysis. --metric_only will be applied by default.
477The output is SMI cycles%, equals to (aperf - unhalted core cycles) / aperf
478
479Users who wants to get the actual value can apply --no-metric-only.
480
481--all-kernel::
482Configure all used events to run in kernel space.
483
484--all-user::
485Configure all used events to run in user space.
486
487--percore-show-thread::
488The event modifier "percore" has supported to sum up the event counts
489for all hardware threads in a core and show the counts per core.
490
491This option with event modifier "percore" enabled also sums up the event
492counts for all hardware threads in a core but show the sum counts per
493hardware thread. This is essentially a replacement for the any bit and
494convenient for post processing.
495
496--summary::
497Print summary for interval mode (-I).
498
499--no-csv-summary::
500Don't print 'summary' at the first column for CVS summary output.
501This option must be used with -x and --summary.
502
503This option can be enabled in perf config by setting the variable
504'stat.no-csv-summary'.
505
506$ perf config stat.no-csv-summary=true
507
508--cputype::
509Only enable events on applying cpu with this type for hybrid platform
510(e.g. core or atom)"
511
512EXAMPLES
513--------
514
515$ perf stat \-- make
516
517   Performance counter stats for 'make':
518
519        83723.452481      task-clock:u (msec)       #    1.004 CPUs utilized
520                   0      context-switches:u        #    0.000 K/sec
521                   0      cpu-migrations:u          #    0.000 K/sec
522           3,228,188      page-faults:u             #    0.039 M/sec
523     229,570,665,834      cycles:u                  #    2.742 GHz
524     313,163,853,778      instructions:u            #    1.36  insn per cycle
525      69,704,684,856      branches:u                #  832.559 M/sec
526       2,078,861,393      branch-misses:u           #    2.98% of all branches
527
528        83.409183620 seconds time elapsed
529
530        74.684747000 seconds user
531         8.739217000 seconds sys
532
533TIMINGS
534-------
535As displayed in the example above we can display 3 types of timings.
536We always display the time the counters were enabled/alive:
537
538        83.409183620 seconds time elapsed
539
540For workload sessions we also display time the workloads spent in
541user/system lands:
542
543        74.684747000 seconds user
544         8.739217000 seconds sys
545
546Those times are the very same as displayed by the 'time' tool.
547
548CSV FORMAT
549----------
550
551With -x, perf stat is able to output a not-quite-CSV format output
552Commas in the output are not put into "". To make it easy to parse
553it is recommended to use a different character like -x \;
554
555The fields are in this order:
556
557	- optional usec time stamp in fractions of second (with -I xxx)
558	- optional CPU, core, or socket identifier
559	- optional number of logical CPUs aggregated
560	- counter value
561	- unit of the counter value or empty
562	- event name
563	- run time of counter
564	- percentage of measurement time the counter was running
565	- optional variance if multiple values are collected with -r
566	- optional metric value
567	- optional unit of metric
568
569Additional metrics may be printed with all earlier fields being empty.
570
571include::intel-hybrid.txt[]
572
573SEE ALSO
574--------
575linkperf:perf-top[1], linkperf:perf-list[1]
576