6 perf-stat - Run a command and gather performance counter statistics
11 'perf stat' [-e <EVENT> | --event=EVENT] [-a] <command>
12 'perf stat' [-e <EVENT> | --event=EVENT] [-a] -- <command> [<options>]
16 This command runs a command and gathers performance counter statistics
23 Any command you can specify in a shell.
28 Select the PMU event. Selection can be a symbolic event name
29 (use 'perf list' to list all events) or a raw PMU
30 event (eventsel+umask) in the form of rNNN where NNN is a
31 hexadecimal event descriptor.
35 child tasks do not inherit counters
38 stat events on existing process id (comma separated list)
42 stat events on existing thread id (comma separated list)
47 system-wide collection from all CPUs
51 scale/normalize counter values
55 repeat command and print average + stddev (max: 100). 0 means forever.
59 print large numbers with thousands' separators according to locale
63 Count only on the list of CPUs provided. Multiple CPUs can be provided as a
64 comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
65 In per-thread mode, this option is ignored. The -a option is still necessary
66 to activate system-wide monitoring. Default is to count on all CPUs.
70 Do not aggregate counts across all monitored CPUs in system-wide mode (-a).
71 This option is only valid in system-wide mode.
75 null run - don't start any counters
79 be more verbose (show counter open errors, etc)
82 --field-separator SEP::
83 print counts using a CSV-style output to make it easy to import directly into
84 spreadsheets. Columns are separated by the string specified in SEP.
88 monitor only in the container (cgroup) called "name". This option is available only
89 in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
90 container "name" are monitored when they run on the monitored CPUs. Multiple cgroups
91 can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup
92 to first event, second cgroup to second event and so on. It is possible to provide
93 an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
94 corresponding events, i.e., they always refer to events defined earlier on the command
99 Print the output into the designated file.
102 Append to the output file designated with the -o option. Ignored if -o is not specified.
106 Log output to fd, instead of stderr. Complementary to --output, and mutually exclusive
107 with it. --append may be used here. Examples:
108 3>results perf stat --log-fd 3 -- $cmd
109 3>>results perf stat --log-fd 3 --append -- $cmd
113 Pre and post measurement hooks, e.g.:
115 perf stat --repeat 10 --null --sync --pre 'make -s O=defconfig-build/clean' -- make -s -j64 O=defconfig-build/ bzImage
118 --interval-print msecs::
119 Print count deltas every N milliseconds (minimum: 100ms)
120 example: perf stat -I 1000 -e cycles -a sleep 5
123 Aggregate counts per processor socket for system-wide mode measurements. This
124 is a useful mode to detect imbalance between sockets. To enable this mode,
125 use --per-socket in addition to -a. (system-wide). The output includes the
126 socket number and the number of online processors on that socket. This is
127 useful to gauge the amount of aggregation.
130 Aggregate counts per physical processor for system-wide mode measurements. This
131 is a useful mode to detect imbalance between physical cores. To enable this mode,
132 use --per-core in addition to -a. (system-wide). The output includes the
133 core number and the number of online logical processors on that physical processor.
137 After starting the program, wait msecs before measuring. This is useful to
138 filter out the startup phase of the program, which is often very different.
143 Print statistics of transactional execution if supported.
148 $ perf stat -- make -j
150 Performance counter stats for 'make -j':
152 8117.370256 task clock ticks # 11.281 CPU utilization factor
153 678 context switches # 0.000 M/sec
154 133 CPU migrations # 0.000 M/sec
155 235724 pagefaults # 0.029 M/sec
156 24821162526 CPU cycles # 3057.784 M/sec
157 18687303457 instructions # 2302.138 M/sec
158 172158895 cache references # 21.209 M/sec
159 27075259 cache misses # 3.335 M/sec
161 Wall-clock time elapsed: 719.554352 msecs
165 linkperf:perf-top[1], linkperf:perf-list[1]