| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 1 | perf-stat(1) |
| Ingo Molnar | 6e6b754 | 2008-04-15 22:39:31 +0200 | [diff] [blame] | 2 | ============ |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 3 | |
| 4 | NAME |
| 5 | ---- |
| 6 | perf-stat - Run a command and gather performance counter statistics |
| 7 | |
| 8 | SYNOPSIS |
| 9 | -------- |
| 10 | [verse] |
| Shawn Bohrer | 8c20769 | 2010-11-30 19:57:19 -0600 | [diff] [blame] | 11 | 'perf stat' [-e <EVENT> | --event=EVENT] [-a] <command> |
| 12 | 'perf stat' [-e <EVENT> | --event=EVENT] [-a] -- <command> [<options>] |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 13 | |
| 14 | DESCRIPTION |
| 15 | ----------- |
| 16 | This command runs a command and gathers performance counter statistics |
| 17 | from it. |
| 18 | |
| 19 | |
| 20 | OPTIONS |
| 21 | ------- |
| 22 | <command>...:: |
| 23 | Any command you can specify in a shell. |
| 24 | |
| Ingo Molnar | 20c84e9 | 2009-06-04 16:33:00 +0200 | [diff] [blame] | 25 | |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 26 | -e:: |
| 27 | --event=:: |
| Thomas Gleixner | 386b05e | 2009-06-06 14:56:33 +0200 | [diff] [blame] | 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. |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 32 | |
| Ingo Molnar | 20c84e9 | 2009-06-04 16:33:00 +0200 | [diff] [blame] | 33 | -i:: |
| Stephane Eranian | 2e6cdf9 | 2010-05-12 10:40:01 +0200 | [diff] [blame] | 34 | --no-inherit:: |
| 35 | child tasks do not inherit counters |
| Ingo Molnar | 20c84e9 | 2009-06-04 16:33:00 +0200 | [diff] [blame] | 36 | -p:: |
| 37 | --pid=<pid>:: |
| Shawn Bohrer | 8c20769 | 2010-11-30 19:57:19 -0600 | [diff] [blame] | 38 | stat events on existing process id |
| 39 | |
| 40 | -t:: |
| 41 | --tid=<tid>:: |
| 42 | stat events on existing thread id |
| 43 | |
| Ingo Molnar | 20c84e9 | 2009-06-04 16:33:00 +0200 | [diff] [blame] | 44 | |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 45 | -a:: |
| Shawn Bohrer | 8c20769 | 2010-11-30 19:57:19 -0600 | [diff] [blame] | 46 | --all-cpus:: |
| 47 | system-wide collection from all CPUs |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 48 | |
| Brice Goglin | b26bc5a | 2009-08-07 10:18:39 +0200 | [diff] [blame] | 49 | -c:: |
| Shawn Bohrer | 8c20769 | 2010-11-30 19:57:19 -0600 | [diff] [blame] | 50 | --scale:: |
| 51 | scale/normalize counter values |
| 52 | |
| 53 | -r:: |
| 54 | --repeat=<n>:: |
| 55 | repeat command and print average + stddev (max: 100) |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 56 | |
| Stephane Eranian | 5af52b5 | 2010-05-18 15:00:01 +0200 | [diff] [blame] | 57 | -B:: |
| Shawn Bohrer | 8c20769 | 2010-11-30 19:57:19 -0600 | [diff] [blame] | 58 | --big-num:: |
| Stephane Eranian | 5af52b5 | 2010-05-18 15:00:01 +0200 | [diff] [blame] | 59 | print large numbers with thousands' separators according to locale |
| 60 | |
| Stephane Eranian | c45c6ea | 2010-05-28 12:00:01 +0200 | [diff] [blame] | 61 | -C:: |
| 62 | --cpu=:: |
| Shawn Bohrer | 8c20769 | 2010-11-30 19:57:19 -0600 | [diff] [blame] | 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. |
| Stephane Eranian | c45c6ea | 2010-05-28 12:00:01 +0200 | [diff] [blame] | 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. |
| 67 | |
| Stephane Eranian | f5b4a9c3 | 2010-11-16 11:05:01 +0200 | [diff] [blame] | 68 | -A:: |
| 69 | --no-aggr:: |
| 70 | Do not aggregate counts across all monitored CPUs in system-wide mode (-a). |
| 71 | This option is only valid in system-wide mode. |
| 72 | |
| Shawn Bohrer | 8c20769 | 2010-11-30 19:57:19 -0600 | [diff] [blame] | 73 | -n:: |
| 74 | --null:: |
| 75 | null run - don't start any counters |
| 76 | |
| 77 | -v:: |
| 78 | --verbose:: |
| 79 | be more verbose (show counter open errors, etc) |
| 80 | |
| Stephane Eranian | d7470b6 | 2010-12-01 18:49:05 +0200 | [diff] [blame^] | 81 | -x SEP:: |
| 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. |
| 85 | |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 86 | EXAMPLES |
| 87 | -------- |
| 88 | |
| Ingo Molnar | 20c84e9 | 2009-06-04 16:33:00 +0200 | [diff] [blame] | 89 | $ perf stat -- make -j |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 90 | |
| Ingo Molnar | 20c84e9 | 2009-06-04 16:33:00 +0200 | [diff] [blame] | 91 | Performance counter stats for 'make -j': |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 92 | |
| Ingo Molnar | 20c84e9 | 2009-06-04 16:33:00 +0200 | [diff] [blame] | 93 | 8117.370256 task clock ticks # 11.281 CPU utilization factor |
| 94 | 678 context switches # 0.000 M/sec |
| 95 | 133 CPU migrations # 0.000 M/sec |
| 96 | 235724 pagefaults # 0.029 M/sec |
| 97 | 24821162526 CPU cycles # 3057.784 M/sec |
| 98 | 18687303457 instructions # 2302.138 M/sec |
| 99 | 172158895 cache references # 21.209 M/sec |
| 100 | 27075259 cache misses # 3.335 M/sec |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 101 | |
| Ingo Molnar | 20c84e9 | 2009-06-04 16:33:00 +0200 | [diff] [blame] | 102 | Wall-clock time elapsed: 719.554352 msecs |
| Ingo Molnar | 1d8c8b2 | 2009-04-20 15:52:29 +0200 | [diff] [blame] | 103 | |
| 104 | SEE ALSO |
| 105 | -------- |
| Thomas Gleixner | 386b05e | 2009-06-06 14:56:33 +0200 | [diff] [blame] | 106 | linkperf:perf-top[1], linkperf:perf-list[1] |