git://bbs.cooldavid.org / net-next-2.6.git / blame
| Commit | Line | Data |
|---|---|---|
| e33e0a43 | 1 | perf-record(1) |
| c1c2365a | 2 | ============== |
| e33e0a43 IM |
3 | |
| 4 | NAME | |
| 5 | ---- | |
| 23ac9cbe | 6 | perf-record - Run a command and record its profile into perf.data |
| e33e0a43 IM |
7 | |
| 8 | SYNOPSIS | |
| 9 | -------- | |
| 10 | [verse] | |
| 11 | 'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] <command> | |
| 9e096753 | 12 | 'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] -- <command> [<options>] |
| e33e0a43 IM |
13 | |
| 14 | DESCRIPTION | |
| 15 | ----------- | |
| 16 | This command runs a command and gathers a performance counter profile | |
| 23ac9cbe | 17 | from it, into perf.data - without displaying anything. |
| e33e0a43 IM |
18 | |
| 19 | This file can then be inspected later on, using 'perf report'. | |
| 20 | ||
| 21 | ||
| 22 | OPTIONS | |
| 23 | ------- | |
| 24 | <command>...:: | |
| 25 | Any command you can specify in a shell. | |
| 26 | ||
| 27 | -e:: | |
| 28 | --event=:: | |
| 1b290d67 | 29 | Select the PMU event. Selection can be: |
| e33e0a43 | 30 | |
| 1b290d67 FW |
31 | - a symbolic event name (use 'perf list' to list all events) |
| 32 | ||
| 33 | - a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a | |
| 34 | hexadecimal event descriptor. | |
| 35 | ||
| 36 | - a hardware breakpoint event in the form of '\mem:addr[:access]' | |
| 37 | where addr is the address in memory you want to break in. | |
| 38 | Access is the memory access type (read, write, execute) it can | |
| 39 | be passed as follows: '\mem:addr[:[r][w][x]]'. | |
| 40 | If you want to profile read-write accesses in 0x1000, just set | |
| 41 | 'mem:0x1000:rw'. | |
| e33e0a43 | 42 | -a:: |
| 386c0b70 | 43 | System-wide collection. |
| e33e0a43 IM |
44 | |
| 45 | -l:: | |
| 386c0b70 ACM |
46 | Scale counter values. |
| 47 | ||
| 48 | -p:: | |
| 49 | --pid=:: | |
| 50 | Record events on existing pid. | |
| 51 | ||
| 52 | -r:: | |
| 53 | --realtime=:: | |
| 54 | Collect data with this RT SCHED_FIFO priority. | |
| 55 | -A:: | |
| 56 | --append:: | |
| 57 | Append to the output file to do incremental profiling. | |
| 58 | ||
| 59 | -f:: | |
| 60 | --force:: | |
| 7865e817 | 61 | Overwrite existing data file. (deprecated) |
| 386c0b70 ACM |
62 | |
| 63 | -c:: | |
| 64 | --count=:: | |
| 65 | Event period to sample. | |
| 66 | ||
| 67 | -o:: | |
| 68 | --output=:: | |
| 69 | Output file name. | |
| 70 | ||
| 71 | -i:: | |
| 2e6cdf99 SE |
72 | --no-inherit:: |
| 73 | Child tasks do not inherit counters. | |
| 386c0b70 ACM |
74 | -F:: |
| 75 | --freq=:: | |
| 76 | Profile at this frequency. | |
| 77 | ||
| 78 | -m:: | |
| 79 | --mmap-pages=:: | |
| 80 | Number of mmap data pages. | |
| 81 | ||
| 82 | -g:: | |
| 83 | --call-graph:: | |
| 84 | Do call-graph (stack chain/backtrace) recording. | |
| 85 | ||
| b44308f5 ACM |
86 | -q:: |
| 87 | --quiet:: | |
| 88 | Don't print any message, useful for scripting. | |
| 89 | ||
| 386c0b70 ACM |
90 | -v:: |
| 91 | --verbose:: | |
| 92 | Be more verbose (show counter open errors, etc). | |
| 93 | ||
| 94 | -s:: | |
| 95 | --stat:: | |
| 96 | Per thread counts. | |
| 97 | ||
| 98 | -d:: | |
| 99 | --data:: | |
| 100 | Sample addresses. | |
| 101 | ||
| 102 | -n:: | |
| 103 | --no-samples:: | |
| 104 | Don't sample. | |
| e33e0a43 | 105 | |
| ec7ba4ea FW |
106 | -R:: |
| 107 | --raw-samples:: | |
| bdef3b02 | 108 | Collect raw sample records from all opened counters (default for tracepoint counters). |
| ec7ba4ea | 109 | |
| c45c6ea2 SE |
110 | -C:: |
| 111 | --cpu:: | |
| 112 | Collect samples only on the list of cpus provided. Multiple CPUs can be provided as a | |
| 113 | comma-sperated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. | |
| 114 | In per-thread mode with inheritance mode on (default), samples are captured only when | |
| 115 | the thread executes on the designated CPUs. Default is to monitor all CPUs. | |
| 116 | ||
| a1ac1d3c SE |
117 | -N:: |
| 118 | --no-buildid-cache:: | |
| 119 | Do not update the builid cache. This saves some overhead in situations | |
| 120 | where the information in the perf.data file (which includes buildids) | |
| 121 | is sufficient. | |
| 122 | ||
| e33e0a43 IM |
123 | SEE ALSO |
| 124 | -------- | |
| 386b05e3 | 125 | linkperf:perf-stat[1], linkperf:perf-list[1] |