perf-record.txt 3.7 KB
Newer Older
1
perf-record(1)
2
==============
3 4 5

NAME
----
6
perf-record - Run a command and record its profile into perf.data
7 8 9 10 11

SYNOPSIS
--------
[verse]
'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] <command>
12
'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] -- <command> [<options>]
13 14 15 16

DESCRIPTION
-----------
This command runs a command and gathers a performance counter profile
17
from it, into perf.data - without displaying anything.
18 19 20 21 22 23 24 25 26 27 28

This file can then be inspected later on, using 'perf report'.


OPTIONS
-------
<command>...::
	Any command you can specify in a shell.

-e::
--event=::
29
	Select the PMU event. Selection can be:
30

31 32 33 34 35 36 37 38 39 40 41
        - a symbolic event name	(use 'perf list' to list all events)

        - a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a
	  hexadecimal event descriptor.

        - a hardware breakpoint event in the form of '\mem:addr[:access]'
          where addr is the address in memory you want to break in.
          Access is the memory access type (read, write, execute) it can
          be passed as follows: '\mem:addr[:[r][w][x]]'.
          If you want to profile read-write accesses in 0x1000, just set
          'mem:0x1000:rw'.
42 43 44 45

--filter=<filter>::
        Event filter.

46
-a::
47 48
--all-cpus::
        System-wide collection from all CPUs.
49 50

-l::
51 52 53 54
        Scale counter values.

-p::
--pid=::
55
	Record events on existing process ID (comma separated list).
56 57 58

-t::
--tid=::
59
        Record events on existing thread ID (comma separated list).
60

61 62 63 64
-u::
--uid=::
        Record events in threads owned by uid. Name or number.

65 66 67
-r::
--realtime=::
	Collect data with this RT SCHED_FIFO priority.
68 69 70
-D::
--no-delay::
	Collect data without buffering.
71 72 73 74 75 76
-A::
--append::
	Append to the output file to do incremental profiling.

-f::
--force::
77
	Overwrite existing data file. (deprecated)
78 79 80 81 82 83 84 85 86 87

-c::
--count=::
	Event period to sample.

-o::
--output=::
	Output file name.

-i::
88 89
--no-inherit::
	Child tasks do not inherit counters.
90 91 92 93 94 95
-F::
--freq=::
	Profile at this frequency.

-m::
--mmap-pages=::
96
	Number of mmap data pages. Must be a power of two.
97 98 99 100 101

-g::
--call-graph::
	Do call-graph (stack chain/backtrace) recording.

102 103 104 105
-q::
--quiet::
	Don't print any message, useful for scripting.

106 107 108 109 110 111 112 113 114 115 116 117
-v::
--verbose::
	Be more verbose (show counter open errors, etc).

-s::
--stat::
	Per thread counts.

-d::
--data::
	Sample addresses.

118 119 120 121 122
-T::
--timestamp::
	Sample timestamps. Use it with 'perf report -D' to see the timestamps,
	for instance.

123 124 125
-n::
--no-samples::
	Don't sample.
126

127 128
-R::
--raw-samples::
129
Collect raw sample records from all opened counters (default for tracepoint counters).
130

131 132
-C::
--cpu::
133 134
Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a
comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
135 136 137
In per-thread mode with inheritance mode on (default), samples are captured only when
the thread executes on the designated CPUs. Default is to monitor all CPUs.

138 139 140 141 142 143
-N::
--no-buildid-cache::
Do not update the builid cache. This saves some overhead in situations
where the information in the perf.data file (which includes buildids)
is sufficient.

S
Stephane Eranian 已提交
144 145 146 147 148 149 150 151 152 153 154
-G name,...::
--cgroup name,...::
monitor only in the container (cgroup) called "name". This option is available only
in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
container "name" are monitored when they run on the monitored CPUs. Multiple cgroups
can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup
to first event, second cgroup to second event and so on. It is possible to provide
an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
corresponding events, i.e., they always refer to events defined earlier on the command
line.

155 156
SEE ALSO
--------
157
linkperf:perf-stat[1], linkperf:perf-list[1]