1perf-record(1)
2==============
3
4NAME
5----
6perf-record - Run a command and record its profile into perf.data
7
8SYNOPSIS
9--------
10[verse]
11'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] <command>
12'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] -- <command> [<options>]
13
14DESCRIPTION
15-----------
16This command runs a command and gathers a performance counter profile
17from it, into perf.data - without displaying anything.
18
19This file can then be inspected later on, using 'perf report'.
20
21
22OPTIONS
23-------
24<command>...::
25	Any command you can specify in a shell.
26
27-e::
28--event=::
29	Select the PMU event. Selection can be:
30
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'.
42
43--filter=<filter>::
44        Event filter.
45
46-a::
47--all-cpus::
48        System-wide collection from all CPUs.
49
50-l::
51        Scale counter values.
52
53-p::
54--pid=::
55	Record events on existing process ID (comma separated list).
56
57-t::
58--tid=::
59        Record events on existing thread ID (comma separated list).
60
61-u::
62--uid=::
63        Record events in threads owned by uid. Name or number.
64
65-r::
66--realtime=::
67	Collect data with this RT SCHED_FIFO priority.
68
69-D::
70--no-delay::
71	Collect data without buffering.
72
73-c::
74--count=::
75	Event period to sample.
76
77-o::
78--output=::
79	Output file name.
80
81-i::
82--no-inherit::
83	Child tasks do not inherit counters.
84-F::
85--freq=::
86	Profile at this frequency.
87
88-m::
89--mmap-pages=::
90	Number of mmap data pages. Must be a power of two.
91
92-g::
93	Enables call-graph (stack chain/backtrace) recording.
94
95--call-graph::
96	Setup and enable call-graph (stack chain/backtrace) recording,
97	implies -g.
98
99	Allows specifying "fp" (frame pointer) or "dwarf"
100	(DWARF's CFI - Call Frame Information) as the method to collect
101	the information used to show the call graphs.
102
103	In some systems, where binaries are build with gcc
104	--fomit-frame-pointer, using the "fp" method will produce bogus
105	call graphs, using "dwarf", if available (perf tools linked to
106	the libunwind library) should be used instead.
107
108-q::
109--quiet::
110	Don't print any message, useful for scripting.
111
112-v::
113--verbose::
114	Be more verbose (show counter open errors, etc).
115
116-s::
117--stat::
118	Per thread counts.
119
120-d::
121--data::
122	Sample addresses.
123
124-T::
125--timestamp::
126	Sample timestamps. Use it with 'perf report -D' to see the timestamps,
127	for instance.
128
129-n::
130--no-samples::
131	Don't sample.
132
133-R::
134--raw-samples::
135Collect raw sample records from all opened counters (default for tracepoint counters).
136
137-C::
138--cpu::
139Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a
140comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
141In per-thread mode with inheritance mode on (default), samples are captured only when
142the thread executes on the designated CPUs. Default is to monitor all CPUs.
143
144-N::
145--no-buildid-cache::
146Do not update the builid cache. This saves some overhead in situations
147where the information in the perf.data file (which includes buildids)
148is sufficient.
149
150-G name,...::
151--cgroup name,...::
152monitor only in the container (cgroup) called "name". This option is available only
153in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
154container "name" are monitored when they run on the monitored CPUs. Multiple cgroups
155can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup
156to first event, second cgroup to second event and so on. It is possible to provide
157an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
158corresponding events, i.e., they always refer to events defined earlier on the command
159line.
160
161-b::
162--branch-any::
163Enable taken branch stack sampling. Any type of taken branch may be sampled.
164This is a shortcut for --branch-filter any. See --branch-filter for more infos.
165
166-j::
167--branch-filter::
168Enable taken branch stack sampling. Each sample captures a series of consecutive
169taken branches. The number of branches captured with each sample depends on the
170underlying hardware, the type of branches of interest, and the executed code.
171It is possible to select the types of branches captured by enabling filters. The
172following filters are defined:
173
174        - any:  any type of branches
175        - any_call: any function call or system call
176        - any_ret: any function return or system call return
177        - ind_call: any indirect branch
178        - u:  only when the branch target is at the user level
179        - k: only when the branch target is in the kernel
180        - hv: only when the target is at the hypervisor level
181
182+
183The option requires at least one branch type among any, any_call, any_ret, ind_call.
184The privilege levels may be omitted, in which case, the privilege levels of the associated
185event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege
186levels are subject to permissions.  When sampling on multiple events, branch stack sampling
187is enabled for all the sampling events. The sampled branch type is the same for all events.
188The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k
189Note that this feature may not be available on all processors.
190
191-W::
192--weight::
193Enable weightened sampling. An additional weight is recorded per sample and can be
194displayed with the weight and local_weight sort keys.  This currently works for TSX
195abort events and some memory events in precise mode on modern Intel CPUs.
196
197SEE ALSO
198--------
199linkperf:perf-stat[1], linkperf:perf-list[1]
200