1PERF-SCRIPT(1) perf Manual PERF-SCRIPT(1)
2
6 perf-script - Read perf.data (created by perf record) and display trace
7 output
8
10 perf script [<options>]
11 perf script [<options>] record <script> [<record-options>] <command>
12 perf script [<options>] report <script> [script-args]
13 perf script [<options>] <script> <required-script-args> [<record-options>] <command>
14 perf script [<options>] <top-script> [script-args]
15
17 This command reads the input file and displays the trace recorded.
18 There are several variants of perf script:
20 'perf script' to see a detailed trace of the workload that was
22 recorded.
23 You can also run a set of pre-canned scripts that aggregate and
25 summarize the raw trace data in various ways (the list of scripts is
26 available via 'perf script -l'). The following variants allow you to
27 record and run those scripts:
28 'perf script record <script> <command>' to record the events required
30 for 'perf script report'. <script> is the name displayed in the
31 output of 'perf script --list' i.e. the actual script name minus any
32 language extension. If <command> is not specified, the events are
33 recorded using the -a (system-wide) 'perf record' option.
34 'perf script report <script> [args]' to run and display the results
36 of <script>. <script> is the name displayed in the output of 'perf
37 script --list' i.e. the actual script name minus any language
38 extension. The perf.data output from a previous run of 'perf script
39 record <script>' is used and should be present for this command to
40 succeed. [args] refers to the (mainly optional) args expected by
41 the script.
42 'perf script <script> <required-script-args> <command>' to both
44 record the events required for <script> and to run the <script>
45 using 'live-mode' i.e. without writing anything to disk. <script>
46 is the name displayed in the output of 'perf script --list' i.e. the
47 actual script name minus any language extension. If <command> is
48 not specified, the events are recorded using the -a (system-wide)
49 'perf record' option. If <script> has any required args, they
50 should be specified before <command>. This mode doesn't allow for
51 optional script args to be specified; if optional script args are
52 desired, they can be specified using separate 'perf script record'
53 and 'perf script report' commands, with the stdout of the record step
54 piped to the stdin of the report script, using the '-o -' and '-i -'
55 options of the corresponding commands.
56 'perf script <top-script>' to both record the events required for
58 <top-script> and to run the <top-script> using 'live-mode'
59 i.e. without writing anything to disk. <top-script> is the name
60 displayed in the output of 'perf script --list' i.e. the actual
61 script name minus any language extension; a <top-script> is defined
62 as any script name ending with the string 'top'.
63 [<record-options>] can be passed to the record steps of 'perf script
65 record' and 'live-mode' variants; this isn't possible however for
66 <top-script> 'live-mode' or 'perf script report' variants.
67 See the 'SEE ALSO' section for links to language-specific
69 information on how to write and run your own trace scripts.
70
72 <command>...
73 Any command you can specify in a shell.
74 -D, --dump-raw-trace=
76 Display verbose dump of the trace data.
77 -L, --Latency=
79 Show latency attributes (irqs/preemption disabled, etc).
80 -l, --list=
82 Display a list of available trace scripts.
83 -s [lang], --script=
85 Process trace data with the given script ([lang]:script[.ext]). If
86 the string lang is specified in place of a script name, a list of
87 supported languages will be displayed instead.
88 -g, --gen-script=
90 Generate perf-script.[ext] starter script for given language, using
91 current perf.data.
92 -a
94 Force system-wide collection. Scripts run without a <command>
95 normally use -a by default, while scripts run with a <command>
96 normally don’t - this option allows the latter to be run in
97 system-wide mode.
98 -i, --input=
100 Input file name. (default: perf.data unless stdin is a fifo)
101 -d, --debug-mode
103 Do various checks like samples ordering and lost events.
104 -F, --fields
106 Comma separated list of fields to print. Options are: comm, tid,
107 pid, time, cpu, event, trace, ip, sym, dso, addr, symoff, srcline,
108 period, iregs, uregs, brstack, brstacksym, flags, brstackinsn,
109 brstackoff, insn, insnlen, synth, phys_addr. Field list can be
110 prepended with the type, trace, sw or hw, to indicate to which
111 event type the field list applies. e.g., -F sw:comm,tid,time,ip,sym
112 and -F trace:time,cpu,trace
113 perf script -F <fields>
115 is equivalent to:
117 perf script -F trace:<fields> -F sw:<fields> -F hw:<fields>
119 i.e., the specified fields apply to all event types if the type string
121 is not given.
122 In addition to overriding fields, it is also possible to add or remove
124 fields from the defaults. For example
125 -F -cpu,+insn
127 removes the cpu field and adds the insn field. Adding/removing fields
129 cannot be mixed with normal overriding.
130 The arguments are processed in the order received. A later usage can
132 reset a prior request. e.g.:
133 -F trace: -F comm,tid,time,ip,sym
135 The first -F suppresses trace events (field list is ""), but then the
137 second invocation sets the fields to comm,tid,time,ip,sym. In this case a
138 warning is given to the user:
139 "Overriding previous field request for all events."
141 Alternatively, consider the order:
143 -F comm,tid,time,ip,sym -F trace:
145 The first -F sets the fields for all events and the second -F
147 suppresses trace events. The user is given a warning message about
148 the override, and the result of the above is that only S/W and H/W
149 events are displayed with the given fields.
150 For the 'wildcard' option if a user selected field is invalid for an
152 event type, a message is displayed to the user that the option is
153 ignored for that type. For example:
154 $ perf script -F comm,tid,trace
156 'trace' not valid for hardware events. Ignoring.
157 'trace' not valid for software events. Ignoring.
158 Alternatively, if the type is given an invalid field is specified it
160 is an error. For example:
161 perf script -v -F sw:comm,tid,trace
163 'trace' not valid for software events.
164 At this point usage is displayed, and perf-script exits.
166 The flags field is synthesized and may have a value when Instruction
168 Trace decoding. The flags are "bcrosyiABEx" which stand for branch,
169 call, return, conditional, system, asynchronous, interrupt,
170 transaction abort, trace begin, trace end, and in transaction,
171 respectively. Known combinations of flags are printed more nicely e.g.
172 "call" for "bc", "return" for "br", "jcc" for "bo", "jmp" for "b",
173 "int" for "bci", "iret" for "bri", "syscall" for "bcs", "sysret" for "brs",
174 "async" for "by", "hw int" for "bcyi", "tx abrt" for "bA", "tr strt" for "bB",
175 "tr end" for "bE". However the "x" flag will be display separately in those
176 cases e.g. "jcc (x)" for a condition branch within a transaction.
177 The callindent field is synthesized and may have a value when
179 Instruction Trace decoding. For calls and returns, it will display the
180 name of the symbol indented with spaces to reflect the stack depth.
181 When doing instruction trace decoding insn and insnlen give the
183 instruction bytes and the instruction length of the current
184 instruction.
185 The synth field is used by synthesized events which may be created when
187 Instruction Trace decoding.
188 Finally, a user may not set fields to none for all event types.
190 i.e., -F "" is not allowed.
191 The brstack output includes branch related information with raw addresses using the
193 /v/v/v/v/cycles syntax in the following order:
194 FROM: branch source instruction
195 TO : branch target instruction
196 M/P/-: M=branch target mispredicted or branch direction was mispredicted, P=target predicted or direction predicted, -=not supported
197 X/- : X=branch inside a transactional region, -=not in transaction region or not supported
198 A/- : A=TSX abort entry, -=not aborted region or not supported
199 cycles
200 The brstacksym is identical to brstack, except that the FROM and TO addresses are printed in a symbolic form if possible.
202 When brstackinsn is specified the full assembler sequences of branch sequences for each sample
204 is printed. This is the full execution path leading to the sample. This is only supported when the
205 sample was recorded with perf record -b or -j any.
206 The brstackoff field will print an offset into a specific dso/binary.
208 -k, --vmlinux=<file>
210 vmlinux pathname
211 --kallsyms=<file>
213 kallsyms pathname
214 --symfs=<directory>
216 Look for files with symbols relative to this directory.
217 -G, --hide-call-graph
219 When printing symbols do not display call chain.
220 --stop-bt
222 Stop display of callgraph at these symbols
223 -C, --cpu
225 Only report samples for the list of CPUs provided. Multiple CPUs
226 can be provided as a comma-separated list with no space: 0,1.
227 Ranges of CPUs are specified with -: 0-2. Default is to report
228 samples on all CPUs.
229 -c, --comms=
231 Only display events for these comms. CSV that understands
232 file://filename entries.
233 --pid=
235 Only show events for given process ID (comma separated list).
236 --tid=
238 Only show events for given thread ID (comma separated list).
239 -I, --show-info
241 Display extended information about the perf.data file. This adds
242 information which may be very large and thus may clutter the
243 display. It currently includes: cpu and numa topology of the host
244 system. It can only be used with the perf script report mode.
245 --show-kernel-path
247 Try to resolve the path of [kernel.kallsyms]
248 --show-task-events Display task related events (e.g. FORK, COMM, EXIT).
250 --show-mmap-events Display mmap related events (e.g. MMAP, MMAP2).
252 --show-switch-events Display context switch events i.e. events of type
254 PERF_RECORD_SWITCH or PERF_RECORD_SWITCH_CPU_WIDE.
255 --demangle
257 Demangle symbol names to human readable form. It’s enabled by
258 default, disable with --no-demangle.
259 --demangle-kernel
261 Demangle kernel symbol names to human readable form (for C++
262 kernels).
263 --header Show perf.data header.
265 --header-only Show only perf.data header.
267 --itrace
269 Options for decoding instruction tracing data. The options are:
270 i synthesize instructions events
272 b synthesize branches events
273 c synthesize branches events (calls only)
274 r synthesize branches events (returns only)
275 x synthesize transactions events
276 w synthesize ptwrite events
277 p synthesize power events
278 e synthesize error events
279 d create a debug log
280 g synthesize a call chain (use with i or x)
281 l synthesize last branch entries (use with i or x)
282 s skip initial number of events
283 The default is all events i.e. the same as --itrace=ibxwpe
285 In addition, the period (default 100000) for instructions events
287 can be specified in units of:
288 i instructions
290 t ticks
291 ms milliseconds
292 us microseconds
293 ns nanoseconds (default)
294 Also the call chain size (default 16, max. 1024) for instructions or
296 transactions events can be specified.
297 Also the number of last branch entries (default 64, max. 1024) for
299 instructions or transactions events can be specified.
300 It is also possible to skip events generated (instructions, branches, transactions,
302 ptwrite, power) at the beginning. This is useful to ignore initialization code.
303 --itrace=i0nss1000000
305 skips the first million instructions.
307 To disable decoding entirely, use --no-itrace.
309 --full-source-path
311 Show the full path for source files for srcline output.
312 --max-stack
314 Set the stack depth limit when parsing the callchain, anything
315 beyond the specified depth will be ignored. This is a trade-off
316 between information loss and faster processing especially for
317 workloads that can have a very long callchain stack. Note that when
318 using the --itrace option the synthesized callchain size will
319 override this value if the synthesized callchain size is bigger.
320 Default: /proc/sys/kernel/perf_event_max_stack when present, 127 otherwise.
322 --ns
324 Use 9 decimal places when displaying time (i.e. show the
325 nanoseconds)
326 -f, --force
328 Don’t do ownership validation.
329 --time
331 Only analyze samples within given time window: <start>,<stop>.
332 Times have the format seconds.microseconds. If start is not given
333 (i.e., time string is ,x.y) then analysis starts at the beginning
334 of the file. If stop time is not given (i.e, time string is x.y,)
335 then analysis goes to end of file.
336 --max-blocks
338 Set the maximum number of program blocks to print with brstackasm
339 for each sample.
340 --per-event-dump
342 Create per event files with a "perf.data.EVENT.dump" name instead
343 of printing to stdout, useful, for instance, for generating
344 flamegraphs.
345 --inline
347 If a callgraph address belongs to an inlined function, the inline
348 stack will be printed. Each entry has function name and file/line.
349 Enabled by default, disable with --no-inline.
350
352 perf-record(1), perf-script-perl(1), perf-script-python(1)
353perf 06/18/2019 PERF-SCRIPT(1)