6 perf-record - Run a command and record its profile into perf.data
11 'perf record' [-e <EVENT> | --event=EVENT] [-a] <command>
12 'perf record' [-e <EVENT> | --event=EVENT] [-a] -- <command> [<options>]
16 This command runs a command and gathers a performance counter profile
17 from it, into perf.data - without displaying anything.
19 This file can then be inspected later on, using 'perf report'.
25 Any command you can specify in a shell.
29 Select the PMU event. Selection can be:
31 - a symbolic event name (use 'perf list' to list all events)
33 - a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a
34 hexadecimal event descriptor.
36 - a symbolically formed PMU event like 'pmu/param1=0x3,param2/' where
37 'param1', 'param2', etc are defined as formats for the PMU in
38 /sys/bus/event_source/devices/<pmu>/format/*.
40 - a symbolically formed event like 'pmu/config=M,config1=N,config3=K/'
42 where M, N, K are numbers (in decimal, hex, octal format). Acceptable
43 values for each of 'config', 'config1' and 'config2' are defined by
44 corresponding entries in /sys/bus/event_source/devices/<pmu>/format/*
45 param1 and param2 are defined as formats for the PMU in:
46 /sys/bus/event_source/devices/<pmu>/format/*
48 There are also some parameters which are not defined in .../<pmu>/format/*.
49 These params can be used to overload default config values per event.
50 Here are some common parameters:
51 - 'period': Set event sampling period
52 - 'freq': Set event sampling frequency
53 - 'time': Disable/enable time stamping. Acceptable values are 1 for
54 enabling time stamping. 0 for disabling time stamping.
56 - 'call-graph': Disable/enable callgraph. Acceptable str are "fp" for
57 FP mode, "dwarf" for DWARF mode, "lbr" for LBR mode and
58 "no" for disable callgraph.
59 - 'stack-size': user stack size for dwarf mode
60 - 'name' : User defined event name. Single quotes (') may be used to
61 escape symbols in the name from parsing by shell and tool
62 like this: name=\'CPU_CLK_UNHALTED.THREAD:cmask=0x1\'.
64 See the linkperf:perf-list[1] man page for more parameters.
66 Note: If user explicitly sets options which conflict with the params,
67 the value set by the parameters will be overridden.
69 Also not defined in .../<pmu>/format/* are PMU driver specific
70 configuration parameters. Any configuration parameter preceded by
71 the letter '@' is not interpreted in user space and sent down directly
72 to the PMU driver. For example:
74 perf record -e some_event/@cfg1,@cfg2=config/ ...
76 will see 'cfg1' and 'cfg2=config' pushed to the PMU driver associated
77 with the event for further processing. There is no restriction on
78 what the configuration parameters are, as long as their semantic is
79 understood and supported by the PMU driver.
81 - a hardware breakpoint event in the form of '\mem:addr[/len][:access]'
82 where addr is the address in memory you want to break in.
83 Access is the memory access type (read, write, execute) it can
84 be passed as follows: '\mem:addr[:[r][w][x]]'. len is the range,
85 number of bytes from specified addr, which the breakpoint will cover.
86 If you want to profile read-write accesses in 0x1000, just set
88 If you want to profile write accesses in [0x1000~1008), just set
91 - a BPF source file (ending in .c) or a precompiled object file (ending
92 in .o) selects one or more BPF events.
93 The BPF program can attach to various perf events based on the ELF section
96 When processing a '.c' file, perf searches an installed LLVM to compile it
97 into an object file first. Optional clang options can be passed via the
98 '--clang-opt' command line option, e.g.:
100 perf record --clang-opt "-DLINUX_VERSION_CODE=0x50000" \
101 -e tests/bpf-script-example.c
103 Note: '--clang-opt' must be placed before '--event/-e'.
105 - a group of events surrounded by a pair of brace ("{event1,event2,...}").
106 Each event is separated by commas and the group should be quoted to
107 prevent the shell interpretation. You also need to use --group on
108 "perf report" to view group events together.
111 Event filter. This option should follow an event selector (-e) which
112 selects either tracepoint event(s) or a hardware trace PMU
113 (e.g. Intel PT or CoreSight).
117 In the case of tracepoints, multiple '--filter' options are combined
122 A hardware trace PMU advertises its ability to accept a number of
123 address filters by specifying a non-zero value in
124 /sys/bus/event_source/devices/<pmu>/nr_addr_filters.
126 Address filters have the format:
128 filter|start|stop|tracestop <start> [/ <size>] [@<file name>]
131 - 'filter': defines a region that will be traced.
132 - 'start': defines an address at which tracing will begin.
133 - 'stop': defines an address at which tracing will stop.
134 - 'tracestop': defines a region in which tracing will stop.
136 <file name> is the name of the object file, <start> is the offset to the
137 code to trace in that file, and <size> is the size of the region to
138 trace. 'start' and 'stop' filters need not specify a <size>.
140 If no object file is specified then the kernel is assumed, in which case
141 the start address must be a current kernel memory address.
143 <start> can also be specified by providing the name of a symbol. If the
144 symbol name is not unique, it can be disambiguated by inserting #n where
145 'n' selects the n'th symbol in address order. Alternately #0, #g or #G
146 select only a global symbol. <size> can also be specified by providing
147 the name of a symbol, in which case the size is calculated to the end
148 of that symbol. For 'filter' and 'tracestop' filters, if <size> is
149 omitted and <start> is a symbol, then the size is calculated to the end
152 If <size> is omitted and <start> is '*', then the start and size will
153 be calculated from the first and last symbols, i.e. to trace the whole
156 If symbol names (or '*') are provided, they must be surrounded by white
159 The filter passed to the kernel is not necessarily the same as entered.
160 To see the filter that is passed, use the -v option.
162 The kernel may not be able to configure a trace region if it is not
163 within a single mapping. MMAP events (or /proc/<pid>/maps) can be
164 examined to determine if that is a possibility.
166 Multiple filters can be separated with space or comma.
169 Don't record events issued by perf itself. This option should follow
170 an event selector (-e) which selects tracepoint event(s). It adds a
171 filter expression 'common_pid != $PERFPID' to filters. If other
172 '--filter' exists, the new filter expression will be combined with
177 System-wide collection from all CPUs (default if no target is specified).
181 Record events on existing process ID (comma separated list).
185 Record events on existing thread ID (comma separated list).
186 This option also disables inheritance by default. Enable it by adding
191 Record events in threads owned by uid. Name or number.
195 Collect data with this RT SCHED_FIFO priority.
198 Collect data without buffering.
202 Event period to sample.
210 Child tasks do not inherit counters.
214 Profile at this frequency. Use 'max' to use the currently maximum
215 allowed frequency, i.e. the value in the kernel.perf_event_max_sample_rate
216 sysctl. Will throttle down to the currently maximum allowed frequency.
220 Fail if the specified frequency can't be used.
224 Number of mmap data pages (must be a power of two) or size
225 specification with appended unit character - B/K/M/G. The
226 size is rounded up to have nearest pages power of two value.
227 Also, by adding a comma, the number of mmap pages for AUX
228 area tracing can be specified.
231 Put all events in a single event group. This precedes the --event
232 option and remains only for backward compatibility. See --event.
235 Enables call-graph (stack chain/backtrace) recording.
238 Setup and enable call-graph (stack chain/backtrace) recording,
239 implies -g. Default is "fp".
241 Allows specifying "fp" (frame pointer) or "dwarf"
242 (DWARF's CFI - Call Frame Information) or "lbr"
243 (Hardware Last Branch Record facility) as the method to collect
244 the information used to show the call graphs.
246 In some systems, where binaries are build with gcc
247 --fomit-frame-pointer, using the "fp" method will produce bogus
248 call graphs, using "dwarf", if available (perf tools linked to
249 the libunwind or libdw library) should be used instead.
250 Using the "lbr" method doesn't require any compiler options. It
251 will produce call graphs from the hardware LBR registers. The
252 main limitation is that it is only available on new Intel
253 platforms, such as Haswell. It can only get user call chain. It
254 doesn't work with branch stack sampling at the same time.
256 When "dwarf" recording is used, perf also records (user) stack dump
257 when sampled. Default size of the stack dump is 8192 (bytes).
258 User can change the size by passing the size after comma like
259 "--call-graph dwarf,4096".
263 Don't print any message, useful for scripting.
267 Be more verbose (show counter open errors, etc).
271 Record per-thread event counts. Use it with 'perf report -T' to see
276 Record the sample virtual addresses.
279 Record the sample physical addresses.
283 Record the sample timestamps. Use it with 'perf report -D' to see the
284 timestamps, for instance.
288 Record the sample period.
291 Record the sample cpu.
299 Collect raw sample records from all opened counters (default for tracepoint counters).
303 Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a
304 comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
305 In per-thread mode with inheritance mode on (default), samples are captured only when
306 the thread executes on the designated CPUs. Default is to monitor all CPUs.
310 Do not save the build ids of binaries in the perf.data files. This skips
311 post processing after recording, which sometimes makes the final step in
312 the recording process to take a long time, as it needs to process all
313 events looking for mmap records. The downside is that it can misresolve
314 symbols if the workload binaries used when recording get locally rebuilt
315 or upgraded, because the only key available in this case is the
316 pathname. You can also set the "record.build-id" config variable to
317 'skip to have this behaviour permanently.
321 Do not update the buildid cache. This saves some overhead in situations
322 where the information in the perf.data file (which includes buildids)
323 is sufficient. You can also set the "record.build-id" config variable to
324 'no-cache' to have the same effect.
328 monitor only in the container (cgroup) called "name". This option is available only
329 in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
330 container "name" are monitored when they run on the monitored CPUs. Multiple cgroups
331 can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup
332 to first event, second cgroup to second event and so on. It is possible to provide
333 an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
334 corresponding events, i.e., they always refer to events defined earlier on the command
335 line. If the user wants to track multiple events for a specific cgroup, the user can
336 use '-e e1 -e e2 -G foo,foo' or just use '-e e1 -e e2 -G foo'.
338 If wanting to monitor, say, 'cycles' for a cgroup and also for system wide, this
339 command line can be used: 'perf stat -e cycles -G cgroup_name -a -e cycles'.
343 Enable taken branch stack sampling. Any type of taken branch may be sampled.
344 This is a shortcut for --branch-filter any. See --branch-filter for more infos.
348 Enable taken branch stack sampling. Each sample captures a series of consecutive
349 taken branches. The number of branches captured with each sample depends on the
350 underlying hardware, the type of branches of interest, and the executed code.
351 It is possible to select the types of branches captured by enabling filters. The
352 following filters are defined:
354 - any: any type of branches
355 - any_call: any function call or system call
356 - any_ret: any function return or system call return
357 - ind_call: any indirect branch
358 - call: direct calls, including far (to/from kernel) calls
359 - u: only when the branch target is at the user level
360 - k: only when the branch target is in the kernel
361 - hv: only when the target is at the hypervisor level
362 - in_tx: only when the target is in a hardware transaction
363 - no_tx: only when the target is not in a hardware transaction
364 - abort_tx: only when the target is a hardware transaction abort
365 - cond: conditional branches
366 - save_type: save branch type during sampling in case binary is not available later
369 The option requires at least one branch type among any, any_call, any_ret, ind_call, cond.
370 The privilege levels may be omitted, in which case, the privilege levels of the associated
371 event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege
372 levels are subject to permissions. When sampling on multiple events, branch stack sampling
373 is enabled for all the sampling events. The sampled branch type is the same for all events.
374 The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k
375 Note that this feature may not be available on all processors.
378 Enable weightened sampling. An additional weight is recorded per sample and can be
379 displayed with the weight and local_weight sort keys. This currently works for TSX
380 abort events and some memory events in precise mode on modern Intel CPUs.
383 Record events of type PERF_RECORD_NAMESPACES.
386 Record transaction flags for transaction related events.
389 Use per-thread mmaps. By default per-cpu mmaps are created. This option
390 overrides that and uses per-thread mmaps. A side-effect of that is that
391 inheritance is automatically disabled. --per-thread is ignored with a warning
392 if combined with -a or -C options.
396 After starting the program, wait msecs before measuring. This is useful to
397 filter out the startup phase of the program, which is often very different.
401 Capture machine state (registers) at interrupt, i.e., on counter overflows for
402 each sample. List of captured registers depends on the architecture. This option
403 is off by default. It is possible to select the registers to sample using their
404 symbolic names, e.g. on x86, ax, si. To list the available registers use
405 --intr-regs=\?. To name registers, pass a comma separated list such as
406 --intr-regs=ax,bx. The list of register is architecture dependent.
409 Similar to -I, but capture user registers at sample time. To list the available
410 user registers use --user-regs=\?.
413 Record running and enabled time for read events (:S)
417 Sets the clock id to use for the various time fields in the perf_event_type
418 records. See clock_gettime(). In particular CLOCK_MONOTONIC and
419 CLOCK_MONOTONIC_RAW are supported, some events might also allow
420 CLOCK_BOOTTIME, CLOCK_REALTIME and CLOCK_TAI.
424 Select AUX area tracing Snapshot Mode. This option is valid only with an
425 AUX area tracing event. Optionally the number of bytes to capture per
426 snapshot can be specified. In Snapshot Mode, trace data is captured only when
427 signal SIGUSR2 is received.
430 When processing pre-existing threads /proc/XXX/mmap, it may take a long time,
431 because the file may be huge. A time out is needed in such cases.
432 This option sets the time out limit. The default value is 500 ms.
435 Record context switch events i.e. events of type PERF_RECORD_SWITCH or
436 PERF_RECORD_SWITCH_CPU_WIDE.
439 Path to clang binary to use for compiling BPF scriptlets.
440 (enabled when BPF support is on)
442 --clang-opt=OPTIONS::
443 Options passed to clang when compiling BPF scriptlets.
444 (enabled when BPF support is on)
447 Specify vmlinux path which has debuginfo.
448 (enabled when BPF prologue is on)
451 Record build-id of all DSOs regardless whether it's actually hit or not.
454 Use <n> control blocks in asynchronous (Posix AIO) trace writing mode (default: 1, max: 4).
455 Asynchronous mode is supported only when linking Perf tool with libc library
456 providing implementation for Posix AIO API.
459 Set affinity mask of trace reading thread according to the policy defined by 'mode' value:
460 node - thread affinity mask is set to NUMA node cpu mask of the processed mmap buffer
461 cpu - thread affinity mask is set to cpu of the processed mmap buffer
463 --mmap-flush=number::
465 Specify minimal number of bytes that is extracted from mmap data pages and
466 processed for output. One can specify the number using B/K/M/G suffixes.
468 The maximal allowed value is a quarter of the size of mmaped data pages.
470 The default option value is 1 byte which means that every time that the output
471 writing thread finds some new data in the mmaped buffer the data is extracted,
472 possibly compressed (-z) and written to the output, perf.data or pipe.
474 Larger data chunks are compressed more effectively in comparison to smaller
475 chunks so extraction of larger chunks from the mmap data pages is preferable
476 from the perspective of output size reduction.
478 Also at some cases executing less output write syscalls with bigger data size
479 can take less time than executing more output write syscalls with smaller data
480 size thus lowering runtime profiling overhead.
483 --compression-level[=n]::
484 Produce compressed trace using specified level n (default: 1 - fastest compression,
488 Configure all used events to run in kernel space.
491 Configure all used events to run in user space.
494 Append timestamp to output file name.
496 --timestamp-boundary::
497 Record timestamp boundary (time of first/last samples).
499 --switch-output[=mode]::
500 Generate multiple perf.data files, timestamp prefixed, switching to a new one
501 based on 'mode' value:
502 "signal" - when receiving a SIGUSR2 (default value) or
503 <size> - when reaching the size threshold, size is expected to
504 be a number with appended unit character - B/K/M/G
505 <time> - when reaching the time threshold, size is expected to
506 be a number with appended unit character - s/m/h/d
508 Note: the precision of the size threshold hugely depends
509 on your configuration - the number and size of your ring
510 buffers (-m). It is generally more precise for higher sizes
511 (like >5M), for lower values expect different sizes.
513 A possible use case is to, given an external event, slice the perf.data file
514 that gets then processed, possibly via a perf script, to decide if that
515 particular perf.data snapshot should be kept or not.
517 Implies --timestamp-filename, --no-buildid and --no-buildid-cache.
518 The reason for the latter two is to reduce the data file switching
519 overhead. You can still switch them on with:
521 --switch-output --no-no-buildid --no-no-buildid-cache
523 --switch-max-files=N::
525 When rotating perf.data with --switch-output, only keep N files.
528 Parse options then exit. --dry-run can be used to detect errors in cmdline
531 'perf record --dry-run -e' can act as a BPF script compiler if llvm.dump-obj
532 in config file is set to true.
535 Instead of collecting non-sample events (for example, fork, comm, mmap) at
536 the beginning of record, collect them during finalizing an output file.
537 The collected non-sample events reflects the status of the system when
541 Makes all events use an overwritable ring buffer. An overwritable ring
542 buffer works like a flight recorder: when it gets full, the kernel will
543 overwrite the oldest records, that thus will never make it to the
546 When '--overwrite' and '--switch-output' are used perf records and drops
547 events until it receives a signal, meaning that something unusual was
548 detected that warrants taking a snapshot of the most current events,
549 those fitting in the ring buffer at that moment.
551 'overwrite' attribute can also be set or canceled for an event using
552 config terms. For example: 'cycles/overwrite/' and 'instructions/no-overwrite/'.
554 Implies --tail-synthesize.
558 linkperf:perf-stat[1], linkperf:perf-list[1]
projects / linux-2.6-microblaze.git / blob