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 in the form of rN where N is a hexadecimal value
34 that represents the raw register encoding with the layout of the
35 event control registers as described by entries in
36 /sys/bus/event_source/devices/cpu/format/*.
38 - a symbolic or raw PMU event followed by an optional colon
39 and a list of event modifiers, e.g., cpu-cycles:p. See the
40 linkperf:perf-list[1] man page for details on event modifiers.
42 - a symbolically formed PMU event like 'pmu/param1=0x3,param2/' where
43 'param1', 'param2', etc are defined as formats for the PMU in
44 /sys/bus/event_source/devices/<pmu>/format/*.
46 - a symbolically formed event like 'pmu/config=M,config1=N,config3=K/'
48 where M, N, K are numbers (in decimal, hex, octal format). Acceptable
49 values for each of 'config', 'config1' and 'config2' are defined by
50 corresponding entries in /sys/bus/event_source/devices/<pmu>/format/*
51 param1 and param2 are defined as formats for the PMU in:
52 /sys/bus/event_source/devices/<pmu>/format/*
54 There are also some parameters which are not defined in .../<pmu>/format/*.
55 These params can be used to overload default config values per event.
56 Here are some common parameters:
57 - 'period': Set event sampling period
58 - 'freq': Set event sampling frequency
59 - 'time': Disable/enable time stamping. Acceptable values are 1 for
60 enabling time stamping. 0 for disabling time stamping.
62 - 'call-graph': Disable/enable callgraph. Acceptable str are "fp" for
63 FP mode, "dwarf" for DWARF mode, "lbr" for LBR mode and
64 "no" for disable callgraph.
65 - 'stack-size': user stack size for dwarf mode
66 - 'name' : User defined event name. Single quotes (') may be used to
67 escape symbols in the name from parsing by shell and tool
68 like this: name=\'CPU_CLK_UNHALTED.THREAD:cmask=0x1\'.
69 - 'aux-output': Generate AUX records instead of events. This requires
70 that an AUX area event is also provided.
71 - 'aux-sample-size': Set sample size for AUX area sampling. If the
72 '--aux-sample' option has been used, set aux-sample-size=0 to disable
73 AUX area sampling for the event.
75 See the linkperf:perf-list[1] man page for more parameters.
77 Note: If user explicitly sets options which conflict with the params,
78 the value set by the parameters will be overridden.
80 Also not defined in .../<pmu>/format/* are PMU driver specific
81 configuration parameters. Any configuration parameter preceded by
82 the letter '@' is not interpreted in user space and sent down directly
83 to the PMU driver. For example:
85 perf record -e some_event/@cfg1,@cfg2=config/ ...
87 will see 'cfg1' and 'cfg2=config' pushed to the PMU driver associated
88 with the event for further processing. There is no restriction on
89 what the configuration parameters are, as long as their semantic is
90 understood and supported by the PMU driver.
92 - a hardware breakpoint event in the form of '\mem:addr[/len][:access]'
93 where addr is the address in memory you want to break in.
94 Access is the memory access type (read, write, execute) it can
95 be passed as follows: '\mem:addr[:[r][w][x]]'. len is the range,
96 number of bytes from specified addr, which the breakpoint will cover.
97 If you want to profile read-write accesses in 0x1000, just set
99 If you want to profile write accesses in [0x1000~1008), just set
102 - a BPF source file (ending in .c) or a precompiled object file (ending
103 in .o) selects one or more BPF events.
104 The BPF program can attach to various perf events based on the ELF section
107 When processing a '.c' file, perf searches an installed LLVM to compile it
108 into an object file first. Optional clang options can be passed via the
109 '--clang-opt' command line option, e.g.:
111 perf record --clang-opt "-DLINUX_VERSION_CODE=0x50000" \
112 -e tests/bpf-script-example.c
114 Note: '--clang-opt' must be placed before '--event/-e'.
116 - a group of events surrounded by a pair of brace ("{event1,event2,...}").
117 Each event is separated by commas and the group should be quoted to
118 prevent the shell interpretation. You also need to use --group on
119 "perf report" to view group events together.
122 Event filter. This option should follow an event selector (-e) which
123 selects either tracepoint event(s) or a hardware trace PMU
124 (e.g. Intel PT or CoreSight).
128 In the case of tracepoints, multiple '--filter' options are combined
133 A hardware trace PMU advertises its ability to accept a number of
134 address filters by specifying a non-zero value in
135 /sys/bus/event_source/devices/<pmu>/nr_addr_filters.
137 Address filters have the format:
139 filter|start|stop|tracestop <start> [/ <size>] [@<file name>]
142 - 'filter': defines a region that will be traced.
143 - 'start': defines an address at which tracing will begin.
144 - 'stop': defines an address at which tracing will stop.
145 - 'tracestop': defines a region in which tracing will stop.
147 <file name> is the name of the object file, <start> is the offset to the
148 code to trace in that file, and <size> is the size of the region to
149 trace. 'start' and 'stop' filters need not specify a <size>.
151 If no object file is specified then the kernel is assumed, in which case
152 the start address must be a current kernel memory address.
154 <start> can also be specified by providing the name of a symbol. If the
155 symbol name is not unique, it can be disambiguated by inserting #n where
156 'n' selects the n'th symbol in address order. Alternately #0, #g or #G
157 select only a global symbol. <size> can also be specified by providing
158 the name of a symbol, in which case the size is calculated to the end
159 of that symbol. For 'filter' and 'tracestop' filters, if <size> is
160 omitted and <start> is a symbol, then the size is calculated to the end
163 If <size> is omitted and <start> is '*', then the start and size will
164 be calculated from the first and last symbols, i.e. to trace the whole
167 If symbol names (or '*') are provided, they must be surrounded by white
170 The filter passed to the kernel is not necessarily the same as entered.
171 To see the filter that is passed, use the -v option.
173 The kernel may not be able to configure a trace region if it is not
174 within a single mapping. MMAP events (or /proc/<pid>/maps) can be
175 examined to determine if that is a possibility.
177 Multiple filters can be separated with space or comma.
180 Don't record events issued by perf itself. This option should follow
181 an event selector (-e) which selects tracepoint event(s). It adds a
182 filter expression 'common_pid != $PERFPID' to filters. If other
183 '--filter' exists, the new filter expression will be combined with
188 System-wide collection from all CPUs (default if no target is specified).
192 Record events on existing process ID (comma separated list).
196 Record events on existing thread ID (comma separated list).
197 This option also disables inheritance by default. Enable it by adding
202 Record events in threads owned by uid. Name or number.
206 Collect data with this RT SCHED_FIFO priority.
209 Collect data without buffering.
213 Event period to sample.
221 Child tasks do not inherit counters.
225 Profile at this frequency. Use 'max' to use the currently maximum
226 allowed frequency, i.e. the value in the kernel.perf_event_max_sample_rate
227 sysctl. Will throttle down to the currently maximum allowed frequency.
231 Fail if the specified frequency can't be used.
235 Number of mmap data pages (must be a power of two) or size
236 specification with appended unit character - B/K/M/G. The
237 size is rounded up to have nearest pages power of two value.
238 Also, by adding a comma, the number of mmap pages for AUX
239 area tracing can be specified.
242 Put all events in a single event group. This precedes the --event
243 option and remains only for backward compatibility. See --event.
246 Enables call-graph (stack chain/backtrace) recording for both
247 kernel space and user space.
250 Setup and enable call-graph (stack chain/backtrace) recording,
251 implies -g. Default is "fp" (for user space).
253 The unwinding method used for kernel space is dependent on the
254 unwinder used by the active kernel configuration, i.e
255 CONFIG_UNWINDER_FRAME_POINTER (fp) or CONFIG_UNWINDER_ORC (orc)
257 Any option specified here controls the method used for user space.
259 Valid options are "fp" (frame pointer), "dwarf" (DWARF's CFI -
260 Call Frame Information) or "lbr" (Hardware Last Branch Record
263 In some systems, where binaries are build with gcc
264 --fomit-frame-pointer, using the "fp" method will produce bogus
265 call graphs, using "dwarf", if available (perf tools linked to
266 the libunwind or libdw library) should be used instead.
267 Using the "lbr" method doesn't require any compiler options. It
268 will produce call graphs from the hardware LBR registers. The
269 main limitation is that it is only available on new Intel
270 platforms, such as Haswell. It can only get user call chain. It
271 doesn't work with branch stack sampling at the same time.
273 When "dwarf" recording is used, perf also records (user) stack dump
274 when sampled. Default size of the stack dump is 8192 (bytes).
275 User can change the size by passing the size after comma like
276 "--call-graph dwarf,4096".
278 When "fp" recording is used, perf tries to save stack enties
279 up to the number specified in sysctl.kernel.perf_event_max_stack
280 by default. User can change the number by passing it after comma
281 like "--call-graph fp,32".
285 Don't print any message, useful for scripting.
289 Be more verbose (show counter open errors, etc).
293 Record per-thread event counts. Use it with 'perf report -T' to see
298 Record the sample virtual addresses.
301 Record the sample physical addresses.
304 Record the sampled data address data page size.
307 Record the sampled code address (ip) page size
311 Record the sample timestamps. Use it with 'perf report -D' to see the
312 timestamps, for instance.
316 Record the sample period.
319 Record the sample cpu.
321 --sample-identifier::
322 Record the sample identifier i.e. PERF_SAMPLE_IDENTIFIER bit set in
323 the sample_type member of the struct perf_event_attr argument to the
324 perf_event_open system call.
332 Collect raw sample records from all opened counters (default for tracepoint counters).
336 Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a
337 comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
338 In per-thread mode with inheritance mode on (default), samples are captured only when
339 the thread executes on the designated CPUs. Default is to monitor all CPUs.
343 Do not save the build ids of binaries in the perf.data files. This skips
344 post processing after recording, which sometimes makes the final step in
345 the recording process to take a long time, as it needs to process all
346 events looking for mmap records. The downside is that it can misresolve
347 symbols if the workload binaries used when recording get locally rebuilt
348 or upgraded, because the only key available in this case is the
349 pathname. You can also set the "record.build-id" config variable to
350 'skip to have this behaviour permanently.
354 Do not update the buildid cache. This saves some overhead in situations
355 where the information in the perf.data file (which includes buildids)
356 is sufficient. You can also set the "record.build-id" config variable to
357 'no-cache' to have the same effect.
361 monitor only in the container (cgroup) called "name". This option is available only
362 in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
363 container "name" are monitored when they run on the monitored CPUs. Multiple cgroups
364 can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup
365 to first event, second cgroup to second event and so on. It is possible to provide
366 an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
367 corresponding events, i.e., they always refer to events defined earlier on the command
368 line. If the user wants to track multiple events for a specific cgroup, the user can
369 use '-e e1 -e e2 -G foo,foo' or just use '-e e1 -e e2 -G foo'.
371 If wanting to monitor, say, 'cycles' for a cgroup and also for system wide, this
372 command line can be used: 'perf stat -e cycles -G cgroup_name -a -e cycles'.
376 Enable taken branch stack sampling. Any type of taken branch may be sampled.
377 This is a shortcut for --branch-filter any. See --branch-filter for more infos.
381 Enable taken branch stack sampling. Each sample captures a series of consecutive
382 taken branches. The number of branches captured with each sample depends on the
383 underlying hardware, the type of branches of interest, and the executed code.
384 It is possible to select the types of branches captured by enabling filters. The
385 following filters are defined:
387 - any: any type of branches
388 - any_call: any function call or system call
389 - any_ret: any function return or system call return
390 - ind_call: any indirect branch
391 - call: direct calls, including far (to/from kernel) calls
392 - u: only when the branch target is at the user level
393 - k: only when the branch target is in the kernel
394 - hv: only when the target is at the hypervisor level
395 - in_tx: only when the target is in a hardware transaction
396 - no_tx: only when the target is not in a hardware transaction
397 - abort_tx: only when the target is a hardware transaction abort
398 - cond: conditional branches
399 - save_type: save branch type during sampling in case binary is not available later
400 For the platforms with Intel Arch LBR support (12th-Gen+ client or
401 4th-Gen Xeon+ server), the save branch type is unconditionally enabled
402 when the taken branch stack sampling is enabled.
405 The option requires at least one branch type among any, any_call, any_ret, ind_call, cond.
406 The privilege levels may be omitted, in which case, the privilege levels of the associated
407 event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege
408 levels are subject to permissions. When sampling on multiple events, branch stack sampling
409 is enabled for all the sampling events. The sampled branch type is the same for all events.
410 The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k
411 Note that this feature may not be available on all processors.
414 Enable weightened sampling. An additional weight is recorded per sample and can be
415 displayed with the weight and local_weight sort keys. This currently works for TSX
416 abort events and some memory events in precise mode on modern Intel CPUs.
419 Record events of type PERF_RECORD_NAMESPACES. This enables 'cgroup_id' sort key.
422 Record events of type PERF_RECORD_CGROUP. This enables 'cgroup' sort key.
425 Record transaction flags for transaction related events.
428 Use per-thread mmaps. By default per-cpu mmaps are created. This option
429 overrides that and uses per-thread mmaps. A side-effect of that is that
430 inheritance is automatically disabled. --per-thread is ignored with a warning
431 if combined with -a or -C options.
435 After starting the program, wait msecs before measuring (-1: start with events
436 disabled). This is useful to filter out the startup phase of the program, which
437 is often very different.
441 Capture machine state (registers) at interrupt, i.e., on counter overflows for
442 each sample. List of captured registers depends on the architecture. This option
443 is off by default. It is possible to select the registers to sample using their
444 symbolic names, e.g. on x86, ax, si. To list the available registers use
445 --intr-regs=\?. To name registers, pass a comma separated list such as
446 --intr-regs=ax,bx. The list of register is architecture dependent.
449 Similar to -I, but capture user registers at sample time. To list the available
450 user registers use --user-regs=\?.
453 Record running and enabled time for read events (:S)
457 Sets the clock id to use for the various time fields in the perf_event_type
458 records. See clock_gettime(). In particular CLOCK_MONOTONIC and
459 CLOCK_MONOTONIC_RAW are supported, some events might also allow
460 CLOCK_BOOTTIME, CLOCK_REALTIME and CLOCK_TAI.
464 Select AUX area tracing Snapshot Mode. This option is valid only with an
465 AUX area tracing event. Optionally, certain snapshot capturing parameters
466 can be specified in a string that follows this option:
467 'e': take one last snapshot on exit; guarantees that there is at least one
468 snapshot in the output file;
469 <size>: if the PMU supports this, specify the desired snapshot size.
471 In Snapshot Mode trace data is captured only when signal SIGUSR2 is received
472 and on exit if the above 'e' option is given.
474 --aux-sample[=OPTIONS]::
475 Select AUX area sampling. At least one of the events selected by the -e option
476 must be an AUX area event. Samples on other events will be created containing
477 data from the AUX area. Optionally sample size may be specified, otherwise it
481 When processing pre-existing threads /proc/XXX/mmap, it may take a long time,
482 because the file may be huge. A time out is needed in such cases.
483 This option sets the time out limit. The default value is 500 ms.
486 Record context switch events i.e. events of type PERF_RECORD_SWITCH or
487 PERF_RECORD_SWITCH_CPU_WIDE. In some cases (e.g. Intel PT, CoreSight or Arm SPE)
488 switch events will be enabled automatically, which can be suppressed by
489 by the option --no-switch-events.
492 Path to clang binary to use for compiling BPF scriptlets.
493 (enabled when BPF support is on)
495 --clang-opt=OPTIONS::
496 Options passed to clang when compiling BPF scriptlets.
497 (enabled when BPF support is on)
500 Specify vmlinux path which has debuginfo.
501 (enabled when BPF prologue is on)
504 Record build-id of all DSOs regardless whether it's actually hit or not.
507 Record build ids in mmap2 events, disables build id cache (implies --no-buildid).
510 Use <n> control blocks in asynchronous (Posix AIO) trace writing mode (default: 1, max: 4).
511 Asynchronous mode is supported only when linking Perf tool with libc library
512 providing implementation for Posix AIO API.
515 Set affinity mask of trace reading thread according to the policy defined by 'mode' value:
516 node - thread affinity mask is set to NUMA node cpu mask of the processed mmap buffer
517 cpu - thread affinity mask is set to cpu of the processed mmap buffer
519 --mmap-flush=number::
521 Specify minimal number of bytes that is extracted from mmap data pages and
522 processed for output. One can specify the number using B/K/M/G suffixes.
524 The maximal allowed value is a quarter of the size of mmaped data pages.
526 The default option value is 1 byte which means that every time that the output
527 writing thread finds some new data in the mmaped buffer the data is extracted,
528 possibly compressed (-z) and written to the output, perf.data or pipe.
530 Larger data chunks are compressed more effectively in comparison to smaller
531 chunks so extraction of larger chunks from the mmap data pages is preferable
532 from the perspective of output size reduction.
534 Also at some cases executing less output write syscalls with bigger data size
535 can take less time than executing more output write syscalls with smaller data
536 size thus lowering runtime profiling overhead.
539 --compression-level[=n]::
540 Produce compressed trace using specified level n (default: 1 - fastest compression,
544 Configure all used events to run in kernel space.
547 Configure all used events to run in user space.
549 --kernel-callchains::
550 Collect callchains only from kernel space. I.e. this option sets
551 perf_event_attr.exclude_callchain_user to 1.
554 Collect callchains only from user space. I.e. this option sets
555 perf_event_attr.exclude_callchain_kernel to 1.
557 Don't use both --kernel-callchains and --user-callchains at the same time or no
558 callchains will be collected.
561 Append timestamp to output file name.
563 --timestamp-boundary::
564 Record timestamp boundary (time of first/last samples).
566 --switch-output[=mode]::
567 Generate multiple perf.data files, timestamp prefixed, switching to a new one
568 based on 'mode' value:
569 "signal" - when receiving a SIGUSR2 (default value) or
570 <size> - when reaching the size threshold, size is expected to
571 be a number with appended unit character - B/K/M/G
572 <time> - when reaching the time threshold, size is expected to
573 be a number with appended unit character - s/m/h/d
575 Note: the precision of the size threshold hugely depends
576 on your configuration - the number and size of your ring
577 buffers (-m). It is generally more precise for higher sizes
578 (like >5M), for lower values expect different sizes.
580 A possible use case is to, given an external event, slice the perf.data file
581 that gets then processed, possibly via a perf script, to decide if that
582 particular perf.data snapshot should be kept or not.
584 Implies --timestamp-filename, --no-buildid and --no-buildid-cache.
585 The reason for the latter two is to reduce the data file switching
586 overhead. You can still switch them on with:
588 --switch-output --no-no-buildid --no-no-buildid-cache
590 --switch-output-event::
591 Events that will cause the switch of the perf.data file, auto-selecting
592 --switch-output=signal, the results are similar as internally the side band
593 thread will also send a SIGUSR2 to the main one.
595 Uses the same syntax as --event, it will just not be recorded, serving only to
596 switch the perf.data file as soon as the --switch-output event is processed by
597 a separate sideband thread.
599 This sideband thread is also used to other purposes, like processing the
600 PERF_RECORD_BPF_EVENT records as they happen, asking the kernel for extra BPF
603 --switch-max-files=N::
605 When rotating perf.data with --switch-output, only keep N files.
608 Parse options then exit. --dry-run can be used to detect errors in cmdline
611 'perf record --dry-run -e' can act as a BPF script compiler if llvm.dump-obj
612 in config file is set to true.
615 Collect and synthesize given type of events (comma separated). Note that
616 this option controls the synthesis from the /proc filesystem which represent
617 task status for pre-existing threads.
619 Kernel (and some other) events are recorded regardless of the
620 choice in this option. For example, --synth=no would have MMAP events for
624 'task' - synthesize FORK and COMM events for each task
625 'mmap' - synthesize MMAP events for each process (implies 'task')
626 'cgroup' - synthesize CGROUP events for each cgroup
627 'all' - synthesize all events (default)
628 'no' - do not synthesize any of the above events
631 Instead of collecting non-sample events (for example, fork, comm, mmap) at
632 the beginning of record, collect them during finalizing an output file.
633 The collected non-sample events reflects the status of the system when
637 Makes all events use an overwritable ring buffer. An overwritable ring
638 buffer works like a flight recorder: when it gets full, the kernel will
639 overwrite the oldest records, that thus will never make it to the
642 When '--overwrite' and '--switch-output' are used perf records and drops
643 events until it receives a signal, meaning that something unusual was
644 detected that warrants taking a snapshot of the most current events,
645 those fitting in the ring buffer at that moment.
647 'overwrite' attribute can also be set or canceled for an event using
648 config terms. For example: 'cycles/overwrite/' and 'instructions/no-overwrite/'.
650 Implies --tail-synthesize.
653 Make a copy of /proc/kcore and place it into a directory with the perf data file.
656 Limit the sample data max size, <size> is expected to be a number with
657 appended unit character - B/K/M/G
659 --num-thread-synthesize::
660 The number of threads to run when synthesizing events for existing processes.
661 By default, the number of threads equals 1.
664 --pfm-events events::
665 Select a PMU event using libpfm4 syntax (see http://perfmon2.sf.net)
666 including support for event filters. For example '--pfm-events
667 inst_retired:any_p:u:c=1:i'. More than one event can be passed to the
668 option using the comma separator. Hardware events and generic hardware
669 events cannot be mixed together. The latter must be used with the -e
670 option. The -e option and this one can be mixed and matched. Events
671 can be grouped using the {} notation.
674 --control=fifo:ctl-fifo[,ack-fifo]::
675 --control=fd:ctl-fd[,ack-fd]::
676 ctl-fifo / ack-fifo are opened and used as ctl-fd / ack-fd as follows.
677 Listen on ctl-fd descriptor for command to control measurement.
680 'enable' : enable events
681 'disable' : disable events
682 'enable name' : enable event 'name'
683 'disable name' : disable event 'name'
684 'snapshot' : AUX area tracing snapshot).
685 'stop' : stop perf record
688 'evlist [-v|-g|-F] : display all events
689 -F Show just the sample frequency used for each event.
691 -g Show event group information.
693 Measurements can be started with events disabled using --delay=-1 option. Optionally
694 send control command completion ('ack\n') to ack-fd descriptor to synchronize with the
695 controlling process. Example of bash shell script to enable and disable events during
702 ctl_fifo=${ctl_dir}perf_ctl.fifo
703 test -p ${ctl_fifo} && unlink ${ctl_fifo}
705 exec {ctl_fd}<>${ctl_fifo}
707 ctl_ack_fifo=${ctl_dir}perf_ctl_ack.fifo
708 test -p ${ctl_ack_fifo} && unlink ${ctl_ack_fifo}
709 mkfifo ${ctl_ack_fifo}
710 exec {ctl_fd_ack}<>${ctl_ack_fifo}
712 perf record -D -1 -e cpu-cycles -a \
713 --control fd:${ctl_fd},${ctl_fd_ack} \
717 sleep 5 && echo 'enable' >&${ctl_fd} && read -u ${ctl_fd_ack} e1 && echo "enabled(${e1})"
718 sleep 10 && echo 'disable' >&${ctl_fd} && read -u ${ctl_fd_ack} d1 && echo "disabled(${d1})"
721 unlink ${ctl_ack_fifo}
730 Write collected trace data into several data files using parallel threads.
731 <spec> value can be user defined list of masks. Masks separated by colon
732 define CPUs to be monitored by a thread and affinity mask of that thread
733 is separated by slash:
735 <cpus mask 1>/<affinity mask 1>:<cpus mask 2>/<affinity mask 2>:...
737 CPUs or affinity masks must not overlap with other corresponding masks.
738 Invalid CPUs are ignored, but masks containing only invalid CPUs are not
741 For example user specification like the following:
745 specifies parallel threads layout that consists of two threads,
746 the first thread monitors CPUs 0 and 2-4 with the affinity mask 2-4,
747 the second monitors CPUs 1 and 5-7 with the affinity mask 5-7.
749 <spec> value can also be a string meaning predefined parallel threads
752 cpu - create new data streaming thread for every monitored cpu
753 core - create new thread to monitor CPUs grouped by a core
754 package - create new thread to monitor CPUs grouped by a package
755 numa - create new threed to monitor CPUs grouped by a NUMA domain
757 Predefined layouts can be used on systems with large number of CPUs in
758 order not to spawn multiple per-cpu streaming threads but still avoid LOST
759 events in data directory files. Option specified with no or empty value
760 defaults to CPU layout. Masks defined or provided by the option value are
761 filtered through the mask provided by -C option.
763 --debuginfod[=URLs]::
764 Specify debuginfod URL to be used when cacheing perf.data binaries,
765 it follows the same syntax as the DEBUGINFOD_URLS variable, like:
767 http://192.168.122.174:8002
769 If the URLs is not specified, the value of DEBUGINFOD_URLS
770 system environment variable is used.
773 Enable off-cpu profiling with BPF. The BPF program will collect
774 task scheduling information with (user) stacktrace and save them
775 as sample data of a software event named "offcpu-time". The
776 sample period will have the time the task slept in nanoseconds.
778 Note that BPF can collect stack traces using frame pointer ("fp")
779 only, as of now. So the applications built without the frame
780 pointer might see bogus addresses.
782 include::intel-hybrid.txt[]
786 linkperf:perf-stat[1], linkperf:perf-list[1], linkperf:perf-intel-pt[1]