cachepc-linux

Fork of AMDESE/linux with modifications for CachePC side-channel attack
git clone https://git.sinitax.com/sinitax/cachepc-linux
Log | Files | Refs | README | LICENSE | sfeed.txt

perf-record.txt (29982B)


      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] [-a] <command>
     12'perf record' [-e <EVENT> | --event=EVENT] [-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 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/*.
     37
     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.
     41
     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/*.
     45
     46	- a symbolically formed event like 'pmu/config=M,config1=N,config3=K/'
     47
     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/*
     53
     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.
     61		    The default is 1.
     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.
     74
     75          See the linkperf:perf-list[1] man page for more parameters.
     76
     77	  Note: If user explicitly sets options which conflict with the params,
     78	  the value set by the parameters will be overridden.
     79
     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:
     84
     85	  perf record -e some_event/@cfg1,@cfg2=config/ ...
     86
     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.
     91
     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
     98          'mem:0x1000:rw'.
     99          If you want to profile write accesses in [0x1000~1008), just set
    100          'mem:0x1000/8:w'.
    101
    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
    105          names.
    106
    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.:
    110
    111            perf record --clang-opt "-DLINUX_VERSION_CODE=0x50000" \
    112                        -e tests/bpf-script-example.c
    113
    114          Note: '--clang-opt' must be placed before '--event/-e'.
    115
    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.
    120
    121--filter=<filter>::
    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).
    125
    126	- tracepoint filters
    127
    128	In the case of tracepoints, multiple '--filter' options are combined
    129	using '&&'.
    130
    131	- address filters
    132
    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.
    136
    137	Address filters have the format:
    138
    139	filter|start|stop|tracestop <start> [/ <size>] [@<file name>]
    140
    141	Where:
    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.
    146
    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>.
    150
    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.
    153
    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
    161	of that symbol.
    162
    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
    165	file.
    166
    167	If symbol names (or '*') are provided, they must be surrounded by white
    168	space.
    169
    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.
    172
    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.
    176
    177	Multiple filters can be separated with space or comma.
    178
    179--exclude-perf::
    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
    184	them by '&&'.
    185
    186-a::
    187--all-cpus::
    188        System-wide collection from all CPUs (default if no target is specified).
    189
    190-p::
    191--pid=::
    192	Record events on existing process ID (comma separated list).
    193
    194-t::
    195--tid=::
    196        Record events on existing thread ID (comma separated list).
    197        This option also disables inheritance by default.  Enable it by adding
    198        --inherit.
    199
    200-u::
    201--uid=::
    202        Record events in threads owned by uid. Name or number.
    203
    204-r::
    205--realtime=::
    206	Collect data with this RT SCHED_FIFO priority.
    207
    208--no-buffering::
    209	Collect data without buffering.
    210
    211-c::
    212--count=::
    213	Event period to sample.
    214
    215-o::
    216--output=::
    217	Output file name.
    218
    219-i::
    220--no-inherit::
    221	Child tasks do not inherit counters.
    222
    223-F::
    224--freq=::
    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.
    228	See --strict-freq.
    229
    230--strict-freq::
    231	Fail if the specified frequency can't be used.
    232
    233-m::
    234--mmap-pages=::
    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.
    240
    241--group::
    242	Put all events in a single event group.  This precedes the --event
    243	option and remains only for backward compatibility.  See --event.
    244
    245-g::
    246	Enables call-graph (stack chain/backtrace) recording for both
    247	kernel space and user space.
    248
    249--call-graph::
    250	Setup and enable call-graph (stack chain/backtrace) recording,
    251	implies -g.  Default is "fp" (for user space).
    252
    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)
    256
    257	Any option specified here controls the method used for user space.
    258
    259	Valid options are "fp" (frame pointer), "dwarf" (DWARF's CFI -
    260	Call Frame Information) or "lbr" (Hardware Last Branch Record
    261	facility).
    262
    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.
    272
    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".
    277
    278-q::
    279--quiet::
    280	Don't print any message, useful for scripting.
    281
    282-v::
    283--verbose::
    284	Be more verbose (show counter open errors, etc).
    285
    286-s::
    287--stat::
    288	Record per-thread event counts.  Use it with 'perf report -T' to see
    289	the values.
    290
    291-d::
    292--data::
    293	Record the sample virtual addresses.
    294
    295--phys-data::
    296	Record the sample physical addresses.
    297
    298--data-page-size::
    299	Record the sampled data address data page size.
    300
    301--code-page-size::
    302	Record the sampled code address (ip) page size
    303
    304-T::
    305--timestamp::
    306	Record the sample timestamps. Use it with 'perf report -D' to see the
    307	timestamps, for instance.
    308
    309-P::
    310--period::
    311	Record the sample period.
    312
    313--sample-cpu::
    314	Record the sample cpu.
    315
    316-n::
    317--no-samples::
    318	Don't sample.
    319
    320-R::
    321--raw-samples::
    322Collect raw sample records from all opened counters (default for tracepoint counters).
    323
    324-C::
    325--cpu::
    326Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a
    327comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
    328In per-thread mode with inheritance mode on (default), samples are captured only when
    329the thread executes on the designated CPUs. Default is to monitor all CPUs.
    330
    331-B::
    332--no-buildid::
    333Do not save the build ids of binaries in the perf.data files. This skips
    334post processing after recording, which sometimes makes the final step in
    335the recording process to take a long time, as it needs to process all
    336events looking for mmap records. The downside is that it can misresolve
    337symbols if the workload binaries used when recording get locally rebuilt
    338or upgraded, because the only key available in this case is the
    339pathname. You can also set the "record.build-id" config variable to
    340'skip to have this behaviour permanently.
    341
    342-N::
    343--no-buildid-cache::
    344Do not update the buildid cache. This saves some overhead in situations
    345where the information in the perf.data file (which includes buildids)
    346is sufficient.  You can also set the "record.build-id" config variable to
    347'no-cache' to have the same effect.
    348
    349-G name,...::
    350--cgroup name,...::
    351monitor only in the container (cgroup) called "name". This option is available only
    352in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
    353container "name" are monitored when they run on the monitored CPUs. Multiple cgroups
    354can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup
    355to first event, second cgroup to second event and so on. It is possible to provide
    356an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
    357corresponding events, i.e., they always refer to events defined earlier on the command
    358line. If the user wants to track multiple events for a specific cgroup, the user can
    359use '-e e1 -e e2 -G foo,foo' or just use '-e e1 -e e2 -G foo'.
    360
    361If wanting to monitor, say, 'cycles' for a cgroup and also for system wide, this
    362command line can be used: 'perf stat -e cycles -G cgroup_name -a -e cycles'.
    363
    364-b::
    365--branch-any::
    366Enable taken branch stack sampling. Any type of taken branch may be sampled.
    367This is a shortcut for --branch-filter any. See --branch-filter for more infos.
    368
    369-j::
    370--branch-filter::
    371Enable taken branch stack sampling. Each sample captures a series of consecutive
    372taken branches. The number of branches captured with each sample depends on the
    373underlying hardware, the type of branches of interest, and the executed code.
    374It is possible to select the types of branches captured by enabling filters. The
    375following filters are defined:
    376
    377        - any:  any type of branches
    378        - any_call: any function call or system call
    379        - any_ret: any function return or system call return
    380        - ind_call: any indirect branch
    381        - call: direct calls, including far (to/from kernel) calls
    382        - u:  only when the branch target is at the user level
    383        - k: only when the branch target is in the kernel
    384        - hv: only when the target is at the hypervisor level
    385	- in_tx: only when the target is in a hardware transaction
    386	- no_tx: only when the target is not in a hardware transaction
    387	- abort_tx: only when the target is a hardware transaction abort
    388	- cond: conditional branches
    389	- save_type: save branch type during sampling in case binary is not available later
    390
    391+
    392The option requires at least one branch type among any, any_call, any_ret, ind_call, cond.
    393The privilege levels may be omitted, in which case, the privilege levels of the associated
    394event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege
    395levels are subject to permissions.  When sampling on multiple events, branch stack sampling
    396is enabled for all the sampling events. The sampled branch type is the same for all events.
    397The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k
    398Note that this feature may not be available on all processors.
    399
    400--weight::
    401Enable weightened sampling. An additional weight is recorded per sample and can be
    402displayed with the weight and local_weight sort keys.  This currently works for TSX
    403abort events and some memory events in precise mode on modern Intel CPUs.
    404
    405--namespaces::
    406Record events of type PERF_RECORD_NAMESPACES.  This enables 'cgroup_id' sort key.
    407
    408--all-cgroups::
    409Record events of type PERF_RECORD_CGROUP.  This enables 'cgroup' sort key.
    410
    411--transaction::
    412Record transaction flags for transaction related events.
    413
    414--per-thread::
    415Use per-thread mmaps.  By default per-cpu mmaps are created.  This option
    416overrides that and uses per-thread mmaps.  A side-effect of that is that
    417inheritance is automatically disabled.  --per-thread is ignored with a warning
    418if combined with -a or -C options.
    419
    420-D::
    421--delay=::
    422After starting the program, wait msecs before measuring (-1: start with events
    423disabled). This is useful to filter out the startup phase of the program, which
    424is often very different.
    425
    426-I::
    427--intr-regs::
    428Capture machine state (registers) at interrupt, i.e., on counter overflows for
    429each sample. List of captured registers depends on the architecture. This option
    430is off by default. It is possible to select the registers to sample using their
    431symbolic names, e.g. on x86, ax, si. To list the available registers use
    432--intr-regs=\?. To name registers, pass a comma separated list such as
    433--intr-regs=ax,bx. The list of register is architecture dependent.
    434
    435--user-regs::
    436Similar to -I, but capture user registers at sample time. To list the available
    437user registers use --user-regs=\?.
    438
    439--running-time::
    440Record running and enabled time for read events (:S)
    441
    442-k::
    443--clockid::
    444Sets the clock id to use for the various time fields in the perf_event_type
    445records. See clock_gettime(). In particular CLOCK_MONOTONIC and
    446CLOCK_MONOTONIC_RAW are supported, some events might also allow
    447CLOCK_BOOTTIME, CLOCK_REALTIME and CLOCK_TAI.
    448
    449-S::
    450--snapshot::
    451Select AUX area tracing Snapshot Mode. This option is valid only with an
    452AUX area tracing event. Optionally, certain snapshot capturing parameters
    453can be specified in a string that follows this option:
    454  'e': take one last snapshot on exit; guarantees that there is at least one
    455       snapshot in the output file;
    456  <size>: if the PMU supports this, specify the desired snapshot size.
    457
    458In Snapshot Mode trace data is captured only when signal SIGUSR2 is received
    459and on exit if the above 'e' option is given.
    460
    461--aux-sample[=OPTIONS]::
    462Select AUX area sampling. At least one of the events selected by the -e option
    463must be an AUX area event. Samples on other events will be created containing
    464data from the AUX area. Optionally sample size may be specified, otherwise it
    465defaults to 4KiB.
    466
    467--proc-map-timeout::
    468When processing pre-existing threads /proc/XXX/mmap, it may take a long time,
    469because the file may be huge. A time out is needed in such cases.
    470This option sets the time out limit. The default value is 500 ms.
    471
    472--switch-events::
    473Record context switch events i.e. events of type PERF_RECORD_SWITCH or
    474PERF_RECORD_SWITCH_CPU_WIDE. In some cases (e.g. Intel PT, CoreSight or Arm SPE)
    475switch events will be enabled automatically, which can be suppressed by
    476by the option --no-switch-events.
    477
    478--clang-path=PATH::
    479Path to clang binary to use for compiling BPF scriptlets.
    480(enabled when BPF support is on)
    481
    482--clang-opt=OPTIONS::
    483Options passed to clang when compiling BPF scriptlets.
    484(enabled when BPF support is on)
    485
    486--vmlinux=PATH::
    487Specify vmlinux path which has debuginfo.
    488(enabled when BPF prologue is on)
    489
    490--buildid-all::
    491Record build-id of all DSOs regardless whether it's actually hit or not.
    492
    493--buildid-mmap::
    494Record build ids in mmap2 events, disables build id cache (implies --no-buildid).
    495
    496--aio[=n]::
    497Use <n> control blocks in asynchronous (Posix AIO) trace writing mode (default: 1, max: 4).
    498Asynchronous mode is supported only when linking Perf tool with libc library
    499providing implementation for Posix AIO API.
    500
    501--affinity=mode::
    502Set affinity mask of trace reading thread according to the policy defined by 'mode' value:
    503  node - thread affinity mask is set to NUMA node cpu mask of the processed mmap buffer
    504  cpu  - thread affinity mask is set to cpu of the processed mmap buffer
    505
    506--mmap-flush=number::
    507
    508Specify minimal number of bytes that is extracted from mmap data pages and
    509processed for output. One can specify the number using B/K/M/G suffixes.
    510
    511The maximal allowed value is a quarter of the size of mmaped data pages.
    512
    513The default option value is 1 byte which means that every time that the output
    514writing thread finds some new data in the mmaped buffer the data is extracted,
    515possibly compressed (-z) and written to the output, perf.data or pipe.
    516
    517Larger data chunks are compressed more effectively in comparison to smaller
    518chunks so extraction of larger chunks from the mmap data pages is preferable
    519from the perspective of output size reduction.
    520
    521Also at some cases executing less output write syscalls with bigger data size
    522can take less time than executing more output write syscalls with smaller data
    523size thus lowering runtime profiling overhead.
    524
    525-z::
    526--compression-level[=n]::
    527Produce compressed trace using specified level n (default: 1 - fastest compression,
    52822 - smallest trace)
    529
    530--all-kernel::
    531Configure all used events to run in kernel space.
    532
    533--all-user::
    534Configure all used events to run in user space.
    535
    536--kernel-callchains::
    537Collect callchains only from kernel space. I.e. this option sets
    538perf_event_attr.exclude_callchain_user to 1.
    539
    540--user-callchains::
    541Collect callchains only from user space. I.e. this option sets
    542perf_event_attr.exclude_callchain_kernel to 1.
    543
    544Don't use both --kernel-callchains and --user-callchains at the same time or no
    545callchains will be collected.
    546
    547--timestamp-filename
    548Append timestamp to output file name.
    549
    550--timestamp-boundary::
    551Record timestamp boundary (time of first/last samples).
    552
    553--switch-output[=mode]::
    554Generate multiple perf.data files, timestamp prefixed, switching to a new one
    555based on 'mode' value:
    556  "signal" - when receiving a SIGUSR2 (default value) or
    557  <size>   - when reaching the size threshold, size is expected to
    558             be a number with appended unit character - B/K/M/G
    559  <time>   - when reaching the time threshold, size is expected to
    560             be a number with appended unit character - s/m/h/d
    561
    562             Note: the precision of  the size  threshold  hugely depends
    563             on your configuration  - the number and size of  your  ring
    564             buffers (-m). It is generally more precise for higher sizes
    565             (like >5M), for lower values expect different sizes.
    566
    567A possible use case is to, given an external event, slice the perf.data file
    568that gets then processed, possibly via a perf script, to decide if that
    569particular perf.data snapshot should be kept or not.
    570
    571Implies --timestamp-filename, --no-buildid and --no-buildid-cache.
    572The reason for the latter two is to reduce the data file switching
    573overhead. You can still switch them on with:
    574
    575  --switch-output --no-no-buildid  --no-no-buildid-cache
    576
    577--switch-output-event::
    578Events that will cause the switch of the perf.data file, auto-selecting
    579--switch-output=signal, the results are similar as internally the side band
    580thread will also send a SIGUSR2 to the main one.
    581
    582Uses the same syntax as --event, it will just not be recorded, serving only to
    583switch the perf.data file as soon as the --switch-output event is processed by
    584a separate sideband thread.
    585
    586This sideband thread is also used to other purposes, like processing the
    587PERF_RECORD_BPF_EVENT records as they happen, asking the kernel for extra BPF
    588information, etc.
    589
    590--switch-max-files=N::
    591
    592When rotating perf.data with --switch-output, only keep N files.
    593
    594--dry-run::
    595Parse options then exit. --dry-run can be used to detect errors in cmdline
    596options.
    597
    598'perf record --dry-run -e' can act as a BPF script compiler if llvm.dump-obj
    599in config file is set to true.
    600
    601--synth=TYPE::
    602Collect and synthesize given type of events (comma separated).  Note that
    603this option controls the synthesis from the /proc filesystem which represent
    604task status for pre-existing threads.
    605
    606Kernel (and some other) events are recorded regardless of the
    607choice in this option.  For example, --synth=no would have MMAP events for
    608kernel and modules.
    609
    610Available types are:
    611  'task'    - synthesize FORK and COMM events for each task
    612  'mmap'    - synthesize MMAP events for each process (implies 'task')
    613  'cgroup'  - synthesize CGROUP events for each cgroup
    614  'all'     - synthesize all events (default)
    615  'no'      - do not synthesize any of the above events
    616
    617--tail-synthesize::
    618Instead of collecting non-sample events (for example, fork, comm, mmap) at
    619the beginning of record, collect them during finalizing an output file.
    620The collected non-sample events reflects the status of the system when
    621record is finished.
    622
    623--overwrite::
    624Makes all events use an overwritable ring buffer. An overwritable ring
    625buffer works like a flight recorder: when it gets full, the kernel will
    626overwrite the oldest records, that thus will never make it to the
    627perf.data file.
    628
    629When '--overwrite' and '--switch-output' are used perf records and drops
    630events until it receives a signal, meaning that something unusual was
    631detected that warrants taking a snapshot of the most current events,
    632those fitting in the ring buffer at that moment.
    633
    634'overwrite' attribute can also be set or canceled for an event using
    635config terms. For example: 'cycles/overwrite/' and 'instructions/no-overwrite/'.
    636
    637Implies --tail-synthesize.
    638
    639--kcore::
    640Make a copy of /proc/kcore and place it into a directory with the perf data file.
    641
    642--max-size=<size>::
    643Limit the sample data max size, <size> is expected to be a number with
    644appended unit character - B/K/M/G
    645
    646--num-thread-synthesize::
    647	The number of threads to run when synthesizing events for existing processes.
    648	By default, the number of threads equals 1.
    649
    650ifdef::HAVE_LIBPFM[]
    651--pfm-events events::
    652Select a PMU event using libpfm4 syntax (see http://perfmon2.sf.net)
    653including support for event filters. For example '--pfm-events
    654inst_retired:any_p:u:c=1:i'. More than one event can be passed to the
    655option using the comma separator. Hardware events and generic hardware
    656events cannot be mixed together. The latter must be used with the -e
    657option. The -e option and this one can be mixed and matched.  Events
    658can be grouped using the {} notation.
    659endif::HAVE_LIBPFM[]
    660
    661--control=fifo:ctl-fifo[,ack-fifo]::
    662--control=fd:ctl-fd[,ack-fd]::
    663ctl-fifo / ack-fifo are opened and used as ctl-fd / ack-fd as follows.
    664Listen on ctl-fd descriptor for command to control measurement.
    665
    666Available commands:
    667  'enable'           : enable events
    668  'disable'          : disable events
    669  'enable name'      : enable event 'name'
    670  'disable name'     : disable event 'name'
    671  'snapshot'         : AUX area tracing snapshot).
    672  'stop'             : stop perf record
    673  'ping'             : ping
    674
    675  'evlist [-v|-g|-F] : display all events
    676                       -F  Show just the sample frequency used for each event.
    677                       -v  Show all fields.
    678                       -g  Show event group information.
    679
    680Measurements can be started with events disabled using --delay=-1 option. Optionally
    681send control command completion ('ack\n') to ack-fd descriptor to synchronize with the
    682controlling process.  Example of bash shell script to enable and disable events during
    683measurements:
    684
    685 #!/bin/bash
    686
    687 ctl_dir=/tmp/
    688
    689 ctl_fifo=${ctl_dir}perf_ctl.fifo
    690 test -p ${ctl_fifo} && unlink ${ctl_fifo}
    691 mkfifo ${ctl_fifo}
    692 exec {ctl_fd}<>${ctl_fifo}
    693
    694 ctl_ack_fifo=${ctl_dir}perf_ctl_ack.fifo
    695 test -p ${ctl_ack_fifo} && unlink ${ctl_ack_fifo}
    696 mkfifo ${ctl_ack_fifo}
    697 exec {ctl_fd_ack}<>${ctl_ack_fifo}
    698
    699 perf record -D -1 -e cpu-cycles -a               \
    700             --control fd:${ctl_fd},${ctl_fd_ack} \
    701             -- sleep 30 &
    702 perf_pid=$!
    703
    704 sleep 5  && echo 'enable' >&${ctl_fd} && read -u ${ctl_fd_ack} e1 && echo "enabled(${e1})"
    705 sleep 10 && echo 'disable' >&${ctl_fd} && read -u ${ctl_fd_ack} d1 && echo "disabled(${d1})"
    706
    707 exec {ctl_fd_ack}>&-
    708 unlink ${ctl_ack_fifo}
    709
    710 exec {ctl_fd}>&-
    711 unlink ${ctl_fifo}
    712
    713 wait -n ${perf_pid}
    714 exit $?
    715
    716--threads=<spec>::
    717Write collected trace data into several data files using parallel threads.
    718<spec> value can be user defined list of masks. Masks separated by colon
    719define CPUs to be monitored by a thread and affinity mask of that thread
    720is separated by slash:
    721
    722    <cpus mask 1>/<affinity mask 1>:<cpus mask 2>/<affinity mask 2>:...
    723
    724CPUs or affinity masks must not overlap with other corresponding masks.
    725Invalid CPUs are ignored, but masks containing only invalid CPUs are not
    726allowed.
    727
    728For example user specification like the following:
    729
    730    0,2-4/2-4:1,5-7/5-7
    731
    732specifies parallel threads layout that consists of two threads,
    733the first thread monitors CPUs 0 and 2-4 with the affinity mask 2-4,
    734the second monitors CPUs 1 and 5-7 with the affinity mask 5-7.
    735
    736<spec> value can also be a string meaning predefined parallel threads
    737layout:
    738
    739    cpu    - create new data streaming thread for every monitored cpu
    740    core   - create new thread to monitor CPUs grouped by a core
    741    package - create new thread to monitor CPUs grouped by a package
    742    numa   - create new threed to monitor CPUs grouped by a NUMA domain
    743
    744Predefined layouts can be used on systems with large number of CPUs in
    745order not to spawn multiple per-cpu streaming threads but still avoid LOST
    746events in data directory files. Option specified with no or empty value
    747defaults to CPU layout. Masks defined or provided by the option value are
    748filtered through the mask provided by -C option.
    749
    750include::intel-hybrid.txt[]
    751
    752--debuginfod[=URLs]::
    753	Specify debuginfod URL to be used when cacheing perf.data binaries,
    754	it follows the same syntax as the DEBUGINFOD_URLS variable, like:
    755
    756	  http://192.168.122.174:8002
    757
    758	If the URLs is not specified, the value of DEBUGINFOD_URLS
    759	system environment variable is used.
    760
    761--off-cpu::
    762	Enable off-cpu profiling with BPF.  The BPF program will collect
    763	task scheduling information with (user) stacktrace and save them
    764	as sample data of a software event named "offcpu-time".  The
    765	sample period will have the time the task slept in nanoseconds.
    766
    767	Note that BPF can collect stack traces using frame pointer ("fp")
    768	only, as of now.  So the applications built without the frame
    769	pointer might see bogus addresses.
    770
    771SEE ALSO
    772--------
    773linkperf:perf-stat[1], linkperf:perf-list[1], linkperf:perf-intel-pt[1]