1perf-record(1) 2============== 3 4NAME 5---- 6perf-record - Run a command and record its profile into perf.data 7 8SYNOPSIS 9-------- 10[verse] 11'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] <command> 12'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] -- <command> [<options>] 13 14DESCRIPTION 15----------- 16This command runs a command and gathers a performance counter profile 17from it, into perf.data - without displaying anything. 18 19This file can then be inspected later on, using 'perf report'. 20 21 22OPTIONS 23------- 24<command>...:: 25 Any command you can specify in a shell. 26 27-e:: 28--event=:: 29 Select the PMU event. Selection can be: 30 31 - a symbolic event name (use 'perf list' to list all events) 32 33 - a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a 34 hexadecimal event descriptor. 35 36 - a 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/*. 39 40 - a symbolically formed event like 'pmu/config=M,config1=N,config3=K/' 41 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/* 47 48 There are also some params which are not defined in .../<pmu>/format/*. 49 These params can be used to overload default config values per event. 50 Here is a list of the params. 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. 55 The default is 1. 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 Note: If user explicitly sets options which conflict with the params, 61 the value set by the params will be overridden. 62 63 Also not defined in .../<pmu>/format/* are PMU driver specific 64 configuration parameters. Any configuration parameter preceded by 65 the letter '@' is not interpreted in user space and sent down directly 66 to the PMU driver. For example: 67 68 perf record -e some_event/@cfg1,@cfg2=config/ ... 69 70 will see 'cfg1' and 'cfg2=config' pushed to the PMU driver associated 71 with the event for further processing. There is no restriction on 72 what the configuration parameters are, as long as their semantic is 73 understood and supported by the PMU driver. 74 75 - a hardware breakpoint event in the form of '\mem:addr[/len][:access]' 76 where addr is the address in memory you want to break in. 77 Access is the memory access type (read, write, execute) it can 78 be passed as follows: '\mem:addr[:[r][w][x]]'. len is the range, 79 number of bytes from specified addr, which the breakpoint will cover. 80 If you want to profile read-write accesses in 0x1000, just set 81 'mem:0x1000:rw'. 82 If you want to profile write accesses in [0x1000~1008), just set 83 'mem:0x1000/8:w'. 84 85 - a group of events surrounded by a pair of brace ("{event1,event2,...}"). 86 Each event is separated by commas and the group should be quoted to 87 prevent the shell interpretation. You also need to use --group on 88 "perf report" to view group events together. 89 90--filter=<filter>:: 91 Event filter. This option should follow a event selector (-e) which 92 selects either tracepoint event(s) or a hardware trace PMU 93 (e.g. Intel PT or CoreSight). 94 95 - tracepoint filters 96 97 In the case of tracepoints, multiple '--filter' options are combined 98 using '&&'. 99 100 - address filters 101 102 A hardware trace PMU advertises its ability to accept a number of 103 address filters by specifying a non-zero value in 104 /sys/bus/event_source/devices/<pmu>/nr_addr_filters. 105 106 Address filters have the format: 107 108 filter|start|stop|tracestop <start> [/ <size>] [@<file name>] 109 110 Where: 111 - 'filter': defines a region that will be traced. 112 - 'start': defines an address at which tracing will begin. 113 - 'stop': defines an address at which tracing will stop. 114 - 'tracestop': defines a region in which tracing will stop. 115 116 <file name> is the name of the object file, <start> is the offset to the 117 code to trace in that file, and <size> is the size of the region to 118 trace. 'start' and 'stop' filters need not specify a <size>. 119 120 If no object file is specified then the kernel is assumed, in which case 121 the start address must be a current kernel memory address. 122 123 <start> can also be specified by providing the name of a symbol. If the 124 symbol name is not unique, it can be disambiguated by inserting #n where 125 'n' selects the n'th symbol in address order. Alternately #0, #g or #G 126 select only a global symbol. <size> can also be specified by providing 127 the name of a symbol, in which case the size is calculated to the end 128 of that symbol. For 'filter' and 'tracestop' filters, if <size> is 129 omitted and <start> is a symbol, then the size is calculated to the end 130 of that symbol. 131 132 If <size> is omitted and <start> is '*', then the start and size will 133 be calculated from the first and last symbols, i.e. to trace the whole 134 file. 135 136 If symbol names (or '*') are provided, they must be surrounded by white 137 space. 138 139 The filter passed to the kernel is not necessarily the same as entered. 140 To see the filter that is passed, use the -v option. 141 142 The kernel may not be able to configure a trace region if it is not 143 within a single mapping. MMAP events (or /proc/<pid>/maps) can be 144 examined to determine if that is a possibility. 145 146 Multiple filters can be separated with space or comma. 147 148--exclude-perf:: 149 Don't record events issued by perf itself. This option should follow 150 a event selector (-e) which selects tracepoint event(s). It adds a 151 filter expression 'common_pid != $PERFPID' to filters. If other 152 '--filter' exists, the new filter expression will be combined with 153 them by '&&'. 154 155-a:: 156--all-cpus:: 157 System-wide collection from all CPUs. 158 159-p:: 160--pid=:: 161 Record events on existing process ID (comma separated list). 162 163-t:: 164--tid=:: 165 Record events on existing thread ID (comma separated list). 166 This option also disables inheritance by default. Enable it by adding 167 --inherit. 168 169-u:: 170--uid=:: 171 Record events in threads owned by uid. Name or number. 172 173-r:: 174--realtime=:: 175 Collect data with this RT SCHED_FIFO priority. 176 177--no-buffering:: 178 Collect data without buffering. 179 180-c:: 181--count=:: 182 Event period to sample. 183 184-o:: 185--output=:: 186 Output file name. 187 188-i:: 189--no-inherit:: 190 Child tasks do not inherit counters. 191-F:: 192--freq=:: 193 Profile at this frequency. 194 195-m:: 196--mmap-pages=:: 197 Number of mmap data pages (must be a power of two) or size 198 specification with appended unit character - B/K/M/G. The 199 size is rounded up to have nearest pages power of two value. 200 Also, by adding a comma, the number of mmap pages for AUX 201 area tracing can be specified. 202 203--group:: 204 Put all events in a single event group. This precedes the --event 205 option and remains only for backward compatibility. See --event. 206 207-g:: 208 Enables call-graph (stack chain/backtrace) recording. 209 210--call-graph:: 211 Setup and enable call-graph (stack chain/backtrace) recording, 212 implies -g. Default is "fp". 213 214 Allows specifying "fp" (frame pointer) or "dwarf" 215 (DWARF's CFI - Call Frame Information) or "lbr" 216 (Hardware Last Branch Record facility) as the method to collect 217 the information used to show the call graphs. 218 219 In some systems, where binaries are build with gcc 220 --fomit-frame-pointer, using the "fp" method will produce bogus 221 call graphs, using "dwarf", if available (perf tools linked to 222 the libunwind or libdw library) should be used instead. 223 Using the "lbr" method doesn't require any compiler options. It 224 will produce call graphs from the hardware LBR registers. The 225 main limition is that it is only available on new Intel 226 platforms, such as Haswell. It can only get user call chain. It 227 doesn't work with branch stack sampling at the same time. 228 229 When "dwarf" recording is used, perf also records (user) stack dump 230 when sampled. Default size of the stack dump is 8192 (bytes). 231 User can change the size by passing the size after comma like 232 "--call-graph dwarf,4096". 233 234-q:: 235--quiet:: 236 Don't print any message, useful for scripting. 237 238-v:: 239--verbose:: 240 Be more verbose (show counter open errors, etc). 241 242-s:: 243--stat:: 244 Record per-thread event counts. Use it with 'perf report -T' to see 245 the values. 246 247-d:: 248--data:: 249 Record the sample addresses. 250 251-T:: 252--timestamp:: 253 Record the sample timestamps. Use it with 'perf report -D' to see the 254 timestamps, for instance. 255 256-P:: 257--period:: 258 Record the sample period. 259 260--sample-cpu:: 261 Record the sample cpu. 262 263-n:: 264--no-samples:: 265 Don't sample. 266 267-R:: 268--raw-samples:: 269Collect raw sample records from all opened counters (default for tracepoint counters). 270 271-C:: 272--cpu:: 273Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a 274comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. 275In per-thread mode with inheritance mode on (default), samples are captured only when 276the thread executes on the designated CPUs. Default is to monitor all CPUs. 277 278-B:: 279--no-buildid:: 280Do not save the build ids of binaries in the perf.data files. This skips 281post processing after recording, which sometimes makes the final step in 282the recording process to take a long time, as it needs to process all 283events looking for mmap records. The downside is that it can misresolve 284symbols if the workload binaries used when recording get locally rebuilt 285or upgraded, because the only key available in this case is the 286pathname. You can also set the "record.build-id" config variable to 287'skip to have this behaviour permanently. 288 289-N:: 290--no-buildid-cache:: 291Do not update the buildid cache. This saves some overhead in situations 292where the information in the perf.data file (which includes buildids) 293is sufficient. You can also set the "record.build-id" config variable to 294'no-cache' to have the same effect. 295 296-G name,...:: 297--cgroup name,...:: 298monitor only in the container (cgroup) called "name". This option is available only 299in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to 300container "name" are monitored when they run on the monitored CPUs. Multiple cgroups 301can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup 302to first event, second cgroup to second event and so on. It is possible to provide 303an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have 304corresponding events, i.e., they always refer to events defined earlier on the command 305line. 306 307-b:: 308--branch-any:: 309Enable taken branch stack sampling. Any type of taken branch may be sampled. 310This is a shortcut for --branch-filter any. See --branch-filter for more infos. 311 312-j:: 313--branch-filter:: 314Enable taken branch stack sampling. Each sample captures a series of consecutive 315taken branches. The number of branches captured with each sample depends on the 316underlying hardware, the type of branches of interest, and the executed code. 317It is possible to select the types of branches captured by enabling filters. The 318following filters are defined: 319 320 - any: any type of branches 321 - any_call: any function call or system call 322 - any_ret: any function return or system call return 323 - ind_call: any indirect branch 324 - call: direct calls, including far (to/from kernel) calls 325 - u: only when the branch target is at the user level 326 - k: only when the branch target is in the kernel 327 - hv: only when the target is at the hypervisor level 328 - in_tx: only when the target is in a hardware transaction 329 - no_tx: only when the target is not in a hardware transaction 330 - abort_tx: only when the target is a hardware transaction abort 331 - cond: conditional branches 332 333+ 334The option requires at least one branch type among any, any_call, any_ret, ind_call, cond. 335The privilege levels may be omitted, in which case, the privilege levels of the associated 336event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege 337levels are subject to permissions. When sampling on multiple events, branch stack sampling 338is enabled for all the sampling events. The sampled branch type is the same for all events. 339The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k 340Note that this feature may not be available on all processors. 341 342--weight:: 343Enable weightened sampling. An additional weight is recorded per sample and can be 344displayed with the weight and local_weight sort keys. This currently works for TSX 345abort events and some memory events in precise mode on modern Intel CPUs. 346 347--transaction:: 348Record transaction flags for transaction related events. 349 350--per-thread:: 351Use per-thread mmaps. By default per-cpu mmaps are created. This option 352overrides that and uses per-thread mmaps. A side-effect of that is that 353inheritance is automatically disabled. --per-thread is ignored with a warning 354if combined with -a or -C options. 355 356-D:: 357--delay=:: 358After starting the program, wait msecs before measuring. This is useful to 359filter out the startup phase of the program, which is often very different. 360 361-I:: 362--intr-regs:: 363Capture machine state (registers) at interrupt, i.e., on counter overflows for 364each sample. List of captured registers depends on the architecture. This option 365is off by default. It is possible to select the registers to sample using their 366symbolic names, e.g. on x86, ax, si. To list the available registers use 367--intr-regs=\?. To name registers, pass a comma separated list such as 368--intr-regs=ax,bx. The list of register is architecture dependent. 369 370 371--running-time:: 372Record running and enabled time for read events (:S) 373 374-k:: 375--clockid:: 376Sets the clock id to use for the various time fields in the perf_event_type 377records. See clock_gettime(). In particular CLOCK_MONOTONIC and 378CLOCK_MONOTONIC_RAW are supported, some events might also allow 379CLOCK_BOOTTIME, CLOCK_REALTIME and CLOCK_TAI. 380 381-S:: 382--snapshot:: 383Select AUX area tracing Snapshot Mode. This option is valid only with an 384AUX area tracing event. Optionally the number of bytes to capture per 385snapshot can be specified. In Snapshot Mode, trace data is captured only when 386signal SIGUSR2 is received. 387 388--proc-map-timeout:: 389When processing pre-existing threads /proc/XXX/mmap, it may take a long time, 390because the file may be huge. A time out is needed in such cases. 391This option sets the time out limit. The default value is 500 ms. 392 393--switch-events:: 394Record context switch events i.e. events of type PERF_RECORD_SWITCH or 395PERF_RECORD_SWITCH_CPU_WIDE. 396 397--clang-path=PATH:: 398Path to clang binary to use for compiling BPF scriptlets. 399(enabled when BPF support is on) 400 401--clang-opt=OPTIONS:: 402Options passed to clang when compiling BPF scriptlets. 403(enabled when BPF support is on) 404 405--vmlinux=PATH:: 406Specify vmlinux path which has debuginfo. 407(enabled when BPF prologue is on) 408 409--buildid-all:: 410Record build-id of all DSOs regardless whether it's actually hit or not. 411 412--all-kernel:: 413Configure all used events to run in kernel space. 414 415--all-user:: 416Configure all used events to run in user space. 417 418--timestamp-filename 419Append timestamp to output file name. 420 421--switch-output:: 422Generate multiple perf.data files, timestamp prefixed, switching to a new one 423when receiving a SIGUSR2. 424 425A possible use case is to, given an external event, slice the perf.data file 426that gets then processed, possibly via a perf script, to decide if that 427particular perf.data snapshot should be kept or not. 428 429Implies --timestamp-filename, --no-buildid and --no-buildid-cache. 430 431--dry-run:: 432Parse options then exit. --dry-run can be used to detect errors in cmdline 433options. 434 435'perf record --dry-run -e' can act as a BPF script compiler if llvm.dump-obj 436in config file is set to true. 437 438--tail-synthesize:: 439Instead of collecting non-sample events (for example, fork, comm, mmap) at 440the beginning of record, collect them during finalizing an output file. 441The collected non-sample events reflects the status of the system when 442record is finished. 443 444--overwrite:: 445Makes all events use an overwritable ring buffer. An overwritable ring 446buffer works like a flight recorder: when it gets full, the kernel will 447overwrite the oldest records, that thus will never make it to the 448perf.data file. 449 450When '--overwrite' and '--switch-output' are used perf records and drops 451events until it receives a signal, meaning that something unusual was 452detected that warrants taking a snapshot of the most current events, 453those fitting in the ring buffer at that moment. 454 455'overwrite' attribute can also be set or canceled for an event using 456config terms. For example: 'cycles/overwrite/' and 'instructions/no-overwrite/'. 457 458Implies --tail-synthesize. 459 460SEE ALSO 461-------- 462linkperf:perf-stat[1], linkperf:perf-list[1] 463