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 symbolic or raw PMU event followed by an optional colon 37 and a list of event modifiers, e.g., cpu-cycles:p. See the 38 linkperf:perf-list[1] man page for details on event modifiers. 39 40 - a symbolically formed PMU event like 'pmu/param1=0x3,param2/' where 41 'param1', 'param2', etc are defined as formats for the PMU in 42 /sys/bus/event_sources/devices/<pmu>/format/*. 43 44 - a symbolically formed event like 'pmu/config=M,config1=N,config3=K/' 45 46 where M, N, K are numbers (in decimal, hex, octal format). Acceptable 47 values for each of 'config', 'config1' and 'config2' are defined by 48 corresponding entries in /sys/bus/event_sources/devices/<pmu>/format/* 49 param1 and param2 are defined as formats for the PMU in: 50 /sys/bus/event_sources/devices/<pmu>/format/* 51 52 There are also some params which are not defined in .../<pmu>/format/*. 53 These params can be used to overload default config values per event. 54 Here is a list of the params. 55 - 'period': Set event sampling period 56 - 'freq': Set event sampling frequency 57 - 'time': Disable/enable time stamping. Acceptable values are 1 for 58 enabling time stamping. 0 for disabling time stamping. 59 The default is 1. 60 - 'call-graph': Disable/enable callgraph. Acceptable str are "fp" for 61 FP mode, "dwarf" for DWARF mode, "lbr" for LBR mode and 62 "no" for disable callgraph. 63 - 'stack-size': user stack size for dwarf mode 64 Note: If user explicitly sets options which conflict with the params, 65 the value set by the params will be overridden. 66 67 - a hardware breakpoint event in the form of '\mem:addr[/len][:access]' 68 where addr is the address in memory you want to break in. 69 Access is the memory access type (read, write, execute) it can 70 be passed as follows: '\mem:addr[:[r][w][x]]'. len is the range, 71 number of bytes from specified addr, which the breakpoint will cover. 72 If you want to profile read-write accesses in 0x1000, just set 73 'mem:0x1000:rw'. 74 If you want to profile write accesses in [0x1000~1008), just set 75 'mem:0x1000/8:w'. 76 77 - a group of events surrounded by a pair of brace ("{event1,event2,...}"). 78 Each event is separated by commas and the group should be quoted to 79 prevent the shell interpretation. You also need to use --group on 80 "perf report" to view group events together. 81 82--filter=<filter>:: 83 Event filter. This option should follow a event selector (-e) which 84 selects tracepoint event(s). Multiple '--filter' options are combined 85 using '&&'. 86 87--exclude-perf:: 88 Don't record events issued by perf itself. This option should follow 89 a event selector (-e) which selects tracepoint event(s). It adds a 90 filter expression 'common_pid != $PERFPID' to filters. If other 91 '--filter' exists, the new filter expression will be combined with 92 them by '&&'. 93 94-a:: 95--all-cpus:: 96 System-wide collection from all CPUs. 97 98-p:: 99--pid=:: 100 Record events on existing process ID (comma separated list). 101 102-t:: 103--tid=:: 104 Record events on existing thread ID (comma separated list). 105 This option also disables inheritance by default. Enable it by adding 106 --inherit. 107 108-u:: 109--uid=:: 110 Record events in threads owned by uid. Name or number. 111 112-r:: 113--realtime=:: 114 Collect data with this RT SCHED_FIFO priority. 115 116--no-buffering:: 117 Collect data without buffering. 118 119-c:: 120--count=:: 121 Event period to sample. 122 123-o:: 124--output=:: 125 Output file name. 126 127-i:: 128--no-inherit:: 129 Child tasks do not inherit counters. 130-F:: 131--freq=:: 132 Profile at this frequency. 133 134-m:: 135--mmap-pages=:: 136 Number of mmap data pages (must be a power of two) or size 137 specification with appended unit character - B/K/M/G. The 138 size is rounded up to have nearest pages power of two value. 139 Also, by adding a comma, the number of mmap pages for AUX 140 area tracing can be specified. 141 142--group:: 143 Put all events in a single event group. This precedes the --event 144 option and remains only for backward compatibility. See --event. 145 146-g:: 147 Enables call-graph (stack chain/backtrace) recording. 148 149--call-graph:: 150 Setup and enable call-graph (stack chain/backtrace) recording, 151 implies -g. Default is "fp". 152 153 Allows specifying "fp" (frame pointer) or "dwarf" 154 (DWARF's CFI - Call Frame Information) or "lbr" 155 (Hardware Last Branch Record facility) as the method to collect 156 the information used to show the call graphs. 157 158 In some systems, where binaries are build with gcc 159 --fomit-frame-pointer, using the "fp" method will produce bogus 160 call graphs, using "dwarf", if available (perf tools linked to 161 the libunwind or libdw library) should be used instead. 162 Using the "lbr" method doesn't require any compiler options. It 163 will produce call graphs from the hardware LBR registers. The 164 main limition is that it is only available on new Intel 165 platforms, such as Haswell. It can only get user call chain. It 166 doesn't work with branch stack sampling at the same time. 167 168 When "dwarf" recording is used, perf also records (user) stack dump 169 when sampled. Default size of the stack dump is 8192 (bytes). 170 User can change the size by passing the size after comma like 171 "--call-graph dwarf,4096". 172 173-q:: 174--quiet:: 175 Don't print any message, useful for scripting. 176 177-v:: 178--verbose:: 179 Be more verbose (show counter open errors, etc). 180 181-s:: 182--stat:: 183 Record per-thread event counts. Use it with 'perf report -T' to see 184 the values. 185 186-d:: 187--data:: 188 Record the sample addresses. 189 190-T:: 191--timestamp:: 192 Record the sample timestamps. Use it with 'perf report -D' to see the 193 timestamps, for instance. 194 195-P:: 196--period:: 197 Record the sample period. 198 199-n:: 200--no-samples:: 201 Don't sample. 202 203-R:: 204--raw-samples:: 205Collect raw sample records from all opened counters (default for tracepoint counters). 206 207-C:: 208--cpu:: 209Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a 210comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. 211In per-thread mode with inheritance mode on (default), samples are captured only when 212the thread executes on the designated CPUs. Default is to monitor all CPUs. 213 214-N:: 215--no-buildid-cache:: 216Do not update the buildid cache. This saves some overhead in situations 217where the information in the perf.data file (which includes buildids) 218is sufficient. 219 220-G name,...:: 221--cgroup name,...:: 222monitor only in the container (cgroup) called "name". This option is available only 223in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to 224container "name" are monitored when they run on the monitored CPUs. Multiple cgroups 225can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup 226to first event, second cgroup to second event and so on. It is possible to provide 227an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have 228corresponding events, i.e., they always refer to events defined earlier on the command 229line. 230 231-b:: 232--branch-any:: 233Enable taken branch stack sampling. Any type of taken branch may be sampled. 234This is a shortcut for --branch-filter any. See --branch-filter for more infos. 235 236-j:: 237--branch-filter:: 238Enable taken branch stack sampling. Each sample captures a series of consecutive 239taken branches. The number of branches captured with each sample depends on the 240underlying hardware, the type of branches of interest, and the executed code. 241It is possible to select the types of branches captured by enabling filters. The 242following filters are defined: 243 244 - any: any type of branches 245 - any_call: any function call or system call 246 - any_ret: any function return or system call return 247 - ind_call: any indirect branch 248 - call: direct calls, including far (to/from kernel) calls 249 - u: only when the branch target is at the user level 250 - k: only when the branch target is in the kernel 251 - hv: only when the target is at the hypervisor level 252 - in_tx: only when the target is in a hardware transaction 253 - no_tx: only when the target is not in a hardware transaction 254 - abort_tx: only when the target is a hardware transaction abort 255 - cond: conditional branches 256 257+ 258The option requires at least one branch type among any, any_call, any_ret, ind_call, cond. 259The privilege levels may be omitted, in which case, the privilege levels of the associated 260event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege 261levels are subject to permissions. When sampling on multiple events, branch stack sampling 262is enabled for all the sampling events. The sampled branch type is the same for all events. 263The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k 264Note that this feature may not be available on all processors. 265 266--weight:: 267Enable weightened sampling. An additional weight is recorded per sample and can be 268displayed with the weight and local_weight sort keys. This currently works for TSX 269abort events and some memory events in precise mode on modern Intel CPUs. 270 271--transaction:: 272Record transaction flags for transaction related events. 273 274--per-thread:: 275Use per-thread mmaps. By default per-cpu mmaps are created. This option 276overrides that and uses per-thread mmaps. A side-effect of that is that 277inheritance is automatically disabled. --per-thread is ignored with a warning 278if combined with -a or -C options. 279 280-D:: 281--delay=:: 282After starting the program, wait msecs before measuring. This is useful to 283filter out the startup phase of the program, which is often very different. 284 285-I:: 286--intr-regs:: 287Capture machine state (registers) at interrupt, i.e., on counter overflows for 288each sample. List of captured registers depends on the architecture. This option 289is off by default. It is possible to select the registers to sample using their 290symbolic names, e.g. on x86, ax, si. To list the available registers use 291--intr-regs=\?. To name registers, pass a comma separated list such as 292--intr-regs=ax,bx. The list of register is architecture dependent. 293 294 295--running-time:: 296Record running and enabled time for read events (:S) 297 298-k:: 299--clockid:: 300Sets the clock id to use for the various time fields in the perf_event_type 301records. See clock_gettime(). In particular CLOCK_MONOTONIC and 302CLOCK_MONOTONIC_RAW are supported, some events might also allow 303CLOCK_BOOTTIME, CLOCK_REALTIME and CLOCK_TAI. 304 305-S:: 306--snapshot:: 307Select AUX area tracing Snapshot Mode. This option is valid only with an 308AUX area tracing event. Optionally the number of bytes to capture per 309snapshot can be specified. In Snapshot Mode, trace data is captured only when 310signal SIGUSR2 is received. 311 312--proc-map-timeout:: 313When processing pre-existing threads /proc/XXX/mmap, it may take a long time, 314because the file may be huge. A time out is needed in such cases. 315This option sets the time out limit. The default value is 500 ms. 316 317--switch-events:: 318Record context switch events i.e. events of type PERF_RECORD_SWITCH or 319PERF_RECORD_SWITCH_CPU_WIDE. 320 321--clang-path:: 322Path to clang binary to use for compiling BPF scriptlets. 323 324--clang-opt:: 325Options passed to clang when compiling BPF scriptlets. 326 327SEE ALSO 328-------- 329linkperf:perf-stat[1], linkperf:perf-list[1] 330