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