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--filter=<filter>:: 59 Event filter. 60 61-a:: 62--all-cpus:: 63 System-wide collection from all CPUs. 64 65-l:: 66 Scale counter values. 67 68-p:: 69--pid=:: 70 Record events on existing process ID (comma separated list). 71 72-t:: 73--tid=:: 74 Record events on existing thread ID (comma separated list). 75 This option also disables inheritance by default. Enable it by adding 76 --inherit. 77 78-u:: 79--uid=:: 80 Record events in threads owned by uid. Name or number. 81 82-r:: 83--realtime=:: 84 Collect data with this RT SCHED_FIFO priority. 85 86--no-buffering:: 87 Collect data without buffering. 88 89-c:: 90--count=:: 91 Event period to sample. 92 93-o:: 94--output=:: 95 Output file name. 96 97-i:: 98--no-inherit:: 99 Child tasks do not inherit counters. 100-F:: 101--freq=:: 102 Profile at this frequency. 103 104-m:: 105--mmap-pages=:: 106 Number of mmap data pages (must be a power of two) or size 107 specification with appended unit character - B/K/M/G. The 108 size is rounded up to have nearest pages power of two value. 109 110-g:: 111 Enables call-graph (stack chain/backtrace) recording. 112 113--call-graph:: 114 Setup and enable call-graph (stack chain/backtrace) recording, 115 implies -g. 116 117 Allows specifying "fp" (frame pointer) or "dwarf" 118 (DWARF's CFI - Call Frame Information) or "lbr" 119 (Hardware Last Branch Record facility) as the method to collect 120 the information used to show the call graphs. 121 122 In some systems, where binaries are build with gcc 123 --fomit-frame-pointer, using the "fp" method will produce bogus 124 call graphs, using "dwarf", if available (perf tools linked to 125 the libunwind library) should be used instead. 126 Using the "lbr" method doesn't require any compiler options. It 127 will produce call graphs from the hardware LBR registers. The 128 main limition is that it is only available on new Intel 129 platforms, such as Haswell. It can only get user call chain. It 130 doesn't work with branch stack sampling at the same time. 131 132-q:: 133--quiet:: 134 Don't print any message, useful for scripting. 135 136-v:: 137--verbose:: 138 Be more verbose (show counter open errors, etc). 139 140-s:: 141--stat:: 142 Per thread counts. 143 144-d:: 145--data:: 146 Sample addresses. 147 148-T:: 149--timestamp:: 150 Sample timestamps. Use it with 'perf report -D' to see the timestamps, 151 for instance. 152 153-n:: 154--no-samples:: 155 Don't sample. 156 157-R:: 158--raw-samples:: 159Collect raw sample records from all opened counters (default for tracepoint counters). 160 161-C:: 162--cpu:: 163Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a 164comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. 165In per-thread mode with inheritance mode on (default), samples are captured only when 166the thread executes on the designated CPUs. Default is to monitor all CPUs. 167 168-N:: 169--no-buildid-cache:: 170Do not update the buildid cache. This saves some overhead in situations 171where the information in the perf.data file (which includes buildids) 172is sufficient. 173 174-G name,...:: 175--cgroup name,...:: 176monitor only in the container (cgroup) called "name". This option is available only 177in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to 178container "name" are monitored when they run on the monitored CPUs. Multiple cgroups 179can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup 180to first event, second cgroup to second event and so on. It is possible to provide 181an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have 182corresponding events, i.e., they always refer to events defined earlier on the command 183line. 184 185-b:: 186--branch-any:: 187Enable taken branch stack sampling. Any type of taken branch may be sampled. 188This is a shortcut for --branch-filter any. See --branch-filter for more infos. 189 190-j:: 191--branch-filter:: 192Enable taken branch stack sampling. Each sample captures a series of consecutive 193taken branches. The number of branches captured with each sample depends on the 194underlying hardware, the type of branches of interest, and the executed code. 195It is possible to select the types of branches captured by enabling filters. The 196following filters are defined: 197 198 - any: any type of branches 199 - any_call: any function call or system call 200 - any_ret: any function return or system call return 201 - ind_call: any indirect branch 202 - u: only when the branch target is at the user level 203 - k: only when the branch target is in the kernel 204 - hv: only when the target is at the hypervisor level 205 - in_tx: only when the target is in a hardware transaction 206 - no_tx: only when the target is not in a hardware transaction 207 - abort_tx: only when the target is a hardware transaction abort 208 - cond: conditional branches 209 210+ 211The option requires at least one branch type among any, any_call, any_ret, ind_call, cond. 212The privilege levels may be omitted, in which case, the privilege levels of the associated 213event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege 214levels are subject to permissions. When sampling on multiple events, branch stack sampling 215is enabled for all the sampling events. The sampled branch type is the same for all events. 216The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k 217Note that this feature may not be available on all processors. 218 219--weight:: 220Enable weightened sampling. An additional weight is recorded per sample and can be 221displayed with the weight and local_weight sort keys. This currently works for TSX 222abort events and some memory events in precise mode on modern Intel CPUs. 223 224--transaction:: 225Record transaction flags for transaction related events. 226 227--per-thread:: 228Use per-thread mmaps. By default per-cpu mmaps are created. This option 229overrides that and uses per-thread mmaps. A side-effect of that is that 230inheritance is automatically disabled. --per-thread is ignored with a warning 231if combined with -a or -C options. 232 233-D:: 234--delay=:: 235After starting the program, wait msecs before measuring. This is useful to 236filter out the startup phase of the program, which is often very different. 237 238-I:: 239--intr-regs:: 240Capture machine state (registers) at interrupt, i.e., on counter overflows for 241each sample. List of captured registers depends on the architecture. This option 242is off by default. 243 244--running-time:: 245Record running and enabled time for read events (:S) 246 247SEE ALSO 248-------- 249linkperf:perf-stat[1], linkperf:perf-list[1] 250