1perf-script(1) 2============= 3 4NAME 5---- 6perf-script - Read perf.data (created by perf record) and display trace output 7 8SYNOPSIS 9-------- 10[verse] 11'perf script' [<options>] 12'perf script' [<options>] record <script> [<record-options>] <command> 13'perf script' [<options>] report <script> [script-args] 14'perf script' [<options>] <script> <required-script-args> [<record-options>] <command> 15'perf script' [<options>] <top-script> [script-args] 16 17DESCRIPTION 18----------- 19This command reads the input file and displays the trace recorded. 20 21There are several variants of perf script: 22 23 'perf script' to see a detailed trace of the workload that was 24 recorded. 25 26 You can also run a set of pre-canned scripts that aggregate and 27 summarize the raw trace data in various ways (the list of scripts is 28 available via 'perf script -l'). The following variants allow you to 29 record and run those scripts: 30 31 'perf script record <script> <command>' to record the events required 32 for 'perf script report'. <script> is the name displayed in the 33 output of 'perf script --list' i.e. the actual script name minus any 34 language extension. If <command> is not specified, the events are 35 recorded using the -a (system-wide) 'perf record' option. 36 37 'perf script report <script> [args]' to run and display the results 38 of <script>. <script> is the name displayed in the output of 'perf 39 script --list' i.e. the actual script name minus any language 40 extension. The perf.data output from a previous run of 'perf script 41 record <script>' is used and should be present for this command to 42 succeed. [args] refers to the (mainly optional) args expected by 43 the script. 44 45 'perf script <script> <required-script-args> <command>' to both 46 record the events required for <script> and to run the <script> 47 using 'live-mode' i.e. without writing anything to disk. <script> 48 is the name displayed in the output of 'perf script --list' i.e. the 49 actual script name minus any language extension. If <command> is 50 not specified, the events are recorded using the -a (system-wide) 51 'perf record' option. If <script> has any required args, they 52 should be specified before <command>. This mode doesn't allow for 53 optional script args to be specified; if optional script args are 54 desired, they can be specified using separate 'perf script record' 55 and 'perf script report' commands, with the stdout of the record step 56 piped to the stdin of the report script, using the '-o -' and '-i -' 57 options of the corresponding commands. 58 59 'perf script <top-script>' to both record the events required for 60 <top-script> and to run the <top-script> using 'live-mode' 61 i.e. without writing anything to disk. <top-script> is the name 62 displayed in the output of 'perf script --list' i.e. the actual 63 script name minus any language extension; a <top-script> is defined 64 as any script name ending with the string 'top'. 65 66 [<record-options>] can be passed to the record steps of 'perf script 67 record' and 'live-mode' variants; this isn't possible however for 68 <top-script> 'live-mode' or 'perf script report' variants. 69 70 See the 'SEE ALSO' section for links to language-specific 71 information on how to write and run your own trace scripts. 72 73OPTIONS 74------- 75<command>...:: 76 Any command you can specify in a shell. 77 78-D:: 79--dump-raw-trace=:: 80 Display verbose dump of the trace data. 81 82-L:: 83--Latency=:: 84 Show latency attributes (irqs/preemption disabled, etc). 85 86-l:: 87--list=:: 88 Display a list of available trace scripts. 89 90-s ['lang']:: 91--script=:: 92 Process trace data with the given script ([lang]:script[.ext]). 93 If the string 'lang' is specified in place of a script name, a 94 list of supported languages will be displayed instead. 95 96-g:: 97--gen-script=:: 98 Generate perf-script.[ext] starter script for given language, 99 using current perf.data. 100 101-a:: 102 Force system-wide collection. Scripts run without a <command> 103 normally use -a by default, while scripts run with a <command> 104 normally don't - this option allows the latter to be run in 105 system-wide mode. 106 107-i:: 108--input=:: 109 Input file name. (default: perf.data unless stdin is a fifo) 110 111-d:: 112--debug-mode:: 113 Do various checks like samples ordering and lost events. 114 115-F:: 116--fields:: 117 Comma separated list of fields to print. Options are: 118 comm, tid, pid, time, cpu, event, trace, ip, sym, dso, addr, symoff, 119 srcline, period, iregs, brstack, brstacksym, flags, bpf-output, 120 callindent, insn, insnlen. Field list can be prepended with the type, trace, sw or hw, 121 to indicate to which event type the field list applies. 122 e.g., -F sw:comm,tid,time,ip,sym and -F trace:time,cpu,trace 123 124 perf script -F <fields> 125 126 is equivalent to: 127 128 perf script -F trace:<fields> -F sw:<fields> -F hw:<fields> 129 130 i.e., the specified fields apply to all event types if the type string 131 is not given. 132 133 The arguments are processed in the order received. A later usage can 134 reset a prior request. e.g.: 135 136 -F trace: -F comm,tid,time,ip,sym 137 138 The first -F suppresses trace events (field list is ""), but then the 139 second invocation sets the fields to comm,tid,time,ip,sym. In this case a 140 warning is given to the user: 141 142 "Overriding previous field request for all events." 143 144 Alternatively, consider the order: 145 146 -F comm,tid,time,ip,sym -F trace: 147 148 The first -F sets the fields for all events and the second -F 149 suppresses trace events. The user is given a warning message about 150 the override, and the result of the above is that only S/W and H/W 151 events are displayed with the given fields. 152 153 For the 'wildcard' option if a user selected field is invalid for an 154 event type, a message is displayed to the user that the option is 155 ignored for that type. For example: 156 157 $ perf script -F comm,tid,trace 158 'trace' not valid for hardware events. Ignoring. 159 'trace' not valid for software events. Ignoring. 160 161 Alternatively, if the type is given an invalid field is specified it 162 is an error. For example: 163 164 perf script -v -F sw:comm,tid,trace 165 'trace' not valid for software events. 166 167 At this point usage is displayed, and perf-script exits. 168 169 The flags field is synthesized and may have a value when Instruction 170 Trace decoding. The flags are "bcrosyiABEx" which stand for branch, 171 call, return, conditional, system, asynchronous, interrupt, 172 transaction abort, trace begin, trace end, and in transaction, 173 respectively. Known combinations of flags are printed more nicely e.g. 174 "call" for "bc", "return" for "br", "jcc" for "bo", "jmp" for "b", 175 "int" for "bci", "iret" for "bri", "syscall" for "bcs", "sysret" for "brs", 176 "async" for "by", "hw int" for "bcyi", "tx abrt" for "bA", "tr strt" for "bB", 177 "tr end" for "bE". However the "x" flag will be display separately in those 178 cases e.g. "jcc (x)" for a condition branch within a transaction. 179 180 The callindent field is synthesized and may have a value when 181 Instruction Trace decoding. For calls and returns, it will display the 182 name of the symbol indented with spaces to reflect the stack depth. 183 184 When doing instruction trace decoding insn and insnlen give the 185 instruction bytes and the instruction length of the current 186 instruction. 187 188 Finally, a user may not set fields to none for all event types. 189 i.e., -F "" is not allowed. 190 191 The brstack output includes branch related information with raw addresses using the 192 /v/v/v/v/ syntax in the following order: 193 FROM: branch source instruction 194 TO : branch target instruction 195 M/P/-: M=branch target mispredicted or branch direction was mispredicted, P=target predicted or direction predicted, -=not supported 196 X/- : X=branch inside a transactional region, -=not in transaction region or not supported 197 A/- : A=TSX abort entry, -=not aborted region or not supported 198 199 The brstacksym is identical to brstack, except that the FROM and TO addresses are printed in a symbolic form if possible. 200 201-k:: 202--vmlinux=<file>:: 203 vmlinux pathname 204 205--kallsyms=<file>:: 206 kallsyms pathname 207 208--symfs=<directory>:: 209 Look for files with symbols relative to this directory. 210 211-G:: 212--hide-call-graph:: 213 When printing symbols do not display call chain. 214 215--stop-bt:: 216 Stop display of callgraph at these symbols 217 218-C:: 219--cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can 220 be provided as a comma-separated list with no space: 0,1. Ranges of 221 CPUs are specified with -: 0-2. Default is to report samples on all 222 CPUs. 223 224-c:: 225--comms=:: 226 Only display events for these comms. CSV that understands 227 file://filename entries. 228 229--pid=:: 230 Only show events for given process ID (comma separated list). 231 232--tid=:: 233 Only show events for given thread ID (comma separated list). 234 235-I:: 236--show-info:: 237 Display extended information about the perf.data file. This adds 238 information which may be very large and thus may clutter the display. 239 It currently includes: cpu and numa topology of the host system. 240 It can only be used with the perf script report mode. 241 242--show-kernel-path:: 243 Try to resolve the path of [kernel.kallsyms] 244 245--show-task-events 246 Display task related events (e.g. FORK, COMM, EXIT). 247 248--show-mmap-events 249 Display mmap related events (e.g. MMAP, MMAP2). 250 251--show-switch-events 252 Display context switch events i.e. events of type PERF_RECORD_SWITCH or 253 PERF_RECORD_SWITCH_CPU_WIDE. 254 255--demangle:: 256 Demangle symbol names to human readable form. It's enabled by default, 257 disable with --no-demangle. 258 259--demangle-kernel:: 260 Demangle kernel symbol names to human readable form (for C++ kernels). 261 262--header 263 Show perf.data header. 264 265--header-only 266 Show only perf.data header. 267 268--itrace:: 269 Options for decoding instruction tracing data. The options are: 270 271include::itrace.txt[] 272 273 To disable decoding entirely, use --no-itrace. 274 275--full-source-path:: 276 Show the full path for source files for srcline output. 277 278--max-stack:: 279 Set the stack depth limit when parsing the callchain, anything 280 beyond the specified depth will be ignored. This is a trade-off 281 between information loss and faster processing especially for 282 workloads that can have a very long callchain stack. 283 Note that when using the --itrace option the synthesized callchain size 284 will override this value if the synthesized callchain size is bigger. 285 286 Default: 127 287 288--ns:: 289 Use 9 decimal places when displaying time (i.e. show the nanoseconds) 290 291-f:: 292--force:: 293 Don't do ownership validation. 294 295--time:: 296 Only analyze samples within given time window: <start>,<stop>. Times 297 have the format seconds.microseconds. If start is not given (i.e., time 298 string is ',x.y') then analysis starts at the beginning of the file. If 299 stop time is not given (i.e, time string is 'x.y,') then analysis goes 300 to end of file. 301 302SEE ALSO 303-------- 304linkperf:perf-record[1], linkperf:perf-script-perl[1], 305linkperf:perf-script-python[1] 306