1perf-top(1) 2=========== 3 4NAME 5---- 6perf-top - System profiling tool. 7 8SYNOPSIS 9-------- 10[verse] 11'perf top' [-e <EVENT> | --event=EVENT] [<options>] 12 13DESCRIPTION 14----------- 15This command generates and displays a performance counter profile in real time. 16 17 18OPTIONS 19------- 20-a:: 21--all-cpus:: 22 System-wide collection. (default) 23 24-c <count>:: 25--count=<count>:: 26 Event period to sample. 27 28-C <cpu-list>:: 29--cpu=<cpu>:: 30Monitor only on the list of CPUs provided. Multiple CPUs can be provided as a 31comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. 32Default is to monitor all CPUS. 33 34-d <seconds>:: 35--delay=<seconds>:: 36 Number of seconds to delay between refreshes. 37 38-e <event>:: 39--event=<event>:: 40 Select the PMU event. Selection can be a symbolic event name 41 (use 'perf list' to list all events) or a raw PMU event in the form 42 of rN where N is a hexadecimal value that represents the raw register 43 encoding with the layout of the event control registers as described 44 by entries in /sys/bus/event_source/devices/cpu/format/*. 45 46--filter=<filter>:: 47 Event filter. This option should follow an event selector (-e). For 48 syntax see linkperf:perf-record[1]. 49 50-E <entries>:: 51--entries=<entries>:: 52 Display this many functions. 53 54-f <count>:: 55--count-filter=<count>:: 56 Only display functions with more events than this. 57 58--group-sort-idx:: 59 Sort the output by the event at the index n in group. If n is invalid, 60 sort by the first event. It can support multiple groups with different 61 amount of events. WARNING: This should be used on grouped events. 62 63-F <freq>:: 64--freq=<freq>:: 65 Profile at this frequency. Use 'max' to use the currently maximum 66 allowed frequency, i.e. the value in the kernel.perf_event_max_sample_rate 67 sysctl. 68 69-i:: 70--inherit:: 71 Child tasks do not inherit counters. 72 73-k <path>:: 74--vmlinux=<path>:: 75 Path to vmlinux. Required for annotation functionality. 76 77--ignore-vmlinux:: 78 Ignore vmlinux files. 79 80--kallsyms=<file>:: 81 kallsyms pathname 82 83-m <pages>:: 84--mmap-pages=<pages>:: 85 Number of mmap data pages (must be a power of two) or size 86 specification in bytes with appended unit character - B/K/M/G. 87 The size is rounded up to the nearest power-of-two page value. 88 89-p <pid>:: 90--pid=<pid>:: 91 Profile events on existing Process ID (comma separated list). 92 93-t <tid>:: 94--tid=<tid>:: 95 Profile events on existing thread ID (comma separated list). 96 97-u:: 98--uid=:: 99 Record events in threads owned by uid. Name or number. 100 101-r <priority>:: 102--realtime=<priority>:: 103 Collect data with this RT SCHED_FIFO priority. 104 105--sym-annotate=<symbol>:: 106 Annotate this symbol. 107 108-K:: 109--hide_kernel_symbols:: 110 Hide kernel symbols. 111 112-U:: 113--hide_user_symbols:: 114 Hide user symbols. 115 116--demangle-kernel:: 117 Demangle kernel symbols. 118 119-D:: 120--dump-symtab:: 121 Dump the symbol table used for profiling. 122 123-v:: 124--verbose:: 125 Be more verbose (show counter open errors, etc). 126 127-z:: 128--zero:: 129 Zero history across display updates. 130 131-s:: 132--sort:: 133 Sort by key(s): pid, comm, dso, symbol, parent, srcline, weight, 134 local_weight, abort, in_tx, transaction, overhead, sample, period. 135 Please see description of --sort in the perf-report man page. 136 137--fields=:: 138 Specify output field - multiple keys can be specified in CSV format. 139 Following fields are available: 140 overhead, overhead_sys, overhead_us, overhead_children, sample and period. 141 Also it can contain any sort key(s). 142 143 By default, every sort keys not specified in --field will be appended 144 automatically. 145 146-n:: 147--show-nr-samples:: 148 Show a column with the number of samples. 149 150--show-total-period:: 151 Show a column with the sum of periods. 152 153--dsos:: 154 Only consider symbols in these dsos. This option will affect the 155 percentage of the overhead column. See --percentage for more info. 156 157--comms:: 158 Only consider symbols in these comms. This option will affect the 159 percentage of the overhead column. See --percentage for more info. 160 161--symbols:: 162 Only consider these symbols. This option will affect the 163 percentage of the overhead column. See --percentage for more info. 164 165-M:: 166--disassembler-style=:: Set disassembler style for objdump. 167 168--addr2line=<path>:: 169 Path to addr2line binary. 170 171--objdump=<path>:: 172 Path to objdump binary. 173 174--prefix=PREFIX:: 175--prefix-strip=N:: 176 Remove first N entries from source file path names in executables 177 and add PREFIX. This allows to display source code compiled on systems 178 with different file system layout. 179 180--source:: 181 Interleave source code with assembly code. Enabled by default, 182 disable with --no-source. 183 184--asm-raw:: 185 Show raw instruction encoding of assembly instructions. 186 187-g:: 188 Enables call-graph (stack chain/backtrace) recording. 189 190--call-graph [mode,type,min[,limit],order[,key][,branch]]:: 191 Setup and enable call-graph (stack chain/backtrace) recording, 192 implies -g. See `--call-graph` section in perf-record and 193 perf-report man pages for details. 194 195--children:: 196 Accumulate callchain of children to parent entry so that then can 197 show up in the output. The output will have a new "Children" column 198 and will be sorted on the data. It requires -g/--call-graph option 199 enabled. See the `overhead calculation' section for more details. 200 Enabled by default, disable with --no-children. 201 202--max-stack:: 203 Set the stack depth limit when parsing the callchain, anything 204 beyond the specified depth will be ignored. This is a trade-off 205 between information loss and faster processing especially for 206 workloads that can have a very long callchain stack. 207 208 Default: /proc/sys/kernel/perf_event_max_stack when present, 127 otherwise. 209 210--ignore-callees=<regex>:: 211 Ignore callees of the function(s) matching the given regex. 212 This has the effect of collecting the callers of each such 213 function into one place in the call-graph tree. 214 215--percent-limit:: 216 Do not show entries which have an overhead under that percent. 217 (Default: 0). 218 219--percentage:: 220 Determine how to display the overhead percentage of filtered entries. 221 Filters can be applied by --comms, --dsos and/or --symbols options and 222 Zoom operations on the TUI (thread, dso, etc). 223 224 "relative" means it's relative to filtered entries only so that the 225 sum of shown entries will be always 100%. "absolute" means it retains 226 the original value before and after the filter is applied. 227 228-w:: 229--column-widths=<width[,width...]>:: 230 Force each column width to the provided list, for large terminal 231 readability. 0 means no limit (default behavior). 232 233--proc-map-timeout:: 234 When processing pre-existing threads /proc/XXX/mmap, it may take 235 a long time, because the file may be huge. A time out is needed 236 in such cases. 237 This option sets the time out limit. The default value is 500 ms. 238 239 240-b:: 241--branch-any:: 242 Enable taken branch stack sampling. Any type of taken branch may be sampled. 243 This is a shortcut for --branch-filter any. See --branch-filter for more infos. 244 245-j:: 246--branch-filter:: 247 Enable taken branch stack sampling. Each sample captures a series of consecutive 248 taken branches. The number of branches captured with each sample depends on the 249 underlying hardware, the type of branches of interest, and the executed code. 250 It is possible to select the types of branches captured by enabling filters. 251 For a full list of modifiers please see the perf record manpage. 252 253 The option requires at least one branch type among any, any_call, any_ret, ind_call, cond. 254 The privilege levels may be omitted, in which case, the privilege levels of the associated 255 event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege 256 levels are subject to permissions. When sampling on multiple events, branch stack sampling 257 is enabled for all the sampling events. The sampled branch type is the same for all events. 258 The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k 259 Note that this feature may not be available on all processors. 260 261--branch-history:: 262 Add the addresses of sampled taken branches to the callstack. 263 This allows to examine the path the program took to each sample. 264 265--raw-trace:: 266 When displaying traceevent output, do not use print fmt or plugins. 267 268-H:: 269--hierarchy:: 270 Enable hierarchical output. In the hierarchy mode, each sort key groups 271 samples based on the criteria and then sub-divide it using the lower 272 level sort key. 273 274 For example, in normal output: 275 276 perf report -s dso,sym 277 # 278 # Overhead Shared Object Symbol 279 # ........ ................. ........... 280 50.00% [kernel.kallsyms] [k] kfunc1 281 20.00% perf [.] foo 282 15.00% [kernel.kallsyms] [k] kfunc2 283 10.00% perf [.] bar 284 5.00% libc.so [.] libcall 285 286 In hierarchy output: 287 288 perf report -s dso,sym --hierarchy 289 # 290 # Overhead Shared Object / Symbol 291 # .......... ...................... 292 65.00% [kernel.kallsyms] 293 50.00% [k] kfunc1 294 15.00% [k] kfunc2 295 30.00% perf 296 20.00% [.] foo 297 10.00% [.] bar 298 5.00% libc.so 299 5.00% [.] libcall 300 301--overwrite:: 302 Enable this to use just the most recent records, which helps in high core count 303 machines such as Knights Landing/Mill, but right now is disabled by default as 304 the pausing used in this technique is leading to loss of metadata events such 305 as PERF_RECORD_MMAP which makes 'perf top' unable to resolve samples, leading 306 to lots of unknown samples appearing on the UI. Enable this if you are in such 307 machines and profiling a workload that doesn't creates short lived threads and/or 308 doesn't uses many executable mmap operations. Work is being planed to solve 309 this situation, till then, this will remain disabled by default. 310 311--force:: 312 Don't do ownership validation. 313 314--num-thread-synthesize:: 315 The number of threads to run when synthesizing events for existing processes. 316 By default, the number of threads equals to the number of online CPUs. 317 318--namespaces:: 319 Record events of type PERF_RECORD_NAMESPACES and display it with the 320 'cgroup_id' sort key. 321 322-G name:: 323--cgroup name:: 324monitor only in the container (cgroup) called "name". This option is available only 325in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to 326container "name" are monitored when they run on the monitored CPUs. Multiple cgroups 327can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup 328to first event, second cgroup to second event and so on. It is possible to provide 329an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have 330corresponding events, i.e., they always refer to events defined earlier on the command 331line. If the user wants to track multiple events for a specific cgroup, the user can 332use '-e e1 -e e2 -G foo,foo' or just use '-e e1 -e e2 -G foo'. 333 334--all-cgroups:: 335 Record events of type PERF_RECORD_CGROUP and display it with the 336 'cgroup' sort key. 337 338--switch-on EVENT_NAME:: 339 Only consider events after this event is found. 340 341 E.g.: 342 343 Find out where broadcast packets are handled 344 345 perf probe -L icmp_rcv 346 347 Insert a probe there: 348 349 perf probe icmp_rcv:59 350 351 Start perf top and ask it to only consider the cycles events when a 352 broadcast packet arrives This will show a menu with two entries and 353 will start counting when a broadcast packet arrives: 354 355 perf top -e cycles,probe:icmp_rcv --switch-on=probe:icmp_rcv 356 357 Alternatively one can ask for a group and then two overhead columns 358 will appear, the first for cycles and the second for the switch-on event. 359 360 perf top -e '{cycles,probe:icmp_rcv}' --switch-on=probe:icmp_rcv 361 362 This may be interesting to measure a workload only after some initialization 363 phase is over, i.e. insert a perf probe at that point and use the above 364 examples replacing probe:icmp_rcv with the just-after-init probe. 365 366--switch-off EVENT_NAME:: 367 Stop considering events after this event is found. 368 369--show-on-off-events:: 370 Show the --switch-on/off events too. This has no effect in 'perf top' now 371 but probably we'll make the default not to show the switch-on/off events 372 on the --group mode and if there is only one event besides the off/on ones, 373 go straight to the histogram browser, just like 'perf top' with no events 374 explicitly specified does. 375 376--stitch-lbr:: 377 Show callgraph with stitched LBRs, which may have more complete 378 callgraph. The option must be used with --call-graph lbr recording. 379 Disabled by default. In common cases with call stack overflows, 380 it can recreate better call stacks than the default lbr call stack 381 output. But this approach is not foolproof. There can be cases 382 where it creates incorrect call stacks from incorrect matches. 383 The known limitations include exception handing such as 384 setjmp/longjmp will have calls/returns not match. 385 386ifdef::HAVE_LIBPFM[] 387--pfm-events events:: 388Select a PMU event using libpfm4 syntax (see http://perfmon2.sf.net) 389including support for event filters. For example '--pfm-events 390inst_retired:any_p:u:c=1:i'. More than one event can be passed to the 391option using the comma separator. Hardware events and generic hardware 392events cannot be mixed together. The latter must be used with the -e 393option. The -e option and this one can be mixed and matched. Events 394can be grouped using the {} notation. 395endif::HAVE_LIBPFM[] 396 397INTERACTIVE PROMPTING KEYS 398-------------------------- 399 400[d]:: 401 Display refresh delay. 402 403[e]:: 404 Number of entries to display. 405 406[E]:: 407 Event to display when multiple counters are active. 408 409[f]:: 410 Profile display filter (>= hit count). 411 412[F]:: 413 Annotation display filter (>= % of total). 414 415[s]:: 416 Annotate symbol. 417 418[S]:: 419 Stop annotation, return to full profile display. 420 421[K]:: 422 Hide kernel symbols. 423 424[U]:: 425 Hide user symbols. 426 427[z]:: 428 Toggle event count zeroing across display updates. 429 430[qQ]:: 431 Quit. 432 433Pressing any unmapped key displays a menu, and prompts for input. 434 435include::callchain-overhead-calculation.txt[] 436 437SEE ALSO 438-------- 439linkperf:perf-stat[1], linkperf:perf-list[1], linkperf:perf-report[1] 440