6 perf-top - System profiling tool.
11 'perf top' [-e <EVENT> | --event=EVENT] [<options>]
15 This command generates and displays a performance counter profile in real time.
22 System-wide collection. (default)
26 Event period to sample.
30 Monitor only on the list of CPUs provided. Multiple CPUs can be provided as a
31 comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
32 Default is to monitor all CPUS.
36 Number of seconds to delay between refreshes.
40 Select the PMU event. Selection can be a symbolic event name
41 (use 'perf list' to list all events) or a raw PMU
42 event (eventsel+umask) in the form of rNNN where NNN is a
43 hexadecimal event descriptor.
47 Display this many functions.
50 --count-filter=<count>::
51 Only display functions with more events than this.
54 Put the counters into a counter group.
58 Profile at this frequency. Use 'max' to use the currently maximum
59 allowed frequency, i.e. the value in the kernel.perf_event_max_sample_rate
64 Child tasks do not inherit counters.
68 Path to vmlinux. Required for annotation functionality.
77 --mmap-pages=<pages>::
78 Number of mmap data pages (must be a power of two) or size
79 specification with appended unit character - B/K/M/G. The
80 size is rounded up to have nearest pages power of two value.
84 Profile events on existing Process ID (comma separated list).
88 Profile events on existing thread ID (comma separated list).
92 Record events in threads owned by uid. Name or number.
95 --realtime=<priority>::
96 Collect data with this RT SCHED_FIFO priority.
98 --sym-annotate=<symbol>::
102 --hide_kernel_symbols::
106 --hide_user_symbols::
110 Demangle kernel symbols.
114 Dump the symbol table used for profiling.
118 Be more verbose (show counter open errors, etc).
122 Zero history across display updates.
126 Sort by key(s): pid, comm, dso, symbol, parent, srcline, weight,
127 local_weight, abort, in_tx, transaction, overhead, sample, period.
128 Please see description of --sort in the perf-report man page.
131 Specify output field - multiple keys can be specified in CSV format.
132 Following fields are available:
133 overhead, overhead_sys, overhead_us, overhead_children, sample and period.
134 Also it can contain any sort key(s).
136 By default, every sort keys not specified in --field will be appended
141 Show a column with the number of samples.
143 --show-total-period::
144 Show a column with the sum of periods.
147 Only consider symbols in these dsos. This option will affect the
148 percentage of the overhead column. See --percentage for more info.
151 Only consider symbols in these comms. This option will affect the
152 percentage of the overhead column. See --percentage for more info.
155 Only consider these symbols. This option will affect the
156 percentage of the overhead column. See --percentage for more info.
159 --disassembler-style=:: Set disassembler style for objdump.
162 Interleave source code with assembly code. Enabled by default,
163 disable with --no-source.
166 Show raw instruction encoding of assembly instructions.
169 Enables call-graph (stack chain/backtrace) recording.
171 --call-graph [mode,type,min[,limit],order[,key][,branch]]::
172 Setup and enable call-graph (stack chain/backtrace) recording,
173 implies -g. See `--call-graph` section in perf-record and
174 perf-report man pages for details.
177 Accumulate callchain of children to parent entry so that then can
178 show up in the output. The output will have a new "Children" column
179 and will be sorted on the data. It requires -g/--call-graph option
180 enabled. See the `overhead calculation' section for more details.
181 Enabled by default, disable with --no-children.
184 Set the stack depth limit when parsing the callchain, anything
185 beyond the specified depth will be ignored. This is a trade-off
186 between information loss and faster processing especially for
187 workloads that can have a very long callchain stack.
189 Default: /proc/sys/kernel/perf_event_max_stack when present, 127 otherwise.
191 --ignore-callees=<regex>::
192 Ignore callees of the function(s) matching the given regex.
193 This has the effect of collecting the callers of each such
194 function into one place in the call-graph tree.
197 Do not show entries which have an overhead under that percent.
201 Determine how to display the overhead percentage of filtered entries.
202 Filters can be applied by --comms, --dsos and/or --symbols options and
203 Zoom operations on the TUI (thread, dso, etc).
205 "relative" means it's relative to filtered entries only so that the
206 sum of shown entries will be always 100%. "absolute" means it retains
207 the original value before and after the filter is applied.
210 --column-widths=<width[,width...]>::
211 Force each column width to the provided list, for large terminal
212 readability. 0 means no limit (default behavior).
215 When processing pre-existing threads /proc/XXX/mmap, it may take
216 a long time, because the file may be huge. A time out is needed
218 This option sets the time out limit. The default value is 500 ms.
223 Enable taken branch stack sampling. Any type of taken branch may be sampled.
224 This is a shortcut for --branch-filter any. See --branch-filter for more infos.
228 Enable taken branch stack sampling. Each sample captures a series of consecutive
229 taken branches. The number of branches captured with each sample depends on the
230 underlying hardware, the type of branches of interest, and the executed code.
231 It is possible to select the types of branches captured by enabling filters.
232 For a full list of modifiers please see the perf record manpage.
234 The option requires at least one branch type among any, any_call, any_ret, ind_call, cond.
235 The privilege levels may be omitted, in which case, the privilege levels of the associated
236 event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege
237 levels are subject to permissions. When sampling on multiple events, branch stack sampling
238 is enabled for all the sampling events. The sampled branch type is the same for all events.
239 The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k
240 Note that this feature may not be available on all processors.
243 When displaying traceevent output, do not use print fmt or plugins.
246 Enable hierarchy output.
249 Enable this to use just the most recent records, which helps in high core count
250 machines such as Knights Landing/Mill, but right now is disabled by default as
251 the pausing used in this technique is leading to loss of metadata events such
252 as PERF_RECORD_MMAP which makes 'perf top' unable to resolve samples, leading
253 to lots of unknown samples appearing on the UI. Enable this if you are in such
254 machines and profiling a workload that doesn't creates short lived threads and/or
255 doesn't uses many executable mmap operations. Work is being planed to solve
256 this situation, till then, this will remain disabled by default.
259 Don't do ownership validation.
261 --num-thread-synthesize::
262 The number of threads to run when synthesizing events for existing processes.
263 By default, the number of threads equals to the number of online CPUs.
265 INTERACTIVE PROMPTING KEYS
266 --------------------------
269 Display refresh delay.
272 Number of entries to display.
275 Event to display when multiple counters are active.
278 Profile display filter (>= hit count).
281 Annotation display filter (>= % of total).
287 Stop annotation, return to full profile display.
296 Toggle event count zeroing across display updates.
301 Pressing any unmapped key displays a menu, and prompts for input.
303 include::callchain-overhead-calculation.txt[]
307 linkperf:perf-stat[1], linkperf:perf-list[1], linkperf:perf-report[1]