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.
62 Child tasks do not inherit counters.
66 Path to vmlinux. Required for annotation functionality.
69 --mmap-pages=<pages>::
70 Number of mmap data pages (must be a power of two) or size
71 specification with appended unit character - B/K/M/G. The
72 size is rounded up to have nearest pages power of two value.
76 Profile events on existing Process ID (comma separated list).
80 Profile events on existing thread ID (comma separated list).
84 Record events in threads owned by uid. Name or number.
87 --realtime=<priority>::
88 Collect data with this RT SCHED_FIFO priority.
91 --sym-annotate=<symbol>::
95 --hide_kernel_symbols::
104 Dump the symbol table used for profiling.
108 Be more verbose (show counter open errors, etc).
112 Zero history across display updates.
116 Sort by key(s): pid, comm, dso, symbol, parent, srcline, weight,
117 local_weight, abort, in_tx, transaction
121 Show a column with the number of samples.
123 --show-total-period::
124 Show a column with the sum of periods.
127 Only consider symbols in these dsos.
130 Only consider symbols in these comms.
133 Only consider these symbols.
136 --disassembler-style=:: Set disassembler style for objdump.
139 Interleave source code with assembly code. Enabled by default,
140 disable with --no-source.
143 Show raw instruction encoding of assembly instructions.
146 Enables call-graph (stack chain/backtrace) recording.
149 Setup and enable call-graph (stack chain/backtrace) recording,
153 Set the stack depth limit when parsing the callchain, anything
154 beyond the specified depth will be ignored. This is a trade-off
155 between information loss and faster processing especially for
156 workloads that can have a very long callchain stack.
160 --ignore-callees=<regex>::
161 Ignore callees of the function(s) matching the given regex.
162 This has the effect of collecting the callers of each such
163 function into one place in the call-graph tree.
166 Do not show entries which have an overhead under that percent.
169 INTERACTIVE PROMPTING KEYS
170 --------------------------
173 Display refresh delay.
176 Number of entries to display.
179 Event to display when multiple counters are active.
182 Profile display filter (>= hit count).
185 Annotation display filter (>= % of total).
191 Stop annotation, return to full profile display.
194 Toggle event count zeroing across display updates.
199 Pressing any unmapped key displays a menu, and prompts for input.
204 linkperf:perf-stat[1], linkperf:perf-list[1]