Lines Matching +full:over +full:- +full:sampling
1 perf-top(1)
5 ----
6 perf-top - System profiling tool.
9 --------
11 'perf top' [-e <EVENT> | --event=EVENT] [<options>]
14 -----------
19 -------
20 -a::
21 --all-cpus::
22 System-wide collection. (default)
24 -c <count>::
25 --count=<count>::
28 -C <cpu-list>::
29 --cpu=<cpu>::
31 comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
34 -d <seconds>::
35 --delay=<seconds>::
38 -e <event>::
39 --event=<event>::
46 -E <entries>::
47 --entries=<entries>::
50 -f <count>::
51 --count-filter=<count>::
54 --group-sort-idx::
59 -F <freq>::
60 --freq=<freq>::
65 -i::
66 --inherit::
69 -k <path>::
70 --vmlinux=<path>::
73 --ignore-vmlinux::
76 --kallsyms=<file>::
79 -m <pages>::
80 --mmap-pages=<pages>::
82 specification with appended unit character - B/K/M/G. The
85 -p <pid>::
86 --pid=<pid>::
89 -t <tid>::
90 --tid=<tid>::
93 -u::
94 --uid=::
97 -r <priority>::
98 --realtime=<priority>::
101 --sym-annotate=<symbol>::
104 -K::
105 --hide_kernel_symbols::
108 -U::
109 --hide_user_symbols::
112 --demangle-kernel::
115 -D::
116 --dump-symtab::
119 -v::
120 --verbose::
123 -z::
124 --zero::
127 -s::
128 --sort::
131 Please see description of --sort in the perf-report man page.
133 --fields=::
134 Specify output field - multiple keys can be specified in CSV format.
139 By default, every sort keys not specified in --field will be appended
142 -n::
143 --show-nr-samples::
146 --show-total-period::
149 --dsos::
151 percentage of the overhead column. See --percentage for more info.
153 --comms::
155 percentage of the overhead column. See --percentage for more info.
157 --symbols::
159 percentage of the overhead column. See --percentage for more info.
161 -M::
162 --disassembler-style=:: Set disassembler style for objdump.
164 --addr2line=<path>::
167 --objdump=<path>::
170 --prefix=PREFIX::
171 --prefix-strip=N::
176 --source::
178 disable with --no-source.
180 --asm-raw::
183 -g::
184 Enables call-graph (stack chain/backtrace) recording.
186 --call-graph [mode,type,min[,limit],order[,key][,branch]]::
187 Setup and enable call-graph (stack chain/backtrace) recording,
188 implies -g. See `--call-graph` section in perf-record and
189 perf-report man pages for details.
191 --children::
194 and will be sorted on the data. It requires -g/--call-graph option
196 Enabled by default, disable with --no-children.
198 --max-stack::
200 beyond the specified depth will be ignored. This is a trade-off
206 --ignore-callees=<regex>::
209 function into one place in the call-graph tree.
211 --percent-limit::
215 --percentage::
217 Filters can be applied by --comms, --dsos and/or --symbols options and
224 -w::
225 --column-widths=<width[,width...]>::
229 --proc-map-timeout::
230 When processing pre-existing threads /proc/XXX/mmap, it may take
236 -b::
237 --branch-any::
238 Enable taken branch stack sampling. Any type of taken branch may be sampled.
239 This is a shortcut for --branch-filter any. See --branch-filter for more infos.
241 -j::
242 --branch-filter::
243 Enable taken branch stack sampling. Each sample captures a series of consecutive
252 levels are subject to permissions. When sampling on multiple events, branch stack sampling
253 is enabled for all the sampling events. The sampled branch type is the same for all events.
254 The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k
257 --branch-history::
261 --raw-trace::
264 --hierarchy::
267 --overwrite::
277 --force::
280 --num-thread-synthesize::
284 --namespaces::
288 -G name::
289 --cgroup name::
291 in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
295 an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
298 use '-e e1 -e e2 -G foo,foo' or just use '-e e1 -e e2 -G foo'.
300 --all-cgroups::
304 --switch-on EVENT_NAME::
311 perf probe -L icmp_rcv
321 perf top -e cycles,probe:icmp_rcv --switch-on=probe:icmp_rcv
324 will appear, the first for cycles and the second for the switch-on event.
326 perf top -e '{cycles,probe:icmp_rcv}' --switch-on=probe:icmp_rcv
329 phase is over, i.e. insert a perf probe at that point and use the above
330 examples replacing probe:icmp_rcv with the just-after-init probe.
332 --switch-off EVENT_NAME::
335 --show-on-off-events::
336 Show the --switch-on/off events too. This has no effect in 'perf top' now
337 but probably we'll make the default not to show the switch-on/off events
338 on the --group mode and if there is only one event besides the off/on ones,
342 --stitch-lbr::
344 callgraph. The option must be used with --call-graph lbr recording.
353 --pfm-events events::
355 including support for event filters. For example '--pfm-events
358 events cannot be mixed together. The latter must be used with the -e
359 option. The -e option and this one can be mixed and matched. Events
364 --------------------------
401 include::callchain-overhead-calculation.txt[]
404 --------
405 linkperf:perf-stat[1], linkperf:perf-list[1], linkperf:perf-report[1]