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 --filter=<filter>::
47 Event filter. This option should follow an event selector (-e). For
48 syntax see linkperf:perf-record[1].
50 -E <entries>::
51 --entries=<entries>::
54 -f <count>::
55 --count-filter=<count>::
58 --group-sort-idx::
63 -F <freq>::
64 --freq=<freq>::
69 -i::
70 --inherit::
73 -k <path>::
74 --vmlinux=<path>::
77 --ignore-vmlinux::
80 --kallsyms=<file>::
83 -m <pages>::
84 --mmap-pages=<pages>::
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.
89 -p <pid>::
90 --pid=<pid>::
93 -t <tid>::
94 --tid=<tid>::
97 -u::
98 --uid=::
101 -r <priority>::
102 --realtime=<priority>::
105 --sym-annotate=<symbol>::
108 -K::
109 --hide_kernel_symbols::
112 -U::
113 --hide_user_symbols::
116 --demangle-kernel::
119 -D::
120 --dump-symtab::
123 -v::
124 --verbose::
127 -z::
128 --zero::
131 -s::
132 --sort::
135 Please see description of --sort in the perf-report man page.
137 --fields=::
138 Specify output field - multiple keys can be specified in CSV format.
143 By default, every sort keys not specified in --field will be appended
146 -n::
147 --show-nr-samples::
150 --show-total-period::
153 --dsos::
155 percentage of the overhead column. See --percentage for more info.
157 --comms::
159 percentage of the overhead column. See --percentage for more info.
161 --symbols::
163 percentage of the overhead column. See --percentage for more info.
165 -M::
166 --disassembler-style=:: Set disassembler style for objdump.
168 --addr2line=<path>::
171 --objdump=<path>::
174 --prefix=PREFIX::
175 --prefix-strip=N::
180 --source::
182 disable with --no-source.
184 --asm-raw::
187 -g::
188 Enables call-graph (stack chain/backtrace) recording.
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.
195 --children::
198 and will be sorted on the data. It requires -g/--call-graph option
200 Enabled by default, disable with --no-children.
202 --max-stack::
204 beyond the specified depth will be ignored. This is a trade-off
210 --ignore-callees=<regex>::
213 function into one place in the call-graph tree.
215 --percent-limit::
219 --percentage::
221 Filters can be applied by --comms, --dsos and/or --symbols options and
228 -w::
229 --column-widths=<width[,width...]>::
233 --proc-map-timeout::
234 When processing pre-existing threads /proc/XXX/mmap, it may take
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.
245 -j::
246 --branch-filter::
247 Enable taken branch stack sampling. Each sample captures a series of consecutive
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
261 --branch-history::
265 --raw-trace::
268 -H::
269 --hierarchy::
271 samples based on the criteria and then sub-divide it using the lower
276 perf report -s dso,sym
288 perf report -s dso,sym --hierarchy
301 --overwrite::
311 --force::
314 --num-thread-synthesize::
318 --namespaces::
322 -G name::
323 --cgroup name::
325 in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
329 an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
332 use '-e e1 -e e2 -G foo,foo' or just use '-e e1 -e e2 -G foo'.
334 --all-cgroups::
338 --switch-on EVENT_NAME::
345 perf probe -L icmp_rcv
355 perf top -e cycles,probe:icmp_rcv --switch-on=probe:icmp_rcv
358 will appear, the first for cycles and the second for the switch-on event.
360 perf top -e '{cycles,probe:icmp_rcv}' --switch-on=probe:icmp_rcv
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.
366 --switch-off EVENT_NAME::
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,
376 --stitch-lbr::
378 callgraph. The option must be used with --call-graph lbr recording.
387 --pfm-events events::
389 including support for event filters. For example '--pfm-events
392 events cannot be mixed together. The latter must be used with the -e
393 option. The -e option and this one can be mixed and matched. Events
398 --------------------------
435 include::callchain-overhead-calculation.txt[]
438 --------
439 linkperf:perf-stat[1], linkperf:perf-list[1], linkperf:perf-report[1]