tjh.dev
Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
1perf-report(1)
2==============
3
4NAME
5----
6perf-report - Read perf.data (created by perf record) and display the profile
7
8SYNOPSIS
9--------
10[verse]
11'perf report' [-i <file> | --input=file]
12
13DESCRIPTION
14-----------
15This command displays the performance counter profile information recorded
16via perf record.
17
18OPTIONS
19-------
20-i::
21--input=::
22 Input file name. (default: perf.data unless stdin is a fifo)
23
24-v::
25--verbose::
26 Be more verbose. (show symbol address, etc)
27
28-d::
29--dsos=::
30 Only consider symbols in these dsos. CSV that understands
31 file://filename entries.
32-n::
33--show-nr-samples::
34 Show the number of samples for each symbol
35
36--showcpuutilization::
37 Show sample percentage for different cpu modes.
38
39-T::
40--threads::
41 Show per-thread event counters
42-c::
43--comms=::
44 Only consider symbols in these comms. CSV that understands
45 file://filename entries.
46-S::
47--symbols=::
48 Only consider these symbols. CSV that understands
49 file://filename entries.
50
51--symbol-filter=::
52 Only show symbols that match (partially) with this filter.
53
54-U::
55--hide-unresolved::
56 Only display entries resolved to a symbol.
57
58-s::
59--sort=::
60 Sort histogram entries by given key(s) - multiple keys can be specified
61 in CSV format. Following sort keys are available:
62 pid, comm, dso, symbol, parent, cpu, srcline, weight, local_weight.
63
64 Each key has following meaning:
65
66 - comm: command (name) of the task which can be read via /proc/<pid>/comm
67 - pid: command and tid of the task
68 - dso: name of library or module executed at the time of sample
69 - symbol: name of function executed at the time of sample
70 - parent: name of function matched to the parent regex filter. Unmatched
71 entries are displayed as "[other]".
72 - cpu: cpu number the task ran at the time of sample
73 - srcline: filename and line number executed at the time of sample. The
74 DWARF debuggin info must be provided.
75
76 By default, comm, dso and symbol keys are used.
77 (i.e. --sort comm,dso,symbol)
78
79 If --branch-stack option is used, following sort keys are also
80 available:
81 dso_from, dso_to, symbol_from, symbol_to, mispredict.
82
83 - dso_from: name of library or module branched from
84 - dso_to: name of library or module branched to
85 - symbol_from: name of function branched from
86 - symbol_to: name of function branched to
87 - mispredict: "N" for predicted branch, "Y" for mispredicted branch
88
89 And default sort keys are changed to comm, dso_from, symbol_from, dso_to
90 and symbol_to, see '--branch-stack'.
91
92-p::
93--parent=<regex>::
94 A regex filter to identify parent. The parent is a caller of this
95 function and searched through the callchain, thus it requires callchain
96 information recorded. The pattern is in the exteneded regex format and
97 defaults to "\^sys_|^do_page_fault", see '--sort parent'.
98
99-x::
100--exclude-other::
101 Only display entries with parent-match.
102
103-w::
104--column-widths=<width[,width...]>::
105 Force each column width to the provided list, for large terminal
106 readability.
107
108-t::
109--field-separator=::
110 Use a special separator character and don't pad with spaces, replacing
111 all occurrences of this separator in symbol names (and other output)
112 with a '.' character, that thus it's the only non valid separator.
113
114-D::
115--dump-raw-trace::
116 Dump raw trace in ASCII.
117
118-g [type,min[,limit],order]::
119--call-graph::
120 Display call chains using type, min percent threshold, optional print
121 limit and order.
122 type can be either:
123 - flat: single column, linear exposure of call chains.
124 - graph: use a graph tree, displaying absolute overhead rates.
125 - fractal: like graph, but displays relative rates. Each branch of
126 the tree is considered as a new profiled object. +
127
128 order can be either:
129 - callee: callee based call graph.
130 - caller: inverted caller based call graph.
131
132 Default: fractal,0.5,callee.
133
134-G::
135--inverted::
136 alias for inverted caller based call graph.
137
138--pretty=<key>::
139 Pretty printing style. key: normal, raw
140
141--stdio:: Use the stdio interface.
142
143--tui:: Use the TUI interface, that is integrated with annotate and allows
144 zooming into DSOs or threads, among other features. Use of --tui
145 requires a tty, if one is not present, as when piping to other
146 commands, the stdio interface is used.
147
148--gtk:: Use the GTK2 interface.
149
150-k::
151--vmlinux=<file>::
152 vmlinux pathname
153
154--kallsyms=<file>::
155 kallsyms pathname
156
157-m::
158--modules::
159 Load module symbols. WARNING: This should only be used with -k and
160 a LIVE kernel.
161
162-f::
163--force::
164 Don't complain, do it.
165
166--symfs=<directory>::
167 Look for files with symbols relative to this directory.
168
169-C::
170--cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can
171 be provided as a comma-separated list with no space: 0,1. Ranges of
172 CPUs are specified with -: 0-2. Default is to report samples on all
173 CPUs.
174
175-M::
176--disassembler-style=:: Set disassembler style for objdump.
177
178--source::
179 Interleave source code with assembly code. Enabled by default,
180 disable with --no-source.
181
182--asm-raw::
183 Show raw instruction encoding of assembly instructions.
184
185--show-total-period:: Show a column with the sum of periods.
186
187-I::
188--show-info::
189 Display extended information about the perf.data file. This adds
190 information which may be very large and thus may clutter the display.
191 It currently includes: cpu and numa topology of the host system.
192
193-b::
194--branch-stack::
195 Use the addresses of sampled taken branches instead of the instruction
196 address to build the histograms. To generate meaningful output, the
197 perf.data file must have been obtained using perf record -b or
198 perf record --branch-filter xxx where xxx is a branch filter option.
199 perf report is able to auto-detect whether a perf.data file contains
200 branch stacks and it will automatically switch to the branch view mode,
201 unless --no-branch-stack is used.
202
203--objdump=<path>::
204 Path to objdump binary.
205
206--group::
207 Show event group information together.
208
209--demangle::
210 Demangle symbol names to human readable form. It's enabled by default,
211 disable with --no-demangle.
212
213--percent-limit::
214 Do not show entries which have an overhead under that percent.
215 (Default: 0).
216
217SEE ALSO
218--------
219linkperf:perf-stat[1], linkperf:perf-annotate[1]