1PERF-ANNOTATE(1) perf Manual PERF-ANNOTATE(1)
2
6 perf-annotate - Read perf.data (created by perf record) and display
7 annotated code
8
10 perf annotate [-i <file> | --input=file] [symbol_name]
11
13 This command reads the input file and displays an annotated version of
14 the code. If the object file has debug symbols then the source code
15 will be displayed alongside assembly code.
16 If there is no debug info in the object, then annotated assembly is
18 displayed.
19
21 -i, --input=<file>
22 Input file name. (default: perf.data unless stdin is a fifo)
23 -d, --dsos=<dso[,dso...]>
25 Only consider symbols in these dsos.
26 -s, --symbol=<symbol>
28 Symbol to annotate.
29 -f, --force
31 Don’t do ownership validation.
32 -v, --verbose
34 Be more verbose. (Show symbol address, etc)
35 -q, --quiet
37 Do not show any message. (Suppress -v)
38 -n, --show-nr-samples
40 Show the number of samples for each symbol
41 -D, --dump-raw-trace
43 Dump raw trace in ASCII.
44 -k, --vmlinux=<file>
46 vmlinux pathname.
47 --ignore-vmlinux
49 Ignore vmlinux files.
50 --itrace
52 Options for decoding instruction tracing data. The options are:
53 i synthesize instructions events
55 b synthesize branches events (branch misses for Arm SPE)
56 c synthesize branches events (calls only)
57 r synthesize branches events (returns only)
58 x synthesize transactions events
59 w synthesize ptwrite events
60 p synthesize power events (incl. PSB events for Intel PT)
61 o synthesize other events recorded due to the use
62 of aux-output (refer to perf record)
63 e synthesize error events
64 d create a debug log
65 f synthesize first level cache events
66 m synthesize last level cache events
67 M synthesize memory events
68 t synthesize TLB events
69 a synthesize remote access events
70 g synthesize a call chain (use with i or x)
71 G synthesize a call chain on existing event records
72 l synthesize last branch entries (use with i or x)
73 L synthesize last branch entries on existing event records
74 s skip initial number of events
75 q quicker (less detailed) decoding
76 Z prefer to ignore timestamps (so-called "timeless" decoding)
77 The default is all events i.e. the same as --itrace=ibxwpe,
79 except for perf script where it is --itrace=ce
80 In addition, the period (default 100000, except for perf script where it is 1)
82 for instructions events can be specified in units of:
83 i instructions
85 t ticks
86 ms milliseconds
87 us microseconds
88 ns nanoseconds (default)
89 Also the call chain size (default 16, max. 1024) for instructions or
91 transactions events can be specified.
92 Also the number of last branch entries (default 64, max. 1024) for
94 instructions or transactions events can be specified.
95 Similar to options g and l, size may also be specified for options G and L.
97 On x86, note that G and L work poorly when data has been recorded with
98 large PEBS. Refer linkperf:perf-intel-pt[1] man page for details.
99 It is also possible to skip events generated (instructions, branches, transactions,
101 ptwrite, power) at the beginning. This is useful to ignore initialization code.
102 --itrace=i0nss1000000
104 skips the first million instructions.
106 The 'e' option may be followed by flags which affect what errors will or
108 will not be reported. Each flag must be preceded by either '+' or '-'.
109 The flags are:
110 o overflow
111 l trace data lost
112 If supported, the 'd' option may be followed by flags which affect what
114 debug messages will or will not be logged. Each flag must be preceded
115 by either '+' or '-'. The flags are:
116 a all perf events
117 If supported, the 'q' option may be repeated to increase the effect.
119 To disable decoding entirely, use --no-itrace.
121 -m, --modules
123 Load module symbols. WARNING: use only with -k and LIVE kernel.
124 -l, --print-line
126 Print matching source lines (may be slow).
127 -P, --full-paths
129 Don’t shorten the displayed pathnames.
130 --stdio
132 Use the stdio interface.
133 --stdio2
135 Use the stdio2 interface, non-interactive, uses the TUI formatting.
136 --stdio-color=<mode>
138 always, never or auto, allowing configuring color output via the
139 command line, in addition to via "color.ui" .perfconfig. Use
140 --stdio-color always to generate color even when redirecting to a
141 pipe or file. Using just --stdio-color is equivalent to using
142 always.
143 --tui
145 Use the TUI interface. Use of --tui requires a tty, if one is not
146 present, as when piping to other commands, the stdio interface is
147 used. This interfaces starts by centering on the line with more
148 samples, TAB/UNTAB cycles through the lines with more samples.
149 --gtk
151 Use the GTK interface.
152 -C, --cpu=<cpu>
154 Only report samples for the list of CPUs provided. Multiple CPUs
155 can be provided as a comma-separated list with no space: 0,1.
156 Ranges of CPUs are specified with -: 0-2. Default is to report
157 samples on all CPUs.
158 --asm-raw
160 Show raw instruction encoding of assembly instructions.
161 --show-total-period
163 Show a column with the sum of periods.
164 --source
166 Interleave source code with assembly code. Enabled by default,
167 disable with --no-source.
168 --symfs=<directory>
170 Look for files with symbols relative to this directory.
171 -M, --disassembler-style=
173 Set disassembler style for objdump.
174 --objdump=<path>
176 Path to objdump binary.
177 --prefix=PREFIX, --prefix-strip=N
179 Remove first N entries from source file path names in executables
180 and add PREFIX. This allows to display source code compiled on
181 systems with different file system layout.
182 --skip-missing
184 Skip symbols that cannot be annotated.
185 --group
187 Show event group information together
188 --demangle
190 Demangle symbol names to human readable form. It’s enabled by
191 default, disable with --no-demangle.
192 --demangle-kernel
194 Demangle kernel symbol names to human readable form (for C++
195 kernels).
196 --percent-type
198 Set annotation percent type from following choices: global-period,
199 local-period, global-hits, local-hits
200 The local/global keywords set if the percentage is computed
202 in the scope of the function (local) or the whole data (global).
203 The period/hits keywords set the base the percentage is computed
204 on - the samples period or the number of samples (hits).
205
207 perf-record(1), perf-report(1)
208perf 11/22/2021 PERF-ANNOTATE(1)