1 perf-config(1) 1 perf-config(1)
2 ============== 2 ==============
3 3
4 NAME 4 NAME
5 ---- 5 ----
6 perf-config - Get and set variables in a confi 6 perf-config - Get and set variables in a configuration file.
7 7
8 SYNOPSIS 8 SYNOPSIS
9 -------- 9 --------
10 [verse] 10 [verse]
11 'perf config' [<file-option>] [section.name[=v 11 'perf config' [<file-option>] [section.name[=value] ...]
12 or 12 or
13 'perf config' [<file-option>] -l | --list 13 'perf config' [<file-option>] -l | --list
14 14
15 DESCRIPTION 15 DESCRIPTION
16 ----------- 16 -----------
17 You can manage variables in a configuration fi 17 You can manage variables in a configuration file with this command.
18 18
19 OPTIONS 19 OPTIONS
20 ------- 20 -------
21 21
22 -l:: 22 -l::
23 --list:: 23 --list::
24 Show current config variables, name an 24 Show current config variables, name and value, for all sections.
25 25
26 --user:: 26 --user::
27 For writing and reading options: write 27 For writing and reading options: write to user
28 '$HOME/.perfconfig' file or read it. 28 '$HOME/.perfconfig' file or read it.
29 29
30 --system:: 30 --system::
31 For writing and reading options: write 31 For writing and reading options: write to system-wide
32 '$(sysconfdir)/perfconfig' or read it. 32 '$(sysconfdir)/perfconfig' or read it.
33 33
34 CONFIGURATION FILE 34 CONFIGURATION FILE
35 ------------------ 35 ------------------
36 36
37 The perf configuration file contains many vari 37 The perf configuration file contains many variables to change various
38 aspects of each of its tools, including output 38 aspects of each of its tools, including output, disk usage, etc.
39 The '$HOME/.perfconfig' file is used to store 39 The '$HOME/.perfconfig' file is used to store a per-user configuration.
40 The file '$(sysconfdir)/perfconfig' can be use 40 The file '$(sysconfdir)/perfconfig' can be used to
41 store a system-wide default configuration. 41 store a system-wide default configuration.
42 42
43 One an disable reading config files by setting 43 One an disable reading config files by setting the PERF_CONFIG environment
44 variable to /dev/null, or provide an alternate 44 variable to /dev/null, or provide an alternate config file by setting that
45 variable. 45 variable.
46 46
47 When reading or writing, the values are read f 47 When reading or writing, the values are read from the system and user
48 configuration files by default, and options '- 48 configuration files by default, and options '--system' and '--user'
49 can be used to tell the command to read from o 49 can be used to tell the command to read from or write to only that location.
50 50
51 Syntax 51 Syntax
52 ~~~~~~ 52 ~~~~~~
53 53
54 The file consist of sections. A section starts 54 The file consist of sections. A section starts with its name
55 surrounded by square brackets and continues ti 55 surrounded by square brackets and continues till the next section
56 begins. Each variable must be in a section, an 56 begins. Each variable must be in a section, and have the form
57 'name = value', for example: 57 'name = value', for example:
58 58
59 [section] 59 [section]
60 name1 = value1 60 name1 = value1
61 name2 = value2 61 name2 = value2
62 62
63 Section names are case sensitive and can conta 63 Section names are case sensitive and can contain any characters except
64 newline (double quote `"` and backslash have t 64 newline (double quote `"` and backslash have to be escaped as `\"` and `\\`,
65 respectively). Section headers can't span mult 65 respectively). Section headers can't span multiple lines.
66 66
67 Example 67 Example
68 ~~~~~~~ 68 ~~~~~~~
69 69
70 Given a $HOME/.perfconfig like this: 70 Given a $HOME/.perfconfig like this:
71 71
72 # 72 #
73 # This is the config file, and 73 # This is the config file, and
74 # a '#' and ';' character indicates a comment 74 # a '#' and ';' character indicates a comment
75 # 75 #
76 76
77 [colors] 77 [colors]
78 # Color variables 78 # Color variables
79 top = red, default 79 top = red, default
80 medium = green, default 80 medium = green, default
81 normal = lightgray, default 81 normal = lightgray, default
82 selected = white, lightgray 82 selected = white, lightgray
83 jump_arrows = blue, default 83 jump_arrows = blue, default
84 addr = magenta, default 84 addr = magenta, default
85 root = white, blue 85 root = white, blue
86 86
87 [tui] 87 [tui]
88 # Defaults if linked with libs 88 # Defaults if linked with libslang
89 report = on 89 report = on
90 annotate = on 90 annotate = on
91 top = on 91 top = on
92 92
93 [buildid] 93 [buildid]
94 # Default, disable using /dev/ 94 # Default, disable using /dev/null
95 dir = ~/.debug 95 dir = ~/.debug
96 96
97 [annotate] 97 [annotate]
98 # Defaults 98 # Defaults
99 hide_src_code = false 99 hide_src_code = false
100 use_offset = true 100 use_offset = true
101 jump_arrows = true 101 jump_arrows = true
102 show_nr_jumps = false 102 show_nr_jumps = false
103 103
104 [help] 104 [help]
105 # Format can be man, info, web 105 # Format can be man, info, web or html
106 format = man 106 format = man
107 autocorrect = 0 107 autocorrect = 0
108 108
109 [ui] 109 [ui]
110 show-headers = true 110 show-headers = true
111 111
112 [call-graph] 112 [call-graph]
113 # fp (framepointer), dwarf 113 # fp (framepointer), dwarf
114 record-mode = fp 114 record-mode = fp
115 print-type = graph 115 print-type = graph
116 order = caller 116 order = caller
117 sort-key = function 117 sort-key = function
118 118
119 [report] 119 [report]
120 # Defaults 120 # Defaults
121 sort_order = comm,dso,symbol 121 sort_order = comm,dso,symbol
122 percent-limit = 0 122 percent-limit = 0
123 queue-size = 0 123 queue-size = 0
124 children = true 124 children = true
125 group = true 125 group = true
126 skip-empty = true <<
127 126
>> 127 [llvm]
>> 128 dump-obj = true
>> 129 clang-opt = -g
128 130
129 You can hide source code of annotate feature s 131 You can hide source code of annotate feature setting the config to false with
130 132
131 % perf config annotate.hide_src_code=t 133 % perf config annotate.hide_src_code=true
132 134
133 If you want to add or modify several config it 135 If you want to add or modify several config items, you can do like
134 136
135 % perf config ui.show-headers=false km 137 % perf config ui.show-headers=false kmem.default=slab
136 138
137 To modify the sort order of report functionali 139 To modify the sort order of report functionality in user config file(i.e. `~/.perfconfig`), do
138 140
139 % perf config --user report.sort-order !! 141 % perf config --user report sort-order=srcline
140 142
141 To change colors of selected line to other for 143 To change colors of selected line to other foreground and background colors
142 in system config file (i.e. `$(sysconf)/perfco 144 in system config file (i.e. `$(sysconf)/perfconfig`), do
143 145
144 % perf config --system colors.selected 146 % perf config --system colors.selected=yellow,green
145 147
146 To query the record mode of call graph, do 148 To query the record mode of call graph, do
147 149
148 % perf config call-graph.record-mode 150 % perf config call-graph.record-mode
149 151
150 If you want to know multiple config key/value 152 If you want to know multiple config key/value pairs, you can do like
151 153
152 % perf config report.queue-size call-g 154 % perf config report.queue-size call-graph.order report.children
153 155
154 To query the config value of sort order of cal 156 To query the config value of sort order of call graph in user config file (i.e. `~/.perfconfig`), do
155 157
156 % perf config --user call-graph.sort-o 158 % perf config --user call-graph.sort-order
157 159
158 To query the config value of buildid directory 160 To query the config value of buildid directory in system config file (i.e. `$(sysconf)/perfconfig`), do
159 161
160 % perf config --system buildid.dir 162 % perf config --system buildid.dir
161 163
162 Variables 164 Variables
163 ~~~~~~~~~ 165 ~~~~~~~~~
164 166
165 colors.*:: 167 colors.*::
166 The variables for customizing the colo 168 The variables for customizing the colors used in the output for the
167 'report', 'top' and 'annotate' in the 169 'report', 'top' and 'annotate' in the TUI. They should specify the
168 foreground and background colors, sepa 170 foreground and background colors, separated by a comma, for example:
169 171
170 medium = green, lightgray 172 medium = green, lightgray
171 173
172 If you want to use the color configure 174 If you want to use the color configured for you terminal, just leave it
173 as 'default', for example: 175 as 'default', for example:
174 176
175 medium = default, lightgray 177 medium = default, lightgray
176 178
177 Available colors: 179 Available colors:
178 red, yellow, green, cyan, gray, black, 180 red, yellow, green, cyan, gray, black, blue,
179 white, default, magenta, lightgray 181 white, default, magenta, lightgray
180 182
181 colors.top:: 183 colors.top::
182 'top' means a overhead percent 184 'top' means a overhead percentage which is more than 5%.
183 And values of this variable sp 185 And values of this variable specify percentage colors.
184 Basic key values are foregroun 186 Basic key values are foreground-color 'red' and
185 background-color 'default'. 187 background-color 'default'.
186 colors.medium:: 188 colors.medium::
187 'medium' means a overhead perc 189 'medium' means a overhead percentage which has more than 0.5%.
188 Default values are 'green' and 190 Default values are 'green' and 'default'.
189 colors.normal:: 191 colors.normal::
190 'normal' means the rest of ove 192 'normal' means the rest of overhead percentages
191 except 'top', 'medium', 'selec 193 except 'top', 'medium', 'selected'.
192 Default values are 'lightgray' 194 Default values are 'lightgray' and 'default'.
193 colors.selected:: 195 colors.selected::
194 This selects the colors for th 196 This selects the colors for the current entry in a list of entries
195 from sub-commands (top, report 197 from sub-commands (top, report, annotate).
196 Default values are 'black' and 198 Default values are 'black' and 'lightgray'.
197 colors.jump_arrows:: 199 colors.jump_arrows::
198 Colors for jump arrows on asse 200 Colors for jump arrows on assembly code listings
199 such as 'jns', 'jmp', 'jane', 201 such as 'jns', 'jmp', 'jane', etc.
200 Default values are 'blue', 'de 202 Default values are 'blue', 'default'.
201 colors.addr:: 203 colors.addr::
202 This selects colors for addres 204 This selects colors for addresses from 'annotate'.
203 Default values are 'magenta', 205 Default values are 'magenta', 'default'.
204 colors.root:: 206 colors.root::
205 Colors for headers in the outp 207 Colors for headers in the output of a sub-commands (top, report).
206 Default values are 'white', 'b 208 Default values are 'white', 'blue'.
207 209
208 core.*:: 210 core.*::
209 core.proc-map-timeout:: 211 core.proc-map-timeout::
210 Sets a timeout (in millisecond 212 Sets a timeout (in milliseconds) for parsing /proc/<pid>/maps files.
211 Can be overridden by the --pro 213 Can be overridden by the --proc-map-timeout option on supported
212 subcommands. The default timeo 214 subcommands. The default timeout is 500ms.
213 215
214 tui.*, gtk.*:: 216 tui.*, gtk.*::
215 Subcommands that can be configured her 217 Subcommands that can be configured here are 'top', 'report' and 'annotate'.
216 These values are booleans, for example 218 These values are booleans, for example:
217 219
218 [tui] 220 [tui]
219 top = true 221 top = true
220 222
221 will make the TUI be the default for t 223 will make the TUI be the default for the 'top' subcommand. Those will be
222 available if the required libs were de 224 available if the required libs were detected at tool build time.
223 225
224 buildid.*:: 226 buildid.*::
225 buildid.dir:: 227 buildid.dir::
226 Each executable and shared lib 228 Each executable and shared library in modern distributions comes with a
227 content based identifier that, 229 content based identifier that, if available, will be inserted in a
228 'perf.data' file header to, at 230 'perf.data' file header to, at analysis time find what is needed to do
229 symbol resolution, code annota 231 symbol resolution, code annotation, etc.
230 232
231 The recording tools also store 233 The recording tools also stores a hard link or copy in a per-user
232 directory, $HOME/.debug/, of b 234 directory, $HOME/.debug/, of binaries, shared libraries, /proc/kallsyms
233 and /proc/kcore files to be us 235 and /proc/kcore files to be used at analysis time.
234 236
235 The buildid.dir variable can b 237 The buildid.dir variable can be used to either change this directory
236 cache location, or to disable 238 cache location, or to disable it altogether. If you want to disable it,
237 set buildid.dir to /dev/null. 239 set buildid.dir to /dev/null. The default is $HOME/.debug
238 240
239 buildid-cache.*:: <<
240 buildid-cache.debuginfod=URLs <<
241 Specify debuginfod URLs to be <<
242 it follows the same syntax as <<
243 <<
244 buildid-cache.debuginfod=htt <<
245 <<
246 annotate.*:: 241 annotate.*::
247 These are in control of addresses, jum 242 These are in control of addresses, jump function, source code
248 in lines of assembly code from a speci 243 in lines of assembly code from a specific program.
249 244
250 annotate.addr2line:: <<
251 addr2line binary to use for fi <<
252 <<
253 annotate.objdump:: <<
254 objdump binary to use for disa <<
255 including in the 'perf test' c <<
256 <<
257 annotate.disassembler_style:: <<
258 Use this to change the default <<
259 supported by binutils, such as <<
260 'objdump' man page. <<
261 <<
262 annotate.hide_src_code:: 245 annotate.hide_src_code::
263 If a program which is analyzed 246 If a program which is analyzed has source code,
264 this option lets 'annotate' pr 247 this option lets 'annotate' print a list of assembly code with the source code.
265 For example, let's see a part 248 For example, let's see a part of a program. There're four lines.
266 If this option is 'true', they 249 If this option is 'true', they can be printed
267 without source code from a pro 250 without source code from a program as below.
268 251
269 │ push %rbp 252 │ push %rbp
270 │ mov %rsp,%rbp 253 │ mov %rsp,%rbp
271 │ sub $0x10,%rsp 254 │ sub $0x10,%rsp
272 │ mov (%rdi),%rdx 255 │ mov (%rdi),%rdx
273 256
274 But if this option is 'false', 257 But if this option is 'false', source code of the part
275 can be also printed as below. 258 can be also printed as below. Default is 'false'.
276 259
277 │ struct rb_node *rb_ne 260 │ struct rb_node *rb_next(const struct rb_node *node)
278 │ { 261 │ {
279 │ push %rbp 262 │ push %rbp
280 │ mov %rsp,%rbp 263 │ mov %rsp,%rbp
281 │ sub $0x10,%rsp 264 │ sub $0x10,%rsp
282 │ struct rb_nod 265 │ struct rb_node *parent;
283 │ 266 │
284 │ if (RB_EMPTY_ 267 │ if (RB_EMPTY_NODE(node))
285 │ mov (%rdi),%rdx 268 │ mov (%rdi),%rdx
286 │ return n; 269 │ return n;
287 270
288 This option works with tui, st 271 This option works with tui, stdio2 browsers.
289 272
290 annotate.use_offset:: 273 annotate.use_offset::
291 Basing on a first address of a 274 Basing on a first address of a loaded function, offset can be used.
292 Instead of using original addr 275 Instead of using original addresses of assembly code,
293 addresses subtracted from a ba 276 addresses subtracted from a base address can be printed.
294 Let's illustrate an example. 277 Let's illustrate an example.
295 If a base address is 0XFFFFFFF 278 If a base address is 0XFFFFFFFF81624d50 as below,
296 279
297 ffffffff81624d50 <load0> 280 ffffffff81624d50 <load0>
298 281
299 an address on assembly code ha 282 an address on assembly code has a specific absolute address as below
300 283
301 ffffffff816250b8:│ mov 0 284 ffffffff816250b8:│ mov 0x8(%r14),%rdi
302 285
303 but if use_offset is 'true', a 286 but if use_offset is 'true', an address subtracted from a base address is printed.
304 Default is true. This option i 287 Default is true. This option is only applied to TUI.
305 288
306 368:│ mov 0 289 368:│ mov 0x8(%r14),%rdi
307 290
308 This option works with tui, st 291 This option works with tui, stdio2 browsers.
309 292
310 annotate.jump_arrows:: 293 annotate.jump_arrows::
311 There can be jump instruction 294 There can be jump instruction among assembly code.
312 Depending on a boolean value o 295 Depending on a boolean value of jump_arrows,
313 arrows can be printed or not w 296 arrows can be printed or not which represent
314 where do the instruction jump 297 where do the instruction jump into as below.
315 298
316 │ ┌──jmp 1333 299 │ ┌──jmp 1333
317 │ │ xchg %ax,%ax 300 │ │ xchg %ax,%ax
318 │1330:│ mov %r15,%r10 301 │1330:│ mov %r15,%r10
319 │1333:└─→cmp %r15,% 302 │1333:└─→cmp %r15,%r14
320 303
321 If jump_arrow is 'false', the 304 If jump_arrow is 'false', the arrows isn't printed as below.
322 Default is 'false'. 305 Default is 'false'.
323 306
324 │ ↓ jmp 1333 307 │ ↓ jmp 1333
325 │ xchg %ax,%ax 308 │ xchg %ax,%ax
326 │1330: mov %r15,%r10 309 │1330: mov %r15,%r10
327 │1333: cmp %r15,%r14 310 │1333: cmp %r15,%r14
328 311
329 This option works with tui bro 312 This option works with tui browser.
330 313
331 annotate.show_linenr:: 314 annotate.show_linenr::
332 When showing source code if th 315 When showing source code if this option is 'true',
333 line numbers are printed as be 316 line numbers are printed as below.
334 317
335 │1628 if (type & PER 318 │1628 if (type & PERF_SAMPLE_IDENTIFIER) {
336 │ ↓ jne 508 319 │ ↓ jne 508
337 │1628 data-> 320 │1628 data->id = *array;
338 │1629 array+ 321 │1629 array++;
339 │1630 } 322 │1630 }
340 323
341 However if this option is 'fal 324 However if this option is 'false', they aren't printed as below.
342 Default is 'false'. 325 Default is 'false'.
343 326
344 │ if (type & PER 327 │ if (type & PERF_SAMPLE_IDENTIFIER) {
345 │ ↓ jne 508 328 │ ↓ jne 508
346 │ data-> 329 │ data->id = *array;
347 │ array+ 330 │ array++;
348 │ } 331 │ }
349 332
350 This option works with tui, st 333 This option works with tui, stdio2 browsers.
351 334
352 annotate.show_nr_jumps:: 335 annotate.show_nr_jumps::
353 Let's see a part of assembly c 336 Let's see a part of assembly code.
354 337
355 │1382: movb $0x1,-0x270( 338 │1382: movb $0x1,-0x270(%rbp)
356 339
357 If use this, the number of bra 340 If use this, the number of branches jumping to that address can be printed as below.
358 Default is 'false'. 341 Default is 'false'.
359 342
360 │1 1382: movb $0x1,-0x27 343 │1 1382: movb $0x1,-0x270(%rbp)
361 344
362 This option works with tui, st 345 This option works with tui, stdio2 browsers.
363 346
364 annotate.show_total_period:: 347 annotate.show_total_period::
365 To compare two records on an i 348 To compare two records on an instruction base, with this option
366 provided, display total number 349 provided, display total number of samples that belong to a line
367 in assembly code. If this opti 350 in assembly code. If this option is 'true', total periods are printed
368 instead of percent values as b 351 instead of percent values as below.
369 352
370 302 │ mov %eax,%ea 353 302 │ mov %eax,%eax
371 354
372 But if this option is 'false', 355 But if this option is 'false', percent values for overhead are printed i.e.
373 Default is 'false'. 356 Default is 'false'.
374 357
375 99.93 │ mov %eax,%ea 358 99.93 │ mov %eax,%eax
376 359
377 This option works with tui, st 360 This option works with tui, stdio2, stdio browsers.
378 361
379 annotate.show_nr_samples:: 362 annotate.show_nr_samples::
380 By default perf annotate shows 363 By default perf annotate shows percentage of samples. This option
381 can be used to print absolute 364 can be used to print absolute number of samples. Ex, when set as
382 false: 365 false:
383 366
384 Percent│ 367 Percent│
385 74.03 │ mov %fs:0x2 368 74.03 │ mov %fs:0x28,%rax
386 369
387 When set as true: 370 When set as true:
388 371
389 Samples│ 372 Samples│
390 6 │ mov %fs:0x2 373 6 │ mov %fs:0x28,%rax
391 374
392 This option works with tui, st 375 This option works with tui, stdio2, stdio browsers.
393 376
394 annotate.offset_level:: 377 annotate.offset_level::
395 Default is '1', meaning just j 378 Default is '1', meaning just jump targets will have offsets show right beside
396 the instruction. When set to ' 379 the instruction. When set to '2' 'call' instructions will also have its offsets
397 shown, 3 or higher will show o 380 shown, 3 or higher will show offsets for all instructions.
398 381
399 This option works with tui, st 382 This option works with tui, stdio2 browsers.
400 383
401 annotate.demangle:: <<
402 Demangle symbol names to human <<
403 <<
404 annotate.demangle_kernel:: <<
405 Demangle kernel symbol names t <<
406 <<
407 hist.*:: 384 hist.*::
408 hist.percentage:: 385 hist.percentage::
409 This option control the way to 386 This option control the way to calculate overhead of filtered entries -
410 that means the value of this o 387 that means the value of this option is effective only if there's a
411 filter (by comm, dso or symbol 388 filter (by comm, dso or symbol name). Suppose a following example:
412 389
413 Overhead Symbols 390 Overhead Symbols
414 ........ ....... 391 ........ .......
415 33.33% foo 392 33.33% foo
416 33.33% bar 393 33.33% bar
417 33.33% baz 394 33.33% baz
418 395
419 This is an original overhead an 396 This is an original overhead and we'll filter out the first 'foo'
420 entry. The value of 'relative' 397 entry. The value of 'relative' would increase the overhead of 'bar'
421 and 'baz' to 50.00% for each, w 398 and 'baz' to 50.00% for each, while 'absolute' would show their
422 current overhead (33.33%). 399 current overhead (33.33%).
423 400
424 ui.*:: 401 ui.*::
425 ui.show-headers:: 402 ui.show-headers::
426 This option controls display o 403 This option controls display of column headers (like 'Overhead' and 'Symbol')
427 in 'report' and 'top'. If this 404 in 'report' and 'top'. If this option is false, they are hidden.
428 This option is only applied to 405 This option is only applied to TUI.
429 406
430 call-graph.*:: 407 call-graph.*::
431 The following controls the handling of 408 The following controls the handling of call-graphs (obtained via the
432 -g/--call-graph options). 409 -g/--call-graph options).
433 410
434 call-graph.record-mode:: 411 call-graph.record-mode::
435 The mode for user space can be 412 The mode for user space can be 'fp' (frame pointer), 'dwarf'
436 and 'lbr'. The value 'dwarf' 413 and 'lbr'. The value 'dwarf' is effective only if libunwind
437 (or a recent version of libdw) 414 (or a recent version of libdw) is present on the system;
438 the value 'lbr' only works for 415 the value 'lbr' only works for certain cpus. The method for
439 kernel space is controlled not 416 kernel space is controlled not by this option but by the
440 kernel config (CONFIG_UNWINDER 417 kernel config (CONFIG_UNWINDER_*).
441 418
442 call-graph.dump-size:: 419 call-graph.dump-size::
443 The size of stack to dump in o 420 The size of stack to dump in order to do post-unwinding. Default is 8192 (byte).
444 When using dwarf into record-m 421 When using dwarf into record-mode, the default size will be used if omitted.
445 422
446 call-graph.print-type:: 423 call-graph.print-type::
447 The print-types can be graph ( 424 The print-types can be graph (graph absolute), fractal (graph relative),
448 flat and folded. This option c 425 flat and folded. This option controls a way to show overhead for each callchain
449 entry. Suppose a following exa 426 entry. Suppose a following example.
450 427
451 Overhead Symbols 428 Overhead Symbols
452 ........ ....... 429 ........ .......
453 40.00% foo 430 40.00% foo
454 | 431 |
455 ---foo 432 ---foo
456 | 433 |
457 |--50.00%--bar 434 |--50.00%--bar
458 | main 435 | main
459 | 436 |
460 --50.00%--baz 437 --50.00%--baz
461 main 438 main
462 439
463 This output is a 'fractal' for 440 This output is a 'fractal' format. The 'foo' came from 'bar' and 'baz' exactly
464 half and half so 'fractal' sho 441 half and half so 'fractal' shows 50.00% for each
465 (meaning that it assumes 100% 442 (meaning that it assumes 100% total overhead of 'foo').
466 443
467 The 'graph' uses absolute over 444 The 'graph' uses absolute overhead value of 'foo' as total so each of
468 'bar' and 'baz' callchain will 445 'bar' and 'baz' callchain will have 20.00% of overhead.
469 If 'flat' is used, single colu 446 If 'flat' is used, single column and linear exposure of call chains.
470 'folded' mean call chains are 447 'folded' mean call chains are displayed in a line, separated by semicolons.
471 448
472 call-graph.order:: 449 call-graph.order::
473 This option controls print ord 450 This option controls print order of callchains. The default is
474 'callee' which means callee is 451 'callee' which means callee is printed at top and then followed by its
475 caller and so on. The 'caller' 452 caller and so on. The 'caller' prints it in reverse order.
476 453
477 If this option is not set and 454 If this option is not set and report.children or top.children is
478 set to true (or the equivalent 455 set to true (or the equivalent command line option is given),
479 the default value of this opti 456 the default value of this option is changed to 'caller' for the
480 execution of 'perf report' or 457 execution of 'perf report' or 'perf top'. Other commands will
481 still default to 'callee'. 458 still default to 'callee'.
482 459
483 call-graph.sort-key:: 460 call-graph.sort-key::
484 The callchains are merged if t 461 The callchains are merged if they contain same information.
485 The sort-key option determines 462 The sort-key option determines a way to compare the callchains.
486 A value of 'sort-key' can be ' 463 A value of 'sort-key' can be 'function' or 'address'.
487 The default is 'function'. 464 The default is 'function'.
488 465
489 call-graph.threshold:: 466 call-graph.threshold::
490 When there're many callchains 467 When there're many callchains it'd print tons of lines. So perf omits
491 small callchains under a certa 468 small callchains under a certain overhead (threshold) and this option
492 control the threshold. Default 469 control the threshold. Default is 0.5 (%). The overhead is calculated
493 by value depends on call-graph 470 by value depends on call-graph.print-type.
494 471
495 call-graph.print-limit:: 472 call-graph.print-limit::
496 This is a maximum number of li 473 This is a maximum number of lines of callchain printed for a single
497 histogram entry. Default is 0 474 histogram entry. Default is 0 which means no limitation.
498 475
499 report.*:: 476 report.*::
500 report.sort_order:: 477 report.sort_order::
501 Allows changing the default so 478 Allows changing the default sort order from "comm,dso,symbol" to
502 some other default, for instan 479 some other default, for instance "sym,dso" may be more fitting for
503 kernel developers. 480 kernel developers.
504 report.percent-limit:: 481 report.percent-limit::
505 This one is mostly the same as 482 This one is mostly the same as call-graph.threshold but works for
506 histogram entries. Entries hav 483 histogram entries. Entries having an overhead lower than this
507 percentage will not be printed 484 percentage will not be printed. Default is '0'. If percent-limit
508 is '10', only entries which ha 485 is '10', only entries which have more than 10% of overhead will be
509 printed. 486 printed.
510 487
511 report.queue-size:: 488 report.queue-size::
512 This option sets up the maximu 489 This option sets up the maximum allocation size of the internal
513 event queue for ordering event 490 event queue for ordering events. Default is 0, meaning no limit.
514 491
515 report.children:: 492 report.children::
516 'Children' means functions cal 493 'Children' means functions called from another function.
517 If this option is true, 'perf 494 If this option is true, 'perf report' cumulates callchains of children
518 and show (accumulated) total o 495 and show (accumulated) total overhead as well as 'Self' overhead.
519 Please refer to the 'perf repo 496 Please refer to the 'perf report' manual. The default is 'true'.
520 497
521 report.group:: 498 report.group::
522 This option is to show event g 499 This option is to show event group information together.
523 Example output with this turne 500 Example output with this turned on, notice that there is one column
524 per event in the group, ref-cy 501 per event in the group, ref-cycles and cycles:
525 502
526 # group: {ref-cycles,cycles} 503 # group: {ref-cycles,cycles}
527 # ======== 504 # ========
528 # 505 #
529 # Samples: 7K of event 'anon g 506 # Samples: 7K of event 'anon group { ref-cycles, cycles }'
530 # Event count (approx.): 68761 507 # Event count (approx.): 6876107743
531 # 508 #
532 # Overhead Command 509 # Overhead Command Shared Object Symbol
533 # ................ ....... . 510 # ................ ....... ................. ...................
534 # 511 #
535 99.84% 99.76% noploop n 512 99.84% 99.76% noploop noploop [.] main
536 0.07% 0.00% noploop l 513 0.07% 0.00% noploop ld-2.15.so [.] strcmp
537 0.03% 0.00% noploop [ 514 0.03% 0.00% noploop [kernel.kallsyms] [k] timerqueue_del
538 515
539 report.skip-empty:: <<
540 This option can change default <<
541 If it's set true, 'perf report <<
542 <<
543 top.*:: 516 top.*::
544 top.children:: 517 top.children::
545 Same as 'report.children'. So 518 Same as 'report.children'. So if it is enabled, the output of 'top'
546 command will have 'Children' o 519 command will have 'Children' overhead column as well as 'Self' overhead
547 column by default. 520 column by default.
548 The default is 'true'. 521 The default is 'true'.
549 522
550 top.call-graph:: 523 top.call-graph::
551 This is identical to 'call-gra 524 This is identical to 'call-graph.record-mode', except it is
552 applicable only for 'top' subc 525 applicable only for 'top' subcommand. This option ONLY setup
553 the unwind method. To enable ' 526 the unwind method. To enable 'perf top' to actually use it,
554 the command line option -g mus 527 the command line option -g must be specified.
555 528
556 man.*:: 529 man.*::
557 man.viewer:: 530 man.viewer::
558 This option can assign a tool 531 This option can assign a tool to view manual pages when 'help'
559 subcommand was invoked. Suppor 532 subcommand was invoked. Supported tools are 'man', 'woman'
560 (with emacs client) and 'konqu 533 (with emacs client) and 'konqueror'. Default is 'man'.
561 534
562 New man viewer tool can be als 535 New man viewer tool can be also added using 'man.<tool>.cmd'
563 or use different path using 'm 536 or use different path using 'man.<tool>.path' config option.
564 537
565 pager.*:: 538 pager.*::
566 pager.<subcommand>:: 539 pager.<subcommand>::
567 When the subcommand is run on 540 When the subcommand is run on stdio, determine whether it uses
568 pager or not based on this val 541 pager or not based on this value. Default is 'unspecified'.
569 542
570 kmem.*:: 543 kmem.*::
571 kmem.default:: 544 kmem.default::
572 This option decides which allo 545 This option decides which allocator is to be analyzed if neither
573 '--slab' nor '--page' option i 546 '--slab' nor '--page' option is used. Default is 'slab'.
574 547
575 record.*:: 548 record.*::
576 record.build-id:: 549 record.build-id::
577 This option can be 'cache', 'n !! 550 This option can be 'cache', 'no-cache' or 'skip'.
578 'cache' is to post-process dat 551 'cache' is to post-process data and save/update the binaries into
579 the build-id cache (in ~/.debu 552 the build-id cache (in ~/.debug). This is the default.
580 But if this option is 'no-cach 553 But if this option is 'no-cache', it will not update the build-id cache.
581 'skip' skips post-processing a 554 'skip' skips post-processing and does not update the cache.
582 'mmap' skips post-processing a <<
583 555
584 record.call-graph:: 556 record.call-graph::
585 This is identical to 'call-gra 557 This is identical to 'call-graph.record-mode', except it is
586 applicable only for 'record' s 558 applicable only for 'record' subcommand. This option ONLY setup
587 the unwind method. To enable ' 559 the unwind method. To enable 'perf record' to actually use it,
588 the command line option -g mus 560 the command line option -g must be specified.
589 561
590 record.aio:: 562 record.aio::
591 Use 'n' control blocks in asyn 563 Use 'n' control blocks in asynchronous (Posix AIO) trace writing
592 mode ('n' default: 1, max: 4). 564 mode ('n' default: 1, max: 4).
593 565
594 record.debuginfod:: <<
595 Specify debuginfod URL to be u <<
596 it follows the same syntax as <<
597 <<
598 http://192.168.122.174:8002 <<
599 <<
600 If the URLs is 'system', the v <<
601 variable is used. <<
602 <<
603 diff.*:: 566 diff.*::
604 diff.order:: 567 diff.order::
605 This option sets the number of 568 This option sets the number of columns to sort the result.
606 The default is 0, which means 569 The default is 0, which means sorting by baseline.
607 Setting it to 1 will sort the 570 Setting it to 1 will sort the result by delta (or other
608 compute method selected). 571 compute method selected).
609 572
610 diff.compute:: 573 diff.compute::
611 This options sets the method f 574 This options sets the method for computing the diff result.
612 Possible values are 'delta', ' 575 Possible values are 'delta', 'delta-abs', 'ratio' and
613 'wdiff'. Default is 'delta'. 576 'wdiff'. Default is 'delta'.
614 577
615 trace.*:: 578 trace.*::
616 trace.add_events:: 579 trace.add_events::
617 Allows adding a set of events 580 Allows adding a set of events to add to the ones specified
618 by the user, or use as a defau 581 by the user, or use as a default one if none was specified.
619 The initial use case is to add 582 The initial use case is to add augmented_raw_syscalls.o to
620 activate the 'perf trace' logi 583 activate the 'perf trace' logic that looks for syscall
621 pointer contents after the nor 584 pointer contents after the normal tracepoint payload.
622 585
623 trace.args_alignment:: 586 trace.args_alignment::
624 Number of columns to align the 587 Number of columns to align the argument list, default is 70,
625 use 40 for the strace default, 588 use 40 for the strace default, zero to no alignment.
626 589
627 trace.no_inherit:: 590 trace.no_inherit::
628 Do not follow children threads 591 Do not follow children threads.
629 592
630 trace.show_arg_names:: 593 trace.show_arg_names::
631 Should syscall argument names 594 Should syscall argument names be printed? If not then trace.show_zeros
632 will be set. 595 will be set.
633 596
634 trace.show_duration:: 597 trace.show_duration::
635 Show syscall duration. 598 Show syscall duration.
636 599
637 trace.show_prefix:: 600 trace.show_prefix::
638 If set to 'yes' will show comm 601 If set to 'yes' will show common string prefixes in tables. The default
639 is to remove the common prefix 602 is to remove the common prefix in things like "MAP_SHARED", showing just "SHARED".
640 603
641 trace.show_timestamp:: 604 trace.show_timestamp::
642 Show syscall start timestamp. 605 Show syscall start timestamp.
643 606
644 trace.show_zeros:: 607 trace.show_zeros::
645 Do not suppress syscall argume 608 Do not suppress syscall arguments that are equal to zero.
646 609
647 trace.tracepoint_beautifiers:: 610 trace.tracepoint_beautifiers::
648 Use "libtraceevent" to use tha 611 Use "libtraceevent" to use that library to augment the tracepoint arguments,
649 "libbeauty", the default, to u 612 "libbeauty", the default, to use the same argument beautifiers used in the
650 strace-like sys_enter+sys_exit 613 strace-like sys_enter+sys_exit lines.
651 614
652 ftrace.*:: 615 ftrace.*::
653 ftrace.tracer:: 616 ftrace.tracer::
654 Can be used to select the defa 617 Can be used to select the default tracer when neither -G nor
655 -F option is not specified. Po 618 -F option is not specified. Possible values are 'function' and
656 'function_graph'. 619 'function_graph'.
657 620
>> 621 llvm.*::
>> 622 llvm.clang-path::
>> 623 Path to clang. If omit, search it from $PATH.
>> 624
>> 625 llvm.clang-bpf-cmd-template::
>> 626 Cmdline template. Below lines show its default value. Environment
>> 627 variable is used to pass options.
>> 628 "$CLANG_EXEC -D__KERNEL__ -D__NR_CPUS__=$NR_CPUS "\
>> 629 "-DLINUX_VERSION_CODE=$LINUX_VERSION_CODE " \
>> 630 "$CLANG_OPTIONS $PERF_BPF_INC_OPTIONS $KERNEL_INC_OPTIONS " \
>> 631 "-Wno-unused-value -Wno-pointer-sign " \
>> 632 "-working-directory $WORKING_DIR " \
>> 633 "-c \"$CLANG_SOURCE\" -target bpf $CLANG_EMIT_LLVM -O2 -o - $LLVM_OPTIONS_PIPE"
>> 634
>> 635 llvm.clang-opt::
>> 636 Options passed to clang.
>> 637
>> 638 llvm.kbuild-dir::
>> 639 kbuild directory. If not set, use /lib/modules/`uname -r`/build.
>> 640 If set to "" deliberately, skip kernel header auto-detector.
>> 641
>> 642 llvm.kbuild-opts::
>> 643 Options passed to 'make' when detecting kernel header options.
>> 644
>> 645 llvm.dump-obj::
>> 646 Enable perf dump BPF object files compiled by LLVM.
>> 647
>> 648 llvm.opts::
>> 649 Options passed to llc.
>> 650
658 samples.*:: 651 samples.*::
659 652
660 samples.context:: 653 samples.context::
661 Define how many ns worth of ti 654 Define how many ns worth of time to show
662 around samples in perf report 655 around samples in perf report sample context browser.
663 656
664 scripts.*:: 657 scripts.*::
665 658
666 Any option defines a script that is ad 659 Any option defines a script that is added to the scripts menu
667 in the interactive perf browser and wh 660 in the interactive perf browser and whose output is displayed.
668 The name of the option is the name, th 661 The name of the option is the name, the value is a script command line.
669 The script gets the same options passe 662 The script gets the same options passed as a full perf script,
670 in particular -i perfdata file, --cpu, 663 in particular -i perfdata file, --cpu, --tid
671 664
672 convert.*:: 665 convert.*::
673 666
674 convert.queue-size:: 667 convert.queue-size::
675 Limit the size of ordered_even 668 Limit the size of ordered_events queue, so we could control
676 allocation size of perf data f 669 allocation size of perf data files without proper finished
677 round events. 670 round events.
678 stat.*:: 671 stat.*::
679 672
680 stat.big-num:: 673 stat.big-num::
681 (boolean) Change the default f 674 (boolean) Change the default for "--big-num". To make
682 "--no-big-num" the default, se 675 "--no-big-num" the default, set "stat.big-num=false".
683 676
684 intel-pt.*:: 677 intel-pt.*::
685 678
686 intel-pt.cache-divisor:: 679 intel-pt.cache-divisor::
687 680
688 intel-pt.mispred-all:: 681 intel-pt.mispred-all::
689 If set, Intel PT decoder will 682 If set, Intel PT decoder will set the mispred flag on all
690 branches. 683 branches.
691 684
692 intel-pt.max-loops:: <<
693 If set and non-zero, the maxim <<
694 branches decoded without consu <<
695 the maximum is exceeded there <<
696 error. The default is 100000. <<
697 <<
698 auxtrace.*:: 685 auxtrace.*::
699 686
700 auxtrace.dumpdir:: 687 auxtrace.dumpdir::
701 s390 only. The directory to sa 688 s390 only. The directory to save the auxiliary trace buffer
702 can be changed using this opti 689 can be changed using this option. Ex, auxtrace.dumpdir=/tmp.
703 If the directory does not exis 690 If the directory does not exist or has the wrong file type,
704 the current directory is used. 691 the current directory is used.
705 <<
706 itrace.*:: <<
707 <<
708 debug-log-buffer-size:: <<
709 Log size in bytes to output wh <<
710 Refer 'itrace' option of linkp <<
711 linkperf:perf-report[1]. The d <<
712 <<
713 daemon.*:: <<
714 <<
715 daemon.base:: <<
716 Base path for daemon data. All <<
717 this path. <<
718 <<
719 session-<NAME>.*:: <<
720 <<
721 session-<NAME>.run:: <<
722 <<
723 Defines new record session for <<
724 command line without the 'reco <<
725 692
726 SEE ALSO 693 SEE ALSO
727 -------- 694 --------
728 linkperf:perf[1] 695 linkperf:perf[1]
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.