1perf-config(1) 2============== 3 4NAME 5---- 6perf-config - Get and set variables in a configuration file. 7 8SYNOPSIS 9-------- 10[verse] 11'perf config' [<file-option>] [section.name[=value] ...] 12or 13'perf config' [<file-option>] -l | --list 14 15DESCRIPTION 16----------- 17You can manage variables in a configuration file with this command. 18 19OPTIONS 20------- 21 22-l:: 23--list:: 24 Show current config variables, name and value, for all sections. 25 26--user:: 27 For writing and reading options: write to user 28 '$HOME/.perfconfig' file or read it. 29 30--system:: 31 For writing and reading options: write to system-wide 32 '$(sysconfdir)/perfconfig' or read it. 33 34CONFIGURATION FILE 35------------------ 36 37The perf configuration file contains many variables to change various 38aspects of each of its tools, including output, disk usage, etc. 39The '$HOME/.perfconfig' file is used to store a per-user configuration. 40The file '$(sysconfdir)/perfconfig' can be used to 41store a system-wide default configuration. 42 43When reading or writing, the values are read from the system and user 44configuration files by default, and options '--system' and '--user' 45can be used to tell the command to read from or write to only that location. 46 47Syntax 48~~~~~~ 49 50The file consist of sections. A section starts with its name 51surrounded by square brackets and continues till the next section 52begins. Each variable must be in a section, and have the form 53'name = value', for example: 54 55 [section] 56 name1 = value1 57 name2 = value2 58 59Section names are case sensitive and can contain any characters except 60newline (double quote `"` and backslash have to be escaped as `\"` and `\\`, 61respectively). Section headers can't span multiple lines. 62 63Example 64~~~~~~~ 65 66Given a $HOME/.perfconfig like this: 67 68# 69# This is the config file, and 70# a '#' and ';' character indicates a comment 71# 72 73 [colors] 74 # Color variables 75 top = red, default 76 medium = green, default 77 normal = lightgray, default 78 selected = white, lightgray 79 jump_arrows = blue, default 80 addr = magenta, default 81 root = white, blue 82 83 [tui] 84 # Defaults if linked with libslang 85 report = on 86 annotate = on 87 top = on 88 89 [buildid] 90 # Default, disable using /dev/null 91 dir = ~/.debug 92 93 [annotate] 94 # Defaults 95 hide_src_code = false 96 use_offset = true 97 jump_arrows = true 98 show_nr_jumps = false 99 100 [help] 101 # Format can be man, info, web or html 102 format = man 103 autocorrect = 0 104 105 [ui] 106 show-headers = true 107 108 [call-graph] 109 # fp (framepointer), dwarf 110 record-mode = fp 111 print-type = graph 112 order = caller 113 sort-key = function 114 115 [report] 116 # Defaults 117 sort-order = comm,dso,symbol 118 percent-limit = 0 119 queue-size = 0 120 children = true 121 group = true 122 123You can hide source code of annotate feature setting the config to false with 124 125 % perf config annotate.hide_src_code=true 126 127If you want to add or modify several config items, you can do like 128 129 % perf config ui.show-headers=false kmem.default=slab 130 131To modify the sort order of report functionality in user config file(i.e. `~/.perfconfig`), do 132 133 % perf config --user report sort-order=srcline 134 135To change colors of selected line to other foreground and background colors 136in system config file (i.e. `$(sysconf)/perfconfig`), do 137 138 % perf config --system colors.selected=yellow,green 139 140To query the record mode of call graph, do 141 142 % perf config call-graph.record-mode 143 144If you want to know multiple config key/value pairs, you can do like 145 146 % perf config report.queue-size call-graph.order report.children 147 148To query the config value of sort order of call graph in user config file (i.e. `~/.perfconfig`), do 149 150 % perf config --user call-graph.sort-order 151 152To query the config value of buildid directory in system config file (i.e. `$(sysconf)/perfconfig`), do 153 154 % perf config --system buildid.dir 155 156Variables 157~~~~~~~~~ 158 159colors.*:: 160 The variables for customizing the colors used in the output for the 161 'report', 'top' and 'annotate' in the TUI. They should specify the 162 foreground and background colors, separated by a comma, for example: 163 164 medium = green, lightgray 165 166 If you want to use the color configured for you terminal, just leave it 167 as 'default', for example: 168 169 medium = default, lightgray 170 171 Available colors: 172 red, yellow, green, cyan, gray, black, blue, 173 white, default, magenta, lightgray 174 175 colors.top:: 176 'top' means a overhead percentage which is more than 5%. 177 And values of this variable specify percentage colors. 178 Basic key values are foreground-color 'red' and 179 background-color 'default'. 180 colors.medium:: 181 'medium' means a overhead percentage which has more than 0.5%. 182 Default values are 'green' and 'default'. 183 colors.normal:: 184 'normal' means the rest of overhead percentages 185 except 'top', 'medium', 'selected'. 186 Default values are 'lightgray' and 'default'. 187 colors.selected:: 188 This selects the colors for the current entry in a list of entries 189 from sub-commands (top, report, annotate). 190 Default values are 'black' and 'lightgray'. 191 colors.jump_arrows:: 192 Colors for jump arrows on assembly code listings 193 such as 'jns', 'jmp', 'jane', etc. 194 Default values are 'blue', 'default'. 195 colors.addr:: 196 This selects colors for addresses from 'annotate'. 197 Default values are 'magenta', 'default'. 198 colors.root:: 199 Colors for headers in the output of a sub-commands (top, report). 200 Default values are 'white', 'blue'. 201 202core.*:: 203 core.proc-map-timeout:: 204 Sets a timeout (in milliseconds) for parsing /proc/<pid>/maps files. 205 Can be overridden by the --proc-map-timeout option on supported 206 subcommands. The default timeout is 500ms. 207 208tui.*, gtk.*:: 209 Subcommands that can be configured here are 'top', 'report' and 'annotate'. 210 These values are booleans, for example: 211 212 [tui] 213 top = true 214 215 will make the TUI be the default for the 'top' subcommand. Those will be 216 available if the required libs were detected at tool build time. 217 218buildid.*:: 219 buildid.dir:: 220 Each executable and shared library in modern distributions comes with a 221 content based identifier that, if available, will be inserted in a 222 'perf.data' file header to, at analysis time find what is needed to do 223 symbol resolution, code annotation, etc. 224 225 The recording tools also stores a hard link or copy in a per-user 226 directory, $HOME/.debug/, of binaries, shared libraries, /proc/kallsyms 227 and /proc/kcore files to be used at analysis time. 228 229 The buildid.dir variable can be used to either change this directory 230 cache location, or to disable it altogether. If you want to disable it, 231 set buildid.dir to /dev/null. The default is $HOME/.debug 232 233annotate.*:: 234 These options work only for TUI. 235 These are in control of addresses, jump function, source code 236 in lines of assembly code from a specific program. 237 238 annotate.hide_src_code:: 239 If a program which is analyzed has source code, 240 this option lets 'annotate' print a list of assembly code with the source code. 241 For example, let's see a part of a program. There're four lines. 242 If this option is 'true', they can be printed 243 without source code from a program as below. 244 245 │ push %rbp 246 │ mov %rsp,%rbp 247 │ sub $0x10,%rsp 248 │ mov (%rdi),%rdx 249 250 But if this option is 'false', source code of the part 251 can be also printed as below. Default is 'false'. 252 253 │ struct rb_node *rb_next(const struct rb_node *node) 254 │ { 255 │ push %rbp 256 │ mov %rsp,%rbp 257 │ sub $0x10,%rsp 258 │ struct rb_node *parent; 259 │ 260 │ if (RB_EMPTY_NODE(node)) 261 │ mov (%rdi),%rdx 262 │ return n; 263 264 annotate.use_offset:: 265 Basing on a first address of a loaded function, offset can be used. 266 Instead of using original addresses of assembly code, 267 addresses subtracted from a base address can be printed. 268 Let's illustrate an example. 269 If a base address is 0XFFFFFFFF81624d50 as below, 270 271 ffffffff81624d50 <load0> 272 273 an address on assembly code has a specific absolute address as below 274 275 ffffffff816250b8:│ mov 0x8(%r14),%rdi 276 277 but if use_offset is 'true', an address subtracted from a base address is printed. 278 Default is true. This option is only applied to TUI. 279 280 368:│ mov 0x8(%r14),%rdi 281 282 annotate.jump_arrows:: 283 There can be jump instruction among assembly code. 284 Depending on a boolean value of jump_arrows, 285 arrows can be printed or not which represent 286 where do the instruction jump into as below. 287 288 │ ┌──jmp 1333 289 │ │ xchg %ax,%ax 290 │1330:│ mov %r15,%r10 291 │1333:└─→cmp %r15,%r14 292 293 If jump_arrow is 'false', the arrows isn't printed as below. 294 Default is 'false'. 295 296 │ ↓ jmp 1333 297 │ xchg %ax,%ax 298 │1330: mov %r15,%r10 299 │1333: cmp %r15,%r14 300 301 annotate.show_linenr:: 302 When showing source code if this option is 'true', 303 line numbers are printed as below. 304 305 │1628 if (type & PERF_SAMPLE_IDENTIFIER) { 306 │ ↓ jne 508 307 │1628 data->id = *array; 308 │1629 array++; 309 │1630 } 310 311 However if this option is 'false', they aren't printed as below. 312 Default is 'false'. 313 314 │ if (type & PERF_SAMPLE_IDENTIFIER) { 315 │ ↓ jne 508 316 │ data->id = *array; 317 │ array++; 318 │ } 319 320 annotate.show_nr_jumps:: 321 Let's see a part of assembly code. 322 323 │1382: movb $0x1,-0x270(%rbp) 324 325 If use this, the number of branches jumping to that address can be printed as below. 326 Default is 'false'. 327 328 │1 1382: movb $0x1,-0x270(%rbp) 329 330 annotate.show_total_period:: 331 To compare two records on an instruction base, with this option 332 provided, display total number of samples that belong to a line 333 in assembly code. If this option is 'true', total periods are printed 334 instead of percent values as below. 335 336 302 │ mov %eax,%eax 337 338 But if this option is 'false', percent values for overhead are printed i.e. 339 Default is 'false'. 340 341 99.93 │ mov %eax,%eax 342 343 annotate.offset_level:: 344 Default is '1', meaning just jump targets will have offsets show right beside 345 the instruction. When set to '2' 'call' instructions will also have its offsets 346 shown, 3 or higher will show offsets for all instructions. 347 348hist.*:: 349 hist.percentage:: 350 This option control the way to calculate overhead of filtered entries - 351 that means the value of this option is effective only if there's a 352 filter (by comm, dso or symbol name). Suppose a following example: 353 354 Overhead Symbols 355 ........ ....... 356 33.33% foo 357 33.33% bar 358 33.33% baz 359 360 This is an original overhead and we'll filter out the first 'foo' 361 entry. The value of 'relative' would increase the overhead of 'bar' 362 and 'baz' to 50.00% for each, while 'absolute' would show their 363 current overhead (33.33%). 364 365ui.*:: 366 ui.show-headers:: 367 This option controls display of column headers (like 'Overhead' and 'Symbol') 368 in 'report' and 'top'. If this option is false, they are hidden. 369 This option is only applied to TUI. 370 371call-graph.*:: 372 When sub-commands 'top' and 'report' work with -g/—-children 373 there're options in control of call-graph. 374 375 call-graph.record-mode:: 376 The record-mode can be 'fp' (frame pointer), 'dwarf' and 'lbr'. 377 The value of 'dwarf' is effective only if perf detect needed library 378 (libunwind or a recent version of libdw). 379 'lbr' only work for cpus that support it. 380 381 call-graph.dump-size:: 382 The size of stack to dump in order to do post-unwinding. Default is 8192 (byte). 383 When using dwarf into record-mode, the default size will be used if omitted. 384 385 call-graph.print-type:: 386 The print-types can be graph (graph absolute), fractal (graph relative), 387 flat and folded. This option controls a way to show overhead for each callchain 388 entry. Suppose a following example. 389 390 Overhead Symbols 391 ........ ....... 392 40.00% foo 393 | 394 ---foo 395 | 396 |--50.00%--bar 397 | main 398 | 399 --50.00%--baz 400 main 401 402 This output is a 'fractal' format. The 'foo' came from 'bar' and 'baz' exactly 403 half and half so 'fractal' shows 50.00% for each 404 (meaning that it assumes 100% total overhead of 'foo'). 405 406 The 'graph' uses absolute overhead value of 'foo' as total so each of 407 'bar' and 'baz' callchain will have 20.00% of overhead. 408 If 'flat' is used, single column and linear exposure of call chains. 409 'folded' mean call chains are displayed in a line, separated by semicolons. 410 411 call-graph.order:: 412 This option controls print order of callchains. The default is 413 'callee' which means callee is printed at top and then followed by its 414 caller and so on. The 'caller' prints it in reverse order. 415 416 If this option is not set and report.children or top.children is 417 set to true (or the equivalent command line option is given), 418 the default value of this option is changed to 'caller' for the 419 execution of 'perf report' or 'perf top'. Other commands will 420 still default to 'callee'. 421 422 call-graph.sort-key:: 423 The callchains are merged if they contain same information. 424 The sort-key option determines a way to compare the callchains. 425 A value of 'sort-key' can be 'function' or 'address'. 426 The default is 'function'. 427 428 call-graph.threshold:: 429 When there're many callchains it'd print tons of lines. So perf omits 430 small callchains under a certain overhead (threshold) and this option 431 control the threshold. Default is 0.5 (%). The overhead is calculated 432 by value depends on call-graph.print-type. 433 434 call-graph.print-limit:: 435 This is a maximum number of lines of callchain printed for a single 436 histogram entry. Default is 0 which means no limitation. 437 438report.*:: 439 report.sort_order:: 440 Allows changing the default sort order from "comm,dso,symbol" to 441 some other default, for instance "sym,dso" may be more fitting for 442 kernel developers. 443 report.percent-limit:: 444 This one is mostly the same as call-graph.threshold but works for 445 histogram entries. Entries having an overhead lower than this 446 percentage will not be printed. Default is '0'. If percent-limit 447 is '10', only entries which have more than 10% of overhead will be 448 printed. 449 450 report.queue-size:: 451 This option sets up the maximum allocation size of the internal 452 event queue for ordering events. Default is 0, meaning no limit. 453 454 report.children:: 455 'Children' means functions called from another function. 456 If this option is true, 'perf report' cumulates callchains of children 457 and show (accumulated) total overhead as well as 'Self' overhead. 458 Please refer to the 'perf report' manual. The default is 'true'. 459 460 report.group:: 461 This option is to show event group information together. 462 Example output with this turned on, notice that there is one column 463 per event in the group, ref-cycles and cycles: 464 465 # group: {ref-cycles,cycles} 466 # ======== 467 # 468 # Samples: 7K of event 'anon group { ref-cycles, cycles }' 469 # Event count (approx.): 6876107743 470 # 471 # Overhead Command Shared Object Symbol 472 # ................ ....... ................. ................... 473 # 474 99.84% 99.76% noploop noploop [.] main 475 0.07% 0.00% noploop ld-2.15.so [.] strcmp 476 0.03% 0.00% noploop [kernel.kallsyms] [k] timerqueue_del 477 478top.*:: 479 top.children:: 480 Same as 'report.children'. So if it is enabled, the output of 'top' 481 command will have 'Children' overhead column as well as 'Self' overhead 482 column by default. 483 The default is 'true'. 484 485man.*:: 486 man.viewer:: 487 This option can assign a tool to view manual pages when 'help' 488 subcommand was invoked. Supported tools are 'man', 'woman' 489 (with emacs client) and 'konqueror'. Default is 'man'. 490 491 New man viewer tool can be also added using 'man.<tool>.cmd' 492 or use different path using 'man.<tool>.path' config option. 493 494pager.*:: 495 pager.<subcommand>:: 496 When the subcommand is run on stdio, determine whether it uses 497 pager or not based on this value. Default is 'unspecified'. 498 499kmem.*:: 500 kmem.default:: 501 This option decides which allocator is to be analyzed if neither 502 '--slab' nor '--page' option is used. Default is 'slab'. 503 504record.*:: 505 record.build-id:: 506 This option can be 'cache', 'no-cache' or 'skip'. 507 'cache' is to post-process data and save/update the binaries into 508 the build-id cache (in ~/.debug). This is the default. 509 But if this option is 'no-cache', it will not update the build-id cache. 510 'skip' skips post-processing and does not update the cache. 511 512diff.*:: 513 diff.order:: 514 This option sets the number of columns to sort the result. 515 The default is 0, which means sorting by baseline. 516 Setting it to 1 will sort the result by delta (or other 517 compute method selected). 518 519 diff.compute:: 520 This options sets the method for computing the diff result. 521 Possible values are 'delta', 'delta-abs', 'ratio' and 522 'wdiff'. Default is 'delta'. 523 524trace.*:: 525 trace.add_events:: 526 Allows adding a set of events to add to the ones specified 527 by the user, or use as a default one if none was specified. 528 The initial use case is to add augmented_raw_syscalls.o to 529 activate the 'perf trace' logic that looks for syscall 530 pointer contents after the normal tracepoint payload. 531 532 trace.args_alignment:: 533 Number of columns to align the argument list, default is 70, 534 use 40 for the strace default, zero to no alignment. 535 536 trace.no_inherit:: 537 Do not follow children threads. 538 539 trace.show_arg_names:: 540 Should syscall argument names be printed? If not then trace.show_zeros 541 will be set. 542 543 trace.show_duration:: 544 Show syscall duration. 545 546 trace.show_prefix:: 547 If set to 'yes' will show common string prefixes in tables. The default 548 is to remove the common prefix in things like "MAP_SHARED", showing just "SHARED". 549 550 trace.show_timestamp:: 551 Show syscall start timestamp. 552 553 trace.show_zeros:: 554 Do not suppress syscall arguments that are equal to zero. 555 556SEE ALSO 557-------- 558linkperf:perf[1] 559