Lines Matching +full:foo +full:- +full:queue
1 perf-config(1)
5 ----
6 perf-config - Get and set variables in a configuration file.
9 --------
11 'perf config' [<file-option>] [section.name[=value] ...]
13 'perf config' [<file-option>] -l | --list
16 -----------
20 -------
22 -l::
23 --list::
26 --user::
30 --system::
31 For writing and reading options: write to system-wide
35 ------------------
39 The '$HOME/.perfconfig' file is used to store a per-user configuration.
41 store a system-wide default configuration.
44 configuration files by default, and options '--system' and '--user'
106 show-headers = true
108 [call-graph]
110 record-mode = fp
111 print-type = graph
113 sort-key = function
118 percent-limit = 0
119 queue-size = 0
129 % perf config ui.show-headers=false kmem.default=slab
133 % perf config --user report sort-order=srcline
138 % perf config --system colors.selected=yellow,green
142 % perf config call-graph.record-mode
146 % perf config report.queue-size call-graph.order report.children
150 % perf config --user call-graph.sort-order
154 % perf config --system buildid.dir
178 Basic key values are foreground-color 'red' and
179 background-color 'default'.
189 from sub-commands (top, report, annotate).
199 Colors for headers in the output of a sub-commands (top, report).
219 The recording tools also stores a hard link or copy in a per-user
301 │1628 data->id = *array;
310 │ data->id = *array;
317 │1382: movb $0x1,-0x270(%rbp)
322 │1 1382: movb $0x1,-0x270(%rbp)
344 This option control the way to calculate overhead of filtered entries -
350 33.33% foo
354 This is an original overhead and we'll filter out the first 'foo'
360 ui.show-headers::
365 call-graph.*::
366 When sub-commands 'top' and 'report' work with -g/—-children
367 there're options in control of call-graph.
369 call-graph.record-mode::
370 The record-mode can be 'fp' (frame pointer), 'dwarf' and 'lbr'.
375 call-graph.dump-size::
376 The size of stack to dump in order to do post-unwinding. Default is 8192 (byte).
377 When using dwarf into record-mode, the default size will be used if omitted.
379 call-graph.print-type::
380 The print-types can be graph (graph absolute), fractal (graph relative),
386 40.00% foo
388 ---foo
390 |--50.00%--bar
393 --50.00%--baz
396 This output is a 'fractal' format. The 'foo' came from 'bar' and 'baz' exactly
398 (meaning that it assumes 100% total overhead of 'foo').
400 The 'graph' uses absolute overhead value of 'foo' as total so each of
405 call-graph.order::
416 call-graph.sort-key::
418 The sort-key option determines a way to compare the callchains.
419 A value of 'sort-key' can be 'function' or 'address'.
422 call-graph.threshold::
426 by value depends on call-graph.print-type.
428 call-graph.print-limit::
437 report.percent-limit::
438 This one is mostly the same as call-graph.threshold but works for
440 percentage will not be printed. Default is '0'. If percent-limit
444 report.queue-size::
446 event queue for ordering events. Default is 0, meaning no limit.
457 per event in the group, ref-cycles and cycles:
459 # group: {ref-cycles,cycles}
462 # Samples: 7K of event 'anon group { ref-cycles, cycles }'
469 0.07% 0.00% noploop ld-2.15.so [.] strcmp
496 '--slab' nor '--page' option is used. Default is 'slab'.
499 record.build-id::
500 This option can be 'cache', 'no-cache' or 'skip'.
501 'cache' is to post-process data and save/update the binaries into
502 the build-id cache (in ~/.debug). This is the default.
503 But if this option is 'no-cache', it will not update the build-id cache.
504 'skip' skips post-processing and does not update the cache.
515 Possible values are 'delta', 'delta-abs', 'ratio' and
519 --------