Commit | Line | Data |
---|---|---|
0bec253c | 1 | perf-report(1) |
c1c2365a | 2 | ============== |
0bec253c IM |
3 | |
4 | NAME | |
5 | ---- | |
23ac9cbe | 6 | perf-report - Read perf.data (created by perf record) and display the profile |
0bec253c IM |
7 | |
8 | SYNOPSIS | |
9 | -------- | |
10 | [verse] | |
11 | 'perf report' [-i <file> | --input=file] | |
12 | ||
13 | DESCRIPTION | |
14 | ----------- | |
15 | This command displays the performance counter profile information recorded | |
b0a28589 | 16 | via perf record. |
0bec253c IM |
17 | |
18 | OPTIONS | |
19 | ------- | |
20 | -i:: | |
21 | --input=:: | |
efad1415 | 22 | Input file name. (default: perf.data unless stdin is a fifo) |
e04fffc3 SB |
23 | |
24 | -v:: | |
25 | --verbose:: | |
26 | Be more verbose. (show symbol address, etc) | |
27 | ||
ca2b900f ZT |
28 | -n:: |
29 | --show-nr-samples:: | |
e3d7e183 | 30 | Show the number of samples for each symbol |
e04fffc3 | 31 | |
b272a59d | 32 | --show-cpu-utilization:: |
e04fffc3 SB |
33 | Show sample percentage for different cpu modes. |
34 | ||
ca2b900f ZT |
35 | -T:: |
36 | --threads:: | |
1f91d5fd NK |
37 | Show per-thread event counters. The input data file should be recorded |
38 | with -s option. | |
c8e66720 | 39 | -c:: |
cc8b88b1 ACM |
40 | --comms=:: |
41 | Only consider symbols in these comms. CSV that understands | |
f2148330 NK |
42 | file://filename entries. This option will affect the percentage of |
43 | the overhead column. See --percentage for more info. | |
e03eaa40 DA |
44 | --pid=:: |
45 | Only show events for given process ID (comma separated list). | |
46 | ||
47 | --tid=:: | |
48 | Only show events for given thread ID (comma separated list). | |
f2148330 NK |
49 | -d:: |
50 | --dsos=:: | |
51 | Only consider symbols in these dsos. CSV that understands | |
52 | file://filename entries. This option will affect the percentage of | |
53 | the overhead column. See --percentage for more info. | |
7bec7a91 ACM |
54 | -S:: |
55 | --symbols=:: | |
56 | Only consider these symbols. CSV that understands | |
f2148330 NK |
57 | file://filename entries. This option will affect the percentage of |
58 | the overhead column. See --percentage for more info. | |
0bec253c | 59 | |
fde0eeab NK |
60 | --symbol-filter=:: |
61 | Only show symbols that match (partially) with this filter. | |
62 | ||
e04fffc3 SB |
63 | -U:: |
64 | --hide-unresolved:: | |
65 | Only display entries resolved to a symbol. | |
66 | ||
c351c281 ACM |
67 | -s:: |
68 | --sort=:: | |
9811360e NK |
69 | Sort histogram entries by given key(s) - multiple keys can be specified |
70 | in CSV format. Following sort keys are available: | |
2e7ea3ab | 71 | pid, comm, dso, symbol, parent, cpu, socket, srcline, weight, local_weight. |
9811360e NK |
72 | |
73 | Each key has following meaning: | |
74 | ||
75 | - comm: command (name) of the task which can be read via /proc/<pid>/comm | |
76 | - pid: command and tid of the task | |
77 | - dso: name of library or module executed at the time of sample | |
78 | - symbol: name of function executed at the time of sample | |
79 | - parent: name of function matched to the parent regex filter. Unmatched | |
80 | entries are displayed as "[other]". | |
81 | - cpu: cpu number the task ran at the time of sample | |
2e7ea3ab | 82 | - socket: processor socket number the task ran at the time of sample |
9811360e | 83 | - srcline: filename and line number executed at the time of sample. The |
f5d05bce | 84 | DWARF debugging info must be provided. |
31191a85 AK |
85 | - srcfile: file name of the source file of the same. Requires dwarf |
86 | information. | |
475eeab9 AK |
87 | - weight: Event specific weight, e.g. memory latency or transaction |
88 | abort cost. This is the global weight. | |
89 | - local_weight: Local weight version of the weight above. | |
90 | - transaction: Transaction abort flags. | |
a2ce067e NK |
91 | - overhead: Overhead percentage of sample |
92 | - overhead_sys: Overhead percentage of sample running in system mode | |
93 | - overhead_us: Overhead percentage of sample running in user mode | |
94 | - overhead_guest_sys: Overhead percentage of sample running in system mode | |
95 | on guest machine | |
96 | - overhead_guest_us: Overhead percentage of sample running in user mode on | |
97 | guest machine | |
98 | - sample: Number of sample | |
99 | - period: Raw number of event count of sample | |
9811360e NK |
100 | |
101 | By default, comm, dso and symbol keys are used. | |
102 | (i.e. --sort comm,dso,symbol) | |
103 | ||
104 | If --branch-stack option is used, following sort keys are also | |
105 | available: | |
9811360e NK |
106 | |
107 | - dso_from: name of library or module branched from | |
108 | - dso_to: name of library or module branched to | |
109 | - symbol_from: name of function branched from | |
110 | - symbol_to: name of function branched to | |
508be0df AK |
111 | - srcline_from: source file and line branched from |
112 | - srcline_to: source file and line branched to | |
9811360e | 113 | - mispredict: "N" for predicted branch, "Y" for mispredicted branch |
f5d05bce AK |
114 | - in_tx: branch in TSX transaction |
115 | - abort: TSX transaction abort. | |
0e332f03 | 116 | - cycles: Cycles in basic block |
9811360e NK |
117 | |
118 | And default sort keys are changed to comm, dso_from, symbol_from, dso_to | |
119 | and symbol_to, see '--branch-stack'. | |
c351c281 | 120 | |
c6f5f6b6 NK |
121 | If the --mem-mode option is used, the following sort keys are also available |
122 | (incompatible with --branch-stack): | |
123 | symbol_daddr, dso_daddr, locked, tlb, mem, snoop, dcacheline. | |
124 | ||
125 | - symbol_daddr: name of data symbol being executed on at the time of sample | |
126 | - dso_daddr: name of library or module containing the data being executed | |
127 | on at the time of the sample | |
128 | - locked: whether the bus was locked at the time of the sample | |
129 | - tlb: type of tlb access for the data at the time of the sample | |
130 | - mem: type of memory access for the data at the time of the sample | |
131 | - snoop: type of snoop (if any) for the data at the time of the sample | |
132 | - dcacheline: the cacheline the data address is on at the time of the sample | |
133 | ||
134 | And the default sort keys are changed to local_weight, mem, sym, dso, | |
135 | symbol_daddr, dso_daddr, snoop, tlb, locked, see '--mem-mode'. | |
136 | ||
4c96bee0 NK |
137 | If the data file has tracepoint event(s), following (dynamic) sort keys |
138 | are also available: | |
139 | trace, trace_fields, [<event>.]<field>[/raw] | |
140 | ||
141 | - trace: pretty printed trace output in a single column | |
142 | - trace_fields: fields in tracepoints in separate columns | |
143 | - <field name>: optional event and field name for a specific field | |
144 | ||
145 | The last form consists of event and field names. If event name is | |
146 | omitted, it searches all events for matching field name. The matched | |
147 | field will be shown only for the event has the field. The event name | |
148 | supports substring match so user doesn't need to specify full subsystem | |
149 | and event name everytime. For example, 'sched:sched_switch' event can | |
150 | be shortened to 'switch' as long as it's not ambiguous. Also event can | |
151 | be specified by its index (starting from 1) preceded by the '%'. | |
152 | So '%1' is the first event, '%2' is the second, and so on. | |
153 | ||
154 | The field name can have '/raw' suffix which disables pretty printing | |
155 | and shows raw field value like hex numbers. The --raw-trace option | |
156 | has the same effect for all dynamic sort keys. | |
157 | ||
158 | The default sort keys are changed to 'trace' if all events in the data | |
159 | file are tracepoint. | |
160 | ||
a7d945bc NK |
161 | -F:: |
162 | --fields=:: | |
163 | Specify output field - multiple keys can be specified in CSV format. | |
164 | Following fields are available: | |
793aaaab | 165 | overhead, overhead_sys, overhead_us, overhead_children, sample and period. |
a7d945bc NK |
166 | Also it can contain any sort key(s). |
167 | ||
168 | By default, every sort keys not specified in -F will be appended | |
169 | automatically. | |
170 | ||
e04fffc3 SB |
171 | -p:: |
172 | --parent=<regex>:: | |
9811360e NK |
173 | A regex filter to identify parent. The parent is a caller of this |
174 | function and searched through the callchain, thus it requires callchain | |
175 | information recorded. The pattern is in the exteneded regex format and | |
176 | defaults to "\^sys_|^do_page_fault", see '--sort parent'. | |
e04fffc3 SB |
177 | |
178 | -x:: | |
179 | --exclude-other:: | |
180 | Only display entries with parent-match. | |
181 | ||
52d422de | 182 | -w:: |
e04fffc3 | 183 | --column-widths=<width[,width...]>:: |
52d422de | 184 | Force each column width to the provided list, for large terminal |
cf59002f | 185 | readability. 0 means no limit (default behavior). |
52d422de ACM |
186 | |
187 | -t:: | |
188 | --field-separator=:: | |
52d422de | 189 | Use a special separator character and don't pad with spaces, replacing |
e04fffc3 | 190 | all occurrences of this separator in symbol names (and other output) |
52d422de ACM |
191 | with a '.' character, that thus it's the only non valid separator. |
192 | ||
e04fffc3 SB |
193 | -D:: |
194 | --dump-raw-trace:: | |
195 | Dump raw trace in ASCII. | |
196 | ||
76a26549 | 197 | -g:: |
f2af0086 | 198 | --call-graph=<print_type,threshold[,print_limit],order,sort_key[,branch],value>:: |
76a26549 | 199 | Display call chains using type, min percent threshold, print limit, |
f2af0086 NK |
200 | call order, sort key, optional branch and value. Note that ordering of |
201 | parameters is not fixed so any parement can be given in an arbitraty order. | |
202 | One exception is the print_limit which should be preceded by threshold. | |
76a26549 NK |
203 | |
204 | print_type can be either: | |
e04fffc3 | 205 | - flat: single column, linear exposure of call chains. |
76a26549 | 206 | - graph: use a graph tree, displaying absolute overhead rates. (default) |
ec7ba4ea | 207 | - fractal: like graph, but displays relative rates. Each branch of |
76a26549 | 208 | the tree is considered as a new profiled object. |
26e77924 | 209 | - folded: call chains are displayed in a line, separated by semicolons |
76a26549 NK |
210 | - none: disable call chain display. |
211 | ||
212 | threshold is a percentage value which specifies a minimum percent to be | |
213 | included in the output call graph. Default is 0.5 (%). | |
214 | ||
215 | print_limit is only applied when stdio interface is used. It's to limit | |
216 | number of call graph entries in a single hist entry. Note that it needs | |
217 | to be given after threshold (but not necessarily consecutive). | |
218 | Default is 0 (unlimited). | |
d797fdc5 SL |
219 | |
220 | order can be either: | |
221 | - callee: callee based call graph. | |
222 | - caller: inverted caller based call graph. | |
76a26549 | 223 | Default is 'caller' when --children is used, otherwise 'callee'. |
d797fdc5 | 224 | |
76a26549 NK |
225 | sort_key can be: |
226 | - function: compare on functions (default) | |
99571ab3 AK |
227 | - address: compare on individual code addresses |
228 | ||
8b7bad58 | 229 | branch can be: |
76a26549 NK |
230 | - branch: include last branch information in callgraph when available. |
231 | Usually more convenient to use --branch-history for this. | |
d797fdc5 | 232 | |
f2af0086 NK |
233 | value can be: |
234 | - percent: diplay overhead percent (default) | |
235 | - period: display event period | |
236 | - count: display event count | |
237 | ||
793aaaab NK |
238 | --children:: |
239 | Accumulate callchain of children to parent entry so that then can | |
240 | show up in the output. The output will have a new "Children" column | |
241 | and will be sorted on the data. It requires callchains are recorded. | |
dd309207 | 242 | See the `overhead calculation' section for more details. |
793aaaab | 243 | |
91e95617 WL |
244 | --max-stack:: |
245 | Set the stack depth limit when parsing the callchain, anything | |
246 | beyond the specified depth will be ignored. This is a trade-off | |
247 | between information loss and faster processing especially for | |
248 | workloads that can have a very long callchain stack. | |
40862a7b AH |
249 | Note that when using the --itrace option the synthesized callchain size |
250 | will override this value if the synthesized callchain size is bigger. | |
91e95617 | 251 | |
fe176085 | 252 | Default: 127 |
91e95617 | 253 | |
d797fdc5 SL |
254 | -G:: |
255 | --inverted:: | |
256 | alias for inverted caller based call graph. | |
ec7ba4ea | 257 | |
b21484f1 GP |
258 | --ignore-callees=<regex>:: |
259 | Ignore callees of the function(s) matching the given regex. | |
260 | This has the effect of collecting the callers of each such | |
261 | function into one place in the call-graph tree. | |
262 | ||
e04fffc3 SB |
263 | --pretty=<key>:: |
264 | Pretty printing style. key: normal, raw | |
265 | ||
8b9e74eb ACM |
266 | --stdio:: Use the stdio interface. |
267 | ||
175b968b ACM |
268 | --stdio-color:: |
269 | 'always', 'never' or 'auto', allowing configuring color output | |
270 | via the command line, in addition to via "color.ui" .perfconfig. | |
271 | Use '--stdio-color always' to generate color even when redirecting | |
272 | to a pipe or file. Using just '--stdio-color' is equivalent to | |
273 | using 'always'. | |
274 | ||
8b9e74eb ACM |
275 | --tui:: Use the TUI interface, that is integrated with annotate and allows |
276 | zooming into DSOs or threads, among other features. Use of --tui | |
277 | requires a tty, if one is not present, as when piping to other | |
278 | commands, the stdio interface is used. | |
279 | ||
c31a9457 PE |
280 | --gtk:: Use the GTK2 interface. |
281 | ||
e04fffc3 SB |
282 | -k:: |
283 | --vmlinux=<file>:: | |
284 | vmlinux pathname | |
285 | ||
b226a5a7 DA |
286 | --kallsyms=<file>:: |
287 | kallsyms pathname | |
288 | ||
e04fffc3 SB |
289 | -m:: |
290 | --modules:: | |
291 | Load module symbols. WARNING: This should only be used with -k and | |
292 | a LIVE kernel. | |
293 | ||
294 | -f:: | |
295 | --force:: | |
e0be62cc | 296 | Don't do ownership validation. |
e04fffc3 | 297 | |
ec5761ea DA |
298 | --symfs=<directory>:: |
299 | Look for files with symbols relative to this directory. | |
300 | ||
c8e66720 | 301 | -C:: |
5d67be97 AB |
302 | --cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can |
303 | be provided as a comma-separated list with no space: 0,1. Ranges of | |
304 | CPUs are specified with -: 0-2. Default is to report samples on all | |
305 | CPUs. | |
306 | ||
f69b64f7 AK |
307 | -M:: |
308 | --disassembler-style=:: Set disassembler style for objdump. | |
309 | ||
64c6f0c7 ACM |
310 | --source:: |
311 | Interleave source code with assembly code. Enabled by default, | |
312 | disable with --no-source. | |
313 | ||
314 | --asm-raw:: | |
315 | Show raw instruction encoding of assembly instructions. | |
316 | ||
3f2728bd ACM |
317 | --show-total-period:: Show a column with the sum of periods. |
318 | ||
fbe96f29 SE |
319 | -I:: |
320 | --show-info:: | |
321 | Display extended information about the perf.data file. This adds | |
322 | information which may be very large and thus may clutter the display. | |
323 | It currently includes: cpu and numa topology of the host system. | |
324 | ||
b50311dc RAV |
325 | -b:: |
326 | --branch-stack:: | |
327 | Use the addresses of sampled taken branches instead of the instruction | |
328 | address to build the histograms. To generate meaningful output, the | |
993ac88d SE |
329 | perf.data file must have been obtained using perf record -b or |
330 | perf record --branch-filter xxx where xxx is a branch filter option. | |
331 | perf report is able to auto-detect whether a perf.data file contains | |
332 | branch stacks and it will automatically switch to the branch view mode, | |
333 | unless --no-branch-stack is used. | |
b50311dc | 334 | |
fa94c36c AK |
335 | --branch-history:: |
336 | Add the addresses of sampled taken branches to the callstack. | |
337 | This allows to examine the path the program took to each sample. | |
338 | The data collection must have used -b (or -j) and -g. | |
339 | ||
7a4ec938 MB |
340 | --objdump=<path>:: |
341 | Path to objdump binary. | |
342 | ||
01d14f16 NK |
343 | --group:: |
344 | Show event group information together. | |
345 | ||
328ccdac NK |
346 | --demangle:: |
347 | Demangle symbol names to human readable form. It's enabled by default, | |
348 | disable with --no-demangle. | |
349 | ||
763122ad AK |
350 | --demangle-kernel:: |
351 | Demangle kernel symbol names to human readable form (for C++ kernels). | |
352 | ||
75e906c9 DZ |
353 | --mem-mode:: |
354 | Use the data addresses of samples in addition to instruction addresses | |
355 | to build the histograms. To generate meaningful output, the perf.data | |
356 | file must have been obtained using perf record -d -W and using a | |
357 | special event -e cpu/mem-loads/ or -e cpu/mem-stores/. See | |
358 | 'perf mem' for simpler access. | |
359 | ||
064f1981 NK |
360 | --percent-limit:: |
361 | Do not show entries which have an overhead under that percent. | |
1ba2fc6d NK |
362 | (Default: 0). Note that this option also sets the percent limit (threshold) |
363 | of callchains. However the default value of callchain threshold is | |
364 | different than the default value of hist entries. Please see the | |
365 | --call-graph option for details. | |
064f1981 | 366 | |
f2148330 NK |
367 | --percentage:: |
368 | Determine how to display the overhead percentage of filtered entries. | |
369 | Filters can be applied by --comms, --dsos and/or --symbols options and | |
370 | Zoom operations on the TUI (thread, dso, etc). | |
371 | ||
372 | "relative" means it's relative to filtered entries only so that the | |
373 | sum of shown entries will be always 100%. "absolute" means it retains | |
374 | the original value before and after the filter is applied. | |
375 | ||
5cfe2c82 JO |
376 | --header:: |
377 | Show header information in the perf.data file. This includes | |
378 | various information like hostname, OS and perf version, cpu/mem | |
379 | info, perf command line, event list and so on. Currently only | |
380 | --stdio output supports this feature. | |
381 | ||
382 | --header-only:: | |
383 | Show only perf.data header (forces --stdio). | |
384 | ||
46690a80 DA |
385 | --time:: |
386 | Only analyze samples within given time window: <start>,<stop>. Times | |
387 | have the format seconds.microseconds. If start is not given (i.e., time | |
388 | string is ',x.y') then analysis starts at the beginning of the file. If | |
389 | stop time is not given (i.e, time string is 'x.y,') then analysis goes | |
390 | to end of file. | |
391 | ||
520a2ebc AH |
392 | --itrace:: |
393 | Options for decoding instruction tracing data. The options are: | |
394 | ||
60b88d87 | 395 | include::itrace.txt[] |
520a2ebc AH |
396 | |
397 | To disable decoding entirely, use --no-itrace. | |
398 | ||
a9710ba0 AK |
399 | --full-source-path:: |
400 | Show the full path for source files for srcline output. | |
64a7e61f | 401 | |
9e207ddf KL |
402 | --show-ref-call-graph:: |
403 | When multiple events are sampled, it may not be needed to collect | |
404 | callgraphs for all of them. The sample sites are usually nearby, | |
405 | and it's enough to collect the callgraphs on a reference event. | |
406 | So user can use "call-graph=no" event modifier to disable callgraph | |
407 | for other events to reduce the overhead. | |
408 | However, perf report cannot show callgraphs for the event which | |
409 | disable the callgraph. | |
410 | This option extends the perf report to show reference callgraphs, | |
411 | which collected by reference event, in no callgraph event. | |
412 | ||
21394d94 KL |
413 | --socket-filter:: |
414 | Only report the samples on the processor socket that match with this filter | |
415 | ||
053a3989 NK |
416 | --raw-trace:: |
417 | When displaying traceevent output, do not use print fmt or plugins. | |
418 | ||
4251446d NK |
419 | --hierarchy:: |
420 | Enable hierarchical output. | |
421 | ||
64a7e61f AH |
422 | include::callchain-overhead-calculation.txt[] |
423 | ||
0bec253c IM |
424 | SEE ALSO |
425 | -------- | |
64c6f0c7 | 426 | linkperf:perf-stat[1], linkperf:perf-annotate[1] |