xref: /linux/tools/perf/Documentation/perf-trace.txt (revision 9ea42ba4411ac502e83baa9c61bb037d0a7355a0)
1perf-trace(1)
2=============
3
4NAME
5----
6perf-trace - strace inspired tool
7
8SYNOPSIS
9--------
10[verse]
11'perf trace'
12'perf trace record'
13
14DESCRIPTION
15-----------
16This command will show the events associated with the target, initially
17syscalls, but other system events like pagefaults, task lifetime events,
18scheduling events, etc.
19
20This is a live mode tool in addition to working with perf.data files like
21the other perf tools. Files can be generated using the 'perf record' command
22but the session needs to include the raw_syscalls events (-e 'raw_syscalls:*').
23Alternatively, 'perf trace record' can be used as a shortcut to
24automatically include the raw_syscalls events when writing events to a file.
25
26The following options apply to perf trace; options to perf trace record are
27found in the perf record man page.
28
29OPTIONS
30-------
31
32-a::
33--all-cpus::
34        System-wide collection from all CPUs.
35
36-e::
37--expr::
38--event::
39	List of syscalls and other perf events (tracepoints, HW cache events,
40	etc) to show. Globbing is supported, e.g.: "epoll_*", "*msg*", etc.
41	See 'perf list' for a complete list of events.
42	Prefixing with ! shows all syscalls but the ones specified.  You may
43	need to escape it.
44
45-D msecs::
46--delay msecs::
47After starting the program, wait msecs before measuring. This is useful to
48filter out the startup phase of the program, which is often very different.
49
50-o::
51--output=::
52	Output file name.
53
54-p::
55--pid=::
56	Record events on existing process ID (comma separated list).
57
58-t::
59--tid=::
60        Record events on existing thread ID (comma separated list).
61
62-u::
63--uid=::
64        Record events in threads owned by uid. Name or number.
65
66-G::
67--cgroup::
68	Record events in threads in a cgroup.
69
70	Look for cgroups to set at the /sys/fs/cgroup/perf_event directory, then
71	remove the /sys/fs/cgroup/perf_event/ part and try:
72
73		perf trace -G A -e sched:*switch
74
75	Will set all raw_syscalls:sys_{enter,exit}, pgfault, vfs_getname, etc
76	_and_ sched:sched_switch to the 'A' cgroup, while:
77
78		perf trace -e sched:*switch -G A
79
80	will only set the sched:sched_switch event to the 'A' cgroup, all the
81	other events (raw_syscalls:sys_{enter,exit}, etc are left "without"
82	a cgroup (on the root cgroup, sys wide, etc).
83
84	Multiple cgroups:
85
86		perf trace -G A -e sched:*switch -G B
87
88	the syscall ones go to the 'A' cgroup, the sched:sched_switch goes
89	to the 'B' cgroup.
90
91--filter-pids=::
92	Filter out events for these pids and for 'trace' itself (comma separated list).
93
94-v::
95--verbose=::
96        Verbosity level.
97
98--no-inherit::
99	Child tasks do not inherit counters.
100
101-m::
102--mmap-pages=::
103	Number of mmap data pages (must be a power of two) or size
104	specification with appended unit character - B/K/M/G. The
105	size is rounded up to have nearest pages power of two value.
106
107-C::
108--cpu::
109Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a
110comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
111In per-thread mode with inheritance mode on (default), Events are captured only when
112the thread executes on the designated CPUs. Default is to monitor all CPUs.
113
114--duration::
115	Show only events that had a duration greater than N.M ms.
116
117--sched::
118	Accrue thread runtime and provide a summary at the end of the session.
119
120-i::
121--input::
122	Process events from a given perf data file.
123
124-T::
125--time::
126	Print full timestamp rather time relative to first sample.
127
128--comm::
129        Show process COMM right beside its ID, on by default, disable with --no-comm.
130
131-s::
132--summary::
133	Show only a summary of syscalls by thread with min, max, and average times
134    (in msec) and relative stddev.
135
136-S::
137--with-summary::
138	Show all syscalls followed by a summary by thread with min, max, and
139    average times (in msec) and relative stddev.
140
141--tool_stats::
142	Show tool stats such as number of times fd->pathname was discovered thru
143	hooking the open syscall return + vfs_getname or via reading /proc/pid/fd, etc.
144
145-f::
146--force::
147	Don't complain, do it.
148
149-F=[all|min|maj]::
150--pf=[all|min|maj]::
151	Trace pagefaults. Optionally, you can specify whether you want minor,
152	major or all pagefaults. Default value is maj.
153
154--syscalls::
155	Trace system calls. This options is enabled by default, disable with
156	--no-syscalls.
157
158--call-graph [mode,type,min[,limit],order[,key][,branch]]::
159        Setup and enable call-graph (stack chain/backtrace) recording.
160        See `--call-graph` section in perf-record and perf-report
161        man pages for details. The ones that are most useful in 'perf trace'
162        are 'dwarf' and 'lbr', where available, try: 'perf trace --call-graph dwarf'.
163
164        Using this will, for the root user, bump the value of --mmap-pages to 4
165        times the maximum for non-root users, based on the kernel.perf_event_mlock_kb
166        sysctl. This is done only if the user doesn't specify a --mmap-pages value.
167
168--kernel-syscall-graph::
169	 Show the kernel callchains on the syscall exit path.
170
171--max-stack::
172        Set the stack depth limit when parsing the callchain, anything
173        beyond the specified depth will be ignored. Note that at this point
174        this is just about the presentation part, i.e. the kernel is still
175        not limiting, the overhead of callchains needs to be set via the
176        knobs in --call-graph dwarf.
177
178        Implies '--call-graph dwarf' when --call-graph not present on the
179        command line, on systems where DWARF unwinding was built in.
180
181        Default: /proc/sys/kernel/perf_event_max_stack when present for
182                 live sessions (without --input/-i), 127 otherwise.
183
184--min-stack::
185        Set the stack depth limit when parsing the callchain, anything
186        below the specified depth will be ignored. Disabled by default.
187
188        Implies '--call-graph dwarf' when --call-graph not present on the
189        command line, on systems where DWARF unwinding was built in.
190
191--print-sample::
192	Print the PERF_RECORD_SAMPLE PERF_SAMPLE_ info for the
193	raw_syscalls:sys_{enter,exit} tracepoints, for debugging.
194
195--proc-map-timeout::
196	When processing pre-existing threads /proc/XXX/mmap, it may take a long time,
197	because the file may be huge. A time out is needed in such cases.
198	This option sets the time out limit. The default value is 500 ms.
199
200PAGEFAULTS
201----------
202
203When tracing pagefaults, the format of the trace is as follows:
204
205<min|maj>fault [<ip.symbol>+<ip.offset>] => <addr.dso@addr.offset> (<map type><addr level>).
206
207- min/maj indicates whether fault event is minor or major;
208- ip.symbol shows symbol for instruction pointer (the code that generated the
209  fault); if no debug symbols available, perf trace will print raw IP;
210- addr.dso shows DSO for the faulted address;
211- map type is either 'd' for non-executable maps or 'x' for executable maps;
212- addr level is either 'k' for kernel dso or '.' for user dso.
213
214For symbols resolution you may need to install debugging symbols.
215
216Please be aware that duration is currently always 0 and doesn't reflect actual
217time it took for fault to be handled!
218
219When --verbose specified, perf trace tries to print all available information
220for both IP and fault address in the form of dso@symbol+offset.
221
222EXAMPLES
223--------
224
225Trace only major pagefaults:
226
227 $ perf trace --no-syscalls -F
228
229Trace syscalls, major and minor pagefaults:
230
231 $ perf trace -F all
232
233  1416.547 ( 0.000 ms): python/20235 majfault [CRYPTO_push_info_+0x0] => /lib/x86_64-linux-gnu/libcrypto.so.1.0.0@0x61be0 (x.)
234
235  As you can see, there was major pagefault in python process, from
236  CRYPTO_push_info_ routine which faulted somewhere in libcrypto.so.
237
238SEE ALSO
239--------
240linkperf:perf-record[1], linkperf:perf-script[1]
241