Commit | Line | Data |
---|---|---|
514f1c67 ACM |
1 | perf-trace(1) |
2 | ============= | |
3 | ||
4 | NAME | |
5 | ---- | |
6 | perf-trace - strace inspired tool | |
7 | ||
8 | SYNOPSIS | |
9 | -------- | |
10 | [verse] | |
11 | 'perf trace' | |
5e2485b1 | 12 | 'perf trace record' |
514f1c67 ACM |
13 | |
14 | DESCRIPTION | |
15 | ----------- | |
16 | This command will show the events associated with the target, initially | |
17 | syscalls, but other system events like pagefaults, task lifetime events, | |
18 | scheduling events, etc. | |
19 | ||
5e2485b1 DA |
20 | This is a live mode tool in addition to working with perf.data files like |
21 | the other perf tools. Files can be generated using the 'perf record' command | |
22 | but the session needs to include the raw_syscalls events (-e 'raw_syscalls:*'). | |
96355f2c | 23 | Alternatively, 'perf trace record' can be used as a shortcut to |
5e2485b1 DA |
24 | automatically include the raw_syscalls events when writing events to a file. |
25 | ||
26 | The following options apply to perf trace; options to perf trace record are | |
27 | found in the perf record man page. | |
514f1c67 ACM |
28 | |
29 | OPTIONS | |
30 | ------- | |
31 | ||
ac9be8ee | 32 | -a:: |
514f1c67 ACM |
33 | --all-cpus:: |
34 | System-wide collection from all CPUs. | |
35 | ||
2ae3a312 ACM |
36 | -e:: |
37 | --expr:: | |
d303e85a | 38 | List of syscalls to show, currently only syscall names. |
b059efdf ACM |
39 | Prefixing with ! shows all syscalls but the ones specified. You may |
40 | need to escape it. | |
2ae3a312 | 41 | |
c24ff998 ACM |
42 | -o:: |
43 | --output=:: | |
44 | Output file name. | |
45 | ||
514f1c67 ACM |
46 | -p:: |
47 | --pid=:: | |
48 | Record events on existing process ID (comma separated list). | |
49 | ||
ac9be8ee | 50 | -t:: |
514f1c67 ACM |
51 | --tid=:: |
52 | Record events on existing thread ID (comma separated list). | |
53 | ||
ac9be8ee | 54 | -u:: |
514f1c67 ACM |
55 | --uid=:: |
56 | Record events in threads owned by uid. Name or number. | |
57 | ||
f078c385 ACM |
58 | --filter-pids=:: |
59 | Filter out events for these pids and for 'trace' itself (comma separated list). | |
60 | ||
7c304ee0 ACM |
61 | -v:: |
62 | --verbose=:: | |
63 | Verbosity level. | |
64 | ||
514f1c67 ACM |
65 | --no-inherit:: |
66 | Child tasks do not inherit counters. | |
67 | ||
ac9be8ee | 68 | -m:: |
514f1c67 | 69 | --mmap-pages=:: |
27050f53 JO |
70 | Number of mmap data pages (must be a power of two) or size |
71 | specification with appended unit character - B/K/M/G. The | |
72 | size is rounded up to have nearest pages power of two value. | |
514f1c67 | 73 | |
ac9be8ee | 74 | -C:: |
514f1c67 ACM |
75 | --cpu:: |
76 | Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a | |
77 | comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. | |
78 | In per-thread mode with inheritance mode on (default), Events are captured only when | |
79 | the thread executes on the designated CPUs. Default is to monitor all CPUs. | |
80 | ||
ae9ed035 ACM |
81 | --duration: |
82 | Show only events that had a duration greater than N.M ms. | |
83 | ||
1302d88e ACM |
84 | --sched: |
85 | Accrue thread runtime and provide a summary at the end of the session. | |
86 | ||
6810fc91 DA |
87 | -i |
88 | --input | |
89 | Process events from a given perf data file. | |
90 | ||
4bb09192 DA |
91 | -T |
92 | --time | |
93 | Print full timestamp rather time relative to first sample. | |
94 | ||
50c95cbd ACM |
95 | --comm:: |
96 | Show process COMM right beside its ID, on by default, disable with --no-comm. | |
97 | ||
fd2eabaf | 98 | -s:: |
bf2575c1 | 99 | --summary:: |
fd2eabaf DA |
100 | Show only a summary of syscalls by thread with min, max, and average times |
101 | (in msec) and relative stddev. | |
102 | ||
103 | -S:: | |
104 | --with-summary:: | |
105 | Show all syscalls followed by a summary by thread with min, max, and | |
106 | average times (in msec) and relative stddev. | |
bf2575c1 | 107 | |
c522739d ACM |
108 | --tool_stats:: |
109 | Show tool stats such as number of times fd->pathname was discovered thru | |
110 | hooking the open syscall return + vfs_getname or via reading /proc/pid/fd, etc. | |
111 | ||
598d02c5 SF |
112 | -F=[all|min|maj]:: |
113 | --pf=[all|min|maj]:: | |
114 | Trace pagefaults. Optionally, you can specify whether you want minor, | |
115 | major or all pagefaults. Default value is maj. | |
116 | ||
e281a960 SF |
117 | --syscalls:: |
118 | Trace system calls. This options is enabled by default. | |
119 | ||
566a0885 MW |
120 | --call-graph [mode,type,min[,limit],order[,key][,branch]]:: |
121 | Setup and enable call-graph (stack chain/backtrace) recording. | |
122 | See `--call-graph` section in perf-record and perf-report | |
123 | man pages for details. The ones that are most useful in 'perf trace' | |
124 | are 'dwarf' and 'lbr', where available, try: 'perf trace --call-graph dwarf'. | |
125 | ||
f3e459d1 ACM |
126 | Using this will, for the root user, bump the value of --mmap-pages to 4 |
127 | times the maximum for non-root users, based on the kernel.perf_event_mlock_kb | |
128 | sysctl. This is done only if the user doesn't specify a --mmap-pages value. | |
129 | ||
44621819 ACM |
130 | --kernel-syscall-graph:: |
131 | Show the kernel callchains on the syscall exit path. | |
132 | ||
77c92582 ACM |
133 | --event:: |
134 | Trace other events, see 'perf list' for a complete list. | |
135 | ||
c6d4a494 ACM |
136 | --max-stack:: |
137 | Set the stack depth limit when parsing the callchain, anything | |
138 | beyond the specified depth will be ignored. Note that at this point | |
139 | this is just about the presentation part, i.e. the kernel is still | |
140 | not limiting, the overhead of callchains needs to be set via the | |
141 | knobs in --call-graph dwarf. | |
142 | ||
05614993 ACM |
143 | Implies '--call-graph dwarf' when --call-graph not present on the |
144 | command line, on systems where DWARF unwinding was built in. | |
145 | ||
4cb93446 | 146 | Default: /proc/sys/kernel/perf_event_max_stack when present, 127 otherwise. |
c6d4a494 | 147 | |
5cf9c84e ACM |
148 | --min-stack:: |
149 | Set the stack depth limit when parsing the callchain, anything | |
150 | below the specified depth will be ignored. Disabled by default. | |
151 | ||
05614993 ACM |
152 | Implies '--call-graph dwarf' when --call-graph not present on the |
153 | command line, on systems where DWARF unwinding was built in. | |
154 | ||
9d9cad76 KL |
155 | --proc-map-timeout:: |
156 | When processing pre-existing threads /proc/XXX/mmap, it may take a long time, | |
157 | because the file may be huge. A time out is needed in such cases. | |
158 | This option sets the time out limit. The default value is 500 ms. | |
159 | ||
598d02c5 SF |
160 | PAGEFAULTS |
161 | ---------- | |
162 | ||
163 | When tracing pagefaults, the format of the trace is as follows: | |
164 | ||
165 | <min|maj>fault [<ip.symbol>+<ip.offset>] => <addr.dso@addr.offset> (<map type><addr level>). | |
166 | ||
167 | - min/maj indicates whether fault event is minor or major; | |
168 | - ip.symbol shows symbol for instruction pointer (the code that generated the | |
169 | fault); if no debug symbols available, perf trace will print raw IP; | |
170 | - addr.dso shows DSO for the faulted address; | |
171 | - map type is either 'd' for non-executable maps or 'x' for executable maps; | |
172 | - addr level is either 'k' for kernel dso or '.' for user dso. | |
173 | ||
174 | For symbols resolution you may need to install debugging symbols. | |
175 | ||
176 | Please be aware that duration is currently always 0 and doesn't reflect actual | |
177 | time it took for fault to be handled! | |
178 | ||
179 | When --verbose specified, perf trace tries to print all available information | |
180 | for both IP and fault address in the form of dso@symbol+offset. | |
181 | ||
182 | EXAMPLES | |
183 | -------- | |
184 | ||
e281a960 SF |
185 | Trace only major pagefaults: |
186 | ||
187 | $ perf trace --no-syscalls -F | |
188 | ||
598d02c5 SF |
189 | Trace syscalls, major and minor pagefaults: |
190 | ||
191 | $ perf trace -F all | |
192 | ||
193 | 1416.547 ( 0.000 ms): python/20235 majfault [CRYPTO_push_info_+0x0] => /lib/x86_64-linux-gnu/libcrypto.so.1.0.0@0x61be0 (x.) | |
194 | ||
195 | As you can see, there was major pagefault in python process, from | |
196 | CRYPTO_push_info_ routine which faulted somewhere in libcrypto.so. | |
197 | ||
514f1c67 ACM |
198 | SEE ALSO |
199 | -------- | |
200 | linkperf:perf-record[1], linkperf:perf-script[1] |