tjh.dev
Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
1perf-lock(1)
2============
3
4NAME
5----
6perf-lock - Analyze lock events
7
8SYNOPSIS
9--------
10[verse]
11'perf lock' {record|report|script|info|contention}
12
13DESCRIPTION
14-----------
15You can analyze various lock behaviours
16and statistics with this 'perf lock' command.
17
18 'perf lock record <command>' records lock events
19 between start and end <command>. And this command
20 produces the file "perf.data" which contains tracing
21 results of lock events.
22
23 'perf lock report' reports statistical data.
24
25 'perf lock script' shows raw lock events.
26
27 'perf lock info' shows metadata like threads or addresses
28 of lock instances.
29
30 'perf lock contention' shows contention statistics.
31
32COMMON OPTIONS
33--------------
34
35-i::
36--input=<file>::
37 Input file name. (default: perf.data unless stdin is a fifo)
38
39-v::
40--verbose::
41 Be more verbose (show symbol address, etc).
42
43-D::
44--dump-raw-trace::
45 Dump raw trace in ASCII.
46
47-f::
48--force::
49 Don't complain, do it.
50
51--vmlinux=<file>::
52 vmlinux pathname
53
54--kallsyms=<file>::
55 kallsyms pathname
56
57
58REPORT OPTIONS
59--------------
60
61-k::
62--key=<value>::
63 Sorting key. Possible values: acquired (default), contended,
64 avg_wait, wait_total, wait_max, wait_min.
65
66-F::
67--field=<value>::
68 Output fields. By default it shows all the fields but users can
69 customize that using this. Possible values: acquired, contended,
70 avg_wait, wait_total, wait_max, wait_min.
71
72-c::
73--combine-locks::
74 Merge lock instances in the same class (based on name).
75
76-t::
77--threads::
78 The -t option is to show per-thread lock stat like below:
79
80 $ perf lock report -t -F acquired,contended,avg_wait
81
82 Name acquired contended avg wait (ns)
83
84 perf 240569 9 5784
85 swapper 106610 19 543
86 :15789 17370 2 14538
87 ContainerMgr 8981 6 874
88 sleep 5275 1 11281
89 ContainerThread 4416 4 944
90 RootPressureThr 3215 5 1215
91 rcu_preempt 2954 0 0
92 ContainerMgr 2560 0 0
93 unnamed 1873 0 0
94 EventManager_De 1845 1 636
95 futex-default-S 1609 0 0
96
97INFO OPTIONS
98------------
99
100-t::
101--threads::
102 dump thread list in perf.data
103
104-m::
105--map::
106 dump map of lock instances (address:name table)
107
108CONTENTION OPTIONS
109--------------
110
111-k::
112--key=<value>::
113 Sorting key. Possible values: contended, wait_total (default),
114 wait_max, wait_min, avg_wait.
115
116-F::
117--field=<value>::
118 Output fields. By default it shows all but the wait_min fields
119 and users can customize that using this. Possible values:
120 contended, wait_total, wait_max, wait_min, avg_wait.
121
122-t::
123--threads::
124 Show per-thread lock contention stat
125
126-b::
127--use-bpf::
128 Use BPF program to collect lock contention stats instead of
129 using the input data.
130
131-a::
132--all-cpus::
133 System-wide collection from all CPUs.
134
135-C::
136--cpu::
137 Collect samples only on the list of CPUs provided. Multiple CPUs can be
138 provided as a comma-separated list with no space: 0,1. Ranges of CPUs
139 are specified with -: 0-2. Default is to monitor all CPUs.
140
141-p::
142--pid=::
143 Record events on existing process ID (comma separated list).
144
145--tid=::
146 Record events on existing thread ID (comma separated list).
147
148--map-nr-entries::
149 Maximum number of BPF map entries (default: 10240).
150
151
152SEE ALSO
153--------
154linkperf:perf[1]