1 perf-timechart(1) 1 perf-timechart(1)
2 ================= 2 =================
3 3
4 NAME 4 NAME
5 ---- 5 ----
6 perf-timechart - Tool to visualize total syste 6 perf-timechart - Tool to visualize total system behavior during a workload
7 7
8 SYNOPSIS 8 SYNOPSIS
9 -------- 9 --------
10 [verse] 10 [verse]
11 'perf timechart' [<timechart options>] {record 11 'perf timechart' [<timechart options>] {record} [<record options>]
12 12
13 DESCRIPTION 13 DESCRIPTION
14 ----------- 14 -----------
15 There are two variants of perf timechart: 15 There are two variants of perf timechart:
16 16
17 'perf timechart record <command>' to record 17 'perf timechart record <command>' to record the system level events
18 of an arbitrary workload. By default timecha 18 of an arbitrary workload. By default timechart records only scheduler
19 and CPU events (task switches, running times 19 and CPU events (task switches, running times, CPU power states, etc),
20 but it's possible to record IO (disk, networ 20 but it's possible to record IO (disk, network) activity using -I argument.
21 21
22 'perf timechart' to turn a trace into a Scal 22 'perf timechart' to turn a trace into a Scalable Vector Graphics file,
23 that can be viewed with popular SVG viewers 23 that can be viewed with popular SVG viewers such as 'Inkscape'. Depending
24 on the events in the perf.data file, timecha 24 on the events in the perf.data file, timechart will contain scheduler/cpu
25 events or IO events. 25 events or IO events.
26 26
27 In IO mode, every bar has two charts: upper 27 In IO mode, every bar has two charts: upper and lower.
28 Upper bar shows incoming events (disk reads, 28 Upper bar shows incoming events (disk reads, ingress network packets).
29 Lower bar shows outgoing events (disk writes 29 Lower bar shows outgoing events (disk writes, egress network packets).
30 There are also poll bars which show how much 30 There are also poll bars which show how much time application spent
31 in poll/epoll/select syscalls. 31 in poll/epoll/select syscalls.
32 32
33 TIMECHART OPTIONS 33 TIMECHART OPTIONS
34 ----------------- 34 -----------------
35 -o:: 35 -o::
36 --output=:: 36 --output=::
37 Select the output file (default: outpu 37 Select the output file (default: output.svg)
38 -i:: 38 -i::
39 --input=:: 39 --input=::
40 Select the input file (default: perf.d 40 Select the input file (default: perf.data unless stdin is a fifo)
41 -w:: 41 -w::
42 --width=:: 42 --width=::
43 Select the width of the SVG file (defa 43 Select the width of the SVG file (default: 1000)
44 -P:: 44 -P::
45 --power-only:: 45 --power-only::
46 Only output the CPU power section of t 46 Only output the CPU power section of the diagram
47 -T:: 47 -T::
48 --tasks-only:: 48 --tasks-only::
49 Don't output processor state transitio 49 Don't output processor state transitions
50 -p:: 50 -p::
51 --process:: 51 --process::
52 Select the processes to display, by na 52 Select the processes to display, by name or PID
53 -f:: 53 -f::
54 --force:: 54 --force::
55 Don't complain, do it. 55 Don't complain, do it.
56 --symfs=<directory>:: 56 --symfs=<directory>::
57 Look for files with symbols relative t 57 Look for files with symbols relative to this directory.
58 -n:: 58 -n::
59 --proc-num:: 59 --proc-num::
60 Print task info for at least given num 60 Print task info for at least given number of tasks.
61 -t:: 61 -t::
62 --topology:: 62 --topology::
63 Sort CPUs according to topology. 63 Sort CPUs according to topology.
64 --highlight=<duration_nsecs|task_name>:: 64 --highlight=<duration_nsecs|task_name>::
65 Highlight tasks (using different color 65 Highlight tasks (using different color) that run more than given
66 duration or tasks with given name. If 66 duration or tasks with given name. If number is given it's interpreted
67 as number of nanoseconds. If non-numer 67 as number of nanoseconds. If non-numeric string is given it's
68 interpreted as task name. 68 interpreted as task name.
69 --io-skip-eagain:: 69 --io-skip-eagain::
70 Don't draw EAGAIN IO events. 70 Don't draw EAGAIN IO events.
71 --io-min-time=<nsecs>:: 71 --io-min-time=<nsecs>::
72 Draw small events as if they lasted mi 72 Draw small events as if they lasted min-time. Useful when you need
73 to see very small and fast IO. It's po 73 to see very small and fast IO. It's possible to specify ms or us
74 suffix to specify time in milliseconds 74 suffix to specify time in milliseconds or microseconds.
75 Default value is 1ms. 75 Default value is 1ms.
76 --io-merge-dist=<nsecs>:: 76 --io-merge-dist=<nsecs>::
77 Merge events that are merge-dist nanos 77 Merge events that are merge-dist nanoseconds apart.
78 Reduces number of figures on the SVG a 78 Reduces number of figures on the SVG and makes it more render-friendly.
79 It's possible to specify ms or us suff 79 It's possible to specify ms or us suffix to specify time in
80 milliseconds or microseconds. 80 milliseconds or microseconds.
81 Default value is 1us. 81 Default value is 1us.
82 82
83 RECORD OPTIONS 83 RECORD OPTIONS
84 -------------- 84 --------------
85 -P:: 85 -P::
86 --power-only:: 86 --power-only::
87 Record only power-related events 87 Record only power-related events
88 -T:: 88 -T::
89 --tasks-only:: 89 --tasks-only::
90 Record only tasks-related events 90 Record only tasks-related events
91 -I:: 91 -I::
92 --io-only:: 92 --io-only::
93 Record only io-related events 93 Record only io-related events
94 -g:: 94 -g::
95 --callchain:: 95 --callchain::
96 Do call-graph (stack chain/backtrace) 96 Do call-graph (stack chain/backtrace) recording
97 97
98 EXAMPLES 98 EXAMPLES
99 -------- 99 --------
100 100
101 $ perf timechart record git pull 101 $ perf timechart record git pull
102 102
103 [ perf record: Woken up 13 times to write da 103 [ perf record: Woken up 13 times to write data ]
104 [ perf record: Captured and wrote 4.253 MB p 104 [ perf record: Captured and wrote 4.253 MB perf.data (~185801 samples) ]
105 105
106 $ perf timechart 106 $ perf timechart
107 107
108 Written 10.2 seconds of trace to output.svg. 108 Written 10.2 seconds of trace to output.svg.
109 109
110 Record system-wide timechart: 110 Record system-wide timechart:
111 111
112 $ perf timechart record 112 $ perf timechart record
113 113
114 then generate timechart and highlight 'gcc' 114 then generate timechart and highlight 'gcc' tasks:
115 115
116 $ perf timechart --highlight gcc 116 $ perf timechart --highlight gcc
117 117
118 Record system-wide IO events: 118 Record system-wide IO events:
119 119
120 $ perf timechart record -I 120 $ perf timechart record -I
121 121
122 then generate timechart: 122 then generate timechart:
123 123
124 $ perf timechart 124 $ perf timechart
125 125
126 SEE ALSO 126 SEE ALSO
127 -------- 127 --------
128 linkperf:perf-record[1] 128 linkperf:perf-record[1]
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.