6 perf-timechart - Tool to visualize total system behavior during a workload
11 'perf timechart' [<timechart options>] {record} [<record options>]
15 There are two variants of perf timechart:
17 'perf timechart record <command>' to record the system level events
18 of an arbitrary workload.
20 'perf timechart' to turn a trace into a Scalable Vector Graphics file,
21 that can be viewed with popular SVG viewers such as 'Inkscape'.
27 Select the output file (default: output.svg)
30 Select the input file (default: perf.data unless stdin is a fifo)
33 Select the width of the SVG file (default: 1000)
36 Only output the CPU power section of the diagram
39 Don't output processor state transitions
42 Select the processes to display, by name or PID
45 Look for files with symbols relative to this directory.
48 Print task info for at least given number of tasks.
51 Sort CPUs according to topology.
52 --highlight=<duration_nsecs|task_name>::
53 Highlight tasks (using different color) that run more than given
54 duration or tasks with given name. If number is given it's interpreted
55 as number of nanoseconds. If non-numeric string is given it's
56 interpreted as task name.
62 Record only power-related events
65 Record only tasks-related events
68 Do call-graph (stack chain/backtrace) recording
73 $ perf timechart record git pull
75 [ perf record: Woken up 13 times to write data ]
76 [ perf record: Captured and wrote 4.253 MB perf.data (~185801 samples) ]
80 Written 10.2 seconds of trace to output.svg.
82 Record system-wide timechart:
84 $ perf timechart record
86 then generate timechart and highlight 'gcc' tasks:
88 $ perf timechart --highlight gcc
92 linkperf:perf-record[1]