1perf-timechart(1) 2================= 3 4NAME 5---- 6perf-timechart - Tool to visualize total system behavior during a workload 7 8SYNOPSIS 9-------- 10[verse] 11'perf timechart' [<timechart options>] {record} [<record options>] 12 13DESCRIPTION 14----------- 15There are two variants of perf timechart: 16 17 'perf timechart record <command>' to record the system level events 18 of an arbitrary workload. By default timechart records only scheduler 19 and CPU events (task switches, running times, CPU power states, etc), 20 but it's possible to record IO (disk, network) activity using -I argument. 21 22 'perf timechart' to turn a trace into a Scalable Vector Graphics file, 23 that can be viewed with popular SVG viewers such as 'Inkscape'. Depending 24 on the events in the perf.data file, timechart will contain scheduler/cpu 25 events or IO events. 26 27 In IO mode, every bar has two charts: upper and lower. 28 Upper bar shows incoming events (disk reads, ingress network packets). 29 Lower bar shows outgoing events (disk writes, egress network packets). 30 There are also poll bars which show how much time application spent 31 in poll/epoll/select syscalls. 32 33TIMECHART OPTIONS 34----------------- 35-o:: 36--output=:: 37 Select the output file (default: output.svg) 38-i:: 39--input=:: 40 Select the input file (default: perf.data unless stdin is a fifo) 41-w:: 42--width=:: 43 Select the width of the SVG file (default: 1000) 44-P:: 45--power-only:: 46 Only output the CPU power section of the diagram 47-T:: 48--tasks-only:: 49 Don't output processor state transitions 50-p:: 51--process:: 52 Select the processes to display, by name or PID 53-f:: 54--force:: 55 Don't complain, do it. 56--symfs=<directory[,layout]>:: 57 Look for files with symbols relative to this directory. The optional 58 layout can be 'hierarchy' (default, matches full path) or 'flat' 59 (only matches base name). This is useful when debug files are stored 60 in a flat directory structure. 61-n:: 62--proc-num:: 63 Print task info for at least given number of tasks. 64-t:: 65--topology:: 66 Sort CPUs according to topology. 67--highlight=<duration_nsecs|task_name>:: 68 Highlight tasks (using different color) that run more than given 69 duration or tasks with given name. If number is given it's interpreted 70 as number of nanoseconds. If non-numeric string is given it's 71 interpreted as task name. 72--io-skip-eagain:: 73 Don't draw EAGAIN IO events. 74--io-min-time=<nsecs>:: 75 Draw small events as if they lasted min-time. Useful when you need 76 to see very small and fast IO. It's possible to specify ms or us 77 suffix to specify time in milliseconds or microseconds. 78 Default value is 1ms. 79--io-merge-dist=<nsecs>:: 80 Merge events that are merge-dist nanoseconds apart. 81 Reduces number of figures on the SVG and makes it more render-friendly. 82 It's possible to specify ms or us suffix to specify time in 83 milliseconds or microseconds. 84 Default value is 1us. 85 86RECORD OPTIONS 87-------------- 88-P:: 89--power-only:: 90 Record only power-related events 91-T:: 92--tasks-only:: 93 Record only tasks-related events 94-I:: 95--io-only:: 96 Record only io-related events 97-g:: 98--callchain:: 99 Do call-graph (stack chain/backtrace) recording 100-o:: 101--output=:: 102 Select the output file (default: perf.data) 103 104EXAMPLES 105-------- 106 107$ perf timechart record git pull 108 109 [ perf record: Woken up 13 times to write data ] 110 [ perf record: Captured and wrote 4.253 MB perf.data (~185801 samples) ] 111 112$ perf timechart 113 114 Written 10.2 seconds of trace to output.svg. 115 116Record system-wide timechart: 117 118 $ perf timechart record 119 120 then generate timechart and highlight 'gcc' tasks: 121 122 $ perf timechart --highlight gcc 123 124Record system-wide IO events: 125 126 $ perf timechart record -I 127 128 then generate timechart: 129 130 $ perf timechart 131 132SEE ALSO 133-------- 134linkperf:perf-record[1] 135
