1perf-ftrace(1) 2============== 3 4NAME 5---- 6perf-ftrace - simple wrapper for kernel's ftrace functionality 7 8 9SYNOPSIS 10-------- 11[verse] 12'perf ftrace' {trace|latency|profile} <command> 13 14DESCRIPTION 15----------- 16The 'perf ftrace' command provides a collection of subcommands which use 17kernel's ftrace infrastructure. 18 19 'perf ftrace trace' is a simple wrapper of the ftrace. It only supports 20 single thread tracing currently and just reads trace_pipe in text and then 21 write it to stdout. 22 23 'perf ftrace latency' calculates execution latency of a given function 24 (optionally with BPF) and display it as a histogram. 25 26 'perf ftrace profile' show a execution profile for each function including 27 total, average, max time and the number of calls. 28 29The following options apply to perf ftrace. 30 31COMMON OPTIONS 32-------------- 33 34-p:: 35--pid=:: 36 Trace on existing process id (comma separated list). 37 38--tid=:: 39 Trace on existing thread id (comma separated list). 40 41-a:: 42--all-cpus:: 43 Force system-wide collection. Scripts run without a <command> 44 normally use -a by default, while scripts run with a <command> 45 normally don't - this option allows the latter to be run in 46 system-wide mode. 47 48-C:: 49--cpu=:: 50 Only trace for the list of CPUs provided. Multiple CPUs can 51 be provided as a comma separated list with no space like: 0,1. 52 Ranges of CPUs are specified with -: 0-2. 53 Default is to trace on all online CPUs. 54 55-v:: 56--verbose:: 57 Increase the verbosity level. 58 59 60OPTIONS for 'perf ftrace trace' 61------------------------------- 62 63-t:: 64--tracer=:: 65 Tracer to use when neither -G nor -F option is not 66 specified: function_graph or function. 67 68-F:: 69--funcs:: 70 List available functions to trace. It accepts a pattern to 71 only list interested functions. 72 73-D:: 74--delay:: 75 Time (ms) to wait before starting tracing after program start. 76 77-m:: 78--buffer-size:: 79 Set the size of per-cpu tracing buffer, <size> is expected to 80 be a number with appended unit character - B/K/M/G. 81 82--inherit:: 83 Trace children processes spawned by our target. 84 85-T:: 86--trace-funcs=:: 87 Select function tracer and set function filter on the given 88 function (or a glob pattern). Multiple functions can be given 89 by using this option more than once. The function argument also 90 can be a glob pattern. It will be passed to 'set_ftrace_filter' 91 in tracefs. 92 93-N:: 94--notrace-funcs=:: 95 Select function tracer and do not trace functions given by the 96 argument. Like -T option, this can be used more than once to 97 specify multiple functions (or glob patterns). It will be 98 passed to 'set_ftrace_notrace' in tracefs. 99 100--func-opts:: 101 List of options allowed to set: 102 103 - call-graph - Display kernel stack trace for function tracer. 104 - irq-info - Display irq context info for function tracer. 105 106-G:: 107--graph-funcs=:: 108 Select function_graph tracer and set graph filter on the given 109 function (or a glob pattern). This is useful to trace for 110 functions executed from the given function. This can be used more 111 than once to specify multiple functions. It will be passed to 112 'set_graph_function' in tracefs. 113 114-g:: 115--nograph-funcs=:: 116 Select function_graph tracer and set graph notrace filter on the 117 given function (or a glob pattern). Like -G option, this is useful 118 for the function_graph tracer only and disables tracing for function 119 executed from the given function. This can be used more than once to 120 specify multiple functions. It will be passed to 'set_graph_notrace' 121 in tracefs. 122 123--graph-opts:: 124 List of options allowed to set: 125 126 - args - Show function arguments. 127 - retval - Show function return value. 128 - retval-hex - Show function return value in hexadecimal format. 129 - retaddr - Show function return address. 130 - nosleep-time - Measure on-CPU time only for function_graph tracer. 131 - noirqs - Ignore functions that happen inside interrupt. 132 - verbose - Show process names, PIDs, timestamps, etc. 133 - thresh=<n> - Setup trace duration threshold in microseconds. 134 - depth=<n> - Set max depth for function graph tracer to follow. 135 - tail - Print function name at the end. 136 137 138OPTIONS for 'perf ftrace latency' 139--------------------------------- 140 141-T:: 142--trace-funcs=:: 143 Set the function name to get the histogram. Unlike perf ftrace trace, 144 it only allows single function to calculate the histogram. 145 146-e:: 147--events=:: 148 Set the pair of events to get the histogram. The histogram is calculated 149 by the time difference between the two events from the same thread. This 150 requires -b/--use-bpf option. 151 152-b:: 153--use-bpf:: 154 Use BPF to measure function latency instead of using the ftrace (it 155 uses function_graph tracer internally). 156 157-n:: 158--use-nsec:: 159 Use nano-second instead of micro-second as a base unit of the histogram. 160 161--bucket-range=:: 162 Bucket range in ms or ns (according to -n/--use-nsec), default is log2() mode. 163 164--min-latency=:: 165 Minimum latency for the start of the first bucket, in ms or ns (according to 166 -n/--use-nsec). 167 168--max-latency=:: 169 Maximum latency for the start of the last bucket, in ms or ns (according to 170 -n/--use-nsec). The setting is ignored if the value results in more than 171 22 buckets. 172 173OPTIONS for 'perf ftrace profile' 174--------------------------------- 175 176-T:: 177--trace-funcs=:: 178 Set function filter on the given function (or a glob pattern). 179 Multiple functions can be given by using this option more than once. 180 The function argument also can be a glob pattern. It will be passed 181 to 'set_ftrace_filter' in tracefs. 182 183-N:: 184--notrace-funcs=:: 185 Do not trace functions given by the argument. Like -T option, this 186 can be used more than once to specify multiple functions (or glob 187 patterns). It will be passed to 'set_ftrace_notrace' in tracefs. 188 189-G:: 190--graph-funcs=:: 191 Set graph filter on the given function (or a glob pattern). This is 192 useful to trace for functions executed from the given function. This 193 can be used more than once to specify multiple functions. It will be 194 passed to 'set_graph_function' in tracefs. 195 196-g:: 197--nograph-funcs=:: 198 Set graph notrace filter on the given function (or a glob pattern). 199 Like -G option, this is useful for the function_graph tracer only and 200 disables tracing for function executed from the given function. This 201 can be used more than once to specify multiple functions. It will be 202 passed to 'set_graph_notrace' in tracefs. 203 204-m:: 205--buffer-size:: 206 Set the size of per-cpu tracing buffer, <size> is expected to 207 be a number with appended unit character - B/K/M/G. 208 209-s:: 210--sort=:: 211 Sort the result by the given field. Available values are: 212 total, avg, max, count, name. Default is 'total'. 213 214--graph-opts:: 215 List of options allowed to set: 216 217 - nosleep-time - Measure on-CPU time only for function_graph tracer. 218 - noirqs - Ignore functions that happen inside interrupt. 219 - thresh=<n> - Setup trace duration threshold in microseconds. 220 - depth=<n> - Set max depth for function graph tracer to follow. 221 222 223SEE ALSO 224-------- 225linkperf:perf-record[1], linkperf:perf-trace[1] 226