1PERF-INJECT(1)                    perf Manual                   PERF-INJECT(1)
2
3
4

NAME

6       perf-inject - Filter to augment the events stream with additional
7       information
8

SYNOPSIS

10       perf inject <options>
11

DESCRIPTION

13       perf-inject reads a perf-record event stream and repipes it to stdout.
14       At any point the processing code can inject other events into the event
15       stream - in this case build-ids (-b option) are read and injected as
16       needed into the event stream.
17
18       Build-ids are just the first user of perf-inject - potentially anything
19       that needs userspace processing to augment the events stream with
20       additional information could make use of this facility.
21

OPTIONS

23       -b, --build-ids=
24           Inject build-ids into the output stream
25
26       -v, --verbose
27           Be more verbose.
28
29       -i, --input=
30           Input file name. (default: stdin)
31
32       -o, --output=
33           Output file name. (default: stdout)
34
35       -s, --sched-stat
36           Merge sched_stat and sched_switch for getting events where and how
37           long tasks slept. sched_switch contains a callchain where a task
38           slept and sched_stat contains a timeslice how long a task slept.
39
40       --kallsyms=<file>
41           kallsyms pathname
42
43       --itrace
44           Decode Instruction Tracing data, replacing it with synthesized
45           events. Options are:
46
47               i       synthesize instructions events
48               b       synthesize branches events
49               c       synthesize branches events (calls only)
50               r       synthesize branches events (returns only)
51               x       synthesize transactions events
52               w       synthesize ptwrite events
53               p       synthesize power events
54               e       synthesize error events
55               d       create a debug log
56               g       synthesize a call chain (use with i or x)
57               l       synthesize last branch entries (use with i or x)
58               s       skip initial number of events
59
60               The default is all events i.e. the same as --itrace=ibxwpe,
61               except for perf script where it is --itrace=ce
62
63               In addition, the period (default 100000, except for perf script where it is 1)
64               for instructions events can be specified in units of:
65
66               i       instructions
67               t       ticks
68               ms      milliseconds
69               us      microseconds
70               ns      nanoseconds (default)
71
72               Also the call chain size (default 16, max. 1024) for instructions or
73               transactions events can be specified.
74
75               Also the number of last branch entries (default 64, max. 1024) for
76               instructions or transactions events can be specified.
77
78               It is also possible to skip events generated (instructions, branches, transactions,
79               ptwrite, power) at the beginning. This is useful to ignore initialization code.
80
81               --itrace=i0nss1000000
82
83               skips the first million instructions.
84
85       --strip
86           Use with --itrace to strip out non-synthesized events.
87
88       -j, --jit
89           Process jitdump files by injecting the mmap records corresponding
90           to jitted functions. This option also generates the ELF images for
91           each jitted function found in the jitdumps files captured in the
92           input perf.data file. Use this option if you are monitoring
93           environment using JIT runtimes, such as Java, DART or V8.
94
95       -f, --force
96           Don’t complain, do it.
97

SEE ALSO

99       perf-record(1), perf-report(1), perf-archive(1)
100
101
102
103perf                              06/03/2019                    PERF-INJECT(1)