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       --buildid-all: Inject build-ids of all DSOs into the output stream
27
28       -v, --verbose
29           Be more verbose.
30
31       -i, --input=
32           Input file name. (default: stdin)
33
34       -o, --output=
35           Output file name. (default: stdout)
36
37       -s, --sched-stat
38           Merge sched_stat and sched_switch for getting events where and how
39           long tasks slept. sched_switch contains a callchain where a task
40           slept and sched_stat contains a timeslice how long a task slept.
41
42       --kallsyms=<file>
43           kallsyms pathname
44
45       --itrace
46           Decode Instruction Tracing data, replacing it with synthesized
47           events. Options are:
48
49               i       synthesize instructions events
50               b       synthesize branches events (branch misses for Arm SPE)
51               c       synthesize branches events (calls only)
52               r       synthesize branches events (returns only)
53               x       synthesize transactions events
54               w       synthesize ptwrite events
55               p       synthesize power events
56               o       synthesize other events recorded due to the use
57                       of aux-output (refer to perf record)
58               e       synthesize error events
59               d       create a debug log
60               f       synthesize first level cache events
61               m       synthesize last level cache events
62               M       synthesize memory events
63               t       synthesize TLB events
64               a       synthesize remote access events
65               g       synthesize a call chain (use with i or x)
66               G       synthesize a call chain on existing event records
67               l       synthesize last branch entries (use with i or x)
68               L       synthesize last branch entries on existing event records
69               s       skip initial number of events
70               q       quicker (less detailed) decoding
71
72               The default is all events i.e. the same as --itrace=ibxwpe,
73               except for perf script where it is --itrace=ce
74
75               In addition, the period (default 100000, except for perf script where it is 1)
76               for instructions events can be specified in units of:
77
78               i       instructions
79               t       ticks
80               ms      milliseconds
81               us      microseconds
82               ns      nanoseconds (default)
83
84               Also the call chain size (default 16, max. 1024) for instructions or
85               transactions events can be specified.
86
87               Also the number of last branch entries (default 64, max. 1024) for
88               instructions or transactions events can be specified.
89
90               Similar to options g and l, size may also be specified for options G and L.
91               On x86, note that G and L work poorly when data has been recorded with
92               large PEBS. Refer linkperf:perf-intel-pt[1] man page for details.
93
94               It is also possible to skip events generated (instructions, branches, transactions,
95               ptwrite, power) at the beginning. This is useful to ignore initialization code.
96
97               --itrace=i0nss1000000
98
99               skips the first million instructions.
100
101               The 'e' option may be followed by flags which affect what errors will or
102               will not be reported. Each flag must be preceded by either '+' or '-'.
103               The flags are:
104                       o       overflow
105                       l       trace data lost
106
107               If supported, the 'd' option may be followed by flags which affect what
108               debug messages will or will not be logged. Each flag must be preceded
109               by either '+' or '-'. The flags are:
110                       a       all perf events
111
112               If supported, the 'q' option may be repeated to increase the effect.
113
114       --strip
115           Use with --itrace to strip out non-synthesized events.
116
117       -j, --jit
118           Process jitdump files by injecting the mmap records corresponding
119           to jitted functions. This option also generates the ELF images for
120           each jitted function found in the jitdumps files captured in the
121           input perf.data file. Use this option if you are monitoring
122           environment using JIT runtimes, such as Java, DART or V8.
123
124       -f, --force
125           Don’t complain, do it.
126

SEE ALSO

128       perf-record(1), perf-report(1), perf-archive(1), perf-intel-pt(1)
129
130
131
132perf                              03/30/2021                    PERF-INJECT(1)