bcc-hardirqs(8)

1hardirqs(8)                 System Manager's Manual                hardirqs(8)
2
3
4

NAME

6       hardirqs  -  Measure  hard  IRQ (hard interrupt) event time. Uses Linux
7       eBPF/bcc.
8

SYNOPSIS

10       hardirqs [-h] [-T] [-N] [-C] [-d] [interval] [outputs]
11

DESCRIPTION

13       This summarizes the time spent servicing hard IRQs  (hard  interrupts),
14       and  can  show this time as either totals or histogram distributions. A
15       system-wide summary of this  time  is  shown  by  the  %irq  column  of
16       mpstat(1),  and  event counts (but not times) are shown by /proc/inter‐
17       rupts.
18
19       WARNING: This currently uses dynamic tracing of  hard  interrupts.  You
20       should  understand  what  this means before use. Try in a test environ‐
21       ment. Future versions should switch to tracepoints.
22
23       Since this uses BPF, only the root user can use this tool.
24

REQUIREMENTS

26       CONFIG_BPF and bcc.
27

OPTIONS

29       -h     Print usage message.
30
31       -T     Include timestamps on output.
32
33       -N     Output in nanoseconds.
34
35       -C     Count events only.
36
37       -d     Show IRQ time distribution as histograms.
38

EXAMPLES

40       Sum hard IRQ event time until Ctrl-C:
41              # hardirqs
42
43       Show hard IRQ event time as histograms:
44              # hardirqs -d
45
46       Print 1 second summaries, 10 times:
47              # hardirqs 1 10
48
49       1 second summaries, printed in nanoseconds, with timestamps:
50              # hardirqs -NT 1
51

FIELDS

53       HARDIRQ
54              The irq action name for this hard IRQ.
55
56       TOTAL_usecs
57              Total time spent in this hard IRQ in microseconds.
58
59       TOTAL_nsecs
60              Total time spent in this hard IRQ in nanoseconds.
61
62       usecs  Range of microseconds for this bucket.
63
64       nsecs  Range of nanoseconds for this bucket.
65
66       count  Number of hard IRQs in this time range.
67
68       distribution
69              ASCII representation of the distribution (the count column).
70

OVERHEAD

72       This traces kernel functions and maintains in-kernel counts, which  are
73       asynchronously  copied  to  user-space. While the rate of interrupts be
74       very high (>1M/sec), this is a relatively efficient way to trace  these
75       events,  and  so  the overhead is expected to be small for normal work‐
76       loads, but could become noticeable for heavy workloads.  Measure  in  a
77       test environment before use.
78

SOURCE

80       This is from bcc.
81
82              https://github.com/iovisor/bcc
83
84       Also  look  in  the bcc distribution for a companion _examples.txt file
85       containing example usage, output, and commentary for this tool.
86

OS

88       Linux
89

STABILITY

91       Unstable - in development.
92

AUTHOR

94       Brendan Gregg
95