1 kvm_stat(1) 1 kvm_stat(1)
2 =========== 2 ===========
3 3
4 NAME 4 NAME
5 ---- 5 ----
6 kvm_stat - Report KVM kernel module event coun 6 kvm_stat - Report KVM kernel module event counters
7 7
8 SYNOPSIS 8 SYNOPSIS
9 -------- 9 --------
10 [verse] 10 [verse]
11 'kvm_stat' [OPTION]... 11 'kvm_stat' [OPTION]...
12 12
13 DESCRIPTION 13 DESCRIPTION
14 ----------- 14 -----------
15 kvm_stat prints counts of KVM kernel module tr 15 kvm_stat prints counts of KVM kernel module trace events. These events signify
16 state transitions such as guest mode entry and 16 state transitions such as guest mode entry and exit.
17 17
18 This tool is useful for observing guest behavi 18 This tool is useful for observing guest behavior from the host perspective.
19 Often conclusions about performance or buggy b 19 Often conclusions about performance or buggy behavior can be drawn from the
20 output. 20 output.
21 While running in regular mode, use any of the 21 While running in regular mode, use any of the keys listed in section
22 'Interactive Commands' below. 22 'Interactive Commands' below.
23 Use batch and logging modes for scripting purp 23 Use batch and logging modes for scripting purposes.
24 24
25 The set of KVM kernel module trace events may 25 The set of KVM kernel module trace events may be specific to the kernel version
26 or architecture. It is best to check the KVM 26 or architecture. It is best to check the KVM kernel module source code for the
27 meaning of events. 27 meaning of events.
28 28
29 INTERACTIVE COMMANDS 29 INTERACTIVE COMMANDS
30 -------------------- 30 --------------------
31 [horizontal] 31 [horizontal]
32 *b*:: toggle events by guests (debugfs only, 32 *b*:: toggle events by guests (debugfs only, honors filters)
33 33
34 *c*:: clear filter 34 *c*:: clear filter
35 35
36 *f*:: filter by regular expression 36 *f*:: filter by regular expression
37 :: *Note*: Child events pull in their par 37 :: *Note*: Child events pull in their parents, and parents' stats summarize
38 all child events, not just the 38 all child events, not just the filtered ones
39 39
40 *g*:: filter by guest name/PID 40 *g*:: filter by guest name/PID
41 41
42 *h*:: display interactive commands reference 42 *h*:: display interactive commands reference
43 43
44 *o*:: toggle sorting order (Total vs CurAvg/ 44 *o*:: toggle sorting order (Total vs CurAvg/s)
45 45
46 *p*:: filter by guest name/PID 46 *p*:: filter by guest name/PID
47 47
48 *q*:: quit 48 *q*:: quit
49 49
50 *r*:: reset stats 50 *r*:: reset stats
51 51
52 *s*:: set delay between refreshs !! 52 *s*:: set update interval
53 53
54 *x*:: toggle reporting of stats for child tr 54 *x*:: toggle reporting of stats for child trace events
55 :: *Note*: The stats for the parents summ 55 :: *Note*: The stats for the parents summarize the respective child trace
56 events 56 events
57 57
58 Press any other key to refresh statistics imme 58 Press any other key to refresh statistics immediately.
59 59
60 OPTIONS 60 OPTIONS
61 ------- 61 -------
62 -1:: 62 -1::
63 --once:: 63 --once::
64 --batch:: 64 --batch::
65 run in batch mode for one second 65 run in batch mode for one second
66 66
67 -c:: !! 67 -l::
68 --csv:: !! 68 --log::
69 log in csv format. Requires option -l/ !! 69 run in logging mode (like vmstat)
70 When used with option -L/--log-to-file !! 70
71 written to start of file to preserve t !! 71 -t::
>> 72 --tracepoints::
>> 73 retrieve statistics from tracepoints
72 74
73 -d:: 75 -d::
74 --debugfs:: 76 --debugfs::
75 retrieve statistics from debugfs 77 retrieve statistics from debugfs
76 78
77 -f<fields>:: <<
78 --fields=<fields>:: <<
79 fields to display (regex), "-f help" f <<
80 <<
81 -g<guest>:: <<
82 --guest=<guest_name>:: <<
83 limit statistics to one virtual machin <<
84 <<
85 -h:: <<
86 --help:: <<
87 show help message <<
88 <<
89 -i:: 79 -i::
90 --debugfs-include-past:: 80 --debugfs-include-past::
91 include all available data on past eve 81 include all available data on past events for debugfs
92 82
93 -l:: <<
94 --log:: <<
95 run in logging mode (like vmstat) <<
96 <<
97 <<
98 -L<file>:: <<
99 --log-to-file=<file>:: <<
100 like -l/--log, but logging to a file. <<
101 <<
102 -p<pid>:: 83 -p<pid>::
103 --pid=<pid>:: 84 --pid=<pid>::
104 limit statistics to one virtual machin 85 limit statistics to one virtual machine (pid)
105 86
106 -s:: !! 87 -g<guest>::
107 --set-delay:: !! 88 --guest=<guest_name>::
108 set delay between refreshs (value rang !! 89 limit statistics to one virtual machine (guest name)
109 90
110 -t:: !! 91 -f<fields>::
111 --tracepoints:: !! 92 --fields=<fields>::
112 retrieve statistics from tracepoints !! 93 fields to display (regex), "-f help" for a list of available events
113 94
114 -z:: !! 95 -h::
115 --skip-zero-records:: !! 96 --help::
116 omit records with all zeros in logging !! 97 show help message
117 98
118 SEE ALSO 99 SEE ALSO
119 -------- 100 --------
120 'perf'(1), 'trace-cmd'(1) 101 'perf'(1), 'trace-cmd'(1)
121 102
122 AUTHOR 103 AUTHOR
123 ------ 104 ------
124 Stefan Hajnoczi <stefanha@redhat.com> 105 Stefan Hajnoczi <stefanha@redhat.com>
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.