6 kvm_stat - Report KVM kernel module event counters
11 'kvm_stat' [OPTION]...
15 kvm_stat prints counts of KVM kernel module trace events. These events signify
16 state transitions such as guest mode entry and exit.
18 This tool is useful for observing guest behavior from the host perspective.
19 Often conclusions about performance or buggy behavior can be drawn from the
21 While running in regular mode, use any of the keys listed in section
22 'Interactive Commands' below.
23 Use batch and logging modes for scripting purposes.
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 kernel module source code for the
32 *b*:: toggle events by guests (debugfs only, honors filters)
36 *f*:: filter by regular expression
37 :: *Note*: Child events pull in their parents, and parents' stats summarize
38 all child events, not just the filtered ones
40 *g*:: filter by guest name/PID
42 *h*:: display interactive commands reference
44 *o*:: toggle sorting order (Total vs CurAvg/s)
46 *p*:: filter by guest name/PID
52 *s*:: set delay between refreshs
54 *x*:: toggle reporting of stats for child trace events
55 :: *Note*: The stats for the parents summarize the respective child trace
58 Press any other key to refresh statistics immediately.
65 run in batch mode for one second
69 log in csv format. Requires option -l/--log or -L/--log-to-file.
70 When used with option -L/--log-to-file, the header is only ever
71 written to start of file to preserve the format.
75 retrieve statistics from debugfs
79 fields to display (regex), "-f help" for a list of available events
82 --guest=<guest_name>::
83 limit statistics to one virtual machine (guest name)
90 --debugfs-include-past::
91 include all available data on past events for debugfs
95 run in logging mode (like vmstat)
99 --log-to-file=<file>::
100 like -l/--log, but logging to a file. Appends to existing files.
104 limit statistics to one virtual machine (pid)
108 set delay between refreshs (value range: 0.1-25.5 secs)
112 retrieve statistics from tracepoints
115 --skip-zero-records::
116 omit records with all zeros in logging mode
120 'perf'(1), 'trace-cmd'(1)
124 Stefan Hajnoczi <stefanha@redhat.com>