RTMON-DUMP(1) RTMON-DUMP(1)
rtmon-dump - dump the contents of events collect from rtmond
rtmon-dump [ options ] [ file ]
rtmon-dump prints binary event data collected from an rtmond(1) server
process using rtmon-client(1) or padc(1). Data can be read from a file
or from the standard input. By default, rtmon-dump displays all
interesting events found in a file using the same format as par(1).
Various options are provided for filtering which events are printed and
how they are displayed.
-d usecs
Use usecs for the threshold for displaying system calls as begin and
end entries. By default, rtmon-dump will display system call events
as single line if there are no events to display between the start
and finish of the system call and if the duration of the system is
less than a nominal threshold (default 2 milliseconds). If a system
call takes longer than this threshold than it is displayed as
separate begin and end events so that it is easy to pick out longrunning
calls.
-f Follow the growth of event data collected in a file. If a file is
specified on the command line and the -f option is specified, then
rtmon-dump will dump the contents of the file and then periodically
monitor it for new data. In this situation rtmon-dump will never
exit voluntarily; it must be terminated with an interrupt signal
(e.g. from a keyboard interrupt).
-i Display ``internal'' event as well as normal events. This option
causes the display of certain events that are normally used only in
decoding other events: CPU configuration, task names, process
execs, etc.
-m event-mask
Display only events specified by event-mask. An event mask
specifies a set of events; it is a set of event classes with each
class specified symbolically as one of the following:
Name Description
all All events (default)
alloc Memory allocation
disk Disk i/o work
intr Hardware interrupts
io I/O-related events (disk+intr)
netflow Network I/O flow
netsched Network I/O scheduling
network Network-related events (netflow+netsched)
none No events
Page 1
RTMON-DUMP(1) RTMON-DUMP(1)
profile Kernel profiling
scheduler Process and thread scheduler
signal Signal delivery and reception
syscall System calls and their arguments
task Process and thread scheduling
taskproc Process scheduling (system threads are not included)
vm Virtual memory operation
Event class names are case insensitive; i.e. ``SIGNAL'' is
interpreted the same as ``signal''. Multiple event classes may be
included by using a ``+'', ``|'', or ``,'' symbol to separate the
names. Event classes may be excluded by using a ``-'' to separate
the name. For example, ``network+io-disk'' indicates all network
and i/o events should be included except for disk-related events.
In addition to the above names, a number may be used to specify a
value, where the various events are selected by bits in the value,
as defined in ``<sys/rtmon.h>''.
-O Print unrecognized event data in octal instead of decimal.
-x Print unrecognized event data in hexadecimal instead of decimal.
-p cpu-list
Display only events from certain CPUs. CPUs are specified as a
comma-separated list of CPU numbers or range of CPU numbers. CPUs
are numbered starting at zero. For example, -p 1,4-7,11 means
processors 1, 4, 5, 6, 7, and 11. An open-ended range is
interpreted as all CPUs to the maximum CPU number; e.g. 2- on an
8-processor system is the same as 2-7. The CPU list must be one
string with no white space. By default, rtmon-dump displays events
from all CPUs.
-r Display event decoding; display all events as raw data.
-u Display event times as milliseconds with a microsecond fraction and
show the number of microseconds between displayed events.
-v Enable verbose output.
padc(1), par(1), rtmond(1), rtmon_log_user_tstamp(3)
PPPPaaaaggggeeee 2222 [ Back ]
|
