--help / -?
Show help message.
--version / -v
Show version.
--list-events / -l
Shows the monitorable events.
--init Load the OProfile module if required and make the OProfile
driver interface available.
--setup
Followed by list options for profiling setup. Store setup
in ~root/.oprofile/daemonrc. Optional.
--status
Show configuration information.
--start-daemon
Start the oprofile daemon without starting profiling.
--start / -s
Start data collection with either arguments provided by
--setup or with information saved in
~root/.oprofile/daemonrc.
--dump / -d
Force a flush of the collected profiling data to the
daemon.
--stop / -t
Stop data collection.
--shutdown / -h
Stop data collection and kill the daemon.
--reset
Clear out data from current session, but leaves saved
sessions.
--save=sessionname
Save data from current session to sessionname.
--deinit
Shut down daemon. Unload the oprofile module and
oprofilefs.
--session-dir=dir_path
Use sample database out of directory dir_path instead of
the default location (/var/lib/oprofile).
--buffer-size=num
Set kernel buffer to num samples. The buffer watershed
needs to be tweaked when changing this value. Rules: A
non-zero value goes into effect after a '--shutdown/start'
sequence. A value of zero sets this parameter back to
default value, but does not go into effect until after
'--deinit/init' sequence.
--buffer-watershed=num
Set kernel buffer watershed to num samples. When buffer-
size - buffer-watershed free entries remain in the kernel
buffer, data will be flushed to the daemon. Most useful
values are in the range [0.25 - 0.5] * buffer-size. Same
rules as defined for buffer-size.
--cpu-buffer-size=num
Set kernel per-cpu buffer to num samples. If you profile
at high rate it can help to increase this if the log file
show excessive count of sample lost cpu buffer overflow.
Same rules as defined for buffer-size.
--event / -e [event|"default"]
Specify an event to measure for the hardware performance
counters, or "default" for the default event. The event is
of the form "CPU_CLK_UNHALTED:30000:0:1:1" where the
numeric values are count, unit mask, kernel-space
counting, user-space counting, respectively. Note that
this over-rides all previous events selected; if you want
to profile with two or more events simultaneously, you
must specify them on the same opcontrol invocation. You
can specify unit mask values using either a numerical
value (hex values must begin with "0x") or a symbolic name
(if the name=<um_name> field is shown in the ophelp
output). For some named unit masks, the hex value is not
unique; thus, OProfile tools enforce specifying such unit
masks value by name.
--separate / -p [none,lib,kernel,thread,cpu,all]
Separate samples based on the given separator. 'lib'
separates dynamically linked library samples per
application. 'kernel' separates kernel and kernel module
samples per application; 'kernel' implies 'library'.
'thread' gives separation for each thread and task. 'cpu'
separates for each CPU. 'all' implies all of the above
options and 'none' turns off separation.
--callgraph / -c [#depth]
Enable callgraph sample collection with a maximum depth.
Use 0 to disable callgraph profiling. This option is
available on x86 using a 2.6+ kernel with callgraph
support enabled. It is also available on PowerPC using a
2.6.17+ kernel.
--image / -i [name,name...|"all"]
Only profile the given absolute paths to binaries, or
"all" to profile everything (the default).
--vmlinux=file
vmlinux kernel image.
--no-vmlinux
Use this when you don't have a kernel vmlinux file, and
you don't want to profile the kernel.
--verbose / -V [options]
Be verbose in the daemon log. This has a high overhead.
--kernel-range=start,end
Set kernel range vma address in hexadecimal.
