latency(1) [debian man page]

LATENCY(1)							      Xenomai								LATENCY(1)

NAME

       latency - Xenomai timer latency benchmark

SYNOPSIS

       latency [options]

DESCRIPTION

       latency	is part of the Xenomai test suite. It is a timer latency benchmark program.  The system must run a suitable Xenomai enabled kernel
       with the respective module (xeno_timerbench).

OPTIONS

       latency accepts the following options:

       -h     print histograms of min, avg, max latencies

       -s     print statistics of min, avg, max latencies

       -H <histogram-size>
	      default = 200, increase if your last bucket is full

       -B <bucket-size>
	      default = 1000ns, decrease for more resolution

       -p <period_us>
	      sampling period

       -l <data-lines per header>
	      default=21, 0 to supress headers

       -T <test_duration_seconds>
	      default=0, so ^C to end

       -q     supresses RTD, RTH lines if -T is used

       -D <testing_device_no>
	      number of testing device, default=0

       -t <test_mode>
	      0=user task (default), 1=kernel task, 2=timer IRQ

       -f     freeze trace for each new max latency

       -c <cpu>
	      pin measuring task down to given CPU

       -P <priority>
	      task priority (test mode 0 and 1 only)

AUTHOR

       latency was written by Philippe Gerum <rpm@xenomai.org>. This man page was written by Roland Stigge <stigge@antcom.de>.

2.6.0								    2008-04-19								LATENCY(1)

LATENCY(1)						    BSD General Commands Manual 						LATENCY(1)

NAME

     latency -- monitors scheduling and interrupt latency

SYNOPSIS

     latency [-rt] [-c codefile] [-l logfile] [-st threshold] [-it threshold] [-s sleep_in_usecs] [-d decrementer_in_usecs] [-n kernel]

DESCRIPTION

     The latency utility provides scheduling and interrupt latency statistics.	It requires root privileges due to the kernel tracing facility it
     uses to operate.

     The arguments are as follows:

     -rt     Set real time scheduling policy.  Default policy is timeshare.

     -c      When the -c option is specified, it takes a path to a codefile that contains the mappings for the system calls.  This option over-
	     rides the default location of the system call codefile which is found in /usr/share/misc/trace.codes.

     -l      Specifies a logfile that is written to when either the interrupt or scheduling latency is exceeded.

     -st     Set the scheduler latency threshold in microseconds.  If latency exceeds this, and a logfile has been specified, a record of what
	     occurred during this time is recorded.

     -it     Set the interrupt latency threshold in microseconds.  If latency exceeds this, and a logfile has been specified, a record of what
	     occurred during this time is recorded.

     -s      The -s option sets the timer.  It takes microseconds as an argument, the default timer is set to 1000 microseconds.

     -d      The -d option sets the decrementer.  It takes microseconds as an argument.  The decrementer is set back to the system default on
	     exit.

     -n      By default, latency acts on the default /mach_kernel.  This option allows you to specify an alternate booted kernel.

     The data columns displayed are as follows:

     SCHEDULER		       The number of context switches that fall within the described delay.
     INTERRUPTS 	       The number of interrupts that fall within the described delay.

     The latency utility is also SIGWINCH savvy, so adjusting your window geometry will change the list of delay values displayed.

SAMPLE USAGE

     latency -rt -st 20000 -it 1000 -l /var/tmp/latency.log

     The latency utility will set the realtime scheduling policy.  The threshold for the scheduler is set to 20000 microseconds.  The threshold
     for interrupts is set to 1000 microseconds.  Latencies that exceed these thresholds will be logged in /var/tmp/latency.log.

SEE ALSO

     top(1), fs_usage(1), sc_usage(1)

Mac OS X							  March 28, 2000							  Mac OS X