perf-sched(1) [centos man page]

PERF-SCHED(1)							    perf Manual 						     PERF-SCHED(1)

NAME

       perf-sched - Tool to trace/measure scheduler properties (latencies)

SYNOPSIS

       perf sched {record|latency|map|replay|script}

DESCRIPTION

       There are five variants of perf sched:

	   'perf sched record <command>' to record the scheduling events
	   of an arbitrary workload.

	   'perf sched latency' to report the per task scheduling latencies
	   and other scheduling properties of the workload.

	   'perf sched script' to see a detailed trace of the workload that
	    was recorded (aliased to 'perf script' for now).

	   'perf sched replay' to simulate the workload that was recorded
	   via perf sched record. (this is done by starting up mockup threads
	   that mimic the workload based on the events in the trace. These
	   threads can then replay the timings (CPU runtime and sleep patterns)
	   of the workload as it occurred when it was recorded - and can repeat
	   it a number of times, measuring its performance.)

	   'perf sched map' to print a textual context-switching outline of
	   workload captured via perf sched record.  Columns stand for
	   individual CPUs, and the two-letter shortcuts stand for tasks that
	   are running on a CPU. A '*' denotes the CPU that had the event, and
	   a dot signals an idle CPU.

OPTIONS

       -i, --input=<file>
	   Input file name. (default: perf.data unless stdin is a fifo)

       -v, --verbose
	   Be more verbose. (show symbol address, etc)

       -D, --dump-raw-trace=
	   Display verbose dump of the sched data.

SEE ALSO

       perf-record(1)

perf								    06/30/2014							     PERF-SCHED(1)

TAPSET::PERF(3stap)													       TAPSET::PERF(3stap)

NAME

       tapset::perf - systemtap perf probe points

DESCRIPTION

       This  family  of probe points is used to probe "perf events" on suitably configured kernels (2.6.33+).  It contains a number of aliases for
       the ABI-specified event type/config tuples.

       The default sampling interval of the translator(1000000) is not overridden, so some of these probe points may fire very infrequently.  Use
       the lower level perf.type(AA).config(BB).sample(CC) probe point if this is a problem.

       perf.hw.*
	      A  variety of hardware events, as generalized by the kernel.  These generally require running on bare hardware with selected proces-
	      sors.
	      probe perf.hw.cpu_cycles
	      probe perf.hw.instructions
	      probe perf.hw.cache_references
	      probe perf.hw.cache_misses
	      probe perf.hw.branch_instructions
	      probe perf.hw.branch_misses
	      probe perf.hw.bus_cycles

       perf.sw.*
	      Special "software" events provided by the kernel.  These sometimes work in virtualized environments and with more  processor  archi-
	      tectures.
	      probe perf.sw.cpu_clock
	      probe perf.sw.task_clock
	      probe perf.sw.page_faults
	      probe perf.sw.context_switches
	      probe perf.sw.cpu_migrations
	      probe perf.sw.page_faults_min
	      probe perf.sw.page_faults_maj
	      probe perf.sw.alignment_faults
	      probe perf.sw.emulation_faults

       perf.hw_cache.TYPE.ACCESS.RESULT
	      Hardware cache events, where available.  A subset of the following cartesian product may be available:
	      probe perf.hw_cache.l1d.*.*
	      probe perf.hw_cache.l1i.*.*
	      probe perf.hw_cache.ll.*.*
	      probe perf.hw_cache.dtlb.*.*
	      probe perf.hw_cache.itlb.*.*
	      probe perf.hw_cache.bpu.*.*

	      probe perf.hw_cache.*.read.*
	      probe perf.hw_cache.*.write.*
	      probe perf.hw_cache.*.prefetch.*

	      probe perf.hw_cache.*.*.access
	      probe perf.hw_cache.*.*.miss

SEE ALSO

       stap(1), stapprobes(3stap)

IBM
															       TAPSET::PERF(3stap)