Lines Matching +full:entry +full:- +full:latency
1 # SPDX-License-Identifier: GPL-2.0-only
21 API, which will be used by other function-entry hooking
27 See Documentation/trace/ftrace-design.rst
32 See Documentation/trace/ftrace-design.rst
45 See Documentation/trace/ftrace-design.rst
92 See Documentation/trace/ftrace-design.rst
97 Arch supports the gcc options -pg with -mfentry
102 Arch supports the gcc options -pg with -mrecord-mcount and -nop-mcount
107 Arch supports objtool --mcount
112 Arch supports the objtool options --mcount with --mnop.
223 bool "Boot-time Tracing support"
242 by using a compiler feature to insert a small, 5-byte No-Operation
247 small and not measurable even in micro-benchmarks (at least on
258 and its entry.
273 enable it via the trace option funcgraph-retval.
284 enable it via the trace option funcgraph-retaddr.
294 enabled via the trace option func-args (for the function tracer) and
295 funcgraph-args (for the function graph tracer)
305 replace them with a No-Op instruction) on boot up. During
399 stack-trace saved. If this is configured with DYNAMIC_FTRACE
418 bool "Interrupts-off Latency Tracer"
428 This option measures the time spent in irqs-off critical
432 disabled by default and can be runtime (re-)started
438 enabled. This option and the preempt-off timing option can be
442 bool "Preemption-off Latency Tracer"
452 This option measures the time spent in preemption-off critical
456 disabled by default and can be runtime (re-)started
462 enabled. This option and the irqs-off timing option can be
466 bool "Scheduling Latency Tracer"
472 This tracer tracks the latency of the highest priority task
491 hwlat_detector/width - time in usecs for how long to spin for
492 hwlat_detector/window - time in usecs between the start of each
497 for "window - width" microseconds, where the system can
508 file. Every time a latency is greater than tracing_thresh, it will
516 In the context of high-performance computing (HPC), the Operating
520 can cause noise to the system. Moreover, hardware-related jobs can
526 note of the entry and exit point of any source of interferences,
527 increasing a per-cpu interference counter. It saves an interference
530 observes these interferences' entry events. When a noise happens
532 hardware noise counter increases, pointing to a hardware-related
552 to find sources of wakeup latencies of real-time threads.
554 The tracer creates a per-cpu kernel thread with real-time priority.
557 then computes a wakeup latency value as the difference between
562 timer latency observed at the hardirq context before the
563 activation of the thread. The second is the timer latency observed
582 default and can be enabled at run-time.
602 Basic tracer to catch the syscall entry and exit events.
646 or irq latency tracers are enabled, as those need to swap as well
663 The "all branch" profiler will profile every if-statement in the
756 bool "Enable fprobe-based dynamic events"
762 This allows user to add tracing events on the function entry and
764 and the kprobe events on function entry and exit will be
776 kernel function entry or a tracepoint.
782 bool "Enable kprobes-based dynamic events"
795 This option is also required by perf-probe subcommand of perf tools.
808 functions are protected from kprobe-events to prevent an infinite
819 bool "Enable uprobes-based dynamic events"
833 This option is required if you plan to use perf-probe subcommand
837 bool "Enable event-based dynamic events"
880 depends on $(cc-option,-mrecord-mcount)
903 tracing_map is a special-purpose lock-free map for tracing,
904 separated out as a stand-alone facility in order to allow it
915 Synthetic events are user-defined trace events that can be
919 by way of an in-kernel API.
931 User trace events are user-defined trace events that
957 Inter-event tracing of quantities such as latencies is also
967 Allow user-space to inject a specific trace event into the ring
1142 TBD - enable a way to actually call the syscalls as we test their
1153 is not done correctly, it will cause non-deterministic failures.
1213 tristate "Test module to create a preempt / IRQ disable delay thread to test latency tracers"
1216 Select this option to build a test module that can help test latency
1222 irq-disabled critical sections for 500us:
1225 What's more, if you want to attach the test on the cpu which the latency
1232 tristate "Test module for in-kernel synthetic event generation"
1236 functionality of in-kernel synthetic event definition and
1245 tristate "Test module for in-kernel kprobe event generation"
1249 functionality of in-kernel kprobe event definition.
1266 - Helps developers verify that nothing is broken.
1268 - Provides educational information to support the details
1270 Documentation/trace/histogram-design.rst.