2 # Architectures that offer an FUNCTION_TRACER implementation should
3 # select HAVE_FUNCTION_TRACER:
6 config USER_STACKTRACE_SUPPORT
12 config HAVE_FTRACE_NMI_ENTER
15 See Documentation/trace/ftrace-design.txt
17 config HAVE_FUNCTION_TRACER
20 See Documentation/trace/ftrace-design.txt
22 config HAVE_FUNCTION_GRAPH_TRACER
25 See Documentation/trace/ftrace-design.txt
27 config HAVE_FUNCTION_GRAPH_FP_TEST
30 See Documentation/trace/ftrace-design.txt
32 config HAVE_FUNCTION_TRACE_MCOUNT_TEST
35 See Documentation/trace/ftrace-design.txt
37 config HAVE_DYNAMIC_FTRACE
40 See Documentation/trace/ftrace-design.txt
42 config HAVE_FTRACE_MCOUNT_RECORD
45 See Documentation/trace/ftrace-design.txt
47 config HAVE_HW_BRANCH_TRACER
50 config HAVE_SYSCALL_TRACEPOINTS
53 See Documentation/trace/ftrace-design.txt
55 config TRACER_MAX_TRACE
61 config FTRACE_NMI_ENTER
63 depends on HAVE_FTRACE_NMI_ENTER
67 select CONTEXT_SWITCH_TRACER
70 config CONTEXT_SWITCH_TRACER
73 config RING_BUFFER_ALLOW_SWAP
76 Allow the use of ring_buffer_swap_cpu.
77 Adds a very slight overhead to tracing when enabled.
79 # All tracer options should select GENERIC_TRACER. For those options that are
80 # enabled by all tracers (context switch and event tracer) they select TRACING.
81 # This allows those options to appear when no other tracer is selected. But the
82 # options do not appear when something else selects it. We need the two options
83 # GENERIC_TRACER and TRACING to avoid circular dependencies to accomplish the
84 # hiding of the automatic options.
90 select STACKTRACE if STACKTRACE_SUPPORT
101 # Minimum requirements an architecture has to meet for us to
102 # be able to offer generic tracing facilities:
104 config TRACING_SUPPORT
106 # PPC32 has no irqflags tracing support, but it can use most of the
107 # tracers anyway, they were tested to build and work. Note that new
108 # exceptions to this list aren't welcomed, better implement the
109 # irqflags tracing for your architecture.
110 depends on TRACE_IRQFLAGS_SUPPORT || PPC32
111 depends on STACKTRACE_SUPPORT
118 default y if DEBUG_KERNEL
120 Enable the kernel tracing infrastructure.
124 config FUNCTION_TRACER
125 bool "Kernel Function Tracer"
126 depends on HAVE_FUNCTION_TRACER
129 select GENERIC_TRACER
130 select CONTEXT_SWITCH_TRACER
132 Enable the kernel to trace every kernel function. This is done
133 by using a compiler feature to insert a small, 5-byte No-Operation
134 instruction at the beginning of every kernel function, which NOP
135 sequence is then dynamically patched into a tracer call when
136 tracing is enabled by the administrator. If it's runtime disabled
137 (the bootup default), then the overhead of the instructions is very
138 small and not measurable even in micro-benchmarks.
140 config FUNCTION_GRAPH_TRACER
141 bool "Kernel Function Graph Tracer"
142 depends on HAVE_FUNCTION_GRAPH_TRACER
143 depends on FUNCTION_TRACER
144 depends on !X86_32 || !CC_OPTIMIZE_FOR_SIZE
147 Enable the kernel to trace a function at both its return
149 Its first purpose is to trace the duration of functions and
150 draw a call graph for each thread with some information like
151 the return value. This is done by setting the current return
152 address on the current task structure into a stack of calls.
155 config IRQSOFF_TRACER
156 bool "Interrupts-off Latency Tracer"
158 depends on TRACE_IRQFLAGS_SUPPORT
159 depends on GENERIC_TIME
160 select TRACE_IRQFLAGS
161 select GENERIC_TRACER
162 select TRACER_MAX_TRACE
163 select RING_BUFFER_ALLOW_SWAP
165 This option measures the time spent in irqs-off critical
166 sections, with microsecond accuracy.
168 The default measurement method is a maximum search, which is
169 disabled by default and can be runtime (re-)started
172 echo 0 > /sys/kernel/debug/tracing/tracing_max_latency
174 (Note that kernel size and overhead increase with this option
175 enabled. This option and the preempt-off timing option can be
176 used together or separately.)
178 config PREEMPT_TRACER
179 bool "Preemption-off Latency Tracer"
181 depends on GENERIC_TIME
183 select GENERIC_TRACER
184 select TRACER_MAX_TRACE
185 select RING_BUFFER_ALLOW_SWAP
187 This option measures the time spent in preemption-off critical
188 sections, with microsecond accuracy.
190 The default measurement method is a maximum search, which is
191 disabled by default and can be runtime (re-)started
194 echo 0 > /sys/kernel/debug/tracing/tracing_max_latency
196 (Note that kernel size and overhead increase with this option
197 enabled. This option and the irqs-off timing option can be
198 used together or separately.)
200 config SYSPROF_TRACER
201 bool "Sysprof Tracer"
203 select GENERIC_TRACER
204 select CONTEXT_SWITCH_TRACER
206 This tracer provides the trace needed by the 'Sysprof' userspace
210 bool "Scheduling Latency Tracer"
211 select GENERIC_TRACER
212 select CONTEXT_SWITCH_TRACER
213 select TRACER_MAX_TRACE
215 This tracer tracks the latency of the highest priority task
216 to be scheduled in, starting from the point it has woken up.
218 config ENABLE_DEFAULT_TRACERS
219 bool "Trace process context switches and events"
220 depends on !GENERIC_TRACER
223 This tracer hooks to various trace points in the kernel,
224 allowing the user to pick and choose which trace point they
225 want to trace. It also includes the sched_switch tracer plugin.
227 config FTRACE_SYSCALLS
228 bool "Trace syscalls"
229 depends on HAVE_SYSCALL_TRACEPOINTS
230 select GENERIC_TRACER
233 Basic tracer to catch the syscall entry and exit events.
236 bool "Trace boot initcalls"
237 select GENERIC_TRACER
238 select CONTEXT_SWITCH_TRACER
240 This tracer helps developers to optimize boot times: it records
241 the timings of the initcalls and traces key events and the identity
242 of tasks that can cause boot delays, such as context-switches.
244 Its aim is to be parsed by the scripts/bootgraph.pl tool to
245 produce pretty graphics about boot inefficiencies, giving a visual
246 representation of the delays during initcalls - but the raw
247 /debug/tracing/trace text output is readable too.
249 You must pass in initcall_debug and ftrace=initcall to the kernel
250 command line to enable this on bootup.
252 config TRACE_BRANCH_PROFILING
254 select GENERIC_TRACER
257 prompt "Branch Profiling"
258 default BRANCH_PROFILE_NONE
260 The branch profiling is a software profiler. It will add hooks
261 into the C conditionals to test which path a branch takes.
263 The likely/unlikely profiler only looks at the conditions that
264 are annotated with a likely or unlikely macro.
266 The "all branch" profiler will profile every if-statement in the
267 kernel. This profiler will also enable the likely/unlikely
270 Either of the above profilers adds a bit of overhead to the system.
271 If unsure, choose "No branch profiling".
273 config BRANCH_PROFILE_NONE
274 bool "No branch profiling"
276 No branch profiling. Branch profiling adds a bit of overhead.
277 Only enable it if you want to analyse the branching behavior.
278 Otherwise keep it disabled.
280 config PROFILE_ANNOTATED_BRANCHES
281 bool "Trace likely/unlikely profiler"
282 select TRACE_BRANCH_PROFILING
284 This tracer profiles all the the likely and unlikely macros
285 in the kernel. It will display the results in:
287 /sys/kernel/debug/tracing/profile_annotated_branch
289 Note: this will add a significant overhead; only turn this
290 on if you need to profile the system's use of these macros.
292 config PROFILE_ALL_BRANCHES
293 bool "Profile all if conditionals"
294 select TRACE_BRANCH_PROFILING
296 This tracer profiles all branch conditions. Every if ()
297 taken in the kernel is recorded whether it hit or miss.
298 The results will be displayed in:
300 /sys/kernel/debug/tracing/profile_branch
302 This option also enables the likely/unlikely profiler.
304 This configuration, when enabled, will impose a great overhead
305 on the system. This should only be enabled when the system
306 is to be analyzed in much detail.
309 config TRACING_BRANCHES
312 Selected by tracers that will trace the likely and unlikely
313 conditions. This prevents the tracers themselves from being
314 profiled. Profiling the tracing infrastructure can only happen
315 when the likelys and unlikelys are not being traced.
318 bool "Trace likely/unlikely instances"
319 depends on TRACE_BRANCH_PROFILING
320 select TRACING_BRANCHES
322 This traces the events of likely and unlikely condition
323 calls in the kernel. The difference between this and the
324 "Trace likely/unlikely profiler" is that this is not a
325 histogram of the callers, but actually places the calling
326 events into a running trace buffer to see when and where the
327 events happened, as well as their results.
332 bool "Trace power consumption behavior"
334 select GENERIC_TRACER
336 This tracer helps developers to analyze and optimize the kernel's
337 power management decisions, specifically the C-state and P-state
341 bool "Trace read and write access on kernel memory locations"
342 depends on HAVE_HW_BREAKPOINT
345 This tracer helps find read and write operations on any given kernel
346 symbol i.e. /proc/kallsyms.
348 config PROFILE_KSYM_TRACER
349 bool "Profile all kernel memory accesses on 'watched' variables"
350 depends on KSYM_TRACER
352 This tracer profiles kernel accesses on variables watched through the
353 ksym tracer ftrace plugin. Depending upon the hardware, all read
354 and write operations on kernel variables can be monitored for
357 The results will be displayed in:
358 /debugfs/tracing/profile_ksym
363 bool "Trace max stack"
364 depends on HAVE_FUNCTION_TRACER
365 select FUNCTION_TRACER
369 This special tracer records the maximum stack footprint of the
370 kernel and displays it in /sys/kernel/debug/tracing/stack_trace.
372 This tracer works by hooking into every function call that the
373 kernel executes, and keeping a maximum stack depth value and
374 stack-trace saved. If this is configured with DYNAMIC_FTRACE
375 then it will not have any overhead while the stack tracer
378 To enable the stack tracer on bootup, pass in 'stacktrace'
379 on the kernel command line.
381 The stack tracer can also be enabled or disabled via the
382 sysctl kernel.stack_tracer_enabled
386 config HW_BRANCH_TRACER
387 depends on HAVE_HW_BRANCH_TRACER
388 bool "Trace hw branches"
389 select GENERIC_TRACER
391 This tracer records all branches on the system in a circular
392 buffer, giving access to the last N branches for each cpu.
395 bool "Trace SLAB allocations"
396 select GENERIC_TRACER
398 kmemtrace provides tracing for slab allocator functions, such as
399 kmalloc, kfree, kmem_cache_alloc, kmem_cache_free, etc. Collected
400 data is then fed to the userspace application in order to analyse
401 allocation hotspots, internal fragmentation and so on, making it
402 possible to see how well an allocator performs, as well as debug
403 and profile kernel code.
405 This requires an userspace application to use. See
406 Documentation/trace/kmemtrace.txt for more information.
408 Saying Y will make the kernel somewhat larger and slower. However,
409 if you disable kmemtrace at run-time or boot-time, the performance
410 impact is minimal (depending on the arch the kernel is built for).
414 config WORKQUEUE_TRACER
415 bool "Trace workqueues"
416 select GENERIC_TRACER
418 The workqueue tracer provides some statistical information
419 about each cpu workqueue thread such as the number of the
420 works inserted and executed since their creation. It can help
421 to evaluate the amount of work each of them has to perform.
422 For example it can help a developer to decide whether he should
423 choose a per-cpu workqueue instead of a singlethreaded one.
425 config BLK_DEV_IO_TRACE
426 bool "Support for tracing block IO actions"
432 select GENERIC_TRACER
435 Say Y here if you want to be able to trace the block layer actions
436 on a given queue. Tracing allows you to see any traffic happening
437 on a block device queue. For more information (and the userspace
438 support tools needed), fetch the blktrace tools from:
440 git://git.kernel.dk/blktrace.git
442 Tracing also is possible using the ftrace interface, e.g.:
444 echo 1 > /sys/block/sda/sda1/trace/enable
445 echo blk > /sys/kernel/debug/tracing/current_tracer
446 cat /sys/kernel/debug/tracing/trace_pipe
453 bool "Enable kprobes-based dynamic events"
457 This allows the user to add tracing events (similar to tracepoints)
458 on the fly via the ftrace interface. See
459 Documentation/trace/kprobetrace.txt for more details.
461 Those events can be inserted wherever kprobes can probe, and record
462 various register and memory values.
464 This option is also required by perf-probe subcommand of perf tools.
465 If you want to use perf tools, this option is strongly recommended.
467 config DYNAMIC_FTRACE
468 bool "enable/disable ftrace tracepoints dynamically"
469 depends on FUNCTION_TRACER
470 depends on HAVE_DYNAMIC_FTRACE
473 This option will modify all the calls to ftrace dynamically
474 (will patch them out of the binary image and replace them
475 with a No-Op instruction) as they are called. A table is
476 created to dynamically enable them again.
478 This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but
479 otherwise has native performance as long as no tracing is active.
481 The changes to the code are done by a kernel thread that
482 wakes up once a second and checks to see if any ftrace calls
483 were made. If so, it runs stop_machine (stops all CPUS)
484 and modifies the code to jump over the call to ftrace.
486 config FUNCTION_PROFILER
487 bool "Kernel function profiler"
488 depends on FUNCTION_TRACER
491 This option enables the kernel function profiler. A file is created
492 in debugfs called function_profile_enabled which defaults to zero.
493 When a 1 is echoed into this file profiling begins, and when a
494 zero is entered, profiling stops. A "functions" file is created in
495 the trace_stats directory; this file shows the list of functions that
496 have been hit and their counters.
500 config FTRACE_MCOUNT_RECORD
502 depends on DYNAMIC_FTRACE
503 depends on HAVE_FTRACE_MCOUNT_RECORD
505 config FTRACE_SELFTEST
508 config FTRACE_STARTUP_TEST
509 bool "Perform a startup test on ftrace"
510 depends on GENERIC_TRACER
511 select FTRACE_SELFTEST
513 This option performs a series of startup tests on ftrace. On bootup
514 a series of tests are made to verify that the tracer is
515 functioning properly. It will do tests on all the configured
518 config EVENT_TRACE_TEST_SYSCALLS
519 bool "Run selftest on syscall events"
520 depends on FTRACE_STARTUP_TEST
522 This option will also enable testing every syscall event.
523 It only enables the event and disables it and runs various loads
524 with the event enabled. This adds a bit more time for kernel boot
525 up since it runs this on every system call defined.
527 TBD - enable a way to actually call the syscalls as we test their
531 bool "Memory mapped IO tracing"
532 depends on HAVE_MMIOTRACE_SUPPORT && PCI
533 select GENERIC_TRACER
535 Mmiotrace traces Memory Mapped I/O access and is meant for
536 debugging and reverse engineering. It is called from the ioremap
537 implementation and works via page faults. Tracing is disabled by
538 default and can be enabled at run-time.
540 See Documentation/trace/mmiotrace.txt.
541 If you are not helping to develop drivers, say N.
543 config MMIOTRACE_TEST
544 tristate "Test module for mmiotrace"
545 depends on MMIOTRACE && m
547 This is a dumb module for testing mmiotrace. It is very dangerous
548 as it will write garbage to IO memory starting at a given address.
549 However, it should be safe to use on e.g. unused portion of VRAM.
551 Say N, unless you absolutely know what you are doing.
553 config RING_BUFFER_BENCHMARK
554 tristate "Ring buffer benchmark stress tester"
555 depends on RING_BUFFER
557 This option creates a test to stress the ring buffer and benchmark it.
558 It creates its own ring buffer such that it will not interfere with
559 any other users of the ring buffer (such as ftrace). It then creates
560 a producer and consumer that will run for 10 seconds and sleep for
561 10 seconds. Each interval it will print out the number of events
562 it recorded and give a rough estimate of how long each iteration took.
564 It does not disable interrupts or raise its priority, so it may be
565 affected by processes that are running.
571 endif # TRACING_SUPPORT