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_DYNAMIC_FTRACE_WITH_REGS
45 config HAVE_FTRACE_MCOUNT_RECORD
48 See Documentation/trace/ftrace-design.txt
50 config HAVE_SYSCALL_TRACEPOINTS
53 See Documentation/trace/ftrace-design.txt
58 Arch supports the gcc options -pg with -mfentry
60 config HAVE_C_RECORDMCOUNT
63 C version of recordmcount available?
65 config TRACER_MAX_TRACE
75 config FTRACE_NMI_ENTER
77 depends on HAVE_FTRACE_NMI_ENTER
81 select CONTEXT_SWITCH_TRACER
84 config EVENT_POWER_TRACING_DEPRECATED
85 depends on EVENT_TRACING
86 bool "Deprecated power event trace API, to be removed"
89 Provides old power event types:
90 C-state/idle accounting events:
93 and old cpufreq accounting event:
95 This is for userspace compatibility
96 and will vanish after 5 kernel iterations,
99 config CONTEXT_SWITCH_TRACER
102 config RING_BUFFER_ALLOW_SWAP
105 Allow the use of ring_buffer_swap_cpu.
106 Adds a very slight overhead to tracing when enabled.
108 # All tracer options should select GENERIC_TRACER. For those options that are
109 # enabled by all tracers (context switch and event tracer) they select TRACING.
110 # This allows those options to appear when no other tracer is selected. But the
111 # options do not appear when something else selects it. We need the two options
112 # GENERIC_TRACER and TRACING to avoid circular dependencies to accomplish the
113 # hiding of the automatic options.
119 select STACKTRACE if STACKTRACE_SUPPORT
127 config GENERIC_TRACER
132 # Minimum requirements an architecture has to meet for us to
133 # be able to offer generic tracing facilities:
135 config TRACING_SUPPORT
137 # PPC32 has no irqflags tracing support, but it can use most of the
138 # tracers anyway, they were tested to build and work. Note that new
139 # exceptions to this list aren't welcomed, better implement the
140 # irqflags tracing for your architecture.
141 depends on TRACE_IRQFLAGS_SUPPORT || PPC32
142 depends on STACKTRACE_SUPPORT
149 default y if DEBUG_KERNEL
151 Enable the kernel tracing infrastructure.
155 config FUNCTION_TRACER
156 bool "Kernel Function Tracer"
157 depends on HAVE_FUNCTION_TRACER
159 select GENERIC_TRACER
160 select CONTEXT_SWITCH_TRACER
162 Enable the kernel to trace every kernel function. This is done
163 by using a compiler feature to insert a small, 5-byte No-Operation
164 instruction at the beginning of every kernel function, which NOP
165 sequence is then dynamically patched into a tracer call when
166 tracing is enabled by the administrator. If it's runtime disabled
167 (the bootup default), then the overhead of the instructions is very
168 small and not measurable even in micro-benchmarks.
170 config FUNCTION_GRAPH_TRACER
171 bool "Kernel Function Graph Tracer"
172 depends on HAVE_FUNCTION_GRAPH_TRACER
173 depends on FUNCTION_TRACER
174 depends on !X86_32 || !CC_OPTIMIZE_FOR_SIZE
177 Enable the kernel to trace a function at both its return
179 Its first purpose is to trace the duration of functions and
180 draw a call graph for each thread with some information like
181 the return value. This is done by setting the current return
182 address on the current task structure into a stack of calls.
185 config IRQSOFF_TRACER
186 bool "Interrupts-off Latency Tracer"
188 depends on TRACE_IRQFLAGS_SUPPORT
189 depends on !ARCH_USES_GETTIMEOFFSET
190 select TRACE_IRQFLAGS
191 select GENERIC_TRACER
192 select TRACER_MAX_TRACE
193 select RING_BUFFER_ALLOW_SWAP
195 This option measures the time spent in irqs-off critical
196 sections, with microsecond accuracy.
198 The default measurement method is a maximum search, which is
199 disabled by default and can be runtime (re-)started
202 echo 0 > /sys/kernel/debug/tracing/tracing_max_latency
204 (Note that kernel size and overhead increase with this option
205 enabled. This option and the preempt-off timing option can be
206 used together or separately.)
208 config PREEMPT_TRACER
209 bool "Preemption-off Latency Tracer"
211 depends on !ARCH_USES_GETTIMEOFFSET
213 select GENERIC_TRACER
214 select TRACER_MAX_TRACE
215 select RING_BUFFER_ALLOW_SWAP
217 This option measures the time spent in preemption-off critical
218 sections, with microsecond accuracy.
220 The default measurement method is a maximum search, which is
221 disabled by default and can be runtime (re-)started
224 echo 0 > /sys/kernel/debug/tracing/tracing_max_latency
226 (Note that kernel size and overhead increase with this option
227 enabled. This option and the irqs-off timing option can be
228 used together or separately.)
231 bool "Scheduling Latency Tracer"
232 select GENERIC_TRACER
233 select CONTEXT_SWITCH_TRACER
234 select TRACER_MAX_TRACE
236 This tracer tracks the latency of the highest priority task
237 to be scheduled in, starting from the point it has woken up.
239 config ENABLE_DEFAULT_TRACERS
240 bool "Trace process context switches and events"
241 depends on !GENERIC_TRACER
244 This tracer hooks to various trace points in the kernel,
245 allowing the user to pick and choose which trace point they
246 want to trace. It also includes the sched_switch tracer plugin.
248 config FTRACE_SYSCALLS
249 bool "Trace syscalls"
250 depends on HAVE_SYSCALL_TRACEPOINTS
251 select GENERIC_TRACER
254 Basic tracer to catch the syscall entry and exit events.
256 config TRACER_SNAPSHOT
257 bool "Create a snapshot trace buffer"
258 select TRACER_MAX_TRACE
260 Allow tracing users to take snapshot of the current buffer using the
261 ftrace interface, e.g.:
263 echo 1 > /sys/kernel/debug/tracing/snapshot
266 config TRACE_BRANCH_PROFILING
268 select GENERIC_TRACER
271 prompt "Branch Profiling"
272 default BRANCH_PROFILE_NONE
274 The branch profiling is a software profiler. It will add hooks
275 into the C conditionals to test which path a branch takes.
277 The likely/unlikely profiler only looks at the conditions that
278 are annotated with a likely or unlikely macro.
280 The "all branch" profiler will profile every if-statement in the
281 kernel. This profiler will also enable the likely/unlikely
284 Either of the above profilers adds a bit of overhead to the system.
285 If unsure, choose "No branch profiling".
287 config BRANCH_PROFILE_NONE
288 bool "No branch profiling"
290 No branch profiling. Branch profiling adds a bit of overhead.
291 Only enable it if you want to analyse the branching behavior.
292 Otherwise keep it disabled.
294 config PROFILE_ANNOTATED_BRANCHES
295 bool "Trace likely/unlikely profiler"
296 select TRACE_BRANCH_PROFILING
298 This tracer profiles all likely and unlikely macros
299 in the kernel. It will display the results in:
301 /sys/kernel/debug/tracing/trace_stat/branch_annotated
303 Note: this will add a significant overhead; only turn this
304 on if you need to profile the system's use of these macros.
306 config PROFILE_ALL_BRANCHES
307 bool "Profile all if conditionals"
308 select TRACE_BRANCH_PROFILING
310 This tracer profiles all branch conditions. Every if ()
311 taken in the kernel is recorded whether it hit or miss.
312 The results will be displayed in:
314 /sys/kernel/debug/tracing/trace_stat/branch_all
316 This option also enables the likely/unlikely profiler.
318 This configuration, when enabled, will impose a great overhead
319 on the system. This should only be enabled when the system
320 is to be analyzed in much detail.
323 config TRACING_BRANCHES
326 Selected by tracers that will trace the likely and unlikely
327 conditions. This prevents the tracers themselves from being
328 profiled. Profiling the tracing infrastructure can only happen
329 when the likelys and unlikelys are not being traced.
332 bool "Trace likely/unlikely instances"
333 depends on TRACE_BRANCH_PROFILING
334 select TRACING_BRANCHES
336 This traces the events of likely and unlikely condition
337 calls in the kernel. The difference between this and the
338 "Trace likely/unlikely profiler" is that this is not a
339 histogram of the callers, but actually places the calling
340 events into a running trace buffer to see when and where the
341 events happened, as well as their results.
346 bool "Trace max stack"
347 depends on HAVE_FUNCTION_TRACER
348 select FUNCTION_TRACER
352 This special tracer records the maximum stack footprint of the
353 kernel and displays it in /sys/kernel/debug/tracing/stack_trace.
355 This tracer works by hooking into every function call that the
356 kernel executes, and keeping a maximum stack depth value and
357 stack-trace saved. If this is configured with DYNAMIC_FTRACE
358 then it will not have any overhead while the stack tracer
361 To enable the stack tracer on bootup, pass in 'stacktrace'
362 on the kernel command line.
364 The stack tracer can also be enabled or disabled via the
365 sysctl kernel.stack_tracer_enabled
369 config BLK_DEV_IO_TRACE
370 bool "Support for tracing block IO actions"
376 select GENERIC_TRACER
379 Say Y here if you want to be able to trace the block layer actions
380 on a given queue. Tracing allows you to see any traffic happening
381 on a block device queue. For more information (and the userspace
382 support tools needed), fetch the blktrace tools from:
384 git://git.kernel.dk/blktrace.git
386 Tracing also is possible using the ftrace interface, e.g.:
388 echo 1 > /sys/block/sda/sda1/trace/enable
389 echo blk > /sys/kernel/debug/tracing/current_tracer
390 cat /sys/kernel/debug/tracing/trace_pipe
396 depends on HAVE_REGS_AND_STACK_ACCESS_API
397 bool "Enable kprobes-based dynamic events"
402 This allows the user to add tracing events (similar to tracepoints)
403 on the fly via the ftrace interface. See
404 Documentation/trace/kprobetrace.txt for more details.
406 Those events can be inserted wherever kprobes can probe, and record
407 various register and memory values.
409 This option is also required by perf-probe subcommand of perf tools.
410 If you want to use perf tools, this option is strongly recommended.
413 bool "Enable uprobes-based dynamic events"
414 depends on ARCH_SUPPORTS_UPROBES
421 This allows the user to add tracing events on top of userspace
422 dynamic events (similar to tracepoints) on the fly via the trace
423 events interface. Those events can be inserted wherever uprobes
424 can probe, and record various registers.
425 This option is required if you plan to use perf-probe subcommand
426 of perf tools on user space applications.
431 config DYNAMIC_FTRACE
432 bool "enable/disable ftrace tracepoints dynamically"
433 depends on FUNCTION_TRACER
434 depends on HAVE_DYNAMIC_FTRACE
437 This option will modify all the calls to ftrace dynamically
438 (will patch them out of the binary image and replace them
439 with a No-Op instruction) as they are called. A table is
440 created to dynamically enable them again.
442 This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but
443 otherwise has native performance as long as no tracing is active.
445 The changes to the code are done by a kernel thread that
446 wakes up once a second and checks to see if any ftrace calls
447 were made. If so, it runs stop_machine (stops all CPUS)
448 and modifies the code to jump over the call to ftrace.
450 config DYNAMIC_FTRACE_WITH_REGS
452 depends on DYNAMIC_FTRACE
453 depends on HAVE_DYNAMIC_FTRACE_WITH_REGS
455 config FUNCTION_PROFILER
456 bool "Kernel function profiler"
457 depends on FUNCTION_TRACER
460 This option enables the kernel function profiler. A file is created
461 in debugfs called function_profile_enabled which defaults to zero.
462 When a 1 is echoed into this file profiling begins, and when a
463 zero is entered, profiling stops. A "functions" file is created in
464 the trace_stats directory; this file shows the list of functions that
465 have been hit and their counters.
469 config FTRACE_MCOUNT_RECORD
471 depends on DYNAMIC_FTRACE
472 depends on HAVE_FTRACE_MCOUNT_RECORD
474 config FTRACE_SELFTEST
477 config FTRACE_STARTUP_TEST
478 bool "Perform a startup test on ftrace"
479 depends on GENERIC_TRACER
480 select FTRACE_SELFTEST
482 This option performs a series of startup tests on ftrace. On bootup
483 a series of tests are made to verify that the tracer is
484 functioning properly. It will do tests on all the configured
487 config EVENT_TRACE_TEST_SYSCALLS
488 bool "Run selftest on syscall events"
489 depends on FTRACE_STARTUP_TEST
491 This option will also enable testing every syscall event.
492 It only enables the event and disables it and runs various loads
493 with the event enabled. This adds a bit more time for kernel boot
494 up since it runs this on every system call defined.
496 TBD - enable a way to actually call the syscalls as we test their
500 bool "Memory mapped IO tracing"
501 depends on HAVE_MMIOTRACE_SUPPORT && PCI
502 select GENERIC_TRACER
504 Mmiotrace traces Memory Mapped I/O access and is meant for
505 debugging and reverse engineering. It is called from the ioremap
506 implementation and works via page faults. Tracing is disabled by
507 default and can be enabled at run-time.
509 See Documentation/trace/mmiotrace.txt.
510 If you are not helping to develop drivers, say N.
512 config MMIOTRACE_TEST
513 tristate "Test module for mmiotrace"
514 depends on MMIOTRACE && m
516 This is a dumb module for testing mmiotrace. It is very dangerous
517 as it will write garbage to IO memory starting at a given address.
518 However, it should be safe to use on e.g. unused portion of VRAM.
520 Say N, unless you absolutely know what you are doing.
522 config RING_BUFFER_BENCHMARK
523 tristate "Ring buffer benchmark stress tester"
524 depends on RING_BUFFER
526 This option creates a test to stress the ring buffer and benchmark it.
527 It creates its own ring buffer such that it will not interfere with
528 any other users of the ring buffer (such as ftrace). It then creates
529 a producer and consumer that will run for 10 seconds and sleep for
530 10 seconds. Each interval it will print out the number of events
531 it recorded and give a rough estimate of how long each iteration took.
533 It does not disable interrupts or raise its priority, so it may be
534 affected by processes that are running.
540 endif # TRACING_SUPPORT