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_SYSCALL_TRACEPOINTS
50 See Documentation/trace/ftrace-design.txt
52 config TRACER_MAX_TRACE
58 config FTRACE_NMI_ENTER
60 depends on HAVE_FTRACE_NMI_ENTER
64 select CONTEXT_SWITCH_TRACER
67 config CONTEXT_SWITCH_TRACER
70 config RING_BUFFER_ALLOW_SWAP
73 Allow the use of ring_buffer_swap_cpu.
74 Adds a very slight overhead to tracing when enabled.
76 # All tracer options should select GENERIC_TRACER. For those options that are
77 # enabled by all tracers (context switch and event tracer) they select TRACING.
78 # This allows those options to appear when no other tracer is selected. But the
79 # options do not appear when something else selects it. We need the two options
80 # GENERIC_TRACER and TRACING to avoid circular dependencies to accomplish the
81 # hiding of the automatic options.
87 select STACKTRACE if STACKTRACE_SUPPORT
98 # Minimum requirements an architecture has to meet for us to
99 # be able to offer generic tracing facilities:
101 config TRACING_SUPPORT
103 # PPC32 has no irqflags tracing support, but it can use most of the
104 # tracers anyway, they were tested to build and work. Note that new
105 # exceptions to this list aren't welcomed, better implement the
106 # irqflags tracing for your architecture.
107 depends on TRACE_IRQFLAGS_SUPPORT || PPC32
108 depends on STACKTRACE_SUPPORT
115 default y if DEBUG_KERNEL
117 Enable the kernel tracing infrastructure.
121 config FUNCTION_TRACER
122 bool "Kernel Function Tracer"
123 depends on HAVE_FUNCTION_TRACER
126 select GENERIC_TRACER
127 select CONTEXT_SWITCH_TRACER
129 Enable the kernel to trace every kernel function. This is done
130 by using a compiler feature to insert a small, 5-byte No-Operation
131 instruction at the beginning of every kernel function, which NOP
132 sequence is then dynamically patched into a tracer call when
133 tracing is enabled by the administrator. If it's runtime disabled
134 (the bootup default), then the overhead of the instructions is very
135 small and not measurable even in micro-benchmarks.
137 config FUNCTION_GRAPH_TRACER
138 bool "Kernel Function Graph Tracer"
139 depends on HAVE_FUNCTION_GRAPH_TRACER
140 depends on FUNCTION_TRACER
141 depends on !X86_32 || !CC_OPTIMIZE_FOR_SIZE
144 Enable the kernel to trace a function at both its return
146 Its first purpose is to trace the duration of functions and
147 draw a call graph for each thread with some information like
148 the return value. This is done by setting the current return
149 address on the current task structure into a stack of calls.
152 config IRQSOFF_TRACER
153 bool "Interrupts-off Latency Tracer"
155 depends on TRACE_IRQFLAGS_SUPPORT
156 depends on !ARCH_USES_GETTIMEOFFSET
157 select TRACE_IRQFLAGS
158 select GENERIC_TRACER
159 select TRACER_MAX_TRACE
160 select RING_BUFFER_ALLOW_SWAP
162 This option measures the time spent in irqs-off critical
163 sections, with microsecond accuracy.
165 The default measurement method is a maximum search, which is
166 disabled by default and can be runtime (re-)started
169 echo 0 > /sys/kernel/debug/tracing/tracing_max_latency
171 (Note that kernel size and overhead increase with this option
172 enabled. This option and the preempt-off timing option can be
173 used together or separately.)
175 config PREEMPT_TRACER
176 bool "Preemption-off Latency Tracer"
178 depends on !ARCH_USES_GETTIMEOFFSET
180 select GENERIC_TRACER
181 select TRACER_MAX_TRACE
182 select RING_BUFFER_ALLOW_SWAP
184 This option measures the time spent in preemption-off critical
185 sections, with microsecond accuracy.
187 The default measurement method is a maximum search, which is
188 disabled by default and can be runtime (re-)started
191 echo 0 > /sys/kernel/debug/tracing/tracing_max_latency
193 (Note that kernel size and overhead increase with this option
194 enabled. This option and the irqs-off timing option can be
195 used together or separately.)
198 bool "Scheduling Latency Tracer"
199 select GENERIC_TRACER
200 select CONTEXT_SWITCH_TRACER
201 select TRACER_MAX_TRACE
203 This tracer tracks the latency of the highest priority task
204 to be scheduled in, starting from the point it has woken up.
206 config ENABLE_DEFAULT_TRACERS
207 bool "Trace process context switches and events"
208 depends on !GENERIC_TRACER
211 This tracer hooks to various trace points in the kernel,
212 allowing the user to pick and choose which trace point they
213 want to trace. It also includes the sched_switch tracer plugin.
215 config FTRACE_SYSCALLS
216 bool "Trace syscalls"
217 depends on HAVE_SYSCALL_TRACEPOINTS
218 select GENERIC_TRACER
221 Basic tracer to catch the syscall entry and exit events.
223 config TRACE_BRANCH_PROFILING
225 select GENERIC_TRACER
228 prompt "Branch Profiling"
229 default BRANCH_PROFILE_NONE
231 The branch profiling is a software profiler. It will add hooks
232 into the C conditionals to test which path a branch takes.
234 The likely/unlikely profiler only looks at the conditions that
235 are annotated with a likely or unlikely macro.
237 The "all branch" profiler will profile every if-statement in the
238 kernel. This profiler will also enable the likely/unlikely
241 Either of the above profilers adds a bit of overhead to the system.
242 If unsure, choose "No branch profiling".
244 config BRANCH_PROFILE_NONE
245 bool "No branch profiling"
247 No branch profiling. Branch profiling adds a bit of overhead.
248 Only enable it if you want to analyse the branching behavior.
249 Otherwise keep it disabled.
251 config PROFILE_ANNOTATED_BRANCHES
252 bool "Trace likely/unlikely profiler"
253 select TRACE_BRANCH_PROFILING
255 This tracer profiles all the the likely and unlikely macros
256 in the kernel. It will display the results in:
258 /sys/kernel/debug/tracing/profile_annotated_branch
260 Note: this will add a significant overhead; only turn this
261 on if you need to profile the system's use of these macros.
263 config PROFILE_ALL_BRANCHES
264 bool "Profile all if conditionals"
265 select TRACE_BRANCH_PROFILING
267 This tracer profiles all branch conditions. Every if ()
268 taken in the kernel is recorded whether it hit or miss.
269 The results will be displayed in:
271 /sys/kernel/debug/tracing/profile_branch
273 This option also enables the likely/unlikely profiler.
275 This configuration, when enabled, will impose a great overhead
276 on the system. This should only be enabled when the system
277 is to be analyzed in much detail.
280 config TRACING_BRANCHES
283 Selected by tracers that will trace the likely and unlikely
284 conditions. This prevents the tracers themselves from being
285 profiled. Profiling the tracing infrastructure can only happen
286 when the likelys and unlikelys are not being traced.
289 bool "Trace likely/unlikely instances"
290 depends on TRACE_BRANCH_PROFILING
291 select TRACING_BRANCHES
293 This traces the events of likely and unlikely condition
294 calls in the kernel. The difference between this and the
295 "Trace likely/unlikely profiler" is that this is not a
296 histogram of the callers, but actually places the calling
297 events into a running trace buffer to see when and where the
298 events happened, as well as their results.
303 bool "Trace max stack"
304 depends on HAVE_FUNCTION_TRACER
305 select FUNCTION_TRACER
309 This special tracer records the maximum stack footprint of the
310 kernel and displays it in /sys/kernel/debug/tracing/stack_trace.
312 This tracer works by hooking into every function call that the
313 kernel executes, and keeping a maximum stack depth value and
314 stack-trace saved. If this is configured with DYNAMIC_FTRACE
315 then it will not have any overhead while the stack tracer
318 To enable the stack tracer on bootup, pass in 'stacktrace'
319 on the kernel command line.
321 The stack tracer can also be enabled or disabled via the
322 sysctl kernel.stack_tracer_enabled
326 config BLK_DEV_IO_TRACE
327 bool "Support for tracing block IO actions"
333 select GENERIC_TRACER
336 Say Y here if you want to be able to trace the block layer actions
337 on a given queue. Tracing allows you to see any traffic happening
338 on a block device queue. For more information (and the userspace
339 support tools needed), fetch the blktrace tools from:
341 git://git.kernel.dk/blktrace.git
343 Tracing also is possible using the ftrace interface, e.g.:
345 echo 1 > /sys/block/sda/sda1/trace/enable
346 echo blk > /sys/kernel/debug/tracing/current_tracer
347 cat /sys/kernel/debug/tracing/trace_pipe
353 depends on HAVE_REGS_AND_STACK_ACCESS_API
354 bool "Enable kprobes-based dynamic events"
358 This allows the user to add tracing events (similar to tracepoints)
359 on the fly via the ftrace interface. See
360 Documentation/trace/kprobetrace.txt for more details.
362 Those events can be inserted wherever kprobes can probe, and record
363 various register and memory values.
365 This option is also required by perf-probe subcommand of perf tools.
366 If you want to use perf tools, this option is strongly recommended.
368 config DYNAMIC_FTRACE
369 bool "enable/disable ftrace tracepoints dynamically"
370 depends on FUNCTION_TRACER
371 depends on HAVE_DYNAMIC_FTRACE
374 This option will modify all the calls to ftrace dynamically
375 (will patch them out of the binary image and replace them
376 with a No-Op instruction) as they are called. A table is
377 created to dynamically enable them again.
379 This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but
380 otherwise has native performance as long as no tracing is active.
382 The changes to the code are done by a kernel thread that
383 wakes up once a second and checks to see if any ftrace calls
384 were made. If so, it runs stop_machine (stops all CPUS)
385 and modifies the code to jump over the call to ftrace.
387 config FUNCTION_PROFILER
388 bool "Kernel function profiler"
389 depends on FUNCTION_TRACER
392 This option enables the kernel function profiler. A file is created
393 in debugfs called function_profile_enabled which defaults to zero.
394 When a 1 is echoed into this file profiling begins, and when a
395 zero is entered, profiling stops. A "functions" file is created in
396 the trace_stats directory; this file shows the list of functions that
397 have been hit and their counters.
401 config FTRACE_MCOUNT_RECORD
403 depends on DYNAMIC_FTRACE
404 depends on HAVE_FTRACE_MCOUNT_RECORD
406 config FTRACE_SELFTEST
409 config FTRACE_STARTUP_TEST
410 bool "Perform a startup test on ftrace"
411 depends on GENERIC_TRACER
412 select FTRACE_SELFTEST
414 This option performs a series of startup tests on ftrace. On bootup
415 a series of tests are made to verify that the tracer is
416 functioning properly. It will do tests on all the configured
419 config EVENT_TRACE_TEST_SYSCALLS
420 bool "Run selftest on syscall events"
421 depends on FTRACE_STARTUP_TEST
423 This option will also enable testing every syscall event.
424 It only enables the event and disables it and runs various loads
425 with the event enabled. This adds a bit more time for kernel boot
426 up since it runs this on every system call defined.
428 TBD - enable a way to actually call the syscalls as we test their
432 bool "Memory mapped IO tracing"
433 depends on HAVE_MMIOTRACE_SUPPORT && PCI
434 select GENERIC_TRACER
436 Mmiotrace traces Memory Mapped I/O access and is meant for
437 debugging and reverse engineering. It is called from the ioremap
438 implementation and works via page faults. Tracing is disabled by
439 default and can be enabled at run-time.
441 See Documentation/trace/mmiotrace.txt.
442 If you are not helping to develop drivers, say N.
444 config MMIOTRACE_TEST
445 tristate "Test module for mmiotrace"
446 depends on MMIOTRACE && m
448 This is a dumb module for testing mmiotrace. It is very dangerous
449 as it will write garbage to IO memory starting at a given address.
450 However, it should be safe to use on e.g. unused portion of VRAM.
452 Say N, unless you absolutely know what you are doing.
454 config RING_BUFFER_BENCHMARK
455 tristate "Ring buffer benchmark stress tester"
456 depends on RING_BUFFER
458 This option creates a test to stress the ring buffer and benchmark it.
459 It creates its own ring buffer such that it will not interfere with
460 any other users of the ring buffer (such as ftrace). It then creates
461 a producer and consumer that will run for 10 seconds and sleep for
462 10 seconds. Each interval it will print out the number of events
463 it recorded and give a rough estimate of how long each iteration took.
465 It does not disable interrupts or raise its priority, so it may be
466 affected by processes that are running.
472 endif # TRACING_SUPPORT