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 config HAVE_FUNCTION_TRACER
18 config HAVE_FUNCTION_GRAPH_TRACER
21 config HAVE_FUNCTION_TRACE_MCOUNT_TEST
24 This gets selected when the arch tests the function_trace_stop
25 variable at the mcount call site. Otherwise, this variable
26 is tested by the called function.
28 config HAVE_DYNAMIC_FTRACE
31 config HAVE_FTRACE_MCOUNT_RECORD
34 config HAVE_HW_BRANCH_TRACER
37 config HAVE_FTRACE_SYSCALLS
40 config TRACER_MAX_TRACE
46 config FTRACE_NMI_ENTER
48 depends on HAVE_FTRACE_NMI_ENTER
55 select STACKTRACE if STACKTRACE_SUPPORT
61 # Minimum requirements an architecture has to meet for us to
62 # be able to offer generic tracing facilities:
64 config TRACING_SUPPORT
66 # PPC32 has no irqflags tracing support, but it can use most of the
67 # tracers anyway, they were tested to build and work. Note that new
68 # exceptions to this list aren't welcomed, better implement the
69 # irqflags tracing for your architecture.
70 depends on TRACE_IRQFLAGS_SUPPORT || PPC32
71 depends on STACKTRACE_SUPPORT
78 config FUNCTION_TRACER
79 bool "Kernel Function Tracer"
80 depends on HAVE_FUNCTION_TRACER
84 select CONTEXT_SWITCH_TRACER
86 Enable the kernel to trace every kernel function. This is done
87 by using a compiler feature to insert a small, 5-byte No-Operation
88 instruction to the beginning of every kernel function, which NOP
89 sequence is then dynamically patched into a tracer call when
90 tracing is enabled by the administrator. If it's runtime disabled
91 (the bootup default), then the overhead of the instructions is very
92 small and not measurable even in micro-benchmarks.
94 config FUNCTION_GRAPH_TRACER
95 bool "Kernel Function Graph Tracer"
96 depends on HAVE_FUNCTION_GRAPH_TRACER
97 depends on FUNCTION_TRACER
100 Enable the kernel to trace a function at both its return
102 Its first purpose is to trace the duration of functions and
103 draw a call graph for each thread with some information like
104 the return value. This is done by setting the current return
105 address on the current task structure into a stack of calls.
107 config IRQSOFF_TRACER
108 bool "Interrupts-off Latency Tracer"
110 depends on TRACE_IRQFLAGS_SUPPORT
111 depends on GENERIC_TIME
112 select TRACE_IRQFLAGS
114 select TRACER_MAX_TRACE
116 This option measures the time spent in irqs-off critical
117 sections, with microsecond accuracy.
119 The default measurement method is a maximum search, which is
120 disabled by default and can be runtime (re-)started
123 echo 0 > /debugfs/tracing/tracing_max_latency
125 (Note that kernel size and overhead increases with this option
126 enabled. This option and the preempt-off timing option can be
127 used together or separately.)
129 config PREEMPT_TRACER
130 bool "Preemption-off Latency Tracer"
132 depends on GENERIC_TIME
135 select TRACER_MAX_TRACE
137 This option measures the time spent in preemption off critical
138 sections, with microsecond accuracy.
140 The default measurement method is a maximum search, which is
141 disabled by default and can be runtime (re-)started
144 echo 0 > /debugfs/tracing/tracing_max_latency
146 (Note that kernel size and overhead increases with this option
147 enabled. This option and the irqs-off timing option can be
148 used together or separately.)
150 config SYSPROF_TRACER
151 bool "Sysprof Tracer"
154 select CONTEXT_SWITCH_TRACER
156 This tracer provides the trace needed by the 'Sysprof' userspace
160 bool "Scheduling Latency Tracer"
162 select CONTEXT_SWITCH_TRACER
163 select TRACER_MAX_TRACE
165 This tracer tracks the latency of the highest priority task
166 to be scheduled in, starting from the point it has woken up.
168 config CONTEXT_SWITCH_TRACER
169 bool "Trace process context switches"
173 This tracer gets called from the context switch and records
174 all switching of tasks.
177 bool "Trace various events in the kernel"
180 This tracer hooks to various trace points in the kernel
181 allowing the user to pick and choose which trace point they
184 config FTRACE_SYSCALLS
185 bool "Trace syscalls"
186 depends on HAVE_FTRACE_SYSCALLS
190 Basic tracer to catch the syscall entry and exit events.
193 bool "Trace boot initcalls"
195 select CONTEXT_SWITCH_TRACER
197 This tracer helps developers to optimize boot times: it records
198 the timings of the initcalls and traces key events and the identity
199 of tasks that can cause boot delays, such as context-switches.
201 Its aim is to be parsed by the /scripts/bootgraph.pl tool to
202 produce pretty graphics about boot inefficiencies, giving a visual
203 representation of the delays during initcalls - but the raw
204 /debug/tracing/trace text output is readable too.
206 You must pass in ftrace=initcall to the kernel command line
207 to enable this on bootup.
209 config TRACE_BRANCH_PROFILING
210 bool "Trace likely/unlikely profiler"
213 This tracer profiles all the the likely and unlikely macros
214 in the kernel. It will display the results in:
216 /debugfs/tracing/profile_annotated_branch
218 Note: this will add a significant overhead, only turn this
219 on if you need to profile the system's use of these macros.
223 config PROFILE_ALL_BRANCHES
224 bool "Profile all if conditionals"
225 depends on TRACE_BRANCH_PROFILING
227 This tracer profiles all branch conditions. Every if ()
228 taken in the kernel is recorded whether it hit or miss.
229 The results will be displayed in:
231 /debugfs/tracing/profile_branch
233 This configuration, when enabled, will impose a great overhead
234 on the system. This should only be enabled when the system
239 config TRACING_BRANCHES
242 Selected by tracers that will trace the likely and unlikely
243 conditions. This prevents the tracers themselves from being
244 profiled. Profiling the tracing infrastructure can only happen
245 when the likelys and unlikelys are not being traced.
248 bool "Trace likely/unlikely instances"
249 depends on TRACE_BRANCH_PROFILING
250 select TRACING_BRANCHES
252 This traces the events of likely and unlikely condition
253 calls in the kernel. The difference between this and the
254 "Trace likely/unlikely profiler" is that this is not a
255 histogram of the callers, but actually places the calling
256 events into a running trace buffer to see when and where the
257 events happened, as well as their results.
262 bool "Trace power consumption behavior"
266 This tracer helps developers to analyze and optimize the kernels
267 power management decisions, specifically the C-state and P-state
272 bool "Trace max stack"
273 depends on HAVE_FUNCTION_TRACER
274 select FUNCTION_TRACER
278 This special tracer records the maximum stack footprint of the
279 kernel and displays it in debugfs/tracing/stack_trace.
281 This tracer works by hooking into every function call that the
282 kernel executes, and keeping a maximum stack depth value and
283 stack-trace saved. If this is configured with DYNAMIC_FTRACE
284 then it will not have any overhead while the stack tracer
287 To enable the stack tracer on bootup, pass in 'stacktrace'
288 on the kernel command line.
290 The stack tracer can also be enabled or disabled via the
291 sysctl kernel.stack_tracer_enabled
295 config HW_BRANCH_TRACER
296 depends on HAVE_HW_BRANCH_TRACER
297 bool "Trace hw branches"
300 This tracer records all branches on the system in a circular
301 buffer giving access to the last N branches for each cpu.
304 bool "Trace SLAB allocations"
307 kmemtrace provides tracing for slab allocator functions, such as
308 kmalloc, kfree, kmem_cache_alloc, kmem_cache_free etc.. Collected
309 data is then fed to the userspace application in order to analyse
310 allocation hotspots, internal fragmentation and so on, making it
311 possible to see how well an allocator performs, as well as debug
312 and profile kernel code.
314 This requires an userspace application to use. See
315 Documentation/vm/kmemtrace.txt for more information.
317 Saying Y will make the kernel somewhat larger and slower. However,
318 if you disable kmemtrace at run-time or boot-time, the performance
319 impact is minimal (depending on the arch the kernel is built for).
323 config WORKQUEUE_TRACER
324 bool "Trace workqueues"
327 The workqueue tracer provides some statistical informations
328 about each cpu workqueue thread such as the number of the
329 works inserted and executed since their creation. It can help
330 to evaluate the amount of work each of them have to perform.
331 For example it can help a developer to decide whether he should
332 choose a per cpu workqueue instead of a singlethreaded one.
334 config BLK_DEV_IO_TRACE
335 bool "Support for tracing block io actions"
344 Say Y here if you want to be able to trace the block layer actions
345 on a given queue. Tracing allows you to see any traffic happening
346 on a block device queue. For more information (and the userspace
347 support tools needed), fetch the blktrace tools from:
349 git://git.kernel.dk/blktrace.git
351 Tracing also is possible using the ftrace interface, e.g.:
353 echo 1 > /sys/block/sda/sda1/trace/enable
354 echo blk > /sys/kernel/debug/tracing/current_tracer
355 cat /sys/kernel/debug/tracing/trace_pipe
359 config DYNAMIC_FTRACE
360 bool "enable/disable ftrace tracepoints dynamically"
361 depends on FUNCTION_TRACER
362 depends on HAVE_DYNAMIC_FTRACE
365 This option will modify all the calls to ftrace dynamically
366 (will patch them out of the binary image and replaces them
367 with a No-Op instruction) as they are called. A table is
368 created to dynamically enable them again.
370 This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but otherwise
371 has native performance as long as no tracing is active.
373 The changes to the code are done by a kernel thread that
374 wakes up once a second and checks to see if any ftrace calls
375 were made. If so, it runs stop_machine (stops all CPUS)
376 and modifies the code to jump over the call to ftrace.
378 config FTRACE_MCOUNT_RECORD
380 depends on DYNAMIC_FTRACE
381 depends on HAVE_FTRACE_MCOUNT_RECORD
383 config FTRACE_SELFTEST
386 config FTRACE_STARTUP_TEST
387 bool "Perform a startup test on ftrace"
389 select FTRACE_SELFTEST
391 This option performs a series of startup tests on ftrace. On bootup
392 a series of tests are made to verify that the tracer is
393 functioning properly. It will do tests on all the configured
397 bool "Memory mapped IO tracing"
398 depends on HAVE_MMIOTRACE_SUPPORT && PCI
401 Mmiotrace traces Memory Mapped I/O access and is meant for
402 debugging and reverse engineering. It is called from the ioremap
403 implementation and works via page faults. Tracing is disabled by
404 default and can be enabled at run-time.
406 See Documentation/tracers/mmiotrace.txt.
407 If you are not helping to develop drivers, say N.
409 config MMIOTRACE_TEST
410 tristate "Test module for mmiotrace"
411 depends on MMIOTRACE && m
413 This is a dumb module for testing mmiotrace. It is very dangerous
414 as it will write garbage to IO memory starting at a given address.
415 However, it should be safe to use on e.g. unused portion of VRAM.
417 Say N, unless you absolutely know what you are doing.
421 endif # TRACING_SUPPORT