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 TRACER_MAX_TRACE
43 config FTRACE_NMI_ENTER
45 depends on HAVE_FTRACE_NMI_ENTER
52 select STACKTRACE if STACKTRACE_SUPPORT
57 # Minimum requirements an architecture has to meet for us to
58 # be able to offer generic tracing facilities:
60 config TRACING_SUPPORT
62 depends on TRACE_IRQFLAGS_SUPPORT
63 depends on STACKTRACE_SUPPORT
69 config FUNCTION_TRACER
70 bool "Kernel Function Tracer"
71 depends on HAVE_FUNCTION_TRACER
75 select CONTEXT_SWITCH_TRACER
77 Enable the kernel to trace every kernel function. This is done
78 by using a compiler feature to insert a small, 5-byte No-Operation
79 instruction to the beginning of every kernel function, which NOP
80 sequence is then dynamically patched into a tracer call when
81 tracing is enabled by the administrator. If it's runtime disabled
82 (the bootup default), then the overhead of the instructions is very
83 small and not measurable even in micro-benchmarks.
85 config FUNCTION_GRAPH_TRACER
86 bool "Kernel Function Graph Tracer"
87 depends on HAVE_FUNCTION_GRAPH_TRACER
88 depends on FUNCTION_TRACER
91 Enable the kernel to trace a function at both its return
93 It's first purpose is to trace the duration of functions and
94 draw a call graph for each thread with some informations like
96 This is done by setting the current return address on the current
97 task structure into a stack of calls.
100 bool "Interrupts-off Latency Tracer"
102 depends on TRACE_IRQFLAGS_SUPPORT
103 depends on GENERIC_TIME
104 select TRACE_IRQFLAGS
106 select TRACER_MAX_TRACE
108 This option measures the time spent in irqs-off critical
109 sections, with microsecond accuracy.
111 The default measurement method is a maximum search, which is
112 disabled by default and can be runtime (re-)started
115 echo 0 > /debugfs/tracing/tracing_max_latency
117 (Note that kernel size and overhead increases with this option
118 enabled. This option and the preempt-off timing option can be
119 used together or separately.)
121 config PREEMPT_TRACER
122 bool "Preemption-off Latency Tracer"
124 depends on GENERIC_TIME
127 select TRACER_MAX_TRACE
129 This option measures the time spent in preemption off critical
130 sections, with microsecond accuracy.
132 The default measurement method is a maximum search, which is
133 disabled by default and can be runtime (re-)started
136 echo 0 > /debugfs/tracing/tracing_max_latency
138 (Note that kernel size and overhead increases with this option
139 enabled. This option and the irqs-off timing option can be
140 used together or separately.)
142 config SYSPROF_TRACER
143 bool "Sysprof Tracer"
146 select CONTEXT_SWITCH_TRACER
148 This tracer provides the trace needed by the 'Sysprof' userspace
152 bool "Scheduling Latency Tracer"
154 select CONTEXT_SWITCH_TRACER
155 select TRACER_MAX_TRACE
157 This tracer tracks the latency of the highest priority task
158 to be scheduled in, starting from the point it has woken up.
160 config CONTEXT_SWITCH_TRACER
161 bool "Trace process context switches"
165 This tracer gets called from the context switch and records
166 all switching of tasks.
169 bool "Trace various events in the kernel"
172 This tracer hooks to various trace points in the kernel
173 allowing the user to pick and choose which trace point they
177 bool "Trace boot initcalls"
179 select CONTEXT_SWITCH_TRACER
181 This tracer helps developers to optimize boot times: it records
182 the timings of the initcalls and traces key events and the identity
183 of tasks that can cause boot delays, such as context-switches.
185 Its aim is to be parsed by the /scripts/bootgraph.pl tool to
186 produce pretty graphics about boot inefficiencies, giving a visual
187 representation of the delays during initcalls - but the raw
188 /debug/tracing/trace text output is readable too.
190 You must pass in ftrace=initcall to the kernel command line
191 to enable this on bootup.
193 config TRACE_BRANCH_PROFILING
194 bool "Trace likely/unlikely profiler"
197 This tracer profiles all the the likely and unlikely macros
198 in the kernel. It will display the results in:
200 /debugfs/tracing/profile_annotated_branch
202 Note: this will add a significant overhead, only turn this
203 on if you need to profile the system's use of these macros.
207 config PROFILE_ALL_BRANCHES
208 bool "Profile all if conditionals"
209 depends on TRACE_BRANCH_PROFILING
211 This tracer profiles all branch conditions. Every if ()
212 taken in the kernel is recorded whether it hit or miss.
213 The results will be displayed in:
215 /debugfs/tracing/profile_branch
217 This configuration, when enabled, will impose a great overhead
218 on the system. This should only be enabled when the system
223 config TRACING_BRANCHES
226 Selected by tracers that will trace the likely and unlikely
227 conditions. This prevents the tracers themselves from being
228 profiled. Profiling the tracing infrastructure can only happen
229 when the likelys and unlikelys are not being traced.
232 bool "Trace likely/unlikely instances"
233 depends on TRACE_BRANCH_PROFILING
234 select TRACING_BRANCHES
236 This traces the events of likely and unlikely condition
237 calls in the kernel. The difference between this and the
238 "Trace likely/unlikely profiler" is that this is not a
239 histogram of the callers, but actually places the calling
240 events into a running trace buffer to see when and where the
241 events happened, as well as their results.
246 bool "Trace power consumption behavior"
250 This tracer helps developers to analyze and optimize the kernels
251 power management decisions, specifically the C-state and P-state
256 bool "Trace max stack"
257 depends on HAVE_FUNCTION_TRACER
258 select FUNCTION_TRACER
262 This special tracer records the maximum stack footprint of the
263 kernel and displays it in debugfs/tracing/stack_trace.
265 This tracer works by hooking into every function call that the
266 kernel executes, and keeping a maximum stack depth value and
267 stack-trace saved. If this is configured with DYNAMIC_FTRACE
268 then it will not have any overhead while the stack tracer
271 To enable the stack tracer on bootup, pass in 'stacktrace'
272 on the kernel command line.
274 The stack tracer can also be enabled or disabled via the
275 sysctl kernel.stack_tracer_enabled
279 config HW_BRANCH_TRACER
280 depends on HAVE_HW_BRANCH_TRACER
281 bool "Trace hw branches"
284 This tracer records all branches on the system in a circular
285 buffer giving access to the last N branches for each cpu.
288 bool "Trace SLAB allocations"
291 kmemtrace provides tracing for slab allocator functions, such as
292 kmalloc, kfree, kmem_cache_alloc, kmem_cache_free etc.. Collected
293 data is then fed to the userspace application in order to analyse
294 allocation hotspots, internal fragmentation and so on, making it
295 possible to see how well an allocator performs, as well as debug
296 and profile kernel code.
298 This requires an userspace application to use. See
299 Documentation/vm/kmemtrace.txt for more information.
301 Saying Y will make the kernel somewhat larger and slower. However,
302 if you disable kmemtrace at run-time or boot-time, the performance
303 impact is minimal (depending on the arch the kernel is built for).
307 config WORKQUEUE_TRACER
308 bool "Trace workqueues"
311 The workqueue tracer provides some statistical informations
312 about each cpu workqueue thread such as the number of the
313 works inserted and executed since their creation. It can help
314 to evaluate the amount of work each of them have to perform.
315 For example it can help a developer to decide whether he should
316 choose a per cpu workqueue instead of a singlethreaded one.
318 config BLK_DEV_IO_TRACE
319 bool "Support for tracing block io actions"
328 Say Y here if you want to be able to trace the block layer actions
329 on a given queue. Tracing allows you to see any traffic happening
330 on a block device queue. For more information (and the userspace
331 support tools needed), fetch the blktrace tools from:
333 git://git.kernel.dk/blktrace.git
335 Tracing also is possible using the ftrace interface, e.g.:
337 echo 1 > /sys/block/sda/sda1/trace/enable
338 echo blk > /sys/kernel/debug/tracing/current_tracer
339 cat /sys/kernel/debug/tracing/trace_pipe
343 config DYNAMIC_FTRACE
344 bool "enable/disable ftrace tracepoints dynamically"
345 depends on FUNCTION_TRACER
346 depends on HAVE_DYNAMIC_FTRACE
349 This option will modify all the calls to ftrace dynamically
350 (will patch them out of the binary image and replaces them
351 with a No-Op instruction) as they are called. A table is
352 created to dynamically enable them again.
354 This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but otherwise
355 has native performance as long as no tracing is active.
357 The changes to the code are done by a kernel thread that
358 wakes up once a second and checks to see if any ftrace calls
359 were made. If so, it runs stop_machine (stops all CPUS)
360 and modifies the code to jump over the call to ftrace.
362 config FTRACE_MCOUNT_RECORD
364 depends on DYNAMIC_FTRACE
365 depends on HAVE_FTRACE_MCOUNT_RECORD
367 config FTRACE_SELFTEST
370 config FTRACE_STARTUP_TEST
371 bool "Perform a startup test on ftrace"
373 select FTRACE_SELFTEST
375 This option performs a series of startup tests on ftrace. On bootup
376 a series of tests are made to verify that the tracer is
377 functioning properly. It will do tests on all the configured
381 bool "Memory mapped IO tracing"
382 depends on HAVE_MMIOTRACE_SUPPORT && PCI
385 Mmiotrace traces Memory Mapped I/O access and is meant for
386 debugging and reverse engineering. It is called from the ioremap
387 implementation and works via page faults. Tracing is disabled by
388 default and can be enabled at run-time.
390 See Documentation/tracers/mmiotrace.txt.
391 If you are not helping to develop drivers, say N.
393 config MMIOTRACE_TEST
394 tristate "Test module for mmiotrace"
395 depends on MMIOTRACE && m
397 This is a dumb module for testing mmiotrace. It is very dangerous
398 as it will write garbage to IO memory starting at a given address.
399 However, it should be safe to use on e.g. unused portion of VRAM.
401 Say N, unless you absolutely know what you are doing.
405 endif # TRACING_SUPPORT