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
58 config FUNCTION_TRACER
59 bool "Kernel Function Tracer"
60 depends on HAVE_FUNCTION_TRACER
61 depends on DEBUG_KERNEL
64 select CONTEXT_SWITCH_TRACER
66 Enable the kernel to trace every kernel function. This is done
67 by using a compiler feature to insert a small, 5-byte No-Operation
68 instruction to the beginning of every kernel function, which NOP
69 sequence is then dynamically patched into a tracer call when
70 tracing is enabled by the administrator. If it's runtime disabled
71 (the bootup default), then the overhead of the instructions is very
72 small and not measurable even in micro-benchmarks.
74 config FUNCTION_GRAPH_TRACER
75 bool "Kernel Function Graph Tracer"
76 depends on HAVE_FUNCTION_GRAPH_TRACER
77 depends on FUNCTION_TRACER
80 Enable the kernel to trace a function at both its return
82 It's first purpose is to trace the duration of functions and
83 draw a call graph for each thread with some informations like
85 This is done by setting the current return address on the current
86 task structure into a stack of calls.
89 bool "Interrupts-off Latency Tracer"
91 depends on TRACE_IRQFLAGS_SUPPORT
92 depends on GENERIC_TIME
93 depends on DEBUG_KERNEL
96 select TRACER_MAX_TRACE
98 This option measures the time spent in irqs-off critical
99 sections, with microsecond accuracy.
101 The default measurement method is a maximum search, which is
102 disabled by default and can be runtime (re-)started
105 echo 0 > /debugfs/tracing/tracing_max_latency
107 (Note that kernel size and overhead increases with this option
108 enabled. This option and the preempt-off timing option can be
109 used together or separately.)
111 config PREEMPT_TRACER
112 bool "Preemption-off Latency Tracer"
114 depends on GENERIC_TIME
116 depends on DEBUG_KERNEL
118 select TRACER_MAX_TRACE
120 This option measures the time spent in preemption off critical
121 sections, with microsecond accuracy.
123 The default measurement method is a maximum search, which is
124 disabled by default and can be runtime (re-)started
127 echo 0 > /debugfs/tracing/tracing_max_latency
129 (Note that kernel size and overhead increases with this option
130 enabled. This option and the irqs-off timing option can be
131 used together or separately.)
133 config SYSPROF_TRACER
134 bool "Sysprof Tracer"
137 select CONTEXT_SWITCH_TRACER
139 This tracer provides the trace needed by the 'Sysprof' userspace
143 bool "Scheduling Latency Tracer"
144 depends on DEBUG_KERNEL
146 select CONTEXT_SWITCH_TRACER
147 select TRACER_MAX_TRACE
149 This tracer tracks the latency of the highest priority task
150 to be scheduled in, starting from the point it has woken up.
152 config CONTEXT_SWITCH_TRACER
153 bool "Trace process context switches"
154 depends on DEBUG_KERNEL
158 This tracer gets called from the context switch and records
159 all switching of tasks.
162 bool "Trace boot initcalls"
163 depends on DEBUG_KERNEL
165 select CONTEXT_SWITCH_TRACER
167 This tracer helps developers to optimize boot times: it records
168 the timings of the initcalls and traces key events and the identity
169 of tasks that can cause boot delays, such as context-switches.
171 Its aim is to be parsed by the /scripts/bootgraph.pl tool to
172 produce pretty graphics about boot inefficiencies, giving a visual
173 representation of the delays during initcalls - but the raw
174 /debug/tracing/trace text output is readable too.
176 You must pass in ftrace=initcall to the kernel command line
177 to enable this on bootup.
179 config TRACE_BRANCH_PROFILING
180 bool "Trace likely/unlikely profiler"
181 depends on DEBUG_KERNEL
184 This tracer profiles all the the likely and unlikely macros
185 in the kernel. It will display the results in:
187 /debugfs/tracing/profile_annotated_branch
189 Note: this will add a significant overhead, only turn this
190 on if you need to profile the system's use of these macros.
194 config PROFILE_ALL_BRANCHES
195 bool "Profile all if conditionals"
196 depends on TRACE_BRANCH_PROFILING
198 This tracer profiles all branch conditions. Every if ()
199 taken in the kernel is recorded whether it hit or miss.
200 The results will be displayed in:
202 /debugfs/tracing/profile_branch
204 This configuration, when enabled, will impose a great overhead
205 on the system. This should only be enabled when the system
210 config TRACING_BRANCHES
213 Selected by tracers that will trace the likely and unlikely
214 conditions. This prevents the tracers themselves from being
215 profiled. Profiling the tracing infrastructure can only happen
216 when the likelys and unlikelys are not being traced.
219 bool "Trace likely/unlikely instances"
220 depends on TRACE_BRANCH_PROFILING
221 select TRACING_BRANCHES
223 This traces the events of likely and unlikely condition
224 calls in the kernel. The difference between this and the
225 "Trace likely/unlikely profiler" is that this is not a
226 histogram of the callers, but actually places the calling
227 events into a running trace buffer to see when and where the
228 events happened, as well as their results.
233 bool "Trace power consumption behavior"
234 depends on DEBUG_KERNEL
238 This tracer helps developers to analyze and optimize the kernels
239 power management decisions, specifically the C-state and P-state
244 bool "Trace max stack"
245 depends on HAVE_FUNCTION_TRACER
246 depends on DEBUG_KERNEL
247 select FUNCTION_TRACER
250 This special tracer records the maximum stack footprint of the
251 kernel and displays it in debugfs/tracing/stack_trace.
253 This tracer works by hooking into every function call that the
254 kernel executes, and keeping a maximum stack depth value and
255 stack-trace saved. If this is configured with DYNAMIC_FTRACE
256 then it will not have any overhead while the stack tracer
259 To enable the stack tracer on bootup, pass in 'stacktrace'
260 on the kernel command line.
262 The stack tracer can also be enabled or disabled via the
263 sysctl kernel.stack_tracer_enabled
267 config HW_BRANCH_TRACER
268 depends on HAVE_HW_BRANCH_TRACER
269 bool "Trace hw branches"
272 This tracer records all branches on the system in a circular
273 buffer giving access to the last N branches for each cpu.
276 bool "Trace SLAB allocations"
279 kmemtrace provides tracing for slab allocator functions, such as
280 kmalloc, kfree, kmem_cache_alloc, kmem_cache_free etc.. Collected
281 data is then fed to the userspace application in order to analyse
282 allocation hotspots, internal fragmentation and so on, making it
283 possible to see how well an allocator performs, as well as debug
284 and profile kernel code.
286 This requires an userspace application to use. See
287 Documentation/vm/kmemtrace.txt for more information.
289 Saying Y will make the kernel somewhat larger and slower. However,
290 if you disable kmemtrace at run-time or boot-time, the performance
291 impact is minimal (depending on the arch the kernel is built for).
295 config WORKQUEUE_TRACER
296 bool "Trace workqueues"
299 The workqueue tracer provides some statistical informations
300 about each cpu workqueue thread such as the number of the
301 works inserted and executed since their creation. It can help
302 to evaluate the amount of work each of them have to perform.
303 For example it can help a developer to decide whether he should
304 choose a per cpu workqueue instead of a singlethreaded one.
306 config BLK_DEV_IO_TRACE
307 bool "Support for tracing block io actions"
316 Say Y here if you want to be able to trace the block layer actions
317 on a given queue. Tracing allows you to see any traffic happening
318 on a block device queue. For more information (and the userspace
319 support tools needed), fetch the blktrace tools from:
321 git://git.kernel.dk/blktrace.git
323 Tracing also is possible using the ftrace interface, e.g.:
325 echo 1 > /sys/block/sda/sda1/trace/enable
326 echo blk > /sys/kernel/debug/tracing/current_tracer
327 cat /sys/kernel/debug/tracing/trace_pipe
331 config DYNAMIC_FTRACE
332 bool "enable/disable ftrace tracepoints dynamically"
333 depends on FUNCTION_TRACER
334 depends on HAVE_DYNAMIC_FTRACE
335 depends on DEBUG_KERNEL
338 This option will modify all the calls to ftrace dynamically
339 (will patch them out of the binary image and replaces them
340 with a No-Op instruction) as they are called. A table is
341 created to dynamically enable them again.
343 This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but otherwise
344 has native performance as long as no tracing is active.
346 The changes to the code are done by a kernel thread that
347 wakes up once a second and checks to see if any ftrace calls
348 were made. If so, it runs stop_machine (stops all CPUS)
349 and modifies the code to jump over the call to ftrace.
351 config FTRACE_MCOUNT_RECORD
353 depends on DYNAMIC_FTRACE
354 depends on HAVE_FTRACE_MCOUNT_RECORD
356 config FTRACE_SELFTEST
359 config FTRACE_STARTUP_TEST
360 bool "Perform a startup test on ftrace"
361 depends on TRACING && DEBUG_KERNEL
362 select FTRACE_SELFTEST
364 This option performs a series of startup tests on ftrace. On bootup
365 a series of tests are made to verify that the tracer is
366 functioning properly. It will do tests on all the configured
370 bool "Memory mapped IO tracing"
371 depends on HAVE_MMIOTRACE_SUPPORT && DEBUG_KERNEL && PCI
374 Mmiotrace traces Memory Mapped I/O access and is meant for
375 debugging and reverse engineering. It is called from the ioremap
376 implementation and works via page faults. Tracing is disabled by
377 default and can be enabled at run-time.
379 See Documentation/tracers/mmiotrace.txt.
380 If you are not helping to develop drivers, say N.
382 config MMIOTRACE_TEST
383 tristate "Test module for mmiotrace"
384 depends on MMIOTRACE && m
386 This is a dumb module for testing mmiotrace. It is very dangerous
387 as it will write garbage to IO memory starting at a given address.
388 However, it should be safe to use on e.g. unused portion of VRAM.
390 Say N, unless you absolutely know what you are doing.