tracing: handle unregistering the current tracer
[linux-2.6-block.git] / kernel / trace / Kconfig
CommitLineData
16444a8a 1#
606576ce
SR
2# Architectures that offer an FUNCTION_TRACER implementation should
3# select HAVE_FUNCTION_TRACER:
16444a8a 4#
2a3a4f66 5
8d26487f
TE
6config USER_STACKTRACE_SUPPORT
7 bool
8
2a3a4f66
FW
9config NOP_TRACER
10 bool
11
78d904b4
SR
12config HAVE_FTRACE_NMI_ENTER
13 bool
14
606576ce 15config HAVE_FUNCTION_TRACER
16444a8a 16 bool
bc0c38d1 17
fb52607a 18config HAVE_FUNCTION_GRAPH_TRACER
15e6cb36
FW
19 bool
20
60a7ecf4
SR
21config HAVE_FUNCTION_TRACE_MCOUNT_TEST
22 bool
23 help
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.
27
677aa9f7
SR
28config HAVE_DYNAMIC_FTRACE
29 bool
30
8da3821b
SR
31config HAVE_FTRACE_MCOUNT_RECORD
32 bool
33
1e9b51c2
MM
34config HAVE_HW_BRANCH_TRACER
35 bool
36
352ad25a
SR
37config TRACER_MAX_TRACE
38 bool
39
7a8e76a3
SR
40config RING_BUFFER
41 bool
42
78d904b4
SR
43config FTRACE_NMI_ENTER
44 bool
45 depends on HAVE_FTRACE_NMI_ENTER
46 default y
47
bc0c38d1
SR
48config TRACING
49 bool
50 select DEBUG_FS
7a8e76a3 51 select RING_BUFFER
c2c80529 52 select STACKTRACE if STACKTRACE_SUPPORT
5f87f112 53 select TRACEPOINTS
f3384b28 54 select NOP_TRACER
bc0c38d1 55
17d80fd0
PZ
56menu "Tracers"
57
606576ce 58config FUNCTION_TRACER
1b29b018 59 bool "Kernel Function Tracer"
606576ce 60 depends on HAVE_FUNCTION_TRACER
d3ee6d99 61 depends on DEBUG_KERNEL
1b29b018
SR
62 select FRAME_POINTER
63 select TRACING
35e8e302 64 select CONTEXT_SWITCH_TRACER
1b29b018
SR
65 help
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.
35e8e302 73
fb52607a
FW
74config FUNCTION_GRAPH_TRACER
75 bool "Kernel Function Graph Tracer"
76 depends on HAVE_FUNCTION_GRAPH_TRACER
15e6cb36 77 depends on FUNCTION_TRACER
764f3b95 78 default y
15e6cb36 79 help
fb52607a
FW
80 Enable the kernel to trace a function at both its return
81 and its entry.
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
84 the return value.
85 This is done by setting the current return address on the current
86 task structure into a stack of calls.
15e6cb36 87
81d68a96
SR
88config IRQSOFF_TRACER
89 bool "Interrupts-off Latency Tracer"
90 default n
91 depends on TRACE_IRQFLAGS_SUPPORT
92 depends on GENERIC_TIME
d3ee6d99 93 depends on DEBUG_KERNEL
81d68a96
SR
94 select TRACE_IRQFLAGS
95 select TRACING
96 select TRACER_MAX_TRACE
97 help
98 This option measures the time spent in irqs-off critical
99 sections, with microsecond accuracy.
100
101 The default measurement method is a maximum search, which is
102 disabled by default and can be runtime (re-)started
103 via:
104
105 echo 0 > /debugfs/tracing/tracing_max_latency
106
6cd8a4bb
SR
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.)
110
111config PREEMPT_TRACER
112 bool "Preemption-off Latency Tracer"
113 default n
114 depends on GENERIC_TIME
115 depends on PREEMPT
d3ee6d99 116 depends on DEBUG_KERNEL
6cd8a4bb
SR
117 select TRACING
118 select TRACER_MAX_TRACE
119 help
120 This option measures the time spent in preemption off critical
121 sections, with microsecond accuracy.
122
123 The default measurement method is a maximum search, which is
124 disabled by default and can be runtime (re-)started
125 via:
126
127 echo 0 > /debugfs/tracing/tracing_max_latency
128
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.)
132
f06c3810
IM
133config SYSPROF_TRACER
134 bool "Sysprof Tracer"
4d2df795 135 depends on X86
f06c3810
IM
136 select TRACING
137 help
138 This tracer provides the trace needed by the 'Sysprof' userspace
139 tool.
140
352ad25a
SR
141config SCHED_TRACER
142 bool "Scheduling Latency Tracer"
d3ee6d99 143 depends on DEBUG_KERNEL
352ad25a
SR
144 select TRACING
145 select CONTEXT_SWITCH_TRACER
146 select TRACER_MAX_TRACE
147 help
148 This tracer tracks the latency of the highest priority task
149 to be scheduled in, starting from the point it has woken up.
150
35e8e302
SR
151config CONTEXT_SWITCH_TRACER
152 bool "Trace process context switches"
d3ee6d99 153 depends on DEBUG_KERNEL
35e8e302
SR
154 select TRACING
155 select MARKERS
156 help
157 This tracer gets called from the context switch and records
158 all switching of tasks.
159
1f5c2abb
FW
160config BOOT_TRACER
161 bool "Trace boot initcalls"
1f5c2abb
FW
162 depends on DEBUG_KERNEL
163 select TRACING
ea31e72d 164 select CONTEXT_SWITCH_TRACER
1f5c2abb
FW
165 help
166 This tracer helps developers to optimize boot times: it records
98d9c66a
IM
167 the timings of the initcalls and traces key events and the identity
168 of tasks that can cause boot delays, such as context-switches.
169
170 Its aim is to be parsed by the /scripts/bootgraph.pl tool to
171 produce pretty graphics about boot inefficiencies, giving a visual
172 representation of the delays during initcalls - but the raw
173 /debug/tracing/trace text output is readable too.
174
79fb0768
SR
175 You must pass in ftrace=initcall to the kernel command line
176 to enable this on bootup.
1f5c2abb 177
2ed84eeb 178config TRACE_BRANCH_PROFILING
1f0d69a9
SR
179 bool "Trace likely/unlikely profiler"
180 depends on DEBUG_KERNEL
181 select TRACING
182 help
183 This tracer profiles all the the likely and unlikely macros
184 in the kernel. It will display the results in:
185
45b79749 186 /debugfs/tracing/profile_annotated_branch
1f0d69a9
SR
187
188 Note: this will add a significant overhead, only turn this
189 on if you need to profile the system's use of these macros.
190
191 Say N if unsure.
192
2bcd521a
SR
193config PROFILE_ALL_BRANCHES
194 bool "Profile all if conditionals"
195 depends on TRACE_BRANCH_PROFILING
196 help
197 This tracer profiles all branch conditions. Every if ()
198 taken in the kernel is recorded whether it hit or miss.
199 The results will be displayed in:
200
201 /debugfs/tracing/profile_branch
202
203 This configuration, when enabled, will impose a great overhead
204 on the system. This should only be enabled when the system
205 is to be analyzed
206
207 Say N if unsure.
208
2ed84eeb 209config TRACING_BRANCHES
52f232cb
SR
210 bool
211 help
212 Selected by tracers that will trace the likely and unlikely
213 conditions. This prevents the tracers themselves from being
214 profiled. Profiling the tracing infrastructure can only happen
215 when the likelys and unlikelys are not being traced.
216
2ed84eeb 217config BRANCH_TRACER
52f232cb 218 bool "Trace likely/unlikely instances"
2ed84eeb
SR
219 depends on TRACE_BRANCH_PROFILING
220 select TRACING_BRANCHES
52f232cb
SR
221 help
222 This traces the events of likely and unlikely condition
223 calls in the kernel. The difference between this and the
224 "Trace likely/unlikely profiler" is that this is not a
225 histogram of the callers, but actually places the calling
226 events into a running trace buffer to see when and where the
227 events happened, as well as their results.
228
229 Say N if unsure.
230
f3f47a67
AV
231config POWER_TRACER
232 bool "Trace power consumption behavior"
233 depends on DEBUG_KERNEL
234 depends on X86
235 select TRACING
236 help
237 This tracer helps developers to analyze and optimize the kernels
238 power management decisions, specifically the C-state and P-state
239 behavior.
240
241
e5a81b62
SR
242config STACK_TRACER
243 bool "Trace max stack"
606576ce 244 depends on HAVE_FUNCTION_TRACER
2ff01c6a 245 depends on DEBUG_KERNEL
606576ce 246 select FUNCTION_TRACER
e5a81b62
SR
247 select STACKTRACE
248 help
4519d9e5
IM
249 This special tracer records the maximum stack footprint of the
250 kernel and displays it in debugfs/tracing/stack_trace.
251
252 This tracer works by hooking into every function call that the
253 kernel executes, and keeping a maximum stack depth value and
f38f1d2a
SR
254 stack-trace saved. If this is configured with DYNAMIC_FTRACE
255 then it will not have any overhead while the stack tracer
256 is disabled.
257
258 To enable the stack tracer on bootup, pass in 'stacktrace'
259 on the kernel command line.
260
261 The stack tracer can also be enabled or disabled via the
262 sysctl kernel.stack_tracer_enabled
4519d9e5
IM
263
264 Say N if unsure.
e5a81b62 265
a93751ca 266config HW_BRANCH_TRACER
1e9b51c2 267 depends on HAVE_HW_BRANCH_TRACER
a93751ca 268 bool "Trace hw branches"
1e9b51c2
MM
269 select TRACING
270 help
271 This tracer records all branches on the system in a circular
272 buffer giving access to the last N branches for each cpu.
273
36994e58
FW
274config KMEMTRACE
275 bool "Trace SLAB allocations"
276 select TRACING
36994e58
FW
277 help
278 kmemtrace provides tracing for slab allocator functions, such as
279 kmalloc, kfree, kmem_cache_alloc, kmem_cache_free etc.. Collected
280 data is then fed to the userspace application in order to analyse
281 allocation hotspots, internal fragmentation and so on, making it
282 possible to see how well an allocator performs, as well as debug
283 and profile kernel code.
284
285 This requires an userspace application to use. See
286 Documentation/vm/kmemtrace.txt for more information.
287
288 Saying Y will make the kernel somewhat larger and slower. However,
289 if you disable kmemtrace at run-time or boot-time, the performance
290 impact is minimal (depending on the arch the kernel is built for).
291
292 If unsure, say N.
293
e1d8aa9f
FW
294config WORKQUEUE_TRACER
295 bool "Trace workqueues"
296 select TRACING
297 help
298 The workqueue tracer provides some statistical informations
299 about each cpu workqueue thread such as the number of the
300 works inserted and executed since their creation. It can help
301 to evaluate the amount of work each of them have to perform.
302 For example it can help a developer to decide whether he should
303 choose a per cpu workqueue instead of a singlethreaded one.
304
2db270a8
FW
305config BLK_DEV_IO_TRACE
306 bool "Support for tracing block io actions"
307 depends on SYSFS
308 select RELAY
309 select DEBUG_FS
310 select TRACEPOINTS
311 select TRACING
312 select STACKTRACE
313 help
314 Say Y here if you want to be able to trace the block layer actions
315 on a given queue. Tracing allows you to see any traffic happening
316 on a block device queue. For more information (and the userspace
317 support tools needed), fetch the blktrace tools from:
318
319 git://git.kernel.dk/blktrace.git
320
321 Tracing also is possible using the ftrace interface, e.g.:
322
323 echo 1 > /sys/block/sda/sda1/trace/enable
324 echo blk > /sys/kernel/debug/tracing/current_tracer
325 cat /sys/kernel/debug/tracing/trace_pipe
326
327 If unsure, say N.
36994e58 328
3d083395
SR
329config DYNAMIC_FTRACE
330 bool "enable/disable ftrace tracepoints dynamically"
606576ce 331 depends on FUNCTION_TRACER
677aa9f7 332 depends on HAVE_DYNAMIC_FTRACE
d3ee6d99 333 depends on DEBUG_KERNEL
3d083395
SR
334 default y
335 help
336 This option will modify all the calls to ftrace dynamically
337 (will patch them out of the binary image and replaces them
338 with a No-Op instruction) as they are called. A table is
339 created to dynamically enable them again.
340
606576ce 341 This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but otherwise
3d083395
SR
342 has native performance as long as no tracing is active.
343
344 The changes to the code are done by a kernel thread that
345 wakes up once a second and checks to see if any ftrace calls
346 were made. If so, it runs stop_machine (stops all CPUS)
347 and modifies the code to jump over the call to ftrace.
60a11774 348
8da3821b
SR
349config FTRACE_MCOUNT_RECORD
350 def_bool y
351 depends on DYNAMIC_FTRACE
352 depends on HAVE_FTRACE_MCOUNT_RECORD
353
60a11774
SR
354config FTRACE_SELFTEST
355 bool
356
357config FTRACE_STARTUP_TEST
358 bool "Perform a startup test on ftrace"
79fb0768 359 depends on TRACING && DEBUG_KERNEL
60a11774
SR
360 select FTRACE_SELFTEST
361 help
362 This option performs a series of startup tests on ftrace. On bootup
363 a series of tests are made to verify that the tracer is
364 functioning properly. It will do tests on all the configured
365 tracers of ftrace.
17d80fd0 366
fe6f90e5
PP
367config MMIOTRACE
368 bool "Memory mapped IO tracing"
369 depends on HAVE_MMIOTRACE_SUPPORT && DEBUG_KERNEL && PCI
370 select TRACING
371 help
372 Mmiotrace traces Memory Mapped I/O access and is meant for
373 debugging and reverse engineering. It is called from the ioremap
374 implementation and works via page faults. Tracing is disabled by
375 default and can be enabled at run-time.
376
377 See Documentation/tracers/mmiotrace.txt.
378 If you are not helping to develop drivers, say N.
379
380config MMIOTRACE_TEST
381 tristate "Test module for mmiotrace"
382 depends on MMIOTRACE && m
383 help
384 This is a dumb module for testing mmiotrace. It is very dangerous
385 as it will write garbage to IO memory starting at a given address.
386 However, it should be safe to use on e.g. unused portion of VRAM.
387
388 Say N, unless you absolutely know what you are doing.
389
17d80fd0 390endmenu