blob: 417d1985e29911784adbf506523462f009933c02 [file] [log] [blame]
Arnaldo Carvalho de Melo16444a82008-05-12 21:20:42 +02001#
Steven Rostedt606576c2008-10-06 19:06:12 -04002# Architectures that offer an FUNCTION_TRACER implementation should
3# select HAVE_FUNCTION_TRACER:
Arnaldo Carvalho de Melo16444a82008-05-12 21:20:42 +02004#
Frédéric Weisbecker2a3a4f62008-09-21 20:12:14 +02005
Török Edwin8d264872008-11-23 12:39:08 +02006config USER_STACKTRACE_SUPPORT
7 bool
8
Frédéric Weisbecker2a3a4f62008-09-21 20:12:14 +02009config NOP_TRACER
10 bool
11
Steven Rostedt78d904b2009-02-05 18:43:07 -050012config HAVE_FTRACE_NMI_ENTER
13 bool
14
Steven Rostedt606576c2008-10-06 19:06:12 -040015config HAVE_FUNCTION_TRACER
Arnaldo Carvalho de Melo16444a82008-05-12 21:20:42 +020016 bool
Steven Rostedtbc0c38d2008-05-12 21:20:42 +020017
Frederic Weisbeckerfb526072008-11-25 21:07:04 +010018config HAVE_FUNCTION_GRAPH_TRACER
Frederic Weisbecker15e6cb32008-11-11 07:14:25 +010019 bool
20
Steven Rostedt60a7ecf2008-11-05 16:05:44 -050021config 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
Steven Rostedt677aa9f2008-05-17 00:01:36 -040028config HAVE_DYNAMIC_FTRACE
29 bool
30
Steven Rostedt8da38212008-08-14 15:45:07 -040031config HAVE_FTRACE_MCOUNT_RECORD
32 bool
33
Markus Metzger1e9b51c2008-11-25 09:24:15 +010034config HAVE_HW_BRANCH_TRACER
35 bool
36
Frederic Weisbeckeree08c6e2009-03-07 05:52:59 +010037config HAVE_FTRACE_SYSCALLS
38 bool
39
Steven Rostedt352ad252008-05-12 21:20:42 +020040config TRACER_MAX_TRACE
41 bool
42
Steven Rostedt7a8e76a2008-09-29 23:02:38 -040043config RING_BUFFER
44 bool
45
Steven Rostedt78d904b2009-02-05 18:43:07 -050046config FTRACE_NMI_ENTER
47 bool
48 depends on HAVE_FTRACE_NMI_ENTER
49 default y
50
Steven Rostedtbc0c38d2008-05-12 21:20:42 +020051config TRACING
52 bool
53 select DEBUG_FS
Steven Rostedt7a8e76a2008-09-29 23:02:38 -040054 select RING_BUFFER
Al Viroc2c80522008-10-31 19:50:41 +000055 select STACKTRACE if STACKTRACE_SUPPORT
Ingo Molnar5f87f112008-07-23 14:15:22 +020056 select TRACEPOINTS
Steven Rostedtf3384b22008-10-29 11:15:57 -040057 select NOP_TRACER
Frederic Weisbecker769b0442009-03-06 17:21:49 +010058 select BINARY_PRINTF
Steven Rostedtbc0c38d2008-05-12 21:20:42 +020059
Ingo Molnar40ada302009-03-05 21:19:55 +010060#
61# Minimum requirements an architecture has to meet for us to
62# be able to offer generic tracing facilities:
63#
64config TRACING_SUPPORT
65 bool
Anton Vorontsov45b95602009-03-24 01:07:24 +030066 # 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
Ingo Molnar40ada302009-03-05 21:19:55 +010071 depends on STACKTRACE_SUPPORT
KOSAKI Motohiro422d3c72009-03-06 10:40:53 +090072 default y
Ingo Molnar40ada302009-03-05 21:19:55 +010073
74if TRACING_SUPPORT
Arnaldo Carvalho de Melo16444a82008-05-12 21:20:42 +020075
Peter Zijlstra17d80fd2008-10-21 16:31:18 +020076menu "Tracers"
77
Steven Rostedt606576c2008-10-06 19:06:12 -040078config FUNCTION_TRACER
Steven Rostedt1b29b012008-05-12 21:20:42 +020079 bool "Kernel Function Tracer"
Steven Rostedt606576c2008-10-06 19:06:12 -040080 depends on HAVE_FUNCTION_TRACER
Steven Rostedt1b29b012008-05-12 21:20:42 +020081 select FRAME_POINTER
Steven Rostedt4d7a0772009-02-18 22:06:18 -050082 select KALLSYMS
Steven Rostedt1b29b012008-05-12 21:20:42 +020083 select TRACING
Steven Rostedt35e8e302008-05-12 21:20:42 +020084 select CONTEXT_SWITCH_TRACER
Steven Rostedt1b29b012008-05-12 21:20:42 +020085 help
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.
Steven Rostedt35e8e302008-05-12 21:20:42 +020093
Frederic Weisbeckerfb526072008-11-25 21:07:04 +010094config FUNCTION_GRAPH_TRACER
95 bool "Kernel Function Graph Tracer"
96 depends on HAVE_FUNCTION_GRAPH_TRACER
Frederic Weisbecker15e6cb32008-11-11 07:14:25 +010097 depends on FUNCTION_TRACER
Ingo Molnar764f3b92008-12-03 10:33:58 +010098 default y
Frederic Weisbecker15e6cb32008-11-11 07:14:25 +010099 help
Frederic Weisbeckerfb526072008-11-25 21:07:04 +0100100 Enable the kernel to trace a function at both its return
101 and its entry.
Matt LaPlante692105b2009-01-26 11:12:25 +0100102 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.
Frederic Weisbecker15e6cb32008-11-11 07:14:25 +0100106
Steven Rostedt81d68a92008-05-12 21:20:42 +0200107config IRQSOFF_TRACER
108 bool "Interrupts-off Latency Tracer"
109 default n
110 depends on TRACE_IRQFLAGS_SUPPORT
111 depends on GENERIC_TIME
112 select TRACE_IRQFLAGS
113 select TRACING
114 select TRACER_MAX_TRACE
115 help
116 This option measures the time spent in irqs-off critical
117 sections, with microsecond accuracy.
118
119 The default measurement method is a maximum search, which is
120 disabled by default and can be runtime (re-)started
121 via:
122
123 echo 0 > /debugfs/tracing/tracing_max_latency
124
Steven Rostedt6cd8a4b2008-05-12 21:20:42 +0200125 (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.)
128
129config PREEMPT_TRACER
130 bool "Preemption-off Latency Tracer"
131 default n
132 depends on GENERIC_TIME
133 depends on PREEMPT
134 select TRACING
135 select TRACER_MAX_TRACE
136 help
137 This option measures the time spent in preemption off critical
138 sections, with microsecond accuracy.
139
140 The default measurement method is a maximum search, which is
141 disabled by default and can be runtime (re-)started
142 via:
143
144 echo 0 > /debugfs/tracing/tracing_max_latency
145
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.)
149
Ingo Molnarf06c3812008-05-12 21:20:47 +0200150config SYSPROF_TRACER
151 bool "Sysprof Tracer"
Thomas Gleixner4d2df792008-05-24 15:00:46 +0200152 depends on X86
Ingo Molnarf06c3812008-05-12 21:20:47 +0200153 select TRACING
Frederic Weisbeckerb22f4852009-02-10 15:49:11 +0100154 select CONTEXT_SWITCH_TRACER
Ingo Molnarf06c3812008-05-12 21:20:47 +0200155 help
156 This tracer provides the trace needed by the 'Sysprof' userspace
157 tool.
158
Steven Rostedt352ad252008-05-12 21:20:42 +0200159config SCHED_TRACER
160 bool "Scheduling Latency Tracer"
Steven Rostedt352ad252008-05-12 21:20:42 +0200161 select TRACING
162 select CONTEXT_SWITCH_TRACER
163 select TRACER_MAX_TRACE
164 help
165 This tracer tracks the latency of the highest priority task
166 to be scheduled in, starting from the point it has woken up.
167
Steven Rostedt35e8e302008-05-12 21:20:42 +0200168config CONTEXT_SWITCH_TRACER
169 bool "Trace process context switches"
Steven Rostedt35e8e302008-05-12 21:20:42 +0200170 select TRACING
171 select MARKERS
172 help
173 This tracer gets called from the context switch and records
174 all switching of tasks.
175
Steven Rostedtb77e38a2009-02-24 10:21:36 -0500176config EVENT_TRACER
177 bool "Trace various events in the kernel"
Steven Rostedtb77e38a2009-02-24 10:21:36 -0500178 select TRACING
179 help
180 This tracer hooks to various trace points in the kernel
181 allowing the user to pick and choose which trace point they
182 want to trace.
183
Frederic Weisbeckeree08c6e2009-03-07 05:52:59 +0100184config FTRACE_SYSCALLS
185 bool "Trace syscalls"
186 depends on HAVE_FTRACE_SYSCALLS
187 select TRACING
Frederic Weisbecker0ea1c412009-03-15 22:10:38 +0100188 select KALLSYMS
Frederic Weisbeckeree08c6e2009-03-07 05:52:59 +0100189 help
190 Basic tracer to catch the syscall entry and exit events.
191
Frédéric Weisbecker1f5c2ab2008-09-23 11:36:20 +0100192config BOOT_TRACER
193 bool "Trace boot initcalls"
Frédéric Weisbecker1f5c2ab2008-09-23 11:36:20 +0100194 select TRACING
Frederic Weisbeckerea31e722008-10-22 19:26:23 +0200195 select CONTEXT_SWITCH_TRACER
Frédéric Weisbecker1f5c2ab2008-09-23 11:36:20 +0100196 help
197 This tracer helps developers to optimize boot times: it records
Ingo Molnar98d9c662008-10-14 14:27:20 +0200198 the timings of the initcalls and traces key events and the identity
199 of tasks that can cause boot delays, such as context-switches.
200
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.
205
Steven Rostedt79fb0762009-02-02 21:38:33 -0500206 You must pass in ftrace=initcall to the kernel command line
207 to enable this on bootup.
Frédéric Weisbecker1f5c2ab2008-09-23 11:36:20 +0100208
Steven Rostedt2ed84ee2008-11-12 15:24:24 -0500209config TRACE_BRANCH_PROFILING
Steven Rostedt1f0d69a2008-11-12 00:14:39 -0500210 bool "Trace likely/unlikely profiler"
Steven Rostedt1f0d69a2008-11-12 00:14:39 -0500211 select TRACING
212 help
213 This tracer profiles all the the likely and unlikely macros
214 in the kernel. It will display the results in:
215
Steven Rostedt45b79742008-11-21 00:40:40 -0500216 /debugfs/tracing/profile_annotated_branch
Steven Rostedt1f0d69a2008-11-12 00:14:39 -0500217
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.
220
221 Say N if unsure.
222
Steven Rostedt2bcd5212008-11-21 01:30:54 -0500223config PROFILE_ALL_BRANCHES
224 bool "Profile all if conditionals"
225 depends on TRACE_BRANCH_PROFILING
226 help
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:
230
231 /debugfs/tracing/profile_branch
232
233 This configuration, when enabled, will impose a great overhead
234 on the system. This should only be enabled when the system
235 is to be analyzed
236
237 Say N if unsure.
238
Steven Rostedt2ed84ee2008-11-12 15:24:24 -0500239config TRACING_BRANCHES
Steven Rostedt52f232c2008-11-12 00:14:40 -0500240 bool
241 help
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.
246
Steven Rostedt2ed84ee2008-11-12 15:24:24 -0500247config BRANCH_TRACER
Steven Rostedt52f232c2008-11-12 00:14:40 -0500248 bool "Trace likely/unlikely instances"
Steven Rostedt2ed84ee2008-11-12 15:24:24 -0500249 depends on TRACE_BRANCH_PROFILING
250 select TRACING_BRANCHES
Steven Rostedt52f232c2008-11-12 00:14:40 -0500251 help
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.
258
259 Say N if unsure.
260
Arjan van de Venf3f47a62008-11-23 16:49:58 -0800261config POWER_TRACER
262 bool "Trace power consumption behavior"
Arjan van de Venf3f47a62008-11-23 16:49:58 -0800263 depends on X86
264 select TRACING
265 help
266 This tracer helps developers to analyze and optimize the kernels
267 power management decisions, specifically the C-state and P-state
268 behavior.
269
270
Steven Rostedte5a81b62008-08-27 23:31:01 -0400271config STACK_TRACER
272 bool "Trace max stack"
Steven Rostedt606576c2008-10-06 19:06:12 -0400273 depends on HAVE_FUNCTION_TRACER
Steven Rostedt606576c2008-10-06 19:06:12 -0400274 select FUNCTION_TRACER
Steven Rostedte5a81b62008-08-27 23:31:01 -0400275 select STACKTRACE
Steven Rostedt4d7a0772009-02-18 22:06:18 -0500276 select KALLSYMS
Steven Rostedte5a81b62008-08-27 23:31:01 -0400277 help
Ingo Molnar4519d9e2008-10-14 14:15:43 +0200278 This special tracer records the maximum stack footprint of the
279 kernel and displays it in debugfs/tracing/stack_trace.
280
281 This tracer works by hooking into every function call that the
282 kernel executes, and keeping a maximum stack depth value and
Steven Rostedtf38f1d22008-12-16 23:06:40 -0500283 stack-trace saved. If this is configured with DYNAMIC_FTRACE
284 then it will not have any overhead while the stack tracer
285 is disabled.
286
287 To enable the stack tracer on bootup, pass in 'stacktrace'
288 on the kernel command line.
289
290 The stack tracer can also be enabled or disabled via the
291 sysctl kernel.stack_tracer_enabled
Ingo Molnar4519d9e2008-10-14 14:15:43 +0200292
293 Say N if unsure.
Steven Rostedte5a81b62008-08-27 23:31:01 -0400294
Markus Metzgera93751c2008-12-11 13:53:26 +0100295config HW_BRANCH_TRACER
Markus Metzger1e9b51c2008-11-25 09:24:15 +0100296 depends on HAVE_HW_BRANCH_TRACER
Markus Metzgera93751c2008-12-11 13:53:26 +0100297 bool "Trace hw branches"
Markus Metzger1e9b51c2008-11-25 09:24:15 +0100298 select TRACING
299 help
300 This tracer records all branches on the system in a circular
301 buffer giving access to the last N branches for each cpu.
302
Frederic Weisbecker36994e52008-12-29 13:42:23 -0800303config KMEMTRACE
304 bool "Trace SLAB allocations"
305 select TRACING
Frederic Weisbecker36994e52008-12-29 13:42:23 -0800306 help
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.
313
314 This requires an userspace application to use. See
Li Zefan4d1f4372009-04-10 08:48:36 +0800315 Documentation/trace/kmemtrace.txt for more information.
Frederic Weisbecker36994e52008-12-29 13:42:23 -0800316
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).
320
321 If unsure, say N.
322
Frederic Weisbeckere1d8aa92009-01-12 23:15:46 +0100323config WORKQUEUE_TRACER
324 bool "Trace workqueues"
325 select TRACING
326 help
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.
333
Frederic Weisbecker2db270a2009-02-07 20:46:45 +0100334config BLK_DEV_IO_TRACE
335 bool "Support for tracing block io actions"
336 depends on SYSFS
Ingo Molnar1dfba052009-02-09 12:06:54 +0100337 depends on BLOCK
Frederic Weisbecker2db270a2009-02-07 20:46:45 +0100338 select RELAY
339 select DEBUG_FS
340 select TRACEPOINTS
341 select TRACING
342 select STACKTRACE
343 help
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:
348
349 git://git.kernel.dk/blktrace.git
350
351 Tracing also is possible using the ftrace interface, e.g.:
352
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
356
357 If unsure, say N.
Frederic Weisbecker36994e52008-12-29 13:42:23 -0800358
Steven Rostedt3d083392008-05-12 21:20:42 +0200359config DYNAMIC_FTRACE
360 bool "enable/disable ftrace tracepoints dynamically"
Steven Rostedt606576c2008-10-06 19:06:12 -0400361 depends on FUNCTION_TRACER
Steven Rostedt677aa9f2008-05-17 00:01:36 -0400362 depends on HAVE_DYNAMIC_FTRACE
Steven Rostedt3d083392008-05-12 21:20:42 +0200363 default y
364 help
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.
369
Steven Rostedt606576c2008-10-06 19:06:12 -0400370 This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but otherwise
Steven Rostedt3d083392008-05-12 21:20:42 +0200371 has native performance as long as no tracing is active.
372
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.
Steven Rostedt60a11772008-05-12 21:20:44 +0200377
Steven Rostedt8da38212008-08-14 15:45:07 -0400378config FTRACE_MCOUNT_RECORD
379 def_bool y
380 depends on DYNAMIC_FTRACE
381 depends on HAVE_FTRACE_MCOUNT_RECORD
382
Steven Rostedt60a11772008-05-12 21:20:44 +0200383config FTRACE_SELFTEST
384 bool
385
386config FTRACE_STARTUP_TEST
387 bool "Perform a startup test on ftrace"
Ingo Molnar40ada302009-03-05 21:19:55 +0100388 depends on TRACING
Steven Rostedt60a11772008-05-12 21:20:44 +0200389 select FTRACE_SELFTEST
390 help
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
394 tracers of ftrace.
Peter Zijlstra17d80fd2008-10-21 16:31:18 +0200395
Pekka Paalanen6bc5c362009-01-03 21:23:51 +0200396config MMIOTRACE
397 bool "Memory mapped IO tracing"
Ingo Molnar40ada302009-03-05 21:19:55 +0100398 depends on HAVE_MMIOTRACE_SUPPORT && PCI
Pekka Paalanen6bc5c362009-01-03 21:23:51 +0200399 select TRACING
400 help
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.
405
Li Zefan4d1f4372009-04-10 08:48:36 +0800406 See Documentation/trace/mmiotrace.txt.
Pekka Paalanen6bc5c362009-01-03 21:23:51 +0200407 If you are not helping to develop drivers, say N.
408
409config MMIOTRACE_TEST
410 tristate "Test module for mmiotrace"
411 depends on MMIOTRACE && m
412 help
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.
416
417 Say N, unless you absolutely know what you are doing.
418
Peter Zijlstra17d80fd2008-10-21 16:31:18 +0200419endmenu
Ingo Molnar40ada302009-03-05 21:19:55 +0100420
421endif # TRACING_SUPPORT
422