]>
Commit | Line | Data |
---|---|---|
16444a8a | 1 | # |
606576ce SR |
2 | # Architectures that offer an FUNCTION_TRACER implementation should |
3 | # select HAVE_FUNCTION_TRACER: | |
16444a8a | 4 | # |
2a3a4f66 | 5 | |
8d26487f TE |
6 | config USER_STACKTRACE_SUPPORT |
7 | bool | |
8 | ||
2a3a4f66 FW |
9 | config NOP_TRACER |
10 | bool | |
11 | ||
78d904b4 SR |
12 | config HAVE_FTRACE_NMI_ENTER |
13 | bool | |
555f386c | 14 | help |
40892367 | 15 | See Documentation/trace/ftrace-design.txt |
78d904b4 | 16 | |
606576ce | 17 | config HAVE_FUNCTION_TRACER |
16444a8a | 18 | bool |
555f386c | 19 | help |
40892367 | 20 | See Documentation/trace/ftrace-design.txt |
bc0c38d1 | 21 | |
fb52607a | 22 | config HAVE_FUNCTION_GRAPH_TRACER |
15e6cb36 | 23 | bool |
555f386c | 24 | help |
40892367 | 25 | See Documentation/trace/ftrace-design.txt |
15e6cb36 | 26 | |
71e308a2 SR |
27 | config HAVE_FUNCTION_GRAPH_FP_TEST |
28 | bool | |
29 | help | |
03688970 | 30 | See Documentation/trace/ftrace-design.txt |
71e308a2 | 31 | |
60a7ecf4 SR |
32 | config HAVE_FUNCTION_TRACE_MCOUNT_TEST |
33 | bool | |
34 | help | |
40892367 | 35 | See Documentation/trace/ftrace-design.txt |
60a7ecf4 | 36 | |
677aa9f7 SR |
37 | config HAVE_DYNAMIC_FTRACE |
38 | bool | |
555f386c | 39 | help |
40892367 | 40 | See Documentation/trace/ftrace-design.txt |
677aa9f7 | 41 | |
8da3821b SR |
42 | config HAVE_FTRACE_MCOUNT_RECORD |
43 | bool | |
555f386c | 44 | help |
40892367 | 45 | See Documentation/trace/ftrace-design.txt |
8da3821b | 46 | |
66700001 | 47 | config HAVE_SYSCALL_TRACEPOINTS |
ee08c6ec | 48 | bool |
555f386c | 49 | help |
40892367 | 50 | See Documentation/trace/ftrace-design.txt |
ee08c6ec | 51 | |
cf4db259 | 52 | config HAVE_C_RECORDMCOUNT |
72441cb1 SR |
53 | bool |
54 | help | |
55 | C version of recordmcount available? | |
56 | ||
352ad25a SR |
57 | config TRACER_MAX_TRACE |
58 | bool | |
59 | ||
7a8e76a3 SR |
60 | config RING_BUFFER |
61 | bool | |
62 | ||
78d904b4 SR |
63 | config FTRACE_NMI_ENTER |
64 | bool | |
65 | depends on HAVE_FTRACE_NMI_ENTER | |
66 | default y | |
67 | ||
5f77a88b | 68 | config EVENT_TRACING |
b11c53e1 Z |
69 | select CONTEXT_SWITCH_TRACER |
70 | bool | |
71 | ||
25e41933 TR |
72 | config EVENT_POWER_TRACING_DEPRECATED |
73 | depends on EVENT_TRACING | |
74 | bool "Deprecated power event trace API, to be removed" | |
75 | default y | |
76 | help | |
77 | Provides old power event types: | |
78 | C-state/idle accounting events: | |
79 | power:power_start | |
80 | power:power_end | |
81 | and old cpufreq accounting event: | |
82 | power:power_frequency | |
83 | This is for userspace compatibility | |
84 | and will vanish after 5 kernel iterations, | |
85 | namely 2.6.41. | |
86 | ||
b11c53e1 | 87 | config CONTEXT_SWITCH_TRACER |
5f77a88b TZ |
88 | bool |
89 | ||
85bac32c SR |
90 | config RING_BUFFER_ALLOW_SWAP |
91 | bool | |
92 | help | |
93 | Allow the use of ring_buffer_swap_cpu. | |
94 | Adds a very slight overhead to tracing when enabled. | |
95 | ||
5e0a0939 SR |
96 | # All tracer options should select GENERIC_TRACER. For those options that are |
97 | # enabled by all tracers (context switch and event tracer) they select TRACING. | |
98 | # This allows those options to appear when no other tracer is selected. But the | |
99 | # options do not appear when something else selects it. We need the two options | |
100 | # GENERIC_TRACER and TRACING to avoid circular dependencies to accomplish the | |
40892367 | 101 | # hiding of the automatic options. |
5e0a0939 | 102 | |
bc0c38d1 SR |
103 | config TRACING |
104 | bool | |
105 | select DEBUG_FS | |
7a8e76a3 | 106 | select RING_BUFFER |
c2c80529 | 107 | select STACKTRACE if STACKTRACE_SUPPORT |
5f87f112 | 108 | select TRACEPOINTS |
f3384b28 | 109 | select NOP_TRACER |
769b0441 | 110 | select BINARY_PRINTF |
5f77a88b | 111 | select EVENT_TRACING |
bc0c38d1 | 112 | |
5e0a0939 SR |
113 | config GENERIC_TRACER |
114 | bool | |
115 | select TRACING | |
116 | ||
40ada30f IM |
117 | # |
118 | # Minimum requirements an architecture has to meet for us to | |
119 | # be able to offer generic tracing facilities: | |
120 | # | |
121 | config TRACING_SUPPORT | |
122 | bool | |
45b95608 AV |
123 | # PPC32 has no irqflags tracing support, but it can use most of the |
124 | # tracers anyway, they were tested to build and work. Note that new | |
125 | # exceptions to this list aren't welcomed, better implement the | |
126 | # irqflags tracing for your architecture. | |
127 | depends on TRACE_IRQFLAGS_SUPPORT || PPC32 | |
40ada30f | 128 | depends on STACKTRACE_SUPPORT |
422d3c7a | 129 | default y |
40ada30f IM |
130 | |
131 | if TRACING_SUPPORT | |
132 | ||
4ed9f071 SR |
133 | menuconfig FTRACE |
134 | bool "Tracers" | |
65b77242 | 135 | default y if DEBUG_KERNEL |
4ed9f071 | 136 | help |
40892367 | 137 | Enable the kernel tracing infrastructure. |
4ed9f071 SR |
138 | |
139 | if FTRACE | |
17d80fd0 | 140 | |
606576ce | 141 | config FUNCTION_TRACER |
1b29b018 | 142 | bool "Kernel Function Tracer" |
606576ce | 143 | depends on HAVE_FUNCTION_TRACER |
becf91f1 | 144 | select FRAME_POINTER if !ARM_UNWIND && !S390 |
4d7a077c | 145 | select KALLSYMS |
5e0a0939 | 146 | select GENERIC_TRACER |
35e8e302 | 147 | select CONTEXT_SWITCH_TRACER |
1b29b018 SR |
148 | help |
149 | Enable the kernel to trace every kernel function. This is done | |
150 | by using a compiler feature to insert a small, 5-byte No-Operation | |
40892367 | 151 | instruction at the beginning of every kernel function, which NOP |
1b29b018 SR |
152 | sequence is then dynamically patched into a tracer call when |
153 | tracing is enabled by the administrator. If it's runtime disabled | |
154 | (the bootup default), then the overhead of the instructions is very | |
155 | small and not measurable even in micro-benchmarks. | |
35e8e302 | 156 | |
fb52607a FW |
157 | config FUNCTION_GRAPH_TRACER |
158 | bool "Kernel Function Graph Tracer" | |
159 | depends on HAVE_FUNCTION_GRAPH_TRACER | |
15e6cb36 | 160 | depends on FUNCTION_TRACER |
eb4a0378 | 161 | depends on !X86_32 || !CC_OPTIMIZE_FOR_SIZE |
764f3b95 | 162 | default y |
15e6cb36 | 163 | help |
fb52607a FW |
164 | Enable the kernel to trace a function at both its return |
165 | and its entry. | |
692105b8 ML |
166 | Its first purpose is to trace the duration of functions and |
167 | draw a call graph for each thread with some information like | |
40892367 | 168 | the return value. This is done by setting the current return |
692105b8 | 169 | address on the current task structure into a stack of calls. |
15e6cb36 | 170 | |
bac429f0 | 171 | |
81d68a96 SR |
172 | config IRQSOFF_TRACER |
173 | bool "Interrupts-off Latency Tracer" | |
174 | default n | |
175 | depends on TRACE_IRQFLAGS_SUPPORT | |
592913ec | 176 | depends on !ARCH_USES_GETTIMEOFFSET |
81d68a96 | 177 | select TRACE_IRQFLAGS |
5e0a0939 | 178 | select GENERIC_TRACER |
81d68a96 | 179 | select TRACER_MAX_TRACE |
85bac32c | 180 | select RING_BUFFER_ALLOW_SWAP |
81d68a96 SR |
181 | help |
182 | This option measures the time spent in irqs-off critical | |
183 | sections, with microsecond accuracy. | |
184 | ||
185 | The default measurement method is a maximum search, which is | |
186 | disabled by default and can be runtime (re-)started | |
187 | via: | |
188 | ||
156f5a78 | 189 | echo 0 > /sys/kernel/debug/tracing/tracing_max_latency |
81d68a96 | 190 | |
40892367 | 191 | (Note that kernel size and overhead increase with this option |
6cd8a4bb SR |
192 | enabled. This option and the preempt-off timing option can be |
193 | used together or separately.) | |
194 | ||
195 | config PREEMPT_TRACER | |
196 | bool "Preemption-off Latency Tracer" | |
197 | default n | |
592913ec | 198 | depends on !ARCH_USES_GETTIMEOFFSET |
6cd8a4bb | 199 | depends on PREEMPT |
5e0a0939 | 200 | select GENERIC_TRACER |
6cd8a4bb | 201 | select TRACER_MAX_TRACE |
85bac32c | 202 | select RING_BUFFER_ALLOW_SWAP |
6cd8a4bb | 203 | help |
40892367 | 204 | This option measures the time spent in preemption-off critical |
6cd8a4bb SR |
205 | sections, with microsecond accuracy. |
206 | ||
207 | The default measurement method is a maximum search, which is | |
208 | disabled by default and can be runtime (re-)started | |
209 | via: | |
210 | ||
156f5a78 | 211 | echo 0 > /sys/kernel/debug/tracing/tracing_max_latency |
6cd8a4bb | 212 | |
40892367 | 213 | (Note that kernel size and overhead increase with this option |
6cd8a4bb SR |
214 | enabled. This option and the irqs-off timing option can be |
215 | used together or separately.) | |
216 | ||
352ad25a SR |
217 | config SCHED_TRACER |
218 | bool "Scheduling Latency Tracer" | |
5e0a0939 | 219 | select GENERIC_TRACER |
352ad25a SR |
220 | select CONTEXT_SWITCH_TRACER |
221 | select TRACER_MAX_TRACE | |
222 | help | |
223 | This tracer tracks the latency of the highest priority task | |
224 | to be scheduled in, starting from the point it has woken up. | |
225 | ||
897f17a6 SR |
226 | config ENABLE_DEFAULT_TRACERS |
227 | bool "Trace process context switches and events" | |
5e0a0939 | 228 | depends on !GENERIC_TRACER |
b77e38aa SR |
229 | select TRACING |
230 | help | |
40892367 | 231 | This tracer hooks to various trace points in the kernel, |
b77e38aa | 232 | allowing the user to pick and choose which trace point they |
897f17a6 | 233 | want to trace. It also includes the sched_switch tracer plugin. |
a7abe97f | 234 | |
ee08c6ec FW |
235 | config FTRACE_SYSCALLS |
236 | bool "Trace syscalls" | |
66700001 | 237 | depends on HAVE_SYSCALL_TRACEPOINTS |
5e0a0939 | 238 | select GENERIC_TRACER |
0ea1c415 | 239 | select KALLSYMS |
ee08c6ec FW |
240 | help |
241 | Basic tracer to catch the syscall entry and exit events. | |
242 | ||
2ed84eeb | 243 | config TRACE_BRANCH_PROFILING |
9ae5b879 | 244 | bool |
5e0a0939 | 245 | select GENERIC_TRACER |
9ae5b879 SR |
246 | |
247 | choice | |
248 | prompt "Branch Profiling" | |
249 | default BRANCH_PROFILE_NONE | |
250 | help | |
251 | The branch profiling is a software profiler. It will add hooks | |
252 | into the C conditionals to test which path a branch takes. | |
253 | ||
254 | The likely/unlikely profiler only looks at the conditions that | |
255 | are annotated with a likely or unlikely macro. | |
256 | ||
40892367 | 257 | The "all branch" profiler will profile every if-statement in the |
9ae5b879 | 258 | kernel. This profiler will also enable the likely/unlikely |
40892367 | 259 | profiler. |
9ae5b879 | 260 | |
40892367 RD |
261 | Either of the above profilers adds a bit of overhead to the system. |
262 | If unsure, choose "No branch profiling". | |
9ae5b879 SR |
263 | |
264 | config BRANCH_PROFILE_NONE | |
265 | bool "No branch profiling" | |
266 | help | |
40892367 RD |
267 | No branch profiling. Branch profiling adds a bit of overhead. |
268 | Only enable it if you want to analyse the branching behavior. | |
269 | Otherwise keep it disabled. | |
9ae5b879 SR |
270 | |
271 | config PROFILE_ANNOTATED_BRANCHES | |
272 | bool "Trace likely/unlikely profiler" | |
273 | select TRACE_BRANCH_PROFILING | |
1f0d69a9 SR |
274 | help |
275 | This tracer profiles all the the likely and unlikely macros | |
276 | in the kernel. It will display the results in: | |
277 | ||
13e5befa | 278 | /sys/kernel/debug/tracing/trace_stat/branch_annotated |
1f0d69a9 | 279 | |
40892367 | 280 | Note: this will add a significant overhead; only turn this |
1f0d69a9 SR |
281 | on if you need to profile the system's use of these macros. |
282 | ||
2bcd521a SR |
283 | config PROFILE_ALL_BRANCHES |
284 | bool "Profile all if conditionals" | |
9ae5b879 | 285 | select TRACE_BRANCH_PROFILING |
2bcd521a SR |
286 | help |
287 | This tracer profiles all branch conditions. Every if () | |
288 | taken in the kernel is recorded whether it hit or miss. | |
289 | The results will be displayed in: | |
290 | ||
13e5befa | 291 | /sys/kernel/debug/tracing/trace_stat/branch_all |
2bcd521a | 292 | |
9ae5b879 SR |
293 | This option also enables the likely/unlikely profiler. |
294 | ||
2bcd521a SR |
295 | This configuration, when enabled, will impose a great overhead |
296 | on the system. This should only be enabled when the system | |
40892367 | 297 | is to be analyzed in much detail. |
9ae5b879 | 298 | endchoice |
2bcd521a | 299 | |
2ed84eeb | 300 | config TRACING_BRANCHES |
52f232cb SR |
301 | bool |
302 | help | |
303 | Selected by tracers that will trace the likely and unlikely | |
304 | conditions. This prevents the tracers themselves from being | |
305 | profiled. Profiling the tracing infrastructure can only happen | |
306 | when the likelys and unlikelys are not being traced. | |
307 | ||
2ed84eeb | 308 | config BRANCH_TRACER |
52f232cb | 309 | bool "Trace likely/unlikely instances" |
2ed84eeb SR |
310 | depends on TRACE_BRANCH_PROFILING |
311 | select TRACING_BRANCHES | |
52f232cb SR |
312 | help |
313 | This traces the events of likely and unlikely condition | |
314 | calls in the kernel. The difference between this and the | |
315 | "Trace likely/unlikely profiler" is that this is not a | |
316 | histogram of the callers, but actually places the calling | |
317 | events into a running trace buffer to see when and where the | |
318 | events happened, as well as their results. | |
319 | ||
320 | Say N if unsure. | |
321 | ||
e5a81b62 SR |
322 | config STACK_TRACER |
323 | bool "Trace max stack" | |
606576ce | 324 | depends on HAVE_FUNCTION_TRACER |
606576ce | 325 | select FUNCTION_TRACER |
e5a81b62 | 326 | select STACKTRACE |
4d7a077c | 327 | select KALLSYMS |
e5a81b62 | 328 | help |
4519d9e5 | 329 | This special tracer records the maximum stack footprint of the |
156f5a78 | 330 | kernel and displays it in /sys/kernel/debug/tracing/stack_trace. |
4519d9e5 IM |
331 | |
332 | This tracer works by hooking into every function call that the | |
333 | kernel executes, and keeping a maximum stack depth value and | |
f38f1d2a SR |
334 | stack-trace saved. If this is configured with DYNAMIC_FTRACE |
335 | then it will not have any overhead while the stack tracer | |
336 | is disabled. | |
337 | ||
338 | To enable the stack tracer on bootup, pass in 'stacktrace' | |
339 | on the kernel command line. | |
340 | ||
341 | The stack tracer can also be enabled or disabled via the | |
342 | sysctl kernel.stack_tracer_enabled | |
4519d9e5 IM |
343 | |
344 | Say N if unsure. | |
e5a81b62 | 345 | |
2db270a8 | 346 | config BLK_DEV_IO_TRACE |
40892367 | 347 | bool "Support for tracing block IO actions" |
2db270a8 | 348 | depends on SYSFS |
1dfba05d | 349 | depends on BLOCK |
2db270a8 FW |
350 | select RELAY |
351 | select DEBUG_FS | |
352 | select TRACEPOINTS | |
5e0a0939 | 353 | select GENERIC_TRACER |
2db270a8 FW |
354 | select STACKTRACE |
355 | help | |
356 | Say Y here if you want to be able to trace the block layer actions | |
357 | on a given queue. Tracing allows you to see any traffic happening | |
358 | on a block device queue. For more information (and the userspace | |
359 | support tools needed), fetch the blktrace tools from: | |
360 | ||
361 | git://git.kernel.dk/blktrace.git | |
362 | ||
363 | Tracing also is possible using the ftrace interface, e.g.: | |
364 | ||
365 | echo 1 > /sys/block/sda/sda1/trace/enable | |
366 | echo blk > /sys/kernel/debug/tracing/current_tracer | |
367 | cat /sys/kernel/debug/tracing/trace_pipe | |
368 | ||
369 | If unsure, say N. | |
36994e58 | 370 | |
77b44d1b | 371 | config KPROBE_EVENT |
413d37d1 | 372 | depends on KPROBES |
f850c30c | 373 | depends on HAVE_REGS_AND_STACK_ACCESS_API |
77b44d1b | 374 | bool "Enable kprobes-based dynamic events" |
413d37d1 | 375 | select TRACING |
77b44d1b | 376 | default y |
413d37d1 | 377 | help |
40892367 RD |
378 | This allows the user to add tracing events (similar to tracepoints) |
379 | on the fly via the ftrace interface. See | |
380 | Documentation/trace/kprobetrace.txt for more details. | |
77b44d1b MH |
381 | |
382 | Those events can be inserted wherever kprobes can probe, and record | |
383 | various register and memory values. | |
384 | ||
40892367 RD |
385 | This option is also required by perf-probe subcommand of perf tools. |
386 | If you want to use perf tools, this option is strongly recommended. | |
413d37d1 | 387 | |
3d083395 SR |
388 | config DYNAMIC_FTRACE |
389 | bool "enable/disable ftrace tracepoints dynamically" | |
606576ce | 390 | depends on FUNCTION_TRACER |
677aa9f7 | 391 | depends on HAVE_DYNAMIC_FTRACE |
3d083395 SR |
392 | default y |
393 | help | |
40892367 RD |
394 | This option will modify all the calls to ftrace dynamically |
395 | (will patch them out of the binary image and replace them | |
396 | with a No-Op instruction) as they are called. A table is | |
397 | created to dynamically enable them again. | |
3d083395 | 398 | |
40892367 RD |
399 | This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but |
400 | otherwise has native performance as long as no tracing is active. | |
3d083395 | 401 | |
40892367 RD |
402 | The changes to the code are done by a kernel thread that |
403 | wakes up once a second and checks to see if any ftrace calls | |
404 | were made. If so, it runs stop_machine (stops all CPUS) | |
405 | and modifies the code to jump over the call to ftrace. | |
60a11774 | 406 | |
bac429f0 SR |
407 | config FUNCTION_PROFILER |
408 | bool "Kernel function profiler" | |
493762fc | 409 | depends on FUNCTION_TRACER |
bac429f0 SR |
410 | default n |
411 | help | |
40892367 RD |
412 | This option enables the kernel function profiler. A file is created |
413 | in debugfs called function_profile_enabled which defaults to zero. | |
414 | When a 1 is echoed into this file profiling begins, and when a | |
415 | zero is entered, profiling stops. A "functions" file is created in | |
416 | the trace_stats directory; this file shows the list of functions that | |
417 | have been hit and their counters. | |
bac429f0 | 418 | |
40892367 | 419 | If in doubt, say N. |
bac429f0 | 420 | |
8da3821b SR |
421 | config FTRACE_MCOUNT_RECORD |
422 | def_bool y | |
423 | depends on DYNAMIC_FTRACE | |
424 | depends on HAVE_FTRACE_MCOUNT_RECORD | |
425 | ||
60a11774 SR |
426 | config FTRACE_SELFTEST |
427 | bool | |
428 | ||
429 | config FTRACE_STARTUP_TEST | |
430 | bool "Perform a startup test on ftrace" | |
5e0a0939 | 431 | depends on GENERIC_TRACER |
60a11774 SR |
432 | select FTRACE_SELFTEST |
433 | help | |
434 | This option performs a series of startup tests on ftrace. On bootup | |
435 | a series of tests are made to verify that the tracer is | |
436 | functioning properly. It will do tests on all the configured | |
437 | tracers of ftrace. | |
17d80fd0 | 438 | |
1f5a6b45 SR |
439 | config EVENT_TRACE_TEST_SYSCALLS |
440 | bool "Run selftest on syscall events" | |
441 | depends on FTRACE_STARTUP_TEST | |
442 | help | |
443 | This option will also enable testing every syscall event. | |
444 | It only enables the event and disables it and runs various loads | |
445 | with the event enabled. This adds a bit more time for kernel boot | |
446 | up since it runs this on every system call defined. | |
447 | ||
448 | TBD - enable a way to actually call the syscalls as we test their | |
449 | events | |
450 | ||
fe6f90e5 PP |
451 | config MMIOTRACE |
452 | bool "Memory mapped IO tracing" | |
40ada30f | 453 | depends on HAVE_MMIOTRACE_SUPPORT && PCI |
5e0a0939 | 454 | select GENERIC_TRACER |
fe6f90e5 PP |
455 | help |
456 | Mmiotrace traces Memory Mapped I/O access and is meant for | |
457 | debugging and reverse engineering. It is called from the ioremap | |
458 | implementation and works via page faults. Tracing is disabled by | |
459 | default and can be enabled at run-time. | |
460 | ||
4d1f4372 | 461 | See Documentation/trace/mmiotrace.txt. |
fe6f90e5 PP |
462 | If you are not helping to develop drivers, say N. |
463 | ||
464 | config MMIOTRACE_TEST | |
465 | tristate "Test module for mmiotrace" | |
466 | depends on MMIOTRACE && m | |
467 | help | |
468 | This is a dumb module for testing mmiotrace. It is very dangerous | |
469 | as it will write garbage to IO memory starting at a given address. | |
470 | However, it should be safe to use on e.g. unused portion of VRAM. | |
471 | ||
472 | Say N, unless you absolutely know what you are doing. | |
473 | ||
5092dbc9 SR |
474 | config RING_BUFFER_BENCHMARK |
475 | tristate "Ring buffer benchmark stress tester" | |
476 | depends on RING_BUFFER | |
477 | help | |
40892367 RD |
478 | This option creates a test to stress the ring buffer and benchmark it. |
479 | It creates its own ring buffer such that it will not interfere with | |
5092dbc9 SR |
480 | any other users of the ring buffer (such as ftrace). It then creates |
481 | a producer and consumer that will run for 10 seconds and sleep for | |
482 | 10 seconds. Each interval it will print out the number of events | |
483 | it recorded and give a rough estimate of how long each iteration took. | |
484 | ||
485 | It does not disable interrupts or raise its priority, so it may be | |
486 | affected by processes that are running. | |
487 | ||
40892367 | 488 | If unsure, say N. |
5092dbc9 | 489 | |
4ed9f071 | 490 | endif # FTRACE |
40ada30f IM |
491 | |
492 | endif # TRACING_SUPPORT | |
493 |