Kconfig 6.7 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235
  1. #
  2. # Architectures that offer an FUNCTION_TRACER implementation should
  3. # select HAVE_FUNCTION_TRACER:
  4. #
  5. config NOP_TRACER
  6. bool
  7. config HAVE_FUNCTION_TRACER
  8. bool
  9. config HAVE_FUNCTION_RET_TRACER
  10. bool
  11. config HAVE_FUNCTION_TRACE_MCOUNT_TEST
  12. bool
  13. help
  14. This gets selected when the arch tests the function_trace_stop
  15. variable at the mcount call site. Otherwise, this variable
  16. is tested by the called function.
  17. config HAVE_DYNAMIC_FTRACE
  18. bool
  19. config HAVE_FTRACE_MCOUNT_RECORD
  20. bool
  21. config TRACER_MAX_TRACE
  22. bool
  23. config RING_BUFFER
  24. bool
  25. config TRACING
  26. bool
  27. select DEBUG_FS
  28. select RING_BUFFER
  29. select STACKTRACE if STACKTRACE_SUPPORT
  30. select TRACEPOINTS
  31. select NOP_TRACER
  32. menu "Tracers"
  33. config FUNCTION_TRACER
  34. bool "Kernel Function Tracer"
  35. depends on HAVE_FUNCTION_TRACER
  36. depends on DEBUG_KERNEL
  37. select FRAME_POINTER
  38. select TRACING
  39. select CONTEXT_SWITCH_TRACER
  40. help
  41. Enable the kernel to trace every kernel function. This is done
  42. by using a compiler feature to insert a small, 5-byte No-Operation
  43. instruction to the beginning of every kernel function, which NOP
  44. sequence is then dynamically patched into a tracer call when
  45. tracing is enabled by the administrator. If it's runtime disabled
  46. (the bootup default), then the overhead of the instructions is very
  47. small and not measurable even in micro-benchmarks.
  48. config FUNCTION_RET_TRACER
  49. bool "Kernel Function return Tracer"
  50. depends on !DYNAMIC_FTRACE
  51. depends on HAVE_FUNCTION_RET_TRACER
  52. depends on FUNCTION_TRACER
  53. help
  54. Enable the kernel to trace a function at its return.
  55. It's first purpose is to trace the duration of functions.
  56. This is done by setting the current return address on the thread
  57. info structure of the current task.
  58. config IRQSOFF_TRACER
  59. bool "Interrupts-off Latency Tracer"
  60. default n
  61. depends on TRACE_IRQFLAGS_SUPPORT
  62. depends on GENERIC_TIME
  63. depends on DEBUG_KERNEL
  64. select TRACE_IRQFLAGS
  65. select TRACING
  66. select TRACER_MAX_TRACE
  67. help
  68. This option measures the time spent in irqs-off critical
  69. sections, with microsecond accuracy.
  70. The default measurement method is a maximum search, which is
  71. disabled by default and can be runtime (re-)started
  72. via:
  73. echo 0 > /debugfs/tracing/tracing_max_latency
  74. (Note that kernel size and overhead increases with this option
  75. enabled. This option and the preempt-off timing option can be
  76. used together or separately.)
  77. config PREEMPT_TRACER
  78. bool "Preemption-off Latency Tracer"
  79. default n
  80. depends on GENERIC_TIME
  81. depends on PREEMPT
  82. depends on DEBUG_KERNEL
  83. select TRACING
  84. select TRACER_MAX_TRACE
  85. help
  86. This option measures the time spent in preemption off critical
  87. sections, with microsecond accuracy.
  88. The default measurement method is a maximum search, which is
  89. disabled by default and can be runtime (re-)started
  90. via:
  91. echo 0 > /debugfs/tracing/tracing_max_latency
  92. (Note that kernel size and overhead increases with this option
  93. enabled. This option and the irqs-off timing option can be
  94. used together or separately.)
  95. config SYSPROF_TRACER
  96. bool "Sysprof Tracer"
  97. depends on X86
  98. select TRACING
  99. help
  100. This tracer provides the trace needed by the 'Sysprof' userspace
  101. tool.
  102. config SCHED_TRACER
  103. bool "Scheduling Latency Tracer"
  104. depends on DEBUG_KERNEL
  105. select TRACING
  106. select CONTEXT_SWITCH_TRACER
  107. select TRACER_MAX_TRACE
  108. help
  109. This tracer tracks the latency of the highest priority task
  110. to be scheduled in, starting from the point it has woken up.
  111. config CONTEXT_SWITCH_TRACER
  112. bool "Trace process context switches"
  113. depends on DEBUG_KERNEL
  114. select TRACING
  115. select MARKERS
  116. help
  117. This tracer gets called from the context switch and records
  118. all switching of tasks.
  119. config BOOT_TRACER
  120. bool "Trace boot initcalls"
  121. depends on DEBUG_KERNEL
  122. select TRACING
  123. select CONTEXT_SWITCH_TRACER
  124. help
  125. This tracer helps developers to optimize boot times: it records
  126. the timings of the initcalls and traces key events and the identity
  127. of tasks that can cause boot delays, such as context-switches.
  128. Its aim is to be parsed by the /scripts/bootgraph.pl tool to
  129. produce pretty graphics about boot inefficiencies, giving a visual
  130. representation of the delays during initcalls - but the raw
  131. /debug/tracing/trace text output is readable too.
  132. ( Note that tracing self tests can't be enabled if this tracer is
  133. selected, because the self-tests are an initcall as well and that
  134. would invalidate the boot trace. )
  135. config TRACE_UNLIKELY_PROFILE
  136. bool "Trace likely/unlikely profiler"
  137. depends on DEBUG_KERNEL
  138. select TRACING
  139. help
  140. This tracer profiles all the the likely and unlikely macros
  141. in the kernel. It will display the results in:
  142. /debugfs/tracing/profile_likely
  143. /debugfs/tracing/profile_unlikely
  144. Note: this will add a significant overhead, only turn this
  145. on if you need to profile the system's use of these macros.
  146. Say N if unsure.
  147. config STACK_TRACER
  148. bool "Trace max stack"
  149. depends on HAVE_FUNCTION_TRACER
  150. depends on DEBUG_KERNEL
  151. select FUNCTION_TRACER
  152. select STACKTRACE
  153. help
  154. This special tracer records the maximum stack footprint of the
  155. kernel and displays it in debugfs/tracing/stack_trace.
  156. This tracer works by hooking into every function call that the
  157. kernel executes, and keeping a maximum stack depth value and
  158. stack-trace saved. Because this logic has to execute in every
  159. kernel function, all the time, this option can slow down the
  160. kernel measurably and is generally intended for kernel
  161. developers only.
  162. Say N if unsure.
  163. config DYNAMIC_FTRACE
  164. bool "enable/disable ftrace tracepoints dynamically"
  165. depends on FUNCTION_TRACER
  166. depends on HAVE_DYNAMIC_FTRACE
  167. depends on DEBUG_KERNEL
  168. default y
  169. help
  170. This option will modify all the calls to ftrace dynamically
  171. (will patch them out of the binary image and replaces them
  172. with a No-Op instruction) as they are called. A table is
  173. created to dynamically enable them again.
  174. This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but otherwise
  175. has native performance as long as no tracing is active.
  176. The changes to the code are done by a kernel thread that
  177. wakes up once a second and checks to see if any ftrace calls
  178. were made. If so, it runs stop_machine (stops all CPUS)
  179. and modifies the code to jump over the call to ftrace.
  180. config FTRACE_MCOUNT_RECORD
  181. def_bool y
  182. depends on DYNAMIC_FTRACE
  183. depends on HAVE_FTRACE_MCOUNT_RECORD
  184. config FTRACE_SELFTEST
  185. bool
  186. config FTRACE_STARTUP_TEST
  187. bool "Perform a startup test on ftrace"
  188. depends on TRACING && DEBUG_KERNEL && !BOOT_TRACER
  189. select FTRACE_SELFTEST
  190. help
  191. This option performs a series of startup tests on ftrace. On bootup
  192. a series of tests are made to verify that the tracer is
  193. functioning properly. It will do tests on all the configured
  194. tracers of ftrace.
  195. endmenu