Commit | Line | Data |
---|---|---|
16444a8a ACM |
1 | # |
2 | # Architectures that offer an FTRACE implementation should select HAVE_FTRACE: | |
3 | # | |
4 | config HAVE_FTRACE | |
5 | bool | |
bc0c38d1 | 6 | |
352ad25a SR |
7 | config TRACER_MAX_TRACE |
8 | bool | |
9 | ||
bc0c38d1 SR |
10 | config TRACING |
11 | bool | |
12 | select DEBUG_FS | |
86387f7e | 13 | select STACKTRACE |
bc0c38d1 | 14 | |
1b29b018 SR |
15 | config FTRACE |
16 | bool "Kernel Function Tracer" | |
694379e9 | 17 | depends on HAVE_FTRACE |
1b29b018 SR |
18 | select FRAME_POINTER |
19 | select TRACING | |
35e8e302 | 20 | select CONTEXT_SWITCH_TRACER |
1b29b018 SR |
21 | help |
22 | Enable the kernel to trace every kernel function. This is done | |
23 | by using a compiler feature to insert a small, 5-byte No-Operation | |
24 | instruction to the beginning of every kernel function, which NOP | |
25 | sequence is then dynamically patched into a tracer call when | |
26 | tracing is enabled by the administrator. If it's runtime disabled | |
27 | (the bootup default), then the overhead of the instructions is very | |
28 | small and not measurable even in micro-benchmarks. | |
35e8e302 | 29 | |
81d68a96 SR |
30 | config IRQSOFF_TRACER |
31 | bool "Interrupts-off Latency Tracer" | |
32 | default n | |
33 | depends on TRACE_IRQFLAGS_SUPPORT | |
34 | depends on GENERIC_TIME | |
c1d2327b | 35 | depends on HAVE_FTRACE |
81d68a96 SR |
36 | select TRACE_IRQFLAGS |
37 | select TRACING | |
38 | select TRACER_MAX_TRACE | |
39 | help | |
40 | This option measures the time spent in irqs-off critical | |
41 | sections, with microsecond accuracy. | |
42 | ||
43 | The default measurement method is a maximum search, which is | |
44 | disabled by default and can be runtime (re-)started | |
45 | via: | |
46 | ||
47 | echo 0 > /debugfs/tracing/tracing_max_latency | |
48 | ||
6cd8a4bb SR |
49 | (Note that kernel size and overhead increases with this option |
50 | enabled. This option and the preempt-off timing option can be | |
51 | used together or separately.) | |
52 | ||
53 | config PREEMPT_TRACER | |
54 | bool "Preemption-off Latency Tracer" | |
55 | default n | |
56 | depends on GENERIC_TIME | |
57 | depends on PREEMPT | |
c1d2327b | 58 | depends on HAVE_FTRACE |
6cd8a4bb SR |
59 | select TRACING |
60 | select TRACER_MAX_TRACE | |
61 | help | |
62 | This option measures the time spent in preemption off critical | |
63 | sections, with microsecond accuracy. | |
64 | ||
65 | The default measurement method is a maximum search, which is | |
66 | disabled by default and can be runtime (re-)started | |
67 | via: | |
68 | ||
69 | echo 0 > /debugfs/tracing/tracing_max_latency | |
70 | ||
71 | (Note that kernel size and overhead increases with this option | |
72 | enabled. This option and the irqs-off timing option can be | |
73 | used together or separately.) | |
74 | ||
352ad25a SR |
75 | config SCHED_TRACER |
76 | bool "Scheduling Latency Tracer" | |
c1d2327b | 77 | depends on HAVE_FTRACE |
352ad25a SR |
78 | select TRACING |
79 | select CONTEXT_SWITCH_TRACER | |
80 | select TRACER_MAX_TRACE | |
81 | help | |
82 | This tracer tracks the latency of the highest priority task | |
83 | to be scheduled in, starting from the point it has woken up. | |
84 | ||
35e8e302 SR |
85 | config CONTEXT_SWITCH_TRACER |
86 | bool "Trace process context switches" | |
c1d2327b | 87 | depends on HAVE_FTRACE |
35e8e302 SR |
88 | select TRACING |
89 | select MARKERS | |
90 | help | |
91 | This tracer gets called from the context switch and records | |
92 | all switching of tasks. | |
93 | ||
3d083395 SR |
94 | config DYNAMIC_FTRACE |
95 | bool "enable/disable ftrace tracepoints dynamically" | |
96 | depends on FTRACE | |
97 | default y | |
98 | help | |
99 | This option will modify all the calls to ftrace dynamically | |
100 | (will patch them out of the binary image and replaces them | |
101 | with a No-Op instruction) as they are called. A table is | |
102 | created to dynamically enable them again. | |
103 | ||
104 | This way a CONFIG_FTRACE kernel is slightly larger, but otherwise | |
105 | has native performance as long as no tracing is active. | |
106 | ||
107 | The changes to the code are done by a kernel thread that | |
108 | wakes up once a second and checks to see if any ftrace calls | |
109 | were made. If so, it runs stop_machine (stops all CPUS) | |
110 | and modifies the code to jump over the call to ftrace. | |
60a11774 SR |
111 | |
112 | config FTRACE_SELFTEST | |
113 | bool | |
114 | ||
115 | config FTRACE_STARTUP_TEST | |
116 | bool "Perform a startup test on ftrace" | |
117 | depends on TRACING | |
118 | select FTRACE_SELFTEST | |
119 | help | |
120 | This option performs a series of startup tests on ftrace. On bootup | |
121 | a series of tests are made to verify that the tracer is | |
122 | functioning properly. It will do tests on all the configured | |
123 | tracers of ftrace. |