1# 2# Architectures that offer an FUNCTION_TRACER implementation should 3# select HAVE_FUNCTION_TRACER: 4# 5 6config USER_STACKTRACE_SUPPORT 7 bool 8 9config NOP_TRACER 10 bool 11 12config HAVE_FUNCTION_TRACER 13 bool 14 15config HAVE_FUNCTION_GRAPH_TRACER 16 bool 17 18config HAVE_FUNCTION_TRACE_MCOUNT_TEST 19 bool 20 help 21 This gets selected when the arch tests the function_trace_stop 22 variable at the mcount call site. Otherwise, this variable 23 is tested by the called function. 24 25config HAVE_DYNAMIC_FTRACE 26 bool 27 28config HAVE_FTRACE_MCOUNT_RECORD 29 bool 30 31config HAVE_HW_BRANCH_TRACER 32 bool 33 34config TRACER_MAX_TRACE 35 bool 36 37config RING_BUFFER 38 bool 39 40config TRACING 41 bool 42 select DEBUG_FS 43 select RING_BUFFER 44 select STACKTRACE if STACKTRACE_SUPPORT 45 select TRACEPOINTS 46 select NOP_TRACER 47 48menu "Tracers" 49 50config FUNCTION_TRACER 51 bool "Kernel Function Tracer" 52 depends on HAVE_FUNCTION_TRACER 53 depends on DEBUG_KERNEL 54 select FRAME_POINTER 55 select TRACING 56 select CONTEXT_SWITCH_TRACER 57 help 58 Enable the kernel to trace every kernel function. This is done 59 by using a compiler feature to insert a small, 5-byte No-Operation 60 instruction to the beginning of every kernel function, which NOP 61 sequence is then dynamically patched into a tracer call when 62 tracing is enabled by the administrator. If it's runtime disabled 63 (the bootup default), then the overhead of the instructions is very 64 small and not measurable even in micro-benchmarks. 65 66config FUNCTION_GRAPH_TRACER 67 bool "Kernel Function Graph Tracer" 68 depends on HAVE_FUNCTION_GRAPH_TRACER 69 depends on FUNCTION_TRACER 70 default y 71 help 72 Enable the kernel to trace a function at both its return 73 and its entry. 74 It's first purpose is to trace the duration of functions and 75 draw a call graph for each thread with some informations like 76 the return value. 77 This is done by setting the current return address on the current 78 task structure into a stack of calls. 79 80config IRQSOFF_TRACER 81 bool "Interrupts-off Latency Tracer" 82 default n 83 depends on TRACE_IRQFLAGS_SUPPORT 84 depends on GENERIC_TIME 85 depends on DEBUG_KERNEL 86 select TRACE_IRQFLAGS 87 select TRACING 88 select TRACER_MAX_TRACE 89 help 90 This option measures the time spent in irqs-off critical 91 sections, with microsecond accuracy. 92 93 The default measurement method is a maximum search, which is 94 disabled by default and can be runtime (re-)started 95 via: 96 97 echo 0 > /debugfs/tracing/tracing_max_latency 98 99 (Note that kernel size and overhead increases with this option 100 enabled. This option and the preempt-off timing option can be 101 used together or separately.) 102 103config PREEMPT_TRACER 104 bool "Preemption-off Latency Tracer" 105 default n 106 depends on GENERIC_TIME 107 depends on PREEMPT 108 depends on DEBUG_KERNEL 109 select TRACING 110 select TRACER_MAX_TRACE 111 help 112 This option measures the time spent in preemption off critical 113 sections, with microsecond accuracy. 114 115 The default measurement method is a maximum search, which is 116 disabled by default and can be runtime (re-)started 117 via: 118 119 echo 0 > /debugfs/tracing/tracing_max_latency 120 121 (Note that kernel size and overhead increases with this option 122 enabled. This option and the irqs-off timing option can be 123 used together or separately.) 124 125config SYSPROF_TRACER 126 bool "Sysprof Tracer" 127 depends on X86 128 select TRACING 129 help 130 This tracer provides the trace needed by the 'Sysprof' userspace 131 tool. 132 133config SCHED_TRACER 134 bool "Scheduling Latency Tracer" 135 depends on DEBUG_KERNEL 136 select TRACING 137 select CONTEXT_SWITCH_TRACER 138 select TRACER_MAX_TRACE 139 help 140 This tracer tracks the latency of the highest priority task 141 to be scheduled in, starting from the point it has woken up. 142 143config CONTEXT_SWITCH_TRACER 144 bool "Trace process context switches" 145 depends on DEBUG_KERNEL 146 select TRACING 147 select MARKERS 148 help 149 This tracer gets called from the context switch and records 150 all switching of tasks. 151 152config BOOT_TRACER 153 bool "Trace boot initcalls" 154 depends on DEBUG_KERNEL 155 select TRACING 156 select CONTEXT_SWITCH_TRACER 157 help 158 This tracer helps developers to optimize boot times: it records 159 the timings of the initcalls and traces key events and the identity 160 of tasks that can cause boot delays, such as context-switches. 161 162 Its aim is to be parsed by the /scripts/bootgraph.pl tool to 163 produce pretty graphics about boot inefficiencies, giving a visual 164 representation of the delays during initcalls - but the raw 165 /debug/tracing/trace text output is readable too. 166 167 ( Note that tracing self tests can't be enabled if this tracer is 168 selected, because the self-tests are an initcall as well and that 169 would invalidate the boot trace. ) 170 171config TRACE_BRANCH_PROFILING 172 bool "Trace likely/unlikely profiler" 173 depends on DEBUG_KERNEL 174 select TRACING 175 help 176 This tracer profiles all the the likely and unlikely macros 177 in the kernel. It will display the results in: 178 179 /debugfs/tracing/profile_annotated_branch 180 181 Note: this will add a significant overhead, only turn this 182 on if you need to profile the system's use of these macros. 183 184 Say N if unsure. 185 186config PROFILE_ALL_BRANCHES 187 bool "Profile all if conditionals" 188 depends on TRACE_BRANCH_PROFILING 189 help 190 This tracer profiles all branch conditions. Every if () 191 taken in the kernel is recorded whether it hit or miss. 192 The results will be displayed in: 193 194 /debugfs/tracing/profile_branch 195 196 This configuration, when enabled, will impose a great overhead 197 on the system. This should only be enabled when the system 198 is to be analyzed 199 200 Say N if unsure. 201 202config TRACING_BRANCHES 203 bool 204 help 205 Selected by tracers that will trace the likely and unlikely 206 conditions. This prevents the tracers themselves from being 207 profiled. Profiling the tracing infrastructure can only happen 208 when the likelys and unlikelys are not being traced. 209 210config BRANCH_TRACER 211 bool "Trace likely/unlikely instances" 212 depends on TRACE_BRANCH_PROFILING 213 select TRACING_BRANCHES 214 help 215 This traces the events of likely and unlikely condition 216 calls in the kernel. The difference between this and the 217 "Trace likely/unlikely profiler" is that this is not a 218 histogram of the callers, but actually places the calling 219 events into a running trace buffer to see when and where the 220 events happened, as well as their results. 221 222 Say N if unsure. 223 224config POWER_TRACER 225 bool "Trace power consumption behavior" 226 depends on DEBUG_KERNEL 227 depends on X86 228 select TRACING 229 help 230 This tracer helps developers to analyze and optimize the kernels 231 power management decisions, specifically the C-state and P-state 232 behavior. 233 234 235config STACK_TRACER 236 bool "Trace max stack" 237 depends on HAVE_FUNCTION_TRACER 238 depends on DEBUG_KERNEL 239 select FUNCTION_TRACER 240 select STACKTRACE 241 help 242 This special tracer records the maximum stack footprint of the 243 kernel and displays it in debugfs/tracing/stack_trace. 244 245 This tracer works by hooking into every function call that the 246 kernel executes, and keeping a maximum stack depth value and 247 stack-trace saved. If this is configured with DYNAMIC_FTRACE 248 then it will not have any overhead while the stack tracer 249 is disabled. 250 251 To enable the stack tracer on bootup, pass in 'stacktrace' 252 on the kernel command line. 253 254 The stack tracer can also be enabled or disabled via the 255 sysctl kernel.stack_tracer_enabled 256 257 Say N if unsure. 258 259config HW_BRANCH_TRACER 260 depends on HAVE_HW_BRANCH_TRACER 261 bool "Trace hw branches" 262 select TRACING 263 help 264 This tracer records all branches on the system in a circular 265 buffer giving access to the last N branches for each cpu. 266 267config DYNAMIC_FTRACE 268 bool "enable/disable ftrace tracepoints dynamically" 269 depends on FUNCTION_TRACER 270 depends on HAVE_DYNAMIC_FTRACE 271 depends on DEBUG_KERNEL 272 default y 273 help 274 This option will modify all the calls to ftrace dynamically 275 (will patch them out of the binary image and replaces them 276 with a No-Op instruction) as they are called. A table is 277 created to dynamically enable them again. 278 279 This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but otherwise 280 has native performance as long as no tracing is active. 281 282 The changes to the code are done by a kernel thread that 283 wakes up once a second and checks to see if any ftrace calls 284 were made. If so, it runs stop_machine (stops all CPUS) 285 and modifies the code to jump over the call to ftrace. 286 287config FTRACE_MCOUNT_RECORD 288 def_bool y 289 depends on DYNAMIC_FTRACE 290 depends on HAVE_FTRACE_MCOUNT_RECORD 291 292config FTRACE_SELFTEST 293 bool 294 295config FTRACE_STARTUP_TEST 296 bool "Perform a startup test on ftrace" 297 depends on TRACING && DEBUG_KERNEL && !BOOT_TRACER 298 select FTRACE_SELFTEST 299 help 300 This option performs a series of startup tests on ftrace. On bootup 301 a series of tests are made to verify that the tracer is 302 functioning properly. It will do tests on all the configured 303 tracers of ftrace. 304 305endmenu 306