Arnaldo Carvalho de Melo | 16444a8 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 1 | # |
Steven Rostedt | 606576c | 2008-10-06 19:06:12 -0400 | [diff] [blame] | 2 | # Architectures that offer an FUNCTION_TRACER implementation should |
| 3 | # select HAVE_FUNCTION_TRACER: |
Arnaldo Carvalho de Melo | 16444a8 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 4 | # |
Frédéric Weisbecker | 2a3a4f6 | 2008-09-21 20:12:14 +0200 | [diff] [blame] | 5 | |
Török Edwin | 8d26487 | 2008-11-23 12:39:08 +0200 | [diff] [blame] | 6 | config USER_STACKTRACE_SUPPORT |
| 7 | bool |
| 8 | |
Frédéric Weisbecker | 2a3a4f6 | 2008-09-21 20:12:14 +0200 | [diff] [blame] | 9 | config NOP_TRACER |
| 10 | bool |
| 11 | |
Steven Rostedt | 78d904b | 2009-02-05 18:43:07 -0500 | [diff] [blame] | 12 | config HAVE_FTRACE_NMI_ENTER |
| 13 | bool |
Mike Frysinger | 555f386 | 2009-09-14 20:10:15 -0400 | [diff] [blame] | 14 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 15 | See Documentation/trace/ftrace-design.txt |
Steven Rostedt | 78d904b | 2009-02-05 18:43:07 -0500 | [diff] [blame] | 16 | |
Steven Rostedt | 606576c | 2008-10-06 19:06:12 -0400 | [diff] [blame] | 17 | config HAVE_FUNCTION_TRACER |
Arnaldo Carvalho de Melo | 16444a8 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 18 | bool |
Mike Frysinger | 555f386 | 2009-09-14 20:10:15 -0400 | [diff] [blame] | 19 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 20 | See Documentation/trace/ftrace-design.txt |
Steven Rostedt | bc0c38d | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 21 | |
Frederic Weisbecker | fb52607 | 2008-11-25 21:07:04 +0100 | [diff] [blame] | 22 | config HAVE_FUNCTION_GRAPH_TRACER |
Frederic Weisbecker | 15e6cb3 | 2008-11-11 07:14:25 +0100 | [diff] [blame] | 23 | bool |
Mike Frysinger | 555f386 | 2009-09-14 20:10:15 -0400 | [diff] [blame] | 24 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 25 | See Documentation/trace/ftrace-design.txt |
Frederic Weisbecker | 15e6cb3 | 2008-11-11 07:14:25 +0100 | [diff] [blame] | 26 | |
Steven Rostedt | 71e308a | 2009-06-18 12:45:08 -0400 | [diff] [blame] | 27 | config HAVE_FUNCTION_GRAPH_FP_TEST |
| 28 | bool |
| 29 | help |
Mike Frysinger | 0368897 | 2010-01-22 08:12:47 -0500 | [diff] [blame] | 30 | See Documentation/trace/ftrace-design.txt |
Steven Rostedt | 71e308a | 2009-06-18 12:45:08 -0400 | [diff] [blame] | 31 | |
Steven Rostedt | 60a7ecf | 2008-11-05 16:05:44 -0500 | [diff] [blame] | 32 | config HAVE_FUNCTION_TRACE_MCOUNT_TEST |
| 33 | bool |
| 34 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 35 | See Documentation/trace/ftrace-design.txt |
Steven Rostedt | 60a7ecf | 2008-11-05 16:05:44 -0500 | [diff] [blame] | 36 | |
Steven Rostedt | 677aa9f | 2008-05-17 00:01:36 -0400 | [diff] [blame] | 37 | config HAVE_DYNAMIC_FTRACE |
| 38 | bool |
Mike Frysinger | 555f386 | 2009-09-14 20:10:15 -0400 | [diff] [blame] | 39 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 40 | See Documentation/trace/ftrace-design.txt |
Steven Rostedt | 677aa9f | 2008-05-17 00:01:36 -0400 | [diff] [blame] | 41 | |
Steven Rostedt | 8da3821 | 2008-08-14 15:45:07 -0400 | [diff] [blame] | 42 | config HAVE_FTRACE_MCOUNT_RECORD |
| 43 | bool |
Mike Frysinger | 555f386 | 2009-09-14 20:10:15 -0400 | [diff] [blame] | 44 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 45 | See Documentation/trace/ftrace-design.txt |
Steven Rostedt | 8da3821 | 2008-08-14 15:45:07 -0400 | [diff] [blame] | 46 | |
Josh Stone | 6670000 | 2009-08-24 14:43:11 -0700 | [diff] [blame] | 47 | config HAVE_SYSCALL_TRACEPOINTS |
Frederic Weisbecker | ee08c6e | 2009-03-07 05:52:59 +0100 | [diff] [blame] | 48 | bool |
Mike Frysinger | 555f386 | 2009-09-14 20:10:15 -0400 | [diff] [blame] | 49 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 50 | See Documentation/trace/ftrace-design.txt |
Frederic Weisbecker | ee08c6e | 2009-03-07 05:52:59 +0100 | [diff] [blame] | 51 | |
Steven Rostedt | cf4db25 | 2010-10-14 23:32:44 -0400 | [diff] [blame] | 52 | config HAVE_C_RECORDMCOUNT |
Steven Rostedt | 72441cb | 2010-10-13 17:12:30 -0400 | [diff] [blame] | 53 | bool |
| 54 | help |
| 55 | C version of recordmcount available? |
| 56 | |
Steven Rostedt | 352ad25 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 57 | config TRACER_MAX_TRACE |
| 58 | bool |
| 59 | |
Steven Rostedt | 7a8e76a | 2008-09-29 23:02:38 -0400 | [diff] [blame] | 60 | config RING_BUFFER |
| 61 | bool |
| 62 | |
Steven Rostedt | 78d904b | 2009-02-05 18:43:07 -0500 | [diff] [blame] | 63 | config FTRACE_NMI_ENTER |
| 64 | bool |
| 65 | depends on HAVE_FTRACE_NMI_ENTER |
| 66 | default y |
| 67 | |
Tom Zanussi | 5f77a88 | 2009-04-08 03:14:01 -0500 | [diff] [blame] | 68 | config EVENT_TRACING |
Zhaolei | b11c53e | 2009-05-25 18:11:59 +0800 | [diff] [blame] | 69 | select CONTEXT_SWITCH_TRACER |
| 70 | bool |
| 71 | |
Thomas Renninger | 25e4193 | 2011-01-03 17:50:44 +0100 | [diff] [blame] | 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, |
Jesper Juhl | f629299 | 2011-07-24 23:15:42 +0200 | [diff] [blame] | 85 | namely 3.1. |
Thomas Renninger | 25e4193 | 2011-01-03 17:50:44 +0100 | [diff] [blame] | 86 | |
Zhaolei | b11c53e | 2009-05-25 18:11:59 +0800 | [diff] [blame] | 87 | config CONTEXT_SWITCH_TRACER |
Tom Zanussi | 5f77a88 | 2009-04-08 03:14:01 -0500 | [diff] [blame] | 88 | bool |
| 89 | |
Steven Rostedt | 85bac32 | 2009-09-04 14:24:40 -0400 | [diff] [blame] | 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 | |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 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 |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 101 | # hiding of the automatic options. |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 102 | |
Steven Rostedt | bc0c38d | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 103 | config TRACING |
| 104 | bool |
| 105 | select DEBUG_FS |
Steven Rostedt | 7a8e76a | 2008-09-29 23:02:38 -0400 | [diff] [blame] | 106 | select RING_BUFFER |
Al Viro | c2c8052 | 2008-10-31 19:50:41 +0000 | [diff] [blame] | 107 | select STACKTRACE if STACKTRACE_SUPPORT |
Ingo Molnar | 5f87f11 | 2008-07-23 14:15:22 +0200 | [diff] [blame] | 108 | select TRACEPOINTS |
Steven Rostedt | f3384b2 | 2008-10-29 11:15:57 -0400 | [diff] [blame] | 109 | select NOP_TRACER |
Frederic Weisbecker | 769b044 | 2009-03-06 17:21:49 +0100 | [diff] [blame] | 110 | select BINARY_PRINTF |
Tom Zanussi | 5f77a88 | 2009-04-08 03:14:01 -0500 | [diff] [blame] | 111 | select EVENT_TRACING |
Steven Rostedt | bc0c38d | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 112 | |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 113 | config GENERIC_TRACER |
| 114 | bool |
| 115 | select TRACING |
| 116 | |
Ingo Molnar | 40ada30 | 2009-03-05 21:19:55 +0100 | [diff] [blame] | 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 |
Anton Vorontsov | 45b9560 | 2009-03-24 01:07:24 +0300 | [diff] [blame] | 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 |
Ingo Molnar | 40ada30 | 2009-03-05 21:19:55 +0100 | [diff] [blame] | 128 | depends on STACKTRACE_SUPPORT |
KOSAKI Motohiro | 422d3c7 | 2009-03-06 10:40:53 +0900 | [diff] [blame] | 129 | default y |
Ingo Molnar | 40ada30 | 2009-03-05 21:19:55 +0100 | [diff] [blame] | 130 | |
| 131 | if TRACING_SUPPORT |
| 132 | |
Steven Rostedt | 4ed9f07 | 2009-04-20 10:47:36 -0400 | [diff] [blame] | 133 | menuconfig FTRACE |
| 134 | bool "Tracers" |
Steven Rostedt | 65b7724 | 2009-05-07 12:49:27 -0400 | [diff] [blame] | 135 | default y if DEBUG_KERNEL |
Steven Rostedt | 4ed9f07 | 2009-04-20 10:47:36 -0400 | [diff] [blame] | 136 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 137 | Enable the kernel tracing infrastructure. |
Steven Rostedt | 4ed9f07 | 2009-04-20 10:47:36 -0400 | [diff] [blame] | 138 | |
| 139 | if FTRACE |
Peter Zijlstra | 17d80fd | 2008-10-21 16:31:18 +0200 | [diff] [blame] | 140 | |
Steven Rostedt | 606576c | 2008-10-06 19:06:12 -0400 | [diff] [blame] | 141 | config FUNCTION_TRACER |
Steven Rostedt | 1b29b01 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 142 | bool "Kernel Function Tracer" |
Steven Rostedt | 606576c | 2008-10-06 19:06:12 -0400 | [diff] [blame] | 143 | depends on HAVE_FUNCTION_TRACER |
Michal Simek | d20ac25 | 2011-04-04 11:20:12 +0200 | [diff] [blame] | 144 | select FRAME_POINTER if !ARM_UNWIND && !S390 && !MICROBLAZE |
Steven Rostedt | 4d7a077 | 2009-02-18 22:06:18 -0500 | [diff] [blame] | 145 | select KALLSYMS |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 146 | select GENERIC_TRACER |
Steven Rostedt | 35e8e30 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 147 | select CONTEXT_SWITCH_TRACER |
Steven Rostedt | 1b29b01 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 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 |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 151 | instruction at the beginning of every kernel function, which NOP |
Steven Rostedt | 1b29b01 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 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. |
Steven Rostedt | 35e8e30 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 156 | |
Frederic Weisbecker | fb52607 | 2008-11-25 21:07:04 +0100 | [diff] [blame] | 157 | config FUNCTION_GRAPH_TRACER |
| 158 | bool "Kernel Function Graph Tracer" |
| 159 | depends on HAVE_FUNCTION_GRAPH_TRACER |
Frederic Weisbecker | 15e6cb3 | 2008-11-11 07:14:25 +0100 | [diff] [blame] | 160 | depends on FUNCTION_TRACER |
Steven Rostedt | eb4a037 | 2009-06-18 12:53:21 -0400 | [diff] [blame] | 161 | depends on !X86_32 || !CC_OPTIMIZE_FOR_SIZE |
Ingo Molnar | 764f3b9 | 2008-12-03 10:33:58 +0100 | [diff] [blame] | 162 | default y |
Frederic Weisbecker | 15e6cb3 | 2008-11-11 07:14:25 +0100 | [diff] [blame] | 163 | help |
Frederic Weisbecker | fb52607 | 2008-11-25 21:07:04 +0100 | [diff] [blame] | 164 | Enable the kernel to trace a function at both its return |
| 165 | and its entry. |
Matt LaPlante | 692105b | 2009-01-26 11:12:25 +0100 | [diff] [blame] | 166 | Its first purpose is to trace the duration of functions and |
| 167 | draw a call graph for each thread with some information like |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 168 | the return value. This is done by setting the current return |
Matt LaPlante | 692105b | 2009-01-26 11:12:25 +0100 | [diff] [blame] | 169 | address on the current task structure into a stack of calls. |
Frederic Weisbecker | 15e6cb3 | 2008-11-11 07:14:25 +0100 | [diff] [blame] | 170 | |
Steven Rostedt | bac429f | 2009-03-20 12:50:56 -0400 | [diff] [blame] | 171 | |
Steven Rostedt | 81d68a9 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 172 | config IRQSOFF_TRACER |
| 173 | bool "Interrupts-off Latency Tracer" |
| 174 | default n |
| 175 | depends on TRACE_IRQFLAGS_SUPPORT |
John Stultz | 592913e | 2010-07-13 17:56:20 -0700 | [diff] [blame] | 176 | depends on !ARCH_USES_GETTIMEOFFSET |
Steven Rostedt | 81d68a9 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 177 | select TRACE_IRQFLAGS |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 178 | select GENERIC_TRACER |
Steven Rostedt | 81d68a9 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 179 | select TRACER_MAX_TRACE |
Steven Rostedt | 85bac32 | 2009-09-04 14:24:40 -0400 | [diff] [blame] | 180 | select RING_BUFFER_ALLOW_SWAP |
Steven Rostedt | 81d68a9 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 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 | |
GeunSik Lim | 156f5a7 | 2009-06-02 15:01:37 +0900 | [diff] [blame] | 189 | echo 0 > /sys/kernel/debug/tracing/tracing_max_latency |
Steven Rostedt | 81d68a9 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 190 | |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 191 | (Note that kernel size and overhead increase with this option |
Steven Rostedt | 6cd8a4b | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 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 |
John Stultz | 592913e | 2010-07-13 17:56:20 -0700 | [diff] [blame] | 198 | depends on !ARCH_USES_GETTIMEOFFSET |
Steven Rostedt | 6cd8a4b | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 199 | depends on PREEMPT |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 200 | select GENERIC_TRACER |
Steven Rostedt | 6cd8a4b | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 201 | select TRACER_MAX_TRACE |
Steven Rostedt | 85bac32 | 2009-09-04 14:24:40 -0400 | [diff] [blame] | 202 | select RING_BUFFER_ALLOW_SWAP |
Steven Rostedt | 6cd8a4b | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 203 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 204 | This option measures the time spent in preemption-off critical |
Steven Rostedt | 6cd8a4b | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 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 | |
GeunSik Lim | 156f5a7 | 2009-06-02 15:01:37 +0900 | [diff] [blame] | 211 | echo 0 > /sys/kernel/debug/tracing/tracing_max_latency |
Steven Rostedt | 6cd8a4b | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 212 | |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 213 | (Note that kernel size and overhead increase with this option |
Steven Rostedt | 6cd8a4b | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 214 | enabled. This option and the irqs-off timing option can be |
| 215 | used together or separately.) |
| 216 | |
Steven Rostedt | 352ad25 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 217 | config SCHED_TRACER |
| 218 | bool "Scheduling Latency Tracer" |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 219 | select GENERIC_TRACER |
Steven Rostedt | 352ad25 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 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 | |
Steven Rostedt | 897f17a | 2009-05-28 16:31:21 -0400 | [diff] [blame] | 226 | config ENABLE_DEFAULT_TRACERS |
| 227 | bool "Trace process context switches and events" |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 228 | depends on !GENERIC_TRACER |
Steven Rostedt | b77e38a | 2009-02-24 10:21:36 -0500 | [diff] [blame] | 229 | select TRACING |
| 230 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 231 | This tracer hooks to various trace points in the kernel, |
Steven Rostedt | b77e38a | 2009-02-24 10:21:36 -0500 | [diff] [blame] | 232 | allowing the user to pick and choose which trace point they |
Steven Rostedt | 897f17a | 2009-05-28 16:31:21 -0400 | [diff] [blame] | 233 | want to trace. It also includes the sched_switch tracer plugin. |
Steven Rostedt | a7abe97 | 2009-04-20 10:59:34 -0400 | [diff] [blame] | 234 | |
Frederic Weisbecker | ee08c6e | 2009-03-07 05:52:59 +0100 | [diff] [blame] | 235 | config FTRACE_SYSCALLS |
| 236 | bool "Trace syscalls" |
Josh Stone | 6670000 | 2009-08-24 14:43:11 -0700 | [diff] [blame] | 237 | depends on HAVE_SYSCALL_TRACEPOINTS |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 238 | select GENERIC_TRACER |
Frederic Weisbecker | 0ea1c41 | 2009-03-15 22:10:38 +0100 | [diff] [blame] | 239 | select KALLSYMS |
Frederic Weisbecker | ee08c6e | 2009-03-07 05:52:59 +0100 | [diff] [blame] | 240 | help |
| 241 | Basic tracer to catch the syscall entry and exit events. |
| 242 | |
Steven Rostedt | 2ed84ee | 2008-11-12 15:24:24 -0500 | [diff] [blame] | 243 | config TRACE_BRANCH_PROFILING |
Steven Rostedt | 9ae5b87 | 2009-04-20 10:27:58 -0400 | [diff] [blame] | 244 | bool |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 245 | select GENERIC_TRACER |
Steven Rostedt | 9ae5b87 | 2009-04-20 10:27:58 -0400 | [diff] [blame] | 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 | |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 257 | The "all branch" profiler will profile every if-statement in the |
Steven Rostedt | 9ae5b87 | 2009-04-20 10:27:58 -0400 | [diff] [blame] | 258 | kernel. This profiler will also enable the likely/unlikely |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 259 | profiler. |
Steven Rostedt | 9ae5b87 | 2009-04-20 10:27:58 -0400 | [diff] [blame] | 260 | |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 261 | Either of the above profilers adds a bit of overhead to the system. |
| 262 | If unsure, choose "No branch profiling". |
Steven Rostedt | 9ae5b87 | 2009-04-20 10:27:58 -0400 | [diff] [blame] | 263 | |
| 264 | config BRANCH_PROFILE_NONE |
| 265 | bool "No branch profiling" |
| 266 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 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. |
Steven Rostedt | 9ae5b87 | 2009-04-20 10:27:58 -0400 | [diff] [blame] | 270 | |
| 271 | config PROFILE_ANNOTATED_BRANCHES |
| 272 | bool "Trace likely/unlikely profiler" |
| 273 | select TRACE_BRANCH_PROFILING |
Steven Rostedt | 1f0d69a | 2008-11-12 00:14:39 -0500 | [diff] [blame] | 274 | help |
| 275 | This tracer profiles all the the likely and unlikely macros |
| 276 | in the kernel. It will display the results in: |
| 277 | |
David Rientjes | 13e5bef | 2011-03-16 17:17:08 -0700 | [diff] [blame] | 278 | /sys/kernel/debug/tracing/trace_stat/branch_annotated |
Steven Rostedt | 1f0d69a | 2008-11-12 00:14:39 -0500 | [diff] [blame] | 279 | |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 280 | Note: this will add a significant overhead; only turn this |
Steven Rostedt | 1f0d69a | 2008-11-12 00:14:39 -0500 | [diff] [blame] | 281 | on if you need to profile the system's use of these macros. |
| 282 | |
Steven Rostedt | 2bcd521 | 2008-11-21 01:30:54 -0500 | [diff] [blame] | 283 | config PROFILE_ALL_BRANCHES |
| 284 | bool "Profile all if conditionals" |
Steven Rostedt | 9ae5b87 | 2009-04-20 10:27:58 -0400 | [diff] [blame] | 285 | select TRACE_BRANCH_PROFILING |
Steven Rostedt | 2bcd521 | 2008-11-21 01:30:54 -0500 | [diff] [blame] | 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 | |
David Rientjes | 13e5bef | 2011-03-16 17:17:08 -0700 | [diff] [blame] | 291 | /sys/kernel/debug/tracing/trace_stat/branch_all |
Steven Rostedt | 2bcd521 | 2008-11-21 01:30:54 -0500 | [diff] [blame] | 292 | |
Steven Rostedt | 9ae5b87 | 2009-04-20 10:27:58 -0400 | [diff] [blame] | 293 | This option also enables the likely/unlikely profiler. |
| 294 | |
Steven Rostedt | 2bcd521 | 2008-11-21 01:30:54 -0500 | [diff] [blame] | 295 | This configuration, when enabled, will impose a great overhead |
| 296 | on the system. This should only be enabled when the system |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 297 | is to be analyzed in much detail. |
Steven Rostedt | 9ae5b87 | 2009-04-20 10:27:58 -0400 | [diff] [blame] | 298 | endchoice |
Steven Rostedt | 2bcd521 | 2008-11-21 01:30:54 -0500 | [diff] [blame] | 299 | |
Steven Rostedt | 2ed84ee | 2008-11-12 15:24:24 -0500 | [diff] [blame] | 300 | config TRACING_BRANCHES |
Steven Rostedt | 52f232c | 2008-11-12 00:14:40 -0500 | [diff] [blame] | 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 | |
Steven Rostedt | 2ed84ee | 2008-11-12 15:24:24 -0500 | [diff] [blame] | 308 | config BRANCH_TRACER |
Steven Rostedt | 52f232c | 2008-11-12 00:14:40 -0500 | [diff] [blame] | 309 | bool "Trace likely/unlikely instances" |
Steven Rostedt | 2ed84ee | 2008-11-12 15:24:24 -0500 | [diff] [blame] | 310 | depends on TRACE_BRANCH_PROFILING |
| 311 | select TRACING_BRANCHES |
Steven Rostedt | 52f232c | 2008-11-12 00:14:40 -0500 | [diff] [blame] | 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 | |
Steven Rostedt | e5a81b6 | 2008-08-27 23:31:01 -0400 | [diff] [blame] | 322 | config STACK_TRACER |
| 323 | bool "Trace max stack" |
Steven Rostedt | 606576c | 2008-10-06 19:06:12 -0400 | [diff] [blame] | 324 | depends on HAVE_FUNCTION_TRACER |
Steven Rostedt | 606576c | 2008-10-06 19:06:12 -0400 | [diff] [blame] | 325 | select FUNCTION_TRACER |
Steven Rostedt | e5a81b6 | 2008-08-27 23:31:01 -0400 | [diff] [blame] | 326 | select STACKTRACE |
Steven Rostedt | 4d7a077 | 2009-02-18 22:06:18 -0500 | [diff] [blame] | 327 | select KALLSYMS |
Steven Rostedt | e5a81b6 | 2008-08-27 23:31:01 -0400 | [diff] [blame] | 328 | help |
Ingo Molnar | 4519d9e | 2008-10-14 14:15:43 +0200 | [diff] [blame] | 329 | This special tracer records the maximum stack footprint of the |
GeunSik Lim | 156f5a7 | 2009-06-02 15:01:37 +0900 | [diff] [blame] | 330 | kernel and displays it in /sys/kernel/debug/tracing/stack_trace. |
Ingo Molnar | 4519d9e | 2008-10-14 14:15:43 +0200 | [diff] [blame] | 331 | |
| 332 | This tracer works by hooking into every function call that the |
| 333 | kernel executes, and keeping a maximum stack depth value and |
Steven Rostedt | f38f1d2 | 2008-12-16 23:06:40 -0500 | [diff] [blame] | 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 |
Ingo Molnar | 4519d9e | 2008-10-14 14:15:43 +0200 | [diff] [blame] | 343 | |
| 344 | Say N if unsure. |
Steven Rostedt | e5a81b6 | 2008-08-27 23:31:01 -0400 | [diff] [blame] | 345 | |
Frederic Weisbecker | 2db270a | 2009-02-07 20:46:45 +0100 | [diff] [blame] | 346 | config BLK_DEV_IO_TRACE |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 347 | bool "Support for tracing block IO actions" |
Frederic Weisbecker | 2db270a | 2009-02-07 20:46:45 +0100 | [diff] [blame] | 348 | depends on SYSFS |
Ingo Molnar | 1dfba05 | 2009-02-09 12:06:54 +0100 | [diff] [blame] | 349 | depends on BLOCK |
Frederic Weisbecker | 2db270a | 2009-02-07 20:46:45 +0100 | [diff] [blame] | 350 | select RELAY |
| 351 | select DEBUG_FS |
| 352 | select TRACEPOINTS |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 353 | select GENERIC_TRACER |
Frederic Weisbecker | 2db270a | 2009-02-07 20:46:45 +0100 | [diff] [blame] | 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. |
Frederic Weisbecker | 36994e5 | 2008-12-29 13:42:23 -0800 | [diff] [blame] | 370 | |
Masami Hiramatsu | 77b44d1 | 2009-11-03 19:12:47 -0500 | [diff] [blame] | 371 | config KPROBE_EVENT |
Masami Hiramatsu | 413d37d | 2009-08-13 16:35:11 -0400 | [diff] [blame] | 372 | depends on KPROBES |
Heiko Carstens | f850c30 | 2010-02-10 17:25:17 +0100 | [diff] [blame] | 373 | depends on HAVE_REGS_AND_STACK_ACCESS_API |
Masami Hiramatsu | 77b44d1 | 2009-11-03 19:12:47 -0500 | [diff] [blame] | 374 | bool "Enable kprobes-based dynamic events" |
Masami Hiramatsu | 413d37d | 2009-08-13 16:35:11 -0400 | [diff] [blame] | 375 | select TRACING |
Masami Hiramatsu | 77b44d1 | 2009-11-03 19:12:47 -0500 | [diff] [blame] | 376 | default y |
Masami Hiramatsu | 413d37d | 2009-08-13 16:35:11 -0400 | [diff] [blame] | 377 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 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. |
Masami Hiramatsu | 77b44d1 | 2009-11-03 19:12:47 -0500 | [diff] [blame] | 381 | |
| 382 | Those events can be inserted wherever kprobes can probe, and record |
| 383 | various register and memory values. |
| 384 | |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 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. |
Masami Hiramatsu | 413d37d | 2009-08-13 16:35:11 -0400 | [diff] [blame] | 387 | |
Steven Rostedt | 3d08339 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 388 | config DYNAMIC_FTRACE |
| 389 | bool "enable/disable ftrace tracepoints dynamically" |
Steven Rostedt | 606576c | 2008-10-06 19:06:12 -0400 | [diff] [blame] | 390 | depends on FUNCTION_TRACER |
Steven Rostedt | 677aa9f | 2008-05-17 00:01:36 -0400 | [diff] [blame] | 391 | depends on HAVE_DYNAMIC_FTRACE |
Steven Rostedt | 3d08339 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 392 | default y |
| 393 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 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. |
Steven Rostedt | 3d08339 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 398 | |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 399 | This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but |
| 400 | otherwise has native performance as long as no tracing is active. |
Steven Rostedt | 3d08339 | 2008-05-12 21:20:42 +0200 | [diff] [blame] | 401 | |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 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. |
Steven Rostedt | 60a1177 | 2008-05-12 21:20:44 +0200 | [diff] [blame] | 406 | |
Steven Rostedt | bac429f | 2009-03-20 12:50:56 -0400 | [diff] [blame] | 407 | config FUNCTION_PROFILER |
| 408 | bool "Kernel function profiler" |
Steven Rostedt | 493762f | 2009-03-23 17:12:36 -0400 | [diff] [blame] | 409 | depends on FUNCTION_TRACER |
Steven Rostedt | bac429f | 2009-03-20 12:50:56 -0400 | [diff] [blame] | 410 | default n |
| 411 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 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. |
Steven Rostedt | bac429f | 2009-03-20 12:50:56 -0400 | [diff] [blame] | 418 | |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 419 | If in doubt, say N. |
Steven Rostedt | bac429f | 2009-03-20 12:50:56 -0400 | [diff] [blame] | 420 | |
Steven Rostedt | 8da3821 | 2008-08-14 15:45:07 -0400 | [diff] [blame] | 421 | config FTRACE_MCOUNT_RECORD |
| 422 | def_bool y |
| 423 | depends on DYNAMIC_FTRACE |
| 424 | depends on HAVE_FTRACE_MCOUNT_RECORD |
| 425 | |
Steven Rostedt | 60a1177 | 2008-05-12 21:20:44 +0200 | [diff] [blame] | 426 | config FTRACE_SELFTEST |
| 427 | bool |
| 428 | |
| 429 | config FTRACE_STARTUP_TEST |
| 430 | bool "Perform a startup test on ftrace" |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 431 | depends on GENERIC_TRACER |
Steven Rostedt | 60a1177 | 2008-05-12 21:20:44 +0200 | [diff] [blame] | 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. |
Peter Zijlstra | 17d80fd | 2008-10-21 16:31:18 +0200 | [diff] [blame] | 438 | |
Steven Rostedt | 1f5a6b4 | 2009-09-14 11:58:24 -0400 | [diff] [blame] | 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 | |
Pekka Paalanen | fe6f90e | 2009-01-03 21:23:51 +0200 | [diff] [blame] | 451 | config MMIOTRACE |
| 452 | bool "Memory mapped IO tracing" |
Ingo Molnar | 40ada30 | 2009-03-05 21:19:55 +0100 | [diff] [blame] | 453 | depends on HAVE_MMIOTRACE_SUPPORT && PCI |
Steven Rostedt | 5e0a093 | 2009-05-28 15:50:13 -0400 | [diff] [blame] | 454 | select GENERIC_TRACER |
Pekka Paalanen | fe6f90e | 2009-01-03 21:23:51 +0200 | [diff] [blame] | 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 | |
Li Zefan | 4d1f437 | 2009-04-10 08:48:36 +0800 | [diff] [blame] | 461 | See Documentation/trace/mmiotrace.txt. |
Pekka Paalanen | fe6f90e | 2009-01-03 21:23:51 +0200 | [diff] [blame] | 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 | |
Steven Rostedt | 5092dbc | 2009-05-05 22:47:18 -0400 | [diff] [blame] | 474 | config RING_BUFFER_BENCHMARK |
| 475 | tristate "Ring buffer benchmark stress tester" |
| 476 | depends on RING_BUFFER |
| 477 | help |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 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 |
Steven Rostedt | 5092dbc | 2009-05-05 22:47:18 -0400 | [diff] [blame] | 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 | |
Randy Dunlap | 4089236 | 2009-12-21 12:01:17 -0800 | [diff] [blame] | 488 | If unsure, say N. |
Steven Rostedt | 5092dbc | 2009-05-05 22:47:18 -0400 | [diff] [blame] | 489 | |
Steven Rostedt | 4ed9f07 | 2009-04-20 10:47:36 -0400 | [diff] [blame] | 490 | endif # FTRACE |
Ingo Molnar | 40ada30 | 2009-03-05 21:19:55 +0100 | [diff] [blame] | 491 | |
| 492 | endif # TRACING_SUPPORT |
| 493 | |