Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1 | ftrace - Function Tracer |
| 2 | ======================== |
| 3 | |
| 4 | Copyright 2008 Red Hat Inc. |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 5 | Author: Steven Rostedt <srostedt@redhat.com> |
| 6 | License: The GNU Free Documentation License, Version 1.2 |
Steven Rostedt | a97762a | 2008-07-31 12:40:52 -0400 | [diff] [blame] | 7 | (dual licensed under the GPL v2) |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 8 | Reviewers: Elias Oltmanns, Randy Dunlap, Andrew Morton, |
| 9 | John Kacur, and David Teigland. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 10 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 11 | Written for: 2.6.27-rc1 |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 12 | |
| 13 | Introduction |
| 14 | ------------ |
| 15 | |
| 16 | Ftrace is an internal tracer designed to help out developers and |
| 17 | designers of systems to find what is going on inside the kernel. |
| 18 | It can be used for debugging or analyzing latencies and performance |
| 19 | issues that take place outside of user-space. |
| 20 | |
| 21 | Although ftrace is the function tracer, it also includes an |
| 22 | infrastructure that allows for other types of tracing. Some of the |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 23 | tracers that are currently in ftrace include a tracer to trace |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 24 | context switches, the time it takes for a high priority task to |
| 25 | run after it was woken up, the time interrupts are disabled, and |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 26 | more (ftrace allows for tracer plugins, which means that the list of |
| 27 | tracers can always grow). |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 28 | |
| 29 | |
| 30 | The File System |
| 31 | --------------- |
| 32 | |
| 33 | Ftrace uses the debugfs file system to hold the control files as well |
| 34 | as the files to display output. |
| 35 | |
| 36 | To mount the debugfs system: |
| 37 | |
| 38 | # mkdir /debug |
| 39 | # mount -t debugfs nodev /debug |
| 40 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 41 | (Note: it is more common to mount at /sys/kernel/debug, but for simplicity |
| 42 | this document will use /debug) |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 43 | |
| 44 | That's it! (assuming that you have ftrace configured into your kernel) |
| 45 | |
| 46 | After mounting the debugfs, you can see a directory called |
| 47 | "tracing". This directory contains the control and output files |
| 48 | of ftrace. Here is a list of some of the key files: |
| 49 | |
| 50 | |
| 51 | Note: all time values are in microseconds. |
| 52 | |
| 53 | current_tracer : This is used to set or display the current tracer |
| 54 | that is configured. |
| 55 | |
| 56 | available_tracers : This holds the different types of tracers that |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 57 | have been compiled into the kernel. The tracers |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 58 | listed here can be configured by echoing their name |
| 59 | into current_tracer. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 60 | |
| 61 | tracing_enabled : This sets or displays whether the current_tracer |
| 62 | is activated and tracing or not. Echo 0 into this |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 63 | file to disable the tracer or 1 to enable it. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 64 | |
| 65 | trace : This file holds the output of the trace in a human readable |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 66 | format (described below). |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 67 | |
| 68 | latency_trace : This file shows the same trace but the information |
| 69 | is organized more to display possible latencies |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 70 | in the system (described below). |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 71 | |
| 72 | trace_pipe : The output is the same as the "trace" file but this |
| 73 | file is meant to be streamed with live tracing. |
| 74 | Reads from this file will block until new data |
| 75 | is retrieved. Unlike the "trace" and "latency_trace" |
| 76 | files, this file is a consumer. This means reading |
| 77 | from this file causes sequential reads to display |
| 78 | more current data. Once data is read from this |
| 79 | file, it is consumed, and will not be read |
| 80 | again with a sequential read. The "trace" and |
| 81 | "latency_trace" files are static, and if the |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 82 | tracer is not adding more data, they will display |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 83 | the same information every time they are read. |
| 84 | |
| 85 | iter_ctrl : This file lets the user control the amount of data |
| 86 | that is displayed in one of the above output |
| 87 | files. |
| 88 | |
| 89 | trace_max_latency : Some of the tracers record the max latency. |
| 90 | For example, the time interrupts are disabled. |
| 91 | This time is saved in this file. The max trace |
| 92 | will also be stored, and displayed by either |
| 93 | "trace" or "latency_trace". A new max trace will |
| 94 | only be recorded if the latency is greater than |
| 95 | the value in this file. (in microseconds) |
| 96 | |
| 97 | trace_entries : This sets or displays the number of trace |
| 98 | entries each CPU buffer can hold. The tracer buffers |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 99 | are the same size for each CPU. The displayed number |
| 100 | is the size of the CPU buffer and not total size. The |
| 101 | trace buffers are allocated in pages (blocks of memory |
| 102 | that the kernel uses for allocation, usually 4 KB in size). |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 103 | Since each entry is smaller than a page, if the last |
| 104 | allocated page has room for more entries than were |
| 105 | requested, the rest of the page is used to allocate |
| 106 | entries. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 107 | |
| 108 | This can only be updated when the current_tracer |
| 109 | is set to "none". |
| 110 | |
| 111 | NOTE: It is planned on changing the allocated buffers |
| 112 | from being the number of possible CPUS to |
| 113 | the number of online CPUS. |
| 114 | |
| 115 | tracing_cpumask : This is a mask that lets the user only trace |
| 116 | on specified CPUS. The format is a hex string |
| 117 | representing the CPUS. |
| 118 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 119 | set_ftrace_filter : When dynamic ftrace is configured in (see the |
| 120 | section below "dynamic ftrace"), the code is dynamically |
| 121 | modified (code text rewrite) to disable calling of the |
| 122 | function profiler (mcount). This lets tracing be configured |
| 123 | in with practically no overhead in performance. This also |
| 124 | has a side effect of enabling or disabling specific functions |
| 125 | to be traced. Echoing names of functions into this file |
| 126 | will limit the trace to only those functions. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 127 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 128 | set_ftrace_notrace: This has an effect opposite to that of |
| 129 | set_ftrace_filter. Any function that is added here will not |
| 130 | be traced. If a function exists in both set_ftrace_filter |
| 131 | and set_ftrace_notrace, the function will _not_ be traced. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 132 | |
| 133 | available_filter_functions : When a function is encountered the first |
| 134 | time by the dynamic tracer, it is recorded and |
| 135 | later the call is converted into a nop. This file |
| 136 | lists the functions that have been recorded |
| 137 | by the dynamic tracer and these functions can |
| 138 | be used to set the ftrace filter by the above |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 139 | "set_ftrace_filter" file. (See the section "dynamic ftrace" |
| 140 | below for more details). |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 141 | |
| 142 | |
| 143 | The Tracers |
| 144 | ----------- |
| 145 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 146 | Here is the list of current tracers that may be configured. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 147 | |
| 148 | ftrace - function tracer that uses mcount to trace all functions. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 149 | |
| 150 | sched_switch - traces the context switches between tasks. |
| 151 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 152 | irqsoff - traces the areas that disable interrupts and saves |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 153 | the trace with the longest max latency. |
| 154 | See tracing_max_latency. When a new max is recorded, |
| 155 | it replaces the old trace. It is best to view this |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 156 | trace via the latency_trace file. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 157 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 158 | preemptoff - Similar to irqsoff but traces and records the amount of |
| 159 | time for which preemption is disabled. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 160 | |
| 161 | preemptirqsoff - Similar to irqsoff and preemptoff, but traces and |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 162 | records the largest time for which irqs and/or preemption |
| 163 | is disabled. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 164 | |
| 165 | wakeup - Traces and records the max latency that it takes for |
| 166 | the highest priority task to get scheduled after |
| 167 | it has been woken up. |
| 168 | |
| 169 | none - This is not a tracer. To remove all tracers from tracing |
| 170 | simply echo "none" into current_tracer. |
| 171 | |
| 172 | |
| 173 | Examples of using the tracer |
| 174 | ---------------------------- |
| 175 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 176 | Here are typical examples of using the tracers when controlling them only |
| 177 | with the debugfs interface (without using any user-land utilities). |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 178 | |
| 179 | Output format: |
| 180 | -------------- |
| 181 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 182 | Here is an example of the output format of the file "trace" |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 183 | |
| 184 | -------- |
| 185 | # tracer: ftrace |
| 186 | # |
| 187 | # TASK-PID CPU# TIMESTAMP FUNCTION |
| 188 | # | | | | | |
| 189 | bash-4251 [01] 10152.583854: path_put <-path_walk |
| 190 | bash-4251 [01] 10152.583855: dput <-path_put |
| 191 | bash-4251 [01] 10152.583855: _atomic_dec_and_lock <-dput |
| 192 | -------- |
| 193 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 194 | A header is printed with the tracer name that is represented by the trace. |
| 195 | In this case the tracer is "ftrace". Then a header showing the format. Task |
| 196 | name "bash", the task PID "4251", the CPU that it was running on |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 197 | "01", the timestamp in <secs>.<usecs> format, the function name that was |
| 198 | traced "path_put" and the parent function that called this function |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 199 | "path_walk". The timestamp is the time at which the function was |
| 200 | entered. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 201 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 202 | The sched_switch tracer also includes tracing of task wakeups and |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 203 | context switches. |
| 204 | |
| 205 | ksoftirqd/1-7 [01] 1453.070013: 7:115:R + 2916:115:S |
| 206 | ksoftirqd/1-7 [01] 1453.070013: 7:115:R + 10:115:S |
| 207 | ksoftirqd/1-7 [01] 1453.070013: 7:115:R ==> 10:115:R |
| 208 | events/1-10 [01] 1453.070013: 10:115:S ==> 2916:115:R |
| 209 | kondemand/1-2916 [01] 1453.070013: 2916:115:S ==> 7:115:R |
| 210 | ksoftirqd/1-7 [01] 1453.070013: 7:115:S ==> 0:140:R |
| 211 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 212 | Wake ups are represented by a "+" and the context switches are shown as |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 213 | "==>". The format is: |
| 214 | |
| 215 | Context switches: |
| 216 | |
| 217 | Previous task Next Task |
| 218 | |
| 219 | <pid>:<prio>:<state> ==> <pid>:<prio>:<state> |
| 220 | |
| 221 | Wake ups: |
| 222 | |
| 223 | Current task Task waking up |
| 224 | |
| 225 | <pid>:<prio>:<state> + <pid>:<prio>:<state> |
| 226 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 227 | The prio is the internal kernel priority, which is the inverse of the |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 228 | priority that is usually displayed by user-space tools. Zero represents |
| 229 | the highest priority (99). Prio 100 starts the "nice" priorities with |
| 230 | 100 being equal to nice -20 and 139 being nice 19. The prio "140" is |
| 231 | reserved for the idle task which is the lowest priority thread (pid 0). |
| 232 | |
| 233 | |
| 234 | Latency trace format |
| 235 | -------------------- |
| 236 | |
| 237 | For traces that display latency times, the latency_trace file gives |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 238 | somewhat more information to see why a latency happened. Here is a typical |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 239 | trace. |
| 240 | |
| 241 | # tracer: irqsoff |
| 242 | # |
| 243 | irqsoff latency trace v1.1.5 on 2.6.26-rc8 |
| 244 | -------------------------------------------------------------------- |
| 245 | latency: 97 us, #3/3, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) |
| 246 | ----------------- |
| 247 | | task: swapper-0 (uid:0 nice:0 policy:0 rt_prio:0) |
| 248 | ----------------- |
| 249 | => started at: apic_timer_interrupt |
| 250 | => ended at: do_softirq |
| 251 | |
| 252 | # _------=> CPU# |
| 253 | # / _-----=> irqs-off |
| 254 | # | / _----=> need-resched |
| 255 | # || / _---=> hardirq/softirq |
| 256 | # ||| / _--=> preempt-depth |
| 257 | # |||| / |
| 258 | # ||||| delay |
| 259 | # cmd pid ||||| time | caller |
| 260 | # \ / ||||| \ | / |
| 261 | <idle>-0 0d..1 0us+: trace_hardirqs_off_thunk (apic_timer_interrupt) |
| 262 | <idle>-0 0d.s. 97us : __do_softirq (do_softirq) |
| 263 | <idle>-0 0d.s1 98us : trace_hardirqs_on (do_softirq) |
| 264 | |
| 265 | |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 266 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 267 | This shows that the current tracer is "irqsoff" tracing the time for which |
| 268 | interrupts were disabled. It gives the trace version and the version |
| 269 | of the kernel upon which this was executed on (2.6.26-rc8). Then it displays |
| 270 | the max latency in microsecs (97 us). The number of trace entries displayed |
| 271 | and the total number recorded (both are three: #3/3). The type of |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 272 | preemption that was used (PREEMPT). VP, KP, SP, and HP are always zero |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 273 | and are reserved for later use. #P is the number of online CPUS (#P:2). |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 274 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 275 | The task is the process that was running when the latency occurred. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 276 | (swapper pid: 0). |
| 277 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 278 | The start and stop (the functions in which the interrupts were disabled and |
| 279 | enabled respectively) that caused the latencies: |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 280 | |
| 281 | apic_timer_interrupt is where the interrupts were disabled. |
| 282 | do_softirq is where they were enabled again. |
| 283 | |
| 284 | The next lines after the header are the trace itself. The header |
| 285 | explains which is which. |
| 286 | |
| 287 | cmd: The name of the process in the trace. |
| 288 | |
| 289 | pid: The PID of that process. |
| 290 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 291 | CPU#: The CPU which the process was running on. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 292 | |
| 293 | irqs-off: 'd' interrupts are disabled. '.' otherwise. |
Steven Rostedt | 9244489 | 2008-10-24 09:42:59 -0400 | [diff] [blame] | 294 | Note: If the architecture does not support a way to |
| 295 | read the irq flags variable, an 'X' will always |
| 296 | be printed here. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 297 | |
| 298 | need-resched: 'N' task need_resched is set, '.' otherwise. |
| 299 | |
| 300 | hardirq/softirq: |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 301 | 'H' - hard irq occurred inside a softirq. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 302 | 'h' - hard irq is running |
| 303 | 's' - soft irq is running |
| 304 | '.' - normal context. |
| 305 | |
| 306 | preempt-depth: The level of preempt_disabled |
| 307 | |
| 308 | The above is mostly meaningful for kernel developers. |
| 309 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 310 | time: This differs from the trace file output. The trace file output |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 311 | includes an absolute timestamp. The timestamp used by the |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 312 | latency_trace file is relative to the start of the trace. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 313 | |
| 314 | delay: This is just to help catch your eye a bit better. And |
| 315 | needs to be fixed to be only relative to the same CPU. |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 316 | The marks are determined by the difference between this |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 317 | current trace and the next trace. |
| 318 | '!' - greater than preempt_mark_thresh (default 100) |
| 319 | '+' - greater than 1 microsecond |
| 320 | ' ' - less than or equal to 1 microsecond. |
| 321 | |
| 322 | The rest is the same as the 'trace' file. |
| 323 | |
| 324 | |
| 325 | iter_ctrl |
| 326 | --------- |
| 327 | |
| 328 | The iter_ctrl file is used to control what gets printed in the trace |
| 329 | output. To see what is available, simply cat the file: |
| 330 | |
| 331 | cat /debug/tracing/iter_ctrl |
| 332 | print-parent nosym-offset nosym-addr noverbose noraw nohex nobin \ |
| 333 | noblock nostacktrace nosched-tree |
| 334 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 335 | To disable one of the options, echo in the option prepended with "no". |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 336 | |
| 337 | echo noprint-parent > /debug/tracing/iter_ctrl |
| 338 | |
| 339 | To enable an option, leave off the "no". |
| 340 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 341 | echo sym-offset > /debug/tracing/iter_ctrl |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 342 | |
| 343 | Here are the available options: |
| 344 | |
| 345 | print-parent - On function traces, display the calling function |
| 346 | as well as the function being traced. |
| 347 | |
| 348 | print-parent: |
| 349 | bash-4000 [01] 1477.606694: simple_strtoul <-strict_strtoul |
| 350 | |
| 351 | noprint-parent: |
| 352 | bash-4000 [01] 1477.606694: simple_strtoul |
| 353 | |
| 354 | |
| 355 | sym-offset - Display not only the function name, but also the offset |
| 356 | in the function. For example, instead of seeing just |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 357 | "ktime_get", you will see "ktime_get+0xb/0x20". |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 358 | |
| 359 | sym-offset: |
| 360 | bash-4000 [01] 1477.606694: simple_strtoul+0x6/0xa0 |
| 361 | |
| 362 | sym-addr - this will also display the function address as well as |
| 363 | the function name. |
| 364 | |
| 365 | sym-addr: |
| 366 | bash-4000 [01] 1477.606694: simple_strtoul <c0339346> |
| 367 | |
| 368 | verbose - This deals with the latency_trace file. |
| 369 | |
| 370 | bash 4000 1 0 00000000 00010a95 [58127d26] 1720.415ms \ |
| 371 | (+0.000ms): simple_strtoul (strict_strtoul) |
| 372 | |
| 373 | raw - This will display raw numbers. This option is best for use with |
| 374 | user applications that can translate the raw numbers better than |
| 375 | having it done in the kernel. |
| 376 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 377 | hex - Similar to raw, but the numbers will be in a hexadecimal format. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 378 | |
| 379 | bin - This will print out the formats in raw binary. |
| 380 | |
| 381 | block - TBD (needs update) |
| 382 | |
| 383 | stacktrace - This is one of the options that changes the trace itself. |
| 384 | When a trace is recorded, so is the stack of functions. |
| 385 | This allows for back traces of trace sites. |
| 386 | |
| 387 | sched-tree - TBD (any users??) |
| 388 | |
| 389 | |
| 390 | sched_switch |
| 391 | ------------ |
| 392 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 393 | This tracer simply records schedule switches. Here is an example |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 394 | of how to use it. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 395 | |
| 396 | # echo sched_switch > /debug/tracing/current_tracer |
| 397 | # echo 1 > /debug/tracing/tracing_enabled |
| 398 | # sleep 1 |
| 399 | # echo 0 > /debug/tracing/tracing_enabled |
| 400 | # cat /debug/tracing/trace |
| 401 | |
| 402 | # tracer: sched_switch |
| 403 | # |
| 404 | # TASK-PID CPU# TIMESTAMP FUNCTION |
| 405 | # | | | | | |
| 406 | bash-3997 [01] 240.132281: 3997:120:R + 4055:120:R |
| 407 | bash-3997 [01] 240.132284: 3997:120:R ==> 4055:120:R |
| 408 | sleep-4055 [01] 240.132371: 4055:120:S ==> 3997:120:R |
| 409 | bash-3997 [01] 240.132454: 3997:120:R + 4055:120:S |
| 410 | bash-3997 [01] 240.132457: 3997:120:R ==> 4055:120:R |
| 411 | sleep-4055 [01] 240.132460: 4055:120:D ==> 3997:120:R |
| 412 | bash-3997 [01] 240.132463: 3997:120:R + 4055:120:D |
| 413 | bash-3997 [01] 240.132465: 3997:120:R ==> 4055:120:R |
| 414 | <idle>-0 [00] 240.132589: 0:140:R + 4:115:S |
| 415 | <idle>-0 [00] 240.132591: 0:140:R ==> 4:115:R |
| 416 | ksoftirqd/0-4 [00] 240.132595: 4:115:S ==> 0:140:R |
| 417 | <idle>-0 [00] 240.132598: 0:140:R + 4:115:S |
| 418 | <idle>-0 [00] 240.132599: 0:140:R ==> 4:115:R |
| 419 | ksoftirqd/0-4 [00] 240.132603: 4:115:S ==> 0:140:R |
| 420 | sleep-4055 [01] 240.133058: 4055:120:S ==> 3997:120:R |
| 421 | [...] |
| 422 | |
| 423 | |
| 424 | As we have discussed previously about this format, the header shows |
| 425 | the name of the trace and points to the options. The "FUNCTION" |
| 426 | is a misnomer since here it represents the wake ups and context |
| 427 | switches. |
| 428 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 429 | The sched_switch file only lists the wake ups (represented with '+') |
| 430 | and context switches ('==>') with the previous task or current task |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 431 | first followed by the next task or task waking up. The format for both |
| 432 | of these is PID:KERNEL-PRIO:TASK-STATE. Remember that the KERNEL-PRIO |
| 433 | is the inverse of the actual priority with zero (0) being the highest |
| 434 | priority and the nice values starting at 100 (nice -20). Below is |
| 435 | a quick chart to map the kernel priority to user land priorities. |
| 436 | |
| 437 | Kernel priority: 0 to 99 ==> user RT priority 99 to 0 |
| 438 | Kernel priority: 100 to 139 ==> user nice -20 to 19 |
| 439 | Kernel priority: 140 ==> idle task priority |
| 440 | |
| 441 | The task states are: |
| 442 | |
| 443 | R - running : wants to run, may not actually be running |
| 444 | S - sleep : process is waiting to be woken up (handles signals) |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 445 | D - disk sleep (uninterruptible sleep) : process must be woken up |
| 446 | (ignores signals) |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 447 | T - stopped : process suspended |
| 448 | t - traced : process is being traced (with something like gdb) |
| 449 | Z - zombie : process waiting to be cleaned up |
| 450 | X - unknown |
| 451 | |
| 452 | |
| 453 | ftrace_enabled |
| 454 | -------------- |
| 455 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 456 | The following tracers (listed below) give different output depending |
| 457 | on whether or not the sysctl ftrace_enabled is set. To set ftrace_enabled, |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 458 | one can either use the sysctl function or set it via the proc |
| 459 | file system interface. |
| 460 | |
| 461 | sysctl kernel.ftrace_enabled=1 |
| 462 | |
| 463 | or |
| 464 | |
| 465 | echo 1 > /proc/sys/kernel/ftrace_enabled |
| 466 | |
| 467 | To disable ftrace_enabled simply replace the '1' with '0' in |
| 468 | the above commands. |
| 469 | |
| 470 | When ftrace_enabled is set the tracers will also record the functions |
| 471 | that are within the trace. The descriptions of the tracers |
| 472 | will also show an example with ftrace enabled. |
| 473 | |
| 474 | |
| 475 | irqsoff |
| 476 | ------- |
| 477 | |
| 478 | When interrupts are disabled, the CPU can not react to any other |
| 479 | external event (besides NMIs and SMIs). This prevents the timer |
| 480 | interrupt from triggering or the mouse interrupt from letting the |
| 481 | kernel know of a new mouse event. The result is a latency with the |
| 482 | reaction time. |
| 483 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 484 | The irqsoff tracer tracks the time for which interrupts are disabled. |
| 485 | When a new maximum latency is hit, the tracer saves the trace leading up |
| 486 | to that latency point so that every time a new maximum is reached, the old |
| 487 | saved trace is discarded and the new trace is saved. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 488 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 489 | To reset the maximum, echo 0 into tracing_max_latency. Here is an |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 490 | example: |
| 491 | |
| 492 | # echo irqsoff > /debug/tracing/current_tracer |
| 493 | # echo 0 > /debug/tracing/tracing_max_latency |
| 494 | # echo 1 > /debug/tracing/tracing_enabled |
| 495 | # ls -ltr |
| 496 | [...] |
| 497 | # echo 0 > /debug/tracing/tracing_enabled |
| 498 | # cat /debug/tracing/latency_trace |
| 499 | # tracer: irqsoff |
| 500 | # |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 501 | irqsoff latency trace v1.1.5 on 2.6.26 |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 502 | -------------------------------------------------------------------- |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 503 | latency: 12 us, #3/3, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 504 | ----------------- |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 505 | | task: bash-3730 (uid:0 nice:0 policy:0 rt_prio:0) |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 506 | ----------------- |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 507 | => started at: sys_setpgid |
| 508 | => ended at: sys_setpgid |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 509 | |
| 510 | # _------=> CPU# |
| 511 | # / _-----=> irqs-off |
| 512 | # | / _----=> need-resched |
| 513 | # || / _---=> hardirq/softirq |
| 514 | # ||| / _--=> preempt-depth |
| 515 | # |||| / |
| 516 | # ||||| delay |
| 517 | # cmd pid ||||| time | caller |
| 518 | # \ / ||||| \ | / |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 519 | bash-3730 1d... 0us : _write_lock_irq (sys_setpgid) |
| 520 | bash-3730 1d..1 1us+: _write_unlock_irq (sys_setpgid) |
| 521 | bash-3730 1d..2 14us : trace_hardirqs_on (sys_setpgid) |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 522 | |
| 523 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 524 | Here we see that that we had a latency of 12 microsecs (which is |
| 525 | very good). The _write_lock_irq in sys_setpgid disabled interrupts. |
| 526 | The difference between the 12 and the displayed timestamp 14us occurred |
| 527 | because the clock was incremented between the time of recording the max |
| 528 | latency and the time of recording the function that had that latency. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 529 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 530 | Note the above example had ftrace_enabled not set. If we set the |
| 531 | ftrace_enabled, we get a much larger output: |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 532 | |
| 533 | # tracer: irqsoff |
| 534 | # |
| 535 | irqsoff latency trace v1.1.5 on 2.6.26-rc8 |
| 536 | -------------------------------------------------------------------- |
| 537 | latency: 50 us, #101/101, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) |
| 538 | ----------------- |
| 539 | | task: ls-4339 (uid:0 nice:0 policy:0 rt_prio:0) |
| 540 | ----------------- |
| 541 | => started at: __alloc_pages_internal |
| 542 | => ended at: __alloc_pages_internal |
| 543 | |
| 544 | # _------=> CPU# |
| 545 | # / _-----=> irqs-off |
| 546 | # | / _----=> need-resched |
| 547 | # || / _---=> hardirq/softirq |
| 548 | # ||| / _--=> preempt-depth |
| 549 | # |||| / |
| 550 | # ||||| delay |
| 551 | # cmd pid ||||| time | caller |
| 552 | # \ / ||||| \ | / |
| 553 | ls-4339 0...1 0us+: get_page_from_freelist (__alloc_pages_internal) |
| 554 | ls-4339 0d..1 3us : rmqueue_bulk (get_page_from_freelist) |
| 555 | ls-4339 0d..1 3us : _spin_lock (rmqueue_bulk) |
| 556 | ls-4339 0d..1 4us : add_preempt_count (_spin_lock) |
| 557 | ls-4339 0d..2 4us : __rmqueue (rmqueue_bulk) |
| 558 | ls-4339 0d..2 5us : __rmqueue_smallest (__rmqueue) |
| 559 | ls-4339 0d..2 5us : __mod_zone_page_state (__rmqueue_smallest) |
| 560 | ls-4339 0d..2 6us : __rmqueue (rmqueue_bulk) |
| 561 | ls-4339 0d..2 6us : __rmqueue_smallest (__rmqueue) |
| 562 | ls-4339 0d..2 7us : __mod_zone_page_state (__rmqueue_smallest) |
| 563 | ls-4339 0d..2 7us : __rmqueue (rmqueue_bulk) |
| 564 | ls-4339 0d..2 8us : __rmqueue_smallest (__rmqueue) |
| 565 | [...] |
| 566 | ls-4339 0d..2 46us : __rmqueue_smallest (__rmqueue) |
| 567 | ls-4339 0d..2 47us : __mod_zone_page_state (__rmqueue_smallest) |
| 568 | ls-4339 0d..2 47us : __rmqueue (rmqueue_bulk) |
| 569 | ls-4339 0d..2 48us : __rmqueue_smallest (__rmqueue) |
| 570 | ls-4339 0d..2 48us : __mod_zone_page_state (__rmqueue_smallest) |
| 571 | ls-4339 0d..2 49us : _spin_unlock (rmqueue_bulk) |
| 572 | ls-4339 0d..2 49us : sub_preempt_count (_spin_unlock) |
| 573 | ls-4339 0d..1 50us : get_page_from_freelist (__alloc_pages_internal) |
| 574 | ls-4339 0d..2 51us : trace_hardirqs_on (__alloc_pages_internal) |
| 575 | |
| 576 | |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 577 | |
| 578 | Here we traced a 50 microsecond latency. But we also see all the |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 579 | functions that were called during that time. Note that by enabling |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 580 | function tracing, we incur an added overhead. This overhead may |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 581 | extend the latency times. But nevertheless, this trace has provided |
| 582 | some very helpful debugging information. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 583 | |
| 584 | |
| 585 | preemptoff |
| 586 | ---------- |
| 587 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 588 | When preemption is disabled, we may be able to receive interrupts but |
| 589 | the task cannot be preempted and a higher priority task must wait |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 590 | for preemption to be enabled again before it can preempt a lower |
| 591 | priority task. |
| 592 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 593 | The preemptoff tracer traces the places that disable preemption. |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 594 | Like the irqsoff tracer, it records the maximum latency for which preemption |
| 595 | was disabled. The control of preemptoff tracer is much like the irqsoff |
| 596 | tracer. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 597 | |
| 598 | # echo preemptoff > /debug/tracing/current_tracer |
| 599 | # echo 0 > /debug/tracing/tracing_max_latency |
| 600 | # echo 1 > /debug/tracing/tracing_enabled |
| 601 | # ls -ltr |
| 602 | [...] |
| 603 | # echo 0 > /debug/tracing/tracing_enabled |
| 604 | # cat /debug/tracing/latency_trace |
| 605 | # tracer: preemptoff |
| 606 | # |
| 607 | preemptoff latency trace v1.1.5 on 2.6.26-rc8 |
| 608 | -------------------------------------------------------------------- |
| 609 | latency: 29 us, #3/3, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) |
| 610 | ----------------- |
| 611 | | task: sshd-4261 (uid:0 nice:0 policy:0 rt_prio:0) |
| 612 | ----------------- |
| 613 | => started at: do_IRQ |
| 614 | => ended at: __do_softirq |
| 615 | |
| 616 | # _------=> CPU# |
| 617 | # / _-----=> irqs-off |
| 618 | # | / _----=> need-resched |
| 619 | # || / _---=> hardirq/softirq |
| 620 | # ||| / _--=> preempt-depth |
| 621 | # |||| / |
| 622 | # ||||| delay |
| 623 | # cmd pid ||||| time | caller |
| 624 | # \ / ||||| \ | / |
| 625 | sshd-4261 0d.h. 0us+: irq_enter (do_IRQ) |
| 626 | sshd-4261 0d.s. 29us : _local_bh_enable (__do_softirq) |
| 627 | sshd-4261 0d.s1 30us : trace_preempt_on (__do_softirq) |
| 628 | |
| 629 | |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 630 | This has some more changes. Preemption was disabled when an interrupt |
| 631 | came in (notice the 'h'), and was enabled while doing a softirq. |
| 632 | (notice the 's'). But we also see that interrupts have been disabled |
| 633 | when entering the preempt off section and leaving it (the 'd'). |
| 634 | We do not know if interrupts were enabled in the mean time. |
| 635 | |
| 636 | # tracer: preemptoff |
| 637 | # |
| 638 | preemptoff latency trace v1.1.5 on 2.6.26-rc8 |
| 639 | -------------------------------------------------------------------- |
| 640 | latency: 63 us, #87/87, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) |
| 641 | ----------------- |
| 642 | | task: sshd-4261 (uid:0 nice:0 policy:0 rt_prio:0) |
| 643 | ----------------- |
| 644 | => started at: remove_wait_queue |
| 645 | => ended at: __do_softirq |
| 646 | |
| 647 | # _------=> CPU# |
| 648 | # / _-----=> irqs-off |
| 649 | # | / _----=> need-resched |
| 650 | # || / _---=> hardirq/softirq |
| 651 | # ||| / _--=> preempt-depth |
| 652 | # |||| / |
| 653 | # ||||| delay |
| 654 | # cmd pid ||||| time | caller |
| 655 | # \ / ||||| \ | / |
| 656 | sshd-4261 0d..1 0us : _spin_lock_irqsave (remove_wait_queue) |
| 657 | sshd-4261 0d..1 1us : _spin_unlock_irqrestore (remove_wait_queue) |
| 658 | sshd-4261 0d..1 2us : do_IRQ (common_interrupt) |
| 659 | sshd-4261 0d..1 2us : irq_enter (do_IRQ) |
| 660 | sshd-4261 0d..1 2us : idle_cpu (irq_enter) |
| 661 | sshd-4261 0d..1 3us : add_preempt_count (irq_enter) |
| 662 | sshd-4261 0d.h1 3us : idle_cpu (irq_enter) |
| 663 | sshd-4261 0d.h. 4us : handle_fasteoi_irq (do_IRQ) |
| 664 | [...] |
| 665 | sshd-4261 0d.h. 12us : add_preempt_count (_spin_lock) |
| 666 | sshd-4261 0d.h1 12us : ack_ioapic_quirk_irq (handle_fasteoi_irq) |
| 667 | sshd-4261 0d.h1 13us : move_native_irq (ack_ioapic_quirk_irq) |
| 668 | sshd-4261 0d.h1 13us : _spin_unlock (handle_fasteoi_irq) |
| 669 | sshd-4261 0d.h1 14us : sub_preempt_count (_spin_unlock) |
| 670 | sshd-4261 0d.h1 14us : irq_exit (do_IRQ) |
| 671 | sshd-4261 0d.h1 15us : sub_preempt_count (irq_exit) |
| 672 | sshd-4261 0d..2 15us : do_softirq (irq_exit) |
| 673 | sshd-4261 0d... 15us : __do_softirq (do_softirq) |
| 674 | sshd-4261 0d... 16us : __local_bh_disable (__do_softirq) |
| 675 | sshd-4261 0d... 16us+: add_preempt_count (__local_bh_disable) |
| 676 | sshd-4261 0d.s4 20us : add_preempt_count (__local_bh_disable) |
| 677 | sshd-4261 0d.s4 21us : sub_preempt_count (local_bh_enable) |
| 678 | sshd-4261 0d.s5 21us : sub_preempt_count (local_bh_enable) |
| 679 | [...] |
| 680 | sshd-4261 0d.s6 41us : add_preempt_count (__local_bh_disable) |
| 681 | sshd-4261 0d.s6 42us : sub_preempt_count (local_bh_enable) |
| 682 | sshd-4261 0d.s7 42us : sub_preempt_count (local_bh_enable) |
| 683 | sshd-4261 0d.s5 43us : add_preempt_count (__local_bh_disable) |
| 684 | sshd-4261 0d.s5 43us : sub_preempt_count (local_bh_enable_ip) |
| 685 | sshd-4261 0d.s6 44us : sub_preempt_count (local_bh_enable_ip) |
| 686 | sshd-4261 0d.s5 44us : add_preempt_count (__local_bh_disable) |
| 687 | sshd-4261 0d.s5 45us : sub_preempt_count (local_bh_enable) |
| 688 | [...] |
| 689 | sshd-4261 0d.s. 63us : _local_bh_enable (__do_softirq) |
| 690 | sshd-4261 0d.s1 64us : trace_preempt_on (__do_softirq) |
| 691 | |
| 692 | |
| 693 | The above is an example of the preemptoff trace with ftrace_enabled |
| 694 | set. Here we see that interrupts were disabled the entire time. |
| 695 | The irq_enter code lets us know that we entered an interrupt 'h'. |
| 696 | Before that, the functions being traced still show that it is not |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 697 | in an interrupt, but we can see from the functions themselves that |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 698 | this is not the case. |
| 699 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 700 | Notice that __do_softirq when called does not have a preempt_count. |
| 701 | It may seem that we missed a preempt enabling. What really happened |
| 702 | is that the preempt count is held on the thread's stack and we |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 703 | switched to the softirq stack (4K stacks in effect). The code |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 704 | does not copy the preempt count, but because interrupts are disabled, |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 705 | we do not need to worry about it. Having a tracer like this is good |
| 706 | for letting people know what really happens inside the kernel. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 707 | |
| 708 | |
| 709 | preemptirqsoff |
| 710 | -------------- |
| 711 | |
| 712 | Knowing the locations that have interrupts disabled or preemption |
| 713 | disabled for the longest times is helpful. But sometimes we would |
| 714 | like to know when either preemption and/or interrupts are disabled. |
| 715 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 716 | Consider the following code: |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 717 | |
| 718 | local_irq_disable(); |
| 719 | call_function_with_irqs_off(); |
| 720 | preempt_disable(); |
| 721 | call_function_with_irqs_and_preemption_off(); |
| 722 | local_irq_enable(); |
| 723 | call_function_with_preemption_off(); |
| 724 | preempt_enable(); |
| 725 | |
| 726 | The irqsoff tracer will record the total length of |
| 727 | call_function_with_irqs_off() and |
| 728 | call_function_with_irqs_and_preemption_off(). |
| 729 | |
| 730 | The preemptoff tracer will record the total length of |
| 731 | call_function_with_irqs_and_preemption_off() and |
| 732 | call_function_with_preemption_off(). |
| 733 | |
| 734 | But neither will trace the time that interrupts and/or preemption |
| 735 | is disabled. This total time is the time that we can not schedule. |
| 736 | To record this time, use the preemptirqsoff tracer. |
| 737 | |
| 738 | Again, using this trace is much like the irqsoff and preemptoff tracers. |
| 739 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 740 | # echo preemptirqsoff > /debug/tracing/current_tracer |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 741 | # echo 0 > /debug/tracing/tracing_max_latency |
| 742 | # echo 1 > /debug/tracing/tracing_enabled |
| 743 | # ls -ltr |
| 744 | [...] |
| 745 | # echo 0 > /debug/tracing/tracing_enabled |
| 746 | # cat /debug/tracing/latency_trace |
| 747 | # tracer: preemptirqsoff |
| 748 | # |
| 749 | preemptirqsoff latency trace v1.1.5 on 2.6.26-rc8 |
| 750 | -------------------------------------------------------------------- |
| 751 | latency: 293 us, #3/3, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) |
| 752 | ----------------- |
| 753 | | task: ls-4860 (uid:0 nice:0 policy:0 rt_prio:0) |
| 754 | ----------------- |
| 755 | => started at: apic_timer_interrupt |
| 756 | => ended at: __do_softirq |
| 757 | |
| 758 | # _------=> CPU# |
| 759 | # / _-----=> irqs-off |
| 760 | # | / _----=> need-resched |
| 761 | # || / _---=> hardirq/softirq |
| 762 | # ||| / _--=> preempt-depth |
| 763 | # |||| / |
| 764 | # ||||| delay |
| 765 | # cmd pid ||||| time | caller |
| 766 | # \ / ||||| \ | / |
| 767 | ls-4860 0d... 0us!: trace_hardirqs_off_thunk (apic_timer_interrupt) |
| 768 | ls-4860 0d.s. 294us : _local_bh_enable (__do_softirq) |
| 769 | ls-4860 0d.s1 294us : trace_preempt_on (__do_softirq) |
| 770 | |
| 771 | |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 772 | |
| 773 | The trace_hardirqs_off_thunk is called from assembly on x86 when |
| 774 | interrupts are disabled in the assembly code. Without the function |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 775 | tracing, we do not know if interrupts were enabled within the preemption |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 776 | points. We do see that it started with preemption enabled. |
| 777 | |
| 778 | Here is a trace with ftrace_enabled set: |
| 779 | |
| 780 | |
| 781 | # tracer: preemptirqsoff |
| 782 | # |
| 783 | preemptirqsoff latency trace v1.1.5 on 2.6.26-rc8 |
| 784 | -------------------------------------------------------------------- |
| 785 | latency: 105 us, #183/183, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) |
| 786 | ----------------- |
| 787 | | task: sshd-4261 (uid:0 nice:0 policy:0 rt_prio:0) |
| 788 | ----------------- |
| 789 | => started at: write_chan |
| 790 | => ended at: __do_softirq |
| 791 | |
| 792 | # _------=> CPU# |
| 793 | # / _-----=> irqs-off |
| 794 | # | / _----=> need-resched |
| 795 | # || / _---=> hardirq/softirq |
| 796 | # ||| / _--=> preempt-depth |
| 797 | # |||| / |
| 798 | # ||||| delay |
| 799 | # cmd pid ||||| time | caller |
| 800 | # \ / ||||| \ | / |
| 801 | ls-4473 0.N.. 0us : preempt_schedule (write_chan) |
| 802 | ls-4473 0dN.1 1us : _spin_lock (schedule) |
| 803 | ls-4473 0dN.1 2us : add_preempt_count (_spin_lock) |
| 804 | ls-4473 0d..2 2us : put_prev_task_fair (schedule) |
| 805 | [...] |
| 806 | ls-4473 0d..2 13us : set_normalized_timespec (ktime_get_ts) |
| 807 | ls-4473 0d..2 13us : __switch_to (schedule) |
| 808 | sshd-4261 0d..2 14us : finish_task_switch (schedule) |
| 809 | sshd-4261 0d..2 14us : _spin_unlock_irq (finish_task_switch) |
| 810 | sshd-4261 0d..1 15us : add_preempt_count (_spin_lock_irqsave) |
| 811 | sshd-4261 0d..2 16us : _spin_unlock_irqrestore (hrtick_set) |
| 812 | sshd-4261 0d..2 16us : do_IRQ (common_interrupt) |
| 813 | sshd-4261 0d..2 17us : irq_enter (do_IRQ) |
| 814 | sshd-4261 0d..2 17us : idle_cpu (irq_enter) |
| 815 | sshd-4261 0d..2 18us : add_preempt_count (irq_enter) |
| 816 | sshd-4261 0d.h2 18us : idle_cpu (irq_enter) |
| 817 | sshd-4261 0d.h. 18us : handle_fasteoi_irq (do_IRQ) |
| 818 | sshd-4261 0d.h. 19us : _spin_lock (handle_fasteoi_irq) |
| 819 | sshd-4261 0d.h. 19us : add_preempt_count (_spin_lock) |
| 820 | sshd-4261 0d.h1 20us : _spin_unlock (handle_fasteoi_irq) |
| 821 | sshd-4261 0d.h1 20us : sub_preempt_count (_spin_unlock) |
| 822 | [...] |
| 823 | sshd-4261 0d.h1 28us : _spin_unlock (handle_fasteoi_irq) |
| 824 | sshd-4261 0d.h1 29us : sub_preempt_count (_spin_unlock) |
| 825 | sshd-4261 0d.h2 29us : irq_exit (do_IRQ) |
| 826 | sshd-4261 0d.h2 29us : sub_preempt_count (irq_exit) |
| 827 | sshd-4261 0d..3 30us : do_softirq (irq_exit) |
| 828 | sshd-4261 0d... 30us : __do_softirq (do_softirq) |
| 829 | sshd-4261 0d... 31us : __local_bh_disable (__do_softirq) |
| 830 | sshd-4261 0d... 31us+: add_preempt_count (__local_bh_disable) |
| 831 | sshd-4261 0d.s4 34us : add_preempt_count (__local_bh_disable) |
| 832 | [...] |
| 833 | sshd-4261 0d.s3 43us : sub_preempt_count (local_bh_enable_ip) |
| 834 | sshd-4261 0d.s4 44us : sub_preempt_count (local_bh_enable_ip) |
| 835 | sshd-4261 0d.s3 44us : smp_apic_timer_interrupt (apic_timer_interrupt) |
| 836 | sshd-4261 0d.s3 45us : irq_enter (smp_apic_timer_interrupt) |
| 837 | sshd-4261 0d.s3 45us : idle_cpu (irq_enter) |
| 838 | sshd-4261 0d.s3 46us : add_preempt_count (irq_enter) |
| 839 | sshd-4261 0d.H3 46us : idle_cpu (irq_enter) |
| 840 | sshd-4261 0d.H3 47us : hrtimer_interrupt (smp_apic_timer_interrupt) |
| 841 | sshd-4261 0d.H3 47us : ktime_get (hrtimer_interrupt) |
| 842 | [...] |
| 843 | sshd-4261 0d.H3 81us : tick_program_event (hrtimer_interrupt) |
| 844 | sshd-4261 0d.H3 82us : ktime_get (tick_program_event) |
| 845 | sshd-4261 0d.H3 82us : ktime_get_ts (ktime_get) |
| 846 | sshd-4261 0d.H3 83us : getnstimeofday (ktime_get_ts) |
| 847 | sshd-4261 0d.H3 83us : set_normalized_timespec (ktime_get_ts) |
| 848 | sshd-4261 0d.H3 84us : clockevents_program_event (tick_program_event) |
| 849 | sshd-4261 0d.H3 84us : lapic_next_event (clockevents_program_event) |
| 850 | sshd-4261 0d.H3 85us : irq_exit (smp_apic_timer_interrupt) |
| 851 | sshd-4261 0d.H3 85us : sub_preempt_count (irq_exit) |
| 852 | sshd-4261 0d.s4 86us : sub_preempt_count (irq_exit) |
| 853 | sshd-4261 0d.s3 86us : add_preempt_count (__local_bh_disable) |
| 854 | [...] |
| 855 | sshd-4261 0d.s1 98us : sub_preempt_count (net_rx_action) |
| 856 | sshd-4261 0d.s. 99us : add_preempt_count (_spin_lock_irq) |
| 857 | sshd-4261 0d.s1 99us+: _spin_unlock_irq (run_timer_softirq) |
| 858 | sshd-4261 0d.s. 104us : _local_bh_enable (__do_softirq) |
| 859 | sshd-4261 0d.s. 104us : sub_preempt_count (_local_bh_enable) |
| 860 | sshd-4261 0d.s. 105us : _local_bh_enable (__do_softirq) |
| 861 | sshd-4261 0d.s1 105us : trace_preempt_on (__do_softirq) |
| 862 | |
| 863 | |
| 864 | This is a very interesting trace. It started with the preemption of |
| 865 | the ls task. We see that the task had the "need_resched" bit set |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 866 | via the 'N' in the trace. Interrupts were disabled before the spin_lock |
| 867 | at the beginning of the trace. We see that a schedule took place to run |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 868 | sshd. When the interrupts were enabled, we took an interrupt. |
| 869 | On return from the interrupt handler, the softirq ran. We took another |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 870 | interrupt while running the softirq as we see from the capital 'H'. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 871 | |
| 872 | |
| 873 | wakeup |
| 874 | ------ |
| 875 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 876 | In a Real-Time environment it is very important to know the wakeup |
| 877 | time it takes for the highest priority task that is woken up to the |
| 878 | time that it executes. This is also known as "schedule latency". |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 879 | I stress the point that this is about RT tasks. It is also important |
| 880 | to know the scheduling latency of non-RT tasks, but the average |
| 881 | schedule latency is better for non-RT tasks. Tools like |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 882 | LatencyTop are more appropriate for such measurements. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 883 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 884 | Real-Time environments are interested in the worst case latency. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 885 | That is the longest latency it takes for something to happen, and |
| 886 | not the average. We can have a very fast scheduler that may only |
| 887 | have a large latency once in a while, but that would not work well |
| 888 | with Real-Time tasks. The wakeup tracer was designed to record |
| 889 | the worst case wakeups of RT tasks. Non-RT tasks are not recorded |
| 890 | because the tracer only records one worst case and tracing non-RT |
| 891 | tasks that are unpredictable will overwrite the worst case latency |
| 892 | of RT tasks. |
| 893 | |
| 894 | Since this tracer only deals with RT tasks, we will run this slightly |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 895 | differently than we did with the previous tracers. Instead of performing |
| 896 | an 'ls', we will run 'sleep 1' under 'chrt' which changes the |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 897 | priority of the task. |
| 898 | |
| 899 | # echo wakeup > /debug/tracing/current_tracer |
| 900 | # echo 0 > /debug/tracing/tracing_max_latency |
| 901 | # echo 1 > /debug/tracing/tracing_enabled |
| 902 | # chrt -f 5 sleep 1 |
| 903 | # echo 0 > /debug/tracing/tracing_enabled |
| 904 | # cat /debug/tracing/latency_trace |
| 905 | # tracer: wakeup |
| 906 | # |
| 907 | wakeup latency trace v1.1.5 on 2.6.26-rc8 |
| 908 | -------------------------------------------------------------------- |
| 909 | latency: 4 us, #2/2, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) |
| 910 | ----------------- |
| 911 | | task: sleep-4901 (uid:0 nice:0 policy:1 rt_prio:5) |
| 912 | ----------------- |
| 913 | |
| 914 | # _------=> CPU# |
| 915 | # / _-----=> irqs-off |
| 916 | # | / _----=> need-resched |
| 917 | # || / _---=> hardirq/softirq |
| 918 | # ||| / _--=> preempt-depth |
| 919 | # |||| / |
| 920 | # ||||| delay |
| 921 | # cmd pid ||||| time | caller |
| 922 | # \ / ||||| \ | / |
| 923 | <idle>-0 1d.h4 0us+: try_to_wake_up (wake_up_process) |
| 924 | <idle>-0 1d..4 4us : schedule (cpu_idle) |
| 925 | |
| 926 | |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 927 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 928 | Running this on an idle system, we see that it only took 4 microseconds |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 929 | to perform the task switch. Note, since the trace marker in the |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 930 | schedule is before the actual "switch", we stop the tracing when |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 931 | the recorded task is about to schedule in. This may change if |
| 932 | we add a new marker at the end of the scheduler. |
| 933 | |
| 934 | Notice that the recorded task is 'sleep' with the PID of 4901 and it |
| 935 | has an rt_prio of 5. This priority is user-space priority and not |
| 936 | the internal kernel priority. The policy is 1 for SCHED_FIFO and 2 |
| 937 | for SCHED_RR. |
| 938 | |
| 939 | Doing the same with chrt -r 5 and ftrace_enabled set. |
| 940 | |
| 941 | # tracer: wakeup |
| 942 | # |
| 943 | wakeup latency trace v1.1.5 on 2.6.26-rc8 |
| 944 | -------------------------------------------------------------------- |
| 945 | latency: 50 us, #60/60, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) |
| 946 | ----------------- |
| 947 | | task: sleep-4068 (uid:0 nice:0 policy:2 rt_prio:5) |
| 948 | ----------------- |
| 949 | |
| 950 | # _------=> CPU# |
| 951 | # / _-----=> irqs-off |
| 952 | # | / _----=> need-resched |
| 953 | # || / _---=> hardirq/softirq |
| 954 | # ||| / _--=> preempt-depth |
| 955 | # |||| / |
| 956 | # ||||| delay |
| 957 | # cmd pid ||||| time | caller |
| 958 | # \ / ||||| \ | / |
| 959 | ksoftirq-7 1d.H3 0us : try_to_wake_up (wake_up_process) |
| 960 | ksoftirq-7 1d.H4 1us : sub_preempt_count (marker_probe_cb) |
| 961 | ksoftirq-7 1d.H3 2us : check_preempt_wakeup (try_to_wake_up) |
| 962 | ksoftirq-7 1d.H3 3us : update_curr (check_preempt_wakeup) |
| 963 | ksoftirq-7 1d.H3 4us : calc_delta_mine (update_curr) |
| 964 | ksoftirq-7 1d.H3 5us : __resched_task (check_preempt_wakeup) |
| 965 | ksoftirq-7 1d.H3 6us : task_wake_up_rt (try_to_wake_up) |
| 966 | ksoftirq-7 1d.H3 7us : _spin_unlock_irqrestore (try_to_wake_up) |
| 967 | [...] |
| 968 | ksoftirq-7 1d.H2 17us : irq_exit (smp_apic_timer_interrupt) |
| 969 | ksoftirq-7 1d.H2 18us : sub_preempt_count (irq_exit) |
| 970 | ksoftirq-7 1d.s3 19us : sub_preempt_count (irq_exit) |
| 971 | ksoftirq-7 1..s2 20us : rcu_process_callbacks (__do_softirq) |
| 972 | [...] |
| 973 | ksoftirq-7 1..s2 26us : __rcu_process_callbacks (rcu_process_callbacks) |
| 974 | ksoftirq-7 1d.s2 27us : _local_bh_enable (__do_softirq) |
| 975 | ksoftirq-7 1d.s2 28us : sub_preempt_count (_local_bh_enable) |
| 976 | ksoftirq-7 1.N.3 29us : sub_preempt_count (ksoftirqd) |
| 977 | ksoftirq-7 1.N.2 30us : _cond_resched (ksoftirqd) |
| 978 | ksoftirq-7 1.N.2 31us : __cond_resched (_cond_resched) |
| 979 | ksoftirq-7 1.N.2 32us : add_preempt_count (__cond_resched) |
| 980 | ksoftirq-7 1.N.2 33us : schedule (__cond_resched) |
| 981 | ksoftirq-7 1.N.2 33us : add_preempt_count (schedule) |
| 982 | ksoftirq-7 1.N.3 34us : hrtick_clear (schedule) |
| 983 | ksoftirq-7 1dN.3 35us : _spin_lock (schedule) |
| 984 | ksoftirq-7 1dN.3 36us : add_preempt_count (_spin_lock) |
| 985 | ksoftirq-7 1d..4 37us : put_prev_task_fair (schedule) |
| 986 | ksoftirq-7 1d..4 38us : update_curr (put_prev_task_fair) |
| 987 | [...] |
| 988 | ksoftirq-7 1d..5 47us : _spin_trylock (tracing_record_cmdline) |
| 989 | ksoftirq-7 1d..5 48us : add_preempt_count (_spin_trylock) |
| 990 | ksoftirq-7 1d..6 49us : _spin_unlock (tracing_record_cmdline) |
| 991 | ksoftirq-7 1d..6 49us : sub_preempt_count (_spin_unlock) |
| 992 | ksoftirq-7 1d..4 50us : schedule (__cond_resched) |
| 993 | |
| 994 | The interrupt went off while running ksoftirqd. This task runs at |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 995 | SCHED_OTHER. Why did not we see the 'N' set early? This may be |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 996 | a harmless bug with x86_32 and 4K stacks. On x86_32 with 4K stacks |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 997 | configured, the interrupt and softirq run with their own stack. |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 998 | Some information is held on the top of the task's stack (need_resched |
| 999 | and preempt_count are both stored there). The setting of the NEED_RESCHED |
| 1000 | bit is done directly to the task's stack, but the reading of the |
| 1001 | NEED_RESCHED is done by looking at the current stack, which in this case |
| 1002 | is the stack for the hard interrupt. This hides the fact that NEED_RESCHED |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1003 | has been set. We do not see the 'N' until we switch back to the task's |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1004 | assigned stack. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1005 | |
| 1006 | ftrace |
| 1007 | ------ |
| 1008 | |
| 1009 | ftrace is not only the name of the tracing infrastructure, but it |
| 1010 | is also a name of one of the tracers. The tracer is the function |
| 1011 | tracer. Enabling the function tracer can be done from the |
| 1012 | debug file system. Make sure the ftrace_enabled is set otherwise |
| 1013 | this tracer is a nop. |
| 1014 | |
| 1015 | # sysctl kernel.ftrace_enabled=1 |
| 1016 | # echo ftrace > /debug/tracing/current_tracer |
| 1017 | # echo 1 > /debug/tracing/tracing_enabled |
| 1018 | # usleep 1 |
| 1019 | # echo 0 > /debug/tracing/tracing_enabled |
| 1020 | # cat /debug/tracing/trace |
| 1021 | # tracer: ftrace |
| 1022 | # |
| 1023 | # TASK-PID CPU# TIMESTAMP FUNCTION |
| 1024 | # | | | | | |
| 1025 | bash-4003 [00] 123.638713: finish_task_switch <-schedule |
| 1026 | bash-4003 [00] 123.638714: _spin_unlock_irq <-finish_task_switch |
| 1027 | bash-4003 [00] 123.638714: sub_preempt_count <-_spin_unlock_irq |
| 1028 | bash-4003 [00] 123.638715: hrtick_set <-schedule |
| 1029 | bash-4003 [00] 123.638715: _spin_lock_irqsave <-hrtick_set |
| 1030 | bash-4003 [00] 123.638716: add_preempt_count <-_spin_lock_irqsave |
| 1031 | bash-4003 [00] 123.638716: _spin_unlock_irqrestore <-hrtick_set |
| 1032 | bash-4003 [00] 123.638717: sub_preempt_count <-_spin_unlock_irqrestore |
| 1033 | bash-4003 [00] 123.638717: hrtick_clear <-hrtick_set |
| 1034 | bash-4003 [00] 123.638718: sub_preempt_count <-schedule |
| 1035 | bash-4003 [00] 123.638718: sub_preempt_count <-preempt_schedule |
| 1036 | bash-4003 [00] 123.638719: wait_for_completion <-__stop_machine_run |
| 1037 | bash-4003 [00] 123.638719: wait_for_common <-wait_for_completion |
| 1038 | bash-4003 [00] 123.638720: _spin_lock_irq <-wait_for_common |
| 1039 | bash-4003 [00] 123.638720: add_preempt_count <-_spin_lock_irq |
| 1040 | [...] |
| 1041 | |
| 1042 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1043 | Note: ftrace uses ring buffers to store the above entries. The newest data |
| 1044 | may overwrite the oldest data. Sometimes using echo to stop the trace |
| 1045 | is not sufficient because the tracing could have overwritten the data |
| 1046 | that you wanted to record. For this reason, it is sometimes better to |
| 1047 | disable tracing directly from a program. This allows you to stop the |
| 1048 | tracing at the point that you hit the part that you are interested in. |
| 1049 | To disable the tracing directly from a C program, something like following |
| 1050 | code snippet can be used: |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1051 | |
| 1052 | int trace_fd; |
| 1053 | [...] |
| 1054 | int main(int argc, char *argv[]) { |
| 1055 | [...] |
| 1056 | trace_fd = open("/debug/tracing/tracing_enabled", O_WRONLY); |
| 1057 | [...] |
| 1058 | if (condition_hit()) { |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1059 | write(trace_fd, "0", 1); |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1060 | } |
| 1061 | [...] |
| 1062 | } |
| 1063 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1064 | Note: Here we hard coded the path name. The debugfs mount is not |
| 1065 | guaranteed to be at /debug (and is more commonly at /sys/kernel/debug). |
| 1066 | For simple one time traces, the above is sufficent. For anything else, |
| 1067 | a search through /proc/mounts may be needed to find where the debugfs |
| 1068 | file-system is mounted. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1069 | |
| 1070 | dynamic ftrace |
| 1071 | -------------- |
| 1072 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1073 | If CONFIG_DYNAMIC_FTRACE is set, the system will run with |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1074 | virtually no overhead when function tracing is disabled. The way |
| 1075 | this works is the mcount function call (placed at the start of |
| 1076 | every kernel function, produced by the -pg switch in gcc), starts |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1077 | of pointing to a simple return. (Enabling FTRACE will include the |
| 1078 | -pg switch in the compiling of the kernel.) |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1079 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1080 | When dynamic ftrace is initialized, it calls kstop_machine to make |
| 1081 | the machine act like a uniprocessor so that it can freely modify code |
| 1082 | without worrying about other processors executing that same code. At |
| 1083 | initialization, the mcount calls are changed to call a "record_ip" |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1084 | function. After this, the first time a kernel function is called, |
| 1085 | it has the calling address saved in a hash table. |
| 1086 | |
| 1087 | Later on the ftraced kernel thread is awoken and will again call |
| 1088 | kstop_machine if new functions have been recorded. The ftraced thread |
| 1089 | will change all calls to mcount to "nop". Just calling mcount |
| 1090 | and having mcount return has shown a 10% overhead. By converting |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1091 | it to a nop, there is no measurable overhead to the system. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1092 | |
| 1093 | One special side-effect to the recording of the functions being |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1094 | traced is that we can now selectively choose which functions we |
| 1095 | wish to trace and which ones we want the mcount calls to remain as |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1096 | nops. |
| 1097 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1098 | Two files are used, one for enabling and one for disabling the tracing |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1099 | of specified functions. They are: |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1100 | |
| 1101 | set_ftrace_filter |
| 1102 | |
| 1103 | and |
| 1104 | |
| 1105 | set_ftrace_notrace |
| 1106 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1107 | A list of available functions that you can add to these files is listed |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1108 | in: |
| 1109 | |
| 1110 | available_filter_functions |
| 1111 | |
| 1112 | # cat /debug/tracing/available_filter_functions |
| 1113 | put_prev_task_idle |
| 1114 | kmem_cache_create |
| 1115 | pick_next_task_rt |
| 1116 | get_online_cpus |
| 1117 | pick_next_task_fair |
| 1118 | mutex_lock |
| 1119 | [...] |
| 1120 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1121 | If I am only interested in sys_nanosleep and hrtimer_interrupt: |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1122 | |
| 1123 | # echo sys_nanosleep hrtimer_interrupt \ |
| 1124 | > /debug/tracing/set_ftrace_filter |
| 1125 | # echo ftrace > /debug/tracing/current_tracer |
| 1126 | # echo 1 > /debug/tracing/tracing_enabled |
| 1127 | # usleep 1 |
| 1128 | # echo 0 > /debug/tracing/tracing_enabled |
| 1129 | # cat /debug/tracing/trace |
| 1130 | # tracer: ftrace |
| 1131 | # |
| 1132 | # TASK-PID CPU# TIMESTAMP FUNCTION |
| 1133 | # | | | | | |
| 1134 | usleep-4134 [00] 1317.070017: hrtimer_interrupt <-smp_apic_timer_interrupt |
| 1135 | usleep-4134 [00] 1317.070111: sys_nanosleep <-syscall_call |
| 1136 | <idle>-0 [00] 1317.070115: hrtimer_interrupt <-smp_apic_timer_interrupt |
| 1137 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1138 | To see which functions are being traced, you can cat the file: |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1139 | |
| 1140 | # cat /debug/tracing/set_ftrace_filter |
| 1141 | hrtimer_interrupt |
| 1142 | sys_nanosleep |
| 1143 | |
| 1144 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1145 | Perhaps this is not enough. The filters also allow simple wild cards. |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1146 | Only the following are currently available |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1147 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1148 | <match>* - will match functions that begin with <match> |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1149 | *<match> - will match functions that end with <match> |
| 1150 | *<match>* - will match functions that have <match> in it |
| 1151 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1152 | These are the only wild cards which are supported. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1153 | |
| 1154 | <match>*<match> will not work. |
| 1155 | |
| 1156 | # echo hrtimer_* > /debug/tracing/set_ftrace_filter |
| 1157 | |
| 1158 | Produces: |
| 1159 | |
| 1160 | # tracer: ftrace |
| 1161 | # |
| 1162 | # TASK-PID CPU# TIMESTAMP FUNCTION |
| 1163 | # | | | | | |
| 1164 | bash-4003 [00] 1480.611794: hrtimer_init <-copy_process |
| 1165 | bash-4003 [00] 1480.611941: hrtimer_start <-hrtick_set |
| 1166 | bash-4003 [00] 1480.611956: hrtimer_cancel <-hrtick_clear |
| 1167 | bash-4003 [00] 1480.611956: hrtimer_try_to_cancel <-hrtimer_cancel |
| 1168 | <idle>-0 [00] 1480.612019: hrtimer_get_next_event <-get_next_timer_interrupt |
| 1169 | <idle>-0 [00] 1480.612025: hrtimer_get_next_event <-get_next_timer_interrupt |
| 1170 | <idle>-0 [00] 1480.612032: hrtimer_get_next_event <-get_next_timer_interrupt |
| 1171 | <idle>-0 [00] 1480.612037: hrtimer_get_next_event <-get_next_timer_interrupt |
| 1172 | <idle>-0 [00] 1480.612382: hrtimer_get_next_event <-get_next_timer_interrupt |
| 1173 | |
| 1174 | |
| 1175 | Notice that we lost the sys_nanosleep. |
| 1176 | |
| 1177 | # cat /debug/tracing/set_ftrace_filter |
| 1178 | hrtimer_run_queues |
| 1179 | hrtimer_run_pending |
| 1180 | hrtimer_init |
| 1181 | hrtimer_cancel |
| 1182 | hrtimer_try_to_cancel |
| 1183 | hrtimer_forward |
| 1184 | hrtimer_start |
| 1185 | hrtimer_reprogram |
| 1186 | hrtimer_force_reprogram |
| 1187 | hrtimer_get_next_event |
| 1188 | hrtimer_interrupt |
| 1189 | hrtimer_nanosleep |
| 1190 | hrtimer_wakeup |
| 1191 | hrtimer_get_remaining |
| 1192 | hrtimer_get_res |
| 1193 | hrtimer_init_sleeper |
| 1194 | |
| 1195 | |
| 1196 | This is because the '>' and '>>' act just like they do in bash. |
| 1197 | To rewrite the filters, use '>' |
| 1198 | To append to the filters, use '>>' |
| 1199 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1200 | To clear out a filter so that all functions will be recorded again: |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1201 | |
| 1202 | # echo > /debug/tracing/set_ftrace_filter |
| 1203 | # cat /debug/tracing/set_ftrace_filter |
| 1204 | # |
| 1205 | |
| 1206 | Again, now we want to append. |
| 1207 | |
| 1208 | # echo sys_nanosleep > /debug/tracing/set_ftrace_filter |
| 1209 | # cat /debug/tracing/set_ftrace_filter |
| 1210 | sys_nanosleep |
| 1211 | # echo hrtimer_* >> /debug/tracing/set_ftrace_filter |
| 1212 | # cat /debug/tracing/set_ftrace_filter |
| 1213 | hrtimer_run_queues |
| 1214 | hrtimer_run_pending |
| 1215 | hrtimer_init |
| 1216 | hrtimer_cancel |
| 1217 | hrtimer_try_to_cancel |
| 1218 | hrtimer_forward |
| 1219 | hrtimer_start |
| 1220 | hrtimer_reprogram |
| 1221 | hrtimer_force_reprogram |
| 1222 | hrtimer_get_next_event |
| 1223 | hrtimer_interrupt |
| 1224 | sys_nanosleep |
| 1225 | hrtimer_nanosleep |
| 1226 | hrtimer_wakeup |
| 1227 | hrtimer_get_remaining |
| 1228 | hrtimer_get_res |
| 1229 | hrtimer_init_sleeper |
| 1230 | |
| 1231 | |
| 1232 | The set_ftrace_notrace prevents those functions from being traced. |
| 1233 | |
| 1234 | # echo '*preempt*' '*lock*' > /debug/tracing/set_ftrace_notrace |
| 1235 | |
| 1236 | Produces: |
| 1237 | |
| 1238 | # tracer: ftrace |
| 1239 | # |
| 1240 | # TASK-PID CPU# TIMESTAMP FUNCTION |
| 1241 | # | | | | | |
| 1242 | bash-4043 [01] 115.281644: finish_task_switch <-schedule |
| 1243 | bash-4043 [01] 115.281645: hrtick_set <-schedule |
| 1244 | bash-4043 [01] 115.281645: hrtick_clear <-hrtick_set |
| 1245 | bash-4043 [01] 115.281646: wait_for_completion <-__stop_machine_run |
| 1246 | bash-4043 [01] 115.281647: wait_for_common <-wait_for_completion |
| 1247 | bash-4043 [01] 115.281647: kthread_stop <-stop_machine_run |
| 1248 | bash-4043 [01] 115.281648: init_waitqueue_head <-kthread_stop |
| 1249 | bash-4043 [01] 115.281648: wake_up_process <-kthread_stop |
| 1250 | bash-4043 [01] 115.281649: try_to_wake_up <-wake_up_process |
| 1251 | |
| 1252 | We can see that there's no more lock or preempt tracing. |
| 1253 | |
| 1254 | ftraced |
| 1255 | ------- |
| 1256 | |
| 1257 | As mentioned above, when dynamic ftrace is configured in, a kernel |
| 1258 | thread wakes up once a second and checks to see if there are mcount |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1259 | calls that need to be converted into nops. If there are not any, then |
| 1260 | it simply goes back to sleep. But if there are some, it will call |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1261 | kstop_machine to convert the calls to nops. |
| 1262 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1263 | There may be a case in which you do not want this added latency. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1264 | Perhaps you are doing some audio recording and this activity might |
| 1265 | cause skips in the playback. There is an interface to disable |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1266 | and enable the "ftraced" kernel thread. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1267 | |
| 1268 | # echo 0 > /debug/tracing/ftraced_enabled |
| 1269 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1270 | This will disable the calling of kstop_machine to update the |
| 1271 | mcount calls to nops. Remember that there is a large overhead |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1272 | to calling mcount. Without this kernel thread, that overhead will |
| 1273 | exist. |
| 1274 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1275 | If there are recorded calls to mcount, any write to the ftraced_enabled |
| 1276 | file will cause the kstop_machine to run. This means that a |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1277 | user can manually perform the updates when they want to by simply |
| 1278 | echoing a '0' into the ftraced_enabled file. |
| 1279 | |
| 1280 | The updates are also done at the beginning of enabling a tracer |
| 1281 | that uses ftrace function recording. |
| 1282 | |
| 1283 | |
| 1284 | trace_pipe |
| 1285 | ---------- |
| 1286 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1287 | The trace_pipe outputs the same content as the trace file, but the effect |
| 1288 | on the tracing is different. Every read from trace_pipe is consumed. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1289 | This means that subsequent reads will be different. The trace |
| 1290 | is live. |
| 1291 | |
| 1292 | # echo ftrace > /debug/tracing/current_tracer |
| 1293 | # cat /debug/tracing/trace_pipe > /tmp/trace.out & |
| 1294 | [1] 4153 |
| 1295 | # echo 1 > /debug/tracing/tracing_enabled |
| 1296 | # usleep 1 |
| 1297 | # echo 0 > /debug/tracing/tracing_enabled |
| 1298 | # cat /debug/tracing/trace |
| 1299 | # tracer: ftrace |
| 1300 | # |
| 1301 | # TASK-PID CPU# TIMESTAMP FUNCTION |
| 1302 | # | | | | | |
| 1303 | |
| 1304 | # |
| 1305 | # cat /tmp/trace.out |
| 1306 | bash-4043 [00] 41.267106: finish_task_switch <-schedule |
| 1307 | bash-4043 [00] 41.267106: hrtick_set <-schedule |
| 1308 | bash-4043 [00] 41.267107: hrtick_clear <-hrtick_set |
| 1309 | bash-4043 [00] 41.267108: wait_for_completion <-__stop_machine_run |
| 1310 | bash-4043 [00] 41.267108: wait_for_common <-wait_for_completion |
| 1311 | bash-4043 [00] 41.267109: kthread_stop <-stop_machine_run |
| 1312 | bash-4043 [00] 41.267109: init_waitqueue_head <-kthread_stop |
| 1313 | bash-4043 [00] 41.267110: wake_up_process <-kthread_stop |
| 1314 | bash-4043 [00] 41.267110: try_to_wake_up <-wake_up_process |
| 1315 | bash-4043 [00] 41.267111: select_task_rq_rt <-try_to_wake_up |
| 1316 | |
| 1317 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1318 | Note, reading the trace_pipe file will block until more input is added. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1319 | By changing the tracer, trace_pipe will issue an EOF. We needed |
| 1320 | to set the ftrace tracer _before_ cating the trace_pipe file. |
| 1321 | |
| 1322 | |
| 1323 | trace entries |
| 1324 | ------------- |
| 1325 | |
| 1326 | Having too much or not enough data can be troublesome in diagnosing |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1327 | an issue in the kernel. The file trace_entries is used to modify |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1328 | the size of the internal trace buffers. The number listed |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1329 | is the number of entries that can be recorded per CPU. To know |
| 1330 | the full size, multiply the number of possible CPUS with the |
| 1331 | number of entries. |
| 1332 | |
| 1333 | # cat /debug/tracing/trace_entries |
| 1334 | 65620 |
| 1335 | |
Steven Rostedt | a41eeba | 2008-07-14 16:41:12 -0400 | [diff] [blame] | 1336 | Note, to modify this, you must have tracing completely disabled. To do that, |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1337 | echo "none" into the current_tracer. If the current_tracer is not set |
| 1338 | to "none", an EINVAL error will be returned. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1339 | |
| 1340 | # echo none > /debug/tracing/current_tracer |
| 1341 | # echo 100000 > /debug/tracing/trace_entries |
| 1342 | # cat /debug/tracing/trace_entries |
| 1343 | 100045 |
| 1344 | |
| 1345 | |
| 1346 | Notice that we echoed in 100,000 but the size is 100,045. The entries |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1347 | are held in individual pages. It allocates the number of pages it takes |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1348 | to fulfill the request. If more entries may fit on the last page |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1349 | then they will be added. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1350 | |
| 1351 | # echo 1 > /debug/tracing/trace_entries |
| 1352 | # cat /debug/tracing/trace_entries |
| 1353 | 85 |
| 1354 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1355 | This shows us that 85 entries can fit in a single page. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1356 | |
Steven Rostedt | f2d9c74 | 2008-07-15 10:57:33 -0400 | [diff] [blame] | 1357 | The number of pages which will be allocated is limited to a percentage |
| 1358 | of available memory. Allocating too much will produce an error. |
Steven Rostedt | eb6d42e | 2008-07-10 12:46:01 -0400 | [diff] [blame] | 1359 | |
| 1360 | # echo 1000000000000 > /debug/tracing/trace_entries |
| 1361 | -bash: echo: write error: Cannot allocate memory |
| 1362 | # cat /debug/tracing/trace_entries |
| 1363 | 85 |
| 1364 | |