Lines Matching refs:trace
89 This sets or displays whether writing to the trace
95 trace:
97 This file holds the output of the trace in a human
102 The output is the same as the "trace" file but this
105 retrieved. Unlike the "trace" file, this file is a
110 "trace" file is static, and if the tracer is not
124 trace option (also in trace_options). Options may also be set
132 This time is saved in this file. The max trace
133 will also be stored, and displayed by "trace".
134 A new max trace will only be recorded if the
140 Some latency tracers will record a trace whenever the
148 buffer holds. By default, the trace buffers are the same size
151 trace buffers are allocated in pages (blocks of memory
161 This displays the total combined size of all the trace buffers.
178 This is a mask that lets the user only trace
191 will limit the trace to only those functions.
205 Have the function tracer only trace a single thread.
209 Have the events only trace a task with a PID listed in this file.
210 Note, sched_switch and sched_wake_up will also trace events
222 has processed and can trace. These are the function
231 Not only does the trace infrastructure use ftrace function
232 trace utility, but other subsystems might too. This file
265 Enable dynamic trace points. See kprobetrace.txt.
269 Dynamic trace points stats. See kprobetrace.txt.
274 it will trace into a function. Setting this to a value of
290 Only the pid of the task is recorded in a trace event unless
299 take a snapshot of the current running trace.
310 This displays the stack back trace of the largest stack
405 This is a way to make multiple trace buffers where different
411 This is the trace event directory. It holds event tracepoints
422 This is a directory that contains the trace per_cpu information.
433 per_cpu/cpu0/trace:
435 This is similar to the "trace" file, but it will only display
500 Function call tracer to trace all kernel functions.
514 the trace with the longest max latency.
516 it replaces the old trace. It is best to view this
517 trace with the latency-format option enabled.
545 This is the "trace nothing" tracer. To remove all
560 Here is an example of the output format of the file "trace"
587 the trace. In this case the tracer is "function". Then it shows the
600 Latency trace format
604 tracers is set, the trace file gives somewhat more information to see
605 why a latency happened. Here is a typical trace.
609 # irqsoff latency trace v1.1.5 on 3.8.0-test+
630 ps-6143 2d..1 306us : <stack trace>
644 for which interrupts were disabled. It gives the trace version (which
647 of trace entries displayed and the total number (both are four: #4/4).
660 The next lines after the header are the trace itself. The header
663 cmd: The name of the process in the trace.
690 time: When the latency-format option is enabled, the trace file
692 trace. This differs from the output when latency-format
698 current trace and the next trace.
707 The rest is the same as the 'trace' file.
709 Note, the latency tracers will usually end with a back trace
716 what gets printed in the trace output, or manipulate the tracers.
745 function-trace
782 verbose - This deals with the trace file when the
799 stacktrace - This is one of the options that changes the trace
800 itself. When a trace is recorded, so is the stack
802 trace sites.
812 the trace is reported, it shows the oldest events first,
825 userstacktrace - This option changes the trace. It records a
836 trace,trace_pipe. Example:
849 latency-format - This option changes the trace. When
850 it is enabled, the trace displays
853 trace format".
867 in the sched_switch trace point to fill comm cache
873 overwrite - This controls what happens when the trace buffer is
883 When disabled, the trace looks like:
901 function-trace - The latency tracers will enable function tracing
903 it is disabled, the latency tracers do not trace
923 the trace leading up to that latency point so that every time a
924 new maximum is reached, the old saved trace is discarded and the
925 new trace is saved.
930 # echo 0 > options/function-trace
937 # cat trace
940 # irqsoff latency trace v1.1.5 on 3.8.0-test+
961 <idle>-0 0dNs3 25us : <stack trace>
984 Note the above example had function-trace not set. If we set
985 function-trace, we get a much larger output:
987 with echo 1 > options/function-trace
991 # irqsoff latency trace v1.1.5 on 3.8.0-test+
1028 bash-2042 3d..1 120us : <stack trace>
1060 trace has provided some very helpful debugging information.
1076 # echo 0 > options/function-trace
1083 # cat trace
1086 # preemptoff latency trace v1.1.5 on 3.8.0-test+
1107 sshd-1991 1d..1 52us : <stack trace>
1123 # preemptoff latency trace v1.1.5 on 3.8.0-test+
1175 bash-1994 1.N.1 104us : <stack trace>
1185 The above is an example of the preemptoff trace with
1186 function-trace set. Here we see that interrupts were not disabled
1218 But neither will trace the time that interrupts and/or
1223 Again, using this trace is much like the irqsoff and preemptoff
1226 # echo 0 > options/function-trace
1233 # cat trace
1236 # preemptirqsoff latency trace v1.1.5 on 3.8.0-test+
1257 ls-2230 3...1 111us : <stack trace>
1285 Here is a trace with function-trace set:
1289 # preemptirqsoff latency trace v1.1.5 on 3.8.0-test+
1368 ls-2269 3d... 186us : <stack trace>
1378 This is an interesting trace. It started with kworker running and
1396 # echo 0 > options/function-trace
1402 # cat trace
1405 # wakeup latency trace v1.1.5 on 3.8.0-test+
1432 trace is to concentrate only on Real-Time tasks.
1462 # echo 0 > options/function-trace
1468 # cat trace
1473 # wakeup_rt latency trace v1.1.5 on 3.8.0-test+
1495 to perform the task switch. Note, since the trace point in the schedule
1505 Note, that the trace data shows the internal priority (99 - rtprio).
1514 Doing the same with chrt -r 5 and function-trace set.
1516 echo 1 > options/function-trace
1520 # wakeup_rt latency trace v1.1.5 on 3.8.0-test+
1621 This isn't that big of a trace, even with function tracing enabled,
1622 so I included the entire trace.
1635 # echo 0 > options/function-trace
1642 # cat trace
1645 # wakeup_rt latency trace v1.1.5 on 3.8.0-test+
1687 # cat trace
1712 Sometimes using echo to stop the trace is not sufficient because
1736 By writing into set_ftrace_pid you can trace a
1745 # cat trace | head
1757 # cat trace |head
1769 If you want to trace a function when executing, you could use
1931 enabled. It is sometimes better to only trace one cpu (see
2053 trace buffer.
2056 enabled for functions whose start is in the trace buffer,
2149 wish to trace and which ones we want the mcount calls to remain
2182 # cat trace
2345 If you want to trace only one function and all of its children,
2350 will produce the following "expanded" trace of the __do_fault()
2393 Now if you want to go back to trace all functions you can clear
2451 Disable tracing for all modules, but still trace kernel:
2502 These commands can enable or disable a trace event. Note, because
2504 are registered, the trace point is activated, but disabled in
2528 something, and want to dump the trace when a certain function
2541 The trace_pipe outputs the same content as the trace file, but
2544 different. The trace is live.
2547 # cat trace_pipe > /tmp/trace.out &
2552 # cat trace
2566 # cat /tmp/trace.out
2581 trace entries
2586 used to modify the size of the internal trace buffers. The
2640 Snapshot preserves a current trace buffer at a particular point
2654 "trace" (described above in the section "The File
2689 # cat trace
2727 set_event snapshot trace trace_clock trace_marker trace_options
2802 # cat instances/zoot/trace
2826 You can see that the trace of the top most trace buffer shows only
2836 Note, if a process has a trace file open in one of the instance
2840 Stack trace
2859 You can also enable it from the kernel command line to trace
2899 kernel/trace/*.c files.