sphinx.addnodesdocument)}( rawsourcechildren]( translations LanguagesNode)}(hhh](h pending_xref)}(hhh]docutils.nodesTextChinese (Simplified)}parenthsba attributes}(ids]classes]names]dupnames]backrefs] refdomainstdreftypedoc reftarget /translations/zh_CN/trace/ftracemodnameN classnameN refexplicitutagnamehhh ubh)}(hhh]hChinese (Traditional)}hh2sbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget /translations/zh_TW/trace/ftracemodnameN classnameN refexplicituh1hhh ubh)}(hhh]hItalian}hhFsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget /translations/it_IT/trace/ftracemodnameN classnameN refexplicituh1hhh ubh)}(hhh]hJapanese}hhZsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget /translations/ja_JP/trace/ftracemodnameN classnameN refexplicituh1hhh ubh)}(hhh]hKorean}hhnsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget /translations/ko_KR/trace/ftracemodnameN classnameN refexplicituh1hhh ubh)}(hhh]hSpanish}hhsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget /translations/sp_SP/trace/ftracemodnameN classnameN refexplicituh1hhh ubeh}(h]h ]h"]h$]h&]current_languageEnglishuh1h hh _documenthsourceNlineNubhsection)}(hhh](htitle)}(hftrace - Function Tracerh]hftrace - Function Tracer}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhhh:/var/lib/git/docbuild/linux/Documentation/trace/ftrace.rsthKubh paragraph)}(hCopyright 2008 Red Hat Inc.h]hCopyright 2008 Red Hat Inc.}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh field_list)}(hhh](hfield)}(hhh](h field_name)}(hAuthorh]hAuthor}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhhhKubh field_body)}(h$Steven Rostedt h]h)}(hhh](hSteven Rostedt <}(hhhhhNhNubh reference)}(hsrostedt@redhat.comh]hsrostedt@redhat.com}(hhhhhNhNubah}(h]h ]h"]h$]h&]refurimailto:srostedt@redhat.comuh1hhhubh>}(hhhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKhhubah}(h]h ]h"]h$]h&]uh1hhhubeh}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh)}(hhh](h)}(hLicenseh]hLicense}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhKubh)}(hPThe GNU Free Documentation License, Version 1.2 (dual licensed under the GPL v2)h]h)}(hPThe GNU Free Documentation License, Version 1.2 (dual licensed under the GPL v2)h]hPThe GNU Free Documentation License, Version 1.2 (dual licensed under the GPL v2)}(hj+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj'ubah}(h]h ]h"]h$]h&]uh1hhjubeh}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh)}(hhh](h)}(hOriginal Reviewersh]hOriginal Reviewers}(hjHhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjEhhhKubh)}(hMElias Oltmanns, Randy Dunlap, Andrew Morton, John Kacur, and David Teigland. h]h)}(hLElias Oltmanns, Randy Dunlap, Andrew Morton, John Kacur, and David Teigland.h]hLElias Oltmanns, Randy Dunlap, Andrew Morton, John Kacur, and David Teigland.}(hjZhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK hjVubah}(h]h ]h"]h$]h&]uh1hhjEubeh}(h]h ]h"]h$]h&]uh1hhhhK hhhhubeh}(h]h ]h"]h$]h&]uh1hhhhhhhhKubh bullet_list)}(hhh](h list_item)}(hWritten for: 2.6.28-rc2h]h)}(hjh]hWritten for: 2.6.28-rc2}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK hjubah}(h]h ]h"]h$]h&]uh1jhj|hhhhhNubj)}(hUpdated for: 3.10h]h)}(hjh]hUpdated for: 3.10}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1jhj|hhhhhNubj)}(h=Updated for: 4.13 - Copyright 2017 VMware Inc. Steven Rostedth]h)}(hjh]h=Updated for: 4.13 - Copyright 2017 VMware Inc. Steven Rostedt}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1jhj|hhhhhNubj)}(h>Converted to rst format - Changbin Du h]h)}(h=Converted to rst format - Changbin Du h](h'Converted to rst format - Changbin Du <}(hjhhhNhNubh)}(hchangbin.du@intel.comh]hchangbin.du@intel.com}(hjhhhNhNubah}(h]h ]h"]h$]h&]refurimailto:changbin.du@intel.comuh1hhjubh>}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1jhj|hhhhhNubeh}(h]h ]h"]h$]h&]bullet-uh1jzhhhK hhhhubh)}(hhh](h)}(h Introductionh]h Introduction}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhKubh)}(hFtrace is an internal tracer designed to help out developers and designers of systems to find what is going on inside the kernel. It can be used for debugging or analyzing latencies and performance issues that take place outside of user-space.h]hFtrace is an internal tracer designed to help out developers and designers of systems to find what is going on inside the kernel. It can be used for debugging or analyzing latencies and performance issues that take place outside of user-space.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjhhubh)}(hX4Although ftrace is typically considered the function tracer, it is really a framework of several assorted tracing utilities. There's latency tracing to examine what occurs between interrupts disabled and enabled, as well as for preemption and from a time a task is woken to the task is actually scheduled in.h]hX6Although ftrace is typically considered the function tracer, it is really a framework of several assorted tracing utilities. There’s latency tracing to examine what occurs between interrupts disabled and enabled, as well as for preemption and from a time a task is woken to the task is actually scheduled in.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjhhubh)}(hOne of the most common uses of ftrace is the event tracing. Throughout the kernel is hundreds of static event points that can be enabled via the tracefs file system to see what is going on in certain parts of the kernel.h]hOne of the most common uses of ftrace is the event tracing. Throughout the kernel is hundreds of static event points that can be enabled via the tracefs file system to see what is going on in certain parts of the kernel.}(hj'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK hjhhubh)}(h$See events.rst for more information.h]h$See events.rst for more information.}(hj5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK%hjhhubeh}(h] introductionah ]h"] introductionah$]h&]uh1hhhhhhhhKubh)}(hhh](h)}(hImplementation Detailsh]hImplementation Details}(hjNhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjKhhhhhK)ubh)}(hPSee Documentation/trace/ftrace-design.rst for details for arch porters and such.h]hPSee Documentation/trace/ftrace-design.rst for details for arch porters and such.}(hj\hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK+hjKhhubeh}(h]implementation-detailsah ]h"]implementation detailsah$]h&]uh1hhhhhhhhK)ubh)}(hhh](h)}(hThe File Systemh]hThe File System}(hjuhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjrhhhhhK/ubh)}(heFtrace uses the tracefs file system to hold the control files as well as the files to display output.h]heFtrace uses the tracefs file system to hold the control files as well as the files to display output.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK1hjrhhubh)}(hWhen tracefs is configured into the kernel (which selecting any ftrace option will do) the directory /sys/kernel/tracing will be created. To mount this directory, you can add to your /etc/fstab file::h]hWhen tracefs is configured into the kernel (which selecting any ftrace option will do) the directory /sys/kernel/tracing will be created. To mount this directory, you can add to your /etc/fstab file:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK4hjrhhubh literal_block)}(hItracefs /sys/kernel/tracing tracefs defaults 0 0h]hItracefs /sys/kernel/tracing tracefs defaults 0 0}hjsbah}(h]h ]h"]h$]h&] xml:spacepreserveuh1jhhhK8hjrhhubh)}(h&Or you can mount it at run time with::h]h%Or you can mount it at run time with:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK:hjrhhubj)}(h*mount -t tracefs nodev /sys/kernel/tracingh]h*mount -t tracefs nodev /sys/kernel/tracing}hjsbah}(h]h ]h"]h$]h&]jjuh1jhhhKhjrhhubj)}(h"ln -s /sys/kernel/tracing /tracingh]h"ln -s /sys/kernel/tracing /tracing}hjsbah}(h]h ]h"]h$]h&]jjuh1jhhhKAhjrhhubh attention)}(hXBefore 4.1, all ftrace tracing control files were within the debugfs file system, which is typically located at /sys/kernel/debug/tracing. For backward compatibility, when mounting the debugfs file system, the tracefs file system will be automatically mounted at: /sys/kernel/debug/tracing All files located in the tracefs file system will be located in that debugfs file system directory as well.h](h)}(hXBefore 4.1, all ftrace tracing control files were within the debugfs file system, which is typically located at /sys/kernel/debug/tracing. For backward compatibility, when mounting the debugfs file system, the tracefs file system will be automatically mounted at:h]hXBefore 4.1, all ftrace tracing control files were within the debugfs file system, which is typically located at /sys/kernel/debug/tracing. For backward compatibility, when mounting the debugfs file system, the tracefs file system will be automatically mounted at:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKEhjubh)}(h/sys/kernel/debug/tracingh]h/sys/kernel/debug/tracing}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKJhjubh)}(hkAll files located in the tracefs file system will be located in that debugfs file system directory as well.h]hkAll files located in the tracefs file system will be located in that debugfs file system directory as well.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKLhjubeh}(h]h ]h"]h$]h&]uh1jhjrhhhhhNubj)}(hX9Any selected ftrace option will also create the tracefs file system. The rest of the document will assume that you are in the ftrace directory (cd /sys/kernel/tracing) and will only concentrate on the files within that directory and not distract from the content with the extended "/sys/kernel/tracing" path name.h]h)}(hX9Any selected ftrace option will also create the tracefs file system. The rest of the document will assume that you are in the ftrace directory (cd /sys/kernel/tracing) and will only concentrate on the files within that directory and not distract from the content with the extended "/sys/kernel/tracing" path name.h]hX=Any selected ftrace option will also create the tracefs file system. The rest of the document will assume that you are in the ftrace directory (cd /sys/kernel/tracing) and will only concentrate on the files within that directory and not distract from the content with the extended “/sys/kernel/tracing” path name.}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKQhjubah}(h]h ]h"]h$]h&]uh1jhjrhhhhhNubh)}(hFThat's it! (assuming that you have ftrace configured into your kernel)h]hHThat’s it! (assuming that you have ftrace configured into your kernel)}(hj7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKWhjrhhubh)}(hAfter mounting tracefs you will have access to the control and output files of ftrace. Here is a list of some of the key files:h]hAfter mounting tracefs you will have access to the control and output files of ftrace. Here is a list of some of the key files:}(hjEhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKYhjrhhubh block_quote)}(hXkNote: all time values are in microseconds. current_tracer: This is used to set or display the current tracer that is configured. Changing the current tracer clears the ring buffer content as well as the "snapshot" buffer. available_tracers: This holds the different types of tracers that have been compiled into the kernel. The tracers listed here can be configured by echoing their name into current_tracer. tracing_on: This sets or displays whether writing to the trace ring buffer is enabled. Echo 0 into this file to disable the tracer or 1 to enable it. Note, this only disables writing to the ring buffer, the tracing overhead may still be occurring. The kernel function tracing_off() can be used within the kernel to disable writing to the ring buffer, which will set this file to "0". User space can re-enable tracing by echoing "1" into the file. Note, the function and event trigger "traceoff" will also set this file to zero and stop tracing. Which can also be re-enabled by user space using this file. trace: This file holds the output of the trace in a human readable format (described below). Opening this file for writing with the O_TRUNC flag clears the ring buffer content. Note, this file is not a consumer. If tracing is off (no tracer running, or tracing_on is zero), it will produce the same output each time it is read. When tracing is on, it may produce inconsistent results as it tries to read the entire buffer without consuming it. trace_pipe: The output is the same as the "trace" file but this file is meant to be streamed with live tracing. Reads from this file will block until new data is retrieved. Unlike the "trace" file, this file is a consumer. This means reading from this file causes sequential reads to display more current data. Once data is read from this file, it is consumed, and will not be read again with a sequential read. The "trace" file is static, and if the tracer is not adding more data, it will display the same information every time it is read. trace_options: This file lets the user control the amount of data that is displayed in one of the above output files. Options also exist to modify how a tracer or events work (stack traces, timestamps, etc). options: This is a directory that has a file for every available trace option (also in trace_options). Options may also be set or cleared by writing a "1" or "0" respectively into the corresponding file with the option name. tracing_max_latency: Some of the tracers record the max latency. For example, the maximum time that interrupts are disabled. The maximum time is saved in this file. The max trace will also be stored, and displayed by "trace". A new max trace will only be recorded if the latency is greater than the value in this file (in microseconds). By echoing in a time into this file, no latency will be recorded unless it is greater than the time in this file. tracing_thresh: Some latency tracers will record a trace whenever the latency is greater than the number in this file. Only active when the file contains a number greater than 0. (in microseconds) buffer_percent: This is the watermark for how much the ring buffer needs to be filled before a waiter is woken up. That is, if an application calls a blocking read syscall on one of the per_cpu trace_pipe_raw files, it will block until the given amount of data specified by buffer_percent is in the ring buffer before it wakes the reader up. This also controls how the splice system calls are blocked on this file:: 0 - means to wake up as soon as there is any data in the ring buffer. 50 - means to wake up when roughly half of the ring buffer sub-buffers are full. 100 - means to block until the ring buffer is totally full and is about to start overwriting the older data. buffer_size_kb: This sets or displays the number of kilobytes each CPU buffer holds. By default, the trace buffers are the same size for each CPU. The displayed number is the size of the CPU buffer and not total size of all buffers. The trace buffers are allocated in pages (blocks of memory that the kernel uses for allocation, usually 4 KB in size). A few extra pages may be allocated to accommodate buffer management meta-data. If the last page allocated has room for more bytes than requested, the rest of the page will be used, making the actual allocation bigger than requested or shown. ( Note, the size may not be a multiple of the page size due to buffer management meta-data. ) Buffer sizes for individual CPUs may vary (see "per_cpu/cpu0/buffer_size_kb" below), and if they do this file will show "X". buffer_total_size_kb: This displays the total combined size of all the trace buffers. buffer_subbuf_size_kb: This sets or displays the sub buffer size. The ring buffer is broken up into several same size "sub buffers". An event can not be bigger than the size of the sub buffer. Normally, the sub buffer is the size of the architecture's page (4K on x86). The sub buffer also contains meta data at the start which also limits the size of an event. That means when the sub buffer is a page size, no event can be larger than the page size minus the sub buffer meta data. Note, the buffer_subbuf_size_kb is a way for the user to specify the minimum size of the subbuffer. The kernel may make it bigger due to the implementation details, or simply fail the operation if the kernel can not handle the request. Changing the sub buffer size allows for events to be larger than the page size. Note: When changing the sub-buffer size, tracing is stopped and any data in the ring buffer and the snapshot buffer will be discarded. free_buffer: If a process is performing tracing, and the ring buffer should be shrunk "freed" when the process is finished, even if it were to be killed by a signal, this file can be used for that purpose. On close of this file, the ring buffer will be resized to its minimum size. Having a process that is tracing also open this file, when the process exits its file descriptor for this file will be closed, and in doing so, the ring buffer will be "freed". It may also stop tracing if disable_on_free option is set. tracing_cpumask: This is a mask that lets the user only trace on specified CPUs. The format is a hex string representing the CPUs. set_ftrace_filter: When dynamic ftrace is configured in (see the section below "dynamic ftrace"), the code is dynamically modified (code text rewrite) to disable calling of the function profiler (mcount). This lets tracing be configured in with practically no overhead in performance. This also has a side effect of enabling or disabling specific functions to be traced. Echoing names of functions into this file will limit the trace to only those functions. This influences the tracers "function" and "function_graph" and thus also function profiling (see "function_profile_enabled"). The functions listed in "available_filter_functions" are what can be written into this file. This interface also allows for commands to be used. See the "Filter commands" section for more details. As a speed up, since processing strings can be quite expensive and requires a check of all functions registered to tracing, instead an index can be written into this file. A number (starting with "1") written will instead select the same corresponding at the line position of the "available_filter_functions" file. set_ftrace_notrace: This has an effect opposite to that of set_ftrace_filter. Any function that is added here will not be traced. If a function exists in both set_ftrace_filter and set_ftrace_notrace, the function will _not_ be traced. set_ftrace_pid: Have the function tracer only trace the threads whose PID are listed in this file. If the "function-fork" option is set, then when a task whose PID is listed in this file forks, the child's PID will automatically be added to this file, and the child will be traced by the function tracer as well. This option will also cause PIDs of tasks that exit to be removed from the file. set_ftrace_notrace_pid: Have the function tracer ignore threads whose PID are listed in this file. If the "function-fork" option is set, then when a task whose PID is listed in this file forks, the child's PID will automatically be added to this file, and the child will not be traced by the function tracer as well. This option will also cause PIDs of tasks that exit to be removed from the file. If a PID is in both this file and "set_ftrace_pid", then this file takes precedence, and the thread will not be traced. set_event_pid: Have the events only trace a task with a PID listed in this file. Note, sched_switch and sched_wake_up will also trace events listed in this file. To have the PIDs of children of tasks with their PID in this file added on fork, enable the "event-fork" option. That option will also cause the PIDs of tasks to be removed from this file when the task exits. set_event_notrace_pid: Have the events not trace a task with a PID listed in this file. Note, sched_switch and sched_wakeup will trace threads not listed in this file, even if a thread's PID is in the file if the sched_switch or sched_wakeup events also trace a thread that should be traced. To have the PIDs of children of tasks with their PID in this file added on fork, enable the "event-fork" option. That option will also cause the PIDs of tasks to be removed from this file when the task exits. set_graph_function: Functions listed in this file will cause the function graph tracer to only trace these functions and the functions that they call. (See the section "dynamic ftrace" for more details). Note, set_ftrace_filter and set_ftrace_notrace still affects what functions are being traced. set_graph_notrace: Similar to set_graph_function, but will disable function graph tracing when the function is hit until it exits the function. This makes it possible to ignore tracing functions that are called by a specific function. available_filter_functions: This lists the functions that ftrace has processed and can trace. These are the function names that you can pass to "set_ftrace_filter", "set_ftrace_notrace", "set_graph_function", or "set_graph_notrace". (See the section "dynamic ftrace" below for more details.) available_filter_functions_addrs: Similar to available_filter_functions, but with address displayed for each function. The displayed address is the patch-site address and can differ from /proc/kallsyms address. dyn_ftrace_total_info: This file is for debugging purposes. The number of functions that have been converted to nops and are available to be traced. enabled_functions: This file is more for debugging ftrace, but can also be useful in seeing if any function has a callback attached to it. Not only does the trace infrastructure use ftrace function trace utility, but other subsystems might too. This file displays all functions that have a callback attached to them as well as the number of callbacks that have been attached. Note, a callback may also call multiple functions which will not be listed in this count. If the callback registered to be traced by a function with the "save regs" attribute (thus even more overhead), a 'R' will be displayed on the same line as the function that is returning registers. If the callback registered to be traced by a function with the "ip modify" attribute (thus the regs->ip can be changed), an 'I' will be displayed on the same line as the function that can be overridden. If a non ftrace trampoline is attached (BPF) a 'D' will be displayed. Note, normal ftrace trampolines can also be attached, but only one "direct" trampoline can be attached to a given function at a time. Some architectures can not call direct trampolines, but instead have the ftrace ops function located above the function entry point. In such cases an 'O' will be displayed. If a function had either the "ip modify" or a "direct" call attached to it in the past, a 'M' will be shown. This flag is never cleared. It is used to know if a function was every modified by the ftrace infrastructure, and can be used for debugging. If the architecture supports it, it will also show what callback is being directly called by the function. If the count is greater than 1 it most likely will be ftrace_ops_list_func(). If the callback of a function jumps to a trampoline that is specific to the callback and which is not the standard trampoline, its address will be printed as well as the function that the trampoline calls. touched_functions: This file contains all the functions that ever had a function callback to it via the ftrace infrastructure. It has the same format as enabled_functions but shows all functions that have every been traced. To see any function that has every been modified by "ip modify" or a direct trampoline, one can perform the following command: grep ' M ' /sys/kernel/tracing/touched_functions function_profile_enabled: When set it will enable all functions with either the function tracer, or if configured, the function graph tracer. It will keep a histogram of the number of functions that were called and if the function graph tracer was configured, it will also keep track of the time spent in those functions. The histogram content can be displayed in the files: trace_stat/function ( function0, function1, etc). trace_stat: A directory that holds different tracing stats. kprobe_events: Enable dynamic trace points. See kprobetrace.rst. kprobe_profile: Dynamic trace points stats. See kprobetrace.rst. max_graph_depth: Used with the function graph tracer. This is the max depth it will trace into a function. Setting this to a value of one will show only the first kernel function that is called from user space. printk_formats: This is for tools that read the raw format files. If an event in the ring buffer references a string, only a pointer to the string is recorded into the buffer and not the string itself. This prevents tools from knowing what that string was. This file displays the string and address for the string allowing tools to map the pointers to what the strings were. saved_cmdlines: Only the pid of the task is recorded in a trace event unless the event specifically saves the task comm as well. Ftrace makes a cache of pid mappings to comms to try to display comms for events. If a pid for a comm is not listed, then "<...>" is displayed in the output. If the option "record-cmd" is set to "0", then comms of tasks will not be saved during recording. By default, it is enabled. saved_cmdlines_size: By default, 128 comms are saved (see "saved_cmdlines" above). To increase or decrease the amount of comms that are cached, echo the number of comms to cache into this file. saved_tgids: If the option "record-tgid" is set, on each scheduling context switch the Task Group ID of a task is saved in a table mapping the PID of the thread to its TGID. By default, the "record-tgid" option is disabled. snapshot: This displays the "snapshot" buffer and also lets the user take a snapshot of the current running trace. See the "Snapshot" section below for more details. stack_max_size: When the stack tracer is activated, this will display the maximum stack size it has encountered. See the "Stack Trace" section below. stack_trace: This displays the stack back trace of the largest stack that was encountered when the stack tracer is activated. See the "Stack Trace" section below. stack_trace_filter: This is similar to "set_ftrace_filter" but it limits what functions the stack tracer will check. trace_clock: Whenever an event is recorded into the ring buffer, a "timestamp" is added. This stamp comes from a specified clock. By default, ftrace uses the "local" clock. This clock is very fast and strictly per cpu, but on some systems it may not be monotonic with respect to other CPUs. In other words, the local clocks may not be in sync with local clocks on other CPUs. Usual clocks for tracing:: # cat trace_clock [local] global counter x86-tsc The clock with the square brackets around it is the one in effect. local: Default clock, but may not be in sync across CPUs global: This clock is in sync with all CPUs but may be a bit slower than the local clock. counter: This is not a clock at all, but literally an atomic counter. It counts up one by one, but is in sync with all CPUs. This is useful when you need to know exactly the order events occurred with respect to each other on different CPUs. uptime: This uses the jiffies counter and the time stamp is relative to the time since boot up. perf: This makes ftrace use the same clock that perf uses. Eventually perf will be able to read ftrace buffers and this will help out in interleaving the data. x86-tsc: Architectures may define their own clocks. For example, x86 uses its own TSC cycle clock here. ppc-tb: This uses the powerpc timebase register value. This is in sync across CPUs and can also be used to correlate events across hypervisor/guest if tb_offset is known. mono: This uses the fast monotonic clock (CLOCK_MONOTONIC) which is monotonic and is subject to NTP rate adjustments. mono_raw: This is the raw monotonic clock (CLOCK_MONOTONIC_RAW) which is monotonic but is not subject to any rate adjustments and ticks at the same rate as the hardware clocksource. boot: This is the boot clock (CLOCK_BOOTTIME) and is based on the fast monotonic clock, but also accounts for time spent in suspend. Since the clock access is designed for use in tracing in the suspend path, some side effects are possible if clock is accessed after the suspend time is accounted before the fast mono clock is updated. In this case, the clock update appears to happen slightly sooner than it normally would have. Also on 32-bit systems, it's possible that the 64-bit boot offset sees a partial update. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_boot_fast_ns() function for more information. tai: This is the tai clock (CLOCK_TAI) and is derived from the wall- clock time. However, this clock does not experience discontinuities and backwards jumps caused by NTP inserting leap seconds. Since the clock access is designed for use in tracing, side effects are possible. The clock access may yield wrong readouts in case the internal TAI offset is updated e.g., caused by setting the system time or using adjtimex() with an offset. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_tai_fast_ns() function for more information. To set a clock, simply echo the clock name into this file:: # echo global > trace_clock Setting a clock clears the ring buffer content as well as the "snapshot" buffer. trace_marker: This is a very useful file for synchronizing user space with events happening in the kernel. Writing strings into this file will be written into the ftrace buffer. It is useful in applications to open this file at the start of the application and just reference the file descriptor for the file:: void trace_write(const char *fmt, ...) { va_list ap; char buf[256]; int n; if (trace_fd < 0) return; va_start(ap, fmt); n = vsnprintf(buf, 256, fmt, ap); va_end(ap); write(trace_fd, buf, n); } start:: trace_fd = open("trace_marker", O_WRONLY); Note: Writing into the trace_marker file can also initiate triggers that are written into /sys/kernel/tracing/events/ftrace/print/trigger See "Event triggers" in Documentation/trace/events.rst and an example in Documentation/trace/histogram.rst (Section 3.) trace_marker_raw: This is similar to trace_marker above, but is meant for binary data to be written to it, where a tool can be used to parse the data from trace_pipe_raw. uprobe_events: Add dynamic tracepoints in programs. See uprobetracer.rst uprobe_profile: Uprobe statistics. See uprobetrace.txt instances: This is a way to make multiple trace buffers where different events can be recorded in different buffers. See "Instances" section below. events: This is the trace event directory. It holds event tracepoints (also known as static tracepoints) that have been compiled into the kernel. It shows what event tracepoints exist and how they are grouped by system. There are "enable" files at various levels that can enable the tracepoints when a "1" is written to them. See events.rst for more information. set_event: By echoing in the event into this file, will enable that event. See events.rst for more information. available_events: A list of events that can be enabled in tracing. See events.rst for more information. timestamp_mode: Certain tracers may change the timestamp mode used when logging trace events into the event buffer. Events with different modes can coexist within a buffer but the mode in effect when an event is logged determines which timestamp mode is used for that event. The default timestamp mode is 'delta'. Usual timestamp modes for tracing: # cat timestamp_mode [delta] absolute The timestamp mode with the square brackets around it is the one in effect. delta: Default timestamp mode - timestamp is a delta against a per-buffer timestamp. absolute: The timestamp is a full timestamp, not a delta against some other value. As such it takes up more space and is less efficient. hwlat_detector: Directory for the Hardware Latency Detector. See "Hardware Latency Detector" section below. per_cpu: This is a directory that contains the trace per_cpu information. per_cpu/cpu0/buffer_size_kb: The ftrace buffer is defined per_cpu. That is, there's a separate buffer for each CPU to allow writes to be done atomically, and free from cache bouncing. These buffers may have different size buffers. This file is similar to the buffer_size_kb file, but it only displays or sets the buffer size for the specific CPU. (here cpu0). per_cpu/cpu0/trace: This is similar to the "trace" file, but it will only display the data specific for the CPU. If written to, it only clears the specific CPU buffer. per_cpu/cpu0/trace_pipe This is similar to the "trace_pipe" file, and is a consuming read, but it will only display (and consume) the data specific for the CPU. per_cpu/cpu0/trace_pipe_raw For tools that can parse the ftrace ring buffer binary format, the trace_pipe_raw file can be used to extract the data from the ring buffer directly. With the use of the splice() system call, the buffer data can be quickly transferred to a file or to the network where a server is collecting the data. Like trace_pipe, this is a consuming reader, where multiple reads will always produce different data. per_cpu/cpu0/snapshot: This is similar to the main "snapshot" file, but will only snapshot the current CPU (if supported). It only displays the content of the snapshot for a given CPU, and if written to, only clears this CPU buffer. per_cpu/cpu0/snapshot_raw: Similar to the trace_pipe_raw, but will read the binary format from the snapshot buffer for the given CPU. per_cpu/cpu0/stats: This displays certain stats about the ring buffer: entries: The number of events that are still in the buffer. overrun: The number of lost events due to overwriting when the buffer was full. commit overrun: Should always be zero. This gets set if so many events happened within a nested event (ring buffer is re-entrant), that it fills the buffer and starts dropping events. bytes: Bytes actually read (not overwritten). oldest event ts: The oldest timestamp in the buffer now ts: The current timestamp dropped events: Events lost due to overwrite option being off. read events: The number of events read. h](h)}(h*Note: all time values are in microseconds.h]h*Note: all time values are in microseconds.}(hjYhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK]hjUubjT)}(hXicurrent_tracer: This is used to set or display the current tracer that is configured. Changing the current tracer clears the ring buffer content as well as the "snapshot" buffer. available_tracers: This holds the different types of tracers that have been compiled into the kernel. The tracers listed here can be configured by echoing their name into current_tracer. tracing_on: This sets or displays whether writing to the trace ring buffer is enabled. Echo 0 into this file to disable the tracer or 1 to enable it. Note, this only disables writing to the ring buffer, the tracing overhead may still be occurring. The kernel function tracing_off() can be used within the kernel to disable writing to the ring buffer, which will set this file to "0". User space can re-enable tracing by echoing "1" into the file. Note, the function and event trigger "traceoff" will also set this file to zero and stop tracing. Which can also be re-enabled by user space using this file. trace: This file holds the output of the trace in a human readable format (described below). Opening this file for writing with the O_TRUNC flag clears the ring buffer content. Note, this file is not a consumer. If tracing is off (no tracer running, or tracing_on is zero), it will produce the same output each time it is read. When tracing is on, it may produce inconsistent results as it tries to read the entire buffer without consuming it. trace_pipe: The output is the same as the "trace" file but this file is meant to be streamed with live tracing. Reads from this file will block until new data is retrieved. Unlike the "trace" file, this file is a consumer. This means reading from this file causes sequential reads to display more current data. Once data is read from this file, it is consumed, and will not be read again with a sequential read. The "trace" file is static, and if the tracer is not adding more data, it will display the same information every time it is read. trace_options: This file lets the user control the amount of data that is displayed in one of the above output files. Options also exist to modify how a tracer or events work (stack traces, timestamps, etc). options: This is a directory that has a file for every available trace option (also in trace_options). Options may also be set or cleared by writing a "1" or "0" respectively into the corresponding file with the option name. tracing_max_latency: Some of the tracers record the max latency. For example, the maximum time that interrupts are disabled. The maximum time is saved in this file. The max trace will also be stored, and displayed by "trace". A new max trace will only be recorded if the latency is greater than the value in this file (in microseconds). By echoing in a time into this file, no latency will be recorded unless it is greater than the time in this file. tracing_thresh: Some latency tracers will record a trace whenever the latency is greater than the number in this file. Only active when the file contains a number greater than 0. (in microseconds) buffer_percent: This is the watermark for how much the ring buffer needs to be filled before a waiter is woken up. That is, if an application calls a blocking read syscall on one of the per_cpu trace_pipe_raw files, it will block until the given amount of data specified by buffer_percent is in the ring buffer before it wakes the reader up. This also controls how the splice system calls are blocked on this file:: 0 - means to wake up as soon as there is any data in the ring buffer. 50 - means to wake up when roughly half of the ring buffer sub-buffers are full. 100 - means to block until the ring buffer is totally full and is about to start overwriting the older data. buffer_size_kb: This sets or displays the number of kilobytes each CPU buffer holds. By default, the trace buffers are the same size for each CPU. The displayed number is the size of the CPU buffer and not total size of all buffers. The trace buffers are allocated in pages (blocks of memory that the kernel uses for allocation, usually 4 KB in size). A few extra pages may be allocated to accommodate buffer management meta-data. If the last page allocated has room for more bytes than requested, the rest of the page will be used, making the actual allocation bigger than requested or shown. ( Note, the size may not be a multiple of the page size due to buffer management meta-data. ) Buffer sizes for individual CPUs may vary (see "per_cpu/cpu0/buffer_size_kb" below), and if they do this file will show "X". buffer_total_size_kb: This displays the total combined size of all the trace buffers. buffer_subbuf_size_kb: This sets or displays the sub buffer size. The ring buffer is broken up into several same size "sub buffers". An event can not be bigger than the size of the sub buffer. Normally, the sub buffer is the size of the architecture's page (4K on x86). The sub buffer also contains meta data at the start which also limits the size of an event. That means when the sub buffer is a page size, no event can be larger than the page size minus the sub buffer meta data. Note, the buffer_subbuf_size_kb is a way for the user to specify the minimum size of the subbuffer. The kernel may make it bigger due to the implementation details, or simply fail the operation if the kernel can not handle the request. Changing the sub buffer size allows for events to be larger than the page size. Note: When changing the sub-buffer size, tracing is stopped and any data in the ring buffer and the snapshot buffer will be discarded. free_buffer: If a process is performing tracing, and the ring buffer should be shrunk "freed" when the process is finished, even if it were to be killed by a signal, this file can be used for that purpose. On close of this file, the ring buffer will be resized to its minimum size. Having a process that is tracing also open this file, when the process exits its file descriptor for this file will be closed, and in doing so, the ring buffer will be "freed". It may also stop tracing if disable_on_free option is set. tracing_cpumask: This is a mask that lets the user only trace on specified CPUs. The format is a hex string representing the CPUs. set_ftrace_filter: When dynamic ftrace is configured in (see the section below "dynamic ftrace"), the code is dynamically modified (code text rewrite) to disable calling of the function profiler (mcount). This lets tracing be configured in with practically no overhead in performance. This also has a side effect of enabling or disabling specific functions to be traced. Echoing names of functions into this file will limit the trace to only those functions. This influences the tracers "function" and "function_graph" and thus also function profiling (see "function_profile_enabled"). The functions listed in "available_filter_functions" are what can be written into this file. This interface also allows for commands to be used. See the "Filter commands" section for more details. As a speed up, since processing strings can be quite expensive and requires a check of all functions registered to tracing, instead an index can be written into this file. A number (starting with "1") written will instead select the same corresponding at the line position of the "available_filter_functions" file. set_ftrace_notrace: This has an effect opposite to that of set_ftrace_filter. Any function that is added here will not be traced. If a function exists in both set_ftrace_filter and set_ftrace_notrace, the function will _not_ be traced. set_ftrace_pid: Have the function tracer only trace the threads whose PID are listed in this file. If the "function-fork" option is set, then when a task whose PID is listed in this file forks, the child's PID will automatically be added to this file, and the child will be traced by the function tracer as well. This option will also cause PIDs of tasks that exit to be removed from the file. set_ftrace_notrace_pid: Have the function tracer ignore threads whose PID are listed in this file. If the "function-fork" option is set, then when a task whose PID is listed in this file forks, the child's PID will automatically be added to this file, and the child will not be traced by the function tracer as well. This option will also cause PIDs of tasks that exit to be removed from the file. If a PID is in both this file and "set_ftrace_pid", then this file takes precedence, and the thread will not be traced. set_event_pid: Have the events only trace a task with a PID listed in this file. Note, sched_switch and sched_wake_up will also trace events listed in this file. To have the PIDs of children of tasks with their PID in this file added on fork, enable the "event-fork" option. That option will also cause the PIDs of tasks to be removed from this file when the task exits. set_event_notrace_pid: Have the events not trace a task with a PID listed in this file. Note, sched_switch and sched_wakeup will trace threads not listed in this file, even if a thread's PID is in the file if the sched_switch or sched_wakeup events also trace a thread that should be traced. To have the PIDs of children of tasks with their PID in this file added on fork, enable the "event-fork" option. That option will also cause the PIDs of tasks to be removed from this file when the task exits. set_graph_function: Functions listed in this file will cause the function graph tracer to only trace these functions and the functions that they call. (See the section "dynamic ftrace" for more details). Note, set_ftrace_filter and set_ftrace_notrace still affects what functions are being traced. set_graph_notrace: Similar to set_graph_function, but will disable function graph tracing when the function is hit until it exits the function. This makes it possible to ignore tracing functions that are called by a specific function. available_filter_functions: This lists the functions that ftrace has processed and can trace. These are the function names that you can pass to "set_ftrace_filter", "set_ftrace_notrace", "set_graph_function", or "set_graph_notrace". (See the section "dynamic ftrace" below for more details.) available_filter_functions_addrs: Similar to available_filter_functions, but with address displayed for each function. The displayed address is the patch-site address and can differ from /proc/kallsyms address. dyn_ftrace_total_info: This file is for debugging purposes. The number of functions that have been converted to nops and are available to be traced. enabled_functions: This file is more for debugging ftrace, but can also be useful in seeing if any function has a callback attached to it. Not only does the trace infrastructure use ftrace function trace utility, but other subsystems might too. This file displays all functions that have a callback attached to them as well as the number of callbacks that have been attached. Note, a callback may also call multiple functions which will not be listed in this count. If the callback registered to be traced by a function with the "save regs" attribute (thus even more overhead), a 'R' will be displayed on the same line as the function that is returning registers. If the callback registered to be traced by a function with the "ip modify" attribute (thus the regs->ip can be changed), an 'I' will be displayed on the same line as the function that can be overridden. If a non ftrace trampoline is attached (BPF) a 'D' will be displayed. Note, normal ftrace trampolines can also be attached, but only one "direct" trampoline can be attached to a given function at a time. Some architectures can not call direct trampolines, but instead have the ftrace ops function located above the function entry point. In such cases an 'O' will be displayed. If a function had either the "ip modify" or a "direct" call attached to it in the past, a 'M' will be shown. This flag is never cleared. It is used to know if a function was every modified by the ftrace infrastructure, and can be used for debugging. If the architecture supports it, it will also show what callback is being directly called by the function. If the count is greater than 1 it most likely will be ftrace_ops_list_func(). If the callback of a function jumps to a trampoline that is specific to the callback and which is not the standard trampoline, its address will be printed as well as the function that the trampoline calls. touched_functions: This file contains all the functions that ever had a function callback to it via the ftrace infrastructure. It has the same format as enabled_functions but shows all functions that have every been traced. To see any function that has every been modified by "ip modify" or a direct trampoline, one can perform the following command: grep ' M ' /sys/kernel/tracing/touched_functions function_profile_enabled: When set it will enable all functions with either the function tracer, or if configured, the function graph tracer. It will keep a histogram of the number of functions that were called and if the function graph tracer was configured, it will also keep track of the time spent in those functions. The histogram content can be displayed in the files: trace_stat/function ( function0, function1, etc). trace_stat: A directory that holds different tracing stats. kprobe_events: Enable dynamic trace points. See kprobetrace.rst. kprobe_profile: Dynamic trace points stats. See kprobetrace.rst. max_graph_depth: Used with the function graph tracer. This is the max depth it will trace into a function. Setting this to a value of one will show only the first kernel function that is called from user space. printk_formats: This is for tools that read the raw format files. If an event in the ring buffer references a string, only a pointer to the string is recorded into the buffer and not the string itself. This prevents tools from knowing what that string was. This file displays the string and address for the string allowing tools to map the pointers to what the strings were. saved_cmdlines: Only the pid of the task is recorded in a trace event unless the event specifically saves the task comm as well. Ftrace makes a cache of pid mappings to comms to try to display comms for events. If a pid for a comm is not listed, then "<...>" is displayed in the output. If the option "record-cmd" is set to "0", then comms of tasks will not be saved during recording. By default, it is enabled. saved_cmdlines_size: By default, 128 comms are saved (see "saved_cmdlines" above). To increase or decrease the amount of comms that are cached, echo the number of comms to cache into this file. saved_tgids: If the option "record-tgid" is set, on each scheduling context switch the Task Group ID of a task is saved in a table mapping the PID of the thread to its TGID. By default, the "record-tgid" option is disabled. snapshot: This displays the "snapshot" buffer and also lets the user take a snapshot of the current running trace. See the "Snapshot" section below for more details. stack_max_size: When the stack tracer is activated, this will display the maximum stack size it has encountered. See the "Stack Trace" section below. stack_trace: This displays the stack back trace of the largest stack that was encountered when the stack tracer is activated. See the "Stack Trace" section below. stack_trace_filter: This is similar to "set_ftrace_filter" but it limits what functions the stack tracer will check. trace_clock: Whenever an event is recorded into the ring buffer, a "timestamp" is added. This stamp comes from a specified clock. By default, ftrace uses the "local" clock. This clock is very fast and strictly per cpu, but on some systems it may not be monotonic with respect to other CPUs. In other words, the local clocks may not be in sync with local clocks on other CPUs. Usual clocks for tracing:: # cat trace_clock [local] global counter x86-tsc The clock with the square brackets around it is the one in effect. local: Default clock, but may not be in sync across CPUs global: This clock is in sync with all CPUs but may be a bit slower than the local clock. counter: This is not a clock at all, but literally an atomic counter. It counts up one by one, but is in sync with all CPUs. This is useful when you need to know exactly the order events occurred with respect to each other on different CPUs. uptime: This uses the jiffies counter and the time stamp is relative to the time since boot up. perf: This makes ftrace use the same clock that perf uses. Eventually perf will be able to read ftrace buffers and this will help out in interleaving the data. x86-tsc: Architectures may define their own clocks. For example, x86 uses its own TSC cycle clock here. ppc-tb: This uses the powerpc timebase register value. This is in sync across CPUs and can also be used to correlate events across hypervisor/guest if tb_offset is known. mono: This uses the fast monotonic clock (CLOCK_MONOTONIC) which is monotonic and is subject to NTP rate adjustments. mono_raw: This is the raw monotonic clock (CLOCK_MONOTONIC_RAW) which is monotonic but is not subject to any rate adjustments and ticks at the same rate as the hardware clocksource. boot: This is the boot clock (CLOCK_BOOTTIME) and is based on the fast monotonic clock, but also accounts for time spent in suspend. Since the clock access is designed for use in tracing in the suspend path, some side effects are possible if clock is accessed after the suspend time is accounted before the fast mono clock is updated. In this case, the clock update appears to happen slightly sooner than it normally would have. Also on 32-bit systems, it's possible that the 64-bit boot offset sees a partial update. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_boot_fast_ns() function for more information. tai: This is the tai clock (CLOCK_TAI) and is derived from the wall- clock time. However, this clock does not experience discontinuities and backwards jumps caused by NTP inserting leap seconds. Since the clock access is designed for use in tracing, side effects are possible. The clock access may yield wrong readouts in case the internal TAI offset is updated e.g., caused by setting the system time or using adjtimex() with an offset. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_tai_fast_ns() function for more information. To set a clock, simply echo the clock name into this file:: # echo global > trace_clock Setting a clock clears the ring buffer content as well as the "snapshot" buffer. trace_marker: This is a very useful file for synchronizing user space with events happening in the kernel. Writing strings into this file will be written into the ftrace buffer. It is useful in applications to open this file at the start of the application and just reference the file descriptor for the file:: void trace_write(const char *fmt, ...) { va_list ap; char buf[256]; int n; if (trace_fd < 0) return; va_start(ap, fmt); n = vsnprintf(buf, 256, fmt, ap); va_end(ap); write(trace_fd, buf, n); } start:: trace_fd = open("trace_marker", O_WRONLY); Note: Writing into the trace_marker file can also initiate triggers that are written into /sys/kernel/tracing/events/ftrace/print/trigger See "Event triggers" in Documentation/trace/events.rst and an example in Documentation/trace/histogram.rst (Section 3.) trace_marker_raw: This is similar to trace_marker above, but is meant for binary data to be written to it, where a tool can be used to parse the data from trace_pipe_raw. uprobe_events: Add dynamic tracepoints in programs. See uprobetracer.rst uprobe_profile: Uprobe statistics. See uprobetrace.txt instances: This is a way to make multiple trace buffers where different events can be recorded in different buffers. See "Instances" section below. events: This is the trace event directory. It holds event tracepoints (also known as static tracepoints) that have been compiled into the kernel. It shows what event tracepoints exist and how they are grouped by system. There are "enable" files at various levels that can enable the tracepoints when a "1" is written to them. See events.rst for more information. set_event: By echoing in the event into this file, will enable that event. See events.rst for more information. available_events: A list of events that can be enabled in tracing. See events.rst for more information. timestamp_mode: Certain tracers may change the timestamp mode used when logging trace events into the event buffer. Events with different modes can coexist within a buffer but the mode in effect when an event is logged determines which timestamp mode is used for that event. The default timestamp mode is 'delta'. Usual timestamp modes for tracing: # cat timestamp_mode [delta] absolute The timestamp mode with the square brackets around it is the one in effect. delta: Default timestamp mode - timestamp is a delta against a per-buffer timestamp. absolute: The timestamp is a full timestamp, not a delta against some other value. As such it takes up more space and is less efficient. hwlat_detector: Directory for the Hardware Latency Detector. See "Hardware Latency Detector" section below. per_cpu: This is a directory that contains the trace per_cpu information. per_cpu/cpu0/buffer_size_kb: The ftrace buffer is defined per_cpu. That is, there's a separate buffer for each CPU to allow writes to be done atomically, and free from cache bouncing. These buffers may have different size buffers. This file is similar to the buffer_size_kb file, but it only displays or sets the buffer size for the specific CPU. (here cpu0). per_cpu/cpu0/trace: This is similar to the "trace" file, but it will only display the data specific for the CPU. If written to, it only clears the specific CPU buffer. per_cpu/cpu0/trace_pipe This is similar to the "trace_pipe" file, and is a consuming read, but it will only display (and consume) the data specific for the CPU. per_cpu/cpu0/trace_pipe_raw For tools that can parse the ftrace ring buffer binary format, the trace_pipe_raw file can be used to extract the data from the ring buffer directly. With the use of the splice() system call, the buffer data can be quickly transferred to a file or to the network where a server is collecting the data. Like trace_pipe, this is a consuming reader, where multiple reads will always produce different data. per_cpu/cpu0/snapshot: This is similar to the main "snapshot" file, but will only snapshot the current CPU (if supported). It only displays the content of the snapshot for a given CPU, and if written to, only clears this CPU buffer. per_cpu/cpu0/snapshot_raw: Similar to the trace_pipe_raw, but will read the binary format from the snapshot buffer for the given CPU. per_cpu/cpu0/stats: This displays certain stats about the ring buffer: entries: The number of events that are still in the buffer. overrun: The number of lost events due to overwriting when the buffer was full. commit overrun: Should always be zero. This gets set if so many events happened within a nested event (ring buffer is re-entrant), that it fills the buffer and starts dropping events. bytes: Bytes actually read (not overwritten). oldest event ts: The oldest timestamp in the buffer now ts: The current timestamp dropped events: Events lost due to overwrite option being off. read events: The number of events read. h](h)}(hcurrent_tracer:h]hcurrent_tracer:}(hjkhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK_hjgubjT)}(hThis is used to set or display the current tracer that is configured. Changing the current tracer clears the ring buffer content as well as the "snapshot" buffer. h]h)}(hThis is used to set or display the current tracer that is configured. Changing the current tracer clears the ring buffer content as well as the "snapshot" buffer.h]hThis is used to set or display the current tracer that is configured. Changing the current tracer clears the ring buffer content as well as the “snapshot” buffer.}(hj}hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKahjyubah}(h]h ]h"]h$]h&]uh1jShhhKahjgubh)}(havailable_tracers:h]havailable_tracers:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKehjgubjT)}(hThis holds the different types of tracers that have been compiled into the kernel. The tracers listed here can be configured by echoing their name into current_tracer. h]h)}(hThis holds the different types of tracers that have been compiled into the kernel. The tracers listed here can be configured by echoing their name into current_tracer.h]hThis holds the different types of tracers that have been compiled into the kernel. The tracers listed here can be configured by echoing their name into current_tracer.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKghjubah}(h]h ]h"]h$]h&]uh1jShhhKghjgubh)}(h tracing_on:h]h tracing_on:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKlhjgubjT)}(hXSThis sets or displays whether writing to the trace ring buffer is enabled. Echo 0 into this file to disable the tracer or 1 to enable it. Note, this only disables writing to the ring buffer, the tracing overhead may still be occurring. The kernel function tracing_off() can be used within the kernel to disable writing to the ring buffer, which will set this file to "0". User space can re-enable tracing by echoing "1" into the file. Note, the function and event trigger "traceoff" will also set this file to zero and stop tracing. Which can also be re-enabled by user space using this file. h](h)}(hThis sets or displays whether writing to the trace ring buffer is enabled. Echo 0 into this file to disable the tracer or 1 to enable it. Note, this only disables writing to the ring buffer, the tracing overhead may still be occurring.h]hThis sets or displays whether writing to the trace ring buffer is enabled. Echo 0 into this file to disable the tracer or 1 to enable it. Note, this only disables writing to the ring buffer, the tracing overhead may still be occurring.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKnhjubh)}(hThe kernel function tracing_off() can be used within the kernel to disable writing to the ring buffer, which will set this file to "0". User space can re-enable tracing by echoing "1" into the file.h]hThe kernel function tracing_off() can be used within the kernel to disable writing to the ring buffer, which will set this file to “0”. User space can re-enable tracing by echoing “1” into the file.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKthjubh)}(hNote, the function and event trigger "traceoff" will also set this file to zero and stop tracing. Which can also be re-enabled by user space using this file.h]hNote, the function and event trigger “traceoff” will also set this file to zero and stop tracing. Which can also be re-enabled by user space using this file.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKyhjubeh}(h]h ]h"]h$]h&]uh1jShhhKnhjgubh)}(htrace:h]htrace:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK}hjgubjT)}(hXThis file holds the output of the trace in a human readable format (described below). Opening this file for writing with the O_TRUNC flag clears the ring buffer content. Note, this file is not a consumer. If tracing is off (no tracer running, or tracing_on is zero), it will produce the same output each time it is read. When tracing is on, it may produce inconsistent results as it tries to read the entire buffer without consuming it. h]h)}(hXThis file holds the output of the trace in a human readable format (described below). Opening this file for writing with the O_TRUNC flag clears the ring buffer content. Note, this file is not a consumer. If tracing is off (no tracer running, or tracing_on is zero), it will produce the same output each time it is read. When tracing is on, it may produce inconsistent results as it tries to read the entire buffer without consuming it.h]hXThis file holds the output of the trace in a human readable format (described below). Opening this file for writing with the O_TRUNC flag clears the ring buffer content. Note, this file is not a consumer. If tracing is off (no tracer running, or tracing_on is zero), it will produce the same output each time it is read. When tracing is on, it may produce inconsistent results as it tries to read the entire buffer without consuming it.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(h trace_pipe:h]h trace_pipe:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(hXThe output is the same as the "trace" file but this file is meant to be streamed with live tracing. Reads from this file will block until new data is retrieved. Unlike the "trace" file, this file is a consumer. This means reading from this file causes sequential reads to display more current data. Once data is read from this file, it is consumed, and will not be read again with a sequential read. The "trace" file is static, and if the tracer is not adding more data, it will display the same information every time it is read. h]h)}(hXThe output is the same as the "trace" file but this file is meant to be streamed with live tracing. Reads from this file will block until new data is retrieved. Unlike the "trace" file, this file is a consumer. This means reading from this file causes sequential reads to display more current data. Once data is read from this file, it is consumed, and will not be read again with a sequential read. The "trace" file is static, and if the tracer is not adding more data, it will display the same information every time it is read.h]hXThe output is the same as the “trace” file but this file is meant to be streamed with live tracing. Reads from this file will block until new data is retrieved. Unlike the “trace” file, this file is a consumer. This means reading from this file causes sequential reads to display more current data. Once data is read from this file, it is consumed, and will not be read again with a sequential read. The “trace” file is static, and if the tracer is not adding more data, it will display the same information every time it is read.}(hj1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj-ubah}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(htrace_options:h]htrace_options:}(hjEhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(hThis file lets the user control the amount of data that is displayed in one of the above output files. Options also exist to modify how a tracer or events work (stack traces, timestamps, etc). h]h)}(hThis file lets the user control the amount of data that is displayed in one of the above output files. Options also exist to modify how a tracer or events work (stack traces, timestamps, etc).h]hThis file lets the user control the amount of data that is displayed in one of the above output files. Options also exist to modify how a tracer or events work (stack traces, timestamps, etc).}(hjWhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjSubah}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(hoptions:h]hoptions:}(hjkhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(hThis is a directory that has a file for every available trace option (also in trace_options). Options may also be set or cleared by writing a "1" or "0" respectively into the corresponding file with the option name. h]h)}(hThis is a directory that has a file for every available trace option (also in trace_options). Options may also be set or cleared by writing a "1" or "0" respectively into the corresponding file with the option name.h]hThis is a directory that has a file for every available trace option (also in trace_options). Options may also be set or cleared by writing a “1” or “0” respectively into the corresponding file with the option name.}(hj}hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjyubah}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(htracing_max_latency:h]htracing_max_latency:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(hXSome of the tracers record the max latency. For example, the maximum time that interrupts are disabled. The maximum time is saved in this file. The max trace will also be stored, and displayed by "trace". A new max trace will only be recorded if the latency is greater than the value in this file (in microseconds). By echoing in a time into this file, no latency will be recorded unless it is greater than the time in this file. h](h)}(hX;Some of the tracers record the max latency. For example, the maximum time that interrupts are disabled. The maximum time is saved in this file. The max trace will also be stored, and displayed by "trace". A new max trace will only be recorded if the latency is greater than the value in this file (in microseconds).h]hX?Some of the tracers record the max latency. For example, the maximum time that interrupts are disabled. The maximum time is saved in this file. The max trace will also be stored, and displayed by “trace”. A new max trace will only be recorded if the latency is greater than the value in this file (in microseconds).}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubh)}(hqBy echoing in a time into this file, no latency will be recorded unless it is greater than the time in this file.h]hqBy echoing in a time into this file, no latency will be recorded unless it is greater than the time in this file.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubeh}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(htracing_thresh:h]htracing_thresh:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(hSome latency tracers will record a trace whenever the latency is greater than the number in this file. Only active when the file contains a number greater than 0. (in microseconds) h]h)}(hSome latency tracers will record a trace whenever the latency is greater than the number in this file. Only active when the file contains a number greater than 0. (in microseconds)h]hSome latency tracers will record a trace whenever the latency is greater than the number in this file. Only active when the file contains a number greater than 0. (in microseconds)}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(hbuffer_percent:h]hbuffer_percent:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(hXThis is the watermark for how much the ring buffer needs to be filled before a waiter is woken up. That is, if an application calls a blocking read syscall on one of the per_cpu trace_pipe_raw files, it will block until the given amount of data specified by buffer_percent is in the ring buffer before it wakes the reader up. This also controls how the splice system calls are blocked on this file:: 0 - means to wake up as soon as there is any data in the ring buffer. 50 - means to wake up when roughly half of the ring buffer sub-buffers are full. 100 - means to block until the ring buffer is totally full and is about to start overwriting the older data. h](h)}(hXThis is the watermark for how much the ring buffer needs to be filled before a waiter is woken up. That is, if an application calls a blocking read syscall on one of the per_cpu trace_pipe_raw files, it will block until the given amount of data specified by buffer_percent is in the ring buffer before it wakes the reader up. This also controls how the splice system calls are blocked on this file::h]hXThis is the watermark for how much the ring buffer needs to be filled before a waiter is woken up. That is, if an application calls a blocking read syscall on one of the per_cpu trace_pipe_raw files, it will block until the given amount of data specified by buffer_percent is in the ring buffer before it wakes the reader up. This also controls how the splice system calls are blocked on this file:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubj)}(hX0 - means to wake up as soon as there is any data in the ring buffer. 50 - means to wake up when roughly half of the ring buffer sub-buffers are full. 100 - means to block until the ring buffer is totally full and is about to start overwriting the older data.h]hX0 - means to wake up as soon as there is any data in the ring buffer. 50 - means to wake up when roughly half of the ring buffer sub-buffers are full. 100 - means to block until the ring buffer is totally full and is about to start overwriting the older data.}hj sbah}(h]h ]h"]h$]h&]jjuh1jhhhKhjubeh}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(hbuffer_size_kb:h]hbuffer_size_kb:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(hXThis sets or displays the number of kilobytes each CPU buffer holds. By default, the trace buffers are the same size for each CPU. The displayed number is the size of the CPU buffer and not total size of all buffers. The trace buffers are allocated in pages (blocks of memory that the kernel uses for allocation, usually 4 KB in size). A few extra pages may be allocated to accommodate buffer management meta-data. If the last page allocated has room for more bytes than requested, the rest of the page will be used, making the actual allocation bigger than requested or shown. ( Note, the size may not be a multiple of the page size due to buffer management meta-data. ) Buffer sizes for individual CPUs may vary (see "per_cpu/cpu0/buffer_size_kb" below), and if they do this file will show "X". h](h)}(hXThis sets or displays the number of kilobytes each CPU buffer holds. By default, the trace buffers are the same size for each CPU. The displayed number is the size of the CPU buffer and not total size of all buffers. The trace buffers are allocated in pages (blocks of memory that the kernel uses for allocation, usually 4 KB in size). A few extra pages may be allocated to accommodate buffer management meta-data. If the last page allocated has room for more bytes than requested, the rest of the page will be used, making the actual allocation bigger than requested or shown. ( Note, the size may not be a multiple of the page size due to buffer management meta-data. )h]hXThis sets or displays the number of kilobytes each CPU buffer holds. By default, the trace buffers are the same size for each CPU. The displayed number is the size of the CPU buffer and not total size of all buffers. The trace buffers are allocated in pages (blocks of memory that the kernel uses for allocation, usually 4 KB in size). A few extra pages may be allocated to accommodate buffer management meta-data. If the last page allocated has room for more bytes than requested, the rest of the page will be used, making the actual allocation bigger than requested or shown. ( Note, the size may not be a multiple of the page size due to buffer management meta-data. )}(hj1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj-ubh)}(h|Buffer sizes for individual CPUs may vary (see "per_cpu/cpu0/buffer_size_kb" below), and if they do this file will show "X".h]hBuffer sizes for individual CPUs may vary (see “per_cpu/cpu0/buffer_size_kb” below), and if they do this file will show “X”.}(hj?hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj-ubeh}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(hbuffer_total_size_kb:h]hbuffer_total_size_kb:}(hjShhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(h@This displays the total combined size of all the trace buffers. h]h)}(h?This displays the total combined size of all the trace buffers.h]h?This displays the total combined size of all the trace buffers.}(hjehhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjaubah}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(hbuffer_subbuf_size_kb:h]hbuffer_subbuf_size_kb:}(hjyhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(hXThis sets or displays the sub buffer size. The ring buffer is broken up into several same size "sub buffers". An event can not be bigger than the size of the sub buffer. Normally, the sub buffer is the size of the architecture's page (4K on x86). The sub buffer also contains meta data at the start which also limits the size of an event. That means when the sub buffer is a page size, no event can be larger than the page size minus the sub buffer meta data. Note, the buffer_subbuf_size_kb is a way for the user to specify the minimum size of the subbuffer. The kernel may make it bigger due to the implementation details, or simply fail the operation if the kernel can not handle the request. Changing the sub buffer size allows for events to be larger than the page size. Note: When changing the sub-buffer size, tracing is stopped and any data in the ring buffer and the snapshot buffer will be discarded. h](h)}(hXThis sets or displays the sub buffer size. The ring buffer is broken up into several same size "sub buffers". An event can not be bigger than the size of the sub buffer. Normally, the sub buffer is the size of the architecture's page (4K on x86). The sub buffer also contains meta data at the start which also limits the size of an event. That means when the sub buffer is a page size, no event can be larger than the page size minus the sub buffer meta data.h]hXThis sets or displays the sub buffer size. The ring buffer is broken up into several same size “sub buffers”. An event can not be bigger than the size of the sub buffer. Normally, the sub buffer is the size of the architecture’s page (4K on x86). The sub buffer also contains meta data at the start which also limits the size of an event. That means when the sub buffer is a page size, no event can be larger than the page size minus the sub buffer meta data.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubh)}(hNote, the buffer_subbuf_size_kb is a way for the user to specify the minimum size of the subbuffer. The kernel may make it bigger due to the implementation details, or simply fail the operation if the kernel can not handle the request.h]hNote, the buffer_subbuf_size_kb is a way for the user to specify the minimum size of the subbuffer. The kernel may make it bigger due to the implementation details, or simply fail the operation if the kernel can not handle the request.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubh)}(hOChanging the sub buffer size allows for events to be larger than the page size.h]hOChanging the sub buffer size allows for events to be larger than the page size.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubh)}(hNote: When changing the sub-buffer size, tracing is stopped and any data in the ring buffer and the snapshot buffer will be discarded.h]hNote: When changing the sub-buffer size, tracing is stopped and any data in the ring buffer and the snapshot buffer will be discarded.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubeh}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(h free_buffer:h]h free_buffer:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(hXIf a process is performing tracing, and the ring buffer should be shrunk "freed" when the process is finished, even if it were to be killed by a signal, this file can be used for that purpose. On close of this file, the ring buffer will be resized to its minimum size. Having a process that is tracing also open this file, when the process exits its file descriptor for this file will be closed, and in doing so, the ring buffer will be "freed". It may also stop tracing if disable_on_free option is set. h](h)}(hXIf a process is performing tracing, and the ring buffer should be shrunk "freed" when the process is finished, even if it were to be killed by a signal, this file can be used for that purpose. On close of this file, the ring buffer will be resized to its minimum size. Having a process that is tracing also open this file, when the process exits its file descriptor for this file will be closed, and in doing so, the ring buffer will be "freed".h]hXIf a process is performing tracing, and the ring buffer should be shrunk “freed” when the process is finished, even if it were to be killed by a signal, this file can be used for that purpose. On close of this file, the ring buffer will be resized to its minimum size. Having a process that is tracing also open this file, when the process exits its file descriptor for this file will be closed, and in doing so, the ring buffer will be “freed”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubh)}(h:It may also stop tracing if disable_on_free option is set.h]h:It may also stop tracing if disable_on_free option is set.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubeh}(h]h ]h"]h$]h&]uh1jShhhKhjgubh)}(htracing_cpumask:h]htracing_cpumask:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjgubjT)}(hrThis is a mask that lets the user only trace on specified CPUs. The format is a hex string representing the CPUs. h]h)}(hqThis is a mask that lets the user only trace on specified CPUs. The format is a hex string representing the CPUs.h]hqThis is a mask that lets the user only trace on specified CPUs. The format is a hex string representing the CPUs.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hset_ftrace_filter:h]hset_ftrace_filter:}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hX;When dynamic ftrace is configured in (see the section below "dynamic ftrace"), the code is dynamically modified (code text rewrite) to disable calling of the function profiler (mcount). This lets tracing be configured in with practically no overhead in performance. This also has a side effect of enabling or disabling specific functions to be traced. Echoing names of functions into this file will limit the trace to only those functions. This influences the tracers "function" and "function_graph" and thus also function profiling (see "function_profile_enabled"). The functions listed in "available_filter_functions" are what can be written into this file. This interface also allows for commands to be used. See the "Filter commands" section for more details. As a speed up, since processing strings can be quite expensive and requires a check of all functions registered to tracing, instead an index can be written into this file. A number (starting with "1") written will instead select the same corresponding at the line position of the "available_filter_functions" file. h](h)}(hX7When dynamic ftrace is configured in (see the section below "dynamic ftrace"), the code is dynamically modified (code text rewrite) to disable calling of the function profiler (mcount). This lets tracing be configured in with practically no overhead in performance. This also has a side effect of enabling or disabling specific functions to be traced. Echoing names of functions into this file will limit the trace to only those functions. This influences the tracers "function" and "function_graph" and thus also function profiling (see "function_profile_enabled").h]hXGWhen dynamic ftrace is configured in (see the section below “dynamic ftrace”), the code is dynamically modified (code text rewrite) to disable calling of the function profiler (mcount). This lets tracing be configured in with practically no overhead in performance. This also has a side effect of enabling or disabling specific functions to be traced. Echoing names of functions into this file will limit the trace to only those functions. This influences the tracers “function” and “function_graph” and thus also function profiling (see “function_profile_enabled”).}(hj5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj1ubh)}(h\The functions listed in "available_filter_functions" are what can be written into this file.h]h`The functions listed in “available_filter_functions” are what can be written into this file.}(hjChhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj1ubh)}(hgThis interface also allows for commands to be used. See the "Filter commands" section for more details.h]hkThis interface also allows for commands to be used. See the “Filter commands” section for more details.}(hjQhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj1ubh)}(hX:As a speed up, since processing strings can be quite expensive and requires a check of all functions registered to tracing, instead an index can be written into this file. A number (starting with "1") written will instead select the same corresponding at the line position of the "available_filter_functions" file.h]hXBAs a speed up, since processing strings can be quite expensive and requires a check of all functions registered to tracing, instead an index can be written into this file. A number (starting with “1”) written will instead select the same corresponding at the line position of the “available_filter_functions” file.}(hj_hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj1ubeh}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hset_ftrace_notrace:h]hset_ftrace_notrace:}(hjshhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hThis has an effect opposite to that of set_ftrace_filter. Any function that is added here will not be traced. If a function exists in both set_ftrace_filter and set_ftrace_notrace, the function will _not_ be traced. h]h)}(hThis has an effect opposite to that of set_ftrace_filter. Any function that is added here will not be traced. If a function exists in both set_ftrace_filter and set_ftrace_notrace, the function will _not_ be traced.h]hThis has an effect opposite to that of set_ftrace_filter. Any function that is added here will not be traced. If a function exists in both set_ftrace_filter and set_ftrace_notrace, the function will _not_ be traced.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hset_ftrace_pid:h]hset_ftrace_pid:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM#hjgubjT)}(hX{Have the function tracer only trace the threads whose PID are listed in this file. If the "function-fork" option is set, then when a task whose PID is listed in this file forks, the child's PID will automatically be added to this file, and the child will be traced by the function tracer as well. This option will also cause PIDs of tasks that exit to be removed from the file. h](h)}(hRHave the function tracer only trace the threads whose PID are listed in this file.h]hRHave the function tracer only trace the threads whose PID are listed in this file.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM%hjubh)}(hX&If the "function-fork" option is set, then when a task whose PID is listed in this file forks, the child's PID will automatically be added to this file, and the child will be traced by the function tracer as well. This option will also cause PIDs of tasks that exit to be removed from the file.h]hX,If the “function-fork” option is set, then when a task whose PID is listed in this file forks, the child’s PID will automatically be added to this file, and the child will be traced by the function tracer as well. This option will also cause PIDs of tasks that exit to be removed from the file.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM(hjubeh}(h]h ]h"]h$]h&]uh1jShhhM%hjgubh)}(hset_ftrace_notrace_pid:h]hset_ftrace_notrace_pid:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM.hjgubjT)}(hXHave the function tracer ignore threads whose PID are listed in this file. If the "function-fork" option is set, then when a task whose PID is listed in this file forks, the child's PID will automatically be added to this file, and the child will not be traced by the function tracer as well. This option will also cause PIDs of tasks that exit to be removed from the file. If a PID is in both this file and "set_ftrace_pid", then this file takes precedence, and the thread will not be traced. h](h)}(hJHave the function tracer ignore threads whose PID are listed in this file.h]hJHave the function tracer ignore threads whose PID are listed in this file.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM0hjubh)}(hX*If the "function-fork" option is set, then when a task whose PID is listed in this file forks, the child's PID will automatically be added to this file, and the child will not be traced by the function tracer as well. This option will also cause PIDs of tasks that exit to be removed from the file.h]hX0If the “function-fork” option is set, then when a task whose PID is listed in this file forks, the child’s PID will automatically be added to this file, and the child will not be traced by the function tracer as well. This option will also cause PIDs of tasks that exit to be removed from the file.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM3hjubh)}(hwIf a PID is in both this file and "set_ftrace_pid", then this file takes precedence, and the thread will not be traced.h]h{If a PID is in both this file and “set_ftrace_pid”, then this file takes precedence, and the thread will not be traced.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM9hjubeh}(h]h ]h"]h$]h&]uh1jShhhM0hjgubh)}(hset_event_pid:h]hset_event_pid:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM<hjgubjT)}(hXeHave the events only trace a task with a PID listed in this file. Note, sched_switch and sched_wake_up will also trace events listed in this file. To have the PIDs of children of tasks with their PID in this file added on fork, enable the "event-fork" option. That option will also cause the PIDs of tasks to be removed from this file when the task exits. h](h)}(hHave the events only trace a task with a PID listed in this file. Note, sched_switch and sched_wake_up will also trace events listed in this file.h]hHave the events only trace a task with a PID listed in this file. Note, sched_switch and sched_wake_up will also trace events listed in this file.}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM>hjubh)}(hTo have the PIDs of children of tasks with their PID in this file added on fork, enable the "event-fork" option. That option will also cause the PIDs of tasks to be removed from this file when the task exits.h]hTo have the PIDs of children of tasks with their PID in this file added on fork, enable the “event-fork” option. That option will also cause the PIDs of tasks to be removed from this file when the task exits.}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMBhjubeh}(h]h ]h"]h$]h&]uh1jShhhM>hjgubh)}(hset_event_notrace_pid:h]hset_event_notrace_pid:}(hjChhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMGhjgubjT)}(hXHave the events not trace a task with a PID listed in this file. Note, sched_switch and sched_wakeup will trace threads not listed in this file, even if a thread's PID is in the file if the sched_switch or sched_wakeup events also trace a thread that should be traced. To have the PIDs of children of tasks with their PID in this file added on fork, enable the "event-fork" option. That option will also cause the PIDs of tasks to be removed from this file when the task exits. h](h)}(hX Have the events not trace a task with a PID listed in this file. Note, sched_switch and sched_wakeup will trace threads not listed in this file, even if a thread's PID is in the file if the sched_switch or sched_wakeup events also trace a thread that should be traced.h]hXHave the events not trace a task with a PID listed in this file. Note, sched_switch and sched_wakeup will trace threads not listed in this file, even if a thread’s PID is in the file if the sched_switch or sched_wakeup events also trace a thread that should be traced.}(hjUhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMIhjQubh)}(hTo have the PIDs of children of tasks with their PID in this file added on fork, enable the "event-fork" option. That option will also cause the PIDs of tasks to be removed from this file when the task exits.h]hTo have the PIDs of children of tasks with their PID in this file added on fork, enable the “event-fork” option. That option will also cause the PIDs of tasks to be removed from this file when the task exits.}(hjchhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMOhjQubeh}(h]h ]h"]h$]h&]uh1jShhhMIhjgubh)}(hset_graph_function:h]hset_graph_function:}(hjwhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMThjgubjT)}(hXFunctions listed in this file will cause the function graph tracer to only trace these functions and the functions that they call. (See the section "dynamic ftrace" for more details). Note, set_ftrace_filter and set_ftrace_notrace still affects what functions are being traced. h]h)}(hXFunctions listed in this file will cause the function graph tracer to only trace these functions and the functions that they call. (See the section "dynamic ftrace" for more details). Note, set_ftrace_filter and set_ftrace_notrace still affects what functions are being traced.h]hXFunctions listed in this file will cause the function graph tracer to only trace these functions and the functions that they call. (See the section “dynamic ftrace” for more details). Note, set_ftrace_filter and set_ftrace_notrace still affects what functions are being traced.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMVhjubah}(h]h ]h"]h$]h&]uh1jShhhMVhjgubh)}(hset_graph_notrace:h]hset_graph_notrace:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM\hjgubjT)}(hSimilar to set_graph_function, but will disable function graph tracing when the function is hit until it exits the function. This makes it possible to ignore tracing functions that are called by a specific function. h]h)}(hSimilar to set_graph_function, but will disable function graph tracing when the function is hit until it exits the function. This makes it possible to ignore tracing functions that are called by a specific function.h]hSimilar to set_graph_function, but will disable function graph tracing when the function is hit until it exits the function. This makes it possible to ignore tracing functions that are called by a specific function.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM^hjubah}(h]h ]h"]h$]h&]uh1jShhhM^hjgubh)}(havailable_filter_functions:h]havailable_filter_functions:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMchjgubjT)}(hXThis lists the functions that ftrace has processed and can trace. These are the function names that you can pass to "set_ftrace_filter", "set_ftrace_notrace", "set_graph_function", or "set_graph_notrace". (See the section "dynamic ftrace" below for more details.) h]h)}(hXThis lists the functions that ftrace has processed and can trace. These are the function names that you can pass to "set_ftrace_filter", "set_ftrace_notrace", "set_graph_function", or "set_graph_notrace". (See the section "dynamic ftrace" below for more details.)h]hXThis lists the functions that ftrace has processed and can trace. These are the function names that you can pass to “set_ftrace_filter”, “set_ftrace_notrace”, “set_graph_function”, or “set_graph_notrace”. (See the section “dynamic ftrace” below for more details.)}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMehjubah}(h]h ]h"]h$]h&]uh1jShhhMehjgubh)}(h!available_filter_functions_addrs:h]h!available_filter_functions_addrs:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMkhjgubjT)}(hSimilar to available_filter_functions, but with address displayed for each function. The displayed address is the patch-site address and can differ from /proc/kallsyms address. h]h)}(hSimilar to available_filter_functions, but with address displayed for each function. The displayed address is the patch-site address and can differ from /proc/kallsyms address.h]hSimilar to available_filter_functions, but with address displayed for each function. The displayed address is the patch-site address and can differ from /proc/kallsyms address.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMmhjubah}(h]h ]h"]h$]h&]uh1jShhhMmhjgubh)}(hdyn_ftrace_total_info:h]hdyn_ftrace_total_info:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMqhjgubjT)}(h~This file is for debugging purposes. The number of functions that have been converted to nops and are available to be traced. h]h)}(h}This file is for debugging purposes. The number of functions that have been converted to nops and are available to be traced.h]h}This file is for debugging purposes. The number of functions that have been converted to nops and are available to be traced.}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMshjubah}(h]h ]h"]h$]h&]uh1jShhhMshjgubh)}(henabled_functions:h]henabled_functions:}(hj5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMvhjgubjT)}(hXQThis file is more for debugging ftrace, but can also be useful in seeing if any function has a callback attached to it. Not only does the trace infrastructure use ftrace function trace utility, but other subsystems might too. This file displays all functions that have a callback attached to them as well as the number of callbacks that have been attached. Note, a callback may also call multiple functions which will not be listed in this count. If the callback registered to be traced by a function with the "save regs" attribute (thus even more overhead), a 'R' will be displayed on the same line as the function that is returning registers. If the callback registered to be traced by a function with the "ip modify" attribute (thus the regs->ip can be changed), an 'I' will be displayed on the same line as the function that can be overridden. If a non ftrace trampoline is attached (BPF) a 'D' will be displayed. Note, normal ftrace trampolines can also be attached, but only one "direct" trampoline can be attached to a given function at a time. Some architectures can not call direct trampolines, but instead have the ftrace ops function located above the function entry point. In such cases an 'O' will be displayed. If a function had either the "ip modify" or a "direct" call attached to it in the past, a 'M' will be shown. This flag is never cleared. It is used to know if a function was every modified by the ftrace infrastructure, and can be used for debugging. If the architecture supports it, it will also show what callback is being directly called by the function. If the count is greater than 1 it most likely will be ftrace_ops_list_func(). If the callback of a function jumps to a trampoline that is specific to the callback and which is not the standard trampoline, its address will be printed as well as the function that the trampoline calls. h](h)}(hXThis file is more for debugging ftrace, but can also be useful in seeing if any function has a callback attached to it. Not only does the trace infrastructure use ftrace function trace utility, but other subsystems might too. This file displays all functions that have a callback attached to them as well as the number of callbacks that have been attached. Note, a callback may also call multiple functions which will not be listed in this count.h]hXThis file is more for debugging ftrace, but can also be useful in seeing if any function has a callback attached to it. Not only does the trace infrastructure use ftrace function trace utility, but other subsystems might too. This file displays all functions that have a callback attached to them as well as the number of callbacks that have been attached. Note, a callback may also call multiple functions which will not be listed in this count.}(hjGhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMxhjCubh)}(hIf the callback registered to be traced by a function with the "save regs" attribute (thus even more overhead), a 'R' will be displayed on the same line as the function that is returning registers.h]hIf the callback registered to be traced by a function with the “save regs” attribute (thus even more overhead), a ‘R’ will be displayed on the same line as the function that is returning registers.}(hjUhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjCubh)}(hIf the callback registered to be traced by a function with the "ip modify" attribute (thus the regs->ip can be changed), an 'I' will be displayed on the same line as the function that can be overridden.h]hIf the callback registered to be traced by a function with the “ip modify” attribute (thus the regs->ip can be changed), an ‘I’ will be displayed on the same line as the function that can be overridden.}(hjchhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjCubh)}(hIf a non ftrace trampoline is attached (BPF) a 'D' will be displayed. Note, normal ftrace trampolines can also be attached, but only one "direct" trampoline can be attached to a given function at a time.h]hIf a non ftrace trampoline is attached (BPF) a ‘D’ will be displayed. Note, normal ftrace trampolines can also be attached, but only one “direct” trampoline can be attached to a given function at a time.}(hjqhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjCubh)}(hSome architectures can not call direct trampolines, but instead have the ftrace ops function located above the function entry point. In such cases an 'O' will be displayed.h]hSome architectures can not call direct trampolines, but instead have the ftrace ops function located above the function entry point. In such cases an ‘O’ will be displayed.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjCubh)}(hIf a function had either the "ip modify" or a "direct" call attached to it in the past, a 'M' will be shown. This flag is never cleared. It is used to know if a function was every modified by the ftrace infrastructure, and can be used for debugging.h]hXIf a function had either the “ip modify” or a “direct” call attached to it in the past, a ‘M’ will be shown. This flag is never cleared. It is used to know if a function was every modified by the ftrace infrastructure, and can be used for debugging.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjCubh)}(hIf the architecture supports it, it will also show what callback is being directly called by the function. If the count is greater than 1 it most likely will be ftrace_ops_list_func().h]hIf the architecture supports it, it will also show what callback is being directly called by the function. If the count is greater than 1 it most likely will be ftrace_ops_list_func().}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjCubh)}(hIf the callback of a function jumps to a trampoline that is specific to the callback and which is not the standard trampoline, its address will be printed as well as the function that the trampoline calls.h]hIf the callback of a function jumps to a trampoline that is specific to the callback and which is not the standard trampoline, its address will be printed as well as the function that the trampoline calls.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjCubeh}(h]h ]h"]h$]h&]uh1jShhhMxhjgubh)}(htouched_functions:h]htouched_functions:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hXThis file contains all the functions that ever had a function callback to it via the ftrace infrastructure. It has the same format as enabled_functions but shows all functions that have every been traced. To see any function that has every been modified by "ip modify" or a direct trampoline, one can perform the following command: grep ' M ' /sys/kernel/tracing/touched_functions h](h)}(hThis file contains all the functions that ever had a function callback to it via the ftrace infrastructure. It has the same format as enabled_functions but shows all functions that have every been traced.h]hThis file contains all the functions that ever had a function callback to it via the ftrace infrastructure. It has the same format as enabled_functions but shows all functions that have every been traced.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubh)}(h~To see any function that has every been modified by "ip modify" or a direct trampoline, one can perform the following command:h]hTo see any function that has every been modified by “ip modify” or a direct trampoline, one can perform the following command:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubh)}(h0grep ' M ' /sys/kernel/tracing/touched_functionsh]h4grep ‘ M ‘ /sys/kernel/tracing/touched_functions}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubeh}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hfunction_profile_enabled:h]hfunction_profile_enabled:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hXWhen set it will enable all functions with either the function tracer, or if configured, the function graph tracer. It will keep a histogram of the number of functions that were called and if the function graph tracer was configured, it will also keep track of the time spent in those functions. The histogram content can be displayed in the files: trace_stat/function ( function0, function1, etc). h](h)}(hX\When set it will enable all functions with either the function tracer, or if configured, the function graph tracer. It will keep a histogram of the number of functions that were called and if the function graph tracer was configured, it will also keep track of the time spent in those functions. The histogram content can be displayed in the files:h]hX\When set it will enable all functions with either the function tracer, or if configured, the function graph tracer. It will keep a histogram of the number of functions that were called and if the function graph tracer was configured, it will also keep track of the time spent in those functions. The histogram content can be displayed in the files:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubh)}(h6trace_stat/function ( function0, function1, etc).h]h6trace_stat/function ( function0, function1, etc).}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubeh}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(h trace_stat:h]h trace_stat:}(hj3 hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(h0A directory that holds different tracing stats. h]h)}(h/A directory that holds different tracing stats.h]h/A directory that holds different tracing stats.}(hjE hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjA ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hkprobe_events:h]hkprobe_events:}(hjY hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(h2Enable dynamic trace points. See kprobetrace.rst. h]h)}(h1Enable dynamic trace points. See kprobetrace.rst.h]h1Enable dynamic trace points. See kprobetrace.rst.}(hjk hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjg ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hkprobe_profile:h]hkprobe_profile:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(h1Dynamic trace points stats. See kprobetrace.rst. h]h)}(h0Dynamic trace points stats. See kprobetrace.rst.h]h0Dynamic trace points stats. See kprobetrace.rst.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hmax_graph_depth:h]hmax_graph_depth:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hUsed with the function graph tracer. This is the max depth it will trace into a function. Setting this to a value of one will show only the first kernel function that is called from user space. h]h)}(hUsed with the function graph tracer. This is the max depth it will trace into a function. Setting this to a value of one will show only the first kernel function that is called from user space.h]hUsed with the function graph tracer. This is the max depth it will trace into a function. Setting this to a value of one will show only the first kernel function that is called from user space.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hprintk_formats:h]hprintk_formats:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hXgThis is for tools that read the raw format files. If an event in the ring buffer references a string, only a pointer to the string is recorded into the buffer and not the string itself. This prevents tools from knowing what that string was. This file displays the string and address for the string allowing tools to map the pointers to what the strings were. h]h)}(hXfThis is for tools that read the raw format files. If an event in the ring buffer references a string, only a pointer to the string is recorded into the buffer and not the string itself. This prevents tools from knowing what that string was. This file displays the string and address for the string allowing tools to map the pointers to what the strings were.h]hXfThis is for tools that read the raw format files. If an event in the ring buffer references a string, only a pointer to the string is recorded into the buffer and not the string itself. This prevents tools from knowing what that string was. This file displays the string and address for the string allowing tools to map the pointers to what the strings were.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hsaved_cmdlines:h]hsaved_cmdlines:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hXOnly the pid of the task is recorded in a trace event unless the event specifically saves the task comm as well. Ftrace makes a cache of pid mappings to comms to try to display comms for events. If a pid for a comm is not listed, then "<...>" is displayed in the output. If the option "record-cmd" is set to "0", then comms of tasks will not be saved during recording. By default, it is enabled. h](h)}(hXOnly the pid of the task is recorded in a trace event unless the event specifically saves the task comm as well. Ftrace makes a cache of pid mappings to comms to try to display comms for events. If a pid for a comm is not listed, then "<...>" is displayed in the output.h]hXOnly the pid of the task is recorded in a trace event unless the event specifically saves the task comm as well. Ftrace makes a cache of pid mappings to comms to try to display comms for events. If a pid for a comm is not listed, then “<...>” is displayed in the output.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubh)}(h|If the option "record-cmd" is set to "0", then comms of tasks will not be saved during recording. By default, it is enabled.h]hIf the option “record-cmd” is set to “0”, then comms of tasks will not be saved during recording. By default, it is enabled.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubeh}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hsaved_cmdlines_size:h]hsaved_cmdlines_size:}(hj% hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hBy default, 128 comms are saved (see "saved_cmdlines" above). To increase or decrease the amount of comms that are cached, echo the number of comms to cache into this file. h]h)}(hBy default, 128 comms are saved (see "saved_cmdlines" above). To increase or decrease the amount of comms that are cached, echo the number of comms to cache into this file.h]hBy default, 128 comms are saved (see “saved_cmdlines” above). To increase or decrease the amount of comms that are cached, echo the number of comms to cache into this file.}(hj7 hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj3 ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(h saved_tgids:h]h saved_tgids:}(hjK hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hIf the option "record-tgid" is set, on each scheduling context switch the Task Group ID of a task is saved in a table mapping the PID of the thread to its TGID. By default, the "record-tgid" option is disabled. h]h)}(hIf the option "record-tgid" is set, on each scheduling context switch the Task Group ID of a task is saved in a table mapping the PID of the thread to its TGID. By default, the "record-tgid" option is disabled.h]hIf the option “record-tgid” is set, on each scheduling context switch the Task Group ID of a task is saved in a table mapping the PID of the thread to its TGID. By default, the “record-tgid” option is disabled.}(hj] hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjY ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(h snapshot:h]h snapshot:}(hjq hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hThis displays the "snapshot" buffer and also lets the user take a snapshot of the current running trace. See the "Snapshot" section below for more details. h]h)}(hThis displays the "snapshot" buffer and also lets the user take a snapshot of the current running trace. See the "Snapshot" section below for more details.h]hThis displays the “snapshot” buffer and also lets the user take a snapshot of the current running trace. See the “Snapshot” section below for more details.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hstack_max_size:h]hstack_max_size:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hWhen the stack tracer is activated, this will display the maximum stack size it has encountered. See the "Stack Trace" section below. h]h)}(hWhen the stack tracer is activated, this will display the maximum stack size it has encountered. See the "Stack Trace" section below.h]hWhen the stack tracer is activated, this will display the maximum stack size it has encountered. See the “Stack Trace” section below.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(h stack_trace:h]h stack_trace:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hThis displays the stack back trace of the largest stack that was encountered when the stack tracer is activated. See the "Stack Trace" section below. h]h)}(hThis displays the stack back trace of the largest stack that was encountered when the stack tracer is activated. See the "Stack Trace" section below.h]hThis displays the stack back trace of the largest stack that was encountered when the stack tracer is activated. See the “Stack Trace” section below.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hstack_trace_filter:h]hstack_trace_filter:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(haThis is similar to "set_ftrace_filter" but it limits what functions the stack tracer will check. h]h)}(h`This is similar to "set_ftrace_filter" but it limits what functions the stack tracer will check.h]hdThis is similar to “set_ftrace_filter” but it limits what functions the stack tracer will check.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(h trace_clock:h]h trace_clock:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hX Whenever an event is recorded into the ring buffer, a "timestamp" is added. This stamp comes from a specified clock. By default, ftrace uses the "local" clock. This clock is very fast and strictly per cpu, but on some systems it may not be monotonic with respect to other CPUs. In other words, the local clocks may not be in sync with local clocks on other CPUs. Usual clocks for tracing:: # cat trace_clock [local] global counter x86-tsc The clock with the square brackets around it is the one in effect. local: Default clock, but may not be in sync across CPUs global: This clock is in sync with all CPUs but may be a bit slower than the local clock. counter: This is not a clock at all, but literally an atomic counter. It counts up one by one, but is in sync with all CPUs. This is useful when you need to know exactly the order events occurred with respect to each other on different CPUs. uptime: This uses the jiffies counter and the time stamp is relative to the time since boot up. perf: This makes ftrace use the same clock that perf uses. Eventually perf will be able to read ftrace buffers and this will help out in interleaving the data. x86-tsc: Architectures may define their own clocks. For example, x86 uses its own TSC cycle clock here. ppc-tb: This uses the powerpc timebase register value. This is in sync across CPUs and can also be used to correlate events across hypervisor/guest if tb_offset is known. mono: This uses the fast monotonic clock (CLOCK_MONOTONIC) which is monotonic and is subject to NTP rate adjustments. mono_raw: This is the raw monotonic clock (CLOCK_MONOTONIC_RAW) which is monotonic but is not subject to any rate adjustments and ticks at the same rate as the hardware clocksource. boot: This is the boot clock (CLOCK_BOOTTIME) and is based on the fast monotonic clock, but also accounts for time spent in suspend. Since the clock access is designed for use in tracing in the suspend path, some side effects are possible if clock is accessed after the suspend time is accounted before the fast mono clock is updated. In this case, the clock update appears to happen slightly sooner than it normally would have. Also on 32-bit systems, it's possible that the 64-bit boot offset sees a partial update. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_boot_fast_ns() function for more information. tai: This is the tai clock (CLOCK_TAI) and is derived from the wall- clock time. However, this clock does not experience discontinuities and backwards jumps caused by NTP inserting leap seconds. Since the clock access is designed for use in tracing, side effects are possible. The clock access may yield wrong readouts in case the internal TAI offset is updated e.g., caused by setting the system time or using adjtimex() with an offset. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_tai_fast_ns() function for more information. To set a clock, simply echo the clock name into this file:: # echo global > trace_clock Setting a clock clears the ring buffer content as well as the "snapshot" buffer. h](h)}(hXjWhenever an event is recorded into the ring buffer, a "timestamp" is added. This stamp comes from a specified clock. By default, ftrace uses the "local" clock. This clock is very fast and strictly per cpu, but on some systems it may not be monotonic with respect to other CPUs. In other words, the local clocks may not be in sync with local clocks on other CPUs.h]hXrWhenever an event is recorded into the ring buffer, a “timestamp” is added. This stamp comes from a specified clock. By default, ftrace uses the “local” clock. This clock is very fast and strictly per cpu, but on some systems it may not be monotonic with respect to other CPUs. In other words, the local clocks may not be in sync with local clocks on other CPUs.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubh)}(hUsual clocks for tracing::h]hUsual clocks for tracing:}(hj) hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj ubj)}(h0# cat trace_clock [local] global counter x86-tsch]h0# cat trace_clock [local] global counter x86-tsc}hj7 sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj ubh)}(hBThe clock with the square brackets around it is the one in effect.h]hBThe clock with the square brackets around it is the one in effect.}(hjE hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubhdefinition_list)}(hhh](hdefinition_list_item)}(h9local: Default clock, but may not be in sync across CPUs h](hterm)}(hlocal:h]hlocal:}(hj` hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjZ ubh definition)}(hhh]h)}(h1Default clock, but may not be in sync across CPUsh]h1Default clock, but may not be in sync across CPUs}(hjs hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjp ubah}(h]h ]h"]h$]h&]uh1jn hjZ ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjU ubjY )}(hZglobal: This clock is in sync with all CPUs but may be a bit slower than the local clock. h](j_ )}(hglobal:h]hglobal:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj ubjo )}(hhh]h)}(hQThis clock is in sync with all CPUs but may be a bit slower than the local clock.h]hQThis clock is in sync with all CPUs but may be a bit slower than the local clock.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubah}(h]h ]h"]h$]h&]uh1jn hj ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjU ubjY )}(hcounter: This is not a clock at all, but literally an atomic counter. It counts up one by one, but is in sync with all CPUs. This is useful when you need to know exactly the order events occurred with respect to each other on different CPUs. h](j_ )}(hcounter:h]hcounter:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM hj ubjo )}(hhh]h)}(hThis is not a clock at all, but literally an atomic counter. It counts up one by one, but is in sync with all CPUs. This is useful when you need to know exactly the order events occurred with respect to each other on different CPUs.h]hThis is not a clock at all, but literally an atomic counter. It counts up one by one, but is in sync with all CPUs. This is useful when you need to know exactly the order events occurred with respect to each other on different CPUs.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubah}(h]h ]h"]h$]h&]uh1jn hj ubeh}(h]h ]h"]h$]h&]uh1jX hhhM hjU ubjY )}(h`uptime: This uses the jiffies counter and the time stamp is relative to the time since boot up. h](j_ )}(huptime:h]huptime:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM$hj ubjo )}(hhh]h)}(hWThis uses the jiffies counter and the time stamp is relative to the time since boot up.h]hWThis uses the jiffies counter and the time stamp is relative to the time since boot up.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM#hj ubah}(h]h ]h"]h$]h&]uh1jn hj ubeh}(h]h ]h"]h$]h&]uh1jX hhhM$hjU ubjY )}(hperf: This makes ftrace use the same clock that perf uses. Eventually perf will be able to read ftrace buffers and this will help out in interleaving the data. h](j_ )}(hperf:h]hperf:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM)hj ubjo )}(hhh]h)}(hThis makes ftrace use the same clock that perf uses. Eventually perf will be able to read ftrace buffers and this will help out in interleaving the data.h]hThis makes ftrace use the same clock that perf uses. Eventually perf will be able to read ftrace buffers and this will help out in interleaving the data.}(hj/ hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM'hj, ubah}(h]h ]h"]h$]h&]uh1jn hj ubeh}(h]h ]h"]h$]h&]uh1jX hhhM)hjU ubjY )}(hhx86-tsc: Architectures may define their own clocks. For example, x86 uses its own TSC cycle clock here. h](j_ )}(hx86-tsc:h]hx86-tsc:}(hjM hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM-hjI ubjo )}(hhh]h)}(h^Architectures may define their own clocks. For example, x86 uses its own TSC cycle clock here.h]h^Architectures may define their own clocks. For example, x86 uses its own TSC cycle clock here.}(hj^ hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM,hj[ ubah}(h]h ]h"]h$]h&]uh1jn hjI ubeh}(h]h ]h"]h$]h&]uh1jX hhhM-hjU ubjY )}(hppc-tb: This uses the powerpc timebase register value. This is in sync across CPUs and can also be used to correlate events across hypervisor/guest if tb_offset is known. h](j_ )}(hppc-tb:h]hppc-tb:}(hj| hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM3hjx ubjo )}(hhh]h)}(hThis uses the powerpc timebase register value. This is in sync across CPUs and can also be used to correlate events across hypervisor/guest if tb_offset is known.h]hThis uses the powerpc timebase register value. This is in sync across CPUs and can also be used to correlate events across hypervisor/guest if tb_offset is known.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM0hj ubah}(h]h ]h"]h$]h&]uh1jn hjx ubeh}(h]h ]h"]h$]h&]uh1jX hhhM3hjU ubjY )}(hvmono: This uses the fast monotonic clock (CLOCK_MONOTONIC) which is monotonic and is subject to NTP rate adjustments. h](j_ )}(hmono:h]hmono:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM7hj ubjo )}(hhh]h)}(hoThis uses the fast monotonic clock (CLOCK_MONOTONIC) which is monotonic and is subject to NTP rate adjustments.h]hoThis uses the fast monotonic clock (CLOCK_MONOTONIC) which is monotonic and is subject to NTP rate adjustments.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM6hj ubah}(h]h ]h"]h$]h&]uh1jn hj ubeh}(h]h ]h"]h$]h&]uh1jX hhhM7hjU ubjY )}(hmono_raw: This is the raw monotonic clock (CLOCK_MONOTONIC_RAW) which is monotonic but is not subject to any rate adjustments and ticks at the same rate as the hardware clocksource. h](j_ )}(h mono_raw:h]h mono_raw:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM<hj ubjo )}(hhh]h)}(hThis is the raw monotonic clock (CLOCK_MONOTONIC_RAW) which is monotonic but is not subject to any rate adjustments and ticks at the same rate as the hardware clocksource.h]hThis is the raw monotonic clock (CLOCK_MONOTONIC_RAW) which is monotonic but is not subject to any rate adjustments and ticks at the same rate as the hardware clocksource.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM:hj ubah}(h]h ]h"]h$]h&]uh1jn hj ubeh}(h]h ]h"]h$]h&]uh1jX hhhM<hjU ubjY )}(hXboot: This is the boot clock (CLOCK_BOOTTIME) and is based on the fast monotonic clock, but also accounts for time spent in suspend. Since the clock access is designed for use in tracing in the suspend path, some side effects are possible if clock is accessed after the suspend time is accounted before the fast mono clock is updated. In this case, the clock update appears to happen slightly sooner than it normally would have. Also on 32-bit systems, it's possible that the 64-bit boot offset sees a partial update. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_boot_fast_ns() function for more information. h](j_ )}(hboot:h]hboot:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMIhj ubjo )}(hhh]h)}(hXThis is the boot clock (CLOCK_BOOTTIME) and is based on the fast monotonic clock, but also accounts for time spent in suspend. Since the clock access is designed for use in tracing in the suspend path, some side effects are possible if clock is accessed after the suspend time is accounted before the fast mono clock is updated. In this case, the clock update appears to happen slightly sooner than it normally would have. Also on 32-bit systems, it's possible that the 64-bit boot offset sees a partial update. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_boot_fast_ns() function for more information.h]hXThis is the boot clock (CLOCK_BOOTTIME) and is based on the fast monotonic clock, but also accounts for time spent in suspend. Since the clock access is designed for use in tracing in the suspend path, some side effects are possible if clock is accessed after the suspend time is accounted before the fast mono clock is updated. In this case, the clock update appears to happen slightly sooner than it normally would have. Also on 32-bit systems, it’s possible that the 64-bit boot offset sees a partial update. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_boot_fast_ns() function for more information.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM?hj ubah}(h]h ]h"]h$]h&]uh1jn hj ubeh}(h]h ]h"]h$]h&]uh1jX hhhMIhjU ubjY )}(hXKtai: This is the tai clock (CLOCK_TAI) and is derived from the wall- clock time. However, this clock does not experience discontinuities and backwards jumps caused by NTP inserting leap seconds. Since the clock access is designed for use in tracing, side effects are possible. The clock access may yield wrong readouts in case the internal TAI offset is updated e.g., caused by setting the system time or using adjtimex() with an offset. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_tai_fast_ns() function for more information. h](j_ )}(htai:h]htai:}(hj8 hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMUhj4 ubjo )}(hhh]h)}(hXEThis is the tai clock (CLOCK_TAI) and is derived from the wall- clock time. However, this clock does not experience discontinuities and backwards jumps caused by NTP inserting leap seconds. Since the clock access is designed for use in tracing, side effects are possible. The clock access may yield wrong readouts in case the internal TAI offset is updated e.g., caused by setting the system time or using adjtimex() with an offset. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_tai_fast_ns() function for more information.h]hXEThis is the tai clock (CLOCK_TAI) and is derived from the wall- clock time. However, this clock does not experience discontinuities and backwards jumps caused by NTP inserting leap seconds. Since the clock access is designed for use in tracing, side effects are possible. The clock access may yield wrong readouts in case the internal TAI offset is updated e.g., caused by setting the system time or using adjtimex() with an offset. These effects are rare and post processing should be able to handle them. See comments in the ktime_get_tai_fast_ns() function for more information.}(hjI hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMLhjF ubah}(h]h ]h"]h$]h&]uh1jn hj4 ubeh}(h]h ]h"]h$]h&]uh1jX hhhMUhjU ubeh}(h]h ]h"]h$]h&]uh1jS hj ubh)}(h;To set a clock, simply echo the clock name into this file::h]h:To set a clock, simply echo the clock name into this file:}(hji hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMWhj ubj)}(h# echo global > trace_clockh]h# echo global > trace_clock}hjw sbah}(h]h ]h"]h$]h&]jjuh1jhhhMYhj ubh)}(hPSetting a clock clears the ring buffer content as well as the "snapshot" buffer.h]hTSetting a clock clears the ring buffer content as well as the “snapshot” buffer.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM[hj ubeh}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(h trace_marker:h]h trace_marker:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM^hjgubjT)}(hXThis is a very useful file for synchronizing user space with events happening in the kernel. Writing strings into this file will be written into the ftrace buffer. It is useful in applications to open this file at the start of the application and just reference the file descriptor for the file:: void trace_write(const char *fmt, ...) { va_list ap; char buf[256]; int n; if (trace_fd < 0) return; va_start(ap, fmt); n = vsnprintf(buf, 256, fmt, ap); va_end(ap); write(trace_fd, buf, n); } start:: trace_fd = open("trace_marker", O_WRONLY); Note: Writing into the trace_marker file can also initiate triggers that are written into /sys/kernel/tracing/events/ftrace/print/trigger See "Event triggers" in Documentation/trace/events.rst and an example in Documentation/trace/histogram.rst (Section 3.) h](h)}(hThis is a very useful file for synchronizing user space with events happening in the kernel. Writing strings into this file will be written into the ftrace buffer.h]hThis is a very useful file for synchronizing user space with events happening in the kernel. Writing strings into this file will be written into the ftrace buffer.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM`hj ubh)}(hIt is useful in applications to open this file at the start of the application and just reference the file descriptor for the file::h]hIt is useful in applications to open this file at the start of the application and just reference the file descriptor for the file:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMdhj ubj)}(hXvoid trace_write(const char *fmt, ...) { va_list ap; char buf[256]; int n; if (trace_fd < 0) return; va_start(ap, fmt); n = vsnprintf(buf, 256, fmt, ap); va_end(ap); write(trace_fd, buf, n); }h]hXvoid trace_write(const char *fmt, ...) { va_list ap; char buf[256]; int n; if (trace_fd < 0) return; va_start(ap, fmt); n = vsnprintf(buf, 256, fmt, ap); va_end(ap); write(trace_fd, buf, n); }}hj sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhhj ubh)}(hstart::h]hstart:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMxhj ubj)}(h*trace_fd = open("trace_marker", O_WRONLY);h]h*trace_fd = open("trace_marker", O_WRONLY);}hj sbah}(h]h ]h"]h$]h&]jjuh1jhhhMzhj ubjT )}(hhh]jY )}(hXNote: Writing into the trace_marker file can also initiate triggers that are written into /sys/kernel/tracing/events/ftrace/print/trigger See "Event triggers" in Documentation/trace/events.rst and an example in Documentation/trace/histogram.rst (Section 3.) h](j_ )}(hCNote: Writing into the trace_marker file can also initiate triggersh]hCNote: Writing into the trace_marker file can also initiate triggers}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj ubjo )}(hhh]h)}(hthat are written into /sys/kernel/tracing/events/ftrace/print/trigger See "Event triggers" in Documentation/trace/events.rst and an example in Documentation/trace/histogram.rst (Section 3.)h]hthat are written into /sys/kernel/tracing/events/ftrace/print/trigger See “Event triggers” in Documentation/trace/events.rst and an example in Documentation/trace/histogram.rst (Section 3.)}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM}hjubah}(h]h ]h"]h$]h&]uh1jn hj ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhj ubah}(h]h ]h"]h$]h&]uh1jS hj ubeh}(h]h ]h"]h$]h&]uh1jShhhM`hjgubh)}(htrace_marker_raw:h]htrace_marker_raw:}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hThis is similar to trace_marker above, but is meant for binary data to be written to it, where a tool can be used to parse the data from trace_pipe_raw. h]h)}(hThis is similar to trace_marker above, but is meant for binary data to be written to it, where a tool can be used to parse the data from trace_pipe_raw.h]hThis is similar to trace_marker above, but is meant for binary data to be written to it, where a tool can be used to parse the data from trace_pipe_raw.}(hjAhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj=ubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(huprobe_events:h]huprobe_events:}(hjUhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(h:Add dynamic tracepoints in programs. See uprobetracer.rst h]h)}(h9Add dynamic tracepoints in programs. See uprobetracer.rsth]h9Add dynamic tracepoints in programs. See uprobetracer.rst}(hjghhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjcubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(huprobe_profile:h]huprobe_profile:}(hj{hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(h'Uprobe statistics. See uprobetrace.txt h]h)}(h&Uprobe statistics. See uprobetrace.txth]h&Uprobe statistics. See uprobetrace.txt}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(h instances:h]h instances:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hThis is a way to make multiple trace buffers where different events can be recorded in different buffers. See "Instances" section below. h]h)}(hThis is a way to make multiple trace buffers where different events can be recorded in different buffers. See "Instances" section below.h]hThis is a way to make multiple trace buffers where different events can be recorded in different buffers. See “Instances” section below.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(hevents:h]hevents:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hXdThis is the trace event directory. It holds event tracepoints (also known as static tracepoints) that have been compiled into the kernel. It shows what event tracepoints exist and how they are grouped by system. There are "enable" files at various levels that can enable the tracepoints when a "1" is written to them. See events.rst for more information. h](h)}(hX=This is the trace event directory. It holds event tracepoints (also known as static tracepoints) that have been compiled into the kernel. It shows what event tracepoints exist and how they are grouped by system. There are "enable" files at various levels that can enable the tracepoints when a "1" is written to them.h]hXEThis is the trace event directory. It holds event tracepoints (also known as static tracepoints) that have been compiled into the kernel. It shows what event tracepoints exist and how they are grouped by system. There are “enable” files at various levels that can enable the tracepoints when a “1” is written to them.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubh)}(h$See events.rst for more information.h]h$See events.rst for more information.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubeh}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(h set_event:h]h set_event:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hfBy echoing in the event into this file, will enable that event. See events.rst for more information. h](h)}(h?By echoing in the event into this file, will enable that event.h]h?By echoing in the event into this file, will enable that event.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubh)}(h$See events.rst for more information.h]h$See events.rst for more information.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubeh}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(havailable_events:h]havailable_events:}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hWA list of events that can be enabled in tracing. See events.rst for more information. h](h)}(h0A list of events that can be enabled in tracing.h]h0A list of events that can be enabled in tracing.}(hjAhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj=ubh)}(h$See events.rst for more information.h]h$See events.rst for more information.}(hjOhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj=ubeh}(h]h ]h"]h$]h&]uh1jShhhMhjgubh)}(htimestamp_mode:h]htimestamp_mode:}(hjchhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjgubjT)}(hXCertain tracers may change the timestamp mode used when logging trace events into the event buffer. Events with different modes can coexist within a buffer but the mode in effect when an event is logged determines which timestamp mode is used for that event. The default timestamp mode is 'delta'. Usual timestamp modes for tracing: # cat timestamp_mode [delta] absolute The timestamp mode with the square brackets around it is the one in effect. delta: Default timestamp mode - timestamp is a delta against a per-buffer timestamp. absolute: The timestamp is a full timestamp, not a delta against some other value. As such it takes up more space and is less efficient. h](h)}(hX+Certain tracers may change the timestamp mode used when logging trace events into the event buffer. Events with different modes can coexist within a buffer but the mode in effect when an event is logged determines which timestamp mode is used for that event. The default timestamp mode is 'delta'.h]hX/Certain tracers may change the timestamp mode used when logging trace events into the event buffer. Events with different modes can coexist within a buffer but the mode in effect when an event is logged determines which timestamp mode is used for that event. The default timestamp mode is ‘delta’.}(hjuhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjqubh)}(h"Usual timestamp modes for tracing:h]h"Usual timestamp modes for tracing:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjqubjT)}(hXi# cat timestamp_mode [delta] absolute The timestamp mode with the square brackets around it is the one in effect. delta: Default timestamp mode - timestamp is a delta against a per-buffer timestamp. absolute: The timestamp is a full timestamp, not a delta against some other value. As such it takes up more space and is less efficient. h](h)}(h%# cat timestamp_mode [delta] absoluteh]h%# cat timestamp_mode [delta] absolute}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubh)}(hKThe timestamp mode with the square brackets around it is the one in effect.h]hKThe timestamp mode with the square brackets around it is the one in effect.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubjT )}(hhh](jY )}(hUdelta: Default timestamp mode - timestamp is a delta against a per-buffer timestamp. h](j_ )}(hhj<ubjT)}(hX*Traces the areas that disable interrupts and saves the trace with the longest max latency. See tracing_max_latency. When a new max is recorded, it replaces the old trace. It is best to view this trace with the latency-format option enabled, which happens automatically when the tracer is selected. h]h)}(hX)Traces the areas that disable interrupts and saves the trace with the longest max latency. See tracing_max_latency. When a new max is recorded, it replaces the old trace. It is best to view this trace with the latency-format option enabled, which happens automatically when the tracer is selected.h]hX)Traces the areas that disable interrupts and saves the trace with the longest max latency. See tracing_max_latency. When a new max is recorded, it replaces the old trace. It is best to view this trace with the latency-format option enabled, which happens automatically when the tracer is selected.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM@hjubah}(h]h ]h"]h$]h&]uh1jShhhM@hj<ubh)}(h "preemptoff"h]h“preemptoff”}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMGhj<ubjT)}(h_Similar to irqsoff but traces and records the amount of time for which preemption is disabled. h]h)}(h^Similar to irqsoff but traces and records the amount of time for which preemption is disabled.h]h^Similar to irqsoff but traces and records the amount of time for which preemption is disabled.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMIhjubah}(h]h ]h"]h$]h&]uh1jShhhMIhj<ubh)}(h"preemptirqsoff"h]h“preemptirqsoff”}(hj2hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMLhj<ubjT)}(hySimilar to irqsoff and preemptoff, but traces and records the largest time for which irqs and/or preemption is disabled. h]h)}(hxSimilar to irqsoff and preemptoff, but traces and records the largest time for which irqs and/or preemption is disabled.h]hxSimilar to irqsoff and preemptoff, but traces and records the largest time for which irqs and/or preemption is disabled.}(hjDhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMNhj@ubah}(h]h ]h"]h$]h&]uh1jShhhMNhj<ubh)}(h"wakeup"h]h “wakeup”}(hjXhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMRhj<ubjT)}(hTraces and records the max latency that it takes for the highest priority task to get scheduled after it has been woken up. Traces all tasks as an average developer would expect. h]h)}(hTraces and records the max latency that it takes for the highest priority task to get scheduled after it has been woken up. Traces all tasks as an average developer would expect.h]hTraces and records the max latency that it takes for the highest priority task to get scheduled after it has been woken up. Traces all tasks as an average developer would expect.}(hjjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMThjfubah}(h]h ]h"]h$]h&]uh1jShhhMThj<ubh)}(h "wakeup_rt"h]h“wakeup_rt”}(hj~hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMYhj<ubjT)}(hTraces and records the max latency that it takes for just RT tasks (as the current "wakeup" does). This is useful for those interested in wake up timings of RT tasks. h]h)}(hTraces and records the max latency that it takes for just RT tasks (as the current "wakeup" does). This is useful for those interested in wake up timings of RT tasks.h]hTraces and records the max latency that it takes for just RT tasks (as the current “wakeup” does). This is useful for those interested in wake up timings of RT tasks.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM[hjubah}(h]h ]h"]h$]h&]uh1jShhhM[hj<ubh)}(h "wakeup_dl"h]h“wakeup_dl”}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM_hj<ubjT)}(hTraces and records the max latency that it takes for a SCHED_DEADLINE task to be woken (as the "wakeup" and "wakeup_rt" does). h]h)}(h~Traces and records the max latency that it takes for a SCHED_DEADLINE task to be woken (as the "wakeup" and "wakeup_rt" does).h]hTraces and records the max latency that it takes for a SCHED_DEADLINE task to be woken (as the “wakeup” and “wakeup_rt” does).}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMahjubah}(h]h ]h"]h$]h&]uh1jShhhMahj<ubh)}(h "mmiotrace"h]h“mmiotrace”}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMehj<ubjT)}(hA special tracer that is used to trace binary module. It will trace all the calls that a module makes to the hardware. Everything it writes and reads from the I/O as well. h]h)}(hA special tracer that is used to trace binary module. It will trace all the calls that a module makes to the hardware. Everything it writes and reads from the I/O as well.h]hA special tracer that is used to trace binary module. It will trace all the calls that a module makes to the hardware. Everything it writes and reads from the I/O as well.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMghjubah}(h]h ]h"]h$]h&]uh1jShhhMghj<ubh)}(h"branch"h]h “branch”}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMlhj<ubjT)}(hThis tracer can be configured when tracing likely/unlikely calls within the kernel. It will trace when a likely and unlikely branch is hit and if it was correct in its prediction of being correct. h]h)}(hThis tracer can be configured when tracing likely/unlikely calls within the kernel. It will trace when a likely and unlikely branch is hit and if it was correct in its prediction of being correct.h]hThis tracer can be configured when tracing likely/unlikely calls within the kernel. It will trace when a likely and unlikely branch is hit and if it was correct in its prediction of being correct.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMnhjubah}(h]h ]h"]h$]h&]uh1jShhhMnhj<ubh)}(h"nop"h]h “nop”}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMshj<ubjT)}(hnThis is the "trace nothing" tracer. To remove all tracers from tracing simply echo "nop" into current_tracer. h]h)}(hmThis is the "trace nothing" tracer. To remove all tracers from tracing simply echo "nop" into current_tracer.h]huThis is the “trace nothing” tracer. To remove all tracers from tracing simply echo “nop” into current_tracer.}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMuhj$ubah}(h]h ]h"]h$]h&]uh1jShhhMuhj<ubeh}(h]h ]h"]h$]h&]uh1jShhhM hjhhubeh}(h] the-tracersah ]h"] the tracersah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(hError conditionsh]hError conditions}(hjMhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjJhhhhhMzubjT)}(hXFor most ftrace commands, failure modes are obvious and communicated using standard return codes. For other more involved commands, extended error information may be available via the tracing/error_log file. For the commands that support it, reading the tracing/error_log file after an error will display more detailed information about what went wrong, if information is available. The tracing/error_log file is a circular error log displaying a small number (currently, 8) of ftrace errors for the last (8) failed commands. The extended error information and usage takes the form shown in this example:: # echo xxx > /sys/kernel/tracing/events/sched/sched_wakeup/trigger echo: write error: Invalid argument # cat /sys/kernel/tracing/error_log [ 5348.887237] location: error: Couldn't yyy: zzz Command: xxx ^ [ 7517.023364] location: error: Bad rrr: sss Command: ppp qqq ^ To clear the error log, echo the empty string into it:: # echo > /sys/kernel/tracing/error_log h](h)}(haFor most ftrace commands, failure modes are obvious and communicated using standard return codes.h]haFor most ftrace commands, failure modes are obvious and communicated using standard return codes.}(hj_hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM|hj[ubh)}(hXFor other more involved commands, extended error information may be available via the tracing/error_log file. For the commands that support it, reading the tracing/error_log file after an error will display more detailed information about what went wrong, if information is available. The tracing/error_log file is a circular error log displaying a small number (currently, 8) of ftrace errors for the last (8) failed commands.h]hXFor other more involved commands, extended error information may be available via the tracing/error_log file. For the commands that support it, reading the tracing/error_log file after an error will display more detailed information about what went wrong, if information is available. The tracing/error_log file is a circular error log displaying a small number (currently, 8) of ftrace errors for the last (8) failed commands.}(hjmhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj[ubh)}(hOThe extended error information and usage takes the form shown in this example::h]hNThe extended error information and usage takes the form shown in this example:}(hj{hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj[ubj)}(hX*# echo xxx > /sys/kernel/tracing/events/sched/sched_wakeup/trigger echo: write error: Invalid argument # cat /sys/kernel/tracing/error_log [ 5348.887237] location: error: Couldn't yyy: zzz Command: xxx ^ [ 7517.023364] location: error: Bad rrr: sss Command: ppp qqq ^h]hX*# echo xxx > /sys/kernel/tracing/events/sched/sched_wakeup/trigger echo: write error: Invalid argument # cat /sys/kernel/tracing/error_log [ 5348.887237] location: error: Couldn't yyy: zzz Command: xxx ^ [ 7517.023364] location: error: Bad rrr: sss Command: ppp qqq ^}hjsbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj[ubh)}(h7To clear the error log, echo the empty string into it::h]h6To clear the error log, echo the empty string into it:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj[ubj)}(h&# echo > /sys/kernel/tracing/error_logh]h&# echo > /sys/kernel/tracing/error_log}hjsbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj[ubeh}(h]h ]h"]h$]h&]uh1jShhhM|hjJhhubeh}(h]error-conditionsah ]h"]error conditionsah$]h&]uh1hhhhhhhhMzubh)}(hhh](h)}(hExamples of using the tracerh]hExamples of using the tracer}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubh)}(hHere are typical examples of using the tracers when controlling them only with the tracefs interface (without using any user-land utilities).h]hHere are typical examples of using the tracers when controlling them only with the tracefs interface (without using any user-land utilities).}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubeh}(h]examples-of-using-the-tracerah ]h"]examples of using the tracerah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(hOutput format:h]hOutput format:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubh)}(h=Here is an example of the output format of the file "trace"::h]h@Here is an example of the output format of the file “trace”:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubj)}(hX# tracer: function # # entries-in-buffer/entries-written: 140080/250280 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | bash-1977 [000] .... 17284.993652: sys_close <-system_call_fastpath bash-1977 [000] .... 17284.993653: __close_fd <-sys_close bash-1977 [000] .... 17284.993653: _raw_spin_lock <-__close_fd sshd-1974 [003] .... 17284.993653: __srcu_read_unlock <-fsnotify bash-1977 [000] .... 17284.993654: add_preempt_count <-_raw_spin_lock bash-1977 [000] ...1 17284.993655: _raw_spin_unlock <-__close_fd bash-1977 [000] ...1 17284.993656: sub_preempt_count <-_raw_spin_unlock bash-1977 [000] .... 17284.993657: filp_close <-__close_fd bash-1977 [000] .... 17284.993657: dnotify_flush <-filp_close sshd-1974 [003] .... 17284.993658: sys_select <-system_call_fastpath ....h]hX# tracer: function # # entries-in-buffer/entries-written: 140080/250280 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | bash-1977 [000] .... 17284.993652: sys_close <-system_call_fastpath bash-1977 [000] .... 17284.993653: __close_fd <-sys_close bash-1977 [000] .... 17284.993653: _raw_spin_lock <-__close_fd sshd-1974 [003] .... 17284.993653: __srcu_read_unlock <-fsnotify bash-1977 [000] .... 17284.993654: add_preempt_count <-_raw_spin_lock bash-1977 [000] ...1 17284.993655: _raw_spin_unlock <-__close_fd bash-1977 [000] ...1 17284.993656: sub_preempt_count <-_raw_spin_unlock bash-1977 [000] .... 17284.993657: filp_close <-__close_fd bash-1977 [000] .... 17284.993657: dnotify_flush <-filp_close sshd-1974 [003] .... 17284.993658: sys_select <-system_call_fastpath ....}hjsbah}(h]h ]h"]h$]h&]jjuh1jhhhMhjhhubh)}(hXXA header is printed with the tracer name that is represented by the trace. In this case the tracer is "function". Then it shows the number of events in the buffer as well as the total number of entries that were written. The difference is the number of entries that were lost due to the buffer filling up (250280 - 140080 = 110200 events lost).h]hX\A header is printed with the tracer name that is represented by the trace. In this case the tracer is “function”. Then it shows the number of events in the buffer as well as the total number of entries that were written. The difference is the number of entries that were lost due to the buffer filling up (250280 - 140080 = 110200 events lost).}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubh)}(hX~The header explains the content of the events. Task name "bash", the task PID "1977", the CPU that it was running on "000", the latency format (explained below), the timestamp in . format, the function name that was traced "sys_close" and the parent function that called this function "system_call_fastpath". The timestamp is the time at which the function was entered.h]hXThe header explains the content of the events. Task name “bash”, the task PID “1977”, the CPU that it was running on “000”, the latency format (explained below), the timestamp in . format, the function name that was traced “sys_close” and the parent function that called this function “system_call_fastpath”. The timestamp is the time at which the function was entered.}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubeh}(h] output-formatah ]h"]output format:ah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(hLatency trace formath]hLatency trace format}(hj<hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj9hhhhhMubh)}(hWhen the latency-format option is enabled or when one of the latency tracers is set, the trace file gives somewhat more information to see why a latency happened. Here is a typical trace::h]hWhen the latency-format option is enabled or when one of the latency tracers is set, the trace file gives somewhat more information to see why a latency happened. Here is a typical trace:}(hjJhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj9hhubj)}(hX# tracer: irqsoff # # irqsoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 259 us, #4/4, CPU#2 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: ps-6143 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: __lock_task_sighand # => ended at: _raw_spin_unlock_irqrestore # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / ps-6143 2d... 0us!: trace_hardirqs_off <-__lock_task_sighand ps-6143 2d..1 259us+: trace_hardirqs_on <-_raw_spin_unlock_irqrestore ps-6143 2d..1 263us+: time_hardirqs_on <-_raw_spin_unlock_irqrestore ps-6143 2d..1 306us : => trace_hardirqs_on_caller => trace_hardirqs_on => _raw_spin_unlock_irqrestore => do_task_stat => proc_tgid_stat => proc_single_show => seq_read => vfs_read => sys_read => system_call_fastpathh]hX# tracer: irqsoff # # irqsoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 259 us, #4/4, CPU#2 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: ps-6143 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: __lock_task_sighand # => ended at: _raw_spin_unlock_irqrestore # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / ps-6143 2d... 0us!: trace_hardirqs_off <-__lock_task_sighand ps-6143 2d..1 259us+: trace_hardirqs_on <-_raw_spin_unlock_irqrestore ps-6143 2d..1 263us+: time_hardirqs_on <-_raw_spin_unlock_irqrestore ps-6143 2d..1 306us : => trace_hardirqs_on_caller => trace_hardirqs_on => _raw_spin_unlock_irqrestore => do_task_stat => proc_tgid_stat => proc_single_show => seq_read => vfs_read => sys_read => system_call_fastpath}hjXsbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj9hhubh)}(hXThis shows that the current tracer is "irqsoff" tracing the time for which interrupts were disabled. It gives the trace version (which never changes) and the version of the kernel upon which this was executed on (3.8). Then it displays the max latency in microseconds (259 us). The number of trace entries displayed and the total number (both are four: #4/4). VP, KP, SP, and HP are always zero and are reserved for later use. #P is the number of online CPUs (#P:4).h]hXThis shows that the current tracer is “irqsoff” tracing the time for which interrupts were disabled. It gives the trace version (which never changes) and the version of the kernel upon which this was executed on (3.8). Then it displays the max latency in microseconds (259 us). The number of trace entries displayed and the total number (both are four: #4/4). VP, KP, SP, and HP are always zero and are reserved for later use. #P is the number of online CPUs (#P:4).}(hjfhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj9hhubh)}(hSThe task is the process that was running when the latency occurred. (ps pid: 6143).h]hSThe task is the process that was running when the latency occurred. (ps pid: 6143).}(hjthhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj9hhubh)}(h|The start and stop (the functions in which the interrupts were disabled and enabled respectively) that caused the latencies:h]h|The start and stop (the functions in which the interrupts were disabled and enabled respectively) that caused the latencies:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj9hhubjT)}(h}- __lock_task_sighand is where the interrupts were disabled. - _raw_spin_unlock_irqrestore is where they were enabled again. h]j{)}(hhh](j)}(h:__lock_task_sighand is where the interrupts were disabled.h]h)}(hjh]h:__lock_task_sighand is where the interrupts were disabled.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(h>_raw_spin_unlock_irqrestore is where they were enabled again. h]h)}(h=_raw_spin_unlock_irqrestore is where they were enabled again.h]h=_raw_spin_unlock_irqrestore is where they were enabled again.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]jjuh1jzhhhMhjubah}(h]h ]h"]h$]h&]uh1jShhhMhj9hhubh)}(hYThe next lines after the header are the trace itself. The header explains which is which.h]hYThe next lines after the header are the trace itself. The header explains which is which.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj9hhubjT)}(hXcmd: The name of the process in the trace. pid: The PID of that process. CPU#: The CPU which the process was running on. irqs-off: 'd' interrupts are disabled. '.' otherwise. need-resched: - 'B' all, TIF_NEED_RESCHED, PREEMPT_NEED_RESCHED and TIF_RESCHED_LAZY is set, - 'N' both TIF_NEED_RESCHED and PREEMPT_NEED_RESCHED is set, - 'n' only TIF_NEED_RESCHED is set, - 'p' only PREEMPT_NEED_RESCHED is set, - 'L' both PREEMPT_NEED_RESCHED and TIF_RESCHED_LAZY is set, - 'b' both TIF_NEED_RESCHED and TIF_RESCHED_LAZY is set, - 'l' only TIF_RESCHED_LAZY is set - '.' otherwise. hardirq/softirq: - 'Z' - NMI occurred inside a hardirq - 'z' - NMI is running - 'H' - hard irq occurred inside a softirq. - 'h' - hard irq is running - 's' - soft irq is running - '.' - normal context. preempt-depth: The level of preempt_disabled h](h)}(h*cmd: The name of the process in the trace.h]h*cmd: The name of the process in the trace.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjubh)}(hpid: The PID of that process.h]hpid: The PID of that process.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjubh)}(h/CPU#: The CPU which the process was running on.h]h/CPU#: The CPU which the process was running on.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjubh)}(h5irqs-off: 'd' interrupts are disabled. '.' otherwise.h]h=irqs-off: ‘d’ interrupts are disabled. ‘.’ otherwise.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubjT )}(hhh](jY )}(hXneed-resched: - 'B' all, TIF_NEED_RESCHED, PREEMPT_NEED_RESCHED and TIF_RESCHED_LAZY is set, - 'N' both TIF_NEED_RESCHED and PREEMPT_NEED_RESCHED is set, - 'n' only TIF_NEED_RESCHED is set, - 'p' only PREEMPT_NEED_RESCHED is set, - 'L' both PREEMPT_NEED_RESCHED and TIF_RESCHED_LAZY is set, - 'b' both TIF_NEED_RESCHED and TIF_RESCHED_LAZY is set, - 'l' only TIF_RESCHED_LAZY is set - '.' otherwise. h](j_ )}(h need-resched:h]h need-resched:}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]j{)}(hhh](j)}(hL'B' all, TIF_NEED_RESCHED, PREEMPT_NEED_RESCHED and TIF_RESCHED_LAZY is set,h]h)}(hj9h]hP‘B’ all, TIF_NEED_RESCHED, PREEMPT_NEED_RESCHED and TIF_RESCHED_LAZY is set,}(hj;hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj7ubah}(h]h ]h"]h$]h&]uh1jhj4ubj)}(h:'N' both TIF_NEED_RESCHED and PREEMPT_NEED_RESCHED is set,h]h)}(hjPh]h>‘N’ both TIF_NEED_RESCHED and PREEMPT_NEED_RESCHED is set,}(hjRhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjNubah}(h]h ]h"]h$]h&]uh1jhj4ubj)}(h!'n' only TIF_NEED_RESCHED is set,h]h)}(hjgh]h%‘n’ only TIF_NEED_RESCHED is set,}(hjihhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjeubah}(h]h ]h"]h$]h&]uh1jhj4ubj)}(h%'p' only PREEMPT_NEED_RESCHED is set,h]h)}(hj~h]h)‘p’ only PREEMPT_NEED_RESCHED is set,}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj|ubah}(h]h ]h"]h$]h&]uh1jhj4ubj)}(h:'L' both PREEMPT_NEED_RESCHED and TIF_RESCHED_LAZY is set,h]h)}(hjh]h>‘L’ both PREEMPT_NEED_RESCHED and TIF_RESCHED_LAZY is set,}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jhj4ubj)}(h6'b' both TIF_NEED_RESCHED and TIF_RESCHED_LAZY is set,h]h)}(hjh]h:‘b’ both TIF_NEED_RESCHED and TIF_RESCHED_LAZY is set,}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jhj4ubj)}(h 'l' only TIF_RESCHED_LAZY is seth]h)}(hjh]h$‘l’ only TIF_RESCHED_LAZY is set}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jhj4ubj)}(h'.' otherwise. h]h)}(h'.' otherwise.h]h‘.’ otherwise.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jhj4ubeh}(h]h ]h"]h$]h&]jjuh1jzhhhMhj1ubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hhardirq/softirq: - 'Z' - NMI occurred inside a hardirq - 'z' - NMI is running - 'H' - hard irq occurred inside a softirq. - 'h' - hard irq is running - 's' - soft irq is running - '.' - normal context. h](j_ )}(hhardirq/softirq:h]hhardirq/softirq:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM!hjubjo )}(hhh]j{)}(hhh](j)}(h#'Z' - NMI occurred inside a hardirqh]h)}(hjh]h'‘Z’ - NMI occurred inside a hardirq}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(h'z' - NMI is runningh]h)}(hj3h]h‘z’ - NMI is running}(hj5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj1ubah}(h]h ]h"]h$]h&]uh1jhjubj)}(h)'H' - hard irq occurred inside a softirq.h]h)}(hjJh]h-‘H’ - hard irq occurred inside a softirq.}(hjLhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjHubah}(h]h ]h"]h$]h&]uh1jhjubj)}(h'h' - hard irq is runningh]h)}(hjah]h‘h’ - hard irq is running}(hjchhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj_ubah}(h]h ]h"]h$]h&]uh1jhjubj)}(h's' - soft irq is runningh]h)}(hjxh]h‘s’ - soft irq is running}(hjzhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjvubah}(h]h ]h"]h$]h&]uh1jhjubj)}(h'.' - normal context. h]h)}(h'.' - normal context.h]h‘.’ - normal context.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM!hjubah}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]jjuh1jzhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhM!hjubeh}(h]h ]h"]h$]h&]uh1jS hjubh)}(h,preempt-depth: The level of preempt_disabledh]h,preempt-depth: The level of preempt_disabled}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM#hjubeh}(h]h ]h"]h$]h&]uh1jShhhM hj9hhubh)}(h5The above is mostly meaningful for kernel developers.h]h5The above is mostly meaningful for kernel developers.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM%hj9hhubjT)}(hXtime: When the latency-format option is enabled, the trace file output includes a timestamp relative to the start of the trace. This differs from the output when latency-format is disabled, which includes an absolute timestamp. delay: This is just to help catch your eye a bit better. And needs to be fixed to be only relative to the same CPU. The marks are determined by the difference between this current trace and the next trace. - '$' - greater than 1 second - '@' - greater than 100 millisecond - '*' - greater than 10 millisecond - '#' - greater than 1000 microsecond - '!' - greater than 100 microsecond - '+' - greater than 10 microsecond - ' ' - less than or equal to 10 microsecond. The rest is the same as the 'trace' file. Note, the latency tracers will usually end with a back trace to easily find where the latency occurred. h](jT )}(hhh](jY )}(htime: When the latency-format option is enabled, the trace file output includes a timestamp relative to the start of the trace. This differs from the output when latency-format is disabled, which includes an absolute timestamp. h](j_ )}(htime:h]htime:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM+hjubjo )}(hhh]h)}(hWhen the latency-format option is enabled, the trace file output includes a timestamp relative to the start of the trace. This differs from the output when latency-format is disabled, which includes an absolute timestamp.h]hWhen the latency-format option is enabled, the trace file output includes a timestamp relative to the start of the trace. This differs from the output when latency-format is disabled, which includes an absolute timestamp.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM(hjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhM+hjubjY )}(hXdelay: This is just to help catch your eye a bit better. And needs to be fixed to be only relative to the same CPU. The marks are determined by the difference between this current trace and the next trace. - '$' - greater than 1 second - '@' - greater than 100 millisecond - '*' - greater than 10 millisecond - '#' - greater than 1000 microsecond - '!' - greater than 100 microsecond - '+' - greater than 10 microsecond - ' ' - less than or equal to 10 microsecond. h](j_ )}(hdelay:h]hdelay:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM9hjubjo )}(hhh](h)}(hThis is just to help catch your eye a bit better. And needs to be fixed to be only relative to the same CPU. The marks are determined by the difference between this current trace and the next trace.h]hThis is just to help catch your eye a bit better. And needs to be fixed to be only relative to the same CPU. The marks are determined by the difference between this current trace and the next trace.}(hj*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM.hj'ubjT)}(hX- '$' - greater than 1 second - '@' - greater than 100 millisecond - '*' - greater than 10 millisecond - '#' - greater than 1000 microsecond - '!' - greater than 100 microsecond - '+' - greater than 10 microsecond - ' ' - less than or equal to 10 microsecond. h]j{)}(hhh](j)}(h'$' - greater than 1 secondh]h)}(hjAh]h‘$’ - greater than 1 second}(hjChhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM3hj?ubah}(h]h ]h"]h$]h&]uh1jhj<ubj)}(h"'@' - greater than 100 millisecondh]h)}(hjXh]h&‘@’ - greater than 100 millisecond}(hjZhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM4hjVubah}(h]h ]h"]h$]h&]uh1jhj<ubj)}(h!'*' - greater than 10 millisecondh]h)}(hjoh]h%‘*’ - greater than 10 millisecond}(hjqhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM5hjmubah}(h]h ]h"]h$]h&]uh1jhj<ubj)}(h#'#' - greater than 1000 microsecondh]h)}(hjh]h'‘#’ - greater than 1000 microsecond}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM6hjubah}(h]h ]h"]h$]h&]uh1jhj<ubj)}(h"'!' - greater than 100 microsecondh]h)}(hjh]h&‘!’ - greater than 100 microsecond}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM7hjubah}(h]h ]h"]h$]h&]uh1jhj<ubj)}(h!'+' - greater than 10 microsecondh]h)}(hjh]h%‘+’ - greater than 10 microsecond}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM8hjubah}(h]h ]h"]h$]h&]uh1jhj<ubj)}(h,' ' - less than or equal to 10 microsecond. h]h)}(h+' ' - less than or equal to 10 microsecond.h]h/‘ ‘ - less than or equal to 10 microsecond.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM9hjubah}(h]h ]h"]h$]h&]uh1jhj<ubeh}(h]h ]h"]h$]h&]jjuh1jzhhhM3hj8ubah}(h]h ]h"]h$]h&]uh1jShhhM3hj'ubeh}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhM9hjubeh}(h]h ]h"]h$]h&]uh1jS hjubh)}(h)The rest is the same as the 'trace' file.h]h-The rest is the same as the ‘trace’ file.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM;hjubh)}(hgNote, the latency tracers will usually end with a back trace to easily find where the latency occurred.h]hgNote, the latency tracers will usually end with a back trace to easily find where the latency occurred.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM=hjubeh}(h]h ]h"]h$]h&]uh1jShhhM'hj9hhubeh}(h]latency-trace-formatah ]h"]latency trace formatah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(h trace_optionsh]h trace_options}(hj,hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj)hhhhhMAubh)}(hThe trace_options file (or the options directory) is used to control what gets printed in the trace output, or manipulate the tracers. To see what is available, simply cat the file::h]hThe trace_options file (or the options directory) is used to control what gets printed in the trace output, or manipulate the tracers. To see what is available, simply cat the file:}(hj:hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMChj)hhubj)}(hXcat trace_options print-parent nosym-offset nosym-addr noverbose noraw nohex nobin noblock nofields trace_printk annotate nouserstacktrace nosym-userobj noprintk-msg-only context-info nolatency-format record-cmd norecord-tgid overwrite nodisable_on_free irq-info markers noevent-fork function-trace nofunction-fork nodisplay-graph nostacktrace nobranchh]hXcat trace_options print-parent nosym-offset nosym-addr noverbose noraw nohex nobin noblock nofields trace_printk annotate nouserstacktrace nosym-userobj noprintk-msg-only context-info nolatency-format record-cmd norecord-tgid overwrite nodisable_on_free irq-info markers noevent-fork function-trace nofunction-fork nodisplay-graph nostacktrace nobranch}hjHsbah}(h]h ]h"]h$]h&]jjuh1jhhhMGhj)hhubh)}(hGTo disable one of the options, echo in the option prepended with "no"::h]hJTo disable one of the options, echo in the option prepended with “no”:}(hjVhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMehj)hhubj)}(h#echo noprint-parent > trace_optionsh]h#echo noprint-parent > trace_options}hjdsbah}(h]h ]h"]h$]h&]jjuh1jhhhMhhj)hhubh)}(h)To enable an option, leave off the "no"::h]h,To enable an option, leave off the “no”:}(hjrhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMjhj)hhubj)}(hecho sym-offset > trace_optionsh]hecho sym-offset > trace_options}hjsbah}(h]h ]h"]h$]h&]jjuh1jhhhMlhj)hhubh)}(hHere are the available options:h]hHere are the available options:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMnhj)hhubjT)}(hX: print-parent On function traces, display the calling (parent) function as well as the function being traced. :: print-parent: bash-4000 [01] 1477.606694: simple_strtoul <-kstrtoul noprint-parent: bash-4000 [01] 1477.606694: simple_strtoul sym-offset Display not only the function name, but also the offset in the function. For example, instead of seeing just "ktime_get", you will see "ktime_get+0xb/0x20". :: sym-offset: bash-4000 [01] 1477.606694: simple_strtoul+0x6/0xa0 sym-addr This will also display the function address as well as the function name. :: sym-addr: bash-4000 [01] 1477.606694: simple_strtoul verbose This deals with the trace file when the latency-format option is enabled. :: bash 4000 1 0 00000000 00010a95 [58127d26] 1720.415ms \ (+0.000ms): simple_strtoul (kstrtoul) raw This will display raw numbers. This option is best for use with user applications that can translate the raw numbers better than having it done in the kernel. hex Similar to raw, but the numbers will be in a hexadecimal format. bin This will print out the formats in raw binary. block When set, reading trace_pipe will not block when polled. fields Print the fields as described by their types. This is a better option than using hex, bin or raw, as it gives a better parsing of the content of the event. trace_printk Can disable trace_printk() from writing into the buffer. trace_printk_dest Set to have trace_printk() and similar internal tracing functions write into this instance. Note, only one trace instance can have this set. By setting this flag, it clears the trace_printk_dest flag of the instance that had it set previously. By default, the top level trace has this set, and will get it set again if another instance has it set then clears it. This flag cannot be cleared by the top level instance, as it is the default instance. The only way the top level instance has this flag cleared, is by it being set in another instance. annotate It is sometimes confusing when the CPU buffers are full and one CPU buffer had a lot of events recently, thus a shorter time frame, were another CPU may have only had a few events, which lets it have older events. When the trace is reported, it shows the oldest events first, and it may look like only one CPU ran (the one with the oldest events). When the annotate option is set, it will display when a new CPU buffer started:: -0 [001] dNs4 21169.031481: wake_up_idle_cpu <-add_timer_on -0 [001] dNs4 21169.031482: _raw_spin_unlock_irqrestore <-add_timer_on -0 [001] .Ns4 21169.031484: sub_preempt_count <-_raw_spin_unlock_irqrestore ##### CPU 2 buffer started #### -0 [002] .N.1 21169.031484: rcu_idle_exit <-cpu_idle -0 [001] .Ns3 21169.031484: _raw_spin_unlock <-clocksource_watchdog -0 [001] .Ns3 21169.031485: sub_preempt_count <-_raw_spin_unlock userstacktrace This option changes the trace. It records a stacktrace of the current user space thread after each trace event. sym-userobj when user stacktrace are enabled, look up which object the address belongs to, and print a relative address. This is especially useful when ASLR is on, otherwise you don't get a chance to resolve the address to object/file/line after the app is no longer running The lookup is performed when you read trace,trace_pipe. Example:: a.out-1623 [000] 40874.465068: /root/a.out[+0x480] <-/root/a.out[+0 x494] <- /root/a.out[+0x4a8] <- /lib/libc-2.7.so[+0x1e1a6] printk-msg-only When set, trace_printk()s will only show the format and not their parameters (if trace_bprintk() or trace_bputs() was used to save the trace_printk()). context-info Show only the event data. Hides the comm, PID, timestamp, CPU, and other useful data. latency-format This option changes the trace output. When it is enabled, the trace displays additional information about the latency, as described in "Latency trace format". pause-on-trace When set, opening the trace file for read, will pause writing to the ring buffer (as if tracing_on was set to zero). This simulates the original behavior of the trace file. When the file is closed, tracing will be enabled again. hash-ptr When set, "%p" in the event printk format displays the hashed pointer value instead of real address. This will be useful if you want to find out which hashed value is corresponding to the real value in trace log. record-cmd When any event or tracer is enabled, a hook is enabled in the sched_switch trace point to fill comm cache with mapped pids and comms. But this may cause some overhead, and if you only care about pids, and not the name of the task, disabling this option can lower the impact of tracing. See "saved_cmdlines". record-tgid When any event or tracer is enabled, a hook is enabled in the sched_switch trace point to fill the cache of mapped Thread Group IDs (TGID) mapping to pids. See "saved_tgids". overwrite This controls what happens when the trace buffer is full. If "1" (default), the oldest events are discarded and overwritten. If "0", then the newest events are discarded. (see per_cpu/cpu0/stats for overrun and dropped) disable_on_free When the free_buffer is closed, tracing will stop (tracing_on set to 0). irq-info Shows the interrupt, preempt count, need resched data. When disabled, the trace looks like:: # tracer: function # # entries-in-buffer/entries-written: 144405/9452052 #P:4 # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | -0 [002] 23636.756054: ttwu_do_activate.constprop.89 <-try_to_wake_up -0 [002] 23636.756054: activate_task <-ttwu_do_activate.constprop.89 -0 [002] 23636.756055: enqueue_task <-activate_task markers When set, the trace_marker is writable (only by root). When disabled, the trace_marker will error with EINVAL on write. event-fork When set, tasks with PIDs listed in set_event_pid will have the PIDs of their children added to set_event_pid when those tasks fork. Also, when tasks with PIDs in set_event_pid exit, their PIDs will be removed from the file. This affects PIDs listed in set_event_notrace_pid as well. function-trace The latency tracers will enable function tracing if this option is enabled (default it is). When it is disabled, the latency tracers do not trace functions. This keeps the overhead of the tracer down when performing latency tests. function-fork When set, tasks with PIDs listed in set_ftrace_pid will have the PIDs of their children added to set_ftrace_pid when those tasks fork. Also, when tasks with PIDs in set_ftrace_pid exit, their PIDs will be removed from the file. This affects PIDs in set_ftrace_notrace_pid as well. display-graph When set, the latency tracers (irqsoff, wakeup, etc) will use function graph tracing instead of function tracing. stacktrace When set, a stack trace is recorded after any trace event is recorded. branch Enable branch tracing with the tracer. This enables branch tracer along with the currently set tracer. Enabling this with the "nop" tracer is the same as just enabling the "branch" tracer. h]jT )}(hhh](jY )}(hXprint-parent On function traces, display the calling (parent) function as well as the function being traced. :: print-parent: bash-4000 [01] 1477.606694: simple_strtoul <-kstrtoul noprint-parent: bash-4000 [01] 1477.606694: simple_strtoul h](j_ )}(h print-parenth]h print-parent}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMzhjubjo )}(hhh](h)}(hbOn function traces, display the calling (parent) function as well as the function being traced. ::h]h_On function traces, display the calling (parent) function as well as the function being traced.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMqhjubj)}(hprint-parent: bash-4000 [01] 1477.606694: simple_strtoul <-kstrtoul noprint-parent: bash-4000 [01] 1477.606694: simple_strtoulh]hprint-parent: bash-4000 [01] 1477.606694: simple_strtoul <-kstrtoul noprint-parent: bash-4000 [01] 1477.606694: simple_strtoul}hjsbah}(h]h ]h"]h$]h&]jjuh1jhhhMuhjubeh}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMzhjubjY )}(hsym-offset Display not only the function name, but also the offset in the function. For example, instead of seeing just "ktime_get", you will see "ktime_get+0xb/0x20". :: sym-offset: bash-4000 [01] 1477.606694: simple_strtoul+0x6/0xa0 h](j_ )}(h sym-offseth]h sym-offset}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh](h)}(hDisplay not only the function name, but also the offset in the function. For example, instead of seeing just "ktime_get", you will see "ktime_get+0xb/0x20". ::h]hDisplay not only the function name, but also the offset in the function. For example, instead of seeing just “ktime_get”, you will see “ktime_get+0xb/0x20”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM}hjubj)}(hBsym-offset: bash-4000 [01] 1477.606694: simple_strtoul+0x6/0xa0h]hBsym-offset: bash-4000 [01] 1477.606694: simple_strtoul+0x6/0xa0}hjsbah}(h]h ]h"]h$]h&]jjuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hsym-addr This will also display the function address as well as the function name. :: sym-addr: bash-4000 [01] 1477.606694: simple_strtoul h](j_ )}(hsym-addrh]hsym-addr}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh](h)}(hLThis will also display the function address as well as the function name. ::h]hIThis will also display the function address as well as the function name.}(hj2hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj/ubj)}(hBsym-addr: bash-4000 [01] 1477.606694: simple_strtoul h]hBsym-addr: bash-4000 [01] 1477.606694: simple_strtoul }hj@sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj/ubeh}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hverbose This deals with the trace file when the latency-format option is enabled. :: bash 4000 1 0 00000000 00010a95 [58127d26] 1720.415ms \ (+0.000ms): simple_strtoul (kstrtoul) h](j_ )}(hverboseh]hverbose}(hj^hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjZubjo )}(hhh](h)}(hLThis deals with the trace file when the latency-format option is enabled. ::h]hIThis deals with the trace file when the latency-format option is enabled.}(hjohhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjlubj)}(h^bash 4000 1 0 00000000 00010a95 [58127d26] 1720.415ms \ (+0.000ms): simple_strtoul (kstrtoul)h]h^bash 4000 1 0 00000000 00010a95 [58127d26] 1720.415ms \ (+0.000ms): simple_strtoul (kstrtoul)}hj}sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhjlubeh}(h]h ]h"]h$]h&]uh1jn hjZubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hraw This will display raw numbers. This option is best for use with user applications that can translate the raw numbers better than having it done in the kernel. h](j_ )}(hrawh]hraw}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]h)}(hThis will display raw numbers. This option is best for use with user applications that can translate the raw numbers better than having it done in the kernel.h]hThis will display raw numbers. This option is best for use with user applications that can translate the raw numbers better than having it done in the kernel.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hEhex Similar to raw, but the numbers will be in a hexadecimal format. h](j_ )}(hhexh]hhex}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]h)}(h@Similar to raw, but the numbers will be in a hexadecimal format.h]h@Similar to raw, but the numbers will be in a hexadecimal format.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(h3bin This will print out the formats in raw binary. h](j_ )}(hbinh]hbin}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]h)}(h.This will print out the formats in raw binary.h]h.This will print out the formats in raw binary.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(h?block When set, reading trace_pipe will not block when polled. h](j_ )}(hblockh]hblock}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj$ubjo )}(hhh]h)}(h8When set, reading trace_pipe will not block when polled.h]h8When set, reading trace_pipe will not block when polled.}(hj9hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj6ubah}(h]h ]h"]h$]h&]uh1jn hj$ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hfields Print the fields as described by their types. This is a better option than using hex, bin or raw, as it gives a better parsing of the content of the event. h](j_ )}(hfieldsh]hfields}(hjWhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjSubjo )}(hhh]h)}(hPrint the fields as described by their types. This is a better option than using hex, bin or raw, as it gives a better parsing of the content of the event.h]hPrint the fields as described by their types. This is a better option than using hex, bin or raw, as it gives a better parsing of the content of the event.}(hjhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjeubah}(h]h ]h"]h$]h&]uh1jn hjSubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hFtrace_printk Can disable trace_printk() from writing into the buffer. h](j_ )}(h trace_printkh]h trace_printk}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]h)}(h8Can disable trace_printk() from writing into the buffer.h]h8Can disable trace_printk() from writing into the buffer.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hX7trace_printk_dest Set to have trace_printk() and similar internal tracing functions write into this instance. Note, only one trace instance can have this set. By setting this flag, it clears the trace_printk_dest flag of the instance that had it set previously. By default, the top level trace has this set, and will get it set again if another instance has it set then clears it. This flag cannot be cleared by the top level instance, as it is the default instance. The only way the top level instance has this flag cleared, is by it being set in another instance. h](j_ )}(htrace_printk_desth]htrace_printk_dest}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh](h)}(hXjSet to have trace_printk() and similar internal tracing functions write into this instance. Note, only one trace instance can have this set. By setting this flag, it clears the trace_printk_dest flag of the instance that had it set previously. By default, the top level trace has this set, and will get it set again if another instance has it set then clears it.h]hXjSet to have trace_printk() and similar internal tracing functions write into this instance. Note, only one trace instance can have this set. By setting this flag, it clears the trace_printk_dest flag of the instance that had it set previously. By default, the top level trace has this set, and will get it set again if another instance has it set then clears it.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubh)}(hThis flag cannot be cleared by the top level instance, as it is the default instance. The only way the top level instance has this flag cleared, is by it being set in another instance.h]hThis flag cannot be cleared by the top level instance, as it is the default instance. The only way the top level instance has this flag cleared, is by it being set in another instance.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubeh}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hXannotate It is sometimes confusing when the CPU buffers are full and one CPU buffer had a lot of events recently, thus a shorter time frame, were another CPU may have only had a few events, which lets it have older events. When the trace is reported, it shows the oldest events first, and it may look like only one CPU ran (the one with the oldest events). When the annotate option is set, it will display when a new CPU buffer started:: -0 [001] dNs4 21169.031481: wake_up_idle_cpu <-add_timer_on -0 [001] dNs4 21169.031482: _raw_spin_unlock_irqrestore <-add_timer_on -0 [001] .Ns4 21169.031484: sub_preempt_count <-_raw_spin_unlock_irqrestore ##### CPU 2 buffer started #### -0 [002] .N.1 21169.031484: rcu_idle_exit <-cpu_idle -0 [001] .Ns3 21169.031484: _raw_spin_unlock <-clocksource_watchdog -0 [001] .Ns3 21169.031485: sub_preempt_count <-_raw_spin_unlock h](j_ )}(hannotateh]hannotate}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh](h)}(hXIt is sometimes confusing when the CPU buffers are full and one CPU buffer had a lot of events recently, thus a shorter time frame, were another CPU may have only had a few events, which lets it have older events. When the trace is reported, it shows the oldest events first, and it may look like only one CPU ran (the one with the oldest events). When the annotate option is set, it will display when a new CPU buffer started::h]hXIt is sometimes confusing when the CPU buffers are full and one CPU buffer had a lot of events recently, thus a shorter time frame, were another CPU may have only had a few events, which lets it have older events. When the trace is reported, it shows the oldest events first, and it may look like only one CPU ran (the one with the oldest events). When the annotate option is set, it will display when a new CPU buffer started:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hX  -0 [001] dNs4 21169.031481: wake_up_idle_cpu <-add_timer_on -0 [001] dNs4 21169.031482: _raw_spin_unlock_irqrestore <-add_timer_on -0 [001] .Ns4 21169.031484: sub_preempt_count <-_raw_spin_unlock_irqrestore ##### CPU 2 buffer started #### -0 [002] .N.1 21169.031484: rcu_idle_exit <-cpu_idle -0 [001] .Ns3 21169.031484: _raw_spin_unlock <-clocksource_watchdog -0 [001] .Ns3 21169.031485: sub_preempt_count <-_raw_spin_unlockh]hX  -0 [001] dNs4 21169.031481: wake_up_idle_cpu <-add_timer_on -0 [001] dNs4 21169.031482: _raw_spin_unlock_irqrestore <-add_timer_on -0 [001] .Ns4 21169.031484: sub_preempt_count <-_raw_spin_unlock_irqrestore ##### CPU 2 buffer started #### -0 [002] .N.1 21169.031484: rcu_idle_exit <-cpu_idle -0 [001] .Ns3 21169.031484: _raw_spin_unlock <-clocksource_watchdog -0 [001] .Ns3 21169.031485: sub_preempt_count <-_raw_spin_unlock}hjsbah}(h]h ]h"]h$]h&]jjuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(huserstacktrace This option changes the trace. It records a stacktrace of the current user space thread after each trace event. h](j_ )}(huserstacktraceh]huserstacktrace}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj+ubjo )}(hhh]h)}(hoThis option changes the trace. It records a stacktrace of the current user space thread after each trace event.h]hoThis option changes the trace. It records a stacktrace of the current user space thread after each trace event.}(hj@hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj=ubah}(h]h ]h"]h$]h&]uh1jn hj+ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hXsym-userobj when user stacktrace are enabled, look up which object the address belongs to, and print a relative address. This is especially useful when ASLR is on, otherwise you don't get a chance to resolve the address to object/file/line after the app is no longer running The lookup is performed when you read trace,trace_pipe. Example:: a.out-1623 [000] 40874.465068: /root/a.out[+0x480] <-/root/a.out[+0 x494] <- /root/a.out[+0x4a8] <- /lib/libc-2.7.so[+0x1e1a6] h](j_ )}(h sym-userobjh]h sym-userobj}(hj^hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjZubjo )}(hhh](h)}(hXwhen user stacktrace are enabled, look up which object the address belongs to, and print a relative address. This is especially useful when ASLR is on, otherwise you don't get a chance to resolve the address to object/file/line after the app is no longer runningh]hXwhen user stacktrace are enabled, look up which object the address belongs to, and print a relative address. This is especially useful when ASLR is on, otherwise you don’t get a chance to resolve the address to object/file/line after the app is no longer running}(hjohhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjlubh)}(hAThe lookup is performed when you read trace,trace_pipe. Example::h]h@The lookup is performed when you read trace,trace_pipe. Example:}(hj}hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjlubj)}(ha.out-1623 [000] 40874.465068: /root/a.out[+0x480] <-/root/a.out[+0 x494] <- /root/a.out[+0x4a8] <- /lib/libc-2.7.so[+0x1e1a6]h]ha.out-1623 [000] 40874.465068: /root/a.out[+0x480] <-/root/a.out[+0 x494] <- /root/a.out[+0x4a8] <- /lib/libc-2.7.so[+0x1e1a6]}hjsbah}(h]h ]h"]h$]h&]jjuh1jhhhMhjlubeh}(h]h ]h"]h$]h&]uh1jn hjZubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hprintk-msg-only When set, trace_printk()s will only show the format and not their parameters (if trace_bprintk() or trace_bputs() was used to save the trace_printk()). h](j_ )}(hprintk-msg-onlyh]hprintk-msg-only}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]h)}(hWhen set, trace_printk()s will only show the format and not their parameters (if trace_bprintk() or trace_bputs() was used to save the trace_printk()).h]hWhen set, trace_printk()s will only show the format and not their parameters (if trace_bprintk() or trace_bputs() was used to save the trace_printk()).}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hccontext-info Show only the event data. Hides the comm, PID, timestamp, CPU, and other useful data. h](j_ )}(h context-infoh]h context-info}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]h)}(hUShow only the event data. Hides the comm, PID, timestamp, CPU, and other useful data.h]hUShow only the event data. Hides the comm, PID, timestamp, CPU, and other useful data.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hlatency-format This option changes the trace output. When it is enabled, the trace displays additional information about the latency, as described in "Latency trace format". h](j_ )}(hlatency-formath]hlatency-format}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]h)}(hThis option changes the trace output. When it is enabled, the trace displays additional information about the latency, as described in "Latency trace format".h]hThis option changes the trace output. When it is enabled, the trace displays additional information about the latency, as described in “Latency trace format”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hpause-on-trace When set, opening the trace file for read, will pause writing to the ring buffer (as if tracing_on was set to zero). This simulates the original behavior of the trace file. When the file is closed, tracing will be enabled again. h](j_ )}(hpause-on-traceh]hpause-on-trace}(hj6hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj2ubjo )}(hhh]h)}(hWhen set, opening the trace file for read, will pause writing to the ring buffer (as if tracing_on was set to zero). This simulates the original behavior of the trace file. When the file is closed, tracing will be enabled again.h]hWhen set, opening the trace file for read, will pause writing to the ring buffer (as if tracing_on was set to zero). This simulates the original behavior of the trace file. When the file is closed, tracing will be enabled again.}(hjGhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjDubah}(h]h ]h"]h$]h&]uh1jn hj2ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hhash-ptr When set, "%p" in the event printk format displays the hashed pointer value instead of real address. This will be useful if you want to find out which hashed value is corresponding to the real value in trace log. h](j_ )}(hhash-ptrh]hhash-ptr}(hjehhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjaubjo )}(hhh]h)}(hWhen set, "%p" in the event printk format displays the hashed pointer value instead of real address. This will be useful if you want to find out which hashed value is corresponding to the real value in trace log.h]hWhen set, “%p” in the event printk format displays the hashed pointer value instead of real address. This will be useful if you want to find out which hashed value is corresponding to the real value in trace log.}(hjvhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjsubah}(h]h ]h"]h$]h&]uh1jn hjaubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hX?record-cmd When any event or tracer is enabled, a hook is enabled in the sched_switch trace point to fill comm cache with mapped pids and comms. But this may cause some overhead, and if you only care about pids, and not the name of the task, disabling this option can lower the impact of tracing. See "saved_cmdlines". h](j_ )}(h record-cmdh]h record-cmd}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]h)}(hX3When any event or tracer is enabled, a hook is enabled in the sched_switch trace point to fill comm cache with mapped pids and comms. But this may cause some overhead, and if you only care about pids, and not the name of the task, disabling this option can lower the impact of tracing. See "saved_cmdlines".h]hX7When any event or tracer is enabled, a hook is enabled in the sched_switch trace point to fill comm cache with mapped pids and comms. But this may cause some overhead, and if you only care about pids, and not the name of the task, disabling this option can lower the impact of tracing. See “saved_cmdlines”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hrecord-tgid When any event or tracer is enabled, a hook is enabled in the sched_switch trace point to fill the cache of mapped Thread Group IDs (TGID) mapping to pids. See "saved_tgids". h](j_ )}(h record-tgidh]h record-tgid}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]h)}(hWhen any event or tracer is enabled, a hook is enabled in the sched_switch trace point to fill the cache of mapped Thread Group IDs (TGID) mapping to pids. See "saved_tgids".h]hWhen any event or tracer is enabled, a hook is enabled in the sched_switch trace point to fill the cache of mapped Thread Group IDs (TGID) mapping to pids. See “saved_tgids”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hoverwrite This controls what happens when the trace buffer is full. If "1" (default), the oldest events are discarded and overwritten. If "0", then the newest events are discarded. (see per_cpu/cpu0/stats for overrun and dropped) h](j_ )}(h overwriteh]h overwrite}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM hjubjo )}(hhh]h)}(hThis controls what happens when the trace buffer is full. If "1" (default), the oldest events are discarded and overwritten. If "0", then the newest events are discarded. (see per_cpu/cpu0/stats for overrun and dropped)h]hThis controls what happens when the trace buffer is full. If “1” (default), the oldest events are discarded and overwritten. If “0”, then the newest events are discarded. (see per_cpu/cpu0/stats for overrun and dropped)}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhM hjubjY )}(hYdisable_on_free When the free_buffer is closed, tracing will stop (tracing_on set to 0). h](j_ )}(hdisable_on_freeh]hdisable_on_free}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjubjo )}(hhh]h)}(hHWhen the free_buffer is closed, tracing will stop (tracing_on set to 0).h]hHWhen the free_buffer is closed, tracing will stop (tracing_on set to 0).}(hj2hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj/ubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hXeirq-info Shows the interrupt, preempt count, need resched data. When disabled, the trace looks like:: # tracer: function # # entries-in-buffer/entries-written: 144405/9452052 #P:4 # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | -0 [002] 23636.756054: ttwu_do_activate.constprop.89 <-try_to_wake_up -0 [002] 23636.756054: activate_task <-ttwu_do_activate.constprop.89 -0 [002] 23636.756055: enqueue_task <-activate_task h](j_ )}(hirq-infoh]hirq-info}(hjPhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjLubjo )}(hhh](h)}(h\Shows the interrupt, preempt count, need resched data. When disabled, the trace looks like::h]h[Shows the interrupt, preempt count, need resched data. When disabled, the trace looks like:}(hjahhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj^ubj)}(hX# tracer: function # # entries-in-buffer/entries-written: 144405/9452052 #P:4 # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | -0 [002] 23636.756054: ttwu_do_activate.constprop.89 <-try_to_wake_up -0 [002] 23636.756054: activate_task <-ttwu_do_activate.constprop.89 -0 [002] 23636.756055: enqueue_task <-activate_taskh]hX# tracer: function # # entries-in-buffer/entries-written: 144405/9452052 #P:4 # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | -0 [002] 23636.756054: ttwu_do_activate.constprop.89 <-try_to_wake_up -0 [002] 23636.756054: activate_task <-ttwu_do_activate.constprop.89 -0 [002] 23636.756055: enqueue_task <-activate_task}hjosbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj^ubeh}(h]h ]h"]h$]h&]uh1jn hjLubeh}(h]h ]h"]h$]h&]uh1jX hhhMhjubjY )}(hmarkers When set, the trace_marker is writable (only by root). When disabled, the trace_marker will error with EINVAL on write. h](j_ )}(hmarkersh]hmarkers}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM#hjubjo )}(hhh]h)}(hwWhen set, the trace_marker is writable (only by root). When disabled, the trace_marker will error with EINVAL on write.h]hwWhen set, the trace_marker is writable (only by root). When disabled, the trace_marker will error with EINVAL on write.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM!hjubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhM#hjubjY )}(hX(event-fork When set, tasks with PIDs listed in set_event_pid will have the PIDs of their children added to set_event_pid when those tasks fork. Also, when tasks with PIDs in set_event_pid exit, their PIDs will be removed from the file. This affects PIDs listed in set_event_notrace_pid as well. h](j_ )}(h event-forkh]h event-fork}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM+hjubjo )}(hhh](h)}(hWhen set, tasks with PIDs listed in set_event_pid will have the PIDs of their children added to set_event_pid when those tasks fork. Also, when tasks with PIDs in set_event_pid exit, their PIDs will be removed from the file.h]hWhen set, tasks with PIDs listed in set_event_pid will have the PIDs of their children added to set_event_pid when those tasks fork. Also, when tasks with PIDs in set_event_pid exit, their PIDs will be removed from the file.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM&hjubh)}(h:This affects PIDs listed in set_event_notrace_pid as well.h]h:This affects PIDs listed in set_event_notrace_pid as well.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM+hjubeh}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhM+hjubjY )}(hfunction-trace The latency tracers will enable function tracing if this option is enabled (default it is). When it is disabled, the latency tracers do not trace functions. This keeps the overhead of the tracer down when performing latency tests. h](j_ )}(hfunction-traceh]hfunction-trace}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM2hjubjo )}(hhh]h)}(hThe latency tracers will enable function tracing if this option is enabled (default it is). When it is disabled, the latency tracers do not trace functions. This keeps the overhead of the tracer down when performing latency tests.h]hThe latency tracers will enable function tracing if this option is enabled (default it is). When it is disabled, the latency tracers do not trace functions. This keeps the overhead of the tracer down when performing latency tests.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM.hj ubah}(h]h ]h"]h$]h&]uh1jn hjubeh}(h]h ]h"]h$]h&]uh1jX hhhM2hjubjY )}(hX(function-fork When set, tasks with PIDs listed in set_ftrace_pid will have the PIDs of their children added to set_ftrace_pid when those tasks fork. Also, when tasks with PIDs in set_ftrace_pid exit, their PIDs will be removed from the file. This affects PIDs in set_ftrace_notrace_pid as well. h](j_ )}(h function-forkh]h function-fork}(hj( hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM;hj$ ubjo )}(hhh](h)}(hWhen set, tasks with PIDs listed in set_ftrace_pid will have the PIDs of their children added to set_ftrace_pid when those tasks fork. Also, when tasks with PIDs in set_ftrace_pid exit, their PIDs will be removed from the file.h]hWhen set, tasks with PIDs listed in set_ftrace_pid will have the PIDs of their children added to set_ftrace_pid when those tasks fork. Also, when tasks with PIDs in set_ftrace_pid exit, their PIDs will be removed from the file.}(hj9 hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM5hj6 ubh)}(h4This affects PIDs in set_ftrace_notrace_pid as well.h]h4This affects PIDs in set_ftrace_notrace_pid as well.}(hjG hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM;hj6 ubeh}(h]h ]h"]h$]h&]uh1jn hj$ ubeh}(h]h ]h"]h$]h&]uh1jX hhhM;hjubjY )}(hdisplay-graph When set, the latency tracers (irqsoff, wakeup, etc) will use function graph tracing instead of function tracing. h](j_ )}(h display-graphh]h display-graph}(hje hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM?hja ubjo )}(hhh]h)}(hqWhen set, the latency tracers (irqsoff, wakeup, etc) will use function graph tracing instead of function tracing.h]hqWhen set, the latency tracers (irqsoff, wakeup, etc) will use function graph tracing instead of function tracing.}(hjv hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM>hjs ubah}(h]h ]h"]h$]h&]uh1jn hja ubeh}(h]h ]h"]h$]h&]uh1jX hhhM?hjubjY )}(hRstacktrace When set, a stack trace is recorded after any trace event is recorded. h](j_ )}(h stacktraceh]h stacktrace}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMChj ubjo )}(hhh]h)}(hFWhen set, a stack trace is recorded after any trace event is recorded.h]hFWhen set, a stack trace is recorded after any trace event is recorded.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMBhj ubah}(h]h ]h"]h$]h&]uh1jn hj ubeh}(h]h ]h"]h$]h&]uh1jX hhhMChjubjY )}(hbranch Enable branch tracing with the tracer. This enables branch tracer along with the currently set tracer. Enabling this with the "nop" tracer is the same as just enabling the "branch" tracer. h](j_ )}(hbranchh]hbranch}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMIhj ubjo )}(hhh]h)}(hEnable branch tracing with the tracer. This enables branch tracer along with the currently set tracer. Enabling this with the "nop" tracer is the same as just enabling the "branch" tracer.h]hEnable branch tracing with the tracer. This enables branch tracer along with the currently set tracer. Enabling this with the “nop” tracer is the same as just enabling the “branch” tracer.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMFhj ubah}(h]h ]h"]h$]h&]uh1jn hj ubeh}(h]h ]h"]h$]h&]uh1jX hhhMIhjubeh}(h]h ]h"]h$]h&]uh1jS hjubah}(h]h ]h"]h$]h&]uh1jShhhMphj)hhubhtip)}(hSome tracers have their own options. They only appear in this file when the tracer is active. They always appear in the options directory.h]h)}(hSome tracers have their own options. They only appear in this file when the tracer is active. They always appear in the options directory.h]hSome tracers have their own options. They only appear in this file when the tracer is active. They always appear in the options directory.}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMKhj ubah}(h]h ]h"]h$]h&]uh1j hj)hhhhhNubh)}(h Here are the per tracer options:h]h Here are the per tracer options:}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMPhj)hhubh)}(hOptions for function tracer:h]hOptions for function tracer:}(hj"!hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMRhj)hhubjT)}(hX^func_stack_trace When set, a stack trace is recorded after every function that is recorded. NOTE! Limit the functions that are recorded before enabling this, with "set_ftrace_filter" otherwise the system performance will be critically degraded. Remember to disable this option before clearing the function filter. h]jT )}(hhh]jY )}(hX:func_stack_trace When set, a stack trace is recorded after every function that is recorded. NOTE! Limit the functions that are recorded before enabling this, with "set_ftrace_filter" otherwise the system performance will be critically degraded. Remember to disable this option before clearing the function filter. h](j_ )}(hfunc_stack_traceh]hfunc_stack_trace}(hj;!hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMZhj7!ubjo )}(hhh]h)}(hX(When set, a stack trace is recorded after every function that is recorded. NOTE! Limit the functions that are recorded before enabling this, with "set_ftrace_filter" otherwise the system performance will be critically degraded. Remember to disable this option before clearing the function filter.h]hX,When set, a stack trace is recorded after every function that is recorded. NOTE! Limit the functions that are recorded before enabling this, with “set_ftrace_filter” otherwise the system performance will be critically degraded. Remember to disable this option before clearing the function filter.}(hjL!hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMUhjI!ubah}(h]h ]h"]h$]h&]uh1jn hj7!ubeh}(h]h ]h"]h$]h&]uh1jX hhhMZhj4!ubah}(h]h ]h"]h$]h&]uh1jS hj0!ubah}(h]h ]h"]h$]h&]uh1jShhhMThj)hhubh)}(h"Options for function_graph tracer:h]h"Options for function_graph tracer:}(hjr!hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM\hj)hhubjT)}(hX Since the function_graph tracer has a slightly different output it has its own options to control what is displayed. funcgraph-overrun When set, the "overrun" of the graph stack is displayed after each function traced. The overrun, is when the stack depth of the calls is greater than what is reserved for each task. Each task has a fixed array of functions to trace in the call graph. If the depth of the calls exceeds that, the function is not traced. The overrun is the number of functions missed due to exceeding this array. funcgraph-cpu When set, the CPU number of the CPU where the trace occurred is displayed. funcgraph-overhead When set, if the function takes longer than A certain amount, then a delay marker is displayed. See "delay" above, under the header description. funcgraph-proc Unlike other tracers, the process' command line is not displayed by default, but instead only when a task is traced in and out during a context switch. Enabling this options has the command of each process displayed at every line. funcgraph-duration At the end of each function (the return) the duration of the amount of time in the function is displayed in microseconds. funcgraph-abstime When set, the timestamp is displayed at each line. funcgraph-irqs When disabled, functions that happen inside an interrupt will not be traced. funcgraph-tail When set, the return event will include the function that it represents. By default this is off, and only a closing curly bracket "}" is displayed for the return of a function. funcgraph-retval When set, the return value of each traced function will be printed after an equal sign "=". By default this is off. funcgraph-retval-hex When set, the return value will always be printed in hexadecimal format. If the option is not set and the return value is an error code, it will be printed in signed decimal format; otherwise it will also be printed in hexadecimal format. By default, this option is off. sleep-time When running function graph tracer, to include the time a task schedules out in its function. When enabled, it will account time the task has been scheduled out as part of the function call. graph-time When running function profiler with function graph tracer, to include the time to call nested functions. When this is not set, the time reported for the function will only include the time the function itself executed for, not the time for functions that it called. h](h)}(htSince the function_graph tracer has a slightly different output it has its own options to control what is displayed.h]htSince the function_graph tracer has a slightly different output it has its own options to control what is displayed.}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM^hj!ubjT)}(hX0 funcgraph-overrun When set, the "overrun" of the graph stack is displayed after each function traced. The overrun, is when the stack depth of the calls is greater than what is reserved for each task. Each task has a fixed array of functions to trace in the call graph. If the depth of the calls exceeds that, the function is not traced. The overrun is the number of functions missed due to exceeding this array. funcgraph-cpu When set, the CPU number of the CPU where the trace occurred is displayed. funcgraph-overhead When set, if the function takes longer than A certain amount, then a delay marker is displayed. See "delay" above, under the header description. funcgraph-proc Unlike other tracers, the process' command line is not displayed by default, but instead only when a task is traced in and out during a context switch. Enabling this options has the command of each process displayed at every line. funcgraph-duration At the end of each function (the return) the duration of the amount of time in the function is displayed in microseconds. funcgraph-abstime When set, the timestamp is displayed at each line. funcgraph-irqs When disabled, functions that happen inside an interrupt will not be traced. funcgraph-tail When set, the return event will include the function that it represents. By default this is off, and only a closing curly bracket "}" is displayed for the return of a function. funcgraph-retval When set, the return value of each traced function will be printed after an equal sign "=". By default this is off. funcgraph-retval-hex When set, the return value will always be printed in hexadecimal format. If the option is not set and the return value is an error code, it will be printed in signed decimal format; otherwise it will also be printed in hexadecimal format. By default, this option is off. sleep-time When running function graph tracer, to include the time a task schedules out in its function. When enabled, it will account time the task has been scheduled out as part of the function call. graph-time When running function profiler with function graph tracer, to include the time to call nested functions. When this is not set, the time reported for the function will only include the time the function itself executed for, not the time for functions that it called. h]jT )}(hhh](jY )}(hXfuncgraph-overrun When set, the "overrun" of the graph stack is displayed after each function traced. The overrun, is when the stack depth of the calls is greater than what is reserved for each task. Each task has a fixed array of functions to trace in the call graph. If the depth of the calls exceeds that, the function is not traced. The overrun is the number of functions missed due to exceeding this array. h](j_ )}(hfuncgraph-overrunh]hfuncgraph-overrun}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMjhj!ubjo )}(hhh]h)}(hXWhen set, the "overrun" of the graph stack is displayed after each function traced. The overrun, is when the stack depth of the calls is greater than what is reserved for each task. Each task has a fixed array of functions to trace in the call graph. If the depth of the calls exceeds that, the function is not traced. The overrun is the number of functions missed due to exceeding this array.h]hXWhen set, the “overrun” of the graph stack is displayed after each function traced. The overrun, is when the stack depth of the calls is greater than what is reserved for each task. Each task has a fixed array of functions to trace in the call graph. If the depth of the calls exceeds that, the function is not traced. The overrun is the number of functions missed due to exceeding this array.}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMbhj!ubah}(h]h ]h"]h$]h&]uh1jn hj!ubeh}(h]h ]h"]h$]h&]uh1jX hhhMjhj!ubjY )}(hYfuncgraph-cpu When set, the CPU number of the CPU where the trace occurred is displayed. h](j_ )}(h funcgraph-cpuh]h funcgraph-cpu}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMnhj!ubjo )}(hhh]h)}(hJWhen set, the CPU number of the CPU where the trace occurred is displayed.h]hJWhen set, the CPU number of the CPU where the trace occurred is displayed.}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMmhj!ubah}(h]h ]h"]h$]h&]uh1jn hj!ubeh}(h]h ]h"]h$]h&]uh1jX hhhMnhj!ubjY )}(hfuncgraph-overhead When set, if the function takes longer than A certain amount, then a delay marker is displayed. See "delay" above, under the header description. h](j_ )}(hfuncgraph-overheadh]hfuncgraph-overhead}(hj!hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMthj!ubjo )}(hhh]h)}(hWhen set, if the function takes longer than A certain amount, then a delay marker is displayed. See "delay" above, under the header description.h]hWhen set, if the function takes longer than A certain amount, then a delay marker is displayed. See “delay” above, under the header description.}(hj "hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMqhj "ubah}(h]h ]h"]h$]h&]uh1jn hj!ubeh}(h]h ]h"]h$]h&]uh1jX hhhMthj!ubjY )}(hfuncgraph-proc Unlike other tracers, the process' command line is not displayed by default, but instead only when a task is traced in and out during a context switch. Enabling this options has the command of each process displayed at every line. h](j_ )}(hfuncgraph-proch]hfuncgraph-proc}(hj*"hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM{hj&"ubjo )}(hhh]h)}(hUnlike other tracers, the process' command line is not displayed by default, but instead only when a task is traced in and out during a context switch. Enabling this options has the command of each process displayed at every line.h]hUnlike other tracers, the process’ command line is not displayed by default, but instead only when a task is traced in and out during a context switch. Enabling this options has the command of each process displayed at every line.}(hj;"hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMwhj8"ubah}(h]h ]h"]h$]h&]uh1jn hj&"ubeh}(h]h ]h"]h$]h&]uh1jX hhhM{hj!ubjY )}(hfuncgraph-duration At the end of each function (the return) the duration of the amount of time in the function is displayed in microseconds. h](j_ )}(hfuncgraph-durationh]hfuncgraph-duration}(hjY"hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjU"ubjo )}(hhh]h)}(hyAt the end of each function (the return) the duration of the amount of time in the function is displayed in microseconds.h]hyAt the end of each function (the return) the duration of the amount of time in the function is displayed in microseconds.}(hjj"hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM~hjg"ubah}(h]h ]h"]h$]h&]uh1jn hjU"ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhj!ubjY )}(hEfuncgraph-abstime When set, the timestamp is displayed at each line. h](j_ )}(hfuncgraph-abstimeh]hfuncgraph-abstime}(hj"hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj"ubjo )}(hhh]h)}(h2When set, the timestamp is displayed at each line.h]h2When set, the timestamp is displayed at each line.}(hj"hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj"ubah}(h]h ]h"]h$]h&]uh1jn hj"ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhj!ubjY )}(h\funcgraph-irqs When disabled, functions that happen inside an interrupt will not be traced. h](j_ )}(hfuncgraph-irqsh]hfuncgraph-irqs}(hj"hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj"ubjo )}(hhh]h)}(hLWhen disabled, functions that happen inside an interrupt will not be traced.h]hLWhen disabled, functions that happen inside an interrupt will not be traced.}(hj"hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj"ubah}(h]h ]h"]h$]h&]uh1jn hj"ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhj!ubjY )}(hfuncgraph-tail When set, the return event will include the function that it represents. By default this is off, and only a closing curly bracket "}" is displayed for the return of a function. h](j_ )}(hfuncgraph-tailh]hfuncgraph-tail}(hj"hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj"ubjo )}(hhh]h)}(hWhen set, the return event will include the function that it represents. By default this is off, and only a closing curly bracket "}" is displayed for the return of a function.h]hWhen set, the return event will include the function that it represents. By default this is off, and only a closing curly bracket “}” is displayed for the return of a function.}(hj"hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj"ubah}(h]h ]h"]h$]h&]uh1jn hj"ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhj!ubjY )}(hfuncgraph-retval When set, the return value of each traced function will be printed after an equal sign "=". By default this is off. h](j_ )}(hfuncgraph-retvalh]hfuncgraph-retval}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj#ubjo )}(hhh]h)}(hsWhen set, the return value of each traced function will be printed after an equal sign "=". By default this is off.h]hwWhen set, the return value of each traced function will be printed after an equal sign “=”. By default this is off.}(hj&#hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj##ubah}(h]h ]h"]h$]h&]uh1jn hj#ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhj!ubjY )}(hX$funcgraph-retval-hex When set, the return value will always be printed in hexadecimal format. If the option is not set and the return value is an error code, it will be printed in signed decimal format; otherwise it will also be printed in hexadecimal format. By default, this option is off. h](j_ )}(hfuncgraph-retval-hexh]hfuncgraph-retval-hex}(hjD#hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj@#ubjo )}(hhh]h)}(hXWhen set, the return value will always be printed in hexadecimal format. If the option is not set and the return value is an error code, it will be printed in signed decimal format; otherwise it will also be printed in hexadecimal format. By default, this option is off.h]hXWhen set, the return value will always be printed in hexadecimal format. If the option is not set and the return value is an error code, it will be printed in signed decimal format; otherwise it will also be printed in hexadecimal format. By default, this option is off.}(hjU#hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjR#ubah}(h]h ]h"]h$]h&]uh1jn hj@#ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhj!ubjY )}(hsleep-time When running function graph tracer, to include the time a task schedules out in its function. When enabled, it will account time the task has been scheduled out as part of the function call. h](j_ )}(h sleep-timeh]h sleep-time}(hjs#hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhjo#ubjo )}(hhh]h)}(hWhen running function graph tracer, to include the time a task schedules out in its function. When enabled, it will account time the task has been scheduled out as part of the function call.h]hWhen running function graph tracer, to include the time a task schedules out in its function. When enabled, it will account time the task has been scheduled out as part of the function call.}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj#ubah}(h]h ]h"]h$]h&]uh1jn hjo#ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhj!ubjY )}(hXgraph-time When running function profiler with function graph tracer, to include the time to call nested functions. When this is not set, the time reported for the function will only include the time the function itself executed for, not the time for functions that it called. h](j_ )}(h graph-timeh]h graph-time}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj#ubjo )}(hhh]h)}(hX When running function profiler with function graph tracer, to include the time to call nested functions. When this is not set, the time reported for the function will only include the time the function itself executed for, not the time for functions that it called.h]hX When running function profiler with function graph tracer, to include the time to call nested functions. When this is not set, the time reported for the function will only include the time the function itself executed for, not the time for functions that it called.}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj#ubah}(h]h ]h"]h$]h&]uh1jn hj#ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhj!ubeh}(h]h ]h"]h$]h&]uh1jS hj!ubah}(h]h ]h"]h$]h&]uh1jShhhMahj!ubeh}(h]h ]h"]h$]h&]uh1jShhhM^hj)hhubh)}(hOptions for blk tracer:h]hOptions for blk tracer:}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj)hhubjT)}(h5blk_classic Shows a more minimalistic output. h]jT )}(hhh]jY )}(h/blk_classic Shows a more minimalistic output. h](j_ )}(h blk_classich]h blk_classic}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMhj#ubjo )}(hhh]h)}(h!Shows a more minimalistic output.h]h!Shows a more minimalistic output.}(hj $hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj$ubah}(h]h ]h"]h$]h&]uh1jn hj#ubeh}(h]h ]h"]h$]h&]uh1jX hhhMhj#ubah}(h]h ]h"]h$]h&]uh1jS hj#ubah}(h]h ]h"]h$]h&]uh1jShhhMhj)hhubeh}(h] trace-optionsah ]h"] trace_optionsah$]h&]uh1hhhhhhhhMAubh)}(hhh](h)}(hirqsoffh]hirqsoff}(hj:$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj7$hhhhhMubh)}(hXWhen interrupts are disabled, the CPU can not react to any other external event (besides NMIs and SMIs). This prevents the timer interrupt from triggering or the mouse interrupt from letting the kernel know of a new mouse event. The result is a latency with the reaction time.h]hXWhen interrupts are disabled, the CPU can not react to any other external event (besides NMIs and SMIs). This prevents the timer interrupt from triggering or the mouse interrupt from letting the kernel know of a new mouse event. The result is a latency with the reaction time.}(hjH$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj7$hhubh)}(hXThe irqsoff tracer tracks the time for which interrupts are disabled. When a new maximum latency is hit, the tracer saves the trace leading up to that latency point so that every time a new maximum is reached, the old saved trace is discarded and the new trace is saved.h]hXThe irqsoff tracer tracks the time for which interrupts are disabled. When a new maximum latency is hit, the tracer saves the trace leading up to that latency point so that every time a new maximum is reached, the old saved trace is discarded and the new trace is saved.}(hjV$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj7$hhubh)}(hKTo reset the maximum, echo 0 into tracing_max_latency. Here is an example::h]hJTo reset the maximum, echo 0 into tracing_max_latency. Here is an example:}(hjd$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj7$hhubj)}(hX# echo 0 > options/function-trace # echo irqsoff > current_tracer # echo 1 > tracing_on # echo 0 > tracing_max_latency # ls -ltr [...] # echo 0 > tracing_on # cat trace # tracer: irqsoff # # irqsoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 16 us, #4/4, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: swapper/0-0 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: run_timer_softirq # => ended at: run_timer_softirq # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 0d.s2 0us+: _raw_spin_lock_irq <-run_timer_softirq -0 0dNs3 17us : _raw_spin_unlock_irq <-run_timer_softirq -0 0dNs3 17us+: trace_hardirqs_on <-run_timer_softirq -0 0dNs3 25us : => _raw_spin_unlock_irq => run_timer_softirq => __do_softirq => call_softirq => do_softirq => irq_exit => smp_apic_timer_interrupt => apic_timer_interrupt => rcu_idle_exit => cpu_idle => rest_init => start_kernel => x86_64_start_reservations => x86_64_start_kernelh]hX# echo 0 > options/function-trace # echo irqsoff > current_tracer # echo 1 > tracing_on # echo 0 > tracing_max_latency # ls -ltr [...] # echo 0 > tracing_on # cat trace # tracer: irqsoff # # irqsoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 16 us, #4/4, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: swapper/0-0 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: run_timer_softirq # => ended at: run_timer_softirq # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 0d.s2 0us+: _raw_spin_lock_irq <-run_timer_softirq -0 0dNs3 17us : _raw_spin_unlock_irq <-run_timer_softirq -0 0dNs3 17us+: trace_hardirqs_on <-run_timer_softirq -0 0dNs3 25us : => _raw_spin_unlock_irq => run_timer_softirq => __do_softirq => call_softirq => do_softirq => irq_exit => smp_apic_timer_interrupt => apic_timer_interrupt => rcu_idle_exit => cpu_idle => rest_init => start_kernel => x86_64_start_reservations => x86_64_start_kernel}hjr$sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj7$hhubh)}(hXaHere we see that we had a latency of 16 microseconds (which is very good). The _raw_spin_lock_irq in run_timer_softirq disabled interrupts. The difference between the 16 and the displayed timestamp 25us occurred because the clock was incremented between the time of recording the max latency and the time of recording the function that had that latency.h]hXaHere we see that we had a latency of 16 microseconds (which is very good). The _raw_spin_lock_irq in run_timer_softirq disabled interrupts. The difference between the 16 and the displayed timestamp 25us occurred because the clock was incremented between the time of recording the max latency and the time of recording the function that had that latency.}(hj$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj7$hhubh)}(hjNote the above example had function-trace not set. If we set function-trace, we get a much larger output::h]hiNote the above example had function-trace not set. If we set function-trace, we get a much larger output:}(hj$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj7$hhubj)}(hX with echo 1 > options/function-trace # tracer: irqsoff # # irqsoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 71 us, #168/168, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: bash-2042 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: ata_scsi_queuecmd # => ended at: ata_scsi_queuecmd # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / bash-2042 3d... 0us : _raw_spin_lock_irqsave <-ata_scsi_queuecmd bash-2042 3d... 0us : add_preempt_count <-_raw_spin_lock_irqsave bash-2042 3d..1 1us : ata_scsi_find_dev <-ata_scsi_queuecmd bash-2042 3d..1 1us : __ata_scsi_find_dev <-ata_scsi_find_dev bash-2042 3d..1 2us : ata_find_dev.part.14 <-__ata_scsi_find_dev bash-2042 3d..1 2us : ata_qc_new_init <-__ata_scsi_queuecmd bash-2042 3d..1 3us : ata_sg_init <-__ata_scsi_queuecmd bash-2042 3d..1 4us : ata_scsi_rw_xlat <-__ata_scsi_queuecmd bash-2042 3d..1 4us : ata_build_rw_tf <-ata_scsi_rw_xlat [...] bash-2042 3d..1 67us : delay_tsc <-__delay bash-2042 3d..1 67us : add_preempt_count <-delay_tsc bash-2042 3d..2 67us : sub_preempt_count <-delay_tsc bash-2042 3d..1 67us : add_preempt_count <-delay_tsc bash-2042 3d..2 68us : sub_preempt_count <-delay_tsc bash-2042 3d..1 68us+: ata_bmdma_start <-ata_bmdma_qc_issue bash-2042 3d..1 71us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd bash-2042 3d..1 71us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd bash-2042 3d..1 72us+: trace_hardirqs_on <-ata_scsi_queuecmd bash-2042 3d..1 120us : => _raw_spin_unlock_irqrestore => ata_scsi_queuecmd => scsi_dispatch_cmd => scsi_request_fn => __blk_run_queue_uncond => __blk_run_queue => blk_queue_bio => submit_bio_noacct => submit_bio => submit_bh => __ext3_get_inode_loc => ext3_iget => ext3_lookup => lookup_real => __lookup_hash => walk_component => lookup_last => path_lookupat => filename_lookup => user_path_at_empty => user_path_at => vfs_fstatat => vfs_stat => sys_newstat => system_call_fastpathh]hX with echo 1 > options/function-trace # tracer: irqsoff # # irqsoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 71 us, #168/168, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: bash-2042 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: ata_scsi_queuecmd # => ended at: ata_scsi_queuecmd # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / bash-2042 3d... 0us : _raw_spin_lock_irqsave <-ata_scsi_queuecmd bash-2042 3d... 0us : add_preempt_count <-_raw_spin_lock_irqsave bash-2042 3d..1 1us : ata_scsi_find_dev <-ata_scsi_queuecmd bash-2042 3d..1 1us : __ata_scsi_find_dev <-ata_scsi_find_dev bash-2042 3d..1 2us : ata_find_dev.part.14 <-__ata_scsi_find_dev bash-2042 3d..1 2us : ata_qc_new_init <-__ata_scsi_queuecmd bash-2042 3d..1 3us : ata_sg_init <-__ata_scsi_queuecmd bash-2042 3d..1 4us : ata_scsi_rw_xlat <-__ata_scsi_queuecmd bash-2042 3d..1 4us : ata_build_rw_tf <-ata_scsi_rw_xlat [...] bash-2042 3d..1 67us : delay_tsc <-__delay bash-2042 3d..1 67us : add_preempt_count <-delay_tsc bash-2042 3d..2 67us : sub_preempt_count <-delay_tsc bash-2042 3d..1 67us : add_preempt_count <-delay_tsc bash-2042 3d..2 68us : sub_preempt_count <-delay_tsc bash-2042 3d..1 68us+: ata_bmdma_start <-ata_bmdma_qc_issue bash-2042 3d..1 71us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd bash-2042 3d..1 71us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd bash-2042 3d..1 72us+: trace_hardirqs_on <-ata_scsi_queuecmd bash-2042 3d..1 120us : => _raw_spin_unlock_irqrestore => ata_scsi_queuecmd => scsi_dispatch_cmd => scsi_request_fn => __blk_run_queue_uncond => __blk_run_queue => blk_queue_bio => submit_bio_noacct => submit_bio => submit_bh => __ext3_get_inode_loc => ext3_iget => ext3_lookup => lookup_real => __lookup_hash => walk_component => lookup_last => path_lookupat => filename_lookup => user_path_at_empty => user_path_at => vfs_fstatat => vfs_stat => sys_newstat => system_call_fastpath}hj$sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj7$hhubh)}(hX0Here we traced a 71 microsecond latency. But we also see all the functions that were called during that time. Note that by enabling function tracing, we incur an added overhead. This overhead may extend the latency times. But nevertheless, this trace has provided some very helpful debugging information.h]hX0Here we traced a 71 microsecond latency. But we also see all the functions that were called during that time. Note that by enabling function tracing, we incur an added overhead. This overhead may extend the latency times. But nevertheless, this trace has provided some very helpful debugging information.}(hj$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM?hj7$hhubh)}(hYIf we prefer function graph output instead of function, we can set display-graph option::h]hXIf we prefer function graph output instead of function, we can set display-graph option:}(hj$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMEhj7$hhubj)}(hX with echo 1 > options/display-graph # tracer: irqsoff # # irqsoff latency trace v1.1.5 on 4.20.0-rc6+ # -------------------------------------------------------------------- # latency: 3751 us, #274/274, CPU#0 | (M:desktop VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: bash-1507 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: free_debug_processing # => ended at: return_to_handler # # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / # REL TIME CPU TASK/PID |||| DURATION FUNCTION CALLS # | | | | |||| | | | | | | 0 us | 0) bash-1507 | d... | 0.000 us | _raw_spin_lock_irqsave(); 0 us | 0) bash-1507 | d..1 | 0.378 us | do_raw_spin_trylock(); 1 us | 0) bash-1507 | d..2 | | set_track() { 2 us | 0) bash-1507 | d..2 | | save_stack_trace() { 2 us | 0) bash-1507 | d..2 | | __save_stack_trace() { 3 us | 0) bash-1507 | d..2 | | __unwind_start() { 3 us | 0) bash-1507 | d..2 | | get_stack_info() { 3 us | 0) bash-1507 | d..2 | 0.351 us | in_task_stack(); 4 us | 0) bash-1507 | d..2 | 1.107 us | } [...] 3750 us | 0) bash-1507 | d..1 | 0.516 us | do_raw_spin_unlock(); 3750 us | 0) bash-1507 | d..1 | 0.000 us | _raw_spin_unlock_irqrestore(); 3764 us | 0) bash-1507 | d..1 | 0.000 us | tracer_hardirqs_on(); bash-1507 0d..1 3792us : => free_debug_processing => __slab_free => kmem_cache_free => vm_area_free => remove_vma => exit_mmap => mmput => begin_new_exec => load_elf_binary => search_binary_handler => __do_execve_file.isra.32 => __x64_sys_execve => do_syscall_64 => entry_SYSCALL_64_after_hwframeh]hX with echo 1 > options/display-graph # tracer: irqsoff # # irqsoff latency trace v1.1.5 on 4.20.0-rc6+ # -------------------------------------------------------------------- # latency: 3751 us, #274/274, CPU#0 | (M:desktop VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: bash-1507 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: free_debug_processing # => ended at: return_to_handler # # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / # REL TIME CPU TASK/PID |||| DURATION FUNCTION CALLS # | | | | |||| | | | | | | 0 us | 0) bash-1507 | d... | 0.000 us | _raw_spin_lock_irqsave(); 0 us | 0) bash-1507 | d..1 | 0.378 us | do_raw_spin_trylock(); 1 us | 0) bash-1507 | d..2 | | set_track() { 2 us | 0) bash-1507 | d..2 | | save_stack_trace() { 2 us | 0) bash-1507 | d..2 | | __save_stack_trace() { 3 us | 0) bash-1507 | d..2 | | __unwind_start() { 3 us | 0) bash-1507 | d..2 | | get_stack_info() { 3 us | 0) bash-1507 | d..2 | 0.351 us | in_task_stack(); 4 us | 0) bash-1507 | d..2 | 1.107 us | } [...] 3750 us | 0) bash-1507 | d..1 | 0.516 us | do_raw_spin_unlock(); 3750 us | 0) bash-1507 | d..1 | 0.000 us | _raw_spin_unlock_irqrestore(); 3764 us | 0) bash-1507 | d..1 | 0.000 us | tracer_hardirqs_on(); bash-1507 0d..1 3792us : => free_debug_processing => __slab_free => kmem_cache_free => vm_area_free => remove_vma => exit_mmap => mmput => begin_new_exec => load_elf_binary => search_binary_handler => __do_execve_file.isra.32 => __x64_sys_execve => do_syscall_64 => entry_SYSCALL_64_after_hwframe}hj$sbah}(h]h ]h"]h$]h&]jjuh1jhhhMHhj7$hhubeh}(h]irqsoffah ]h"]irqsoffah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(h preemptoffh]h preemptoff}(hj$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj$hhhhhM{ubh)}(hWhen preemption is disabled, we may be able to receive interrupts but the task cannot be preempted and a higher priority task must wait for preemption to be enabled again before it can preempt a lower priority task.h]hWhen preemption is disabled, we may be able to receive interrupts but the task cannot be preempted and a higher priority task must wait for preemption to be enabled again before it can preempt a lower priority task.}(hj$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM}hj$hhubh)}(hThe preemptoff tracer traces the places that disable preemption. Like the irqsoff tracer, it records the maximum latency for which preemption was disabled. The control of preemptoff tracer is much like the irqsoff tracer. ::h]hThe preemptoff tracer traces the places that disable preemption. Like the irqsoff tracer, it records the maximum latency for which preemption was disabled. The control of preemptoff tracer is much like the irqsoff tracer.}(hj$hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj$hhubj)}(hXb# echo 0 > options/function-trace # echo preemptoff > current_tracer # echo 1 > tracing_on # echo 0 > tracing_max_latency # ls -ltr [...] # echo 0 > tracing_on # cat trace # tracer: preemptoff # # preemptoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 46 us, #4/4, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: sshd-1991 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: do_IRQ # => ended at: do_IRQ # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / sshd-1991 1d.h. 0us+: irq_enter <-do_IRQ sshd-1991 1d..1 46us : irq_exit <-do_IRQ sshd-1991 1d..1 47us+: trace_preempt_on <-do_IRQ sshd-1991 1d..1 52us : => sub_preempt_count => irq_exit => do_IRQ => ret_from_intrh]hXb# echo 0 > options/function-trace # echo preemptoff > current_tracer # echo 1 > tracing_on # echo 0 > tracing_max_latency # ls -ltr [...] # echo 0 > tracing_on # cat trace # tracer: preemptoff # # preemptoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 46 us, #4/4, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: sshd-1991 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: do_IRQ # => ended at: do_IRQ # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / sshd-1991 1d.h. 0us+: irq_enter <-do_IRQ sshd-1991 1d..1 46us : irq_exit <-do_IRQ sshd-1991 1d..1 47us+: trace_preempt_on <-do_IRQ sshd-1991 1d..1 52us : => sub_preempt_count => irq_exit => do_IRQ => ret_from_intr}hj %sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj$hhubh)}(hXIThis has some more changes. Preemption was disabled when an interrupt came in (notice the 'h'), and was enabled on exit. But we also see that interrupts have been disabled when entering the preempt off section and leaving it (the 'd'). We do not know if interrupts were enabled in the mean time or shortly after this was over. ::h]hXNThis has some more changes. Preemption was disabled when an interrupt came in (notice the ‘h’), and was enabled on exit. But we also see that interrupts have been disabled when entering the preempt off section and leaving it (the ‘d’). We do not know if interrupts were enabled in the mean time or shortly after this was over.}(hj%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj$hhubj)}(hX # tracer: preemptoff # # preemptoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 83 us, #241/241, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: bash-1994 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: wake_up_new_task # => ended at: task_rq_unlock # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / bash-1994 1d..1 0us : _raw_spin_lock_irqsave <-wake_up_new_task bash-1994 1d..1 0us : select_task_rq_fair <-select_task_rq bash-1994 1d..1 1us : __rcu_read_lock <-select_task_rq_fair bash-1994 1d..1 1us : source_load <-select_task_rq_fair bash-1994 1d..1 1us : source_load <-select_task_rq_fair [...] bash-1994 1d..1 12us : irq_enter <-smp_apic_timer_interrupt bash-1994 1d..1 12us : rcu_irq_enter <-irq_enter bash-1994 1d..1 13us : add_preempt_count <-irq_enter bash-1994 1d.h1 13us : exit_idle <-smp_apic_timer_interrupt bash-1994 1d.h1 13us : hrtimer_interrupt <-smp_apic_timer_interrupt bash-1994 1d.h1 13us : _raw_spin_lock <-hrtimer_interrupt bash-1994 1d.h1 14us : add_preempt_count <-_raw_spin_lock bash-1994 1d.h2 14us : ktime_get_update_offsets <-hrtimer_interrupt [...] bash-1994 1d.h1 35us : lapic_next_event <-clockevents_program_event bash-1994 1d.h1 35us : irq_exit <-smp_apic_timer_interrupt bash-1994 1d.h1 36us : sub_preempt_count <-irq_exit bash-1994 1d..2 36us : do_softirq <-irq_exit bash-1994 1d..2 36us : __do_softirq <-call_softirq bash-1994 1d..2 36us : __local_bh_disable <-__do_softirq bash-1994 1d.s2 37us : add_preempt_count <-_raw_spin_lock_irq bash-1994 1d.s3 38us : _raw_spin_unlock <-run_timer_softirq bash-1994 1d.s3 39us : sub_preempt_count <-_raw_spin_unlock bash-1994 1d.s2 39us : call_timer_fn <-run_timer_softirq [...] bash-1994 1dNs2 81us : cpu_needs_another_gp <-rcu_process_callbacks bash-1994 1dNs2 82us : __local_bh_enable <-__do_softirq bash-1994 1dNs2 82us : sub_preempt_count <-__local_bh_enable bash-1994 1dN.2 82us : idle_cpu <-irq_exit bash-1994 1dN.2 83us : rcu_irq_exit <-irq_exit bash-1994 1dN.2 83us : sub_preempt_count <-irq_exit bash-1994 1.N.1 84us : _raw_spin_unlock_irqrestore <-task_rq_unlock bash-1994 1.N.1 84us+: trace_preempt_on <-task_rq_unlock bash-1994 1.N.1 104us : => sub_preempt_count => _raw_spin_unlock_irqrestore => task_rq_unlock => wake_up_new_task => do_fork => sys_clone => stub_cloneh]hX # tracer: preemptoff # # preemptoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 83 us, #241/241, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: bash-1994 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: wake_up_new_task # => ended at: task_rq_unlock # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / bash-1994 1d..1 0us : _raw_spin_lock_irqsave <-wake_up_new_task bash-1994 1d..1 0us : select_task_rq_fair <-select_task_rq bash-1994 1d..1 1us : __rcu_read_lock <-select_task_rq_fair bash-1994 1d..1 1us : source_load <-select_task_rq_fair bash-1994 1d..1 1us : source_load <-select_task_rq_fair [...] bash-1994 1d..1 12us : irq_enter <-smp_apic_timer_interrupt bash-1994 1d..1 12us : rcu_irq_enter <-irq_enter bash-1994 1d..1 13us : add_preempt_count <-irq_enter bash-1994 1d.h1 13us : exit_idle <-smp_apic_timer_interrupt bash-1994 1d.h1 13us : hrtimer_interrupt <-smp_apic_timer_interrupt bash-1994 1d.h1 13us : _raw_spin_lock <-hrtimer_interrupt bash-1994 1d.h1 14us : add_preempt_count <-_raw_spin_lock bash-1994 1d.h2 14us : ktime_get_update_offsets <-hrtimer_interrupt [...] bash-1994 1d.h1 35us : lapic_next_event <-clockevents_program_event bash-1994 1d.h1 35us : irq_exit <-smp_apic_timer_interrupt bash-1994 1d.h1 36us : sub_preempt_count <-irq_exit bash-1994 1d..2 36us : do_softirq <-irq_exit bash-1994 1d..2 36us : __do_softirq <-call_softirq bash-1994 1d..2 36us : __local_bh_disable <-__do_softirq bash-1994 1d.s2 37us : add_preempt_count <-_raw_spin_lock_irq bash-1994 1d.s3 38us : _raw_spin_unlock <-run_timer_softirq bash-1994 1d.s3 39us : sub_preempt_count <-_raw_spin_unlock bash-1994 1d.s2 39us : call_timer_fn <-run_timer_softirq [...] bash-1994 1dNs2 81us : cpu_needs_another_gp <-rcu_process_callbacks bash-1994 1dNs2 82us : __local_bh_enable <-__do_softirq bash-1994 1dNs2 82us : sub_preempt_count <-__local_bh_enable bash-1994 1dN.2 82us : idle_cpu <-irq_exit bash-1994 1dN.2 83us : rcu_irq_exit <-irq_exit bash-1994 1dN.2 83us : sub_preempt_count <-irq_exit bash-1994 1.N.1 84us : _raw_spin_unlock_irqrestore <-task_rq_unlock bash-1994 1.N.1 84us+: trace_preempt_on <-task_rq_unlock bash-1994 1.N.1 104us : => sub_preempt_count => _raw_spin_unlock_irqrestore => task_rq_unlock => wake_up_new_task => do_fork => sys_clone => stub_clone}hj%%sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj$hhubh)}(hXdThe above is an example of the preemptoff trace with function-trace set. Here we see that interrupts were not disabled the entire time. The irq_enter code lets us know that we entered an interrupt 'h'. Before that, the functions being traced still show that it is not in an interrupt, but we can see from the functions themselves that this is not the case.h]hXhThe above is an example of the preemptoff trace with function-trace set. Here we see that interrupts were not disabled the entire time. The irq_enter code lets us know that we entered an interrupt ‘h’. Before that, the functions being traced still show that it is not in an interrupt, but we can see from the functions themselves that this is not the case.}(hj3%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj$hhubeh}(h] preemptoffah ]h"] preemptoffah$]h&]uh1hhhhhhhhM{ubh)}(hhh](h)}(hpreemptirqsoffh]hpreemptirqsoff}(hjL%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjI%hhhhhMubh)}(hKnowing the locations that have interrupts disabled or preemption disabled for the longest times is helpful. But sometimes we would like to know when either preemption and/or interrupts are disabled.h]hKnowing the locations that have interrupts disabled or preemption disabled for the longest times is helpful. But sometimes we would like to know when either preemption and/or interrupts are disabled.}(hjZ%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjI%hhubh)}(hConsider the following code::h]hConsider the following code:}(hjh%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjI%hhubj)}(hlocal_irq_disable(); call_function_with_irqs_off(); preempt_disable(); call_function_with_irqs_and_preemption_off(); local_irq_enable(); call_function_with_preemption_off(); preempt_enable();h]hlocal_irq_disable(); call_function_with_irqs_off(); preempt_disable(); call_function_with_irqs_and_preemption_off(); local_irq_enable(); call_function_with_preemption_off(); preempt_enable();}hjv%sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhjI%hhubh)}(hThe irqsoff tracer will record the total length of call_function_with_irqs_off() and call_function_with_irqs_and_preemption_off().h]hThe irqsoff tracer will record the total length of call_function_with_irqs_off() and call_function_with_irqs_and_preemption_off().}(hj%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjI%hhubh)}(hThe preemptoff tracer will record the total length of call_function_with_irqs_and_preemption_off() and call_function_with_preemption_off().h]hThe preemptoff tracer will record the total length of call_function_with_irqs_and_preemption_off() and call_function_with_preemption_off().}(hj%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjI%hhubh)}(hBut neither will trace the time that interrupts and/or preemption is disabled. This total time is the time that we can not schedule. To record this time, use the preemptirqsoff tracer.h]hBut neither will trace the time that interrupts and/or preemption is disabled. This total time is the time that we can not schedule. To record this time, use the preemptirqsoff tracer.}(hj%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjI%hhubh)}(hKAgain, using this trace is much like the irqsoff and preemptoff tracers. ::h]hHAgain, using this trace is much like the irqsoff and preemptoff tracers.}(hj%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjI%hhubj)}(hX# echo 0 > options/function-trace # echo preemptirqsoff > current_tracer # echo 1 > tracing_on # echo 0 > tracing_max_latency # ls -ltr [...] # echo 0 > tracing_on # cat trace # tracer: preemptirqsoff # # preemptirqsoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 100 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: ls-2230 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: ata_scsi_queuecmd # => ended at: ata_scsi_queuecmd # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / ls-2230 3d... 0us+: _raw_spin_lock_irqsave <-ata_scsi_queuecmd ls-2230 3...1 100us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd ls-2230 3...1 101us+: trace_preempt_on <-ata_scsi_queuecmd ls-2230 3...1 111us : => sub_preempt_count => _raw_spin_unlock_irqrestore => ata_scsi_queuecmd => scsi_dispatch_cmd => scsi_request_fn => __blk_run_queue_uncond => __blk_run_queue => blk_queue_bio => submit_bio_noacct => submit_bio => submit_bh => ext3_bread => ext3_dir_bread => htree_dirblock_to_tree => ext3_htree_fill_tree => ext3_readdir => vfs_readdir => sys_getdents => system_call_fastpathh]hX# echo 0 > options/function-trace # echo preemptirqsoff > current_tracer # echo 1 > tracing_on # echo 0 > tracing_max_latency # ls -ltr [...] # echo 0 > tracing_on # cat trace # tracer: preemptirqsoff # # preemptirqsoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 100 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: ls-2230 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: ata_scsi_queuecmd # => ended at: ata_scsi_queuecmd # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / ls-2230 3d... 0us+: _raw_spin_lock_irqsave <-ata_scsi_queuecmd ls-2230 3...1 100us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd ls-2230 3...1 101us+: trace_preempt_on <-ata_scsi_queuecmd ls-2230 3...1 111us : => sub_preempt_count => _raw_spin_unlock_irqrestore => ata_scsi_queuecmd => scsi_dispatch_cmd => scsi_request_fn => __blk_run_queue_uncond => __blk_run_queue => blk_queue_bio => submit_bio_noacct => submit_bio => submit_bh => ext3_bread => ext3_dir_bread => htree_dirblock_to_tree => ext3_htree_fill_tree => ext3_readdir => vfs_readdir => sys_getdents => system_call_fastpath}hj%sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hjI%hhubh)}(hXThe trace_hardirqs_off_thunk is called from assembly on x86 when interrupts are disabled in the assembly code. Without the function tracing, we do not know if interrupts were enabled within the preemption points. We do see that it started with preemption enabled.h]hXThe trace_hardirqs_off_thunk is called from assembly on x86 when interrupts are disabled in the assembly code. Without the function tracing, we do not know if interrupts were enabled within the preemption points. We do see that it started with preemption enabled.}(hj%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMUhjI%hhubh)}(h)Here is a trace with function-trace set::h]h(Here is a trace with function-trace set:}(hj%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM[hjI%hhubj)}(hX# tracer: preemptirqsoff # # preemptirqsoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 161 us, #339/339, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: ls-2269 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: schedule # => ended at: mutex_unlock # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / kworker/-59 3...1 0us : __schedule <-schedule kworker/-59 3d..1 0us : rcu_preempt_qs <-rcu_note_context_switch kworker/-59 3d..1 1us : add_preempt_count <-_raw_spin_lock_irq kworker/-59 3d..2 1us : deactivate_task <-__schedule kworker/-59 3d..2 1us : dequeue_task <-deactivate_task kworker/-59 3d..2 2us : update_rq_clock <-dequeue_task kworker/-59 3d..2 2us : dequeue_task_fair <-dequeue_task kworker/-59 3d..2 2us : update_curr <-dequeue_task_fair kworker/-59 3d..2 2us : update_min_vruntime <-update_curr kworker/-59 3d..2 3us : cpuacct_charge <-update_curr kworker/-59 3d..2 3us : __rcu_read_lock <-cpuacct_charge kworker/-59 3d..2 3us : __rcu_read_unlock <-cpuacct_charge kworker/-59 3d..2 3us : update_cfs_rq_blocked_load <-dequeue_task_fair kworker/-59 3d..2 4us : clear_buddies <-dequeue_task_fair kworker/-59 3d..2 4us : account_entity_dequeue <-dequeue_task_fair kworker/-59 3d..2 4us : update_min_vruntime <-dequeue_task_fair kworker/-59 3d..2 4us : update_cfs_shares <-dequeue_task_fair kworker/-59 3d..2 5us : hrtick_update <-dequeue_task_fair kworker/-59 3d..2 5us : wq_worker_sleeping <-__schedule kworker/-59 3d..2 5us : kthread_data <-wq_worker_sleeping kworker/-59 3d..2 5us : put_prev_task_fair <-__schedule kworker/-59 3d..2 6us : pick_next_task_fair <-pick_next_task kworker/-59 3d..2 6us : clear_buddies <-pick_next_task_fair kworker/-59 3d..2 6us : set_next_entity <-pick_next_task_fair kworker/-59 3d..2 6us : update_stats_wait_end <-set_next_entity ls-2269 3d..2 7us : finish_task_switch <-__schedule ls-2269 3d..2 7us : _raw_spin_unlock_irq <-finish_task_switch ls-2269 3d..2 8us : do_IRQ <-ret_from_intr ls-2269 3d..2 8us : irq_enter <-do_IRQ ls-2269 3d..2 8us : rcu_irq_enter <-irq_enter ls-2269 3d..2 9us : add_preempt_count <-irq_enter ls-2269 3d.h2 9us : exit_idle <-do_IRQ [...] ls-2269 3d.h3 20us : sub_preempt_count <-_raw_spin_unlock ls-2269 3d.h2 20us : irq_exit <-do_IRQ ls-2269 3d.h2 21us : sub_preempt_count <-irq_exit ls-2269 3d..3 21us : do_softirq <-irq_exit ls-2269 3d..3 21us : __do_softirq <-call_softirq ls-2269 3d..3 21us+: __local_bh_disable <-__do_softirq ls-2269 3d.s4 29us : sub_preempt_count <-_local_bh_enable_ip ls-2269 3d.s5 29us : sub_preempt_count <-_local_bh_enable_ip ls-2269 3d.s5 31us : do_IRQ <-ret_from_intr ls-2269 3d.s5 31us : irq_enter <-do_IRQ ls-2269 3d.s5 31us : rcu_irq_enter <-irq_enter [...] ls-2269 3d.s5 31us : rcu_irq_enter <-irq_enter ls-2269 3d.s5 32us : add_preempt_count <-irq_enter ls-2269 3d.H5 32us : exit_idle <-do_IRQ ls-2269 3d.H5 32us : handle_irq <-do_IRQ ls-2269 3d.H5 32us : irq_to_desc <-handle_irq ls-2269 3d.H5 33us : handle_fasteoi_irq <-handle_irq [...] ls-2269 3d.s5 158us : _raw_spin_unlock_irqrestore <-rtl8139_poll ls-2269 3d.s3 158us : net_rps_action_and_irq_enable.isra.65 <-net_rx_action ls-2269 3d.s3 159us : __local_bh_enable <-__do_softirq ls-2269 3d.s3 159us : sub_preempt_count <-__local_bh_enable ls-2269 3d..3 159us : idle_cpu <-irq_exit ls-2269 3d..3 159us : rcu_irq_exit <-irq_exit ls-2269 3d..3 160us : sub_preempt_count <-irq_exit ls-2269 3d... 161us : __mutex_unlock_slowpath <-mutex_unlock ls-2269 3d... 162us+: trace_hardirqs_on <-mutex_unlock ls-2269 3d... 186us : => __mutex_unlock_slowpath => mutex_unlock => process_output => n_tty_write => tty_write => vfs_write => sys_write => system_call_fastpathh]hX# tracer: preemptirqsoff # # preemptirqsoff latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 161 us, #339/339, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: ls-2269 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: schedule # => ended at: mutex_unlock # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / kworker/-59 3...1 0us : __schedule <-schedule kworker/-59 3d..1 0us : rcu_preempt_qs <-rcu_note_context_switch kworker/-59 3d..1 1us : add_preempt_count <-_raw_spin_lock_irq kworker/-59 3d..2 1us : deactivate_task <-__schedule kworker/-59 3d..2 1us : dequeue_task <-deactivate_task kworker/-59 3d..2 2us : update_rq_clock <-dequeue_task kworker/-59 3d..2 2us : dequeue_task_fair <-dequeue_task kworker/-59 3d..2 2us : update_curr <-dequeue_task_fair kworker/-59 3d..2 2us : update_min_vruntime <-update_curr kworker/-59 3d..2 3us : cpuacct_charge <-update_curr kworker/-59 3d..2 3us : __rcu_read_lock <-cpuacct_charge kworker/-59 3d..2 3us : __rcu_read_unlock <-cpuacct_charge kworker/-59 3d..2 3us : update_cfs_rq_blocked_load <-dequeue_task_fair kworker/-59 3d..2 4us : clear_buddies <-dequeue_task_fair kworker/-59 3d..2 4us : account_entity_dequeue <-dequeue_task_fair kworker/-59 3d..2 4us : update_min_vruntime <-dequeue_task_fair kworker/-59 3d..2 4us : update_cfs_shares <-dequeue_task_fair kworker/-59 3d..2 5us : hrtick_update <-dequeue_task_fair kworker/-59 3d..2 5us : wq_worker_sleeping <-__schedule kworker/-59 3d..2 5us : kthread_data <-wq_worker_sleeping kworker/-59 3d..2 5us : put_prev_task_fair <-__schedule kworker/-59 3d..2 6us : pick_next_task_fair <-pick_next_task kworker/-59 3d..2 6us : clear_buddies <-pick_next_task_fair kworker/-59 3d..2 6us : set_next_entity <-pick_next_task_fair kworker/-59 3d..2 6us : update_stats_wait_end <-set_next_entity ls-2269 3d..2 7us : finish_task_switch <-__schedule ls-2269 3d..2 7us : _raw_spin_unlock_irq <-finish_task_switch ls-2269 3d..2 8us : do_IRQ <-ret_from_intr ls-2269 3d..2 8us : irq_enter <-do_IRQ ls-2269 3d..2 8us : rcu_irq_enter <-irq_enter ls-2269 3d..2 9us : add_preempt_count <-irq_enter ls-2269 3d.h2 9us : exit_idle <-do_IRQ [...] ls-2269 3d.h3 20us : sub_preempt_count <-_raw_spin_unlock ls-2269 3d.h2 20us : irq_exit <-do_IRQ ls-2269 3d.h2 21us : sub_preempt_count <-irq_exit ls-2269 3d..3 21us : do_softirq <-irq_exit ls-2269 3d..3 21us : __do_softirq <-call_softirq ls-2269 3d..3 21us+: __local_bh_disable <-__do_softirq ls-2269 3d.s4 29us : sub_preempt_count <-_local_bh_enable_ip ls-2269 3d.s5 29us : sub_preempt_count <-_local_bh_enable_ip ls-2269 3d.s5 31us : do_IRQ <-ret_from_intr ls-2269 3d.s5 31us : irq_enter <-do_IRQ ls-2269 3d.s5 31us : rcu_irq_enter <-irq_enter [...] ls-2269 3d.s5 31us : rcu_irq_enter <-irq_enter ls-2269 3d.s5 32us : add_preempt_count <-irq_enter ls-2269 3d.H5 32us : exit_idle <-do_IRQ ls-2269 3d.H5 32us : handle_irq <-do_IRQ ls-2269 3d.H5 32us : irq_to_desc <-handle_irq ls-2269 3d.H5 33us : handle_fasteoi_irq <-handle_irq [...] ls-2269 3d.s5 158us : _raw_spin_unlock_irqrestore <-rtl8139_poll ls-2269 3d.s3 158us : net_rps_action_and_irq_enable.isra.65 <-net_rx_action ls-2269 3d.s3 159us : __local_bh_enable <-__do_softirq ls-2269 3d.s3 159us : sub_preempt_count <-__local_bh_enable ls-2269 3d..3 159us : idle_cpu <-irq_exit ls-2269 3d..3 159us : rcu_irq_exit <-irq_exit ls-2269 3d..3 160us : sub_preempt_count <-irq_exit ls-2269 3d... 161us : __mutex_unlock_slowpath <-mutex_unlock ls-2269 3d... 162us+: trace_hardirqs_on <-mutex_unlock ls-2269 3d... 186us : => __mutex_unlock_slowpath => mutex_unlock => process_output => n_tty_write => tty_write => vfs_write => sys_write => system_call_fastpath}hj%sbah}(h]h ]h"]h$]h&]jjuh1jhhhM]hjI%hhubh)}(hXThis is an interesting trace. It started with kworker running and scheduling out and ls taking over. But as soon as ls released the rq lock and enabled interrupts (but not preemption) an interrupt triggered. When the interrupt finished, it started running softirqs. But while the softirq was running, another interrupt triggered. When an interrupt is running inside a softirq, the annotation is 'H'.h]hXThis is an interesting trace. It started with kworker running and scheduling out and ls taking over. But as soon as ls released the rq lock and enabled interrupts (but not preemption) an interrupt triggered. When the interrupt finished, it started running softirqs. But while the softirq was running, another interrupt triggered. When an interrupt is running inside a softirq, the annotation is ‘H’.}(hj%hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjI%hhubeh}(h]preemptirqsoffah ]h"]preemptirqsoffah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(hwakeuph]hwakeup}(hj &hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj &hhhhhMubh)}(hOne common case that people are interested in tracing is the time it takes for a task that is woken to actually wake up. Now for non Real-Time tasks, this can be arbitrary. But tracing it nonetheless can be interesting.h]hOne common case that people are interested in tracing is the time it takes for a task that is woken to actually wake up. Now for non Real-Time tasks, this can be arbitrary. But tracing it nonetheless can be interesting.}(hj&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj &hhubh)}(hWithout function tracing::h]hWithout function tracing:}(hj)&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj &hhubj)}(hXG# echo 0 > options/function-trace # echo wakeup > current_tracer # echo 1 > tracing_on # echo 0 > tracing_max_latency # chrt -f 5 sleep 1 # echo 0 > tracing_on # cat trace # tracer: wakeup # # wakeup latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 15 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: kworker/3:1H-312 (uid:0 nice:-20 policy:0 rt_prio:0) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 3dNs7 0us : 0:120:R + [003] 312:100:R kworker/3:1H -0 3dNs7 1us+: ttwu_do_activate.constprop.87 <-try_to_wake_up -0 3d..3 15us : __schedule <-schedule -0 3d..3 15us : 0:120:R ==> [003] 312:100:R kworker/3:1Hh]hXG# echo 0 > options/function-trace # echo wakeup > current_tracer # echo 1 > tracing_on # echo 0 > tracing_max_latency # chrt -f 5 sleep 1 # echo 0 > tracing_on # cat trace # tracer: wakeup # # wakeup latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 15 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: kworker/3:1H-312 (uid:0 nice:-20 policy:0 rt_prio:0) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 3dNs7 0us : 0:120:R + [003] 312:100:R kworker/3:1H -0 3dNs7 1us+: ttwu_do_activate.constprop.87 <-try_to_wake_up -0 3d..3 15us : __schedule <-schedule -0 3d..3 15us : 0:120:R ==> [003] 312:100:R kworker/3:1H}hj7&sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj &hhubh)}(hThe tracer only traces the highest priority task in the system to avoid tracing the normal circumstances. Here we see that the kworker with a nice priority of -20 (not very nice), took just 15 microseconds from the time it woke up, to the time it ran.h]hThe tracer only traces the highest priority task in the system to avoid tracing the normal circumstances. Here we see that the kworker with a nice priority of -20 (not very nice), took just 15 microseconds from the time it woke up, to the time it ran.}(hjE&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj &hhubh)}(hqNon Real-Time tasks are not that interesting. A more interesting trace is to concentrate only on Real-Time tasks.h]hqNon Real-Time tasks are not that interesting. A more interesting trace is to concentrate only on Real-Time tasks.}(hjS&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj &hhubeh}(h]wakeupah ]h"]wakeupah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(h wakeup_rth]h wakeup_rt}(hjl&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhji&hhhhhMubh)}(hXIn a Real-Time environment it is very important to know the wakeup time it takes for the highest priority task that is woken up to the time that it executes. This is also known as "schedule latency". I stress the point that this is about RT tasks. It is also important to know the scheduling latency of non-RT tasks, but the average schedule latency is better for non-RT tasks. Tools like LatencyTop are more appropriate for such measurements.h]hXIn a Real-Time environment it is very important to know the wakeup time it takes for the highest priority task that is woken up to the time that it executes. This is also known as “schedule latency”. I stress the point that this is about RT tasks. It is also important to know the scheduling latency of non-RT tasks, but the average schedule latency is better for non-RT tasks. Tools like LatencyTop are more appropriate for such measurements.}(hjz&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhji&hhubh)}(hX`Real-Time environments are interested in the worst case latency. That is the longest latency it takes for something to happen, and not the average. We can have a very fast scheduler that may only have a large latency once in a while, but that would not work well with Real-Time tasks. The wakeup_rt tracer was designed to record the worst case wakeups of RT tasks. Non-RT tasks are not recorded because the tracer only records one worst case and tracing non-RT tasks that are unpredictable will overwrite the worst case latency of RT tasks (just run the normal wakeup tracer for a while to see that effect).h]hX`Real-Time environments are interested in the worst case latency. That is the longest latency it takes for something to happen, and not the average. We can have a very fast scheduler that may only have a large latency once in a while, but that would not work well with Real-Time tasks. The wakeup_rt tracer was designed to record the worst case wakeups of RT tasks. Non-RT tasks are not recorded because the tracer only records one worst case and tracing non-RT tasks that are unpredictable will overwrite the worst case latency of RT tasks (just run the normal wakeup tracer for a while to see that effect).}(hj&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhji&hhubh)}(hSince this tracer only deals with RT tasks, we will run this slightly differently than we did with the previous tracers. Instead of performing an 'ls', we will run 'sleep 1' under 'chrt' which changes the priority of the task. ::h]hSince this tracer only deals with RT tasks, we will run this slightly differently than we did with the previous tracers. Instead of performing an ‘ls’, we will run ‘sleep 1’ under ‘chrt’ which changes the priority of the task.}(hj&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhji&hhubj)}(hXL# echo 0 > options/function-trace # echo wakeup_rt > current_tracer # echo 1 > tracing_on # echo 0 > tracing_max_latency # chrt -f 5 sleep 1 # echo 0 > tracing_on # cat trace # tracer: wakeup # # tracer: wakeup_rt # # wakeup_rt latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 5 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: sleep-2389 (uid:0 nice:0 policy:1 rt_prio:5) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 3d.h4 0us : 0:120:R + [003] 2389: 94:R sleep -0 3d.h4 1us+: ttwu_do_activate.constprop.87 <-try_to_wake_up -0 3d..3 5us : __schedule <-schedule -0 3d..3 5us : 0:120:R ==> [003] 2389: 94:R sleeph]hXL# echo 0 > options/function-trace # echo wakeup_rt > current_tracer # echo 1 > tracing_on # echo 0 > tracing_max_latency # chrt -f 5 sleep 1 # echo 0 > tracing_on # cat trace # tracer: wakeup # # tracer: wakeup_rt # # wakeup_rt latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 5 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: sleep-2389 (uid:0 nice:0 policy:1 rt_prio:5) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 3d.h4 0us : 0:120:R + [003] 2389: 94:R sleep -0 3d.h4 1us+: ttwu_do_activate.constprop.87 <-try_to_wake_up -0 3d..3 5us : __schedule <-schedule -0 3d..3 5us : 0:120:R ==> [003] 2389: 94:R sleep}hj&sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hji&hhubh)}(hX7Running this on an idle system, we see that it only took 5 microseconds to perform the task switch. Note, since the trace point in the schedule is before the actual "switch", we stop the tracing when the recorded task is about to schedule in. This may change if we add a new marker at the end of the scheduler.h]hX;Running this on an idle system, we see that it only took 5 microseconds to perform the task switch. Note, since the trace point in the schedule is before the actual “switch”, we stop the tracing when the recorded task is about to schedule in. This may change if we add a new marker at the end of the scheduler.}(hj&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM-hji&hhubh)}(hNotice that the recorded task is 'sleep' with the PID of 2389 and it has an rt_prio of 5. This priority is user-space priority and not the internal kernel priority. The policy is 1 for SCHED_FIFO and 2 for SCHED_RR.h]hNotice that the recorded task is ‘sleep’ with the PID of 2389 and it has an rt_prio of 5. This priority is user-space priority and not the internal kernel priority. The policy is 1 for SCHED_FIFO and 2 for SCHED_RR.}(hj&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM3hji&hhubh)}(hGNote, that the trace data shows the internal priority (99 - rtprio). ::h]hDNote, that the trace data shows the internal priority (99 - rtprio).}(hj&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM8hji&hhubj)}(hF-0 3d..3 5us : 0:120:R ==> [003] 2389: 94:R sleeph]hF-0 3d..3 5us : 0:120:R ==> [003] 2389: 94:R sleep}hj&sbah}(h]h ]h"]h$]h&]jjuh1jhhhM;hji&hhubh)}(hThe 0:120:R means idle was running with a nice priority of 0 (120 - 120) and in the running state 'R'. The sleep task was scheduled in with 2389: 94:R. That is the priority is the kernel rtprio (99 - 5 = 94) and it too is in the running state.h]hThe 0:120:R means idle was running with a nice priority of 0 (120 - 120) and in the running state ‘R’. The sleep task was scheduled in with 2389: 94:R. That is the priority is the kernel rtprio (99 - 5 = 94) and it too is in the running state.}(hj&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM=hji&hhubh)}(h8Doing the same with chrt -r 5 and function-trace set. ::h]h5Doing the same with chrt -r 5 and function-trace set.}(hj&hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMBhji&hhubj)}(hXPecho 1 > options/function-trace # tracer: wakeup_rt # # wakeup_rt latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 29 us, #85/85, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: sleep-2448 (uid:0 nice:0 policy:1 rt_prio:5) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 3d.h4 1us+: 0:120:R + [003] 2448: 94:R sleep -0 3d.h4 2us : ttwu_do_activate.constprop.87 <-try_to_wake_up -0 3d.h3 3us : check_preempt_curr <-ttwu_do_wakeup -0 3d.h3 3us : resched_curr <-check_preempt_curr -0 3dNh3 4us : task_woken_rt <-ttwu_do_wakeup -0 3dNh3 4us : _raw_spin_unlock <-try_to_wake_up -0 3dNh3 4us : sub_preempt_count <-_raw_spin_unlock -0 3dNh2 5us : ttwu_stat <-try_to_wake_up -0 3dNh2 5us : _raw_spin_unlock_irqrestore <-try_to_wake_up -0 3dNh2 6us : sub_preempt_count <-_raw_spin_unlock_irqrestore -0 3dNh1 6us : _raw_spin_lock <-__run_hrtimer -0 3dNh1 6us : add_preempt_count <-_raw_spin_lock -0 3dNh2 7us : _raw_spin_unlock <-hrtimer_interrupt -0 3dNh2 7us : sub_preempt_count <-_raw_spin_unlock -0 3dNh1 7us : tick_program_event <-hrtimer_interrupt -0 3dNh1 7us : clockevents_program_event <-tick_program_event -0 3dNh1 8us : ktime_get <-clockevents_program_event -0 3dNh1 8us : lapic_next_event <-clockevents_program_event -0 3dNh1 8us : irq_exit <-smp_apic_timer_interrupt -0 3dNh1 9us : sub_preempt_count <-irq_exit -0 3dN.2 9us : idle_cpu <-irq_exit -0 3dN.2 9us : rcu_irq_exit <-irq_exit -0 3dN.2 10us : rcu_eqs_enter_common.isra.45 <-rcu_irq_exit -0 3dN.2 10us : sub_preempt_count <-irq_exit -0 3.N.1 11us : rcu_idle_exit <-cpu_idle -0 3dN.1 11us : rcu_eqs_exit_common.isra.43 <-rcu_idle_exit -0 3.N.1 11us : tick_nohz_idle_exit <-cpu_idle -0 3dN.1 12us : menu_hrtimer_cancel <-tick_nohz_idle_exit -0 3dN.1 12us : ktime_get <-tick_nohz_idle_exit -0 3dN.1 12us : tick_do_update_jiffies64 <-tick_nohz_idle_exit -0 3dN.1 13us : cpu_load_update_nohz <-tick_nohz_idle_exit -0 3dN.1 13us : _raw_spin_lock <-cpu_load_update_nohz -0 3dN.1 13us : add_preempt_count <-_raw_spin_lock -0 3dN.2 13us : __cpu_load_update <-cpu_load_update_nohz -0 3dN.2 14us : sched_avg_update <-__cpu_load_update -0 3dN.2 14us : _raw_spin_unlock <-cpu_load_update_nohz -0 3dN.2 14us : sub_preempt_count <-_raw_spin_unlock -0 3dN.1 15us : calc_load_nohz_stop <-tick_nohz_idle_exit -0 3dN.1 15us : touch_softlockup_watchdog <-tick_nohz_idle_exit -0 3dN.1 15us : hrtimer_cancel <-tick_nohz_idle_exit -0 3dN.1 15us : hrtimer_try_to_cancel <-hrtimer_cancel -0 3dN.1 16us : lock_hrtimer_base.isra.18 <-hrtimer_try_to_cancel -0 3dN.1 16us : _raw_spin_lock_irqsave <-lock_hrtimer_base.isra.18 -0 3dN.1 16us : add_preempt_count <-_raw_spin_lock_irqsave -0 3dN.2 17us : __remove_hrtimer <-remove_hrtimer.part.16 -0 3dN.2 17us : hrtimer_force_reprogram <-__remove_hrtimer -0 3dN.2 17us : tick_program_event <-hrtimer_force_reprogram -0 3dN.2 18us : clockevents_program_event <-tick_program_event -0 3dN.2 18us : ktime_get <-clockevents_program_event -0 3dN.2 18us : lapic_next_event <-clockevents_program_event -0 3dN.2 19us : _raw_spin_unlock_irqrestore <-hrtimer_try_to_cancel -0 3dN.2 19us : sub_preempt_count <-_raw_spin_unlock_irqrestore -0 3dN.1 19us : hrtimer_forward <-tick_nohz_idle_exit -0 3dN.1 20us : ktime_add_safe <-hrtimer_forward -0 3dN.1 20us : ktime_add_safe <-hrtimer_forward -0 3dN.1 20us : hrtimer_start_range_ns <-hrtimer_start_expires.constprop.11 -0 3dN.1 20us : __hrtimer_start_range_ns <-hrtimer_start_range_ns -0 3dN.1 21us : lock_hrtimer_base.isra.18 <-__hrtimer_start_range_ns -0 3dN.1 21us : _raw_spin_lock_irqsave <-lock_hrtimer_base.isra.18 -0 3dN.1 21us : add_preempt_count <-_raw_spin_lock_irqsave -0 3dN.2 22us : ktime_add_safe <-__hrtimer_start_range_ns -0 3dN.2 22us : enqueue_hrtimer <-__hrtimer_start_range_ns -0 3dN.2 22us : tick_program_event <-__hrtimer_start_range_ns -0 3dN.2 23us : clockevents_program_event <-tick_program_event -0 3dN.2 23us : ktime_get <-clockevents_program_event -0 3dN.2 23us : lapic_next_event <-clockevents_program_event -0 3dN.2 24us : _raw_spin_unlock_irqrestore <-__hrtimer_start_range_ns -0 3dN.2 24us : sub_preempt_count <-_raw_spin_unlock_irqrestore -0 3dN.1 24us : account_idle_ticks <-tick_nohz_idle_exit -0 3dN.1 24us : account_idle_time <-account_idle_ticks -0 3.N.1 25us : sub_preempt_count <-cpu_idle -0 3.N.. 25us : schedule <-cpu_idle -0 3.N.. 25us : __schedule <-preempt_schedule -0 3.N.. 26us : add_preempt_count <-__schedule -0 3.N.1 26us : rcu_note_context_switch <-__schedule -0 3.N.1 26us : rcu_sched_qs <-rcu_note_context_switch -0 3dN.1 27us : rcu_preempt_qs <-rcu_note_context_switch -0 3.N.1 27us : _raw_spin_lock_irq <-__schedule -0 3dN.1 27us : add_preempt_count <-_raw_spin_lock_irq -0 3dN.2 28us : put_prev_task_idle <-__schedule -0 3dN.2 28us : pick_next_task_stop <-pick_next_task -0 3dN.2 28us : pick_next_task_rt <-pick_next_task -0 3dN.2 29us : dequeue_pushable_task <-pick_next_task_rt -0 3d..3 29us : __schedule <-preempt_schedule -0 3d..3 30us : 0:120:R ==> [003] 2448: 94:R sleeph]hXPecho 1 > options/function-trace # tracer: wakeup_rt # # wakeup_rt latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 29 us, #85/85, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: sleep-2448 (uid:0 nice:0 policy:1 rt_prio:5) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 3d.h4 1us+: 0:120:R + [003] 2448: 94:R sleep -0 3d.h4 2us : ttwu_do_activate.constprop.87 <-try_to_wake_up -0 3d.h3 3us : check_preempt_curr <-ttwu_do_wakeup -0 3d.h3 3us : resched_curr <-check_preempt_curr -0 3dNh3 4us : task_woken_rt <-ttwu_do_wakeup -0 3dNh3 4us : _raw_spin_unlock <-try_to_wake_up -0 3dNh3 4us : sub_preempt_count <-_raw_spin_unlock -0 3dNh2 5us : ttwu_stat <-try_to_wake_up -0 3dNh2 5us : _raw_spin_unlock_irqrestore <-try_to_wake_up -0 3dNh2 6us : sub_preempt_count <-_raw_spin_unlock_irqrestore -0 3dNh1 6us : _raw_spin_lock <-__run_hrtimer -0 3dNh1 6us : add_preempt_count <-_raw_spin_lock -0 3dNh2 7us : _raw_spin_unlock <-hrtimer_interrupt -0 3dNh2 7us : sub_preempt_count <-_raw_spin_unlock -0 3dNh1 7us : tick_program_event <-hrtimer_interrupt -0 3dNh1 7us : clockevents_program_event <-tick_program_event -0 3dNh1 8us : ktime_get <-clockevents_program_event -0 3dNh1 8us : lapic_next_event <-clockevents_program_event -0 3dNh1 8us : irq_exit <-smp_apic_timer_interrupt -0 3dNh1 9us : sub_preempt_count <-irq_exit -0 3dN.2 9us : idle_cpu <-irq_exit -0 3dN.2 9us : rcu_irq_exit <-irq_exit -0 3dN.2 10us : rcu_eqs_enter_common.isra.45 <-rcu_irq_exit -0 3dN.2 10us : sub_preempt_count <-irq_exit -0 3.N.1 11us : rcu_idle_exit <-cpu_idle -0 3dN.1 11us : rcu_eqs_exit_common.isra.43 <-rcu_idle_exit -0 3.N.1 11us : tick_nohz_idle_exit <-cpu_idle -0 3dN.1 12us : menu_hrtimer_cancel <-tick_nohz_idle_exit -0 3dN.1 12us : ktime_get <-tick_nohz_idle_exit -0 3dN.1 12us : tick_do_update_jiffies64 <-tick_nohz_idle_exit -0 3dN.1 13us : cpu_load_update_nohz <-tick_nohz_idle_exit -0 3dN.1 13us : _raw_spin_lock <-cpu_load_update_nohz -0 3dN.1 13us : add_preempt_count <-_raw_spin_lock -0 3dN.2 13us : __cpu_load_update <-cpu_load_update_nohz -0 3dN.2 14us : sched_avg_update <-__cpu_load_update -0 3dN.2 14us : _raw_spin_unlock <-cpu_load_update_nohz -0 3dN.2 14us : sub_preempt_count <-_raw_spin_unlock -0 3dN.1 15us : calc_load_nohz_stop <-tick_nohz_idle_exit -0 3dN.1 15us : touch_softlockup_watchdog <-tick_nohz_idle_exit -0 3dN.1 15us : hrtimer_cancel <-tick_nohz_idle_exit -0 3dN.1 15us : hrtimer_try_to_cancel <-hrtimer_cancel -0 3dN.1 16us : lock_hrtimer_base.isra.18 <-hrtimer_try_to_cancel -0 3dN.1 16us : _raw_spin_lock_irqsave <-lock_hrtimer_base.isra.18 -0 3dN.1 16us : add_preempt_count <-_raw_spin_lock_irqsave -0 3dN.2 17us : __remove_hrtimer <-remove_hrtimer.part.16 -0 3dN.2 17us : hrtimer_force_reprogram <-__remove_hrtimer -0 3dN.2 17us : tick_program_event <-hrtimer_force_reprogram -0 3dN.2 18us : clockevents_program_event <-tick_program_event -0 3dN.2 18us : ktime_get <-clockevents_program_event -0 3dN.2 18us : lapic_next_event <-clockevents_program_event -0 3dN.2 19us : _raw_spin_unlock_irqrestore <-hrtimer_try_to_cancel -0 3dN.2 19us : sub_preempt_count <-_raw_spin_unlock_irqrestore -0 3dN.1 19us : hrtimer_forward <-tick_nohz_idle_exit -0 3dN.1 20us : ktime_add_safe <-hrtimer_forward -0 3dN.1 20us : ktime_add_safe <-hrtimer_forward -0 3dN.1 20us : hrtimer_start_range_ns <-hrtimer_start_expires.constprop.11 -0 3dN.1 20us : __hrtimer_start_range_ns <-hrtimer_start_range_ns -0 3dN.1 21us : lock_hrtimer_base.isra.18 <-__hrtimer_start_range_ns -0 3dN.1 21us : _raw_spin_lock_irqsave <-lock_hrtimer_base.isra.18 -0 3dN.1 21us : add_preempt_count <-_raw_spin_lock_irqsave -0 3dN.2 22us : ktime_add_safe <-__hrtimer_start_range_ns -0 3dN.2 22us : enqueue_hrtimer <-__hrtimer_start_range_ns -0 3dN.2 22us : tick_program_event <-__hrtimer_start_range_ns -0 3dN.2 23us : clockevents_program_event <-tick_program_event -0 3dN.2 23us : ktime_get <-clockevents_program_event -0 3dN.2 23us : lapic_next_event <-clockevents_program_event -0 3dN.2 24us : _raw_spin_unlock_irqrestore <-__hrtimer_start_range_ns -0 3dN.2 24us : sub_preempt_count <-_raw_spin_unlock_irqrestore -0 3dN.1 24us : account_idle_ticks <-tick_nohz_idle_exit -0 3dN.1 24us : account_idle_time <-account_idle_ticks -0 3.N.1 25us : sub_preempt_count <-cpu_idle -0 3.N.. 25us : schedule <-cpu_idle -0 3.N.. 25us : __schedule <-preempt_schedule -0 3.N.. 26us : add_preempt_count <-__schedule -0 3.N.1 26us : rcu_note_context_switch <-__schedule -0 3.N.1 26us : rcu_sched_qs <-rcu_note_context_switch -0 3dN.1 27us : rcu_preempt_qs <-rcu_note_context_switch -0 3.N.1 27us : _raw_spin_lock_irq <-__schedule -0 3dN.1 27us : add_preempt_count <-_raw_spin_lock_irq -0 3dN.2 28us : put_prev_task_idle <-__schedule -0 3dN.2 28us : pick_next_task_stop <-pick_next_task -0 3dN.2 28us : pick_next_task_rt <-pick_next_task -0 3dN.2 29us : dequeue_pushable_task <-pick_next_task_rt -0 3d..3 29us : __schedule <-preempt_schedule -0 3d..3 30us : 0:120:R ==> [003] 2448: 94:R sleep}hj'sbah}(h]h ]h"]h$]h&]jjuh1jhhhMEhji&hhubh)}(hcThis isn't that big of a trace, even with function tracing enabled, so I included the entire trace.h]heThis isn’t that big of a trace, even with function tracing enabled, so I included the entire trace.}(hj'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhji&hhubh)}(hThe interrupt went off while when the system was idle. Somewhere before task_woken_rt() was called, the NEED_RESCHED flag was set, this is indicated by the first occurrence of the 'N' flag.h]hThe interrupt went off while when the system was idle. Somewhere before task_woken_rt() was called, the NEED_RESCHED flag was set, this is indicated by the first occurrence of the ‘N’ flag.}(hj"'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhji&hhubeh}(h] wakeup-rtah ]h"] wakeup_rtah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(hLatency tracing and eventsh]hLatency tracing and events}(hj;'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj8'hhhhhMubh)}(hAs function tracing can induce a much larger latency, but without seeing what happens within the latency it is hard to know what caused it. There is a middle ground, and that is with enabling events. ::h]hAs function tracing can induce a much larger latency, but without seeing what happens within the latency it is hard to know what caused it. There is a middle ground, and that is with enabling events.}(hjI'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj8'hhubj)}(hX# echo 0 > options/function-trace # echo wakeup_rt > current_tracer # echo 1 > events/enable # echo 1 > tracing_on # echo 0 > tracing_max_latency # chrt -f 5 sleep 1 # echo 0 > tracing_on # cat trace # tracer: wakeup_rt # # wakeup_rt latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 6 us, #12/12, CPU#2 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: sleep-5882 (uid:0 nice:0 policy:1 rt_prio:5) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 2d.h4 0us : 0:120:R + [002] 5882: 94:R sleep -0 2d.h4 0us : ttwu_do_activate.constprop.87 <-try_to_wake_up -0 2d.h4 1us : sched_wakeup: comm=sleep pid=5882 prio=94 success=1 target_cpu=002 -0 2dNh2 1us : hrtimer_expire_exit: hrtimer=ffff88007796feb8 -0 2.N.2 2us : power_end: cpu_id=2 -0 2.N.2 3us : cpu_idle: state=4294967295 cpu_id=2 -0 2dN.3 4us : hrtimer_cancel: hrtimer=ffff88007d50d5e0 -0 2dN.3 4us : hrtimer_start: hrtimer=ffff88007d50d5e0 function=tick_sched_timer expires=34311211000000 softexpires=34311211000000 -0 2.N.2 5us : rcu_utilization: Start context switch -0 2.N.2 5us : rcu_utilization: End context switch -0 2d..3 6us : __schedule <-schedule -0 2d..3 6us : 0:120:R ==> [002] 5882: 94:R sleeph]hX# echo 0 > options/function-trace # echo wakeup_rt > current_tracer # echo 1 > events/enable # echo 1 > tracing_on # echo 0 > tracing_max_latency # chrt -f 5 sleep 1 # echo 0 > tracing_on # cat trace # tracer: wakeup_rt # # wakeup_rt latency trace v1.1.5 on 3.8.0-test+ # -------------------------------------------------------------------- # latency: 6 us, #12/12, CPU#2 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: sleep-5882 (uid:0 nice:0 policy:1 rt_prio:5) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 2d.h4 0us : 0:120:R + [002] 5882: 94:R sleep -0 2d.h4 0us : ttwu_do_activate.constprop.87 <-try_to_wake_up -0 2d.h4 1us : sched_wakeup: comm=sleep pid=5882 prio=94 success=1 target_cpu=002 -0 2dNh2 1us : hrtimer_expire_exit: hrtimer=ffff88007796feb8 -0 2.N.2 2us : power_end: cpu_id=2 -0 2.N.2 3us : cpu_idle: state=4294967295 cpu_id=2 -0 2dN.3 4us : hrtimer_cancel: hrtimer=ffff88007d50d5e0 -0 2dN.3 4us : hrtimer_start: hrtimer=ffff88007d50d5e0 function=tick_sched_timer expires=34311211000000 softexpires=34311211000000 -0 2.N.2 5us : rcu_utilization: Start context switch -0 2.N.2 5us : rcu_utilization: End context switch -0 2d..3 6us : __schedule <-schedule -0 2d..3 6us : 0:120:R ==> [002] 5882: 94:R sleep}hjW'sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj8'hhubeh}(h]latency-tracing-and-eventsah ]h"]latency tracing and eventsah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(hHardware Latency Detectorh]hHardware Latency Detector}(hjp'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjm'hhhhhMubh)}(hIThe hardware latency detector is executed by enabling the "hwlat" tracer.h]hMThe hardware latency detector is executed by enabling the “hwlat” tracer.}(hj~'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjm'hhubh)}(hNOTE, this tracer will affect the performance of the system as it will periodically make a CPU constantly busy with interrupts disabled. ::h]hNOTE, this tracer will affect the performance of the system as it will periodically make a CPU constantly busy with interrupts disabled.}(hj'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjm'hhubj)}(hX}# echo hwlat > current_tracer # sleep 100 # cat trace # tracer: hwlat # # entries-in-buffer/entries-written: 13/13 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | <...>-1729 [001] d... 678.473449: #1 inner/outer(us): 11/12 ts:1581527483.343962693 count:6 <...>-1729 [004] d... 689.556542: #2 inner/outer(us): 16/9 ts:1581527494.889008092 count:1 <...>-1729 [005] d... 714.756290: #3 inner/outer(us): 16/16 ts:1581527519.678961629 count:5 <...>-1729 [001] d... 718.788247: #4 inner/outer(us): 9/17 ts:1581527523.889012713 count:1 <...>-1729 [002] d... 719.796341: #5 inner/outer(us): 13/9 ts:1581527524.912872606 count:1 <...>-1729 [006] d... 844.787091: #6 inner/outer(us): 9/12 ts:1581527649.889048502 count:2 <...>-1729 [003] d... 849.827033: #7 inner/outer(us): 18/9 ts:1581527654.889013793 count:1 <...>-1729 [007] d... 853.859002: #8 inner/outer(us): 9/12 ts:1581527658.889065736 count:1 <...>-1729 [001] d... 855.874978: #9 inner/outer(us): 9/11 ts:1581527660.861991877 count:1 <...>-1729 [001] d... 863.938932: #10 inner/outer(us): 9/11 ts:1581527668.970010500 count:1 nmi-total:7 nmi-count:1 <...>-1729 [007] d... 878.050780: #11 inner/outer(us): 9/12 ts:1581527683.385002600 count:1 nmi-total:5 nmi-count:1 <...>-1729 [007] d... 886.114702: #12 inner/outer(us): 9/12 ts:1581527691.385001600 count:1h]hX}# echo hwlat > current_tracer # sleep 100 # cat trace # tracer: hwlat # # entries-in-buffer/entries-written: 13/13 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | <...>-1729 [001] d... 678.473449: #1 inner/outer(us): 11/12 ts:1581527483.343962693 count:6 <...>-1729 [004] d... 689.556542: #2 inner/outer(us): 16/9 ts:1581527494.889008092 count:1 <...>-1729 [005] d... 714.756290: #3 inner/outer(us): 16/16 ts:1581527519.678961629 count:5 <...>-1729 [001] d... 718.788247: #4 inner/outer(us): 9/17 ts:1581527523.889012713 count:1 <...>-1729 [002] d... 719.796341: #5 inner/outer(us): 13/9 ts:1581527524.912872606 count:1 <...>-1729 [006] d... 844.787091: #6 inner/outer(us): 9/12 ts:1581527649.889048502 count:2 <...>-1729 [003] d... 849.827033: #7 inner/outer(us): 18/9 ts:1581527654.889013793 count:1 <...>-1729 [007] d... 853.859002: #8 inner/outer(us): 9/12 ts:1581527658.889065736 count:1 <...>-1729 [001] d... 855.874978: #9 inner/outer(us): 9/11 ts:1581527660.861991877 count:1 <...>-1729 [001] d... 863.938932: #10 inner/outer(us): 9/11 ts:1581527668.970010500 count:1 nmi-total:7 nmi-count:1 <...>-1729 [007] d... 878.050780: #11 inner/outer(us): 9/12 ts:1581527683.385002600 count:1 nmi-total:5 nmi-count:1 <...>-1729 [007] d... 886.114702: #12 inner/outer(us): 9/12 ts:1581527691.385001600 count:1}hj'sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhjm'hhubh)}(hThe above output is somewhat the same in the header. All events will have interrupts disabled 'd'. Under the FUNCTION title there is:h]hThe above output is somewhat the same in the header. All events will have interrupts disabled ‘d’. Under the FUNCTION title there is:}(hj'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjm'hhubjT)}(hX#1 This is the count of events recorded that were greater than the tracing_threshold (See below). inner/outer(us): 11/11 This shows two numbers as "inner latency" and "outer latency". The test runs in a loop checking a timestamp twice. The latency detected within the two timestamps is the "inner latency" and the latency detected after the previous timestamp and the next timestamp in the loop is the "outer latency". ts:1581527483.343962693 The absolute timestamp that the first latency was recorded in the window. count:6 The number of times a latency was detected during the window. nmi-total:7 nmi-count:1 On architectures that support it, if an NMI comes in during the test, the time spent in NMI is reported in "nmi-total" (in microseconds). All architectures that have NMIs will show the "nmi-count" if an NMI comes in during the test. h](jT )}(hhh]jY )}(hb#1 This is the count of events recorded that were greater than the tracing_threshold (See below). h](j_ )}(h#1h]h#1}(hj'hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM hj'ubjo )}(hhh]h)}(h^This is the count of events recorded that were greater than the tracing_threshold (See below).h]h^This is the count of events recorded that were greater than the tracing_threshold (See below).}(hj'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj'ubah}(h]h ]h"]h$]h&]uh1jn hj'ubeh}(h]h ]h"]h$]h&]uh1jX hhhM hj'ubah}(h]h ]h"]h$]h&]uh1jS hj'ubh)}(hinner/outer(us): 11/11h]hinner/outer(us): 11/11}(hj'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj'ubjT)}(hX*This shows two numbers as "inner latency" and "outer latency". The test runs in a loop checking a timestamp twice. The latency detected within the two timestamps is the "inner latency" and the latency detected after the previous timestamp and the next timestamp in the loop is the "outer latency". h]h)}(hX)This shows two numbers as "inner latency" and "outer latency". The test runs in a loop checking a timestamp twice. The latency detected within the two timestamps is the "inner latency" and the latency detected after the previous timestamp and the next timestamp in the loop is the "outer latency".h]hX9This shows two numbers as “inner latency” and “outer latency”. The test runs in a loop checking a timestamp twice. The latency detected within the two timestamps is the “inner latency” and the latency detected after the previous timestamp and the next timestamp in the loop is the “outer latency”.}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj(ubah}(h]h ]h"]h$]h&]uh1jShhhM hj'ubh)}(hts:1581527483.343962693h]hts:1581527483.343962693}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj'ubjT)}(hJThe absolute timestamp that the first latency was recorded in the window. h]h)}(hIThe absolute timestamp that the first latency was recorded in the window.h]hIThe absolute timestamp that the first latency was recorded in the window.}(hj*(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj&(ubah}(h]h ]h"]h$]h&]uh1jShhhM hj'ubh)}(hcount:6h]hcount:6}(hj>(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj'ubjT)}(h>The number of times a latency was detected during the window. h]h)}(h=The number of times a latency was detected during the window.h]h=The number of times a latency was detected during the window.}(hjP(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjL(ubah}(h]h ]h"]h$]h&]uh1jShhhM hj'ubh)}(hnmi-total:7 nmi-count:1h]hnmi-total:7 nmi-count:1}(hjd(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj'ubjT)}(hOn architectures that support it, if an NMI comes in during the test, the time spent in NMI is reported in "nmi-total" (in microseconds). All architectures that have NMIs will show the "nmi-count" if an NMI comes in during the test. h](h)}(hOn architectures that support it, if an NMI comes in during the test, the time spent in NMI is reported in "nmi-total" (in microseconds).h]hOn architectures that support it, if an NMI comes in during the test, the time spent in NMI is reported in “nmi-total” (in microseconds).}(hjv(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM" hjr(ubh)}(h^All architectures that have NMIs will show the "nmi-count" if an NMI comes in during the test.h]hbAll architectures that have NMIs will show the “nmi-count” if an NMI comes in during the test.}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM& hjr(ubeh}(h]h ]h"]h$]h&]uh1jShhhM" hj'ubeh}(h]h ]h"]h$]h&]uh1jShhhM hjm'hhubh)}(h hwlat files:h]h hwlat files:}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM) hjm'hhubjT)}(hXtracing_threshold This gets automatically set to "10" to represent 10 microseconds. This is the threshold of latency that needs to be detected before the trace will be recorded. Note, when hwlat tracer is finished (another tracer is written into "current_tracer"), the original value for tracing_threshold is placed back into this file. hwlat_detector/width The length of time the test runs with interrupts disabled. hwlat_detector/window The length of time of the window which the test runs. That is, the test will run for "width" microseconds per "window" microseconds tracing_cpumask When the test is started. A kernel thread is created that runs the test. This thread will alternate between CPUs listed in the tracing_cpumask between each period (one "window"). To limit the test to specific CPUs set the mask in this file to only the CPUs that the test should run on. h]jT )}(hhh](jY )}(hXRtracing_threshold This gets automatically set to "10" to represent 10 microseconds. This is the threshold of latency that needs to be detected before the trace will be recorded. Note, when hwlat tracer is finished (another tracer is written into "current_tracer"), the original value for tracing_threshold is placed back into this file. h](j_ )}(htracing_thresholdh]htracing_threshold}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM2 hj(ubjo )}(hhh](h)}(hThis gets automatically set to "10" to represent 10 microseconds. This is the threshold of latency that needs to be detected before the trace will be recorded.h]hThis gets automatically set to “10” to represent 10 microseconds. This is the threshold of latency that needs to be detected before the trace will be recorded.}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM, hj(ubh)}(hNote, when hwlat tracer is finished (another tracer is written into "current_tracer"), the original value for tracing_threshold is placed back into this file.h]hNote, when hwlat tracer is finished (another tracer is written into “current_tracer”), the original value for tracing_threshold is placed back into this file.}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM0 hj(ubeh}(h]h ]h"]h$]h&]uh1jn hj(ubeh}(h]h ]h"]h$]h&]uh1jX hhhM2 hj(ubjY )}(hPhwlat_detector/width The length of time the test runs with interrupts disabled. h](j_ )}(hhwlat_detector/widthh]hhwlat_detector/width}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM5 hj(ubjo )}(hhh]h)}(h:The length of time the test runs with interrupts disabled.h]h:The length of time the test runs with interrupts disabled.}(hj)hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM5 hj)ubah}(h]h ]h"]h$]h&]uh1jn hj(ubeh}(h]h ]h"]h$]h&]uh1jX hhhM5 hj(ubjY )}(hhwlat_detector/window The length of time of the window which the test runs. That is, the test will run for "width" microseconds per "window" microseconds h](j_ )}(hhwlat_detector/windowh]hhwlat_detector/window}(hj#)hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhM: hj)ubjo )}(hhh]h)}(hThe length of time of the window which the test runs. That is, the test will run for "width" microseconds per "window" microsecondsh]hThe length of time of the window which the test runs. That is, the test will run for “width” microseconds per “window” microseconds}(hj4)hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM8 hj1)ubah}(h]h ]h"]h$]h&]uh1jn hj)ubeh}(h]h ]h"]h$]h&]uh1jX hhhM: hj(ubjY )}(hX.tracing_cpumask When the test is started. A kernel thread is created that runs the test. This thread will alternate between CPUs listed in the tracing_cpumask between each period (one "window"). To limit the test to specific CPUs set the mask in this file to only the CPUs that the test should run on. h](j_ )}(htracing_cpumaskh]htracing_cpumask}(hjR)hhhNhNubah}(h]h ]h"]h$]h&]uh1j^ hhhMB hjN)ubjo )}(hhh]h)}(hXWhen the test is started. A kernel thread is created that runs the test. This thread will alternate between CPUs listed in the tracing_cpumask between each period (one "window"). To limit the test to specific CPUs set the mask in this file to only the CPUs that the test should run on.h]hX!When the test is started. A kernel thread is created that runs the test. This thread will alternate between CPUs listed in the tracing_cpumask between each period (one “window”). To limit the test to specific CPUs set the mask in this file to only the CPUs that the test should run on.}(hjc)hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM= hj`)ubah}(h]h ]h"]h$]h&]uh1jn hjN)ubeh}(h]h ]h"]h$]h&]uh1jX hhhMB hj(ubeh}(h]h ]h"]h$]h&]uh1jS hj(ubah}(h]h ]h"]h$]h&]uh1jShhhM+ hjm'hhubeh}(h]hardware-latency-detectorah ]h"]hardware latency detectorah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(hfunctionh]hfunction}(hj)hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj)hhhhhME ubh)}(hThis tracer is the function tracer. Enabling the function tracer can be done from the debug file system. Make sure the ftrace_enabled is set; otherwise this tracer is a nop. See the "ftrace_enabled" section below. ::h]hThis tracer is the function tracer. Enabling the function tracer can be done from the debug file system. Make sure the ftrace_enabled is set; otherwise this tracer is a nop. See the “ftrace_enabled” section below.}(hj)hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMG hj)hhubj)}(hX# sysctl kernel.ftrace_enabled=1 # echo function > current_tracer # echo 1 > tracing_on # usleep 1 # echo 0 > tracing_on # cat trace # tracer: function # # entries-in-buffer/entries-written: 24799/24799 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | bash-1994 [002] .... 3082.063030: mutex_unlock <-rb_simple_write bash-1994 [002] .... 3082.063031: __mutex_unlock_slowpath <-mutex_unlock bash-1994 [002] .... 3082.063031: __fsnotify_parent <-fsnotify_modify bash-1994 [002] .... 3082.063032: fsnotify <-fsnotify_modify bash-1994 [002] .... 3082.063032: __srcu_read_lock <-fsnotify bash-1994 [002] .... 3082.063032: add_preempt_count <-__srcu_read_lock bash-1994 [002] ...1 3082.063032: sub_preempt_count <-__srcu_read_lock bash-1994 [002] .... 3082.063033: __srcu_read_unlock <-fsnotify [...]h]hX# sysctl kernel.ftrace_enabled=1 # echo function > current_tracer # echo 1 > tracing_on # usleep 1 # echo 0 > tracing_on # cat trace # tracer: function # # entries-in-buffer/entries-written: 24799/24799 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | bash-1994 [002] .... 3082.063030: mutex_unlock <-rb_simple_write bash-1994 [002] .... 3082.063031: __mutex_unlock_slowpath <-mutex_unlock bash-1994 [002] .... 3082.063031: __fsnotify_parent <-fsnotify_modify bash-1994 [002] .... 3082.063032: fsnotify <-fsnotify_modify bash-1994 [002] .... 3082.063032: __srcu_read_lock <-fsnotify bash-1994 [002] .... 3082.063032: add_preempt_count <-__srcu_read_lock bash-1994 [002] ...1 3082.063032: sub_preempt_count <-__srcu_read_lock bash-1994 [002] .... 3082.063033: __srcu_read_unlock <-fsnotify [...]}hj)sbah}(h]h ]h"]h$]h&]jjuh1jhhhMM hj)hhubh)}(hXNote: function tracer uses ring buffers to store the above entries. The newest data may overwrite the oldest data. Sometimes using echo to stop the trace is not sufficient because the tracing could have overwritten the data that you wanted to record. For this reason, it is sometimes better to disable tracing directly from a program. This allows you to stop the tracing at the point that you hit the part that you are interested in. To disable the tracing directly from a C program, something like following code snippet can be used::h]hXNote: function tracer uses ring buffers to store the above entries. The newest data may overwrite the oldest data. Sometimes using echo to stop the trace is not sufficient because the tracing could have overwritten the data that you wanted to record. For this reason, it is sometimes better to disable tracing directly from a program. This allows you to stop the tracing at the point that you hit the part that you are interested in. To disable the tracing directly from a C program, something like following code snippet can be used:}(hj)hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMi hj)hhubj)}(hint trace_fd; [...] int main(int argc, char *argv[]) { [...] trace_fd = open(tracing_file("tracing_on"), O_WRONLY); [...] if (condition_hit()) { write(trace_fd, "0", 1); } [...] }h]hint trace_fd; [...] int main(int argc, char *argv[]) { [...] trace_fd = open(tracing_file("tracing_on"), O_WRONLY); [...] if (condition_hit()) { write(trace_fd, "0", 1); } [...] }}hj)sbah}(h]h ]h"]h$]h&]jjuh1jhhhMs hj)hhubeh}(h]functionah ]h"]functionah$]h&]uh1hhhhhhhhME ubh)}(hhh](h)}(hSingle thread tracingh]hSingle thread tracing}(hj)hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj)hhhhhM ubh)}(hKBy writing into set_ftrace_pid you can trace a single thread. For example::h]hJBy writing into set_ftrace_pid you can trace a single thread. For example:}(hj)hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj)hhubj)}(hX# cat set_ftrace_pid no pid # echo 3111 > set_ftrace_pid # cat set_ftrace_pid 3111 # echo function > current_tracer # cat trace | head # tracer: function # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | yum-updatesd-3111 [003] 1637.254676: finish_task_switch <-thread_return yum-updatesd-3111 [003] 1637.254681: hrtimer_cancel <-schedule_hrtimeout_range yum-updatesd-3111 [003] 1637.254682: hrtimer_try_to_cancel <-hrtimer_cancel yum-updatesd-3111 [003] 1637.254683: lock_hrtimer_base <-hrtimer_try_to_cancel yum-updatesd-3111 [003] 1637.254685: fget_light <-do_sys_poll yum-updatesd-3111 [003] 1637.254686: pipe_poll <-do_sys_poll # echo > set_ftrace_pid # cat trace |head # tracer: function # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | ##### CPU 3 buffer started #### yum-updatesd-3111 [003] 1701.957688: free_poll_entry <-poll_freewait yum-updatesd-3111 [003] 1701.957689: remove_wait_queue <-free_poll_entry yum-updatesd-3111 [003] 1701.957691: fput <-free_poll_entry yum-updatesd-3111 [003] 1701.957692: audit_syscall_exit <-sysret_audit yum-updatesd-3111 [003] 1701.957693: path_put <-audit_syscall_exith]hX# cat set_ftrace_pid no pid # echo 3111 > set_ftrace_pid # cat set_ftrace_pid 3111 # echo function > current_tracer # cat trace | head # tracer: function # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | yum-updatesd-3111 [003] 1637.254676: finish_task_switch <-thread_return yum-updatesd-3111 [003] 1637.254681: hrtimer_cancel <-schedule_hrtimeout_range yum-updatesd-3111 [003] 1637.254682: hrtimer_try_to_cancel <-hrtimer_cancel yum-updatesd-3111 [003] 1637.254683: lock_hrtimer_base <-hrtimer_try_to_cancel yum-updatesd-3111 [003] 1637.254685: fget_light <-do_sys_poll yum-updatesd-3111 [003] 1637.254686: pipe_poll <-do_sys_poll # echo > set_ftrace_pid # cat trace |head # tracer: function # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | ##### CPU 3 buffer started #### yum-updatesd-3111 [003] 1701.957688: free_poll_entry <-poll_freewait yum-updatesd-3111 [003] 1701.957689: remove_wait_queue <-free_poll_entry yum-updatesd-3111 [003] 1701.957691: fput <-free_poll_entry yum-updatesd-3111 [003] 1701.957692: audit_syscall_exit <-sysret_audit yum-updatesd-3111 [003] 1701.957693: path_put <-audit_syscall_exit}hj*sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj)hhubh)}(hdIf you want to trace a function when executing, you could use something like this simple program. ::h]haIf you want to trace a function when executing, you could use something like this simple program.}(hj*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj)hhubj)}(hX#include #include #include #include #include #include #include #define _STR(x) #x #define STR(x) _STR(x) #define MAX_PATH 256 const char *find_tracefs(void) { static char tracefs[MAX_PATH+1]; static int tracefs_found; char type[100]; FILE *fp; if (tracefs_found) return tracefs; if ((fp = fopen("/proc/mounts","r")) == NULL) { perror("/proc/mounts"); return NULL; } while (fscanf(fp, "%*s %" STR(MAX_PATH) "s %99s %*s %*d %*d\n", tracefs, type) == 2) { if (strcmp(type, "tracefs") == 0) break; } fclose(fp); if (strcmp(type, "tracefs") != 0) { fprintf(stderr, "tracefs not mounted"); return NULL; } strcat(tracefs, "/tracing/"); tracefs_found = 1; return tracefs; } const char *tracing_file(const char *file_name) { static char trace_file[MAX_PATH+1]; snprintf(trace_file, MAX_PATH, "%s/%s", find_tracefs(), file_name); return trace_file; } int main (int argc, char **argv) { if (argc < 1) exit(-1); if (fork() > 0) { int fd, ffd; char line[64]; int s; ffd = open(tracing_file("current_tracer"), O_WRONLY); if (ffd < 0) exit(-1); write(ffd, "nop", 3); fd = open(tracing_file("set_ftrace_pid"), O_WRONLY); s = sprintf(line, "%d\n", getpid()); write(fd, line, s); write(ffd, "function", 8); close(fd); close(ffd); execvp(argv[1], argv+1); } return 0; }h]hX#include #include #include #include #include #include #include #define _STR(x) #x #define STR(x) _STR(x) #define MAX_PATH 256 const char *find_tracefs(void) { static char tracefs[MAX_PATH+1]; static int tracefs_found; char type[100]; FILE *fp; if (tracefs_found) return tracefs; if ((fp = fopen("/proc/mounts","r")) == NULL) { perror("/proc/mounts"); return NULL; } while (fscanf(fp, "%*s %" STR(MAX_PATH) "s %99s %*s %*d %*d\n", tracefs, type) == 2) { if (strcmp(type, "tracefs") == 0) break; } fclose(fp); if (strcmp(type, "tracefs") != 0) { fprintf(stderr, "tracefs not mounted"); return NULL; } strcat(tracefs, "/tracing/"); tracefs_found = 1; return tracefs; } const char *tracing_file(const char *file_name) { static char trace_file[MAX_PATH+1]; snprintf(trace_file, MAX_PATH, "%s/%s", find_tracefs(), file_name); return trace_file; } int main (int argc, char **argv) { if (argc < 1) exit(-1); if (fork() > 0) { int fd, ffd; char line[64]; int s; ffd = open(tracing_file("current_tracer"), O_WRONLY); if (ffd < 0) exit(-1); write(ffd, "nop", 3); fd = open(tracing_file("set_ftrace_pid"), O_WRONLY); s = sprintf(line, "%d\n", getpid()); write(fd, line, s); write(ffd, "function", 8); close(fd); close(ffd); execvp(argv[1], argv+1); } return 0; }}hj*sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj)hhubh)}(hOr this simple script! ::h]hOr this simple script!}(hj+*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj)hhubj)}(h#!/bin/bash tracefs=`sed -ne 's/^tracefs \(.*\) tracefs.*/\1/p' /proc/mounts` echo 0 > $tracefs/tracing_on echo $$ > $tracefs/set_ftrace_pid echo function > $tracefs/current_tracer echo 1 > $tracefs/tracing_on exec "$@"h]h#!/bin/bash tracefs=`sed -ne 's/^tracefs \(.*\) tracefs.*/\1/p' /proc/mounts` echo 0 > $tracefs/tracing_on echo $$ > $tracefs/set_ftrace_pid echo function > $tracefs/current_tracer echo 1 > $tracefs/tracing_on exec "$@"}hj9*sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj)hhubeh}(h]single-thread-tracingah ]h"]single thread tracingah$]h&]uh1hhhhhhhhM ubh)}(hhh](h)}(hfunction graph tracerh]hfunction graph tracer}(hjR*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjO*hhhhhM ubh)}(hXThis tracer is similar to the function tracer except that it probes a function on its entry and its exit. This is done by using a dynamically allocated stack of return addresses in each task_struct. On function entry the tracer overwrites the return address of each function traced to set a custom probe. Thus the original return address is stored on the stack of return address in the task_struct.h]hXThis tracer is similar to the function tracer except that it probes a function on its entry and its exit. This is done by using a dynamically allocated stack of return addresses in each task_struct. On function entry the tracer overwrites the return address of each function traced to set a custom probe. Thus the original return address is stored on the stack of return address in the task_struct.}(hj`*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjO*hhubh)}(hEProbing on both ends of a function leads to special features such as:h]hEProbing on both ends of a function leads to special features such as:}(hjn*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjO*hhubj{)}(hhh](j)}(h&measure of a function's time executionh]h)}(hj*h]h(measure of a function’s time execution}(hj*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj*ubah}(h]h ]h"]h$]h&]uh1jhj|*hhhhhNubj)}(h:having a reliable call stack to draw function calls graph h]h)}(h9having a reliable call stack to draw function calls graphh]h9having a reliable call stack to draw function calls graph}(hj*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj*ubah}(h]h ]h"]h$]h&]uh1jhj|*hhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jzhhhM hjO*hhubh)}(h,This tracer is useful in several situations:h]h,This tracer is useful in several situations:}(hj*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjO*hhubj{)}(hhh](j)}(hyou want to find the reason of a strange kernel behavior and need to see what happens in detail on any areas (or specific ones). h]h)}(hyou want to find the reason of a strange kernel behavior and need to see what happens in detail on any areas (or specific ones).h]hyou want to find the reason of a strange kernel behavior and need to see what happens in detail on any areas (or specific ones).}(hj*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj*ubah}(h]h ]h"]h$]h&]uh1jhj*hhhhhNubj)}(hLyou are experiencing weird latencies but it's difficult to find its origin. h]h)}(hKyou are experiencing weird latencies but it's difficult to find its origin.h]hMyou are experiencing weird latencies but it’s difficult to find its origin.}(hj*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj*ubah}(h]h ]h"]h$]h&]uh1jhj*hhhhhNubj)}(hDyou want to find quickly which path is taken by a specific function h]h)}(hCyou want to find quickly which path is taken by a specific functionh]hCyou want to find quickly which path is taken by a specific function}(hj*hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM# hj*ubah}(h]h ]h"]h$]h&]uh1jhj*hhhhhNubj)}(hRyou just want to peek inside a working kernel and want to see what happens there. h]h)}(hQyou just want to peek inside a working kernel and want to see what happens there.h]hQyou just want to peek inside a working kernel and want to see what happens there.}(hj+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM& hj +ubah}(h]h ]h"]h$]h&]uh1jhj*hhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jzhhhM hjO*hhubj)}(hX# tracer: function_graph # # CPU DURATION FUNCTION CALLS # | | | | | | | 0) | sys_open() { 0) | do_sys_open() { 0) | getname() { 0) | kmem_cache_alloc() { 0) 1.382 us | __might_sleep(); 0) 2.478 us | } 0) | strncpy_from_user() { 0) | might_fault() { 0) 1.389 us | __might_sleep(); 0) 2.553 us | } 0) 3.807 us | } 0) 7.876 us | } 0) | alloc_fd() { 0) 0.668 us | _spin_lock(); 0) 0.570 us | expand_files(); 0) 0.586 us | _spin_unlock();h]hX# tracer: function_graph # # CPU DURATION FUNCTION CALLS # | | | | | | | 0) | sys_open() { 0) | do_sys_open() { 0) | getname() { 0) | kmem_cache_alloc() { 0) 1.382 us | __might_sleep(); 0) 2.478 us | } 0) | strncpy_from_user() { 0) | might_fault() { 0) 1.389 us | __might_sleep(); 0) 2.553 us | } 0) 3.807 us | } 0) 7.876 us | } 0) | alloc_fd() { 0) 0.668 us | _spin_lock(); 0) 0.570 us | expand_files(); 0) 0.586 us | _spin_unlock();}hj++sbah}(h]h ]h"]h$]h&]jjuh1jhhhM+ hjO*hhubh)}(hThere are several columns that can be dynamically enabled/disabled. You can use every combination of options you want, depending on your needs.h]hThere are several columns that can be dynamically enabled/disabled. You can use every combination of options you want, depending on your needs.}(hj9+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMB hjO*hhubj{)}(hhh](j)}(hX>The cpu number on which the function executed is default enabled. It is sometimes better to only trace one cpu (see tracing_cpumask file) or you might sometimes see unordered function calls while cpu tracing switch. - hide: echo nofuncgraph-cpu > trace_options - show: echo funcgraph-cpu > trace_options h](h)}(hThe cpu number on which the function executed is default enabled. It is sometimes better to only trace one cpu (see tracing_cpumask file) or you might sometimes see unordered function calls while cpu tracing switch.h]hThe cpu number on which the function executed is default enabled. It is sometimes better to only trace one cpu (see tracing_cpumask file) or you might sometimes see unordered function calls while cpu tracing switch.}(hjN+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMF hjJ+ubjT)}(hX- hide: echo nofuncgraph-cpu > trace_options - show: echo funcgraph-cpu > trace_options h]j{)}(hhh](j)}(h*hide: echo nofuncgraph-cpu > trace_optionsh]h)}(hje+h]h*hide: echo nofuncgraph-cpu > trace_options}(hjg+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMK hjc+ubah}(h]h ]h"]h$]h&]uh1jhj`+ubj)}(h)show: echo funcgraph-cpu > trace_options h]h)}(h(show: echo funcgraph-cpu > trace_optionsh]h(show: echo funcgraph-cpu > trace_options}(hj~+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhML hjz+ubah}(h]h ]h"]h$]h&]uh1jhj`+ubeh}(h]h ]h"]h$]h&]jjuh1jzhhhMK hj\+ubah}(h]h ]h"]h$]h&]uh1jShhhMK hjJ+ubeh}(h]h ]h"]h$]h&]uh1jhjG+hhhhhNubj)}(hX.The duration (function's time of execution) is displayed on the closing bracket line of a function or on the same line than the current function in case of a leaf one. It is default enabled. - hide: echo nofuncgraph-duration > trace_options - show: echo funcgraph-duration > trace_options h](h)}(hThe duration (function's time of execution) is displayed on the closing bracket line of a function or on the same line than the current function in case of a leaf one. It is default enabled.h]hThe duration (function’s time of execution) is displayed on the closing bracket line of a function or on the same line than the current function in case of a leaf one. It is default enabled.}(hj+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMN hj+ubjT)}(hb- hide: echo nofuncgraph-duration > trace_options - show: echo funcgraph-duration > trace_options h]j{)}(hhh](j)}(h/hide: echo nofuncgraph-duration > trace_optionsh]h)}(hj+h]h/hide: echo nofuncgraph-duration > trace_options}(hj+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMS hj+ubah}(h]h ]h"]h$]h&]uh1jhj+ubj)}(h.show: echo funcgraph-duration > trace_options h]h)}(h-show: echo funcgraph-duration > trace_optionsh]h-show: echo funcgraph-duration > trace_options}(hj+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMT hj+ubah}(h]h ]h"]h$]h&]uh1jhj+ubeh}(h]h ]h"]h$]h&]jjuh1jzhhhMS hj+ubah}(h]h ]h"]h$]h&]uh1jShhhMS hj+ubeh}(h]h ]h"]h$]h&]uh1jhjG+hhhhhNubj)}(hXThe overhead field precedes the duration field in case of reached duration thresholds. - hide: echo nofuncgraph-overhead > trace_options - show: echo funcgraph-overhead > trace_options - depends on: funcgraph-duration ie:: 3) # 1837.709 us | } /* __switch_to */ 3) | finish_task_switch() { 3) 0.313 us | _raw_spin_unlock_irq(); 3) 3.177 us | } 3) # 1889.063 us | } /* __schedule */ 3) ! 140.417 us | } /* __schedule */ 3) # 2034.948 us | } /* schedule */ 3) * 33998.59 us | } /* schedule_preempt_disabled */ [...] 1) 0.260 us | msecs_to_jiffies(); 1) 0.313 us | __rcu_read_unlock(); 1) + 61.770 us | } 1) + 64.479 us | } 1) 0.313 us | rcu_bh_qs(); 1) 0.313 us | __local_bh_enable(); 1) ! 217.240 us | } 1) 0.365 us | idle_cpu(); 1) | rcu_irq_exit() { 1) 0.417 us | rcu_eqs_enter_common.isra.47(); 1) 3.125 us | } 1) ! 227.812 us | } 1) ! 457.395 us | } 1) @ 119760.2 us | } [...] 2) | handle_IPI() { 1) 6.979 us | } 2) 0.417 us | scheduler_ipi(); 1) 9.791 us | } 1) + 12.917 us | } 2) 3.490 us | } 1) + 15.729 us | } 1) + 18.542 us | } 2) $ 3594274 us | } h](h)}(hVThe overhead field precedes the duration field in case of reached duration thresholds.h]hVThe overhead field precedes the duration field in case of reached duration thresholds.}(hj,hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMV hj+ubjT)}(h- hide: echo nofuncgraph-overhead > trace_options - show: echo funcgraph-overhead > trace_options - depends on: funcgraph-duration h]j{)}(hhh](j)}(h/hide: echo nofuncgraph-overhead > trace_optionsh]h)}(hj,h]h/hide: echo nofuncgraph-overhead > trace_options}(hj,hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMY hj,ubah}(h]h ]h"]h$]h&]uh1jhj,ubj)}(h-show: echo funcgraph-overhead > trace_optionsh]h)}(hj0,h]h-show: echo funcgraph-overhead > trace_options}(hj2,hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMZ hj.,ubah}(h]h ]h"]h$]h&]uh1jhj,ubj)}(hdepends on: funcgraph-duration h]h)}(hdepends on: funcgraph-durationh]hdepends on: funcgraph-duration}(hjI,hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM[ hjE,ubah}(h]h ]h"]h$]h&]uh1jhj,ubeh}(h]h ]h"]h$]h&]jjuh1jzhhhMY hj,ubah}(h]h ]h"]h$]h&]uh1jShhhMY hj+ubh)}(hie::h]hie:}(hji,hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM] hj+ubj)}(hX3) # 1837.709 us | } /* __switch_to */ 3) | finish_task_switch() { 3) 0.313 us | _raw_spin_unlock_irq(); 3) 3.177 us | } 3) # 1889.063 us | } /* __schedule */ 3) ! 140.417 us | } /* __schedule */ 3) # 2034.948 us | } /* schedule */ 3) * 33998.59 us | } /* schedule_preempt_disabled */ [...] 1) 0.260 us | msecs_to_jiffies(); 1) 0.313 us | __rcu_read_unlock(); 1) + 61.770 us | } 1) + 64.479 us | } 1) 0.313 us | rcu_bh_qs(); 1) 0.313 us | __local_bh_enable(); 1) ! 217.240 us | } 1) 0.365 us | idle_cpu(); 1) | rcu_irq_exit() { 1) 0.417 us | rcu_eqs_enter_common.isra.47(); 1) 3.125 us | } 1) ! 227.812 us | } 1) ! 457.395 us | } 1) @ 119760.2 us | } [...] 2) | handle_IPI() { 1) 6.979 us | } 2) 0.417 us | scheduler_ipi(); 1) 9.791 us | } 1) + 12.917 us | } 2) 3.490 us | } 1) + 15.729 us | } 1) + 18.542 us | } 2) $ 3594274 us | }h]hX3) # 1837.709 us | } /* __switch_to */ 3) | finish_task_switch() { 3) 0.313 us | _raw_spin_unlock_irq(); 3) 3.177 us | } 3) # 1889.063 us | } /* __schedule */ 3) ! 140.417 us | } /* __schedule */ 3) # 2034.948 us | } /* schedule */ 3) * 33998.59 us | } /* schedule_preempt_disabled */ [...] 1) 0.260 us | msecs_to_jiffies(); 1) 0.313 us | __rcu_read_unlock(); 1) + 61.770 us | } 1) + 64.479 us | } 1) 0.313 us | rcu_bh_qs(); 1) 0.313 us | __local_bh_enable(); 1) ! 217.240 us | } 1) 0.365 us | idle_cpu(); 1) | rcu_irq_exit() { 1) 0.417 us | rcu_eqs_enter_common.isra.47(); 1) 3.125 us | } 1) ! 227.812 us | } 1) ! 457.395 us | } 1) @ 119760.2 us | } [...] 2) | handle_IPI() { 1) 6.979 us | } 2) 0.417 us | scheduler_ipi(); 1) 9.791 us | } 1) + 12.917 us | } 2) 3.490 us | } 1) + 15.729 us | } 1) + 18.542 us | } 2) $ 3594274 us | }}hjw,sbah}(h]h ]h"]h$]h&]jjuh1jhhhM_ hj+ubeh}(h]h ]h"]h$]h&]uh1jhjG+hhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jzhhhMF hjO*hhubh)}(hFlags::h]hFlags:}(hj,hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjO*hhubj)}(hX+ means that the function exceeded 10 usecs. ! means that the function exceeded 100 usecs. # means that the function exceeded 1000 usecs. * means that the function exceeded 10 msecs. @ means that the function exceeded 100 msecs. $ means that the function exceeded 1 sec.h]hX+ means that the function exceeded 10 usecs. ! means that the function exceeded 100 usecs. # means that the function exceeded 1000 usecs. * means that the function exceeded 10 msecs. @ means that the function exceeded 100 msecs. $ means that the function exceeded 1 sec.}hj,sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hjO*hhubj{)}(hhh](j)}(hXThe task/pid field displays the thread cmdline and pid which executed the function. It is default disabled. - hide: echo nofuncgraph-proc > trace_options - show: echo funcgraph-proc > trace_options ie:: # tracer: function_graph # # CPU TASK/PID DURATION FUNCTION CALLS # | | | | | | | | | 0) sh-4802 | | d_free() { 0) sh-4802 | | call_rcu() { 0) sh-4802 | | __call_rcu() { 0) sh-4802 | 0.616 us | rcu_process_gp_end(); 0) sh-4802 | 0.586 us | check_for_new_grace_period(); 0) sh-4802 | 2.899 us | } 0) sh-4802 | 4.040 us | } 0) sh-4802 | 5.151 us | } 0) sh-4802 | + 49.370 us | } h](h)}(hkThe task/pid field displays the thread cmdline and pid which executed the function. It is default disabled.h]hkThe task/pid field displays the thread cmdline and pid which executed the function. It is default disabled.}(hj,hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj,ubjT)}(hZ- hide: echo nofuncgraph-proc > trace_options - show: echo funcgraph-proc > trace_options h]j{)}(hhh](j)}(h+hide: echo nofuncgraph-proc > trace_optionsh]h)}(hj,h]h+hide: echo nofuncgraph-proc > trace_options}(hj,hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj,ubah}(h]h ]h"]h$]h&]uh1jhj,ubj)}(h*show: echo funcgraph-proc > trace_options h]h)}(h)show: echo funcgraph-proc > trace_optionsh]h)show: echo funcgraph-proc > trace_options}(hj,hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj,ubah}(h]h ]h"]h$]h&]uh1jhj,ubeh}(h]h ]h"]h$]h&]jjuh1jzhhhM hj,ubah}(h]h ]h"]h$]h&]uh1jShhhM hj,ubh)}(hie::h]hie:}(hj-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj,ubj)}(hX# tracer: function_graph # # CPU TASK/PID DURATION FUNCTION CALLS # | | | | | | | | | 0) sh-4802 | | d_free() { 0) sh-4802 | | call_rcu() { 0) sh-4802 | | __call_rcu() { 0) sh-4802 | 0.616 us | rcu_process_gp_end(); 0) sh-4802 | 0.586 us | check_for_new_grace_period(); 0) sh-4802 | 2.899 us | } 0) sh-4802 | 4.040 us | } 0) sh-4802 | 5.151 us | } 0) sh-4802 | + 49.370 us | }h]hX# tracer: function_graph # # CPU TASK/PID DURATION FUNCTION CALLS # | | | | | | | | | 0) sh-4802 | | d_free() { 0) sh-4802 | | call_rcu() { 0) sh-4802 | | __call_rcu() { 0) sh-4802 | 0.616 us | rcu_process_gp_end(); 0) sh-4802 | 0.586 us | check_for_new_grace_period(); 0) sh-4802 | 2.899 us | } 0) sh-4802 | 4.040 us | } 0) sh-4802 | 5.151 us | } 0) sh-4802 | + 49.370 us | }}hj-sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj,ubeh}(h]h ]h"]h$]h&]uh1jhj,hhhhhNubj)}(hXThe absolute time field is an absolute timestamp given by the system clock since it started. A snapshot of this time is given on each entry/exit of functions - hide: echo nofuncgraph-abstime > trace_options - show: echo funcgraph-abstime > trace_options ie:: # # TIME CPU DURATION FUNCTION CALLS # | | | | | | | | 360.774522 | 1) 0.541 us | } 360.774522 | 1) 4.663 us | } 360.774523 | 1) 0.541 us | __wake_up_bit(); 360.774524 | 1) 6.796 us | } 360.774524 | 1) 7.952 us | } 360.774525 | 1) 9.063 us | } 360.774525 | 1) 0.615 us | journal_mark_dirty(); 360.774527 | 1) 0.578 us | __brelse(); 360.774528 | 1) | reiserfs_prepare_for_journal() { 360.774528 | 1) | unlock_buffer() { 360.774529 | 1) | wake_up_bit() { 360.774529 | 1) | bit_waitqueue() { 360.774530 | 1) 0.594 us | __phys_addr(); h](h)}(hThe absolute time field is an absolute timestamp given by the system clock since it started. A snapshot of this time is given on each entry/exit of functionsh]hThe absolute time field is an absolute timestamp given by the system clock since it started. A snapshot of this time is given on each entry/exit of functions}(hj*-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj&-ubjT)}(h`- hide: echo nofuncgraph-abstime > trace_options - show: echo funcgraph-abstime > trace_options h]j{)}(hhh](j)}(h.hide: echo nofuncgraph-abstime > trace_optionsh]h)}(hjA-h]h.hide: echo nofuncgraph-abstime > trace_options}(hjC-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj?-ubah}(h]h ]h"]h$]h&]uh1jhj<-ubj)}(h-show: echo funcgraph-abstime > trace_options h]h)}(h,show: echo funcgraph-abstime > trace_optionsh]h,show: echo funcgraph-abstime > trace_options}(hjZ-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjV-ubah}(h]h ]h"]h$]h&]uh1jhj<-ubeh}(h]h ]h"]h$]h&]jjuh1jzhhhM hj8-ubah}(h]h ]h"]h$]h&]uh1jShhhM hj&-ubh)}(hie::h]hie:}(hjz-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj&-ubj)}(hX# # TIME CPU DURATION FUNCTION CALLS # | | | | | | | | 360.774522 | 1) 0.541 us | } 360.774522 | 1) 4.663 us | } 360.774523 | 1) 0.541 us | __wake_up_bit(); 360.774524 | 1) 6.796 us | } 360.774524 | 1) 7.952 us | } 360.774525 | 1) 9.063 us | } 360.774525 | 1) 0.615 us | journal_mark_dirty(); 360.774527 | 1) 0.578 us | __brelse(); 360.774528 | 1) | reiserfs_prepare_for_journal() { 360.774528 | 1) | unlock_buffer() { 360.774529 | 1) | wake_up_bit() { 360.774529 | 1) | bit_waitqueue() { 360.774530 | 1) 0.594 us | __phys_addr();h]hX# # TIME CPU DURATION FUNCTION CALLS # | | | | | | | | 360.774522 | 1) 0.541 us | } 360.774522 | 1) 4.663 us | } 360.774523 | 1) 0.541 us | __wake_up_bit(); 360.774524 | 1) 6.796 us | } 360.774524 | 1) 7.952 us | } 360.774525 | 1) 9.063 us | } 360.774525 | 1) 0.615 us | journal_mark_dirty(); 360.774527 | 1) 0.578 us | __brelse(); 360.774528 | 1) | reiserfs_prepare_for_journal() { 360.774528 | 1) | unlock_buffer() { 360.774529 | 1) | wake_up_bit() { 360.774529 | 1) | bit_waitqueue() { 360.774530 | 1) 0.594 us | __phys_addr();}hj-sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj&-ubeh}(h]h ]h"]h$]h&]uh1jhj,hhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jzhhhM hjO*hhubh)}(hThe function name is always displayed after the closing bracket for a function if the start of that function is not in the trace buffer.h]hThe function name is always displayed after the closing bracket for a function if the start of that function is not in the trace buffer.}(hj-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjO*hhubh)}(hDisplay of the function name after the closing bracket may be enabled for functions whose start is in the trace buffer, allowing easier searching with grep for function durations. It is default disabled.h]hDisplay of the function name after the closing bracket may be enabled for functions whose start is in the trace buffer, allowing easier searching with grep for function durations. It is default disabled.}(hj-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjO*hhubjT)}(hXT - hide: echo nofuncgraph-tail > trace_options - show: echo funcgraph-tail > trace_options Example with nofuncgraph-tail (default):: 0) | putname() { 0) | kmem_cache_free() { 0) 0.518 us | __phys_addr(); 0) 1.757 us | } 0) 2.861 us | } Example with funcgraph-tail:: 0) | putname() { 0) | kmem_cache_free() { 0) 0.518 us | __phys_addr(); 0) 1.757 us | } /* kmem_cache_free() */ 0) 2.861 us | } /* putname() */ h](jT)}(hZ- hide: echo nofuncgraph-tail > trace_options - show: echo funcgraph-tail > trace_options h]j{)}(hhh](j)}(h+hide: echo nofuncgraph-tail > trace_optionsh]h)}(hj-h]h+hide: echo nofuncgraph-tail > trace_options}(hj-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj-ubah}(h]h ]h"]h$]h&]uh1jhj-ubj)}(h*show: echo funcgraph-tail > trace_options h]h)}(h)show: echo funcgraph-tail > trace_optionsh]h)show: echo funcgraph-tail > trace_options}(hj-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj-ubah}(h]h ]h"]h$]h&]uh1jhj-ubeh}(h]h ]h"]h$]h&]jjuh1jzhhhM hj-ubah}(h]h ]h"]h$]h&]uh1jShhhM hj-ubh)}(h)Example with nofuncgraph-tail (default)::h]h(Example with nofuncgraph-tail (default):}(hj.hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj-ubj)}(h0) | putname() { 0) | kmem_cache_free() { 0) 0.518 us | __phys_addr(); 0) 1.757 us | } 0) 2.861 us | }h]h0) | putname() { 0) | kmem_cache_free() { 0) 0.518 us | __phys_addr(); 0) 1.757 us | } 0) 2.861 us | }}hj.sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj-ubh)}(hExample with funcgraph-tail::h]hExample with funcgraph-tail:}(hj .hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj-ubj)}(h0) | putname() { 0) | kmem_cache_free() { 0) 0.518 us | __phys_addr(); 0) 1.757 us | } /* kmem_cache_free() */ 0) 2.861 us | } /* putname() */h]h0) | putname() { 0) | kmem_cache_free() { 0) 0.518 us | __phys_addr(); 0) 1.757 us | } /* kmem_cache_free() */ 0) 2.861 us | } /* putname() */}hj..sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj-ubeh}(h]h ]h"]h$]h&]uh1jShhhM hjO*hhubh)}(hThe return value of each traced function can be displayed after an equal sign "=". When encountering system call failures, it can be very helpful to quickly locate the function that first returns an error code.h]hThe return value of each traced function can be displayed after an equal sign “=”. When encountering system call failures, it can be very helpful to quickly locate the function that first returns an error code.}(hjB.hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjO*hhubjT)}(hX@ - hide: echo nofuncgraph-retval > trace_options - show: echo funcgraph-retval > trace_options Example with funcgraph-retval:: 1) | cgroup_migrate() { 1) 0.651 us | cgroup_migrate_add_task(); /* = 0xffff93fcfd346c00 */ 1) | cgroup_migrate_execute() { 1) | cpu_cgroup_can_attach() { 1) | cgroup_taskset_first() { 1) 0.732 us | cgroup_taskset_next(); /* = 0xffff93fc8fb20000 */ 1) 1.232 us | } /* cgroup_taskset_first = 0xffff93fc8fb20000 */ 1) 0.380 us | sched_rt_can_attach(); /* = 0x0 */ 1) 2.335 us | } /* cpu_cgroup_can_attach = -22 */ 1) 4.369 us | } /* cgroup_migrate_execute = -22 */ 1) 7.143 us | } /* cgroup_migrate = -22 */ h](jT)}(h^- hide: echo nofuncgraph-retval > trace_options - show: echo funcgraph-retval > trace_options h]j{)}(hhh](j)}(h-hide: echo nofuncgraph-retval > trace_optionsh]h)}(hj].h]h-hide: echo nofuncgraph-retval > trace_options}(hj_.hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj[.ubah}(h]h ]h"]h$]h&]uh1jhjX.ubj)}(h,show: echo funcgraph-retval > trace_options h]h)}(h+show: echo funcgraph-retval > trace_optionsh]h+show: echo funcgraph-retval > trace_options}(hjv.hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjr.ubah}(h]h ]h"]h$]h&]uh1jhjX.ubeh}(h]h ]h"]h$]h&]jjuh1jzhhhM hjT.ubah}(h]h ]h"]h$]h&]uh1jShhhM hjP.ubh)}(hExample with funcgraph-retval::h]hExample with funcgraph-retval:}(hj.hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjP.ubj)}(hX1) | cgroup_migrate() { 1) 0.651 us | cgroup_migrate_add_task(); /* = 0xffff93fcfd346c00 */ 1) | cgroup_migrate_execute() { 1) | cpu_cgroup_can_attach() { 1) | cgroup_taskset_first() { 1) 0.732 us | cgroup_taskset_next(); /* = 0xffff93fc8fb20000 */ 1) 1.232 us | } /* cgroup_taskset_first = 0xffff93fc8fb20000 */ 1) 0.380 us | sched_rt_can_attach(); /* = 0x0 */ 1) 2.335 us | } /* cpu_cgroup_can_attach = -22 */ 1) 4.369 us | } /* cgroup_migrate_execute = -22 */ 1) 7.143 us | } /* cgroup_migrate = -22 */h]hX1) | cgroup_migrate() { 1) 0.651 us | cgroup_migrate_add_task(); /* = 0xffff93fcfd346c00 */ 1) | cgroup_migrate_execute() { 1) | cpu_cgroup_can_attach() { 1) | cgroup_taskset_first() { 1) 0.732 us | cgroup_taskset_next(); /* = 0xffff93fc8fb20000 */ 1) 1.232 us | } /* cgroup_taskset_first = 0xffff93fc8fb20000 */ 1) 0.380 us | sched_rt_can_attach(); /* = 0x0 */ 1) 2.335 us | } /* cpu_cgroup_can_attach = -22 */ 1) 4.369 us | } /* cgroup_migrate_execute = -22 */ 1) 7.143 us | } /* cgroup_migrate = -22 */}hj.sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hjP.ubeh}(h]h ]h"]h$]h&]uh1jShhhM hjO*hhubh)}(hThe above example shows that the function cpu_cgroup_can_attach returned the error code -22 firstly, then we can read the code of this function to get the root cause.h]hThe above example shows that the function cpu_cgroup_can_attach returned the error code -22 firstly, then we can read the code of this function to get the root cause.}(hj.hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjO*hhubh)}(hWhen the option funcgraph-retval-hex is not set, the return value can be displayed in a smart way. Specifically, if it is an error code, it will be printed in signed decimal format, otherwise it will printed in hexadecimal format.h]hWhen the option funcgraph-retval-hex is not set, the return value can be displayed in a smart way. Specifically, if it is an error code, it will be printed in signed decimal format, otherwise it will printed in hexadecimal format.}(hj.hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjO*hhubjT)}(hX - smart: echo nofuncgraph-retval-hex > trace_options - hexadecimal: echo funcgraph-retval-hex > trace_options Example with funcgraph-retval-hex:: 1) | cgroup_migrate() { 1) 0.651 us | cgroup_migrate_add_task(); /* = 0xffff93fcfd346c00 */ 1) | cgroup_migrate_execute() { 1) | cpu_cgroup_can_attach() { 1) | cgroup_taskset_first() { 1) 0.732 us | cgroup_taskset_next(); /* = 0xffff93fc8fb20000 */ 1) 1.232 us | } /* cgroup_taskset_first = 0xffff93fc8fb20000 */ 1) 0.380 us | sched_rt_can_attach(); /* = 0x0 */ 1) 2.335 us | } /* cpu_cgroup_can_attach = 0xffffffea */ 1) 4.369 us | } /* cgroup_migrate_execute = 0xffffffea */ 1) 7.143 us | } /* cgroup_migrate = 0xffffffea */ h](jT)}(hn- smart: echo nofuncgraph-retval-hex > trace_options - hexadecimal: echo funcgraph-retval-hex > trace_options h]j{)}(hhh](j)}(h2smart: echo nofuncgraph-retval-hex > trace_optionsh]h)}(hj.h]h2smart: echo nofuncgraph-retval-hex > trace_options}(hj.hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj.ubah}(h]h ]h"]h$]h&]uh1jhj.ubj)}(h7hexadecimal: echo funcgraph-retval-hex > trace_options h]h)}(h6hexadecimal: echo funcgraph-retval-hex > trace_optionsh]h6hexadecimal: echo funcgraph-retval-hex > trace_options}(hj.hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj.ubah}(h]h ]h"]h$]h&]uh1jhj.ubeh}(h]h ]h"]h$]h&]jjuh1jzhhhM hj.ubah}(h]h ]h"]h$]h&]uh1jShhhM hj.ubh)}(h#Example with funcgraph-retval-hex::h]h"Example with funcgraph-retval-hex:}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj.ubj)}(hX1) | cgroup_migrate() { 1) 0.651 us | cgroup_migrate_add_task(); /* = 0xffff93fcfd346c00 */ 1) | cgroup_migrate_execute() { 1) | cpu_cgroup_can_attach() { 1) | cgroup_taskset_first() { 1) 0.732 us | cgroup_taskset_next(); /* = 0xffff93fc8fb20000 */ 1) 1.232 us | } /* cgroup_taskset_first = 0xffff93fc8fb20000 */ 1) 0.380 us | sched_rt_can_attach(); /* = 0x0 */ 1) 2.335 us | } /* cpu_cgroup_can_attach = 0xffffffea */ 1) 4.369 us | } /* cgroup_migrate_execute = 0xffffffea */ 1) 7.143 us | } /* cgroup_migrate = 0xffffffea */h]hX1) | cgroup_migrate() { 1) 0.651 us | cgroup_migrate_add_task(); /* = 0xffff93fcfd346c00 */ 1) | cgroup_migrate_execute() { 1) | cpu_cgroup_can_attach() { 1) | cgroup_taskset_first() { 1) 0.732 us | cgroup_taskset_next(); /* = 0xffff93fc8fb20000 */ 1) 1.232 us | } /* cgroup_taskset_first = 0xffff93fc8fb20000 */ 1) 0.380 us | sched_rt_can_attach(); /* = 0x0 */ 1) 2.335 us | } /* cpu_cgroup_can_attach = 0xffffffea */ 1) 4.369 us | } /* cgroup_migrate_execute = 0xffffffea */ 1) 7.143 us | } /* cgroup_migrate = 0xffffffea */}hj(/sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj.ubeh}(h]h ]h"]h$]h&]uh1jShhhM hjO*hhubh)}(hAt present, there are some limitations when using the funcgraph-retval option, and these limitations will be eliminated in the future:h]hAt present, there are some limitations when using the funcgraph-retval option, and these limitations will be eliminated in the future:}(hj RET If you pass 0x123456789abcdef to this function and want to narrow it, it may be recorded as 0x123456789abcdef instead of 0xef. **Case Two**: The function error_if_not_4g_aligned is defined as follows:: int error_if_not_4g_aligned(u64 val) { if (val & GENMASK(31, 0)) return -EINVAL; return 0; } It could be compiled to:: error_if_not_4g_aligned: CBNZ w0, .Lnot_aligned RET // bits [31:0] are zero, bits // [63:32] are UNKNOWN .Lnot_aligned: MOV x0, #-EINVAL RET When passing 0x2_0000_0000 to it, the return value may be recorded as 0x2_0000_0000 instead of 0. h](h)}(hXIn certain procedure call standards, such as arm64's AAPCS64, when a type is smaller than a GPR, it is the responsibility of the consumer to perform the narrowing, and the upper bits may contain UNKNOWN values. Therefore, it is advisable to check the code for such cases. For instance, when using a u8 in a 64-bit GPR, bits [63:8] may contain arbitrary values, especially when larger types are truncated, whether explicitly or implicitly. Here are some specific cases to illustrate this point:h]hXIn certain procedure call standards, such as arm64’s AAPCS64, when a type is smaller than a GPR, it is the responsibility of the consumer to perform the narrowing, and the upper bits may contain UNKNOWN values. Therefore, it is advisable to check the code for such cases. For instance, when using a u8 in a 64-bit GPR, bits [63:8] may contain arbitrary values, especially when larger types are truncated, whether explicitly or implicitly. Here are some specific cases to illustrate this point:}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj}/ubh)}(h **Case One**:h](hstrong)}(h **Case One**h]hCase One}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1j/hj/ubh:}(hj/hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM" hj}/ubh)}(h1The function narrow_to_u8 is defined as follows::h]h0The function narrow_to_u8 is defined as follows:}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM$ hj}/ubj)}(hPu8 narrow_to_u8(u64 val) { // implicitly truncated return val; }h]hPu8 narrow_to_u8(u64 val) { // implicitly truncated return val; }}hj/sbah}(h]h ]h"]h$]h&]jjuh1jhhhM& hj}/ubh)}(hIt may be compiled to::h]hIt may be compiled to:}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM, hj}/ubj)}(hDnarrow_to_u8: < ... ftrace instrumentation ... > RETh]hDnarrow_to_u8: < ... ftrace instrumentation ... > RET}hj/sbah}(h]h ]h"]h$]h&]jjuh1jhhhM. hj}/ubh)}(h~If you pass 0x123456789abcdef to this function and want to narrow it, it may be recorded as 0x123456789abcdef instead of 0xef.h]h~If you pass 0x123456789abcdef to this function and want to narrow it, it may be recorded as 0x123456789abcdef instead of 0xef.}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM2 hj}/ubh)}(h **Case Two**:h](j/)}(h **Case Two**h]hCase Two}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1j/hj/ubh:}(hj/hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM5 hj}/ubh)}(h and call trace_printk() inside __might_sleep()::h]hYou can put some comments on specific functions by using trace_printk() For example, if you want to put a comment inside the __might_sleep() function, you just have to include and call trace_printk() inside __might_sleep():}(hja0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMN hjO*hhubj)}(h trace_printk("I'm a comment!\n")h]h trace_printk("I'm a comment!\n")}hjo0sbah}(h]h ]h"]h$]h&]jjuh1jhhhMS hjO*hhubh)}(hwill produce::h]h will produce:}(hj}0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMU hjO*hhubj)}(h1) | __might_sleep() { 1) | /* I'm a comment! */ 1) 1.449 us | }h]h1) | __might_sleep() { 1) | /* I'm a comment! */ 1) 1.449 us | }}hj0sbah}(h]h ]h"]h$]h&]jjuh1jhhhMW hjO*hhubh)}(hYou might find other useful features for this tracer in the following "dynamic ftrace" section such as tracing only specific functions or tasks.h]hYou might find other useful features for this tracer in the following “dynamic ftrace” section such as tracing only specific functions or tasks.}(hj0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM\ hjO*hhubeh}(h]function-graph-tracerah ]h"]function graph tracerah$]h&]uh1hhhhhhhhM ubh)}(hhh](h)}(hdynamic ftraceh]hdynamic ftrace}(hj0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj0hhhhhMa ubh)}(hXiIf CONFIG_DYNAMIC_FTRACE is set, the system will run with virtually no overhead when function tracing is disabled. The way this works is the mcount function call (placed at the start of every kernel function, produced by the -pg switch in gcc), starts of pointing to a simple return. (Enabling FTRACE will include the -pg switch in the compiling of the kernel.)h]hXiIf CONFIG_DYNAMIC_FTRACE is set, the system will run with virtually no overhead when function tracing is disabled. The way this works is the mcount function call (placed at the start of every kernel function, produced by the -pg switch in gcc), starts of pointing to a simple return. (Enabling FTRACE will include the -pg switch in the compiling of the kernel.)}(hj0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMc hj0hhubh)}(hXAt compile time every C file object is run through the recordmcount program (located in the scripts directory). This program will parse the ELF headers in the C object to find all the locations in the .text section that call mcount. Starting with gcc version 4.6, the -mfentry has been added for x86, which calls "__fentry__" instead of "mcount". Which is called before the creation of the stack frame.h]hXAt compile time every C file object is run through the recordmcount program (located in the scripts directory). This program will parse the ELF headers in the C object to find all the locations in the .text section that call mcount. Starting with gcc version 4.6, the -mfentry has been added for x86, which calls “__fentry__” instead of “mcount”. Which is called before the creation of the stack frame.}(hj0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMj hj0hhubh)}(hNote, not all sections are traced. They may be prevented by either a notrace, or blocked another way and all inline functions are not traced. Check the "available_filter_functions" file to see what functions can be traced.h]hNote, not all sections are traced. They may be prevented by either a notrace, or blocked another way and all inline functions are not traced. Check the “available_filter_functions” file to see what functions can be traced.}(hj0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMr hj0hhubh)}(hX!A section called "__mcount_loc" is created that holds references to all the mcount/fentry call sites in the .text section. The recordmcount program re-links this section back into the original object. The final linking stage of the kernel will add all these references into a single table.h]hX%A section called “__mcount_loc” is created that holds references to all the mcount/fentry call sites in the .text section. The recordmcount program re-links this section back into the original object. The final linking stage of the kernel will add all these references into a single table.}(hj0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMw hj0hhubh)}(hXOn boot up, before SMP is initialized, the dynamic ftrace code scans this table and updates all the locations into nops. It also records the locations, which are added to the available_filter_functions list. Modules are processed as they are loaded and before they are executed. When a module is unloaded, it also removes its functions from the ftrace function list. This is automatic in the module unload code, and the module author does not need to worry about it.h]hXOn boot up, before SMP is initialized, the dynamic ftrace code scans this table and updates all the locations into nops. It also records the locations, which are added to the available_filter_functions list. Modules are processed as they are loaded and before they are executed. When a module is unloaded, it also removes its functions from the ftrace function list. This is automatic in the module unload code, and the module author does not need to worry about it.}(hj0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM} hj0hhubh)}(hXWhen tracing is enabled, the process of modifying the function tracepoints is dependent on architecture. The old method is to use kstop_machine to prevent races with the CPUs executing code being modified (which can cause the CPU to do undesirable things, especially if the modified code crosses cache (or page) boundaries), and the nops are patched back to calls. But this time, they do not call mcount (which is just a function stub). They now call into the ftrace infrastructure.h]hXWhen tracing is enabled, the process of modifying the function tracepoints is dependent on architecture. The old method is to use kstop_machine to prevent races with the CPUs executing code being modified (which can cause the CPU to do undesirable things, especially if the modified code crosses cache (or page) boundaries), and the nops are patched back to calls. But this time, they do not call mcount (which is just a function stub). They now call into the ftrace infrastructure.}(hj1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubh)}(hX%The new method of modifying the function tracepoints is to place a breakpoint at the location to be modified, sync all CPUs, modify the rest of the instruction not covered by the breakpoint. Sync all CPUs again, and then remove the breakpoint with the finished version to the ftrace call site.h]hX%The new method of modifying the function tracepoints is to place a breakpoint at the location to be modified, sync all CPUs, modify the rest of the instruction not covered by the breakpoint. Sync all CPUs again, and then remove the breakpoint with the finished version to the ftrace call site.}(hj1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubh)}(hSome archs do not even need to monkey around with the synchronization, and can just slap the new code on top of the old without any problems with other CPUs executing it at the same time.h]hSome archs do not even need to monkey around with the synchronization, and can just slap the new code on top of the old without any problems with other CPUs executing it at the same time.}(hj"1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubh)}(hOne special side-effect to the recording of the functions being traced is that we can now selectively choose which functions we wish to trace and which ones we want the mcount calls to remain as nops.h]hOne special side-effect to the recording of the functions being traced is that we can now selectively choose which functions we wish to trace and which ones we want the mcount calls to remain as nops.}(hj01hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubh)}(hhTwo files are used, one for enabling and one for disabling the tracing of specified functions. They are:h]hhTwo files are used, one for enabling and one for disabling the tracing of specified functions. They are:}(hj>1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubjT)}(hset_ftrace_filter h]h)}(hset_ftrace_filterh]hset_ftrace_filter}(hjP1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjL1ubah}(h]h ]h"]h$]h&]uh1jShhhM hj0hhubh)}(handh]hand}(hjd1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubjT)}(hset_ftrace_notrace h]h)}(hset_ftrace_notraceh]hset_ftrace_notrace}(hjv1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjr1ubah}(h]h ]h"]h$]h&]uh1jShhhM hj0hhubh)}(hKA list of available functions that you can add to these files is listed in:h]hKA list of available functions that you can add to these files is listed in:}(hj1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubjT)}(havailable_filter_functions h]h)}(havailable_filter_functionsh]havailable_filter_functions}(hj1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj1ubah}(h]h ]h"]h$]h&]uh1jShhhM hj0hhubj)}(h# cat available_filter_functions put_prev_task_idle kmem_cache_create pick_next_task_rt cpus_read_lock pick_next_task_fair mutex_lock [...]h]h# cat available_filter_functions put_prev_task_idle kmem_cache_create pick_next_task_rt cpus_read_lock pick_next_task_fair mutex_lock [...]}hj1sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj0hhubh)}(h@If I am only interested in sys_nanosleep and hrtimer_interrupt::h]h?If I am only interested in sys_nanosleep and hrtimer_interrupt:}(hj1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubj)}(hX# echo sys_nanosleep hrtimer_interrupt > set_ftrace_filter # echo function > current_tracer # echo 1 > tracing_on # usleep 1 # echo 0 > tracing_on # cat trace # tracer: function # # entries-in-buffer/entries-written: 5/5 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | usleep-2665 [001] .... 4186.475355: sys_nanosleep <-system_call_fastpath -0 [001] d.h1 4186.475409: hrtimer_interrupt <-smp_apic_timer_interrupt usleep-2665 [001] d.h1 4186.475426: hrtimer_interrupt <-smp_apic_timer_interrupt -0 [003] d.h1 4186.475426: hrtimer_interrupt <-smp_apic_timer_interrupt -0 [002] d.h1 4186.475427: hrtimer_interrupt <-smp_apic_timer_interrupth]hX# echo sys_nanosleep hrtimer_interrupt > set_ftrace_filter # echo function > current_tracer # echo 1 > tracing_on # usleep 1 # echo 0 > tracing_on # cat trace # tracer: function # # entries-in-buffer/entries-written: 5/5 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | usleep-2665 [001] .... 4186.475355: sys_nanosleep <-system_call_fastpath -0 [001] d.h1 4186.475409: hrtimer_interrupt <-smp_apic_timer_interrupt usleep-2665 [001] d.h1 4186.475426: hrtimer_interrupt <-smp_apic_timer_interrupt -0 [003] d.h1 4186.475426: hrtimer_interrupt <-smp_apic_timer_interrupt -0 [002] d.h1 4186.475427: hrtimer_interrupt <-smp_apic_timer_interrupt}hj1sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj0hhubh)}(hATo see which functions are being traced, you can cat the file: ::h]h>To see which functions are being traced, you can cat the file:}(hj1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubj)}(h7# cat set_ftrace_filter hrtimer_interrupt sys_nanosleeph]h7# cat set_ftrace_filter hrtimer_interrupt sys_nanosleep}hj1sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj0hhubh)}(hDPerhaps this is not enough. The filters also allow glob(7) matching.h]hDPerhaps this is not enough. The filters also allow glob(7) matching.}(hj1hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubjT)}(hX``*`` will match functions that begin with ``*`` will match functions that end with ``**`` will match functions that have in it ``*`` will match functions that begin with and end with h]jT )}(hhh](jY )}(h9``*`` will match functions that begin with h](j_ )}(h ``*``h]hliteral)}(hj2h]h*}(hj2hhhNhNubah}(h]h ]h"]h$]h&]uh1j2hj2ubah}(h]h ]h"]h$]h&]uh1j^ hhhM hj 2ubjo )}(hhh]h)}(h,will match functions that begin with h]h,will match functions that begin with }(hj+2hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj(2ubah}(h]h ]h"]h$]h&]uh1jn hj 2ubeh}(h]h ]h"]h$]h&]uh1jX hhhM hj2ubjY )}(h7``*`` will match functions that end with h](j_ )}(h ``*``h]j2)}(hjK2h]h*}(hjM2hhhNhNubah}(h]h ]h"]h$]h&]uh1j2hjI2ubah}(h]h ]h"]h$]h&]uh1j^ hhhM hjE2ubjo )}(hhh]h)}(h*will match functions that end with h]h*will match functions that end with }(hjc2hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj`2ubah}(h]h ]h"]h$]h&]uh1jn hjE2ubeh}(h]h ]h"]h$]h&]uh1jX hhhM hj2ubjY )}(h:``**`` will match functions that have in ith](j_ )}(h ``**``h]j2)}(hj2h]h **}(hj2hhhNhNubah}(h]h ]h"]h$]h&]uh1j2hj2ubah}(h]h ]h"]h$]h&]uh1j^ hhhM hj}2ubjo )}(hhh]h)}(h,will match functions that have in ith]h,will match functions that have in it}(hj2hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj2ubah}(h]h ]h"]h$]h&]uh1jn hj}2ubeh}(h]h ]h"]h$]h&]uh1jX hhhM hj2ubjY )}(hZ``*`` will match functions that begin with and end with h](j_ )}(h``*``h]j2)}(hj2h]h*}(hj2hhhNhNubah}(h]h ]h"]h$]h&]uh1j2hj2ubah}(h]h ]h"]h$]h&]uh1j^ hhhM hj2ubjo )}(hhh]h)}(hCwill match functions that begin with and end with h]hCwill match functions that begin with and end with }(hj2hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj2ubah}(h]h ]h"]h$]h&]uh1jn hj2ubeh}(h]h ]h"]h$]h&]uh1jX hhhM hj2ubeh}(h]h ]h"]h$]h&]uh1jS hj2ubah}(h]h ]h"]h$]h&]uh1jShhhM hj0hhubhnote)}(hIt is better to use quotes to enclose the wild cards, otherwise the shell may expand the parameters into names of files in the local directory.h]h)}(hIt is better to use quotes to enclose the wild cards, otherwise the shell may expand the parameters into names of files in the local directory.h]hIt is better to use quotes to enclose the wild cards, otherwise the shell may expand the parameters into names of files in the local directory.}(hj2hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj2ubah}(h]h ]h"]h$]h&]uh1j2hj0hhhhhNubj)}(h&# echo 'hrtimer_*' > set_ftrace_filterh]h&# echo 'hrtimer_*' > set_ftrace_filter}hj3sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj0hhubh)}(h Produces::h]h Produces:}(hj!3hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubj)}(hX# tracer: function # # entries-in-buffer/entries-written: 897/897 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | -0 [003] dN.1 4228.547803: hrtimer_cancel <-tick_nohz_idle_exit -0 [003] dN.1 4228.547804: hrtimer_try_to_cancel <-hrtimer_cancel -0 [003] dN.2 4228.547805: hrtimer_force_reprogram <-__remove_hrtimer -0 [003] dN.1 4228.547805: hrtimer_forward <-tick_nohz_idle_exit -0 [003] dN.1 4228.547805: hrtimer_start_range_ns <-hrtimer_start_expires.constprop.11 -0 [003] d..1 4228.547858: hrtimer_get_next_event <-get_next_timer_interrupt -0 [003] d..1 4228.547859: hrtimer_start <-__tick_nohz_idle_enter -0 [003] d..2 4228.547860: hrtimer_force_reprogram <-__remh]hX# tracer: function # # entries-in-buffer/entries-written: 897/897 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | -0 [003] dN.1 4228.547803: hrtimer_cancel <-tick_nohz_idle_exit -0 [003] dN.1 4228.547804: hrtimer_try_to_cancel <-hrtimer_cancel -0 [003] dN.2 4228.547805: hrtimer_force_reprogram <-__remove_hrtimer -0 [003] dN.1 4228.547805: hrtimer_forward <-tick_nohz_idle_exit -0 [003] dN.1 4228.547805: hrtimer_start_range_ns <-hrtimer_start_expires.constprop.11 -0 [003] d..1 4228.547858: hrtimer_get_next_event <-get_next_timer_interrupt -0 [003] d..1 4228.547859: hrtimer_start <-__tick_nohz_idle_enter -0 [003] d..2 4228.547860: hrtimer_force_reprogram <-__rem}hj/3sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj0hhubh)}(h)Notice that we lost the sys_nanosleep. ::h]h&Notice that we lost the sys_nanosleep.}(hj=3hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubj)}(hX># cat set_ftrace_filter hrtimer_run_queues hrtimer_run_pending hrtimer_setup hrtimer_cancel hrtimer_try_to_cancel hrtimer_forward hrtimer_start hrtimer_reprogram hrtimer_force_reprogram hrtimer_get_next_event hrtimer_interrupt hrtimer_nanosleep hrtimer_wakeup hrtimer_get_remaining hrtimer_get_res hrtimer_init_sleeperh]hX># cat set_ftrace_filter hrtimer_run_queues hrtimer_run_pending hrtimer_setup hrtimer_cancel hrtimer_try_to_cancel hrtimer_forward hrtimer_start hrtimer_reprogram hrtimer_force_reprogram hrtimer_get_next_event hrtimer_interrupt hrtimer_nanosleep hrtimer_wakeup hrtimer_get_remaining hrtimer_get_res hrtimer_init_sleeper}hjK3sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj0hhubh)}(hThis is because the '>' and '>>' act just like they do in bash. To rewrite the filters, use '>' To append to the filters, use '>>'h]hThis is because the ‘>’ and ‘>>’ act just like they do in bash. To rewrite the filters, use ‘>’ To append to the filters, use ‘>>’}(hjY3hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubh)}(hDTo clear out a filter so that all functions will be recorded again::h]hCTo clear out a filter so that all functions will be recorded again:}(hjg3hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj0hhubj)}(h4# echo > set_ftrace_filter # cat set_ftrace_filter #h]h4# echo > set_ftrace_filter # cat set_ftrace_filter #}hju3sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj0hhubh)}(hAgain, now we want to append.h]hAgain, now we want to append.}(hj3hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM# hj0hhubj)}(hX# echo sys_nanosleep > set_ftrace_filter # cat set_ftrace_filter sys_nanosleep # echo 'hrtimer_*' >> set_ftrace_filter # cat set_ftrace_filter hrtimer_run_queues hrtimer_run_pending hrtimer_setup hrtimer_cancel hrtimer_try_to_cancel hrtimer_forward hrtimer_start hrtimer_reprogram hrtimer_force_reprogram hrtimer_get_next_event hrtimer_interrupt sys_nanosleep hrtimer_nanosleep hrtimer_wakeup hrtimer_get_remaining hrtimer_get_res hrtimer_init_sleeperh]hX# echo sys_nanosleep > set_ftrace_filter # cat set_ftrace_filter sys_nanosleep # echo 'hrtimer_*' >> set_ftrace_filter # cat set_ftrace_filter hrtimer_run_queues hrtimer_run_pending hrtimer_setup hrtimer_cancel hrtimer_try_to_cancel hrtimer_forward hrtimer_start hrtimer_reprogram hrtimer_force_reprogram hrtimer_get_next_event hrtimer_interrupt sys_nanosleep hrtimer_nanosleep hrtimer_wakeup hrtimer_get_remaining hrtimer_get_res hrtimer_init_sleeper}hj3sbah}(h]h ]h"]h$]h&]jjuh1jhhhM' hj0hhubh)}(hEThe set_ftrace_notrace prevents those functions from being traced. ::h]hBThe set_ftrace_notrace prevents those functions from being traced.}(hj3hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM? hj0hhubj)}(h0# echo '*preempt*' '*lock*' > set_ftrace_notraceh]h0# echo '*preempt*' '*lock*' > set_ftrace_notrace}hj3sbah}(h]h ]h"]h$]h&]jjuh1jhhhMC hj0hhubh)}(h Produces::h]h Produces:}(hj3hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhME hj0hhubj)}(hX# tracer: function # # entries-in-buffer/entries-written: 39608/39608 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | bash-1994 [000] .... 4342.324896: file_ra_state_init <-do_dentry_open bash-1994 [000] .... 4342.324897: open_check_o_direct <-do_last bash-1994 [000] .... 4342.324897: ima_file_check <-do_last bash-1994 [000] .... 4342.324898: process_measurement <-ima_file_check bash-1994 [000] .... 4342.324898: ima_get_action <-process_measurement bash-1994 [000] .... 4342.324898: ima_match_policy <-ima_get_action bash-1994 [000] .... 4342.324899: do_truncate <-do_last bash-1994 [000] .... 4342.324899: setattr_should_drop_suidgid <-do_truncate bash-1994 [000] .... 4342.324899: notify_change <-do_truncate bash-1994 [000] .... 4342.324900: current_fs_time <-notify_change bash-1994 [000] .... 4342.324900: current_kernel_time <-current_fs_time bash-1994 [000] .... 4342.324900: timespec_trunc <-current_fs_timeh]hX# tracer: function # # entries-in-buffer/entries-written: 39608/39608 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | bash-1994 [000] .... 4342.324896: file_ra_state_init <-do_dentry_open bash-1994 [000] .... 4342.324897: open_check_o_direct <-do_last bash-1994 [000] .... 4342.324897: ima_file_check <-do_last bash-1994 [000] .... 4342.324898: process_measurement <-ima_file_check bash-1994 [000] .... 4342.324898: ima_get_action <-process_measurement bash-1994 [000] .... 4342.324898: ima_match_policy <-ima_get_action bash-1994 [000] .... 4342.324899: do_truncate <-do_last bash-1994 [000] .... 4342.324899: setattr_should_drop_suidgid <-do_truncate bash-1994 [000] .... 4342.324899: notify_change <-do_truncate bash-1994 [000] .... 4342.324900: current_fs_time <-notify_change bash-1994 [000] .... 4342.324900: current_kernel_time <-current_fs_time bash-1994 [000] .... 4342.324900: timespec_trunc <-current_fs_time}hj3sbah}(h]h ]h"]h$]h&]jjuh1jhhhMG hj0hhubh)}(h8We can see that there's no more lock or preempt tracing.h]h:We can see that there’s no more lock or preempt tracing.}(hj3hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM_ hj0hhubeh}(h]dynamic-ftraceah ]h"]dynamic ftraceah$]h&]uh1hhhhhhhhMa ubh)}(hhh](h)}(h$Selecting function filters via indexh]h$Selecting function filters via index}(hj3hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj3hhhhhMb ubh)}(hXBecause processing of strings is expensive (the address of the function needs to be looked up before comparing to the string being passed in), an index can be used as well to enable functions. This is useful in the case of setting thousands of specific functions at a time. By passing in a list of numbers, no string processing will occur. Instead, the function at the specific location in the internal array (which corresponds to the functions in the "available_filter_functions" file), is selected.h]hXBecause processing of strings is expensive (the address of the function needs to be looked up before comparing to the string being passed in), an index can be used as well to enable functions. This is useful in the case of setting thousands of specific functions at a time. By passing in a list of numbers, no string processing will occur. Instead, the function at the specific location in the internal array (which corresponds to the functions in the “available_filter_functions” file), is selected.}(hj3hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMd hj3hhubj)}(h# echo 1 > set_ftrace_filterh]h# echo 1 > set_ftrace_filter}hj 4sbah}(h]h ]h"]h$]h&]jjuh1jhhhMn hj3hhubh)}(hEWill select the first function listed in "available_filter_functions"h]hIWill select the first function listed in “available_filter_functions”}(hj4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMp hj3hhubj)}(hX# head -1 available_filter_functions trace_initcall_finish_cb # cat set_ftrace_filter trace_initcall_finish_cb # head -50 available_filter_functions | tail -1 x86_pmu_commit_txn # echo 1 50 > set_ftrace_filter # cat set_ftrace_filter trace_initcall_finish_cb x86_pmu_commit_txnh]hX# head -1 available_filter_functions trace_initcall_finish_cb # cat set_ftrace_filter trace_initcall_finish_cb # head -50 available_filter_functions | tail -1 x86_pmu_commit_txn # echo 1 50 > set_ftrace_filter # cat set_ftrace_filter trace_initcall_finish_cb x86_pmu_commit_txn}hj(4sbah}(h]h ]h"]h$]h&]jjuh1jhhhMt hj3hhubeh}(h]$selecting-function-filters-via-indexah ]h"]$selecting function filters via indexah$]h&]uh1hhhhhhhhMb ubh)}(hhh](h)}(h-Dynamic ftrace with the function graph tracerh]h-Dynamic ftrace with the function graph tracer}(hjA4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj>4hhhhhM ubh)}(hAlthough what has been explained above concerns both the function tracer and the function-graph-tracer, there are some special features only available in the function-graph tracer.h]hAlthough what has been explained above concerns both the function tracer and the function-graph-tracer, there are some special features only available in the function-graph tracer.}(hjO4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj>4hhubh)}(hxIf you want to trace only one function and all of its children, you just have to echo its name into set_graph_function::h]hwIf you want to trace only one function and all of its children, you just have to echo its name into set_graph_function:}(hj]4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj>4hhubj)}(h$echo __do_fault > set_graph_functionh]h$echo __do_fault > set_graph_function}hjk4sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj>4hhubh)}(hJwill produce the following "expanded" trace of the __do_fault() function::h]hMwill produce the following “expanded” trace of the __do_fault() function:}(hjy4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj>4hhubj)}(hX0) | __do_fault() { 0) | filemap_fault() { 0) | find_lock_page() { 0) 0.804 us | find_get_page(); 0) | __might_sleep() { 0) 1.329 us | } 0) 3.904 us | } 0) 4.979 us | } 0) 0.653 us | _spin_lock(); 0) 0.578 us | page_add_file_rmap(); 0) 0.525 us | native_set_pte_at(); 0) 0.585 us | _spin_unlock(); 0) | unlock_page() { 0) 0.541 us | page_waitqueue(); 0) 0.639 us | __wake_up_bit(); 0) 2.786 us | } 0) + 14.237 us | } 0) | __do_fault() { 0) | filemap_fault() { 0) | find_lock_page() { 0) 0.698 us | find_get_page(); 0) | __might_sleep() { 0) 1.412 us | } 0) 3.950 us | } 0) 5.098 us | } 0) 0.631 us | _spin_lock(); 0) 0.571 us | page_add_file_rmap(); 0) 0.526 us | native_set_pte_at(); 0) 0.586 us | _spin_unlock(); 0) | unlock_page() { 0) 0.533 us | page_waitqueue(); 0) 0.638 us | __wake_up_bit(); 0) 2.793 us | } 0) + 14.012 us | }h]hX0) | __do_fault() { 0) | filemap_fault() { 0) | find_lock_page() { 0) 0.804 us | find_get_page(); 0) | __might_sleep() { 0) 1.329 us | } 0) 3.904 us | } 0) 4.979 us | } 0) 0.653 us | _spin_lock(); 0) 0.578 us | page_add_file_rmap(); 0) 0.525 us | native_set_pte_at(); 0) 0.585 us | _spin_unlock(); 0) | unlock_page() { 0) 0.541 us | page_waitqueue(); 0) 0.639 us | __wake_up_bit(); 0) 2.786 us | } 0) + 14.237 us | } 0) | __do_fault() { 0) | filemap_fault() { 0) | find_lock_page() { 0) 0.698 us | find_get_page(); 0) | __might_sleep() { 0) 1.412 us | } 0) 3.950 us | } 0) 5.098 us | } 0) 0.631 us | _spin_lock(); 0) 0.571 us | page_add_file_rmap(); 0) 0.526 us | native_set_pte_at(); 0) 0.586 us | _spin_unlock(); 0) | unlock_page() { 0) 0.533 us | page_waitqueue(); 0) 0.638 us | __wake_up_bit(); 0) 2.793 us | } 0) + 14.012 us | }}hj4sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj>4hhubh)}(h/You can also expand several functions at once::h]h.You can also expand several functions at once:}(hj4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj>4hhubj)}(hGecho sys_open > set_graph_function echo sys_close >> set_graph_functionh]hGecho sys_open > set_graph_function echo sys_close >> set_graph_function}hj4sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj>4hhubh)}(hYNow if you want to go back to trace all functions you can clear this special filter via::h]hXNow if you want to go back to trace all functions you can clear this special filter via:}(hj4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj>4hhubj)}(hecho > set_graph_functionh]hecho > set_graph_function}hj4sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj>4hhubeh}(h]-dynamic-ftrace-with-the-function-graph-tracerah ]h"]-dynamic ftrace with the function graph tracerah$]h&]uh1hhhhhhhhM ubh)}(hhh](h)}(hftrace_enabledh]hftrace_enabled}(hj4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj4hhhhhM ubh)}(hXNote, the proc sysctl ftrace_enable is a big on/off switch for the function tracer. By default it is enabled (when function tracing is enabled in the kernel). If it is disabled, all function tracing is disabled. This includes not only the function tracers for ftrace, but also for any other uses (perf, kprobes, stack tracing, profiling, etc). It cannot be disabled if there is a callback with FTRACE_OPS_FL_PERMANENT set registered.h]hXNote, the proc sysctl ftrace_enable is a big on/off switch for the function tracer. By default it is enabled (when function tracing is enabled in the kernel). If it is disabled, all function tracing is disabled. This includes not only the function tracers for ftrace, but also for any other uses (perf, kprobes, stack tracing, profiling, etc). It cannot be disabled if there is a callback with FTRACE_OPS_FL_PERMANENT set registered.}(hj4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj4hhubh)}(hPlease disable this with care.h]hPlease disable this with care.}(hj4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj4hhubh)}(h(This can be disable (and enabled) with::h]h'This can be disable (and enabled) with:}(hj5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj4hhubj)}(h sysctl kernel.ftrace_enabled=0 sysctl kernel.ftrace_enabled=1 or echo 0 > /proc/sys/kernel/ftrace_enabled echo 1 > /proc/sys/kernel/ftrace_enabledh]h sysctl kernel.ftrace_enabled=0 sysctl kernel.ftrace_enabled=1 or echo 0 > /proc/sys/kernel/ftrace_enabled echo 1 > /proc/sys/kernel/ftrace_enabled}hj5sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj4hhubeh}(h]ftrace-enabledah ]h"]ftrace_enabledah$]h&]uh1hhhhhhhhM ubh)}(hhh](h)}(hFilter commandsh]hFilter commands}(hj)5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj&5hhhhhM ubh)}(hkA few commands are supported by the set_ftrace_filter interface. Trace commands have the following format::h]hjA few commands are supported by the set_ftrace_filter interface. Trace commands have the following format:}(hj75hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj&5hhubj)}(h ::h]h ::}hjE5sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj&5hhubh)}(h%The following commands are supported:h]h%The following commands are supported:}(hjS5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj&5hhubj{)}(hhh](j)}(hXmod: This command enables function filtering per module. The parameter defines the module. For example, if only the write* functions in the ext3 module are desired, run: echo 'write*:mod:ext3' > set_ftrace_filter This command interacts with the filter in the same way as filtering based on function names. Thus, adding more functions in a different module is accomplished by appending (>>) to the filter file. Remove specific module functions by prepending '!':: echo '!writeback*:mod:ext3' >> set_ftrace_filter Mod command supports module globbing. Disable tracing for all functions except a specific module:: echo '!*:mod:!ext3' >> set_ftrace_filter Disable tracing for all modules, but still trace kernel:: echo '!*:mod:*' >> set_ftrace_filter Enable filter only for kernel:: echo '*write*:mod:!*' >> set_ftrace_filter Enable filter for module globbing:: echo '*write*:mod:*snd*' >> set_ftrace_filter h](h)}(hmod: This command enables function filtering per module. The parameter defines the module. For example, if only the write* functions in the ext3 module are desired, run:h]hmod: This command enables function filtering per module. The parameter defines the module. For example, if only the write* functions in the ext3 module are desired, run:}(hjh5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjd5ubjT)}(h+echo 'write*:mod:ext3' > set_ftrace_filter h]h)}(h*echo 'write*:mod:ext3' > set_ftrace_filterh]h.echo ‘write*:mod:ext3’ > set_ftrace_filter}(hjz5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjv5ubah}(h]h ]h"]h$]h&]uh1jShhhM hjd5ubh)}(hThis command interacts with the filter in the same way as filtering based on function names. Thus, adding more functions in a different module is accomplished by appending (>>) to the filter file. Remove specific module functions by prepending '!'::h]hThis command interacts with the filter in the same way as filtering based on function names. Thus, adding more functions in a different module is accomplished by appending (>>) to the filter file. Remove specific module functions by prepending ‘!’:}(hj5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjd5ubj)}(h0echo '!writeback*:mod:ext3' >> set_ftrace_filterh]h0echo '!writeback*:mod:ext3' >> set_ftrace_filter}hj5sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hjd5ubh)}(hbMod command supports module globbing. Disable tracing for all functions except a specific module::h]haMod command supports module globbing. Disable tracing for all functions except a specific module:}(hj5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjd5ubj)}(h(echo '!*:mod:!ext3' >> set_ftrace_filterh]h(echo '!*:mod:!ext3' >> set_ftrace_filter}hj5sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hjd5ubh)}(h9Disable tracing for all modules, but still trace kernel::h]h8Disable tracing for all modules, but still trace kernel:}(hj5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjd5ubj)}(h$echo '!*:mod:*' >> set_ftrace_filterh]h$echo '!*:mod:*' >> set_ftrace_filter}hj5sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hjd5ubh)}(hEnable filter only for kernel::h]hEnable filter only for kernel:}(hj5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjd5ubj)}(h*echo '*write*:mod:!*' >> set_ftrace_filterh]h*echo '*write*:mod:!*' >> set_ftrace_filter}hj5sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hjd5ubh)}(h#Enable filter for module globbing::h]h"Enable filter for module globbing:}(hj5hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjd5ubj)}(h-echo '*write*:mod:*snd*' >> set_ftrace_filterh]h-echo '*write*:mod:*snd*' >> set_ftrace_filter}hj 6sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hjd5ubeh}(h]h ]h"]h$]h&]uh1jhja5hhhhhNubj)}(hXEtraceon/traceoff: These commands turn tracing on and off when the specified functions are hit. The parameter determines how many times the tracing system is turned on and off. If unspecified, there is no limit. For example, to disable tracing when a schedule bug is hit the first 5 times, run:: echo '__schedule_bug:traceoff:5' > set_ftrace_filter To always disable tracing when __schedule_bug is hit:: echo '__schedule_bug:traceoff' > set_ftrace_filter These commands are cumulative whether or not they are appended to set_ftrace_filter. To remove a command, prepend it by '!' and drop the parameter:: echo '!__schedule_bug:traceoff:0' > set_ftrace_filter The above removes the traceoff command for __schedule_bug that have a counter. To remove commands without counters:: echo '!__schedule_bug:traceoff' > set_ftrace_filter h](h)}(hX&traceon/traceoff: These commands turn tracing on and off when the specified functions are hit. The parameter determines how many times the tracing system is turned on and off. If unspecified, there is no limit. For example, to disable tracing when a schedule bug is hit the first 5 times, run::h]hX%traceon/traceoff: These commands turn tracing on and off when the specified functions are hit. The parameter determines how many times the tracing system is turned on and off. If unspecified, there is no limit. For example, to disable tracing when a schedule bug is hit the first 5 times, run:}(hj$6hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj 6ubj)}(h4echo '__schedule_bug:traceoff:5' > set_ftrace_filterh]h4echo '__schedule_bug:traceoff:5' > set_ftrace_filter}hj26sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj 6ubh)}(h6To always disable tracing when __schedule_bug is hit::h]h5To always disable tracing when __schedule_bug is hit:}(hj@6hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj 6ubj)}(h2echo '__schedule_bug:traceoff' > set_ftrace_filterh]h2echo '__schedule_bug:traceoff' > set_ftrace_filter}hjN6sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj 6ubh)}(hThese commands are cumulative whether or not they are appended to set_ftrace_filter. To remove a command, prepend it by '!' and drop the parameter::h]hThese commands are cumulative whether or not they are appended to set_ftrace_filter. To remove a command, prepend it by ‘!’ and drop the parameter:}(hj\6hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj 6ubj)}(h5echo '!__schedule_bug:traceoff:0' > set_ftrace_filterh]h5echo '!__schedule_bug:traceoff:0' > set_ftrace_filter}hjj6sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj 6ubh)}(htThe above removes the traceoff command for __schedule_bug that have a counter. To remove commands without counters::h]hsThe above removes the traceoff command for __schedule_bug that have a counter. To remove commands without counters:}(hjx6hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj 6ubj)}(h3echo '!__schedule_bug:traceoff' > set_ftrace_filterh]h3echo '!__schedule_bug:traceoff' > set_ftrace_filter}hj6sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj 6ubeh}(h]h ]h"]h$]h&]uh1jhja5hhhhhNubj)}(hXsnapshot: Will cause a snapshot to be triggered when the function is hit. :: echo 'native_flush_tlb_others:snapshot' > set_ftrace_filter To only snapshot once: :: echo 'native_flush_tlb_others:snapshot:1' > set_ftrace_filter To remove the above commands:: echo '!native_flush_tlb_others:snapshot' > set_ftrace_filter echo '!native_flush_tlb_others:snapshot:0' > set_ftrace_filter h](h)}(hLsnapshot: Will cause a snapshot to be triggered when the function is hit. ::h]hIsnapshot: Will cause a snapshot to be triggered when the function is hit.}(hj6hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj6ubj)}(h;echo 'native_flush_tlb_others:snapshot' > set_ftrace_filterh]h;echo 'native_flush_tlb_others:snapshot' > set_ftrace_filter}hj6sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj6ubh)}(hTo only snapshot once: ::h]hTo only snapshot once:}(hj6hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj6ubj)}(h=echo 'native_flush_tlb_others:snapshot:1' > set_ftrace_filterh]h=echo 'native_flush_tlb_others:snapshot:1' > set_ftrace_filter}hj6sbah}(h]h ]h"]h$]h&]jjuh1jhhhM" hj6ubh)}(hTo remove the above commands::h]hTo remove the above commands:}(hj6hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM$ hj6ubj)}(h{echo '!native_flush_tlb_others:snapshot' > set_ftrace_filter echo '!native_flush_tlb_others:snapshot:0' > set_ftrace_filterh]h{echo '!native_flush_tlb_others:snapshot' > set_ftrace_filter echo '!native_flush_tlb_others:snapshot:0' > set_ftrace_filter}hj6sbah}(h]h ]h"]h$]h&]jjuh1jhhhM& hj6ubeh}(h]h ]h"]h$]h&]uh1jhja5hhhhhNubj)}(hX&enable_event/disable_event: These commands can enable or disable a trace event. Note, because function tracing callbacks are very sensitive, when these commands are registered, the trace point is activated, but disabled in a "soft" mode. That is, the tracepoint will be called, but just will not be traced. The event tracepoint stays in this mode as long as there's a command that triggers it. :: echo 'try_to_wake_up:enable_event:sched:sched_switch:2' > \ set_ftrace_filter The format is:: :enable_event::[:count] :disable_event::[:count] To remove the events commands:: echo '!try_to_wake_up:enable_event:sched:sched_switch:0' > \ set_ftrace_filter echo '!schedule:disable_event:sched:sched_switch' > \ set_ftrace_filter h](h)}(hXenable_event/disable_event: These commands can enable or disable a trace event. Note, because function tracing callbacks are very sensitive, when these commands are registered, the trace point is activated, but disabled in a "soft" mode. That is, the tracepoint will be called, but just will not be traced. The event tracepoint stays in this mode as long as there's a command that triggers it. ::h]hXenable_event/disable_event: These commands can enable or disable a trace event. Note, because function tracing callbacks are very sensitive, when these commands are registered, the trace point is activated, but disabled in a “soft” mode. That is, the tracepoint will be called, but just will not be traced. The event tracepoint stays in this mode as long as there’s a command that triggers it.}(hj6hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM) hj6ubj)}(hSecho 'try_to_wake_up:enable_event:sched:sched_switch:2' > \ set_ftrace_filterh]hSecho 'try_to_wake_up:enable_event:sched:sched_switch:2' > \ set_ftrace_filter}hj 7sbah}(h]h ]h"]h$]h&]jjuh1jhhhM2 hj6ubh)}(hThe format is::h]hThe format is:}(hj7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM5 hj6ubj)}(hb:enable_event::[:count] :disable_event::[:count]h]hb:enable_event::[:count] :disable_event::[:count]}hj&7sbah}(h]h ]h"]h$]h&]jjuh1jhhhM7 hj6ubh)}(hTo remove the events commands::h]hTo remove the events commands:}(hj47hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM: hj6ubj)}(hecho '!try_to_wake_up:enable_event:sched:sched_switch:0' > \ set_ftrace_filter echo '!schedule:disable_event:sched:sched_switch' > \ set_ftrace_filterh]hecho '!try_to_wake_up:enable_event:sched:sched_switch:0' > \ set_ftrace_filter echo '!schedule:disable_event:sched:sched_switch' > \ set_ftrace_filter}hjB7sbah}(h]h ]h"]h$]h&]jjuh1jhhhM< hj6ubeh}(h]h ]h"]h$]h&]uh1jhja5hhhhhNubj)}(hXAdump: When the function is hit, it will dump the contents of the ftrace ring buffer to the console. This is useful if you need to debug something, and want to dump the trace when a certain function is hit. Perhaps it's a function that is called before a triple fault happens and does not allow you to get a regular dump. h]h)}(hX@dump: When the function is hit, it will dump the contents of the ftrace ring buffer to the console. This is useful if you need to debug something, and want to dump the trace when a certain function is hit. Perhaps it's a function that is called before a triple fault happens and does not allow you to get a regular dump.h]hXBdump: When the function is hit, it will dump the contents of the ftrace ring buffer to the console. This is useful if you need to debug something, and want to dump the trace when a certain function is hit. Perhaps it’s a function that is called before a triple fault happens and does not allow you to get a regular dump.}(hjZ7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMA hjV7ubah}(h]h ]h"]h$]h&]uh1jhja5hhhhhNubj)}(hX cpudump: When the function is hit, it will dump the contents of the ftrace ring buffer for the current CPU to the console. Unlike the "dump" command, it only prints out the contents of the ring buffer for the CPU that executed the function that triggered the dump. h]h)}(hXcpudump: When the function is hit, it will dump the contents of the ftrace ring buffer for the current CPU to the console. Unlike the "dump" command, it only prints out the contents of the ring buffer for the CPU that executed the function that triggered the dump.h]hX cpudump: When the function is hit, it will dump the contents of the ftrace ring buffer for the current CPU to the console. Unlike the “dump” command, it only prints out the contents of the ring buffer for the CPU that executed the function that triggered the dump.}(hjr7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMH hjn7ubah}(h]h ]h"]h$]h&]uh1jhja5hhhhhNubj)}(hAstacktrace: When the function is hit, a stack trace is recorded. h]h)}(h@stacktrace: When the function is hit, a stack trace is recorded.h]h@stacktrace: When the function is hit, a stack trace is recorded.}(hj7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMN hj7ubah}(h]h ]h"]h$]h&]uh1jhja5hhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jzhhhM hj&5hhubeh}(h]filter-commandsah ]h"]filter commandsah$]h&]uh1hhhhhhhhM ubh)}(hhh](h)}(h trace_pipeh]h trace_pipe}(hj7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj7hhhhhMR ubh)}(hThe trace_pipe outputs the same content as the trace file, but the effect on the tracing is different. Every read from trace_pipe is consumed. This means that subsequent reads will be different. The trace is live. ::h]hThe trace_pipe outputs the same content as the trace file, but the effect on the tracing is different. Every read from trace_pipe is consumed. This means that subsequent reads will be different. The trace is live.}(hj7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMT hj7hhubj)}(hX-# echo function > current_tracer # cat trace_pipe > /tmp/trace.out & [1] 4153 # echo 1 > tracing_on # usleep 1 # echo 0 > tracing_on # cat trace # tracer: function # # entries-in-buffer/entries-written: 0/0 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | # # cat /tmp/trace.out bash-1994 [000] .... 5281.568961: mutex_unlock <-rb_simple_write bash-1994 [000] .... 5281.568963: __mutex_unlock_slowpath <-mutex_unlock bash-1994 [000] .... 5281.568963: __fsnotify_parent <-fsnotify_modify bash-1994 [000] .... 5281.568964: fsnotify <-fsnotify_modify bash-1994 [000] .... 5281.568964: __srcu_read_lock <-fsnotify bash-1994 [000] .... 5281.568964: add_preempt_count <-__srcu_read_lock bash-1994 [000] ...1 5281.568965: sub_preempt_count <-__srcu_read_lock bash-1994 [000] .... 5281.568965: __srcu_read_unlock <-fsnotify bash-1994 [000] .... 5281.568967: sys_dup2 <-system_call_fastpathh]hX-# echo function > current_tracer # cat trace_pipe > /tmp/trace.out & [1] 4153 # echo 1 > tracing_on # usleep 1 # echo 0 > tracing_on # cat trace # tracer: function # # entries-in-buffer/entries-written: 0/0 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | # # cat /tmp/trace.out bash-1994 [000] .... 5281.568961: mutex_unlock <-rb_simple_write bash-1994 [000] .... 5281.568963: __mutex_unlock_slowpath <-mutex_unlock bash-1994 [000] .... 5281.568963: __fsnotify_parent <-fsnotify_modify bash-1994 [000] .... 5281.568964: fsnotify <-fsnotify_modify bash-1994 [000] .... 5281.568964: __srcu_read_lock <-fsnotify bash-1994 [000] .... 5281.568964: add_preempt_count <-__srcu_read_lock bash-1994 [000] ...1 5281.568965: sub_preempt_count <-__srcu_read_lock bash-1994 [000] .... 5281.568965: __srcu_read_unlock <-fsnotify bash-1994 [000] .... 5281.568967: sys_dup2 <-system_call_fastpath}hj7sbah}(h]h ]h"]h$]h&]jjuh1jhhhMZ hj7hhubh)}(hXNote, reading the trace_pipe file will block until more input is added. This is contrary to the trace file. If any process opened the trace file for reading, it will actually disable tracing and prevent new entries from being added. The trace_pipe file does not have this limitation.h]hXNote, reading the trace_pipe file will block until more input is added. This is contrary to the trace file. If any process opened the trace file for reading, it will actually disable tracing and prevent new entries from being added. The trace_pipe file does not have this limitation.}(hj7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMz hj7hhubeh}(h] trace-pipeah ]h"] trace_pipeah$]h&]uh1hhhhhhhhMR ubh)}(hhh](h)}(h trace entriesh]h trace entries}(hj7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj7hhhhhM ubh)}(hXQHaving too much or not enough data can be troublesome in diagnosing an issue in the kernel. The file buffer_size_kb is used to modify the size of the internal trace buffers. The number listed is the number of entries that can be recorded per CPU. To know the full size, multiply the number of possible CPUs with the number of entries. ::h]hXNHaving too much or not enough data can be troublesome in diagnosing an issue in the kernel. The file buffer_size_kb is used to modify the size of the internal trace buffers. The number listed is the number of entries that can be recorded per CPU. To know the full size, multiply the number of possible CPUs with the number of entries.}(hj8hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj7hhubj)}(h+# cat buffer_size_kb 1408 (units kilobytes)h]h+# cat buffer_size_kb 1408 (units kilobytes)}hj8sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj7hhubh)}(h&Or simply read buffer_total_size_kb ::h]h#Or simply read buffer_total_size_kb}(hj8hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj7hhubj)}(h# cat buffer_total_size_kb 5632h]h# cat buffer_total_size_kb 5632}hj*8sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj7hhubh)}(hITo modify the buffer, simple echo in a number (in 1024 byte segments). ::h]hFTo modify the buffer, simple echo in a number (in 1024 byte segments).}(hj88hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj7hhubj)}(hJ# echo 10000 > buffer_size_kb # cat buffer_size_kb 10000 (units kilobytes)h]hJ# echo 10000 > buffer_size_kb # cat buffer_size_kb 10000 (units kilobytes)}hjF8sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj7hhubh)}(hpIt will try to allocate as much as possible. If you allocate too much, it can cause Out-Of-Memory to trigger. ::h]hmIt will try to allocate as much as possible. If you allocate too much, it can cause Out-Of-Memory to trigger.}(hjT8hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj7hhubj)}(hn# echo 1000000000000 > buffer_size_kb -bash: echo: write error: Cannot allocate memory # cat buffer_size_kb 85h]hn# echo 1000000000000 > buffer_size_kb -bash: echo: write error: Cannot allocate memory # cat buffer_size_kb 85}hjb8sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj7hhubh)}(h;The per_cpu buffers can be changed individually as well: ::h]h8The per_cpu buffers can be changed individually as well:}(hjp8hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj7hhubj)}(hS# echo 10000 > per_cpu/cpu0/buffer_size_kb # echo 100 > per_cpu/cpu1/buffer_size_kbh]hS# echo 10000 > per_cpu/cpu0/buffer_size_kb # echo 100 > per_cpu/cpu1/buffer_size_kb}hj~8sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj7hhubh)}(heWhen the per_cpu buffers are not the same, the buffer_size_kb at the top level will just show an X ::h]hbWhen the per_cpu buffers are not the same, the buffer_size_kb at the top level will just show an X}(hj8hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj7hhubj)}(h# cat buffer_size_kb Xh]h# cat buffer_size_kb X}hj8sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj7hhubh)}(h4This is where the buffer_total_size_kb is useful: ::h]h1This is where the buffer_total_size_kb is useful:}(hj8hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj7hhubj)}(h # cat buffer_total_size_kb 12916h]h # cat buffer_total_size_kb 12916}hj8sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj7hhubh)}(hXWriting to the top level buffer_size_kb will reset all the buffers to be the same again.h]hXWriting to the top level buffer_size_kb will reset all the buffers to be the same again.}(hj8hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj7hhubeh}(h] trace-entriesah ]h"] trace entriesah$]h&]uh1hhhhhhhhM ubh)}(hhh](h)}(hSnapshoth]hSnapshot}(hj8hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj8hhhhhM ubh)}(hXCONFIG_TRACER_SNAPSHOT makes a generic snapshot feature available to all non latency tracers. (Latency tracers which record max latency, such as "irqsoff" or "wakeup", can't use this feature, since those are already using the snapshot mechanism internally.)h]hX CONFIG_TRACER_SNAPSHOT makes a generic snapshot feature available to all non latency tracers. (Latency tracers which record max latency, such as “irqsoff” or “wakeup”, can’t use this feature, since those are already using the snapshot mechanism internally.)}(hj8hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj8hhubh)}(hSnapshot preserves a current trace buffer at a particular point in time without stopping tracing. Ftrace swaps the current buffer with a spare buffer, and tracing continues in the new current (=previous spare) buffer.h]hSnapshot preserves a current trace buffer at a particular point in time without stopping tracing. Ftrace swaps the current buffer with a spare buffer, and tracing continues in the new current (=previous spare) buffer.}(hj8hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj8hhubh)}(hEThe following tracefs files in "tracing" are related to this feature:h]hIThe following tracefs files in “tracing” are related to this feature:}(hj9hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj8hhubjT)}(hXsnapshot: This is used to take a snapshot and to read the output of the snapshot. Echo 1 into this file to allocate a spare buffer and to take a snapshot (swap), then read the snapshot from this file in the same format as "trace" (described above in the section "The File System"). Both reads snapshot and tracing are executable in parallel. When the spare buffer is allocated, echoing 0 frees it, and echoing else (positive) values clear the snapshot contents. More details are shown in the table below. +--------------+------------+------------+------------+ |status\\input | 0 | 1 | else | +==============+============+============+============+ |not allocated |(do nothing)| alloc+swap |(do nothing)| +--------------+------------+------------+------------+ |allocated | free | swap | clear | +--------------+------------+------------+------------+ h](h)}(h snapshot:h]h snapshot:}(hj9hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj9ubjT)}(hXxThis is used to take a snapshot and to read the output of the snapshot. Echo 1 into this file to allocate a spare buffer and to take a snapshot (swap), then read the snapshot from this file in the same format as "trace" (described above in the section "The File System"). Both reads snapshot and tracing are executable in parallel. When the spare buffer is allocated, echoing 0 frees it, and echoing else (positive) values clear the snapshot contents. More details are shown in the table below. +--------------+------------+------------+------------+ |status\\input | 0 | 1 | else | +==============+============+============+============+ |not allocated |(do nothing)| alloc+swap |(do nothing)| +--------------+------------+------------+------------+ |allocated | free | swap | clear | +--------------+------------+------------+------------+ h](h)}(hXThis is used to take a snapshot and to read the output of the snapshot. Echo 1 into this file to allocate a spare buffer and to take a snapshot (swap), then read the snapshot from this file in the same format as "trace" (described above in the section "The File System"). Both reads snapshot and tracing are executable in parallel. When the spare buffer is allocated, echoing 0 frees it, and echoing else (positive) values clear the snapshot contents. More details are shown in the table below.h]hXThis is used to take a snapshot and to read the output of the snapshot. Echo 1 into this file to allocate a spare buffer and to take a snapshot (swap), then read the snapshot from this file in the same format as “trace” (described above in the section “The File System”). Both reads snapshot and tracing are executable in parallel. When the spare buffer is allocated, echoing 0 frees it, and echoing else (positive) values clear the snapshot contents. More details are shown in the table below.}(hj+9hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj'9ubhtable)}(hhh]htgroup)}(hhh](hcolspec)}(hhh]h}(h]h ]h"]h$]h&]colwidthKuh1jC9hj@9ubjD9)}(hhh]h}(h]h ]h"]h$]h&]colwidthK uh1jC9hj@9ubjD9)}(hhh]h}(h]h ]h"]h$]h&]colwidthK uh1jC9hj@9ubjD9)}(hhh]h}(h]h ]h"]h$]h&]colwidthK uh1jC9hj@9ubhthead)}(hhh]hrow)}(hhh](hentry)}(hhh]h)}(h status\\inputh]h status\input}(hj|9hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjy9ubah}(h]h ]h"]h$]h&]uh1jw9hjt9ubjx9)}(hhh]h)}(h0h]h0}(hj9hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj9ubah}(h]h ]h"]h$]h&]uh1jw9hjt9ubjx9)}(hhh]h)}(h1h]h1}(hj9hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj9ubah}(h]h ]h"]h$]h&]uh1jw9hjt9ubjx9)}(hhh]h)}(helseh]helse}(hj9hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj9ubah}(h]h ]h"]h$]h&]uh1jw9hjt9ubeh}(h]h ]h"]h$]h&]uh1jr9hjo9ubah}(h]h ]h"]h$]h&]uh1jm9hj@9ubhtbody)}(hhh](js9)}(hhh](jx9)}(hhh]h)}(h not allocatedh]h not allocated}(hj9hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj9ubah}(h]h ]h"]h$]h&]uh1jw9hj9ubjx9)}(hhh]h)}(h (do nothing)h]h (do nothing)}(hj:hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj:ubah}(h]h ]h"]h$]h&]uh1jw9hj9ubjx9)}(hhh]h)}(h alloc+swaph]h alloc+swap}(hj:hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj:ubah}(h]h ]h"]h$]h&]uh1jw9hj9ubjx9)}(hhh]h)}(h (do nothing)h]h (do nothing)}(hj1:hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj.:ubah}(h]h ]h"]h$]h&]uh1jw9hj9ubeh}(h]h ]h"]h$]h&]uh1jr9hj9ubjs9)}(hhh](jx9)}(hhh]h)}(h allocatedh]h allocated}(hjQ:hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjN:ubah}(h]h ]h"]h$]h&]uh1jw9hjK:ubjx9)}(hhh]h)}(hfreeh]hfree}(hjh:hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hje:ubah}(h]h ]h"]h$]h&]uh1jw9hjK:ubjx9)}(hhh]h)}(hswaph]hswap}(hj:hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj|:ubah}(h]h ]h"]h$]h&]uh1jw9hjK:ubjx9)}(hhh]h)}(hclearh]hclear}(hj:hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj:ubah}(h]h ]h"]h$]h&]uh1jw9hjK:ubeh}(h]h ]h"]h$]h&]uh1jr9hj9ubeh}(h]h ]h"]h$]h&]uh1j9hj@9ubeh}(h]h ]h"]h$]h&]colsKuh1j>9hj;9ubah}(h]h ]h"]h$]h&]uh1j99hj'9ubeh}(h]h ]h"]h$]h&]uh1jShhhM hj9ubeh}(h]h ]h"]h$]h&]uh1jShhhM hj8hhubh)}(h4Here is an example of using the snapshot feature. ::h]h1Here is an example of using the snapshot feature.}(hj:hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj8hhubj)}(hX=# echo 1 > events/sched/enable # echo 1 > snapshot # cat snapshot # tracer: nop # # entries-in-buffer/entries-written: 71/71 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | -0 [005] d... 2440.603828: sched_switch: prev_comm=swapper/5 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=snapshot-test-2 next_pid=2242 next_prio=120 sleep-2242 [005] d... 2440.603846: sched_switch: prev_comm=snapshot-test-2 prev_pid=2242 prev_prio=120 prev_state=R ==> next_comm=kworker/5:1 next_pid=60 next_prio=120 [...] -0 [002] d... 2440.707230: sched_switch: prev_comm=swapper/2 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=snapshot-test-2 next_pid=2229 next_prio=120 # cat trace # tracer: nop # # entries-in-buffer/entries-written: 77/77 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | -0 [007] d... 2440.707395: sched_switch: prev_comm=swapper/7 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=snapshot-test-2 next_pid=2243 next_prio=120 snapshot-test-2-2229 [002] d... 2440.707438: sched_switch: prev_comm=snapshot-test-2 prev_pid=2229 prev_prio=120 prev_state=S ==> next_comm=swapper/2 next_pid=0 next_prio=120 [...]h]hX=# echo 1 > events/sched/enable # echo 1 > snapshot # cat snapshot # tracer: nop # # entries-in-buffer/entries-written: 71/71 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | -0 [005] d... 2440.603828: sched_switch: prev_comm=swapper/5 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=snapshot-test-2 next_pid=2242 next_prio=120 sleep-2242 [005] d... 2440.603846: sched_switch: prev_comm=snapshot-test-2 prev_pid=2242 prev_prio=120 prev_state=R ==> next_comm=kworker/5:1 next_pid=60 next_prio=120 [...] -0 [002] d... 2440.707230: sched_switch: prev_comm=swapper/2 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=snapshot-test-2 next_pid=2229 next_prio=120 # cat trace # tracer: nop # # entries-in-buffer/entries-written: 77/77 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | -0 [007] d... 2440.707395: sched_switch: prev_comm=swapper/7 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=snapshot-test-2 next_pid=2243 next_prio=120 snapshot-test-2-2229 [002] d... 2440.707438: sched_switch: prev_comm=snapshot-test-2 prev_pid=2229 prev_prio=120 prev_state=S ==> next_comm=swapper/2 next_pid=0 next_prio=120 [...]}hj:sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj8hhubh)}(hIf you try to use this snapshot feature when current tracer is one of the latency tracers, you will get the following results. ::h]h~If you try to use this snapshot feature when current tracer is one of the latency tracers, you will get the following results.}(hj:hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj8hhubj)}(h# echo wakeup > current_tracer # echo 1 > snapshot bash: echo: write error: Device or resource busy # cat snapshot cat: snapshot: Device or resource busyh]h# echo wakeup > current_tracer # echo 1 > snapshot bash: echo: write error: Device or resource busy # cat snapshot cat: snapshot: Device or resource busy}hj:sbah}(h]h ]h"]h$]h&]jjuh1jhhhM hj8hhubeh}(h]snapshotah ]h"]snapshotah$]h&]uh1hhhhhhhhM ubh)}(hhh](h)}(h Instancesh]h Instances}(hj;hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj;hhhhhMubh)}(hX5In the tracefs tracing directory, there is a directory called "instances". This directory can have new directories created inside of it using mkdir, and removing directories with rmdir. The directory created with mkdir in this directory will already contain files and other directories after it is created. ::h]hX6In the tracefs tracing directory, there is a directory called “instances”. This directory can have new directories created inside of it using mkdir, and removing directories with rmdir. The directory created with mkdir in this directory will already contain files and other directories after it is created.}(hj ;hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj;hhubj)}(h# mkdir instances/foo # ls instances/foo buffer_size_kb buffer_total_size_kb events free_buffer per_cpu set_event snapshot trace trace_clock trace_marker trace_options trace_pipe tracing_onh]h# mkdir instances/foo # ls instances/foo buffer_size_kb buffer_total_size_kb events free_buffer per_cpu set_event snapshot trace trace_clock trace_marker trace_options trace_pipe tracing_on}hj.;sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj;hhubh)}(hAs you can see, the new directory looks similar to the tracing directory itself. In fact, it is very similar, except that the buffer and events are agnostic from the main directory, or from any other instances that are created.h]hAs you can see, the new directory looks similar to the tracing directory itself. In fact, it is very similar, except that the buffer and events are agnostic from the main directory, or from any other instances that are created.}(hj<;hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hj;hhubh)}(hXThe files in the new directory work just like the files with the same name in the tracing directory except the buffer that is used is a separate and new buffer. The files affect that buffer but do not affect the main buffer with the exception of trace_options. Currently, the trace_options affect all instances and the top level buffer the same, but this may change in future releases. That is, options may become specific to the instance they reside in.h]hXThe files in the new directory work just like the files with the same name in the tracing directory except the buffer that is used is a separate and new buffer. The files affect that buffer but do not affect the main buffer with the exception of trace_options. Currently, the trace_options affect all instances and the top level buffer the same, but this may change in future releases. That is, options may become specific to the instance they reside in.}(hjJ;hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM%hj;hhubh)}(hNotice that none of the function tracer files are there, nor is current_tracer and available_tracers. This is because the buffers can currently only have events enabled for them. ::h]hNotice that none of the function tracer files are there, nor is current_tracer and available_tracers. This is because the buffers can currently only have events enabled for them.}(hjX;hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM-hj;hhubj)}(hX# mkdir instances/foo # mkdir instances/bar # mkdir instances/zoot # echo 100000 > buffer_size_kb # echo 1000 > instances/foo/buffer_size_kb # echo 5000 > instances/bar/per_cpu/cpu1/buffer_size_kb # echo function > current_trace # echo 1 > instances/foo/events/sched/sched_wakeup/enable # echo 1 > instances/foo/events/sched/sched_wakeup_new/enable # echo 1 > instances/foo/events/sched/sched_switch/enable # echo 1 > instances/bar/events/irq/enable # echo 1 > instances/zoot/events/syscalls/enable # cat trace_pipe CPU:2 [LOST 11745 EVENTS] bash-2044 [002] .... 10594.481032: _raw_spin_lock_irqsave <-get_page_from_freelist bash-2044 [002] d... 10594.481032: add_preempt_count <-_raw_spin_lock_irqsave bash-2044 [002] d..1 10594.481032: __rmqueue <-get_page_from_freelist bash-2044 [002] d..1 10594.481033: _raw_spin_unlock <-get_page_from_freelist bash-2044 [002] d..1 10594.481033: sub_preempt_count <-_raw_spin_unlock bash-2044 [002] d... 10594.481033: get_pageblock_flags_group <-get_pageblock_migratetype bash-2044 [002] d... 10594.481034: __mod_zone_page_state <-get_page_from_freelist bash-2044 [002] d... 10594.481034: zone_statistics <-get_page_from_freelist bash-2044 [002] d... 10594.481034: __inc_zone_state <-zone_statistics bash-2044 [002] d... 10594.481034: __inc_zone_state <-zone_statistics bash-2044 [002] .... 10594.481035: arch_dup_task_struct <-copy_process [...] # cat instances/foo/trace_pipe bash-1998 [000] d..4 136.676759: sched_wakeup: comm=kworker/0:1 pid=59 prio=120 success=1 target_cpu=000 bash-1998 [000] dN.4 136.676760: sched_wakeup: comm=bash pid=1998 prio=120 success=1 target_cpu=000 -0 [003] d.h3 136.676906: sched_wakeup: comm=rcu_preempt pid=9 prio=120 success=1 target_cpu=003 -0 [003] d..3 136.676909: sched_switch: prev_comm=swapper/3 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=rcu_preempt next_pid=9 next_prio=120 rcu_preempt-9 [003] d..3 136.676916: sched_switch: prev_comm=rcu_preempt prev_pid=9 prev_prio=120 prev_state=S ==> next_comm=swapper/3 next_pid=0 next_prio=120 bash-1998 [000] d..4 136.677014: sched_wakeup: comm=kworker/0:1 pid=59 prio=120 success=1 target_cpu=000 bash-1998 [000] dN.4 136.677016: sched_wakeup: comm=bash pid=1998 prio=120 success=1 target_cpu=000 bash-1998 [000] d..3 136.677018: sched_switch: prev_comm=bash prev_pid=1998 prev_prio=120 prev_state=R+ ==> next_comm=kworker/0:1 next_pid=59 next_prio=120 kworker/0:1-59 [000] d..4 136.677022: sched_wakeup: comm=sshd pid=1995 prio=120 success=1 target_cpu=001 kworker/0:1-59 [000] d..3 136.677025: sched_switch: prev_comm=kworker/0:1 prev_pid=59 prev_prio=120 prev_state=S ==> next_comm=bash next_pid=1998 next_prio=120 [...] # cat instances/bar/trace_pipe migration/1-14 [001] d.h3 138.732674: softirq_raise: vec=3 [action=NET_RX] -0 [001] dNh3 138.732725: softirq_raise: vec=3 [action=NET_RX] bash-1998 [000] d.h1 138.733101: softirq_raise: vec=1 [action=TIMER] bash-1998 [000] d.h1 138.733102: softirq_raise: vec=9 [action=RCU] bash-1998 [000] ..s2 138.733105: softirq_entry: vec=1 [action=TIMER] bash-1998 [000] ..s2 138.733106: softirq_exit: vec=1 [action=TIMER] bash-1998 [000] ..s2 138.733106: softirq_entry: vec=9 [action=RCU] bash-1998 [000] ..s2 138.733109: softirq_exit: vec=9 [action=RCU] sshd-1995 [001] d.h1 138.733278: irq_handler_entry: irq=21 name=uhci_hcd:usb4 sshd-1995 [001] d.h1 138.733280: irq_handler_exit: irq=21 ret=unhandled sshd-1995 [001] d.h1 138.733281: irq_handler_entry: irq=21 name=eth0 sshd-1995 [001] d.h1 138.733283: irq_handler_exit: irq=21 ret=handled [...] # cat instances/zoot/trace # tracer: nop # # entries-in-buffer/entries-written: 18996/18996 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | bash-1998 [000] d... 140.733501: sys_write -> 0x2 bash-1998 [000] d... 140.733504: sys_dup2(oldfd: a, newfd: 1) bash-1998 [000] d... 140.733506: sys_dup2 -> 0x1 bash-1998 [000] d... 140.733508: sys_fcntl(fd: a, cmd: 1, arg: 0) bash-1998 [000] d... 140.733509: sys_fcntl -> 0x1 bash-1998 [000] d... 140.733510: sys_close(fd: a) bash-1998 [000] d... 140.733510: sys_close -> 0x0 bash-1998 [000] d... 140.733514: sys_rt_sigprocmask(how: 0, nset: 0, oset: 6e2768, sigsetsize: 8) bash-1998 [000] d... 140.733515: sys_rt_sigprocmask -> 0x0 bash-1998 [000] d... 140.733516: sys_rt_sigaction(sig: 2, act: 7fff718846f0, oact: 7fff71884650, sigsetsize: 8) bash-1998 [000] d... 140.733516: sys_rt_sigaction -> 0x0h]hX# mkdir instances/foo # mkdir instances/bar # mkdir instances/zoot # echo 100000 > buffer_size_kb # echo 1000 > instances/foo/buffer_size_kb # echo 5000 > instances/bar/per_cpu/cpu1/buffer_size_kb # echo function > current_trace # echo 1 > instances/foo/events/sched/sched_wakeup/enable # echo 1 > instances/foo/events/sched/sched_wakeup_new/enable # echo 1 > instances/foo/events/sched/sched_switch/enable # echo 1 > instances/bar/events/irq/enable # echo 1 > instances/zoot/events/syscalls/enable # cat trace_pipe CPU:2 [LOST 11745 EVENTS] bash-2044 [002] .... 10594.481032: _raw_spin_lock_irqsave <-get_page_from_freelist bash-2044 [002] d... 10594.481032: add_preempt_count <-_raw_spin_lock_irqsave bash-2044 [002] d..1 10594.481032: __rmqueue <-get_page_from_freelist bash-2044 [002] d..1 10594.481033: _raw_spin_unlock <-get_page_from_freelist bash-2044 [002] d..1 10594.481033: sub_preempt_count <-_raw_spin_unlock bash-2044 [002] d... 10594.481033: get_pageblock_flags_group <-get_pageblock_migratetype bash-2044 [002] d... 10594.481034: __mod_zone_page_state <-get_page_from_freelist bash-2044 [002] d... 10594.481034: zone_statistics <-get_page_from_freelist bash-2044 [002] d... 10594.481034: __inc_zone_state <-zone_statistics bash-2044 [002] d... 10594.481034: __inc_zone_state <-zone_statistics bash-2044 [002] .... 10594.481035: arch_dup_task_struct <-copy_process [...] # cat instances/foo/trace_pipe bash-1998 [000] d..4 136.676759: sched_wakeup: comm=kworker/0:1 pid=59 prio=120 success=1 target_cpu=000 bash-1998 [000] dN.4 136.676760: sched_wakeup: comm=bash pid=1998 prio=120 success=1 target_cpu=000 -0 [003] d.h3 136.676906: sched_wakeup: comm=rcu_preempt pid=9 prio=120 success=1 target_cpu=003 -0 [003] d..3 136.676909: sched_switch: prev_comm=swapper/3 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=rcu_preempt next_pid=9 next_prio=120 rcu_preempt-9 [003] d..3 136.676916: sched_switch: prev_comm=rcu_preempt prev_pid=9 prev_prio=120 prev_state=S ==> next_comm=swapper/3 next_pid=0 next_prio=120 bash-1998 [000] d..4 136.677014: sched_wakeup: comm=kworker/0:1 pid=59 prio=120 success=1 target_cpu=000 bash-1998 [000] dN.4 136.677016: sched_wakeup: comm=bash pid=1998 prio=120 success=1 target_cpu=000 bash-1998 [000] d..3 136.677018: sched_switch: prev_comm=bash prev_pid=1998 prev_prio=120 prev_state=R+ ==> next_comm=kworker/0:1 next_pid=59 next_prio=120 kworker/0:1-59 [000] d..4 136.677022: sched_wakeup: comm=sshd pid=1995 prio=120 success=1 target_cpu=001 kworker/0:1-59 [000] d..3 136.677025: sched_switch: prev_comm=kworker/0:1 prev_pid=59 prev_prio=120 prev_state=S ==> next_comm=bash next_pid=1998 next_prio=120 [...] # cat instances/bar/trace_pipe migration/1-14 [001] d.h3 138.732674: softirq_raise: vec=3 [action=NET_RX] -0 [001] dNh3 138.732725: softirq_raise: vec=3 [action=NET_RX] bash-1998 [000] d.h1 138.733101: softirq_raise: vec=1 [action=TIMER] bash-1998 [000] d.h1 138.733102: softirq_raise: vec=9 [action=RCU] bash-1998 [000] ..s2 138.733105: softirq_entry: vec=1 [action=TIMER] bash-1998 [000] ..s2 138.733106: softirq_exit: vec=1 [action=TIMER] bash-1998 [000] ..s2 138.733106: softirq_entry: vec=9 [action=RCU] bash-1998 [000] ..s2 138.733109: softirq_exit: vec=9 [action=RCU] sshd-1995 [001] d.h1 138.733278: irq_handler_entry: irq=21 name=uhci_hcd:usb4 sshd-1995 [001] d.h1 138.733280: irq_handler_exit: irq=21 ret=unhandled sshd-1995 [001] d.h1 138.733281: irq_handler_entry: irq=21 name=eth0 sshd-1995 [001] d.h1 138.733283: irq_handler_exit: irq=21 ret=handled [...] # cat instances/zoot/trace # tracer: nop # # entries-in-buffer/entries-written: 18996/18996 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | bash-1998 [000] d... 140.733501: sys_write -> 0x2 bash-1998 [000] d... 140.733504: sys_dup2(oldfd: a, newfd: 1) bash-1998 [000] d... 140.733506: sys_dup2 -> 0x1 bash-1998 [000] d... 140.733508: sys_fcntl(fd: a, cmd: 1, arg: 0) bash-1998 [000] d... 140.733509: sys_fcntl -> 0x1 bash-1998 [000] d... 140.733510: sys_close(fd: a) bash-1998 [000] d... 140.733510: sys_close -> 0x0 bash-1998 [000] d... 140.733514: sys_rt_sigprocmask(how: 0, nset: 0, oset: 6e2768, sigsetsize: 8) bash-1998 [000] d... 140.733515: sys_rt_sigprocmask -> 0x0 bash-1998 [000] d... 140.733516: sys_rt_sigaction(sig: 2, act: 7fff718846f0, oact: 7fff71884650, sigsetsize: 8) bash-1998 [000] d... 140.733516: sys_rt_sigaction -> 0x0}hjf;sbah}(h]h ]h"]h$]h&]jjuh1jhhhM2hj;hhubh)}(hYou can see that the trace of the top most trace buffer shows only the function tracing. The foo instance displays wakeups and task switches.h]hYou can see that the trace of the top most trace buffer shows only the function tracing. The foo instance displays wakeups and task switches.}(hjt;hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj;hhubh)}(h /proc/sys/kernel/stack_tracer_enabledh]h0# echo 1 > /proc/sys/kernel/stack_tracer_enabled}hj;sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj;hhubh)}(hYou can also enable it from the kernel command line to trace the stack size of the kernel during boot up, by adding "stacktrace" to the kernel command line parameter.h]hYou can also enable it from the kernel command line to trace the stack size of the kernel during boot up, by adding “stacktrace” to the kernel command line parameter.}(hj;hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj;hhubh)}(h=After running it for a few minutes, the output looks like: ::h]h:After running it for a few minutes, the output looks like:}(hj <hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj;hhubj)}(hX# cat stack_max_size 2928 # cat stack_trace Depth Size Location (18 entries) ----- ---- -------- 0) 2928 224 update_sd_lb_stats+0xbc/0x4ac 1) 2704 160 find_busiest_group+0x31/0x1f1 2) 2544 256 load_balance+0xd9/0x662 3) 2288 80 idle_balance+0xbb/0x130 4) 2208 128 __schedule+0x26e/0x5b9 5) 2080 16 schedule+0x64/0x66 6) 2064 128 schedule_timeout+0x34/0xe0 7) 1936 112 wait_for_common+0x97/0xf1 8) 1824 16 wait_for_completion+0x1d/0x1f 9) 1808 128 flush_work+0xfe/0x119 10) 1680 16 tty_flush_to_ldisc+0x1e/0x20 11) 1664 48 input_available_p+0x1d/0x5c 12) 1616 48 n_tty_poll+0x6d/0x134 13) 1568 64 tty_poll+0x64/0x7f 14) 1504 880 do_select+0x31e/0x511 15) 624 400 core_sys_select+0x177/0x216 16) 224 96 sys_select+0x91/0xb9 17) 128 128 system_call_fastpath+0x16/0x1bh]hX# cat stack_max_size 2928 # cat stack_trace Depth Size Location (18 entries) ----- ---- -------- 0) 2928 224 update_sd_lb_stats+0xbc/0x4ac 1) 2704 160 find_busiest_group+0x31/0x1f1 2) 2544 256 load_balance+0xd9/0x662 3) 2288 80 idle_balance+0xbb/0x130 4) 2208 128 __schedule+0x26e/0x5b9 5) 2080 16 schedule+0x64/0x66 6) 2064 128 schedule_timeout+0x34/0xe0 7) 1936 112 wait_for_common+0x97/0xf1 8) 1824 16 wait_for_completion+0x1d/0x1f 9) 1808 128 flush_work+0xfe/0x119 10) 1680 16 tty_flush_to_ldisc+0x1e/0x20 11) 1664 48 input_available_p+0x1d/0x5c 12) 1616 48 n_tty_poll+0x6d/0x134 13) 1568 64 tty_poll+0x64/0x7f 14) 1504 880 do_select+0x31e/0x511 15) 624 400 core_sys_select+0x177/0x216 16) 224 96 sys_select+0x91/0xb9 17) 128 128 system_call_fastpath+0x16/0x1b}hj<sbah}(h]h ]h"]h$]h&]jjuh1jhhhMhj;hhubh)}(hNote, if -mfentry is being used by gcc, functions get traced before they set up the stack frame. This means that leaf level functions are not tested by the stack tracer when -mfentry is used.h]hNote, if -mfentry is being used by gcc, functions get traced before they set up the stack frame. This means that leaf level functions are not tested by the stack tracer when -mfentry is used.}(hj'<hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj;hhubh)}(h?Currently, -mfentry is used by gcc 4.6.0 and above on x86 only.h]h?Currently, -mfentry is used by gcc 4.6.0 and above on x86 only.}(hj5<hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj;hhubeh}(h] stack-traceah ]h"] stack traceah$]h&]uh1hhhhhhhhMubh)}(hhh](h)}(hMoreh]hMore}(hjN<hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjK<hhhhhMubh)}(hNMore details can be found in the source code, in the `kernel/trace/*.c` files.h](h5More details can be found in the source code, in the }(hj\<hhhNhNubhtitle_reference)}(h`kernel/trace/*.c`h]hkernel/trace/*.c}(hjf<hhhNhNubah}(h]h ]h"]h$]h&]uh1jd<hj\<ubh files.}(hj\<hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjK<hhubeh}(h]moreah ]h"]moreah$]h&]uh1hhhhhhhhMubeh}(h]ftrace-function-tracerah ]h"]ftrace - function tracerah$]h&]uh1hhhhhhhhKubeh}(h]h ]h"]h$]h&]sourcehuh1hcurrent_sourceN current_lineNsettingsdocutils.frontendValues)}(hN generatorN datestampN source_linkN source_urlN toc_backlinksjw9footnote_backlinksK sectnum_xformKstrip_commentsNstrip_elements_with_classesN strip_classesN report_levelK halt_levelKexit_status_levelKdebugNwarning_streamN tracebackinput_encoding utf-8-siginput_encoding_error_handlerstrictoutput_encodingutf-8output_encoding_error_handlerj<error_encodingutf-8error_encoding_error_handlerbackslashreplace language_codeenrecord_dependenciesNconfigN id_prefixhauto_id_prefixid dump_settingsNdump_internalsNdump_transformsNdump_pseudo_xmlNexpose_internalsNstrict_visitorN_disable_configN_sourceh _destinationN _config_files]7/var/lib/git/docbuild/linux/Documentation/docutils.confafile_insertion_enabled raw_enabledKline_length_limitM'pep_referencesN pep_base_urlhttps://peps.python.org/pep_file_url_templatepep-%04drfc_referencesN rfc_base_url&https://datatracker.ietf.org/doc/html/ tab_widthKtrim_footnote_reference_spacesyntax_highlightlong smart_quotessmartquotes_locales]character_level_inline_markupdoctitle_xform docinfo_xformKsectsubtitle_xform image_loadinglinkembed_stylesheetcloak_email_addressessection_self_linkenvNubreporterNindirect_targets]substitution_defs}substitution_names}refnames}refids}nameids}(j<j<jHjEjojljjjGjDjjjjj6j3j&j#j4$j1$j$j$jF%jC%j&j&jf&jc&j5'j2'jj'jg'j)j)j)j)jL*jI*j0j0j3j3j;4j84j4j4j#5j 5j7j7j7j7j8j8j ;j ;j;j;jH<jE<j<j<u nametypes}(j<jHjojjGjjj6j&j4$j$jF%j&jf&j5'jj'j)j)jL*j0j3j;4j4j#5j7j7j8j ;j;jH<j<uh}(j<hjEjjljKjjrjDjjjJjjj3jj#j9j1$j)j$j7$jC%j$j&jI%jc&j &j2'ji&jg'j8'j)jm'j)j)jI*j)j0jO*j3j0j84j3j4j>4j 5j4j7j&5j7j7j8j7j ;j8j;j;jE<j;j<jK<u footnote_refs} citation_refs} autofootnotes]autofootnote_refs]symbol_footnotes]symbol_footnote_refs] footnotes] citations]autofootnote_startKsymbol_footnote_startK id_counter collectionsCounter}Rparse_messages](hsystem_message)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hj=hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj=ubah}(h]h ]h"]h$]h&]levelKtypeINFOlineM9sourcehuh1j=hji&hhhhhM9ubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hj3=hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj0=ubah}(h]h ]h"]h$]h&]levelKtypej-=lineMCsourcehuh1j=hji&hhhhhMCubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hjN=hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjK=ubah}(h]h ]h"]h$]h&]levelKtypej-=lineM sourcehuh1j=hj)hhhhhM ubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hji=hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjf=ubah}(h]h ]h"]h$]h&]levelKtypej-=lineM sourcehuh1j=hj0hhhhhM ubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hj=hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj=ubah}(h]h ]h"]h$]h&]levelKtypej-=lineM sourcehuh1j=hj0hhhhhM ubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hj=hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj=ubah}(h]h ]h"]h$]h&]levelKtypej-=lineM sourcehuh1j=hj7hhhhhM ubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hj=hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj=ubah}(h]h ]h"]h$]h&]levelKtypej-=lineM sourcehuh1j=hj7hhhhhM ubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hj=hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj=ubah}(h]h ]h"]h$]h&]levelKtypej-=lineM sourcehuh1j=hj7hhhhhM ubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hj=hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj=ubah}(h]h ]h"]h$]h&]levelKtypej-=lineM sourcehuh1j=hj7hhhhhM ubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hj >hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj>ubah}(h]h ]h"]h$]h&]levelKtypej-=lineM sourcehuh1j=hj8hhhhhM ubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hj&>hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj#>ubah}(h]h ]h"]h$]h&]levelKtypej-=lineMsourcehuh1j=hj;hhhhhMubj=)}(hhh]h)}(hfPossible title underline, too short for the title. Treating it as ordinary text because it's so short.h]hhPossible title underline, too short for the title. Treating it as ordinary text because it’s so short.}(hjA>hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj>>ubah}(h]h ]h"]h$]h&]levelKtypej-=lineMsourcehuh1j=hj;hhhhhMubetransform_messages] transformerN include_log] decorationNhhub.