6 perf-probe - Define new dynamic tracepoints
11 'perf probe' [options] --add='PROBE' [...]
13 'perf probe' [options] PROBE
15 'perf probe' [options] --del='[GROUP:]EVENT' [...]
17 'perf probe' --list[=[GROUP:]EVENT]
19 'perf probe' [options] --line='LINE'
21 'perf probe' [options] --vars='PROBEPOINT'
23 'perf probe' [options] --funcs
27 This command defines dynamic tracepoint events, by symbol and registers
28 without debuginfo, or by C expressions (C line numbers, C function names,
29 and C local variables) with debuginfo.
36 Specify vmlinux path which has debuginfo (Dwarf binary).
39 --module=MODNAME|PATH::
40 Specify module name in which perf-probe searches probe points
41 or lines. If a path of module file is passed, perf-probe
42 treat it as an offline module (this means you can add a probe on
43 a module which has not been loaded yet).
47 Specify path to kernel source.
51 Be more verbose (show parsed arguments, etc).
56 Be quiet (do not show any messages including errors).
61 Define a probe event (see PROBE SYNTAX for detail).
65 Delete probe events. This accepts glob wildcards('*', '?') and character
66 classes(e.g. [a-z], [!A-Z]).
69 --list[=[GROUP:]EVENT]::
70 List up current probe events. This can also accept filtering patterns of
72 When this is used with --cache, perf shows all cached probes instead of
77 Show source code lines which can be probed. This needs an argument
78 which specifies a range of the source code. (see LINE SYNTAX for detail)
82 Show available local variables at given probe point. The argument
83 syntax is same as PROBE SYNTAX, but NO ARGs.
86 (Only for --vars) Show external defined variables in addition to local
90 (Only for --add) Search only for non-inlined functions. The functions
91 which do not have instances are ignored.
95 Show available functions in given module or kernel. With -x/--exec,
96 can also list functions in a user space executable / shared library.
97 This also can accept a FILTER rule argument.
100 (Only for --vars and --funcs) Set filter. FILTER is a combination of glob
101 pattern, see FILTER PATTERN for detail.
102 Default FILTER is "!__k???tab_* & !__crc_*" for --vars, and "!_*"
104 If several filters are specified, only the last filter is used.
108 Forcibly add events with existing name.
112 Dry run. With this option, --add and --del doesn't execute actual
113 adding and removal operations.
116 (With --add) Cache the probes. Any events which successfully added
117 are also stored in the cache file.
118 (With --list) Show cached probes.
119 (With --del) Remove cached probes.
122 Set the maximum number of probe points for an event. Default is 128.
126 Specify path to the executable or shared library file for user
127 space tracing. Can also be used with --funcs option.
130 Demangle application symbols. --no-demangle is also available
131 for disabling demangling.
134 Demangle kernel symbols. --no-demangle-kernel is also available
135 for disabling kernel demangling.
137 In absence of -m/-x options, perf probe checks if the first argument after
138 the options is an absolute path name. If its an absolute path, perf probe
139 uses it as a target module/target user space binary to probe.
143 Probe points are defined by following syntax.
145 1) Define event based on function name
146 [[GROUP:]EVENT=]FUNC[@SRC][:RLN|+OFFS|%return|;PTN] [ARG ...]
148 2) Define event based on source file with line number
149 [[GROUP:]EVENT=]SRC:ALN [ARG ...]
151 3) Define event based on source file with lazy pattern
152 [[GROUP:]EVENT=]SRC;PTN [ARG ...]
154 4) Pre-defined SDT events or cached event with name
155 %[sdt_PROVIDER:]SDTEVENT
157 sdt_PROVIDER:SDTEVENT
159 'EVENT' specifies the name of new event, if omitted, it will be set the name of the probed function. You can also specify a group name by 'GROUP', if omitted, set 'probe' is used for kprobe and 'probe_<bin>' is used for uprobe.
160 Note that using existing group name can conflict with other events. Especially, using the group name reserved for kernel modules can hide embedded events in the
162 'FUNC' specifies a probed function name, and it may have one of the following options; '+OFFS' is the offset from function entry address in bytes, ':RLN' is the relative-line number from function entry line, and '%return' means that it probes function return. And ';PTN' means lazy matching pattern (see LAZY MATCHING). Note that ';PTN' must be the end of the probe point definition. In addition, '@SRC' specifies a source file which has that function.
163 It is also possible to specify a probe point by the source line number or lazy matching by using 'SRC:ALN' or 'SRC;PTN' syntax, where 'SRC' is the source file path, ':ALN' is the line number and ';PTN' is the lazy matching pattern.
164 'ARG' specifies the arguments of this probe point, (see PROBE ARGUMENT).
165 'SDTEVENT' and 'PROVIDER' is the pre-defined event name which is defined by user SDT (Statically Defined Tracing) or the pre-cached probes with event name.
166 Note that before using the SDT event, the target binary (on which SDT events are defined) must be scanned by linkperf:perf-buildid-cache[1] to make SDT events as cached events.
168 For details of the SDT, see below.
169 https://sourceware.org/gdb/onlinedocs/gdb/Static-Probe-Points.html
173 Each probe argument follows below syntax.
175 [NAME=]LOCALVAR|$retval|%REG|@SYMBOL[:TYPE]
177 'NAME' specifies the name of this argument (optional). You can use the name of local variable, local data structure member (e.g. var->field, var.field2), local array with fixed index (e.g. array[1], var->array[0], var->pointer[2]), or kprobe-tracer argument format (e.g. $retval, %ax, etc). Note that the name of this argument will be set as the last member name if you specify a local data structure member (e.g. field2 for 'var->field1.field2'.)
178 '$vars' and '$params' special arguments are also available for NAME, '$vars' is expanded to the local variables (including function parameters) which can access at given probe point. '$params' is expanded to only the function parameters.
179 'TYPE' casts the type of this argument (optional). If omitted, perf probe automatically set the type based on debuginfo. Currently, basic types (u8/u16/u32/u64/s8/s16/s32/s64), signedness casting (u/s), "string" and bitfield are supported. (see TYPES for detail)
181 On x86 systems %REG is always the short form of the register: for example %AX. %RAX or %EAX is not valid.
185 Basic types (u8/u16/u32/u64/s8/s16/s32/s64) are integer types. Prefix 's' and 'u' means those types are signed and unsigned respectively. Traced arguments are shown in decimal (signed) or hex (unsigned). You can also use 's' or 'u' to specify only signedness and leave its size auto-detected by perf probe.
186 String type is a special type, which fetches a "null-terminated" string from kernel space. This means it will fail and store NULL if the string container has been paged out. You can specify 'string' type only for the local variable or structure member which is an array of or a pointer to 'char' or 'unsigned char' type.
187 Bitfield is another special type, which takes 3 parameters, bit-width, bit-offset, and container-size (usually 32). The syntax is;
189 b<bit-width>@<bit-offset>/<container-size>
193 Line range is described by following syntax.
195 "FUNC[@SRC][:RLN[+NUM|-RLN2]]|SRC[:ALN[+NUM|-ALN2]]"
197 FUNC specifies the function name of showing lines. 'RLN' is the start line
198 number from function entry line, and 'RLN2' is the end line number. As same as
199 probe syntax, 'SRC' means the source file path, 'ALN' is start line number,
200 and 'ALN2' is end line number in the file. It is also possible to specify how
201 many lines to show by using 'NUM'. Moreover, 'FUNC@SRC' combination is good
202 for searching a specific function when several functions share same name.
203 So, "source.c:100-120" shows lines between 100th to l20th in source.c file. And "func:10+20" shows 20 lines from 10th line of func function.
207 The lazy line matching is similar to glob matching but ignoring spaces in both of pattern and target. So this accepts wildcards('*', '?') and character classes(e.g. [a-z], [!A-Z]).
210 'a=*' can matches 'a=b', 'a = b', 'a == b' and so on.
212 This provides some sort of flexibility and robustness to probe point definitions against minor code changes. For example, actual 10th line of schedule() can be moved easily by modifying schedule(), but the same line matching 'rq=cpu_rq*' may still exist in the function.)
216 The filter pattern is a glob matching pattern(s) to filter variables.
217 In addition, you can use "!" for specifying filter-out rule. You also can give several rules combined with "&" or "|", and fold those rules as one rule by using "(" ")".
220 With --filter "foo* | bar*", perf probe -V shows variables which start with "foo" or "bar".
221 With --filter "!foo* & *bar", perf probe -V shows variables which don't start with "foo" and end with "bar", like "fizzbar". But "foobar" is filtered out.
225 Display which lines in schedule() can be probed:
227 ./perf probe --line schedule
229 Add a probe on schedule() function 12th line with recording cpu local variable:
231 ./perf probe schedule:12 cpu
233 ./perf probe --add='schedule:12 cpu'
235 this will add one or more probes which has the name start with "schedule".
237 Add probes on lines in schedule() function which calls update_rq_clock().
239 ./perf probe 'schedule;update_rq_clock*'
241 ./perf probe --add='schedule;update_rq_clock*'
243 Delete all probes on schedule().
245 ./perf probe --del='schedule*'
247 Add probes at zfree() function on /bin/zsh
249 ./perf probe -x /bin/zsh zfree or ./perf probe /bin/zsh zfree
251 Add probes at malloc() function on libc
253 ./perf probe -x /lib/libc.so.6 malloc or ./perf probe /lib/libc.so.6 malloc
257 linkperf:perf-trace[1], linkperf:perf-record[1], linkperf:perf-buildid-cache[1]