1 .. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
6 -------------------------------------------------------------------------------
7 tool for inspection and simple manipulation of eBPF progs
8 -------------------------------------------------------------------------------
12 .. include:: substitutions.rst
17 **bpftool** [*OPTIONS*] **prog** *COMMAND*
19 *OPTIONS* := { |COMMON_OPTIONS| |
20 { **-f** | **--bpffs** } | { **-m** | **--mapcompat** } | { **-n** | **--nomount** } |
21 { **-L** | **--use-loader** } }
24 { **show** | **list** | **dump xlated** | **dump jited** | **pin** | **load** |
25 **loadall** | **help** }
30 | **bpftool** **prog** { **show** | **list** } [*PROG*]
31 | **bpftool** **prog dump xlated** *PROG* [{ **file** *FILE* | [**opcodes**] [**linum**] [**visual**] }]
32 | **bpftool** **prog dump jited** *PROG* [{ **file** *FILE* | [**opcodes**] [**linum**] }]
33 | **bpftool** **prog pin** *PROG* *FILE*
34 | **bpftool** **prog** { **load** | **loadall** } *OBJ* *PATH* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*] [**pinmaps** *MAP_DIR*] [**autoattach**]
35 | **bpftool** **prog attach** *PROG* *ATTACH_TYPE* [*MAP*]
36 | **bpftool** **prog detach** *PROG* *ATTACH_TYPE* [*MAP*]
37 | **bpftool** **prog tracelog**
38 | **bpftool** **prog run** *PROG* **data_in** *FILE* [**data_out** *FILE* [**data_size_out** *L*]] [**ctx_in** *FILE* [**ctx_out** *FILE* [**ctx_size_out** *M*]]] [**repeat** *N*]
39 | **bpftool** **prog profile** *PROG* [**duration** *DURATION*] *METRICs*
40 | **bpftool** **prog help**
42 | *MAP* := { **id** *MAP_ID* | **pinned** *FILE* }
43 | *PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* | **name** *PROG_NAME* }
45 | **socket** | **kprobe** | **kretprobe** | **classifier** | **action** |
46 | **tracepoint** | **raw_tracepoint** | **xdp** | **perf_event** | **cgroup/skb** |
47 | **cgroup/sock** | **cgroup/dev** | **lwt_in** | **lwt_out** | **lwt_xmit** |
48 | **lwt_seg6local** | **sockops** | **sk_skb** | **sk_msg** | **lirc_mode2** |
49 | **cgroup/bind4** | **cgroup/bind6** | **cgroup/post_bind4** | **cgroup/post_bind6** |
50 | **cgroup/connect4** | **cgroup/connect6** | **cgroup/getpeername4** | **cgroup/getpeername6** |
51 | **cgroup/getsockname4** | **cgroup/getsockname6** | **cgroup/sendmsg4** | **cgroup/sendmsg6** |
52 | **cgroup/recvmsg4** | **cgroup/recvmsg6** | **cgroup/sysctl** |
53 | **cgroup/getsockopt** | **cgroup/setsockopt** | **cgroup/sock_release** |
54 | **struct_ops** | **fentry** | **fexit** | **freplace** | **sk_lookup**
57 | **sk_msg_verdict** | **sk_skb_verdict** | **sk_skb_stream_verdict** |
58 | **sk_skb_stream_parser** | **flow_dissector**
61 | **cycles** | **instructions** | **l1d_loads** | **llc_misses** |
62 | **itlb_misses** | **dtlb_misses**
68 **bpftool prog { show | list }** [*PROG*]
69 Show information about loaded programs. If *PROG* is
70 specified show information only about given programs,
71 otherwise list all programs currently loaded on the system.
72 In case of **tag** or **name**, *PROG* may match several
73 programs which will all be shown.
75 Output will start with program ID followed by program type and
76 zero or more named attributes (depending on kernel version).
78 Since Linux 5.1 the kernel can collect statistics on BPF
79 programs (such as the total time spent running the program,
80 and the number of times it was run). If available, bpftool
81 shows such statistics. However, the kernel does not collect
82 them by defaults, as it slightly impacts performance on each
83 program run. Activation or deactivation of the feature is
84 performed via the **kernel.bpf_stats_enabled** sysctl knob.
86 Since Linux 5.8 bpftool is able to discover information about
87 processes that hold open file descriptors (FDs) against BPF
88 programs. On such kernels bpftool will automatically emit this
91 **bpftool prog dump xlated** *PROG* [{ **file** *FILE* | [**opcodes**] [**linum**] [**visual**] }]
92 Dump eBPF instructions of the programs from the kernel. By
93 default, eBPF will be disassembled and printed to standard
94 output in human-readable format. In this case, **opcodes**
95 controls if raw opcodes should be printed as well.
97 In case of **tag** or **name**, *PROG* may match several
98 programs which will all be dumped. However, if **file** or
99 **visual** is specified, *PROG* must match a single program.
101 If **file** is specified, the binary image will instead be
104 If **visual** is specified, control flow graph (CFG) will be
105 built instead, and eBPF instructions will be presented with
106 CFG in DOT format, on standard output.
108 If the programs have line_info available, the source line will
109 be displayed. If **linum** is specified, the filename, line
110 number and line column will also be displayed.
112 **bpftool prog dump jited** *PROG* [{ **file** *FILE* | [**opcodes**] [**linum**] }]
113 Dump jited image (host machine code) of the program.
115 If *FILE* is specified image will be written to a file,
116 otherwise it will be disassembled and printed to stdout.
117 *PROG* must match a single program when **file** is specified.
119 **opcodes** controls if raw opcodes will be printed.
121 If the prog has line_info available, the source line will
122 be displayed. If **linum** is specified, the filename, line
123 number and line column will also be displayed.
125 **bpftool prog pin** *PROG* *FILE*
126 Pin program *PROG* as *FILE*.
128 Note: *FILE* must be located in *bpffs* mount. It must not
129 contain a dot character ('.'), which is reserved for future
130 extensions of *bpffs*.
132 **bpftool prog { load | loadall }** *OBJ* *PATH* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*] [**pinmaps** *MAP_DIR*] [**autoattach**]
133 Load bpf program(s) from binary *OBJ* and pin as *PATH*.
134 **bpftool prog load** pins only the first program from the
135 *OBJ* as *PATH*. **bpftool prog loadall** pins all programs
136 from the *OBJ* under *PATH* directory.
137 **type** is optional, if not specified program type will be
138 inferred from section names.
139 By default bpftool will create new maps as declared in the ELF
140 object being loaded. **map** parameter allows for the reuse
141 of existing maps. It can be specified multiple times, each
142 time for a different map. *IDX* refers to index of the map
143 to be replaced in the ELF file counting from 0, while *NAME*
144 allows to replace a map by name. *MAP* specifies the map to
145 use, referring to it by **id** or through a **pinned** file.
146 If **dev** *NAME* is specified program will be loaded onto
147 given networking device (offload).
148 Optional **pinmaps** argument can be provided to pin all
149 maps under *MAP_DIR* directory.
151 If **autoattach** is specified program will be attached
152 before pin. In that case, only the link (representing the
153 program attached to its hook) is pinned, not the program as
154 such, so the path won't show in **bpftool prog show -f**,
155 only show in **bpftool link show -f**. Also, this only works
156 when bpftool (libbpf) is able to infer all necessary
157 information from the object file, in particular, it's not
158 supported for all program types. If a program does not
159 support autoattach, bpftool falls back to regular pinning
160 for that program instead.
162 Note: *PATH* must be located in *bpffs* mount. It must not
163 contain a dot character ('.'), which is reserved for future
164 extensions of *bpffs*.
166 **bpftool prog attach** *PROG* *ATTACH_TYPE* [*MAP*]
167 Attach bpf program *PROG* (with type specified by
168 *ATTACH_TYPE*). Most *ATTACH_TYPEs* require a *MAP*
169 parameter, with the exception of *flow_dissector* which is
170 attached to current networking name space.
172 **bpftool prog detach** *PROG* *ATTACH_TYPE* [*MAP*]
173 Detach bpf program *PROG* (with type specified by
174 *ATTACH_TYPE*). Most *ATTACH_TYPEs* require a *MAP*
175 parameter, with the exception of *flow_dissector* which is
176 detached from the current networking name space.
178 **bpftool prog tracelog**
179 Dump the trace pipe of the system to the console (stdout).
180 Hit <Ctrl+C> to stop printing. BPF programs can write to this
181 trace pipe at runtime with the **bpf_trace_printk**\ () helper.
182 This should be used only for debugging purposes. For
183 streaming data from BPF programs to user space, one can use
184 perf events (see also **bpftool-map**\ (8)).
186 **bpftool prog run** *PROG* **data_in** *FILE* [**data_out** *FILE* [**data_size_out** *L*]] [**ctx_in** *FILE* [**ctx_out** *FILE* [**ctx_size_out** *M*]]] [**repeat** *N*]
187 Run BPF program *PROG* in the kernel testing infrastructure
188 for BPF, meaning that the program works on the data and
189 context provided by the user, and not on actual packets or
190 monitored functions etc. Return value and duration for the
191 test run are printed out to the console.
193 Input data is read from the *FILE* passed with **data_in**.
194 If this *FILE* is "**-**", input data is read from standard
195 input. Input context, if any, is read from *FILE* passed with
196 **ctx_in**. Again, "**-**" can be used to read from standard
197 input, but only if standard input is not already in use for
198 input data. If a *FILE* is passed with **data_out**, output
199 data is written to that file. Similarly, output context is
200 written to the *FILE* passed with **ctx_out**. For both
201 output flows, "**-**" can be used to print to the standard
202 output (as plain text, or JSON if relevant option was
203 passed). If output keywords are omitted, output data and
204 context are discarded. Keywords **data_size_out** and
205 **ctx_size_out** are used to pass the size (in bytes) for the
206 output buffers to the kernel, although the default of 32 kB
207 should be more than enough for most cases.
209 Keyword **repeat** is used to indicate the number of
210 consecutive runs to perform. Note that output data and
211 context printed to files correspond to the last of those
212 runs. The duration printed out at the end of the runs is an
213 average over all runs performed by the command.
215 Not all program types support test run. Among those which do,
216 not all of them can take the **ctx_in**/**ctx_out**
217 arguments. bpftool does not perform checks on program types.
219 **bpftool prog profile** *PROG* [**duration** *DURATION*] *METRICs*
220 Profile *METRICs* for bpf program *PROG* for *DURATION*
221 seconds or until user hits <Ctrl+C>. *DURATION* is optional.
222 If *DURATION* is not specified, the profiling will run up to
223 **UINT_MAX** seconds.
225 **bpftool prog help**
226 Print short help message.
230 .. include:: common_options.rst
233 When showing BPF programs, show file names of pinned
237 Allow loading maps with unknown map definitions.
240 Do not automatically attempt to mount any virtual file system
241 (such as tracefs or BPF virtual file system) when necessary.
244 Load program as a "loader" program. This is useful to debug
245 the generation of such programs. When this option is in
246 use, bpftool attempts to load the programs from the object
247 file into the kernel, but does not pin them (therefore, the
248 *PATH* must not be provided).
250 When combined with the **-d**\ \|\ **--debug** option,
251 additional debug messages are generated, and the execution
252 of the loader program will use the **bpf_trace_printk**\ ()
253 helper to log each step of loading BTF, creating the maps,
254 and loading the programs (see **bpftool prog tracelog** as
255 a way to dump those messages).
259 **# bpftool prog show**
263 10: xdp name some_prog tag 005a3d2123620c8b gpl run_time_ns 81632 run_cnt 10
264 loaded_at 2017-09-29T20:11:00+0000 uid 0
265 xlated 528B jited 370B memlock 4096B map_ids 10
268 **# bpftool --json --pretty prog show**
275 "tag": "005a3d2123620c8b",
276 "gpl_compatible": true,
277 "run_time_ns": 81632,
279 "loaded_at": 1506715860,
284 "bytes_memlock": 4096,
296 | **# bpftool prog dump xlated id 10 file /tmp/t**
301 -rw------- 1 root root 560 Jul 22 01:42 /tmp/t
303 **# bpftool prog dump jited tag 005a3d2123620c8b**
311 4: mov %rbx,0x0(%rbp)
314 | **# mount -t bpf none /sys/fs/bpf/**
315 | **# bpftool prog pin id 10 /sys/fs/bpf/prog**
316 | **# bpftool prog load ./my_prog.o /sys/fs/bpf/prog2**
317 | **# ls -l /sys/fs/bpf/**
321 -rw------- 1 root root 0 Jul 22 01:43 prog
322 -rw------- 1 root root 0 Jul 22 01:44 prog2
324 **# bpftool prog dump jited pinned /sys/fs/bpf/prog opcodes**
336 f: mov %rbx,0x0(%rbp)
340 | **# bpftool prog load xdp1_kern.o /sys/fs/bpf/xdp1 type xdp map name rxcnt id 7**
341 | **# bpftool prog show pinned /sys/fs/bpf/xdp1**
345 9: xdp name xdp_prog1 tag 539ec6ce11b52f98 gpl
346 loaded_at 2018-06-25T16:17:31-0700 uid 0
347 xlated 488B jited 336B memlock 4096B map_ids 7
349 **# rm /sys/fs/bpf/xdp1**
352 | **# bpftool prog profile id 337 duration 10 cycles instructions llc_misses**
357 40176203 cycles (83.05%)
358 42518139 instructions # 1.06 insns per cycle (83.39%)
359 123 llc_misses # 2.89 LLC misses per million insns (83.15%)
362 | Output below is for the trace logs.
363 | Run in separate terminals:
364 | **# bpftool prog tracelog**
365 | **# bpftool prog load -L -d file.o**
369 bpftool-620059 [004] d... 2634685.517903: bpf_trace_printk: btf_load size 665 r=5
370 bpftool-620059 [004] d... 2634685.517912: bpf_trace_printk: map_create sample_map idx 0 type 2 value_size 4 value_btf_id 0 r=6
371 bpftool-620059 [004] d... 2634685.517997: bpf_trace_printk: prog_load sample insn_cnt 13 r=7
372 bpftool-620059 [004] d... 2634685.517999: bpf_trace_printk: close(5) = 0