1================ 2bpftool-prog 3================ 4------------------------------------------------------------------------------- 5tool for inspection and simple manipulation of eBPF progs 6------------------------------------------------------------------------------- 7 8:Manual section: 8 9 10SYNOPSIS 11======== 12 13 **bpftool** [*OPTIONS*] **prog** *COMMAND* 14 15 *OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] | { **-f** | **--bpffs** } } 16 17 *COMMANDS* := 18 { **show** | **list** | **dump xlated** | **dump jited** | **pin** | **load** 19 | **loadall** | **help** } 20 21PROG COMMANDS 22============= 23 24| **bpftool** **prog { show | list }** [*PROG*] 25| **bpftool** **prog dump xlated** *PROG* [{**file** *FILE* | **opcodes** | **visual** | **linum**}] 26| **bpftool** **prog dump jited** *PROG* [{**file** *FILE* | **opcodes** | **linum**}] 27| **bpftool** **prog pin** *PROG* *FILE* 28| **bpftool** **prog { load | loadall }** *OBJ* *PATH* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*] [**pinmaps** *MAP_DIR*] 29| **bpftool** **prog attach** *PROG* *ATTACH_TYPE* [*MAP*] 30| **bpftool** **prog detach** *PROG* *ATTACH_TYPE* [*MAP*] 31| **bpftool** **prog tracelog** 32| **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*] 33| **bpftool** **prog help** 34| 35| *MAP* := { **id** *MAP_ID* | **pinned** *FILE* } 36| *PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* | **name** *PROG_NAME* } 37| *TYPE* := { 38| **socket** | **kprobe** | **kretprobe** | **classifier** | **action** | 39| **tracepoint** | **raw_tracepoint** | **xdp** | **perf_event** | **cgroup/skb** | 40| **cgroup/sock** | **cgroup/dev** | **lwt_in** | **lwt_out** | **lwt_xmit** | 41| **lwt_seg6local** | **sockops** | **sk_skb** | **sk_msg** | **lirc_mode2** | 42| **cgroup/bind4** | **cgroup/bind6** | **cgroup/post_bind4** | **cgroup/post_bind6** | 43| **cgroup/connect4** | **cgroup/connect6** | **cgroup/sendmsg4** | **cgroup/sendmsg6** | 44| **cgroup/recvmsg4** | **cgroup/recvmsg6** | **cgroup/sysctl** | 45| **cgroup/getsockopt** | **cgroup/setsockopt** 46| } 47| *ATTACH_TYPE* := { 48| **msg_verdict** | **stream_verdict** | **stream_parser** | **flow_dissector** 49| } 50 51 52DESCRIPTION 53=========== 54 **bpftool prog { show | list }** [*PROG*] 55 Show information about loaded programs. If *PROG* is 56 specified show information only about given programs, 57 otherwise list all programs currently loaded on the system. 58 In case of **tag** or **name**, *PROG* may match several 59 programs which will all be shown. 60 61 Output will start with program ID followed by program type and 62 zero or more named attributes (depending on kernel version). 63 64 Since Linux 5.1 the kernel can collect statistics on BPF 65 programs (such as the total time spent running the program, 66 and the number of times it was run). If available, bpftool 67 shows such statistics. However, the kernel does not collect 68 them by defaults, as it slightly impacts performance on each 69 program run. Activation or deactivation of the feature is 70 performed via the **kernel.bpf_stats_enabled** sysctl knob. 71 72 **bpftool prog dump xlated** *PROG* [{ **file** *FILE* | **opcodes** | **visual** | **linum** }] 73 Dump eBPF instructions of the programs from the kernel. By 74 default, eBPF will be disassembled and printed to standard 75 output in human-readable format. In this case, **opcodes** 76 controls if raw opcodes should be printed as well. 77 78 In case of **tag** or **name**, *PROG* may match several 79 programs which will all be dumped. However, if **file** or 80 **visual** is specified, *PROG* must match a single program. 81 82 If **file** is specified, the binary image will instead be 83 written to *FILE*. 84 85 If **visual** is specified, control flow graph (CFG) will be 86 built instead, and eBPF instructions will be presented with 87 CFG in DOT format, on standard output. 88 89 If the programs have line_info available, the source line will 90 be displayed by default. If **linum** is specified, 91 the filename, line number and line column will also be 92 displayed on top of the source line. 93 94 **bpftool prog dump jited** *PROG* [{ **file** *FILE* | **opcodes** | **linum** }] 95 Dump jited image (host machine code) of the program. 96 97 If *FILE* is specified image will be written to a file, 98 otherwise it will be disassembled and printed to stdout. 99 *PROG* must match a single program when **file** is specified. 100 101 **opcodes** controls if raw opcodes will be printed. 102 103 If the prog has line_info available, the source line will 104 be displayed by default. If **linum** is specified, 105 the filename, line number and line column will also be 106 displayed on top of the source line. 107 108 **bpftool prog pin** *PROG* *FILE* 109 Pin program *PROG* as *FILE*. 110 111 Note: *FILE* must be located in *bpffs* mount. It must not 112 contain a dot character ('.'), which is reserved for future 113 extensions of *bpffs*. 114 115 **bpftool prog { load | loadall }** *OBJ* *PATH* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*] [**pinmaps** *MAP_DIR*] 116 Load bpf program(s) from binary *OBJ* and pin as *PATH*. 117 **bpftool prog load** pins only the first program from the 118 *OBJ* as *PATH*. **bpftool prog loadall** pins all programs 119 from the *OBJ* under *PATH* directory. 120 **type** is optional, if not specified program type will be 121 inferred from section names. 122 By default bpftool will create new maps as declared in the ELF 123 object being loaded. **map** parameter allows for the reuse 124 of existing maps. It can be specified multiple times, each 125 time for a different map. *IDX* refers to index of the map 126 to be replaced in the ELF file counting from 0, while *NAME* 127 allows to replace a map by name. *MAP* specifies the map to 128 use, referring to it by **id** or through a **pinned** file. 129 If **dev** *NAME* is specified program will be loaded onto 130 given networking device (offload). 131 Optional **pinmaps** argument can be provided to pin all 132 maps under *MAP_DIR* directory. 133 134 Note: *PATH* must be located in *bpffs* mount. It must not 135 contain a dot character ('.'), which is reserved for future 136 extensions of *bpffs*. 137 138 **bpftool prog attach** *PROG* *ATTACH_TYPE* [*MAP*] 139 Attach bpf program *PROG* (with type specified by 140 *ATTACH_TYPE*). Most *ATTACH_TYPEs* require a *MAP* 141 parameter, with the exception of *flow_dissector* which is 142 attached to current networking name space. 143 144 **bpftool prog detach** *PROG* *ATTACH_TYPE* [*MAP*] 145 Detach bpf program *PROG* (with type specified by 146 *ATTACH_TYPE*). Most *ATTACH_TYPEs* require a *MAP* 147 parameter, with the exception of *flow_dissector* which is 148 detached from the current networking name space. 149 150 **bpftool prog tracelog** 151 Dump the trace pipe of the system to the console (stdout). 152 Hit <Ctrl+C> to stop printing. BPF programs can write to this 153 trace pipe at runtime with the **bpf_trace_printk()** helper. 154 This should be used only for debugging purposes. For 155 streaming data from BPF programs to user space, one can use 156 perf events (see also **bpftool-map**\ (8)). 157 158 **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*] 159 Run BPF program *PROG* in the kernel testing infrastructure 160 for BPF, meaning that the program works on the data and 161 context provided by the user, and not on actual packets or 162 monitored functions etc. Return value and duration for the 163 test run are printed out to the console. 164 165 Input data is read from the *FILE* passed with **data_in**. 166 If this *FILE* is "**-**", input data is read from standard 167 input. Input context, if any, is read from *FILE* passed with 168 **ctx_in**. Again, "**-**" can be used to read from standard 169 input, but only if standard input is not already in use for 170 input data. If a *FILE* is passed with **data_out**, output 171 data is written to that file. Similarly, output context is 172 written to the *FILE* passed with **ctx_out**. For both 173 output flows, "**-**" can be used to print to the standard 174 output (as plain text, or JSON if relevant option was 175 passed). If output keywords are omitted, output data and 176 context are discarded. Keywords **data_size_out** and 177 **ctx_size_out** are used to pass the size (in bytes) for the 178 output buffers to the kernel, although the default of 32 kB 179 should be more than enough for most cases. 180 181 Keyword **repeat** is used to indicate the number of 182 consecutive runs to perform. Note that output data and 183 context printed to files correspond to the last of those 184 runs. The duration printed out at the end of the runs is an 185 average over all runs performed by the command. 186 187 Not all program types support test run. Among those which do, 188 not all of them can take the **ctx_in**/**ctx_out** 189 arguments. bpftool does not perform checks on program types. 190 191 **bpftool prog help** 192 Print short help message. 193 194OPTIONS 195======= 196 -h, --help 197 Print short generic help message (similar to **bpftool help**). 198 199 -V, --version 200 Print version number (similar to **bpftool version**). 201 202 -j, --json 203 Generate JSON output. For commands that cannot produce JSON, this 204 option has no effect. 205 206 -p, --pretty 207 Generate human-readable JSON output. Implies **-j**. 208 209 -f, --bpffs 210 When showing BPF programs, show file names of pinned 211 programs. 212 213 -m, --mapcompat 214 Allow loading maps with unknown map definitions. 215 216 -n, --nomount 217 Do not automatically attempt to mount any virtual file system 218 (such as tracefs or BPF virtual file system) when necessary. 219 220 -d, --debug 221 Print all logs available, even debug-level information. This 222 includes logs from libbpf as well as from the verifier, when 223 attempting to load programs. 224 225EXAMPLES 226======== 227**# bpftool prog show** 228 229:: 230 231 10: xdp name some_prog tag 005a3d2123620c8b gpl run_time_ns 81632 run_cnt 10 232 loaded_at 2017-09-29T20:11:00+0000 uid 0 233 xlated 528B jited 370B memlock 4096B map_ids 10 234 235**# bpftool --json --pretty prog show** 236 237:: 238 239 [{ 240 "id": 10, 241 "type": "xdp", 242 "tag": "005a3d2123620c8b", 243 "gpl_compatible": true, 244 "run_time_ns": 81632, 245 "run_cnt": 10, 246 "loaded_at": 1506715860, 247 "uid": 0, 248 "bytes_xlated": 528, 249 "jited": true, 250 "bytes_jited": 370, 251 "bytes_memlock": 4096, 252 "map_ids": [10 253 ] 254 } 255 ] 256 257| 258| **# bpftool prog dump xlated id 10 file /tmp/t** 259| **# ls -l /tmp/t** 260 261:: 262 263 -rw------- 1 root root 560 Jul 22 01:42 /tmp/t 264 265**# bpftool prog dump jited tag 005a3d2123620c8b** 266 267:: 268 269 0: push %rbp 270 1: mov %rsp,%rbp 271 2: sub $0x228,%rsp 272 3: sub $0x28,%rbp 273 4: mov %rbx,0x0(%rbp) 274 275| 276| **# mount -t bpf none /sys/fs/bpf/** 277| **# bpftool prog pin id 10 /sys/fs/bpf/prog** 278| **# bpftool prog load ./my_prog.o /sys/fs/bpf/prog2** 279| **# ls -l /sys/fs/bpf/** 280 281:: 282 283 -rw------- 1 root root 0 Jul 22 01:43 prog 284 -rw------- 1 root root 0 Jul 22 01:44 prog2 285 286**# bpftool prog dump jited pinned /sys/fs/bpf/prog opcodes** 287 288:: 289 290 0: push %rbp 291 55 292 1: mov %rsp,%rbp 293 48 89 e5 294 4: sub $0x228,%rsp 295 48 81 ec 28 02 00 00 296 b: sub $0x28,%rbp 297 48 83 ed 28 298 f: mov %rbx,0x0(%rbp) 299 48 89 5d 00 300 301| 302| **# bpftool prog load xdp1_kern.o /sys/fs/bpf/xdp1 type xdp map name rxcnt id 7** 303| **# bpftool prog show pinned /sys/fs/bpf/xdp1** 304 305:: 306 307 9: xdp name xdp_prog1 tag 539ec6ce11b52f98 gpl 308 loaded_at 2018-06-25T16:17:31-0700 uid 0 309 xlated 488B jited 336B memlock 4096B map_ids 7 310 311**# rm /sys/fs/bpf/xdp1** 312 313SEE ALSO 314======== 315 **bpf**\ (2), 316 **bpf-helpers**\ (7), 317 **bpftool**\ (8), 318 **bpftool-map**\ (8), 319 **bpftool-cgroup**\ (8), 320 **bpftool-feature**\ (8), 321 **bpftool-net**\ (8), 322 **bpftool-perf**\ (8), 323 **bpftool-btf**\ (8) 324