1================ 2bpftool-net 3================ 4------------------------------------------------------------------------------- 5tool for inspection of netdev/tc related bpf prog attachments 6------------------------------------------------------------------------------- 7 8:Manual section: 8 9 10SYNOPSIS 11======== 12 13 **bpftool** [*OPTIONS*] **net** *COMMAND* 14 15 *OPTIONS* := { [{ **-j** | **--json** }] [{ **-p** | **--pretty** }] } 16 17 *COMMANDS* := 18 { **show** | **list** | **attach** | **detach** | **help** } 19 20NET COMMANDS 21============ 22 23| **bpftool** **net** { **show** | **list** } [ **dev** *NAME* ] 24| **bpftool** **net attach** *ATTACH_TYPE* *PROG* **dev** *NAME* [ **overwrite** ] 25| **bpftool** **net detach** *ATTACH_TYPE* **dev** *NAME* 26| **bpftool** **net help** 27| 28| *PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* } 29| *ATTACH_TYPE* := { **xdp** | **xdpgeneric** | **xdpdrv** | **xdpoffload** } 30 31DESCRIPTION 32=========== 33 **bpftool net { show | list }** [ **dev** *NAME* ] 34 List bpf program attachments in the kernel networking subsystem. 35 36 Currently, only device driver xdp attachments and tc filter 37 classification/action attachments are implemented, i.e., for 38 program types **BPF_PROG_TYPE_SCHED_CLS**, 39 **BPF_PROG_TYPE_SCHED_ACT** and **BPF_PROG_TYPE_XDP**. 40 For programs attached to a particular cgroup, e.g., 41 **BPF_PROG_TYPE_CGROUP_SKB**, **BPF_PROG_TYPE_CGROUP_SOCK**, 42 **BPF_PROG_TYPE_SOCK_OPS** and **BPF_PROG_TYPE_CGROUP_SOCK_ADDR**, 43 users can use **bpftool cgroup** to dump cgroup attachments. 44 For sk_{filter, skb, msg, reuseport} and lwt/seg6 45 bpf programs, users should consult other tools, e.g., iproute2. 46 47 The current output will start with all xdp program attachments, followed by 48 all tc class/qdisc bpf program attachments. Both xdp programs and 49 tc programs are ordered based on ifindex number. If multiple bpf 50 programs attached to the same networking device through **tc filter**, 51 the order will be first all bpf programs attached to tc classes, then 52 all bpf programs attached to non clsact qdiscs, and finally all 53 bpf programs attached to root and clsact qdisc. 54 55 **bpftool** **net attach** *ATTACH_TYPE* *PROG* **dev** *NAME* [ **overwrite** ] 56 Attach bpf program *PROG* to network interface *NAME* with 57 type specified by *ATTACH_TYPE*. Previously attached bpf program 58 can be replaced by the command used with **overwrite** option. 59 Currently, only XDP-related modes are supported for *ATTACH_TYPE*. 60 61 *ATTACH_TYPE* can be of: 62 **xdp** - try native XDP and fallback to generic XDP if NIC driver does not support it; 63 **xdpgeneric** - Generic XDP. runs at generic XDP hook when packet already enters receive path as skb; 64 **xdpdrv** - Native XDP. runs earliest point in driver's receive path; 65 **xdpoffload** - Offload XDP. runs directly on NIC on each packet reception; 66 67 **bpftool** **net detach** *ATTACH_TYPE* **dev** *NAME* 68 Detach bpf program attached to network interface *NAME* with 69 type specified by *ATTACH_TYPE*. To detach bpf program, same 70 *ATTACH_TYPE* previously used for attach must be specified. 71 Currently, only XDP-related modes are supported for *ATTACH_TYPE*. 72 73 **bpftool net help** 74 Print short help message. 75 76OPTIONS 77======= 78 -h, --help 79 Print short generic help message (similar to **bpftool help**). 80 81 -V, --version 82 Print version number (similar to **bpftool version**). 83 84 -j, --json 85 Generate JSON output. For commands that cannot produce JSON, this 86 option has no effect. 87 88 -p, --pretty 89 Generate human-readable JSON output. Implies **-j**. 90 91 -d, --debug 92 Print all logs available from libbpf, including debug-level 93 information. 94 95EXAMPLES 96======== 97 98| **# bpftool net** 99 100:: 101 102 xdp: 103 eth0(2) driver id 198 104 105 tc: 106 eth0(2) htb name prefix_matcher.o:[cls_prefix_matcher_htb] id 111727 act [] 107 eth0(2) clsact/ingress fbflow_icmp id 130246 act [] 108 eth0(2) clsact/egress prefix_matcher.o:[cls_prefix_matcher_clsact] id 111726 109 eth0(2) clsact/egress cls_fg_dscp id 108619 act [] 110 eth0(2) clsact/egress fbflow_egress id 130245 111 112| 113| **# bpftool -jp net** 114 115:: 116 117 [{ 118 "xdp": [{ 119 "devname": "eth0", 120 "ifindex": 2, 121 "mode": "driver", 122 "id": 198 123 } 124 ], 125 "tc": [{ 126 "devname": "eth0", 127 "ifindex": 2, 128 "kind": "htb", 129 "name": "prefix_matcher.o:[cls_prefix_matcher_htb]", 130 "id": 111727, 131 "act": [] 132 },{ 133 "devname": "eth0", 134 "ifindex": 2, 135 "kind": "clsact/ingress", 136 "name": "fbflow_icmp", 137 "id": 130246, 138 "act": [] 139 },{ 140 "devname": "eth0", 141 "ifindex": 2, 142 "kind": "clsact/egress", 143 "name": "prefix_matcher.o:[cls_prefix_matcher_clsact]", 144 "id": 111726, 145 },{ 146 "devname": "eth0", 147 "ifindex": 2, 148 "kind": "clsact/egress", 149 "name": "cls_fg_dscp", 150 "id": 108619, 151 "act": [] 152 },{ 153 "devname": "eth0", 154 "ifindex": 2, 155 "kind": "clsact/egress", 156 "name": "fbflow_egress", 157 "id": 130245, 158 } 159 ] 160 } 161 ] 162 163| 164| **# bpftool net attach xdpdrv id 16 dev enp6s0np0** 165| **# bpftool net** 166 167:: 168 169 xdp: 170 enp6s0np0(4) driver id 16 171 172| 173| **# bpftool net attach xdpdrv id 16 dev enp6s0np0** 174| **# bpftool net attach xdpdrv id 20 dev enp6s0np0 overwrite** 175| **# bpftool net** 176 177:: 178 179 xdp: 180 enp6s0np0(4) driver id 20 181 182| 183| **# bpftool net attach xdpdrv id 16 dev enp6s0np0** 184| **# bpftool net detach xdpdrv dev enp6s0np0** 185| **# bpftool net** 186 187:: 188 189 xdp: 190 191 192SEE ALSO 193======== 194 **bpf**\ (2), 195 **bpf-helpers**\ (7), 196 **bpftool**\ (8), 197 **bpftool-btf**\ (8), 198 **bpftool-cgroup**\ (8), 199 **bpftool-feature**\ (8), 200 **bpftool-gen**\ (8), 201 **bpftool-iter**\ (8), 202 **bpftool-link**\ (8), 203 **bpftool-map**\ (8), 204 **bpftool-perf**\ (8), 205 **bpftool-prog**\ (8), 206 **bpftool-struct_ops**\ (8) 207