1BPFTOOL-CGROUP(8) BPFTOOL-CGROUP(8)
2
3
4
6 bpftool-cgroup - tool for inspection and simple manipulation of eBPF
7 progs
8
10 bpftool [OPTIONS] cgroup COMMAND
11
12 OPTIONS := { { -j | --json } [{ -p | --pretty }] | { -d | --debug }
13 | { -l | --legacy } | { -f | --bpffs } }
14
15 COMMANDS := { show | list | tree | attach | detach | help }
16
18 bpftool cgroup { show | list } CGROUP [effective]
19 bpftool cgroup tree [CGROUP_ROOT] [effective]
20 bpftool cgroup attach CGROUP ATTACH_TYPE PROG [ATTACH_FLAGS]
21 bpftool cgroup detach CGROUP ATTACH_TYPE PROG
22 bpftool cgroup help
23
24 PROG := { id PROG_ID | pinned FILE | tag PROG_TAG }
25 ATTACH_TYPE := { ingress | egress | sock_create | sock_ops | device |
26 bind4 | bind6 | post_bind4 | post_bind6 | connect4 | connect6 |
27 getpeername4 | getpeername6 | getsockname4 | getsockname6 | sendmsg4 |
28 sendmsg6 | recvmsg4 | recvmsg6 | sysctl | getsockopt | setsockopt |
29 sock_release }
30 ATTACH_FLAGS := { multi | override }
31
32
34 bpftool cgroup { show | list } CGROUP [effective]
35 List all programs attached to the cgroup CGROUP.
36
37 Output will start with program ID followed by attach type,
38 attach flags and program name.
39
40 If effective is specified retrieve effective programs that
41 will execute for events within a cgroup. This includes inher‐
42 ited along with attached ones.
43
44 bpftool cgroup tree [CGROUP_ROOT] [effective]
45 Iterate over all cgroups in CGROUP_ROOT and list all attached
46 programs. If CGROUP_ROOT is not specified, bpftool uses
47 cgroup v2 mountpoint.
48
49 The output is similar to the output of cgroup show/list com‐
50 mands: it starts with absolute cgroup path, followed by pro‐
51 gram ID, attach type, attach flags and program name.
52
53 If effective is specified retrieve effective programs that
54 will execute for events within a cgroup. This includes inher‐
55 ited along with attached ones.
56
57 bpftool cgroup attach CGROUP ATTACH_TYPE PROG [ATTACH_FLAGS]
58 Attach program PROG to the cgroup CGROUP with attach type AT‐
59 TACH_TYPE and optional ATTACH_FLAGS.
60
61 ATTACH_FLAGS can be one of: override if a sub-cgroup installs
62 some bpf program, the program in this cgroup yields to
63 sub-cgroup program; multi if a sub-cgroup installs some bpf
64 program, that cgroup program gets run in addition to the pro‐
65 gram in this cgroup.
66
67 Only one program is allowed to be attached to a cgroup with
68 no attach flags or the override flag. Attaching another pro‐
69 gram will release old program and attach the new one.
70
71 Multiple programs are allowed to be attached to a cgroup with
72 multi. They are executed in FIFO order (those that were at‐
73 tached first, run first).
74
75 Non-default ATTACH_FLAGS are supported by kernel version 4.14
76 and later.
77
78 ATTACH_TYPE can be on of: ingress ingress path of the inet
79 socket (since 4.10); egress egress path of the inet socket
80 (since 4.10); sock_create opening of an inet socket (since
81 4.10); sock_ops various socket operations (since 4.12); de‐
82 vice device access (since 4.15); bind4 call to bind(2) for an
83 inet4 socket (since 4.17); bind6 call to bind(2) for an inet6
84 socket (since 4.17); post_bind4 return from bind(2) for an
85 inet4 socket (since 4.17); post_bind6 return from bind(2) for
86 an inet6 socket (since 4.17); connect4 call to connect(2) for
87 an inet4 socket (since 4.17); connect6 call to connect(2) for
88 an inet6 socket (since 4.17); sendmsg4 call to sendto(2),
89 sendmsg(2), sendmmsg(2) for an unconnected udp4 socket (since
90 4.18); sendmsg6 call to sendto(2), sendmsg(2), sendmmsg(2)
91 for an unconnected udp6 socket (since 4.18); recvmsg4 call to
92 recvfrom(2), recvmsg(2), recvmmsg(2) for an unconnected udp4
93 socket (since 5.2); recvmsg6 call to recvfrom(2), recvmsg(2),
94 recvmmsg(2) for an unconnected udp6 socket (since 5.2);
95 sysctl sysctl access (since 5.2); getsockopt call to getsock‐
96 opt (since 5.3); setsockopt call to setsockopt (since 5.3);
97 getpeername4 call to getpeername(2) for an inet4 socket
98 (since 5.8); getpeername6 call to getpeername(2) for an inet6
99 socket (since 5.8); getsockname4 call to getsockname(2) for
100 an inet4 socket (since 5.8); getsockname6 call to getsock‐
101 name(2) for an inet6 socket (since 5.8). sock_release clos‐
102 ing an userspace inet socket (since 5.9).
103
104 bpftool cgroup detach CGROUP ATTACH_TYPE PROG
105 Detach PROG from the cgroup CGROUP and attach type AT‐
106 TACH_TYPE.
107
108 bpftool prog help
109 Print short help message.
110
112 -h, --help
113 Print short help message (similar to bpftool help).
114
115 -V, --version
116 Print bpftool's version number (similar to bpftool version),
117 the number of the libbpf version in use, and optional fea‐
118 tures that were included when bpftool was compiled. Optional
119 features include linking against libbfd to provide the disas‐
120 sembler for JIT-ted programs (bpftool prog dump jited) and
121 usage of BPF skeletons (some features like bpftool prog pro‐
122 file or showing pids associated to BPF objects may rely on
123 it).
124
125 -j, --json
126 Generate JSON output. For commands that cannot produce JSON,
127 this option has no effect.
128
129 -p, --pretty
130 Generate human-readable JSON output. Implies -j.
131
132 -d, --debug
133 Print all logs available, even debug-level information. This
134 includes logs from libbpf as well as from the verifier, when
135 attempting to load programs.
136
137 -l, --legacy
138 Use legacy libbpf mode which has more relaxed BPF program re‐
139 quirements. By default, bpftool has more strict requirements
140 about section names, changes pinning logic and doesn't sup‐
141 port some of the older non-BTF map declarations.
142
143 See
144 https://github.com/libbpf/libbpf/wiki/Libbpf:-the-road-to-v1.0
145 for details.
146
147 -f, --bpffs
148 Show file names of pinned programs.
149
151 # mount -t bpf none /sys/fs/bpf/
152 # mkdir /sys/fs/cgroup/test.slice
153 # bpftool prog load ./device_cgroup.o /sys/fs/bpf/prog
154 # bpftool cgroup attach /sys/fs/cgroup/test.slice/ device id 1 allow_multi
155
156
157 # bpftool cgroup list /sys/fs/cgroup/test.slice/
158
159 ID AttachType AttachFlags Name
160 1 device allow_multi bpf_prog1
161
162 # bpftool cgroup detach /sys/fs/cgroup/test.slice/ device id 1
163 # bpftool cgroup list /sys/fs/cgroup/test.slice/
164
165
166 ID AttachType AttachFlags Name
167
169 bpf(2), bpf-helpers(7), bpftool(8), bpftool-btf(8), bpftool-fea‐
170 ture(8), bpftool-gen(8), bpftool-iter(8), bpftool-link(8),
171 bpftool-map(8), bpftool-net(8), bpftool-perf(8), bpftool-prog(8),
172 bpftool-struct_ops(8)
173
174
175
176
177 BPFTOOL-CGROUP(8)