1 .. SPDX-License-Identifier: GPL-2.0
7 Kernel connector - new netlink based userspace <-> kernel space easy
8 to use communication module.
10 The Connector driver makes it easy to connect various agents using a
11 netlink based network. One must register a callback and an identifier.
12 When the driver receives a special netlink message with the appropriate
13 identifier, the appropriate callback will be called.
15 From the userspace point of view it's quite straightforward:
22 But if kernelspace wants to use the full power of such connections, the
23 driver writer must create special sockets, must know about struct sk_buff
24 handling, etc... The Connector driver allows any kernelspace agents to use
25 netlink based networking for inter-process communication in a significantly
28 int cn_add_callback(const struct cb_id *id, char *name, void (*callback) (struct cn_msg *, struct netlink_skb_parms *));
29 void cn_netlink_send_mult(struct cn_msg *msg, u16 len, u32 portid, u32 __group, int gfp_mask);
30 void cn_netlink_send(struct cn_msg *msg, u32 portid, u32 __group, int gfp_mask);
38 idx and val are unique identifiers which must be registered in the
39 connector.h header for in-kernel usage. `void (*callback) (void *)` is a
40 callback function which will be called when a message with above idx.val
41 is received by the connector core. The argument for that function must
42 be dereferenced to `struct cn_msg *`::
51 __u16 len; /* Length of the following data */
59 .. kernel-doc:: include/linux/connector.h
62 When registering new callback user, connector core assigns
63 netlink group to the user which is equal to its id.idx.
68 The current framework offers a transport layer with fixed headers. The
69 recommended protocol which uses such a header is as following:
71 msg->seq and msg->ack are used to determine message genealogy. When
72 someone sends a message, they use a locally unique sequence and random
73 acknowledge number. The sequence number may be copied into
74 nlmsghdr->nlmsg_seq too.
76 The sequence number is incremented with each message sent.
78 If you expect a reply to the message, then the sequence number in the
79 received message MUST be the same as in the original message, and the
80 acknowledge number MUST be the same + 1.
82 If we receive a message and its sequence number is not equal to one we
83 are expecting, then it is a new message. If we receive a message and
84 its sequence number is the same as one we are expecting, but its
85 acknowledge is not equal to the sequence number in the original
86 message + 1, then it is a new message.
88 Obviously, the protocol header contains the above id.
90 The connector allows event notification in the following form: kernel
91 driver or userspace process can ask connector to notify it when
92 selected ids will be turned on or off (registered or unregistered its
93 callback). It is done by sending a special command to the connector
94 driver (it also registers itself with id={-1, -1}).
96 As example of this usage can be found in the cn_test.c module which
97 uses the connector to request notification and to send messages.
102 Netlink itself is not a reliable protocol. That means that messages can
103 be lost due to memory pressure or process' receiving queue overflowed,
104 so caller is warned that it must be prepared. That is why the struct
105 cn_msg [main connector's message header] contains u32 seq and u32 ack
111 2.6.14 has a new netlink socket implementation, which by default does not
112 allow people to send data to netlink groups other than 1.
113 So, if you wish to use a netlink socket (for example using connector)
114 with a different group number, the userspace application must subscribe to
115 that group first. It can be achieved by the following pseudocode::
117 s = socket(PF_NETLINK, SOCK_DGRAM, NETLINK_CONNECTOR);
119 l_local.nl_family = AF_NETLINK;
120 l_local.nl_groups = 12345;
123 if (bind(s, (struct sockaddr *)&l_local, sizeof(struct sockaddr_nl)) == -1) {
130 int on = l_local.nl_groups;
131 setsockopt(s, 270, 1, &on, sizeof(on));
134 Where 270 above is SOL_NETLINK, and 1 is a NETLINK_ADD_MEMBERSHIP socket
135 option. To drop a multicast subscription, one should call the above socket
136 option with the NETLINK_DROP_MEMBERSHIP parameter which is defined as 0.
138 2.6.14 netlink code only allows to select a group which is less or equal to
139 the maximum group number, which is used at netlink_kernel_create() time.
140 In case of connector it is CN_NETLINK_USERS + 0xf, so if you want to use
141 group number 12345, you must increment CN_NETLINK_USERS to that number.
142 Additional 0xf numbers are allocated to be used by non-in-kernel users.
144 Due to this limitation, group 0xffffffff does not work now, so one can
145 not use add/remove connector's group notifications, but as far as I know,
146 only cn_test.c test module used it.
148 Some work in netlink area is still being done, so things can be changed in
149 2.6.15 timeframe, if it will happen, documentation will be updated for that
155 Sample code for a connector test module and user space can be found
156 in samples/connector/. To build this code, enable CONFIG_CONNECTOR