1 /*****************************************/
3 /*****************************************/
5 Kernel connector - new netlink based userspace <-> kernel space easy
6 to use communication module.
8 The Connector driver makes it easy to connect various agents using a
9 netlink based network. One must register a callback and an identifier.
10 When the driver receives a special netlink message with the appropriate
11 identifier, the appropriate callback will be called.
13 From the userspace point of view it's quite straightforward:
20 But if kernelspace wants to use the full power of such connections, the
21 driver writer must create special sockets, must know about struct sk_buff
22 handling, etc... The Connector driver allows any kernelspace agents to use
23 netlink based networking for inter-process communication in a significantly
26 int cn_add_callback(struct cb_id *id, char *name, void (*callback) (struct cn_msg *, struct netlink_skb_parms *));
27 void cn_netlink_send_multi(struct cn_msg *msg, u16 len, u32 portid, u32 __group, int gfp_mask);
28 void cn_netlink_send(struct cn_msg *msg, u32 portid, u32 __group, int gfp_mask);
36 idx and val are unique identifiers which must be registered in the
37 connector.h header for in-kernel usage. void (*callback) (void *) is a
38 callback function which will be called when a message with above idx.val
39 is received by the connector core. The argument for that function must
40 be dereferenced to struct cn_msg *.
49 __u32 len; /* Length of the following data */
53 /*****************************************/
55 /*****************************************/
57 int cn_add_callback(struct cb_id *id, char *name, void (*callback) (struct cn_msg *, struct netlink_skb_parms *));
59 Registers new callback with connector core.
61 struct cb_id *id - unique connector's user identifier.
62 It must be registered in connector.h for legal in-kernel users.
63 char *name - connector's callback symbolic name.
64 void (*callback) (struct cn..) - connector's callback.
65 cn_msg and the sender's credentials
68 void cn_del_callback(struct cb_id *id);
70 Unregisters new callback with connector core.
72 struct cb_id *id - unique connector's user identifier.
75 int cn_netlink_send_multi(struct cn_msg *msg, u16 len, u32 portid, u32 __groups, int gfp_mask);
76 int cn_netlink_send(struct cn_msg *msg, u32 portid, u32 __groups, int gfp_mask);
78 Sends message to the specified groups. It can be safely called from
79 softirq context, but may silently fail under strong memory pressure.
80 If there are no listeners for given group -ESRCH can be returned.
82 struct cn_msg * - message header(with attached data).
83 u16 len - for *_multi multiple cn_msg messages can be sent
84 u32 port - destination port.
85 If non-zero the message will be sent to the
86 given port, which should be set to the
88 u32 __group - destination group.
89 If port and __group is zero, then appropriate group will
90 be searched through all registered connector users,
91 and message will be delivered to the group which was
92 created for user with the same ID as in msg.
93 If __group is not zero, then message will be delivered
94 to the specified group.
95 int gfp_mask - GFP mask.
97 Note: When registering new callback user, connector core assigns
98 netlink group to the user which is equal to its id.idx.
100 /*****************************************/
101 Protocol description.
102 /*****************************************/
104 The current framework offers a transport layer with fixed headers. The
105 recommended protocol which uses such a header is as following:
107 msg->seq and msg->ack are used to determine message genealogy. When
108 someone sends a message, they use a locally unique sequence and random
109 acknowledge number. The sequence number may be copied into
110 nlmsghdr->nlmsg_seq too.
112 The sequence number is incremented with each message sent.
114 If you expect a reply to the message, then the sequence number in the
115 received message MUST be the same as in the original message, and the
116 acknowledge number MUST be the same + 1.
118 If we receive a message and its sequence number is not equal to one we
119 are expecting, then it is a new message. If we receive a message and
120 its sequence number is the same as one we are expecting, but its
121 acknowledge is not equal to the sequence number in the original
122 message + 1, then it is a new message.
124 Obviously, the protocol header contains the above id.
126 The connector allows event notification in the following form: kernel
127 driver or userspace process can ask connector to notify it when
128 selected ids will be turned on or off (registered or unregistered its
129 callback). It is done by sending a special command to the connector
130 driver (it also registers itself with id={-1, -1}).
132 As example of this usage can be found in the cn_test.c module which
133 uses the connector to request notification and to send messages.
135 /*****************************************/
137 /*****************************************/
139 Netlink itself is not a reliable protocol. That means that messages can
140 be lost due to memory pressure or process' receiving queue overflowed,
141 so caller is warned that it must be prepared. That is why the struct
142 cn_msg [main connector's message header] contains u32 seq and u32 ack
145 /*****************************************/
147 /*****************************************/
149 2.6.14 has a new netlink socket implementation, which by default does not
150 allow people to send data to netlink groups other than 1.
151 So, if you wish to use a netlink socket (for example using connector)
152 with a different group number, the userspace application must subscribe to
153 that group first. It can be achieved by the following pseudocode:
155 s = socket(PF_NETLINK, SOCK_DGRAM, NETLINK_CONNECTOR);
157 l_local.nl_family = AF_NETLINK;
158 l_local.nl_groups = 12345;
161 if (bind(s, (struct sockaddr *)&l_local, sizeof(struct sockaddr_nl)) == -1) {
168 int on = l_local.nl_groups;
169 setsockopt(s, 270, 1, &on, sizeof(on));
172 Where 270 above is SOL_NETLINK, and 1 is a NETLINK_ADD_MEMBERSHIP socket
173 option. To drop a multicast subscription, one should call the above socket
174 option with the NETLINK_DROP_MEMBERSHIP parameter which is defined as 0.
176 2.6.14 netlink code only allows to select a group which is less or equal to
177 the maximum group number, which is used at netlink_kernel_create() time.
178 In case of connector it is CN_NETLINK_USERS + 0xf, so if you want to use
179 group number 12345, you must increment CN_NETLINK_USERS to that number.
180 Additional 0xf numbers are allocated to be used by non-in-kernel users.
182 Due to this limitation, group 0xffffffff does not work now, so one can
183 not use add/remove connector's group notifications, but as far as I know,
184 only cn_test.c test module used it.
186 Some work in netlink area is still being done, so things can be changed in
187 2.6.15 timeframe, if it will happen, documentation will be updated for that
190 /*****************************************/
192 /*****************************************/
194 Sample code for a connector test module and user space can be found
195 in samples/connector/. To build this code, enable CONFIG_CONNECTOR