blob: 78c9466a9aa8005c4171bdb160c60183042318d6 [file] [log] [blame]
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -07001/*****************************************/
2Kernel Connector.
3/*****************************************/
4
5Kernel connector - new netlink based userspace <-> kernel space easy
6to use communication module.
7
Mike Frysinger41144ca2009-07-17 10:13:58 -07008The Connector driver makes it easy to connect various agents using a
9netlink based network. One must register a callback and an identifier.
10When the driver receives a special netlink message with the appropriate
11identifier, the appropriate callback will be called.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070012
13From the userspace point of view it's quite straightforward:
14
15 socket();
16 bind();
17 send();
18 recv();
19
Mike Frysinger41144ca2009-07-17 10:13:58 -070020But if kernelspace wants to use the full power of such connections, the
21driver writer must create special sockets, must know about struct sk_buff
22handling, etc... The Connector driver allows any kernelspace agents to use
23netlink based networking for inter-process communication in a significantly
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070024easier way:
25
Philipp Reisner70693312009-10-02 02:40:05 +000026int cn_add_callback(struct cb_id *id, char *name, void (*callback) (struct cn_msg *, struct netlink_skb_parms *));
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070027void cn_netlink_send(struct cn_msg *msg, u32 __group, int gfp_mask);
28
29struct cb_id
30{
31 __u32 idx;
32 __u32 val;
33};
34
Mike Frysinger41144ca2009-07-17 10:13:58 -070035idx and val are unique identifiers which must be registered in the
36connector.h header for in-kernel usage. void (*callback) (void *) is a
37callback function which will be called when a message with above idx.val
38is received by the connector core. The argument for that function must
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070039be dereferenced to struct cn_msg *.
40
41struct cn_msg
42{
Mike Frysinger41144ca2009-07-17 10:13:58 -070043 struct cb_id id;
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070044
45 __u32 seq;
46 __u32 ack;
47
48 __u32 len; /* Length of the following data */
49 __u8 data[0];
50};
51
52/*****************************************/
53Connector interfaces.
54/*****************************************/
55
Philipp Reisner70693312009-10-02 02:40:05 +000056int cn_add_callback(struct cb_id *id, char *name, void (*callback) (struct cn_msg *, struct netlink_skb_parms *));
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070057
Mike Frysinger41144ca2009-07-17 10:13:58 -070058 Registers new callback with connector core.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070059
Mike Frysinger41144ca2009-07-17 10:13:58 -070060 struct cb_id *id - unique connector's user identifier.
61 It must be registered in connector.h for legal in-kernel users.
62 char *name - connector's callback symbolic name.
Philipp Reisner70693312009-10-02 02:40:05 +000063 void (*callback) (struct cn..) - connector's callback.
64 cn_msg and the sender's credentials
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070065
Mike Frysinger41144ca2009-07-17 10:13:58 -070066
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070067void cn_del_callback(struct cb_id *id);
68
Mike Frysinger41144ca2009-07-17 10:13:58 -070069 Unregisters new callback with connector core.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070070
Mike Frysinger41144ca2009-07-17 10:13:58 -070071 struct cb_id *id - unique connector's user identifier.
72
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070073
Evgeniy Polyakovb191ba02006-03-20 22:21:40 -080074int cn_netlink_send(struct cn_msg *msg, u32 __groups, int gfp_mask);
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070075
Mike Frysinger41144ca2009-07-17 10:13:58 -070076 Sends message to the specified groups. It can be safely called from
77 softirq context, but may silently fail under strong memory pressure.
78 If there are no listeners for given group -ESRCH can be returned.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070079
Mike Frysinger41144ca2009-07-17 10:13:58 -070080 struct cn_msg * - message header(with attached data).
81 u32 __group - destination group.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070082 If __group is zero, then appropriate group will
83 be searched through all registered connector users,
84 and message will be delivered to the group which was
85 created for user with the same ID as in msg.
86 If __group is not zero, then message will be delivered
87 to the specified group.
Mike Frysinger41144ca2009-07-17 10:13:58 -070088 int gfp_mask - GFP mask.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070089
Mike Frysinger41144ca2009-07-17 10:13:58 -070090 Note: When registering new callback user, connector core assigns
91 netlink group to the user which is equal to it's id.idx.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070092
93/*****************************************/
94Protocol description.
95/*****************************************/
96
Mike Frysinger41144ca2009-07-17 10:13:58 -070097The current framework offers a transport layer with fixed headers. The
98recommended protocol which uses such a header is as following:
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -070099
100msg->seq and msg->ack are used to determine message genealogy. When
Mike Frysinger41144ca2009-07-17 10:13:58 -0700101someone sends a message, they use a locally unique sequence and random
102acknowledge number. The sequence number may be copied into
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -0700103nlmsghdr->nlmsg_seq too.
104
Mike Frysinger41144ca2009-07-17 10:13:58 -0700105The sequence number is incremented with each message sent.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -0700106
Mike Frysinger41144ca2009-07-17 10:13:58 -0700107If you expect a reply to the message, then the sequence number in the
108received message MUST be the same as in the original message, and the
109acknowledge number MUST be the same + 1.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -0700110
Mike Frysinger41144ca2009-07-17 10:13:58 -0700111If we receive a message and its sequence number is not equal to one we
112are expecting, then it is a new message. If we receive a message and
113its sequence number is the same as one we are expecting, but its
114acknowledge is not equal to the acknowledge number in the original
115message + 1, then it is a new message.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -0700116
Mike Frysinger41144ca2009-07-17 10:13:58 -0700117Obviously, the protocol header contains the above id.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -0700118
Mike Frysinger41144ca2009-07-17 10:13:58 -0700119The connector allows event notification in the following form: kernel
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -0700120driver or userspace process can ask connector to notify it when
Mike Frysinger41144ca2009-07-17 10:13:58 -0700121selected ids will be turned on or off (registered or unregistered its
122callback). It is done by sending a special command to the connector
123driver (it also registers itself with id={-1, -1}).
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -0700124
Mike Frysinger41144ca2009-07-17 10:13:58 -0700125As example of this usage can be found in the cn_test.c module which
126uses the connector to request notification and to send messages.
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -0700127
128/*****************************************/
129Reliability.
130/*****************************************/
131
Mike Frysinger41144ca2009-07-17 10:13:58 -0700132Netlink itself is not a reliable protocol. That means that messages can
Evgeniy Polyakov7672d0b2005-09-11 19:15:07 -0700133be lost due to memory pressure or process' receiving queue overflowed,
Mike Frysinger41144ca2009-07-17 10:13:58 -0700134so caller is warned that it must be prepared. That is why the struct
135cn_msg [main connector's message header] contains u32 seq and u32 ack
136fields.
Evgeniy Polyakoveb0d6042005-10-13 14:42:04 -0700137
138/*****************************************/
139Userspace usage.
140/*****************************************/
Mike Frysinger41144ca2009-07-17 10:13:58 -0700141
Evgeniy Polyakoveb0d6042005-10-13 14:42:04 -07001422.6.14 has a new netlink socket implementation, which by default does not
Mike Frysinger41144ca2009-07-17 10:13:58 -0700143allow people to send data to netlink groups other than 1.
144So, if you wish to use a netlink socket (for example using connector)
145with a different group number, the userspace application must subscribe to
146that group first. It can be achieved by the following pseudocode:
Evgeniy Polyakoveb0d6042005-10-13 14:42:04 -0700147
148s = socket(PF_NETLINK, SOCK_DGRAM, NETLINK_CONNECTOR);
149
150l_local.nl_family = AF_NETLINK;
151l_local.nl_groups = 12345;
152l_local.nl_pid = 0;
153
154if (bind(s, (struct sockaddr *)&l_local, sizeof(struct sockaddr_nl)) == -1) {
155 perror("bind");
156 close(s);
157 return -1;
158}
159
160{
161 int on = l_local.nl_groups;
162 setsockopt(s, 270, 1, &on, sizeof(on));
163}
164
165Where 270 above is SOL_NETLINK, and 1 is a NETLINK_ADD_MEMBERSHIP socket
Mike Frysinger41144ca2009-07-17 10:13:58 -0700166option. To drop a multicast subscription, one should call the above socket
167option with the NETLINK_DROP_MEMBERSHIP parameter which is defined as 0.
Evgeniy Polyakoveb0d6042005-10-13 14:42:04 -0700168
1692.6.14 netlink code only allows to select a group which is less or equal to
170the maximum group number, which is used at netlink_kernel_create() time.
171In case of connector it is CN_NETLINK_USERS + 0xf, so if you want to use
172group number 12345, you must increment CN_NETLINK_USERS to that number.
173Additional 0xf numbers are allocated to be used by non-in-kernel users.
174
175Due to this limitation, group 0xffffffff does not work now, so one can
176not use add/remove connector's group notifications, but as far as I know,
177only cn_test.c test module used it.
178
179Some work in netlink area is still being done, so things can be changed in
1802.6.15 timeframe, if it will happen, documentation will be updated for that
181kernel.