#include <net/scm.h>
#define Nprintk(a...)
+#define NLGRPSZ(x) (ALIGN(x, sizeof(unsigned long) * 8) / 8)
struct netlink_sock {
/* struct sock has to be the first member of netlink_sock */
struct sock sk;
u32 pid;
- unsigned int groups;
u32 dst_pid;
u32 dst_group;
+ u32 flags;
+ u32 subscriptions;
+ u32 ngroups;
+ unsigned long *groups;
unsigned long state;
wait_queue_head_t wait;
struct netlink_callback *cb;
spinlock_t cb_lock;
void (*data_ready)(struct sock *sk, int bytes);
struct module *module;
- u32 flags;
};
#define NETLINK_KERNEL_SOCKET 0x1
+#define NETLINK_RECV_PKTINFO 0x2
static inline struct netlink_sock *nlk_sk(struct sock *sk)
{
struct nl_pid_hash hash;
struct hlist_head mc_list;
unsigned int nl_nonroot;
+ unsigned int groups;
struct module *module;
+ int registered;
};
static struct netlink_table *nl_table;
BUG_TRAP(!atomic_read(&sk->sk_rmem_alloc));
BUG_TRAP(!atomic_read(&sk->sk_wmem_alloc));
BUG_TRAP(!nlk_sk(sk)->cb);
+ BUG_TRAP(!nlk_sk(sk)->groups);
}
/* This lock without WQ_FLAG_EXCLUSIVE is good on UP and it is _very_ bad on SMP.
netlink_table_grab();
if (sk_del_node_init(sk))
nl_table[sk->sk_protocol].hash.entries--;
- if (nlk_sk(sk)->groups)
+ if (nlk_sk(sk)->subscriptions)
__sk_del_bind_node(sk);
netlink_table_ungrab();
}
.obj_size = sizeof(struct netlink_sock),
};
-static int netlink_create(struct socket *sock, int protocol)
+static int __netlink_create(struct socket *sock, int protocol)
{
struct sock *sk;
struct netlink_sock *nlk;
- struct module *module;
+
+ sock->ops = &netlink_ops;
+
+ sk = sk_alloc(PF_NETLINK, GFP_KERNEL, &netlink_proto, 1);
+ if (!sk)
+ return -ENOMEM;
+
+ sock_init_data(sock, sk);
+
+ nlk = nlk_sk(sk);
+ spin_lock_init(&nlk->cb_lock);
+ init_waitqueue_head(&nlk->wait);
+
+ sk->sk_destruct = netlink_sock_destruct;
+ sk->sk_protocol = protocol;
+ return 0;
+}
+
+static int netlink_create(struct socket *sock, int protocol)
+{
+ struct module *module = NULL;
+ struct netlink_sock *nlk;
+ unsigned int groups;
+ int err = 0;
sock->state = SS_UNCONNECTED;
return -EPROTONOSUPPORT;
netlink_lock_table();
- if (!nl_table[protocol].hash.entries) {
#ifdef CONFIG_KMOD
- /* We do 'best effort'. If we find a matching module,
- * it is loaded. If not, we don't return an error to
- * allow pure userspace<->userspace communication. -HW
- */
+ if (!nl_table[protocol].registered) {
netlink_unlock_table();
request_module("net-pf-%d-proto-%d", PF_NETLINK, protocol);
netlink_lock_table();
-#endif
}
- module = nl_table[protocol].module;
- if (!try_module_get(module))
- module = NULL;
+#endif
+ if (nl_table[protocol].registered &&
+ try_module_get(nl_table[protocol].module))
+ module = nl_table[protocol].module;
+ groups = nl_table[protocol].groups;
netlink_unlock_table();
- sock->ops = &netlink_ops;
-
- sk = sk_alloc(PF_NETLINK, GFP_KERNEL, &netlink_proto, 1);
- if (!sk) {
- module_put(module);
- return -ENOMEM;
- }
-
- sock_init_data(sock, sk);
-
- nlk = nlk_sk(sk);
+ if ((err = __netlink_create(sock, protocol) < 0))
+ goto out_module;
+ nlk = nlk_sk(sock->sk);
nlk->module = module;
- spin_lock_init(&nlk->cb_lock);
- init_waitqueue_head(&nlk->wait);
- sk->sk_destruct = netlink_sock_destruct;
+out:
+ return err;
- sk->sk_protocol = protocol;
- return 0;
+out_module:
+ module_put(module);
+ goto out;
}
static int netlink_release(struct socket *sock)
skb_queue_purge(&sk->sk_write_queue);
- if (nlk->pid && !nlk->groups) {
+ if (nlk->pid && !nlk->subscriptions) {
struct netlink_notify n = {
.protocol = sk->sk_protocol,
.pid = nlk->pid,
if (nlk->flags & NETLINK_KERNEL_SOCKET) {
netlink_table_grab();
nl_table[sk->sk_protocol].module = NULL;
+ nl_table[sk->sk_protocol].registered = 0;
netlink_table_ungrab();
}
+ kfree(nlk->groups);
+ nlk->groups = NULL;
+
sock_put(sk);
return 0;
}
capable(CAP_NET_ADMIN);
}
+static void
+netlink_update_subscriptions(struct sock *sk, unsigned int subscriptions)
+{
+ struct netlink_sock *nlk = nlk_sk(sk);
+
+ if (nlk->subscriptions && !subscriptions)
+ __sk_del_bind_node(sk);
+ else if (!nlk->subscriptions && subscriptions)
+ sk_add_bind_node(sk, &nl_table[sk->sk_protocol].mc_list);
+ nlk->subscriptions = subscriptions;
+}
+
+static int netlink_alloc_groups(struct sock *sk)
+{
+ struct netlink_sock *nlk = nlk_sk(sk);
+ unsigned int groups;
+ int err = 0;
+
+ netlink_lock_table();
+ groups = nl_table[sk->sk_protocol].groups;
+ if (!nl_table[sk->sk_protocol].registered)
+ err = -ENOENT;
+ netlink_unlock_table();
+
+ if (err)
+ return err;
+
+ nlk->groups = kmalloc(NLGRPSZ(groups), GFP_KERNEL);
+ if (nlk->groups == NULL)
+ return -ENOMEM;
+ memset(nlk->groups, 0, NLGRPSZ(groups));
+ nlk->ngroups = groups;
+ return 0;
+}
+
static int netlink_bind(struct socket *sock, struct sockaddr *addr, int addr_len)
{
struct sock *sk = sock->sk;
return -EINVAL;
/* Only superuser is allowed to listen multicasts */
- if (nladdr->nl_groups && !netlink_capable(sock, NL_NONROOT_RECV))
- return -EPERM;
+ if (nladdr->nl_groups) {
+ if (!netlink_capable(sock, NL_NONROOT_RECV))
+ return -EPERM;
+ if (nlk->groups == NULL) {
+ err = netlink_alloc_groups(sk);
+ if (err)
+ return err;
+ }
+ }
if (nlk->pid) {
if (nladdr->nl_pid != nlk->pid)
return err;
}
- if (!nladdr->nl_groups && !nlk->groups)
+ if (!nladdr->nl_groups && (nlk->groups == NULL || !(u32)nlk->groups[0]))
return 0;
netlink_table_grab();
- if (nlk->groups && !nladdr->nl_groups)
- __sk_del_bind_node(sk);
- else if (!nlk->groups && nladdr->nl_groups)
- sk_add_bind_node(sk, &nl_table[sk->sk_protocol].mc_list);
- nlk->groups = nladdr->nl_groups;
+ netlink_update_subscriptions(sk, nlk->subscriptions +
+ hweight32(nladdr->nl_groups) -
+ hweight32(nlk->groups[0]));
+ nlk->groups[0] = (nlk->groups[0] & ~0xffffffffUL) | nladdr->nl_groups;
netlink_table_ungrab();
return 0;
nladdr->nl_groups = netlink_group_mask(nlk->dst_group);
} else {
nladdr->nl_pid = nlk->pid;
- nladdr->nl_groups = nlk->groups;
+ nladdr->nl_groups = nlk->groups ? nlk->groups[0] : 0;
}
return 0;
}
if (p->exclude_sk == sk)
goto out;
- if (nlk->pid == p->pid || !(nlk->groups & netlink_group_mask(p->group)))
+ if (nlk->pid == p->pid || p->group - 1 >= nlk->ngroups ||
+ !test_bit(p->group - 1, nlk->groups))
goto out;
if (p->failure) {
}
int netlink_broadcast(struct sock *ssk, struct sk_buff *skb, u32 pid,
- u32 group, int allocation)
+ u32 group, unsigned int __nocast allocation)
{
struct netlink_broadcast_data info;
struct hlist_node *node;
if (sk == p->exclude_sk)
goto out;
- if (nlk->pid == p->pid || !(nlk->groups & netlink_group_mask(p->group)))
+ if (nlk->pid == p->pid || p->group - 1 >= nlk->ngroups ||
+ !test_bit(p->group - 1, nlk->groups))
goto out;
sk->sk_err = p->code;
read_unlock(&nl_table_lock);
}
+static int netlink_setsockopt(struct socket *sock, int level, int optname,
+ char __user *optval, int optlen)
+{
+ struct sock *sk = sock->sk;
+ struct netlink_sock *nlk = nlk_sk(sk);
+ int val = 0, err;
+
+ if (level != SOL_NETLINK)
+ return -ENOPROTOOPT;
+
+ if (optlen >= sizeof(int) &&
+ get_user(val, (int __user *)optval))
+ return -EFAULT;
+
+ switch (optname) {
+ case NETLINK_PKTINFO:
+ if (val)
+ nlk->flags |= NETLINK_RECV_PKTINFO;
+ else
+ nlk->flags &= ~NETLINK_RECV_PKTINFO;
+ err = 0;
+ break;
+ case NETLINK_ADD_MEMBERSHIP:
+ case NETLINK_DROP_MEMBERSHIP: {
+ unsigned int subscriptions;
+ int old, new = optname == NETLINK_ADD_MEMBERSHIP ? 1 : 0;
+
+ if (!netlink_capable(sock, NL_NONROOT_RECV))
+ return -EPERM;
+ if (nlk->groups == NULL) {
+ err = netlink_alloc_groups(sk);
+ if (err)
+ return err;
+ }
+ if (!val || val - 1 >= nlk->ngroups)
+ return -EINVAL;
+ netlink_table_grab();
+ old = test_bit(val - 1, nlk->groups);
+ subscriptions = nlk->subscriptions - old + new;
+ if (new)
+ __set_bit(val - 1, nlk->groups);
+ else
+ __clear_bit(val - 1, nlk->groups);
+ netlink_update_subscriptions(sk, subscriptions);
+ netlink_table_ungrab();
+ err = 0;
+ break;
+ }
+ default:
+ err = -ENOPROTOOPT;
+ }
+ return err;
+}
+
+static int netlink_getsockopt(struct socket *sock, int level, int optname,
+ char __user *optval, int __user *optlen)
+{
+ struct sock *sk = sock->sk;
+ struct netlink_sock *nlk = nlk_sk(sk);
+ int len, val, err;
+
+ if (level != SOL_NETLINK)
+ return -ENOPROTOOPT;
+
+ if (get_user(len, optlen))
+ return -EFAULT;
+ if (len < 0)
+ return -EINVAL;
+
+ switch (optname) {
+ case NETLINK_PKTINFO:
+ if (len < sizeof(int))
+ return -EINVAL;
+ len = sizeof(int);
+ val = nlk->flags & NETLINK_RECV_PKTINFO ? 1 : 0;
+ put_user(len, optlen);
+ put_user(val, optval);
+ err = 0;
+ break;
+ default:
+ err = -ENOPROTOOPT;
+ }
+ return err;
+}
+
+static void netlink_cmsg_recv_pktinfo(struct msghdr *msg, struct sk_buff *skb)
+{
+ struct nl_pktinfo info;
+
+ info.group = NETLINK_CB(skb).dst_group;
+ put_cmsg(msg, SOL_NETLINK, NETLINK_PKTINFO, sizeof(info), &info);
+}
+
static inline void netlink_rcv_wake(struct sock *sk)
{
struct netlink_sock *nlk = nlk_sk(sk);
netlink_dump(sk);
scm_recv(sock, msg, siocb->scm, flags);
+ if (nlk->flags & NETLINK_RECV_PKTINFO)
+ netlink_cmsg_recv_pktinfo(msg, skb);
out:
netlink_rcv_wake(sk);
*/
struct sock *
-netlink_kernel_create(int unit, void (*input)(struct sock *sk, int len), struct module *module)
+netlink_kernel_create(int unit, unsigned int groups,
+ void (*input)(struct sock *sk, int len),
+ struct module *module)
{
struct socket *sock;
struct sock *sk;
if (sock_create_lite(PF_NETLINK, SOCK_DGRAM, unit, &sock))
return NULL;
- if (netlink_create(sock, unit) < 0)
+ if (__netlink_create(sock, unit) < 0)
goto out_sock_release;
sk = sock->sk;
nlk->flags |= NETLINK_KERNEL_SOCKET;
netlink_table_grab();
+ nl_table[unit].groups = groups < 32 ? 32 : groups;
nl_table[unit].module = module;
+ nl_table[unit].registered = 1;
netlink_table_ungrab();
return sk;
s,
s->sk_protocol,
nlk->pid,
- nlk->groups,
+ nlk->groups ? (u32)nlk->groups[0] : 0,
atomic_read(&s->sk_rmem_alloc),
atomic_read(&s->sk_wmem_alloc),
nlk->cb,
.ioctl = sock_no_ioctl,
.listen = sock_no_listen,
.shutdown = sock_no_shutdown,
- .setsockopt = sock_no_setsockopt,
- .getsockopt = sock_no_getsockopt,
+ .setsockopt = netlink_setsockopt,
+ .getsockopt = netlink_getsockopt,
.sendmsg = netlink_sendmsg,
.recvmsg = netlink_recvmsg,
.mmap = sock_no_mmap,
projects / linux-2.6 / blobdiff