return 0;
}
+/*
+ * Special lock-class for __icmpv6_socket:
+ */
+static struct lock_class_key icmpv6_socket_sk_dst_lock_key;
+
int __init icmpv6_init(struct net_proto_family *ops)
{
struct sock *sk;
int err, i, j;
- for_each_cpu(i) {
+ for_each_possible_cpu(i) {
err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6,
&per_cpu(__icmpv6_socket, i));
if (err < 0) {
sk = per_cpu(__icmpv6_socket, i)->sk;
sk->sk_allocation = GFP_ATOMIC;
+ /*
+ * Split off their lock-class, because sk->sk_dst_lock
+ * gets used from softirqs, which is safe for
+ * __icmpv6_socket (because those never get directly used
+ * via userspace syscalls), but unsafe for normal sockets.
+ */
+ lockdep_set_class(&sk->sk_dst_lock,
+ &icmpv6_socket_sk_dst_lock_key);
/* Enough space for 2 64K ICMP packets, including
* sk_buff struct overhead.
{
int i;
- for_each_cpu(i) {
+ for_each_possible_cpu(i) {
sock_release(per_cpu(__icmpv6_socket, i));
}
inet6_del_protocol(&icmpv6_protocol, IPPROTO_ICMPV6);