Refactor initialization of socket fields after sock_copy() into its
own function, so that it can be used elsewhere.
Signed-off-by: Daniel Borkmann <dborkman@xxxxxxxxxx>
---
net/core/sock.c | 194 +++++++++++++++++++++++++++++---------------------------
1 file changed, 102 insertions(+), 92 deletions(-)
diff --git a/net/core/sock.c b/net/core/sock.c
index 5393b4b..68a4251 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1231,6 +1231,104 @@ static void sock_copy(struct sock *nsk, const struct sock *osk)
#endif
}
+static void sk_update_clone(const struct sock *sk, struct sock *newsk)
+{
+ if (mem_cgroup_sockets_enabled && sk->sk_cgrp)
+ sock_update_memcg(newsk);
+}
+
+/* After going through a rough flat copy of socket fields through sock_copy(),
+ * make it pretty again, so that we have necessary fields properly initialized.
+ */
+static struct sock *sock_sanitize(struct sock *newsk, const struct sock *sk)
+{
+ struct sk_filter *filter;
+
+ get_net(sock_net(newsk));
+
+ sk_node_init(&newsk->sk_node);
+ sock_lock_init(newsk);
+ bh_lock_sock(newsk);
+
+ newsk->sk_backlog.head = newsk->sk_backlog.tail = NULL;
+ newsk->sk_backlog.len = 0;
+
+ atomic_set(&newsk->sk_rmem_alloc, 0);
+ /*
+ * sk_wmem_alloc set to one (see sk_free() and sock_wfree())
+ */
+ atomic_set(&newsk->sk_wmem_alloc, 1);
+ atomic_set(&newsk->sk_omem_alloc, 0);
+
+ skb_queue_head_init(&newsk->sk_receive_queue);
+ skb_queue_head_init(&newsk->sk_write_queue);
+#ifdef CONFIG_NET_DMA
+ skb_queue_head_init(&newsk->sk_async_wait_queue);
+#endif
+
+ spin_lock_init(&newsk->sk_dst_lock);
+ rwlock_init(&newsk->sk_callback_lock);
+ lockdep_set_class_and_name(&newsk->sk_callback_lock,
+ af_callback_keys + newsk->sk_family,
+ af_family_clock_key_strings[newsk->sk_family]);
+
+ newsk->sk_dst_cache = NULL;
+ newsk->sk_wmem_queued = 0;
+ newsk->sk_forward_alloc = 0;
+ newsk->sk_send_head = NULL;
+ newsk->sk_userlocks = sk->sk_userlocks & ~SOCK_BINDPORT_LOCK;
+
+ sock_reset_flag(newsk, SOCK_DONE);
+ skb_queue_head_init(&newsk->sk_error_queue);
+
+ filter = rcu_dereference_protected(newsk->sk_filter, 1);
+ if (filter != NULL)
+ sk_filter_charge(newsk, filter);
+
+ if (unlikely(xfrm_sk_clone_policy(newsk))) {
+ /* It is still raw copy of parent, so invalidate
+ * destructor and make plain sk_free() */
+ newsk->sk_destruct = NULL;
+ bh_unlock_sock(newsk);
+ sk_free(newsk);
+ return NULL;
+ }
+
+ newsk->sk_err = 0;
+ newsk->sk_priority = 0;
+ /*
+ * Before updating sk_refcnt, we must commit prior changes to memory
+ * (Documentation/RCU/rculist_nulls.txt for details)
+ */
+ smp_wmb();
+ atomic_set(&newsk->sk_refcnt, 2);
+
+ /*
+ * Increment the counter in the same struct proto as the master
+ * sock (sk_refcnt_debug_inc uses newsk->sk_prot->socks, that
+ * is the same as sk->sk_prot->socks, as this field was copied
+ * with memcpy).
+ *
+ * This _changes_ the previous behaviour, where
+ * tcp_create_openreq_child always was incrementing the
+ * equivalent to tcp_prot->socks (inet_sock_nr), so this have
+ * to be taken into account in all callers. -acme
+ */
+ sk_refcnt_debug_inc(newsk);
+ sk_set_socket(newsk, NULL);
+ newsk->sk_wq = NULL;
+
+ sk_update_clone(sk, newsk);
+
+ if (newsk->sk_prot->sockets_allocated)
+ sk_sockets_allocated_inc(newsk);
+
+ if (newsk->sk_flags & SK_FLAGS_TIMESTAMP)
+ net_enable_timestamp();
+
+ return newsk;
+}
+
void sk_prot_clear_portaddr_nulls(struct sock *sk, int size)
{
unsigned long nulls1, nulls2;
@@ -1422,12 +1520,6 @@ void sk_release_kernel(struct sock *sk)
}
EXPORT_SYMBOL(sk_release_kernel);
-static void sk_update_clone(const struct sock *sk, struct sock *newsk)
-{
- if (mem_cgroup_sockets_enabled && sk->sk_cgrp)
- sock_update_memcg(newsk);
-}
-
/**
* sk_clone_lock - clone a socket, and lock its clone
* @sk: the socket to clone
@@ -1440,94 +1532,12 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority)
struct sock *newsk;
newsk = sk_prot_alloc(sk->sk_prot, priority, sk->sk_family);
- if (newsk != NULL) {
- struct sk_filter *filter;
-
- sock_copy(newsk, sk);
-
- /* SANITY */
- get_net(sock_net(newsk));
- sk_node_init(&newsk->sk_node);
- sock_lock_init(newsk);
- bh_lock_sock(newsk);
- newsk->sk_backlog.head = newsk->sk_backlog.tail = NULL;
- newsk->sk_backlog.len = 0;
-
- atomic_set(&newsk->sk_rmem_alloc, 0);
- /*
- * sk_wmem_alloc set to one (see sk_free() and sock_wfree())
- */
- atomic_set(&newsk->sk_wmem_alloc, 1);
- atomic_set(&newsk->sk_omem_alloc, 0);
- skb_queue_head_init(&newsk->sk_receive_queue);
- skb_queue_head_init(&newsk->sk_write_queue);
-#ifdef CONFIG_NET_DMA
- skb_queue_head_init(&newsk->sk_async_wait_queue);
-#endif
-
- spin_lock_init(&newsk->sk_dst_lock);
- rwlock_init(&newsk->sk_callback_lock);
- lockdep_set_class_and_name(&newsk->sk_callback_lock,
- af_callback_keys + newsk->sk_family,
- af_family_clock_key_strings[newsk->sk_family]);
-
- newsk->sk_dst_cache = NULL;
- newsk->sk_wmem_queued = 0;
- newsk->sk_forward_alloc = 0;
- newsk->sk_send_head = NULL;
- newsk->sk_userlocks = sk->sk_userlocks & ~SOCK_BINDPORT_LOCK;
-
- sock_reset_flag(newsk, SOCK_DONE);
- skb_queue_head_init(&newsk->sk_error_queue);
-
- filter = rcu_dereference_protected(newsk->sk_filter, 1);
- if (filter != NULL)
- sk_filter_charge(newsk, filter);
-
- if (unlikely(xfrm_sk_clone_policy(newsk))) {
- /* It is still raw copy of parent, so invalidate
- * destructor and make plain sk_free() */
- newsk->sk_destruct = NULL;
- bh_unlock_sock(newsk);
- sk_free(newsk);
- newsk = NULL;
- goto out;
- }
-
- newsk->sk_err = 0;
- newsk->sk_priority = 0;
- /*
- * Before updating sk_refcnt, we must commit prior changes to memory
- * (Documentation/RCU/rculist_nulls.txt for details)
- */
- smp_wmb();
- atomic_set(&newsk->sk_refcnt, 2);
-
- /*
- * Increment the counter in the same struct proto as the master
- * sock (sk_refcnt_debug_inc uses newsk->sk_prot->socks, that
- * is the same as sk->sk_prot->socks, as this field was copied
- * with memcpy).
- *
- * This _changes_ the previous behaviour, where
- * tcp_create_openreq_child always was incrementing the
- * equivalent to tcp_prot->socks (inet_sock_nr), so this have
- * to be taken into account in all callers. -acme
- */
- sk_refcnt_debug_inc(newsk);
- sk_set_socket(newsk, NULL);
- newsk->sk_wq = NULL;
-
- sk_update_clone(sk, newsk);
+ if (newsk == NULL)
+ return NULL;
- if (newsk->sk_prot->sockets_allocated)
- sk_sockets_allocated_inc(newsk);
+ sock_copy(newsk, sk);
- if (newsk->sk_flags & SK_FLAGS_TIMESTAMP)
- net_enable_timestamp();
- }
-out:
- return newsk;
+ return sock_sanitize(newsk, sk);
}
EXPORT_SYMBOL_GPL(sk_clone_lock);
--
1.8.3.1
--
To unsubscribe from this list: send the line "unsubscribe linux-sctp" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
