[bpf-next,v4,01/12] bpf: sockmap: only check ULP for TCP sockets

Message ID 20200309111243.6982-2-lmb@cloudflare.com
State New
Headers show
Series
  • bpf: sockmap, sockhash: support storing UDP sockets
Related show

Commit Message

Lorenz Bauer March 9, 2020, 11:12 a.m.
The sock map code checks that a socket does not have an active upper
layer protocol before inserting it into the map. This requires casting
via inet_csk, which isn't valid for UDP sockets.

Guard checks for ULP by checking inet_sk(sk)->is_icsk first.

Signed-off-by: Lorenz Bauer <lmb@cloudflare.com>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
---
 include/linux/skmsg.h              | 8 +++++++-
 include/net/inet_connection_sock.h | 6 ++++++
 net/core/sock_map.c                | 6 ++----
 net/ipv4/tcp_ulp.c                 | 7 -------
 4 files changed, 15 insertions(+), 12 deletions(-)

Patch

diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h
index 112765bd146d..4d3d75d63066 100644
--- a/include/linux/skmsg.h
+++ b/include/linux/skmsg.h
@@ -360,7 +360,13 @@  static inline void sk_psock_restore_proto(struct sock *sk,
 					  struct sk_psock *psock)
 {
 	sk->sk_prot->unhash = psock->saved_unhash;
-	tcp_update_ulp(sk, psock->sk_proto, psock->saved_write_space);
+	if (inet_csk_has_ulp(sk)) {
+		tcp_update_ulp(sk, psock->sk_proto, psock->saved_write_space);
+	} else {
+		sk->sk_write_space = psock->saved_write_space;
+		/* Pairs with lockless read in sk_clone_lock() */
+		WRITE_ONCE(sk->sk_prot, psock->sk_proto);
+	}
 }
 
 static inline void sk_psock_set_state(struct sk_psock *psock,
diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index 895546058a20..a3f076befa4f 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -335,4 +335,10 @@  static inline void inet_csk_inc_pingpong_cnt(struct sock *sk)
 	if (icsk->icsk_ack.pingpong < U8_MAX)
 		icsk->icsk_ack.pingpong++;
 }
+
+static inline bool inet_csk_has_ulp(struct sock *sk)
+{
+	return inet_sk(sk)->is_icsk && !!inet_csk(sk)->icsk_ulp_ops;
+}
+
 #endif /* _INET_CONNECTION_SOCK_H */
diff --git a/net/core/sock_map.c b/net/core/sock_map.c
index 2e0f465295c3..cb8f740f7949 100644
--- a/net/core/sock_map.c
+++ b/net/core/sock_map.c
@@ -384,7 +384,6 @@  static int sock_map_update_common(struct bpf_map *map, u32 idx,
 				  struct sock *sk, u64 flags)
 {
 	struct bpf_stab *stab = container_of(map, struct bpf_stab, map);
-	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct sk_psock_link *link;
 	struct sk_psock *psock;
 	struct sock *osk;
@@ -395,7 +394,7 @@  static int sock_map_update_common(struct bpf_map *map, u32 idx,
 		return -EINVAL;
 	if (unlikely(idx >= map->max_entries))
 		return -E2BIG;
-	if (unlikely(rcu_access_pointer(icsk->icsk_ulp_data)))
+	if (inet_csk_has_ulp(sk))
 		return -EINVAL;
 
 	link = sk_psock_init_link();
@@ -738,7 +737,6 @@  static int sock_hash_update_common(struct bpf_map *map, void *key,
 				   struct sock *sk, u64 flags)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
-	struct inet_connection_sock *icsk = inet_csk(sk);
 	u32 key_size = map->key_size, hash;
 	struct bpf_htab_elem *elem, *elem_new;
 	struct bpf_htab_bucket *bucket;
@@ -749,7 +747,7 @@  static int sock_hash_update_common(struct bpf_map *map, void *key,
 	WARN_ON_ONCE(!rcu_read_lock_held());
 	if (unlikely(flags > BPF_EXIST))
 		return -EINVAL;
-	if (unlikely(icsk->icsk_ulp_data))
+	if (inet_csk_has_ulp(sk))
 		return -EINVAL;
 
 	link = sk_psock_init_link();
diff --git a/net/ipv4/tcp_ulp.c b/net/ipv4/tcp_ulp.c
index 2703f24c5d1a..7c27aa629af1 100644
--- a/net/ipv4/tcp_ulp.c
+++ b/net/ipv4/tcp_ulp.c
@@ -105,13 +105,6 @@  void tcp_update_ulp(struct sock *sk, struct proto *proto,
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 
-	if (!icsk->icsk_ulp_ops) {
-		sk->sk_write_space = write_space;
-		/* Pairs with lockless read in sk_clone_lock() */
-		WRITE_ONCE(sk->sk_prot, proto);
-		return;
-	}
-
 	if (icsk->icsk_ulp_ops->update)
 		icsk->icsk_ulp_ops->update(sk, proto, write_space);
 }