Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf

Alexei Starovoitov says:

====================
pull-request: bpf 2020-06-12

The following pull-request contains BPF updates for your *net* tree.

We've added 26 non-merge commits during the last 10 day(s) which contain
a total of 27 files changed, 348 insertions(+), 93 deletions(-).

The main changes are:

1) sock_hash accounting fix, from Andrey.

2) libbpf fix and probe_mem sanitizing, from Andrii.

3) sock_hash fixes, from Jakub.

4) devmap_val fix, from Jesper.

5) load_bytes_relative fix, from YiFei.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

4 files changed, 48 insertions, 19 deletions

diff --git a/net/core/filter.c b/net/core/filter.c
index 209482a4eaa2..73395384afe2 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1755,25 +1755,27 @@ BPF_CALL_5(bpf_skb_load_bytes_relative, const struct sk_buff *, skb,
 	   u32, offset, void *, to, u32, len, u32, start_header)
 {
 	u8 *end = skb_tail_pointer(skb);
-	u8 *net = skb_network_header(skb);
-	u8 *mac = skb_mac_header(skb);
-	u8 *ptr;
+	u8 *start, *ptr;
 
-	if (unlikely(offset > 0xffff || len > (end - mac)))
+	if (unlikely(offset > 0xffff))
 		goto err_clear;
 
 	switch (start_header) {
 	case BPF_HDR_START_MAC:
-		ptr = mac + offset;
+		if (unlikely(!skb_mac_header_was_set(skb)))
+			goto err_clear;
+		start = skb_mac_header(skb);
 		break;
 	case BPF_HDR_START_NET:
-		ptr = net + offset;
+		start = skb_network_header(skb);
 		break;
 	default:
 		goto err_clear;
 	}
 
-	if (likely(ptr >= mac && ptr + len <= end)) {
+	ptr = start + offset;
+
+	if (likely(ptr + len <= end)) {
 		memcpy(to, ptr, len);
 		return 0;
 	}
@@ -4340,8 +4342,6 @@ static int _bpf_setsockopt(struct sock *sk, int level, int optname,
 			}
 			break;
 		case SO_BINDTODEVICE:
-			ret = -ENOPROTOOPT;
-#ifdef CONFIG_NETDEVICES
 			optlen = min_t(long, optlen, IFNAMSIZ - 1);
 			strncpy(devname, optval, optlen);
 			devname[optlen] = 0;
@@ -4360,7 +4360,6 @@ static int _bpf_setsockopt(struct sock *sk, int level, int optname,
 				dev_put(dev);
 			}
 			ret = sock_bindtoindex(sk, ifindex, false);
-#endif
 			break;
 		default:
 			ret = -EINVAL;
diff --git a/net/core/sock_map.c b/net/core/sock_map.c
index 00a26cf2cfe9..4059f94e9bb5 100644
--- a/net/core/sock_map.c
+++ b/net/core/sock_map.c
@@ -424,10 +424,7 @@ static int sock_map_get_next_key(struct bpf_map *map, void *key, void *next)
 	return 0;
 }
 
-static bool sock_map_redirect_allowed(const struct sock *sk)
-{
-	return sk->sk_state != TCP_LISTEN;
-}
+static bool sock_map_redirect_allowed(const struct sock *sk);
 
 static int sock_map_update_common(struct bpf_map *map, u32 idx,
 				  struct sock *sk, u64 flags)
@@ -508,6 +505,11 @@ static bool sk_is_udp(const struct sock *sk)
 	       sk->sk_protocol == IPPROTO_UDP;
 }
 
+static bool sock_map_redirect_allowed(const struct sock *sk)
+{
+	return sk_is_tcp(sk) && sk->sk_state != TCP_LISTEN;
+}
+
 static bool sock_map_sk_is_suitable(const struct sock *sk)
 {
 	return sk_is_tcp(sk) || sk_is_udp(sk);
@@ -989,11 +991,15 @@ static struct bpf_map *sock_hash_alloc(union bpf_attr *attr)
 		err = -EINVAL;
 		goto free_htab;
 	}
+	err = bpf_map_charge_init(&htab->map.memory, cost);
+	if (err)
+		goto free_htab;
 
 	htab->buckets = bpf_map_area_alloc(htab->buckets_num *
 					   sizeof(struct bpf_htab_bucket),
 					   htab->map.numa_node);
 	if (!htab->buckets) {
+		bpf_map_charge_finish(&htab->map.memory);
 		err = -ENOMEM;
 		goto free_htab;
 	}
@@ -1013,6 +1019,7 @@ static void sock_hash_free(struct bpf_map *map)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct bpf_htab_bucket *bucket;
+	struct hlist_head unlink_list;
 	struct bpf_htab_elem *elem;
 	struct hlist_node *node;
 	int i;
@@ -1024,13 +1031,32 @@ static void sock_hash_free(struct bpf_map *map)
 	synchronize_rcu();
 	for (i = 0; i < htab->buckets_num; i++) {
 		bucket = sock_hash_select_bucket(htab, i);
-		hlist_for_each_entry_safe(elem, node, &bucket->head, node) {
-			hlist_del_rcu(&elem->node);
+
+		/* We are racing with sock_hash_delete_from_link to
+		 * enter the spin-lock critical section. Every socket on
+		 * the list is still linked to sockhash. Since link
+		 * exists, psock exists and holds a ref to socket. That
+		 * lets us to grab a socket ref too.
+		 */
+		raw_spin_lock_bh(&bucket->lock);
+		hlist_for_each_entry(elem, &bucket->head, node)
+			sock_hold(elem->sk);
+		hlist_move_list(&bucket->head, &unlink_list);
+		raw_spin_unlock_bh(&bucket->lock);
+
+		/* Process removed entries out of atomic context to
+		 * block for socket lock before deleting the psock's
+		 * link to sockhash.
+		 */
+		hlist_for_each_entry_safe(elem, node, &unlink_list, node) {
+			hlist_del(&elem->node);
 			lock_sock(elem->sk);
 			rcu_read_lock();
 			sock_map_unref(elem->sk, elem);
 			rcu_read_unlock();
 			release_sock(elem->sk);
+			sock_put(elem->sk);
+			sock_hash_free_elem(htab, elem);
 		}
 	}
 
diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c
index 629aaa9a1eb9..7aa68f4aae6c 100644
--- a/net/ipv4/tcp_bpf.c
+++ b/net/ipv4/tcp_bpf.c
@@ -64,6 +64,9 @@ int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock,
 		} while (i != msg_rx->sg.end);
 
 		if (unlikely(peek)) {
+			if (msg_rx == list_last_entry(&psock->ingress_msg,
+						      struct sk_msg, list))
+				break;
 			msg_rx = list_next_entry(msg_rx, list);
 			continue;
 		}
@@ -242,6 +245,9 @@ static int tcp_bpf_wait_data(struct sock *sk, struct sk_psock *psock,
 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
 	int ret = 0;
 
+	if (sk->sk_shutdown & RCV_SHUTDOWN)
+		return 1;
+
 	if (!timeo)
 		return ret;
 
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index b6c0f08bd80d..3700266229f6 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -352,10 +352,8 @@ static int xsk_generic_xmit(struct sock *sk)
 
 		len = desc.len;
 		skb = sock_alloc_send_skb(sk, len, 1, &err);
-		if (unlikely(!skb)) {
-			err = -EAGAIN;
+		if (unlikely(!skb))
 			goto out;
-		}
 
 		skb_put(skb, len);
 		addr = desc.addr;