bpf: Hooks for sys_connect

== The problem ==

See description of the problem in the initial patch of this patch set.

== The solution ==

The patch provides much more reliable in-kernel solution for the 2nd
part of the problem: making outgoing connecttion from desired IP.

It adds new attach types `BPF_CGROUP_INET4_CONNECT` and
`BPF_CGROUP_INET6_CONNECT` for program type
`BPF_PROG_TYPE_CGROUP_SOCK_ADDR` that can be used to override both
source and destination of a connection at connect(2) time.

Local end of connection can be bound to desired IP using newly
introduced BPF-helper `bpf_bind()`. It allows to bind to only IP though,
and doesn't support binding to port, i.e. leverages
`IP_BIND_ADDRESS_NO_PORT` socket option. There are two reasons for this:
* looking for a free port is expensive and can affect performance
  significantly;
* there is no use-case for port.

As for remote end (`struct sockaddr *` passed by user), both parts of it
can be overridden, remote IP and remote port. It's useful if an
application inside cgroup wants to connect to another application inside
same cgroup or to itself, but knows nothing about IP assigned to the
cgroup.

Support is added for IPv4 and IPv6, for TCP and UDP.

IPv4 and IPv6 have separate attach types for same reason as sys_bind
hooks, i.e. to prevent reading from / writing to e.g. user_ip6 fields
when user passes sockaddr_in since it'd be out-of-bound.

== Implementation notes ==

The patch introduces new field in `struct proto`: `pre_connect` that is
a pointer to a function with same signature as `connect` but is called
before it. The reason is in some cases BPF hooks should be called way
before control is passed to `sk->sk_prot->connect`. Specifically
`inet_dgram_connect` autobinds socket before calling
`sk->sk_prot->connect` and there is no way to call `bpf_bind()` from
hooks from e.g. `ip4_datagram_connect` or `ip6_datagram_connect` since
it'd cause double-bind. On the other hand `proto.pre_connect` provides a
flexible way to add BPF hooks for connect only for necessary `proto` and
call them at desired time before `connect`. Since `bpf_bind()` is
allowed to bind only to IP and autobind in `inet_dgram_connect` binds
only port there is no chance of double-bind.

bpf_bind() sets `force_bind_address_no_port` to bind to only IP despite
of value of `bind_address_no_port` socket field.

bpf_bind() sets `with_lock` to `false` when calling to __inet_bind()
and __inet6_bind() since all call-sites, where bpf_bind() is called,
already hold socket lock.

Signed-off-by: Andrey Ignatov <[email protected]>
Signed-off-by: Alexei Starovoitov <[email protected]>
Signed-off-by: Daniel Borkmann <[email protected]>

Author: rdna

include/linux/bpf-cgroup.h

@@ -116,12 +116,38 @@ int __cgroup_bpf_check_dev_permission(short dev_type, u32 major, u32 minor,
 	__ret;								       \
 })
 
+#define BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, type)			       \
+({									       \
+	int __ret = 0;							       \
+	if (cgroup_bpf_enabled)	{					       \
+		lock_sock(sk);						       \
+		__ret = __cgroup_bpf_run_filter_sock_addr(sk, uaddr, type);    \
+		release_sock(sk);					       \
+	}								       \
+	__ret;								       \
+})
+
 #define BPF_CGROUP_RUN_PROG_INET4_BIND(sk, uaddr)			       \
 	BPF_CGROUP_RUN_SA_PROG(sk, uaddr, BPF_CGROUP_INET4_BIND)
 
 #define BPF_CGROUP_RUN_PROG_INET6_BIND(sk, uaddr)			       \
 	BPF_CGROUP_RUN_SA_PROG(sk, uaddr, BPF_CGROUP_INET6_BIND)
 
+#define BPF_CGROUP_PRE_CONNECT_ENABLED(sk) (cgroup_bpf_enabled && \
+					    sk->sk_prot->pre_connect)
+
+#define BPF_CGROUP_RUN_PROG_INET4_CONNECT(sk, uaddr)			       \
+	BPF_CGROUP_RUN_SA_PROG(sk, uaddr, BPF_CGROUP_INET4_CONNECT)
+
+#define BPF_CGROUP_RUN_PROG_INET6_CONNECT(sk, uaddr)			       \
+	BPF_CGROUP_RUN_SA_PROG(sk, uaddr, BPF_CGROUP_INET6_CONNECT)
+
+#define BPF_CGROUP_RUN_PROG_INET4_CONNECT_LOCK(sk, uaddr)		       \
+	BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, BPF_CGROUP_INET4_CONNECT)
+
+#define BPF_CGROUP_RUN_PROG_INET6_CONNECT_LOCK(sk, uaddr)		       \
+	BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, BPF_CGROUP_INET6_CONNECT)
+
 #define BPF_CGROUP_RUN_PROG_SOCK_OPS(sock_ops)				       \
 ({									       \
 	int __ret = 0;							       \
@@ -151,11 +177,16 @@ struct cgroup_bpf {};
 static inline void cgroup_bpf_put(struct cgroup *cgrp) {}
 static inline int cgroup_bpf_inherit(struct cgroup *cgrp) { return 0; }
 
+#define BPF_CGROUP_PRE_CONNECT_ENABLED(sk) (0)
 #define BPF_CGROUP_RUN_PROG_INET_INGRESS(sk,skb) ({ 0; })
 #define BPF_CGROUP_RUN_PROG_INET_EGRESS(sk,skb) ({ 0; })
 #define BPF_CGROUP_RUN_PROG_INET_SOCK(sk) ({ 0; })
 #define BPF_CGROUP_RUN_PROG_INET4_BIND(sk, uaddr) ({ 0; })
 #define BPF_CGROUP_RUN_PROG_INET6_BIND(sk, uaddr) ({ 0; })
+#define BPF_CGROUP_RUN_PROG_INET4_CONNECT(sk, uaddr) ({ 0; })
+#define BPF_CGROUP_RUN_PROG_INET4_CONNECT_LOCK(sk, uaddr) ({ 0; })
+#define BPF_CGROUP_RUN_PROG_INET6_CONNECT(sk, uaddr) ({ 0; })
+#define BPF_CGROUP_RUN_PROG_INET6_CONNECT_LOCK(sk, uaddr) ({ 0; })
 #define BPF_CGROUP_RUN_PROG_SOCK_OPS(sock_ops) ({ 0; })
 #define BPF_CGROUP_RUN_PROG_DEVICE_CGROUP(type,major,minor,access) ({ 0; })
 

include/net/addrconf.h

@@ -231,6 +231,13 @@ struct ipv6_stub {
 };
 extern const struct ipv6_stub *ipv6_stub __read_mostly;
 
+/* A stub used by bpf helpers. Similarly ugly as ipv6_stub */
+struct ipv6_bpf_stub {
+	int (*inet6_bind)(struct sock *sk, struct sockaddr *uaddr, int addr_len,
+			  bool force_bind_address_no_port, bool with_lock);
+};
+extern const struct ipv6_bpf_stub *ipv6_bpf_stub __read_mostly;
+
 /*
  * identify MLD packets for MLD filter exceptions
  */

include/net/sock.h

@@ -1026,6 +1026,9 @@ static inline void sk_prot_clear_nulls(struct sock *sk, int size)
 struct proto {
 	void			(*close)(struct sock *sk,
 					long timeout);
+	int			(*pre_connect)(struct sock *sk,
+					struct sockaddr *uaddr,
+					int addr_len);
 	int			(*connect)(struct sock *sk,
 					struct sockaddr *uaddr,
 					int addr_len);

include/net/udp.h

@@ -273,6 +273,7 @@ void udp4_hwcsum(struct sk_buff *skb, __be32 src, __be32 dst);
 int udp_rcv(struct sk_buff *skb);
 int udp_ioctl(struct sock *sk, int cmd, unsigned long arg);
 int udp_init_sock(struct sock *sk);
+int udp_pre_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len);
 int __udp_disconnect(struct sock *sk, int flags);
 int udp_disconnect(struct sock *sk, int flags);
 __poll_t udp_poll(struct file *file, struct socket *sock, poll_table *wait);

include/uapi/linux/bpf.h

@@ -150,6 +150,8 @@ enum bpf_attach_type {
 	BPF_SK_MSG_VERDICT,
 	BPF_CGROUP_INET4_BIND,
 	BPF_CGROUP_INET6_BIND,
+	BPF_CGROUP_INET4_CONNECT,
+	BPF_CGROUP_INET6_CONNECT,
 	__MAX_BPF_ATTACH_TYPE
 };
 
@@ -744,6 +746,13 @@ union bpf_attr {
  *     @flags: reserved for future use
  *     Return: SK_PASS
  *
+ * int bpf_bind(ctx, addr, addr_len)
+ *     Bind socket to address. Only binding to IP is supported, no port can be
+ *     set in addr.
+ *     @ctx: pointer to context of type bpf_sock_addr
+ *     @addr: pointer to struct sockaddr to bind socket to
+ *     @addr_len: length of sockaddr structure
+ *     Return: 0 on success or negative error code
  */
 #define __BPF_FUNC_MAPPER(FN)		\
 	FN(unspec),			\
@@ -809,7 +818,8 @@ union bpf_attr {
 	FN(msg_redirect_map),		\
 	FN(msg_apply_bytes),		\
 	FN(msg_cork_bytes),		\
-	FN(msg_pull_data),
+	FN(msg_pull_data),		\
+	FN(bind),
 
 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
  * function eBPF program intends to call

kernel/bpf/syscall.c

@@ -1180,6 +1180,8 @@ bpf_prog_load_check_attach_type(enum bpf_prog_type prog_type,
 		switch (expected_attach_type) {
 		case BPF_CGROUP_INET4_BIND:
 		case BPF_CGROUP_INET6_BIND:
+		case BPF_CGROUP_INET4_CONNECT:
+		case BPF_CGROUP_INET6_CONNECT:
 			return 0;
 		default:
 			return -EINVAL;
@@ -1491,6 +1493,8 @@ static int bpf_prog_attach(const union bpf_attr *attr)
 		break;
 	case BPF_CGROUP_INET4_BIND:
 	case BPF_CGROUP_INET6_BIND:
+	case BPF_CGROUP_INET4_CONNECT:
+	case BPF_CGROUP_INET6_CONNECT:
 		ptype = BPF_PROG_TYPE_CGROUP_SOCK_ADDR;
 		break;
 	case BPF_CGROUP_SOCK_OPS:
@@ -1557,6 +1561,8 @@ static int bpf_prog_detach(const union bpf_attr *attr)
 		break;
 	case BPF_CGROUP_INET4_BIND:
 	case BPF_CGROUP_INET6_BIND:
+	case BPF_CGROUP_INET4_CONNECT:
+	case BPF_CGROUP_INET6_CONNECT:
 		ptype = BPF_PROG_TYPE_CGROUP_SOCK_ADDR;
 		break;
 	case BPF_CGROUP_SOCK_OPS:
@@ -1610,6 +1616,8 @@ static int bpf_prog_query(const union bpf_attr *attr,
 	case BPF_CGROUP_INET_SOCK_CREATE:
 	case BPF_CGROUP_INET4_BIND:
 	case BPF_CGROUP_INET6_BIND:
+	case BPF_CGROUP_INET4_CONNECT:
+	case BPF_CGROUP_INET6_CONNECT:
 	case BPF_CGROUP_SOCK_OPS:
 	case BPF_CGROUP_DEVICE:
 		break;

net/core/filter.c

@@ -33,6 +33,7 @@
 #include <linux/if_packet.h>
 #include <linux/if_arp.h>
 #include <linux/gfp.h>
+#include <net/inet_common.h>
 #include <net/ip.h>
 #include <net/protocol.h>
 #include <net/netlink.h>
@@ -3656,6 +3657,52 @@ static const struct bpf_func_proto bpf_sock_ops_cb_flags_set_proto = {
 	.arg2_type	= ARG_ANYTHING,
 };
 
+const struct ipv6_bpf_stub *ipv6_bpf_stub __read_mostly;
+EXPORT_SYMBOL_GPL(ipv6_bpf_stub);
+
+BPF_CALL_3(bpf_bind, struct bpf_sock_addr_kern *, ctx, struct sockaddr *, addr,
+	   int, addr_len)
+{
+#ifdef CONFIG_INET
+	struct sock *sk = ctx->sk;
+	int err;
+
+	/* Binding to port can be expensive so it's prohibited in the helper.
+	 * Only binding to IP is supported.
+	 */
+	err = -EINVAL;
+	if (addr->sa_family == AF_INET) {
+		if (addr_len < sizeof(struct sockaddr_in))
+			return err;
+		if (((struct sockaddr_in *)addr)->sin_port != htons(0))
+			return err;
+		return __inet_bind(sk, addr, addr_len, true, false);
+#if IS_ENABLED(CONFIG_IPV6)
+	} else if (addr->sa_family == AF_INET6) {
+		if (addr_len < SIN6_LEN_RFC2133)
+			return err;
+		if (((struct sockaddr_in6 *)addr)->sin6_port != htons(0))
+			return err;
+		/* ipv6_bpf_stub cannot be NULL, since it's called from
+		 * bpf_cgroup_inet6_connect hook and ipv6 is already loaded
+		 */
+		return ipv6_bpf_stub->inet6_bind(sk, addr, addr_len, true, false);
+#endif /* CONFIG_IPV6 */
+	}
+#endif /* CONFIG_INET */
+
+	return -EAFNOSUPPORT;
+}
+
+static const struct bpf_func_proto bpf_bind_proto = {
+	.func		= bpf_bind,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_PTR_TO_CTX,
+	.arg2_type	= ARG_PTR_TO_MEM,
+	.arg3_type	= ARG_CONST_SIZE,
+};
+
 static const struct bpf_func_proto *
 bpf_base_func_proto(enum bpf_func_id func_id)
 {
@@ -3707,6 +3754,14 @@ sock_addr_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 	 */
 	case BPF_FUNC_get_current_uid_gid:
 		return &bpf_get_current_uid_gid_proto;
+	case BPF_FUNC_bind:
+		switch (prog->expected_attach_type) {
+		case BPF_CGROUP_INET4_CONNECT:
+		case BPF_CGROUP_INET6_CONNECT:
+			return &bpf_bind_proto;
+		default:
+			return NULL;
+		}
 	default:
 		return bpf_base_func_proto(func_id);
 	}
@@ -4213,6 +4268,7 @@ static bool sock_addr_is_valid_access(int off, int size,
 	case bpf_ctx_range(struct bpf_sock_addr, user_ip4):
 		switch (prog->expected_attach_type) {
 		case BPF_CGROUP_INET4_BIND:
+		case BPF_CGROUP_INET4_CONNECT:
 			break;
 		default:
 			return false;
@@ -4221,6 +4277,7 @@ static bool sock_addr_is_valid_access(int off, int size,
 	case bpf_ctx_range_till(struct bpf_sock_addr, user_ip6[0], user_ip6[3]):
 		switch (prog->expected_attach_type) {
 		case BPF_CGROUP_INET6_BIND:
+		case BPF_CGROUP_INET6_CONNECT:
 			break;
 		default:
 			return false;

net/ipv4/af_inet.c

@@ -547,12 +547,19 @@ int inet_dgram_connect(struct socket *sock, struct sockaddr *uaddr,
 		       int addr_len, int flags)
 {
 	struct sock *sk = sock->sk;
+	int err;
 
 	if (addr_len < sizeof(uaddr->sa_family))
 		return -EINVAL;
 	if (uaddr->sa_family == AF_UNSPEC)
 		return sk->sk_prot->disconnect(sk, flags);
 
+	if (BPF_CGROUP_PRE_CONNECT_ENABLED(sk)) {
+		err = sk->sk_prot->pre_connect(sk, uaddr, addr_len);
+		if (err)
+			return err;
+	}
+
 	if (!inet_sk(sk)->inet_num && inet_autobind(sk))
 		return -EAGAIN;
 	return sk->sk_prot->connect(sk, uaddr, addr_len);
@@ -633,6 +640,12 @@ int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr,
 		if (sk->sk_state != TCP_CLOSE)
 			goto out;
 
+		if (BPF_CGROUP_PRE_CONNECT_ENABLED(sk)) {
+			err = sk->sk_prot->pre_connect(sk, uaddr, addr_len);
+			if (err)
+				goto out;
+		}
+
 		err = sk->sk_prot->connect(sk, uaddr, addr_len);
 		if (err < 0)
 			goto out;

net/ipv4/tcp_ipv4.c

@@ -140,6 +140,21 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp)
 }
 EXPORT_SYMBOL_GPL(tcp_twsk_unique);
 
+static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr,
+			      int addr_len)
+{
+	/* This check is replicated from tcp_v4_connect() and intended to
+	 * prevent BPF program called below from accessing bytes that are out
+	 * of the bound specified by user in addr_len.
+	 */
+	if (addr_len < sizeof(struct sockaddr_in))
+		return -EINVAL;
+
+	sock_owned_by_me(sk);
+
+	return BPF_CGROUP_RUN_PROG_INET4_CONNECT(sk, uaddr);
+}
+
 /* This will initiate an outgoing connection. */
 int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
@@ -2409,6 +2424,7 @@ struct proto tcp_prot = {
 	.name			= "TCP",
 	.owner			= THIS_MODULE,
 	.close			= tcp_close,
+	.pre_connect		= tcp_v4_pre_connect,
 	.connect		= tcp_v4_connect,
 	.disconnect		= tcp_disconnect,
 	.accept			= inet_csk_accept,

net/ipv4/udp.c

@@ -1658,6 +1658,19 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int noblock,
 	goto try_again;
 }
 
+int udp_pre_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
+{
+	/* This check is replicated from __ip4_datagram_connect() and
+	 * intended to prevent BPF program called below from accessing bytes
+	 * that are out of the bound specified by user in addr_len.
+	 */
+	if (addr_len < sizeof(struct sockaddr_in))
+		return -EINVAL;
+
+	return BPF_CGROUP_RUN_PROG_INET4_CONNECT_LOCK(sk, uaddr);
+}
+EXPORT_SYMBOL(udp_pre_connect);
+
 int __udp_disconnect(struct sock *sk, int flags)
 {
 	struct inet_sock *inet = inet_sk(sk);
@@ -2530,6 +2543,7 @@ struct proto udp_prot = {
 	.name			= "UDP",
 	.owner			= THIS_MODULE,
 	.close			= udp_lib_close,
+	.pre_connect		= udp_pre_connect,
 	.connect		= ip4_datagram_connect,
 	.disconnect		= udp_disconnect,
 	.ioctl			= udp_ioctl,

net/ipv6/af_inet6.c

@@ -887,6 +887,10 @@ static const struct ipv6_stub ipv6_stub_impl = {
 	.nd_tbl	= &nd_tbl,
 };
 
+static const struct ipv6_bpf_stub ipv6_bpf_stub_impl = {
+	.inet6_bind = __inet6_bind,
+};
+
 static int __init inet6_init(void)
 {
 	struct list_head *r;
@@ -1043,6 +1047,7 @@ static int __init inet6_init(void)
 	/* ensure that ipv6 stubs are visible only after ipv6 is ready */
 	wmb();
 	ipv6_stub = &ipv6_stub_impl;
+	ipv6_bpf_stub = &ipv6_bpf_stub_impl;
 out:
 	return err;
 

net/ipv6/tcp_ipv6.c

@@ -117,6 +117,21 @@ static u32 tcp_v6_init_ts_off(const struct net *net, const struct sk_buff *skb)
 				   ipv6_hdr(skb)->saddr.s6_addr32);
 }
 
+static int tcp_v6_pre_connect(struct sock *sk, struct sockaddr *uaddr,
+			      int addr_len)
+{
+	/* This check is replicated from tcp_v6_connect() and intended to
+	 * prevent BPF program called below from accessing bytes that are out
+	 * of the bound specified by user in addr_len.
+	 */
+	if (addr_len < SIN6_LEN_RFC2133)
+		return -EINVAL;
+
+	sock_owned_by_me(sk);
+
+	return BPF_CGROUP_RUN_PROG_INET6_CONNECT(sk, uaddr);
+}
+
 static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 			  int addr_len)
 {
@@ -1925,6 +1940,7 @@ struct proto tcpv6_prot = {
 	.name			= "TCPv6",
 	.owner			= THIS_MODULE,
 	.close			= tcp_close,
+	.pre_connect		= tcp_v6_pre_connect,
 	.connect		= tcp_v6_connect,
 	.disconnect		= tcp_disconnect,
 	.accept			= inet_csk_accept,