Merge tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next

Daniel Borkmann says:

====================
pull-request: bpf-next 2024-08-23

We've added 10 non-merge commits during the last 15 day(s) which contain
a total of 10 files changed, 222 insertions(+), 190 deletions(-).

The main changes are:

1) Add TCP_BPF_SOCK_OPS_CB_FLAGS to bpf_*sockopt() to address the case
   when long-lived sockets miss a chance to set additional callbacks
   if a sockops program was not attached early in their lifetime,
   from Alan Maguire.

2) Add a batch of BPF selftest improvements which fix a few bugs and add
   missing features to improve the test coverage of sockmap/sockhash,
   from Michal Luczaj.

3) Fix a false-positive Smatch-reported off-by-one in tcp_validate_cookie()
   which is part of the test_tcp_custom_syncookie BPF selftest,
   from Kuniyuki Iwashima.

4) Fix the flow_dissector BPF selftest which had a bug in IP header's
   tot_len calculation doing subtraction after htons() instead of inside
   htons(), from Asbjørn Sloth Tønnesen.

* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next:
  selftest: bpf: Remove mssind boundary check in test_tcp_custom_syncookie.c.
  selftests/bpf: Introduce __attribute__((cleanup)) in create_pair()
  selftests/bpf: Exercise SOCK_STREAM unix_inet_redir_to_connected()
  selftests/bpf: Honour the sotype of af_unix redir tests
  selftests/bpf: Simplify inet_socketpair() and vsock_socketpair_connectible()
  selftests/bpf: Socket pair creation, cleanups
  selftests/bpf: Support more socket types in create_pair()
  selftests/bpf: Avoid subtraction after htons() in ipip tests
  selftests/bpf: add sockopt tests for TCP_BPF_SOCK_OPS_CB_FLAGS
  bpf/bpf_get,set_sockopt: add option to set TCP-BPF sock ops flags
====================

Link: https://patch.msgid.link/20240823134959.1091-1-daniel@iogearbox.netSigned-off-by: Jakub Kicinski <kuba@kernel.org>

include/uapi/linux/bpf.h

@@ -2851,7 +2851,7 @@ union bpf_attr {
  * 		  **TCP_SYNCNT**, **TCP_USER_TIMEOUT**, **TCP_NOTSENT_LOWAT**,
  * 		  **TCP_NODELAY**, **TCP_MAXSEG**, **TCP_WINDOW_CLAMP**,
  * 		  **TCP_THIN_LINEAR_TIMEOUTS**, **TCP_BPF_DELACK_MAX**,
- * 		  **TCP_BPF_RTO_MIN**.
+ *		  **TCP_BPF_RTO_MIN**, **TCP_BPF_SOCK_OPS_CB_FLAGS**.
  * 		* **IPPROTO_IP**, which supports *optname* **IP_TOS**.
  * 		* **IPPROTO_IPV6**, which supports the following *optname*\ s:
  * 		  **IPV6_TCLASS**, **IPV6_AUTOFLOWLABEL**.
@@ -7080,6 +7080,7 @@ enum {
 	TCP_BPF_SYN		= 1005, /* Copy the TCP header */
 	TCP_BPF_SYN_IP		= 1006, /* Copy the IP[46] and TCP header */
 	TCP_BPF_SYN_MAC         = 1007, /* Copy the MAC, IP[46], and TCP header */
+	TCP_BPF_SOCK_OPS_CB_FLAGS = 1008, /* Get or Set TCP sock ops flags */
 };
 
 enum {

net/core/filter.c

@@ -5279,6 +5279,11 @@ static int bpf_sol_tcp_setsockopt(struct sock *sk, int optname,
 			return -EINVAL;
 		inet_csk(sk)->icsk_rto_min = timeout;
 		break;
+	case TCP_BPF_SOCK_OPS_CB_FLAGS:
+		if (val & ~(BPF_SOCK_OPS_ALL_CB_FLAGS))
+			return -EINVAL;
+		tp->bpf_sock_ops_cb_flags = val;
+		break;
 	default:
 		return -EINVAL;
 	}
@@ -5367,6 +5372,17 @@ static int sol_tcp_sockopt(struct sock *sk, int optname,
 		if (*optlen < 1)
 			return -EINVAL;
 		break;
+	case TCP_BPF_SOCK_OPS_CB_FLAGS:
+		if (*optlen != sizeof(int))
+			return -EINVAL;
+		if (getopt) {
+			struct tcp_sock *tp = tcp_sk(sk);
+			int cb_flags = tp->bpf_sock_ops_cb_flags;
+
+			memcpy(optval, &cb_flags, *optlen);
+			return 0;
+		}
+		return bpf_sol_tcp_setsockopt(sk, optname, optval, *optlen);
 	default:
 		if (getopt)
 			return -EINVAL;

tools/include/uapi/linux/bpf.h

@@ -2851,7 +2851,7 @@ union bpf_attr {
  * 		  **TCP_SYNCNT**, **TCP_USER_TIMEOUT**, **TCP_NOTSENT_LOWAT**,
  * 		  **TCP_NODELAY**, **TCP_MAXSEG**, **TCP_WINDOW_CLAMP**,
  * 		  **TCP_THIN_LINEAR_TIMEOUTS**, **TCP_BPF_DELACK_MAX**,
- * 		  **TCP_BPF_RTO_MIN**.
+ *		  **TCP_BPF_RTO_MIN**, **TCP_BPF_SOCK_OPS_CB_FLAGS**.
  * 		* **IPPROTO_IP**, which supports *optname* **IP_TOS**.
  * 		* **IPPROTO_IPV6**, which supports the following *optname*\ s:
  * 		  **IPV6_TCLASS**, **IPV6_AUTOFLOWLABEL**.
@@ -7080,6 +7080,7 @@ enum {
 	TCP_BPF_SYN		= 1005, /* Copy the TCP header */
 	TCP_BPF_SYN_IP		= 1006, /* Copy the IP[46] and TCP header */
 	TCP_BPF_SYN_MAC         = 1007, /* Copy the MAC, IP[46], and TCP header */
+	TCP_BPF_SOCK_OPS_CB_FLAGS = 1008, /* Get or Set TCP sock ops flags */
 };
 
 enum {

tools/testing/selftests/bpf/prog_tests/flow_dissector.c

@@ -378,8 +378,8 @@ struct test tests[] = {
 			.iph_inner.ihl = 5,
 			.iph_inner.protocol = IPPROTO_TCP,
 			.iph_inner.tot_len =
-				__bpf_constant_htons(MAGIC_BYTES) -
-				sizeof(struct iphdr),
+				__bpf_constant_htons(MAGIC_BYTES -
+				sizeof(struct iphdr)),
 			.tcp.doff = 5,
 			.tcp.source = 80,
 			.tcp.dest = 8080,
@@ -407,8 +407,8 @@ struct test tests[] = {
 			.iph_inner.ihl = 5,
 			.iph_inner.protocol = IPPROTO_TCP,
 			.iph_inner.tot_len =
-				__bpf_constant_htons(MAGIC_BYTES) -
-				sizeof(struct iphdr),
+				__bpf_constant_htons(MAGIC_BYTES -
+				sizeof(struct iphdr)),
 			.tcp.doff = 5,
 			.tcp.source = 80,
 			.tcp.dest = 8080,
@@ -436,8 +436,8 @@ struct test tests[] = {
 			.iph_inner.ihl = 5,
 			.iph_inner.protocol = IPPROTO_TCP,
 			.iph_inner.tot_len =
-				__bpf_constant_htons(MAGIC_BYTES) -
-				sizeof(struct iphdr),
+				__bpf_constant_htons(MAGIC_BYTES -
+				sizeof(struct iphdr)),
 			.tcp.doff = 5,
 			.tcp.source = 99,
 			.tcp.dest = 9090,

tools/testing/selftests/bpf/prog_tests/setget_sockopt.c

@@ -154,6 +154,51 @@ static void test_ktls(int family)
 	close(sfd);
 }
 
+static void test_nonstandard_opt(int family)
+{
+	struct setget_sockopt__bss *bss = skel->bss;
+	struct bpf_link *getsockopt_link = NULL;
+	int sfd = -1, fd = -1, cfd = -1, flags;
+	socklen_t flagslen = sizeof(flags);
+
+	memset(bss, 0, sizeof(*bss));
+
+	sfd = start_server(family, SOCK_STREAM,
+			   family == AF_INET6 ? addr6_str : addr4_str, 0, 0);
+	if (!ASSERT_GE(sfd, 0, "start_server"))
+		return;
+
+	fd = connect_to_fd(sfd, 0);
+	if (!ASSERT_GE(fd, 0, "connect_to_fd_server"))
+		goto err_out;
+
+	/* cgroup/getsockopt prog will intercept getsockopt() below and
+	 * retrieve the tcp socket bpf_sock_ops_cb_flags value for the
+	 * accept()ed socket; this was set earlier in the passive established
+	 * callback for the accept()ed socket via bpf_setsockopt().
+	 */
+	getsockopt_link = bpf_program__attach_cgroup(skel->progs._getsockopt, cg_fd);
+	if (!ASSERT_OK_PTR(getsockopt_link, "getsockopt prog"))
+		goto err_out;
+
+	cfd = accept(sfd, NULL, 0);
+	if (!ASSERT_GE(cfd, 0, "accept"))
+		goto err_out;
+
+	if (!ASSERT_OK(getsockopt(cfd, SOL_TCP, TCP_BPF_SOCK_OPS_CB_FLAGS, &flags, &flagslen),
+		       "getsockopt_flags"))
+		goto err_out;
+	ASSERT_EQ(flags & BPF_SOCK_OPS_STATE_CB_FLAG, BPF_SOCK_OPS_STATE_CB_FLAG,
+		  "cb_flags_set");
+err_out:
+	close(sfd);
+	if (fd != -1)
+		close(fd);
+	if (cfd != -1)
+		close(cfd);
+	bpf_link__destroy(getsockopt_link);
+}
+
 void test_setget_sockopt(void)
 {
 	cg_fd = test__join_cgroup(CG_NAME);
@@ -191,6 +236,8 @@ void test_setget_sockopt(void)
 	test_udp(AF_INET);
 	test_ktls(AF_INET6);
 	test_ktls(AF_INET);
+	test_nonstandard_opt(AF_INET);
+	test_nonstandard_opt(AF_INET6);
 
 done:
 	setget_sockopt__destroy(skel);

tools/testing/selftests/bpf/prog_tests/sockmap_basic.c

@@ -451,11 +451,11 @@ static void test_sockmap_progs_query(enum bpf_attach_type attach_type)
 #define MAX_EVENTS 10
 static void test_sockmap_skb_verdict_shutdown(void)
 {
+	int n, err, map, verdict, c1 = -1, p1 = -1;
 	struct epoll_event ev, events[MAX_EVENTS];
-	int n, err, map, verdict, s, c1 = -1, p1 = -1;
 	struct test_sockmap_pass_prog *skel;
-	int epollfd;
 	int zero = 0;
+	int epollfd;
 	char b;
 
 	skel = test_sockmap_pass_prog__open_and_load();
@@ -469,10 +469,7 @@ static void test_sockmap_skb_verdict_shutdown(void)
 	if (!ASSERT_OK(err, "bpf_prog_attach"))
 		goto out;
 
-	s = socket_loopback(AF_INET, SOCK_STREAM);
-	if (s < 0)
-		goto out;
-	err = create_pair(s, AF_INET, SOCK_STREAM, &c1, &p1);
+	err = create_pair(AF_INET, SOCK_STREAM, &c1, &p1);
 	if (err < 0)
 		goto out;
 
@@ -506,8 +503,8 @@ static void test_sockmap_skb_verdict_shutdown(void)
 
 static void test_sockmap_skb_verdict_fionread(bool pass_prog)
 {
+	int err, map, verdict, c0 = -1, c1 = -1, p0 = -1, p1 = -1;
 	int expected, zero = 0, sent, recvd, avail;
-	int err, map, verdict, s, c0 = -1, c1 = -1, p0 = -1, p1 = -1;
 	struct test_sockmap_pass_prog *pass = NULL;
 	struct test_sockmap_drop_prog *drop = NULL;
 	char buf[256] = "0123456789";
@@ -534,11 +531,8 @@ static void test_sockmap_skb_verdict_fionread(bool pass_prog)
 	if (!ASSERT_OK(err, "bpf_prog_attach"))
 		goto out;
 
-	s = socket_loopback(AF_INET, SOCK_STREAM);
-	if (!ASSERT_GT(s, -1, "socket_loopback(s)"))
-		goto out;
-	err = create_socket_pairs(s, AF_INET, SOCK_STREAM, &c0, &c1, &p0, &p1);
-	if (!ASSERT_OK(err, "create_socket_pairs(s)"))
+	err = create_socket_pairs(AF_INET, SOCK_STREAM, &c0, &c1, &p0, &p1);
+	if (!ASSERT_OK(err, "create_socket_pairs()"))
 		goto out;
 
 	err = bpf_map_update_elem(map, &zero, &c1, BPF_NOEXIST);
@@ -570,16 +564,12 @@ static void test_sockmap_skb_verdict_fionread(bool pass_prog)
 
 static void test_sockmap_skb_verdict_peek_helper(int map)
 {
-	int err, s, c1, p1, zero = 0, sent, recvd, avail;
+	int err, c1, p1, zero = 0, sent, recvd, avail;
 	char snd[256] = "0123456789";
 	char rcv[256] = "0";
 
-	s = socket_loopback(AF_INET, SOCK_STREAM);
-	if (!ASSERT_GT(s, -1, "socket_loopback(s)"))
-		return;
-
-	err = create_pair(s, AF_INET, SOCK_STREAM, &c1, &p1);
-	if (!ASSERT_OK(err, "create_pairs(s)"))
+	err = create_pair(AF_INET, SOCK_STREAM, &c1, &p1);
+	if (!ASSERT_OK(err, "create_pair()"))
 		return;
 
 	err = bpf_map_update_elem(map, &zero, &c1, BPF_NOEXIST);

tools/testing/selftests/bpf/prog_tests/sockmap_helpers.h

@@ -3,6 +3,9 @@
 
 #include <linux/vm_sockets.h>
 
+/* include/linux/net.h */
+#define SOCK_TYPE_MASK 0xf
+
 #define IO_TIMEOUT_SEC 30
 #define MAX_STRERR_LEN 256
 #define MAX_TEST_NAME 80
@@ -14,6 +17,17 @@
 
 #define __always_unused	__attribute__((__unused__))
 
+/* include/linux/cleanup.h */
+#define __get_and_null(p, nullvalue)                                           \
+	({                                                                     \
+		__auto_type __ptr = &(p);                                      \
+		__auto_type __val = *__ptr;                                    \
+		*__ptr = nullvalue;                                            \
+		__val;                                                         \
+	})
+
+#define take_fd(fd) __get_and_null(fd, -EBADF)
+
 #define _FAIL(errnum, fmt...)                                                  \
 	({                                                                     \
 		error_at_line(0, (errnum), __func__, __LINE__, fmt);           \
@@ -179,6 +193,14 @@
 		__ret;                                                         \
 	})
 
+static inline void close_fd(int *fd)
+{
+	if (*fd >= 0)
+		xclose(*fd);
+}
+
+#define __close_fd __attribute__((cleanup(close_fd)))
+
 static inline int poll_connect(int fd, unsigned int timeout_sec)
 {
 	struct timeval timeout = { .tv_sec = timeout_sec };
@@ -312,54 +334,6 @@ static inline int add_to_sockmap(int sock_mapfd, int fd1, int fd2)
 	return xbpf_map_update_elem(sock_mapfd, &key, &value, BPF_NOEXIST);
 }
 
-static inline int create_pair(int s, int family, int sotype, int *c, int *p)
-{
-	struct sockaddr_storage addr;
-	socklen_t len;
-	int err = 0;
-
-	len = sizeof(addr);
-	err = xgetsockname(s, sockaddr(&addr), &len);
-	if (err)
-		return err;
-
-	*c = xsocket(family, sotype, 0);
-	if (*c < 0)
-		return errno;
-	err = xconnect(*c, sockaddr(&addr), len);
-	if (err) {
-		err = errno;
-		goto close_cli0;
-	}
-
-	*p = xaccept_nonblock(s, NULL, NULL);
-	if (*p < 0) {
-		err = errno;
-		goto close_cli0;
-	}
-	return err;
-close_cli0:
-	close(*c);
-	return err;
-}
-
-static inline int create_socket_pairs(int s, int family, int sotype,
-				      int *c0, int *c1, int *p0, int *p1)
-{
-	int err;
-
-	err = create_pair(s, family, sotype, c0, p0);
-	if (err)
-		return err;
-
-	err = create_pair(s, family, sotype, c1, p1);
-	if (err) {
-		close(*c0);
-		close(*p0);
-	}
-	return err;
-}
-
 static inline int enable_reuseport(int s, int progfd)
 {
 	int err, one = 1;
@@ -412,5 +386,84 @@ static inline int socket_loopback(int family, int sotype)
 	return socket_loopback_reuseport(family, sotype, -1);
 }
 
+static inline int create_pair(int family, int sotype, int *p0, int *p1)
+{
+	__close_fd int s, c = -1, p = -1;
+	struct sockaddr_storage addr;
+	socklen_t len = sizeof(addr);
+	int err;
+
+	s = socket_loopback(family, sotype);
+	if (s < 0)
+		return s;
+
+	err = xgetsockname(s, sockaddr(&addr), &len);
+	if (err)
+		return err;
+
+	c = xsocket(family, sotype, 0);
+	if (c < 0)
+		return c;
+
+	err = connect(c, sockaddr(&addr), len);
+	if (err) {
+		if (errno != EINPROGRESS) {
+			FAIL_ERRNO("connect");
+			return err;
+		}
+
+		err = poll_connect(c, IO_TIMEOUT_SEC);
+		if (err) {
+			FAIL_ERRNO("poll_connect");
+			return err;
+		}
+	}
+
+	switch (sotype & SOCK_TYPE_MASK) {
+	case SOCK_DGRAM:
+		err = xgetsockname(c, sockaddr(&addr), &len);
+		if (err)
+			return err;
+
+		err = xconnect(s, sockaddr(&addr), len);
+		if (err)
+			return err;
+
+		*p0 = take_fd(s);
+		break;
+	case SOCK_STREAM:
+	case SOCK_SEQPACKET:
+		p = xaccept_nonblock(s, NULL, NULL);
+		if (p < 0)
+			return p;
+
+		*p0 = take_fd(p);
+		break;
+	default:
+		FAIL("Unsupported socket type %#x", sotype);
+		return -EOPNOTSUPP;
+	}
+
+	*p1 = take_fd(c);
+	return 0;
+}
+
+static inline int create_socket_pairs(int family, int sotype, int *c0, int *c1,
+				      int *p0, int *p1)
+{
+	int err;
+
+	err = create_pair(family, sotype, c0, p0);
+	if (err)
+		return err;
+
+	err = create_pair(family, sotype, c1, p1);
+	if (err) {
+		close(*c0);
+		close(*p0);
+	}
+
+	return err;
+}
 
 #endif // __SOCKMAP_HELPERS__

tools/testing/selftests/bpf/prog_tests/sockmap_listen.c

@@ -677,7 +677,7 @@ static void redir_to_connected(int family, int sotype, int sock_mapfd,
 			       int verd_mapfd, enum redir_mode mode)
 {
 	const char *log_prefix = redir_mode_str(mode);
-	int s, c0, c1, p0, p1;
+	int c0, c1, p0, p1;
 	unsigned int pass;
 	int err, n;
 	u32 key;
@@ -685,13 +685,10 @@ static void redir_to_connected(int family, int sotype, int sock_mapfd,
 
 	zero_verdict_count(verd_mapfd);
 
-	s = socket_loopback(family, sotype | SOCK_NONBLOCK);
-	if (s < 0)
-		return;
-
-	err = create_socket_pairs(s, family, sotype, &c0, &c1, &p0, &p1);
+	err = create_socket_pairs(family, sotype | SOCK_NONBLOCK, &c0, &c1,
+				  &p0, &p1);
 	if (err)
-		goto close_srv;
+		return;
 
 	err = add_to_sockmap(sock_mapfd, p0, p1);
 	if (err)
@@ -722,8 +719,6 @@ static void redir_to_connected(int family, int sotype, int sock_mapfd,
 	xclose(c1);
 	xclose(p0);
 	xclose(c0);
-close_srv:
-	xclose(s);
 }
 
 static void test_skb_redir_to_connected(struct test_sockmap_listen *skel,
@@ -909,7 +904,7 @@ static void test_msg_redir_to_listening_with_link(struct test_sockmap_listen *sk
 
 static void redir_partial(int family, int sotype, int sock_map, int parser_map)
 {
-	int s, c0 = -1, c1 = -1, p0 = -1, p1 = -1;
+	int c0 = -1, c1 = -1, p0 = -1, p1 = -1;
 	int err, n, key, value;
 	char buf[] = "abc";
 
@@ -919,13 +914,10 @@ static void redir_partial(int family, int sotype, int sock_map, int parser_map)
 	if (err)
 		return;
 
-	s = socket_loopback(family, sotype | SOCK_NONBLOCK);
-	if (s < 0)
-		goto clean_parser_map;
-
-	err = create_socket_pairs(s, family, sotype, &c0, &c1, &p0, &p1);
+	err = create_socket_pairs(family, sotype | SOCK_NONBLOCK, &c0, &c1,
+				  &p0, &p1);
 	if (err)
-		goto close_srv;
+		goto clean_parser_map;
 
 	err = add_to_sockmap(sock_map, p0, p1);
 	if (err)
@@ -944,8 +936,6 @@ static void redir_partial(int family, int sotype, int sock_map, int parser_map)
 	xclose(p0);
 	xclose(c1);
 	xclose(p1);
-close_srv:
-	xclose(s);
 
 clean_parser_map:
 	key = 0;
@@ -1500,49 +1490,7 @@ static void test_unix_redir(struct test_sockmap_listen *skel, struct bpf_map *ma
 /* Returns two connected loopback vsock sockets */
 static int vsock_socketpair_connectible(int sotype, int *v0, int *v1)
 {
-	struct sockaddr_storage addr;
-	socklen_t len = sizeof(addr);
-	int s, p, c;
-
-	s = socket_loopback(AF_VSOCK, sotype);
-	if (s < 0)
-		return -1;
-
-	c = xsocket(AF_VSOCK, sotype | SOCK_NONBLOCK, 0);
-	if (c == -1)
-		goto close_srv;
-
-	if (getsockname(s, sockaddr(&addr), &len) < 0)
-		goto close_cli;
-
-	if (connect(c, sockaddr(&addr), len) < 0 && errno != EINPROGRESS) {
-		FAIL_ERRNO("connect");
-		goto close_cli;
-	}
-
-	len = sizeof(addr);
-	p = accept_timeout(s, sockaddr(&addr), &len, IO_TIMEOUT_SEC);
-	if (p < 0)
-		goto close_cli;
-
-	if (poll_connect(c, IO_TIMEOUT_SEC) < 0) {
-		FAIL_ERRNO("poll_connect");
-		goto close_acc;
-	}
-
-	*v0 = p;
-	*v1 = c;
-
-	return 0;
-
-close_acc:
-	close(p);
-close_cli:
-	close(c);
-close_srv:
-	close(s);
-
-	return -1;
+	return create_pair(AF_VSOCK, sotype | SOCK_NONBLOCK, v0, v1);
 }
 
 static void vsock_unix_redir_connectible(int sock_mapfd, int verd_mapfd,
@@ -1691,44 +1639,7 @@ static void test_reuseport(struct test_sockmap_listen *skel,
 
 static int inet_socketpair(int family, int type, int *s, int *c)
 {
-	struct sockaddr_storage addr;
-	socklen_t len;
-	int p0, c0;
-	int err;
-
-	p0 = socket_loopback(family, type | SOCK_NONBLOCK);
-	if (p0 < 0)
-		return p0;
-
-	len = sizeof(addr);
-	err = xgetsockname(p0, sockaddr(&addr), &len);
-	if (err)
-		goto close_peer0;
-
-	c0 = xsocket(family, type | SOCK_NONBLOCK, 0);
-	if (c0 < 0) {
-		err = c0;
-		goto close_peer0;
-	}
-	err = xconnect(c0, sockaddr(&addr), len);
-	if (err)
-		goto close_cli0;
-	err = xgetsockname(c0, sockaddr(&addr), &len);
-	if (err)
-		goto close_cli0;
-	err = xconnect(p0, sockaddr(&addr), len);
-	if (err)
-		goto close_cli0;
-
-	*s = p0;
-	*c = c0;
-	return 0;
-
-close_cli0:
-	xclose(c0);
-close_peer0:
-	xclose(p0);
-	return err;
+	return create_pair(family, type | SOCK_NONBLOCK, s, c);
 }
 
 static void udp_redir_to_connected(int family, int sock_mapfd, int verd_mapfd,
@@ -1795,11 +1706,11 @@ static void inet_unix_redir_to_connected(int family, int type, int sock_mapfd,
 	int sfd[2];
 	int err;
 
-	if (socketpair(AF_UNIX, SOCK_DGRAM | SOCK_NONBLOCK, 0, sfd))
+	if (socketpair(AF_UNIX, type | SOCK_NONBLOCK, 0, sfd))
 		return;
 	c0 = sfd[0], p0 = sfd[1];
 
-	err = inet_socketpair(family, SOCK_DGRAM, &p1, &c1);
+	err = inet_socketpair(family, type, &p1, &c1);
 	if (err)
 		goto close;
 
@@ -1847,7 +1758,7 @@ static void unix_inet_redir_to_connected(int family, int type, int sock_mapfd,
 	int sfd[2];
 	int err;
 
-	err = inet_socketpair(family, SOCK_DGRAM, &p0, &c0);
+	err = inet_socketpair(family, type, &p0, &c0);
 	if (err)
 		return;
 
@@ -1882,7 +1793,7 @@ static void unix_inet_skb_redir_to_connected(struct test_sockmap_listen *skel,
 	unix_inet_redir_to_connected(family, SOCK_DGRAM,
 				     sock_map, -1, verdict_map,
 				     REDIR_EGRESS, NO_FLAGS);
-	unix_inet_redir_to_connected(family, SOCK_DGRAM,
+	unix_inet_redir_to_connected(family, SOCK_STREAM,
 				     sock_map, -1, verdict_map,
 				     REDIR_EGRESS, NO_FLAGS);
 

tools/testing/selftests/bpf/progs/setget_sockopt.c

@@ -59,6 +59,8 @@ static const struct sockopt_test sol_tcp_tests[] = {
 	{ .opt = TCP_THIN_LINEAR_TIMEOUTS, .flip = 1, },
 	{ .opt = TCP_USER_TIMEOUT, .new = 123400, .expected = 123400, },
 	{ .opt = TCP_NOTSENT_LOWAT, .new = 1314, .expected = 1314, },
+	{ .opt = TCP_BPF_SOCK_OPS_CB_FLAGS, .new = BPF_SOCK_OPS_ALL_CB_FLAGS,
+	  .expected = BPF_SOCK_OPS_ALL_CB_FLAGS, },
 	{ .opt = 0, },
 };
 
@@ -353,11 +355,30 @@ int BPF_PROG(socket_post_create, struct socket *sock, int family,
 	return 1;
 }
 
+SEC("cgroup/getsockopt")
+int _getsockopt(struct bpf_sockopt *ctx)
+{
+	struct bpf_sock *sk = ctx->sk;
+	int *optval = ctx->optval;
+	struct tcp_sock *tp;
+
+	if (!sk || ctx->level != SOL_TCP || ctx->optname != TCP_BPF_SOCK_OPS_CB_FLAGS)
+		return 1;
+
+	tp = bpf_core_cast(sk, struct tcp_sock);
+	if (ctx->optval + sizeof(int) <= ctx->optval_end) {
+		*optval = tp->bpf_sock_ops_cb_flags;
+		ctx->retval = 0;
+	}
+	return 1;
+}
+
 SEC("sockops")
 int skops_sockopt(struct bpf_sock_ops *skops)
 {
 	struct bpf_sock *bpf_sk = skops->sk;
 	struct sock *sk;
+	int flags;
 
 	if (!bpf_sk)
 		return 1;
@@ -384,9 +405,8 @@ int skops_sockopt(struct bpf_sock_ops *skops)
 		nr_passive += !(bpf_test_sockopt(skops, sk) ||
 				test_tcp_maxseg(skops, sk) ||
 				test_tcp_saved_syn(skops, sk));
-		bpf_sock_ops_cb_flags_set(skops,
-					  skops->bpf_sock_ops_cb_flags |
-					  BPF_SOCK_OPS_STATE_CB_FLAG);
+		flags = skops->bpf_sock_ops_cb_flags | BPF_SOCK_OPS_STATE_CB_FLAG;
+		bpf_setsockopt(skops, SOL_TCP, TCP_BPF_SOCK_OPS_CB_FLAGS, &flags, sizeof(flags));
 		break;
 	case BPF_SOCK_OPS_STATE_CB:
 		if (skops->args[1] == BPF_TCP_CLOSE_WAIT)

tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.c

@@ -486,17 +486,10 @@ static int tcp_validate_cookie(struct tcp_syncookie *ctx)
 		goto err;
 
 	mssind = (cookie & (3 << 6)) >> 6;
-	if (ctx->ipv4) {
-		if (mssind > ARRAY_SIZE(msstab4))
-			goto err;
-
+	if (ctx->ipv4)
 		ctx->attrs.mss = msstab4[mssind];
-	} else {
-		if (mssind > ARRAY_SIZE(msstab6))
-			goto err;
-
+	else
 		ctx->attrs.mss = msstab6[mssind];
-	}
 
 	ctx->attrs.snd_wscale = cookie & BPF_SYNCOOKIE_WSCALE_MASK;
 	ctx->attrs.rcv_wscale = ctx->attrs.snd_wscale;