[PATCH mptcp-next v3 00/14] selftests: consolidate TCP_INQ testing into sockopt

[PATCH mptcp-next v3 00/14] selftests: consolidate TCP_INQ testing into sockopt

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

v3:
 - No longer using process_one_client_inq() and connect_one_server_inq()
   for switching; instead, the inq-related code has been merged into
   process_one_client() and connect_one_server().

v2:
 - patch 4, a new patch, add do_setsockopt_inq helper.
 - cleanups.
 - https://patchwork.kernel.org/project/mptcp/cover/cover.1754664106.git.tanggeliang@kylinos.cn/

v1:
 - https://patchwork.kernel.org/project/mptcp/cover/cover.1754620968.git.tanggeliang@kylinos.cn/

This series consolidates the TCP_INQ test functionality from mptcp_inq
into mptcp_sockopt, simplifying the test suite while maintaining full
test coverage.

Geliang Tang (14):
  selftests: mptcp: sockopt: replace /dev/urandom with getrandom
  selftests: mptcp: sockopt: add protocol arguments
  selftests: mptcp: sockopt: add inq argument
  selftests: mptcp: sockopt: set TCP_INQ on accepted sockets
  selftests: mptcp: sockopt: use unix socket instead of pipe
  selftests: mptcp: sockopt: read eof at the end of process_one_client
  selftests: mptcp: sockopt: use recvmsg instead of read
  selftests: mptcp: sockopt: read one byte first
  selftests: mptcp: sockopt: add large data transfer tests
  selftests: mptcp: sockopt: add wait_for_ack
  selftests: mptcp: sockopt: set FIONREAD ioctl
  selftests: mptcp: sockopt: add check_tcp_inq helper
  selftests: mptcp: sockopt: replace mptcp_inq with sockopt
  selftests: mptcp: sockopt: remove obsolete mptcp_inq

 tools/testing/selftests/net/mptcp/.gitignore  |   1 -
 tools/testing/selftests/net/mptcp/Makefile    |   2 +-
 tools/testing/selftests/net/mptcp/mptcp_inq.c | 608 ------------------
 .../selftests/net/mptcp/mptcp_sockopt.c       | 346 ++++++++--
 .../selftests/net/mptcp/mptcp_sockopt.sh      |   8 +-
 5 files changed, 296 insertions(+), 669 deletions(-)
 delete mode 100644 tools/testing/selftests/net/mptcp/mptcp_inq.c

-- 
2.48.1

[PATCH mptcp-next v3 01/14] selftests: mptcp: sockopt: replace /dev/urandom with getrandom

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Replace /dev/urandom with getrandom() for initializing the RNG. This
simplifies the code and avoids potential failures from opening device
files while maintaining cryptographic quality randomness.

These codes are from mptcp_inq.c.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/net/mptcp/mptcp_sockopt.c       | 20 +++++++------------
 1 file changed, 7 insertions(+), 13 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index e934dd26a59d..2bd75f731dfd 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -20,6 +20,7 @@
 #include <sys/socket.h>
 #include <sys/types.h>
 #include <sys/wait.h>
+#include <sys/random.h>
 
 #include <netdb.h>
 #include <netinet/in.h>
@@ -815,21 +816,14 @@ static int rcheck(int wstatus, const char *what)
 
 static void init_rng(void)
 {
-	int fd = open("/dev/urandom", O_RDONLY);
+	unsigned int foo;
 
-	if (fd >= 0) {
-		unsigned int foo;
-		ssize_t ret;
-
-		/* can't fail */
-		ret = read(fd, &foo, sizeof(foo));
-		assert(ret == sizeof(foo));
-
-		close(fd);
-		srand(foo);
-	} else {
-		srand(time(NULL));
+	if (getrandom(&foo, sizeof(foo), 0) == -1) {
+		perror("getrandom");
+		exit(1);
 	}
+
+	srand(foo);
 }
 
 int main(int argc, char *argv[])
-- 
2.48.1

[PATCH mptcp-next v3 02/14] selftests: mptcp: sockopt: add protocol arguments

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Add -t and -r options to specify tx/rx protocols (TCP/MPTCP). This
increases testing flexibility by allowing explicit protocol selection
for both transmission and reception paths.

These codes are from mptcp_inq.c.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/net/mptcp/mptcp_sockopt.c       | 36 +++++++++++++++----
 1 file changed, 29 insertions(+), 7 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index 2bd75f731dfd..05caa9a57ac4 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -28,6 +28,8 @@
 #include <linux/tcp.h>
 
 static int pf = AF_INET;
+static int proto_tx = IPPROTO_MPTCP;
+static int proto_rx = IPPROTO_MPTCP;
 
 #ifndef IPPROTO_MPTCP
 #define IPPROTO_MPTCP 262
@@ -136,7 +138,7 @@ static void die_perror(const char *msg)
 
 static void die_usage(int r)
 {
-	fprintf(stderr, "Usage: mptcp_sockopt [-6]\n");
+	fprintf(stderr, "Usage: mptcp_sockopt [-6] [-t tcp|mptcp] [-r tcp|mptcp]\n");
 	exit(r);
 }
 
@@ -202,7 +204,7 @@ static int sock_listen_mptcp(const char * const listenaddr,
 	hints.ai_family = pf;
 
 	for (a = addr; a; a = a->ai_next) {
-		sock = socket(a->ai_family, a->ai_socktype, IPPROTO_MPTCP);
+		sock = socket(a->ai_family, a->ai_socktype, proto_rx);
 		if (sock < 0)
 			continue;
 
@@ -260,11 +262,22 @@ static int sock_connect_mptcp(const char * const remoteaddr,
 	return sock;
 }
 
+static int protostr_to_num(const char *s)
+{
+	if (strcasecmp(s, "tcp") == 0)
+		return IPPROTO_TCP;
+	if (strcasecmp(s, "mptcp") == 0)
+		return IPPROTO_MPTCP;
+
+	die_usage(1);
+	return 0;
+}
+
 static void parse_opts(int argc, char **argv)
 {
 	int c;
 
-	while ((c = getopt(argc, argv, "h6")) != -1) {
+	while ((c = getopt(argc, argv, "h6t:r:")) != -1) {
 		switch (c) {
 		case 'h':
 			die_usage(0);
@@ -272,6 +285,12 @@ static void parse_opts(int argc, char **argv)
 		case '6':
 			pf = AF_INET6;
 			break;
+		case 't':
+			proto_tx = protostr_to_num(optarg);
+			break;
+		case 'r':
+			proto_rx = protostr_to_num(optarg);
+			break;
 		default:
 			die_usage(1);
 			break;
@@ -557,6 +576,9 @@ static void do_getsockopt_mptcp_full_info(struct so_state *s, int fd)
 
 static void do_getsockopts(struct so_state *s, int fd, size_t r, size_t w)
 {
+	if (proto_tx != IPPROTO_MPTCP || proto_rx != IPPROTO_MPTCP)
+		return;
+
 	do_getsockopt_mptcp_info(s, fd, w);
 
 	do_getsockopt_tcp_info(s, fd, r, w);
@@ -632,7 +654,7 @@ static void connect_one_server(int fd, int pipefd)
 	if (eof)
 		total += 1; /* sequence advances due to FIN */
 
-	assert(s.mptcpi_rcv_delta == (uint64_t)total);
+	assert(s.mptcpi_rcv_delta ? s.mptcpi_rcv_delta == (uint64_t)total : true);
 	close(fd);
 }
 
@@ -667,7 +689,7 @@ static void process_one_client(int fd, int pipefd)
 		xerror("expected EOF, got %lu", ret3);
 
 	do_getsockopts(&s, fd, ret, ret2);
-	if (s.mptcpi_rcv_delta != (uint64_t)ret + 1)
+	if (s.mptcpi_rcv_delta && s.mptcpi_rcv_delta != (uint64_t)ret + 1)
 		xerror("mptcpi_rcv_delta %" PRIu64 ", expect %" PRIu64, s.mptcpi_rcv_delta, ret + 1, s.mptcpi_rcv_delta - ret);
 
 	/* be nice when running on top of older kernel */
@@ -772,10 +794,10 @@ static int client(int pipefd)
 
 	switch (pf) {
 	case AF_INET:
-		fd = sock_connect_mptcp("127.0.0.1", "15432", IPPROTO_MPTCP);
+		fd = sock_connect_mptcp("127.0.0.1", "15432", proto_tx);
 		break;
 	case AF_INET6:
-		fd = sock_connect_mptcp("::1", "15432", IPPROTO_MPTCP);
+		fd = sock_connect_mptcp("::1", "15432", proto_tx);
 		break;
 	default:
 		xerror("Unknown pf %d\n", pf);
-- 
2.48.1

[PATCH mptcp-next v3 03/14] selftests: mptcp: sockopt: add inq argument

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Introduce -i option to enable TCP_INQ testing. This prepares for
consolidating TCP_INQ tests into a single binary by adding the necessary
command-line interface.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/mptcp_sockopt.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index 05caa9a57ac4..cb3d8dc34997 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -30,6 +30,7 @@
 static int pf = AF_INET;
 static int proto_tx = IPPROTO_MPTCP;
 static int proto_rx = IPPROTO_MPTCP;
+static bool inq;
 
 #ifndef IPPROTO_MPTCP
 #define IPPROTO_MPTCP 262
@@ -138,7 +139,7 @@ static void die_perror(const char *msg)
 
 static void die_usage(int r)
 {
-	fprintf(stderr, "Usage: mptcp_sockopt [-6] [-t tcp|mptcp] [-r tcp|mptcp]\n");
+	fprintf(stderr, "Usage: mptcp_sockopt [-6] [-t tcp|mptcp] [-r tcp|mptcp] [-i]\n");
 	exit(r);
 }
 
@@ -277,7 +278,7 @@ static void parse_opts(int argc, char **argv)
 {
 	int c;
 
-	while ((c = getopt(argc, argv, "h6t:r:")) != -1) {
+	while ((c = getopt(argc, argv, "h6t:r:i")) != -1) {
 		switch (c) {
 		case 'h':
 			die_usage(0);
@@ -291,6 +292,9 @@ static void parse_opts(int argc, char **argv)
 		case 'r':
 			proto_rx = protostr_to_num(optarg);
 			break;
+		case 'i':
+			inq = true;
+			break;
 		default:
 			die_usage(1);
 			break;
-- 
2.48.1

[PATCH mptcp-next v3 04/14] selftests: mptcp: sockopt: set TCP_INQ on accepted sockets

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Add helper functions to conditionally set TCP_INQ sockopt on accepted
sockets when running in inq mode. This ensures proper testing of TCP_CM_INQ
functionality by enabling the required socket option on both ends of the
connection. This setup is necessary for validating TCP_CM_INQ behavior
during TCP_INQ-specific tests.

These codes are from mptcp_inq.c.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/mptcp_sockopt.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index cb3d8dc34997..366e4aa4df0b 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -715,6 +715,14 @@ static void process_one_client(int fd, int pipefd)
 	close(fd);
 }
 
+static void do_setsockopt_inq(int fd)
+{
+	int on = 1;
+
+	if (setsockopt(fd, IPPROTO_TCP, TCP_INQ, &on, sizeof(on)))
+		die_perror("setsockopt(TCP_INQ)");
+}
+
 static int xaccept(int s)
 {
 	int fd = accept(s, NULL, 0);
@@ -722,6 +730,9 @@ static int xaccept(int s)
 	if (fd < 0)
 		die_perror("accept");
 
+	if (inq)
+		do_setsockopt_inq(fd);
+
 	return fd;
 }
 
-- 
2.48.1

[PATCH mptcp-next v3 05/14] selftests: mptcp: sockopt: use unix socket instead of pipe

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Replaces pipe-based synchronization with Unix domain socket pairs
(SOCK_DGRAM) for better bidirectional communication reliability
between server and client processes.

These codes are from mptcp_inq.c.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/net/mptcp/mptcp_sockopt.c       | 36 +++++++++----------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index 366e4aa4df0b..add29bdda9e5 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -593,7 +593,7 @@ static void do_getsockopts(struct so_state *s, int fd, size_t r, size_t w)
 		do_getsockopt_mptcp_full_info(s, fd);
 }
 
-static void connect_one_server(int fd, int pipefd)
+static void connect_one_server(int fd, int unixfd)
 {
 	char buf[4096], buf2[4096];
 	size_t len, i, total;
@@ -618,9 +618,9 @@ static void connect_one_server(int fd, int pipefd)
 	do_getsockopts(&s, fd, 0, 0);
 
 	/* un-block server */
-	ret = read(pipefd, buf2, 4);
+	ret = read(unixfd, buf2, 4);
 	assert(ret == 4);
-	close(pipefd);
+	close(unixfd);
 
 	assert(strncmp(buf2, "xmit", 4) == 0);
 
@@ -662,7 +662,7 @@ static void connect_one_server(int fd, int pipefd)
 	close(fd);
 }
 
-static void process_one_client(int fd, int pipefd)
+static void process_one_client(int fd, int unixfd)
 {
 	ssize_t ret, ret2, ret3;
 	struct so_state s;
@@ -671,7 +671,7 @@ static void process_one_client(int fd, int pipefd)
 	memset(&s, 0, sizeof(s));
 	do_getsockopts(&s, fd, 0, 0);
 
-	ret = write(pipefd, "xmit", 4);
+	ret = write(unixfd, "xmit", 4);
 	assert(ret == 4);
 
 	ret = read(fd, buf, sizeof(buf));
@@ -736,7 +736,7 @@ static int xaccept(int s)
 	return fd;
 }
 
-static int server(int pipefd)
+static int server(int unixfd)
 {
 	int fd = -1, r;
 
@@ -752,13 +752,13 @@ static int server(int pipefd)
 		break;
 	}
 
-	r = write(pipefd, "conn", 4);
+	r = write(unixfd, "conn", 4);
 	assert(r == 4);
 
 	alarm(15);
 	r = xaccept(fd);
 
-	process_one_client(r, pipefd);
+	process_one_client(r, unixfd);
 
 	return 0;
 }
@@ -801,7 +801,7 @@ static void test_ip_tos_sockopt(int fd)
 		xerror("expect socklen_t == -1");
 }
 
-static int client(int pipefd)
+static int client(int unixfd)
 {
 	int fd = -1;
 
@@ -820,7 +820,7 @@ static int client(int pipefd)
 
 	test_ip_tos_sockopt(fd);
 
-	connect_one_server(fd, pipefd);
+	connect_one_server(fd, unixfd);
 
 	return 0;
 }
@@ -867,31 +867,31 @@ int main(int argc, char *argv[])
 {
 	int e1, e2, wstatus;
 	pid_t s, c, ret;
-	int pipefds[2];
+	int unixfds[2];
 
 	parse_opts(argc, argv);
 
 	init_rng();
 
-	e1 = pipe(pipefds);
+	e1 = socketpair(AF_UNIX, SOCK_DGRAM, 0, unixfds);
 	if (e1 < 0)
-		die_perror("pipe");
+		die_perror("socketpair");
 
 	s = xfork();
 	if (s == 0)
-		return server(pipefds[1]);
+		return server(unixfds[1]);
 
-	close(pipefds[1]);
+	close(unixfds[1]);
 
 	/* wait until server bound a socket */
-	e1 = read(pipefds[0], &e1, 4);
+	e1 = read(unixfds[0], &e1, 4);
 	assert(e1 == 4);
 
 	c = xfork();
 	if (c == 0)
-		return client(pipefds[0]);
+		return client(unixfds[0]);
 
-	close(pipefds[0]);
+	close(unixfds[0]);
 
 	ret = waitpid(s, &wstatus, 0);
 	if (ret == -1)
-- 
2.48.1

[PATCH mptcp-next v3 06/14] selftests: mptcp: sockopt: read eof at the end of process_one_client

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Moves EOF read handling to the end of process_one_client to ensure proper
synchronization after processing all data transfers, preparing for TCP_INQ
support in subsequent patches.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/mptcp_sockopt.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index add29bdda9e5..d69b10d2dbe3 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -687,14 +687,9 @@ static void process_one_client(int fd, int unixfd)
 	if (ret2 < 0)
 		die_perror("write");
 
-	/* wait for hangup */
-	ret3 = read(fd, buf, 1);
-	if (ret3 != 0)
-		xerror("expected EOF, got %lu", ret3);
-
 	do_getsockopts(&s, fd, ret, ret2);
-	if (s.mptcpi_rcv_delta && s.mptcpi_rcv_delta != (uint64_t)ret + 1)
-		xerror("mptcpi_rcv_delta %" PRIu64 ", expect %" PRIu64, s.mptcpi_rcv_delta, ret + 1, s.mptcpi_rcv_delta - ret);
+	if (s.mptcpi_rcv_delta && s.mptcpi_rcv_delta != (uint64_t)ret)
+		xerror("mptcpi_rcv_delta %" PRIu64 ", expect %" PRIu64, s.mptcpi_rcv_delta, ret, s.mptcpi_rcv_delta - ret);
 
 	/* be nice when running on top of older kernel */
 	if (s.pkt_stats_avail) {
@@ -712,6 +707,11 @@ static void process_one_client(int fd, int unixfd)
 			       s.last_sample.mptcpi_bytes_acked - ret2);
 	}
 
+	/* wait for hangup */
+	ret3 = read(fd, buf, 1);
+	if (ret3 != 0)
+		xerror("expected EOF, got %lu", ret3);
+
 	close(fd);
 }
 
-- 
2.48.1

[PATCH mptcp-next v3 07/14] selftests: mptcp: sockopt: use recvmsg instead of read

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Replaces read() with recvmsg() to enable control message (CMSG) handling,
preparing for TCP_INQ support in subsequent patches.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/net/mptcp/mptcp_sockopt.c       | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index d69b10d2dbe3..71cb311fcc50 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -665,8 +665,19 @@ static void connect_one_server(int fd, int unixfd)
 static void process_one_client(int fd, int unixfd)
 {
 	ssize_t ret, ret2, ret3;
+	char msg_buf[4096];
 	struct so_state s;
 	char buf[4096];
+	struct iovec iov = {
+		.iov_base = buf,
+		.iov_len = 1,
+	};
+	struct msghdr msg = {
+		.msg_iov = &iov,
+		.msg_iovlen = 1,
+		.msg_control = msg_buf,
+		.msg_controllen = sizeof(msg_buf),
+	};
 
 	memset(&s, 0, sizeof(s));
 	do_getsockopts(&s, fd, 0, 0);
@@ -674,9 +685,10 @@ static void process_one_client(int fd, int unixfd)
 	ret = write(unixfd, "xmit", 4);
 	assert(ret == 4);
 
-	ret = read(fd, buf, sizeof(buf));
+	iov.iov_len = sizeof(buf);
+	ret = recvmsg(fd, &msg, 0);
 	if (ret < 0)
-		die_perror("read");
+		die_perror("recvmsg");
 
 	assert(s.mptcpi_rcv_delta <= (uint64_t)ret);
 
@@ -708,7 +720,8 @@ static void process_one_client(int fd, int unixfd)
 	}
 
 	/* wait for hangup */
-	ret3 = read(fd, buf, 1);
+	iov.iov_len = 1;
+	ret3 = recvmsg(fd, &msg, 0);
 	if (ret3 != 0)
 		xerror("expected EOF, got %lu", ret3);
 
-- 
2.48.1

[PATCH mptcp-next v3 08/14] selftests: mptcp: sockopt: read one byte first

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Modifies data reading to first read one byte, then the remainder, ensuring
accurate TCP_INQ validation and data consistency checks.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/net/mptcp/mptcp_sockopt.c       | 28 ++++++++++++-------
 1 file changed, 18 insertions(+), 10 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index 71cb311fcc50..675ed2dcee9f 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -642,18 +642,18 @@ static void connect_one_server(int fd, int unixfd)
 		}
 
 		total += ret;
-	} while (total < len);
+	} while (total < len - 1);
 
-	if (total != len)
-		xerror("total %lu, len %lu eof %d\n", total, len, eof);
+	if (total != len - 1)
+		xerror("total %lu, len - 1 %lu eof %d\n", total, len - 1, eof);
 
-	if (memcmp(buf, buf2, len))
+	if (memcmp(buf + 1, buf2, len - 1))
 		xerror("data corruption");
 
 	if (s.tcpi_rcv_delta)
 		assert(s.tcpi_rcv_delta <= total);
 
-	do_getsockopts(&s, fd, ret, ret);
+	do_getsockopts(&s, fd, ret, ret + 1);
 
 	if (eof)
 		total += 1; /* sequence advances due to FIN */
@@ -685,6 +685,14 @@ static void process_one_client(int fd, int unixfd)
 	ret = write(unixfd, "xmit", 4);
 	assert(ret == 4);
 
+	/* read one byte, expect cmsg to return expected - 1 */
+	ret = recvmsg(fd, &msg, 0);
+	if (ret < 0)
+		die_perror("recvmsg");
+
+	if (inq && msg.msg_controllen == 0)
+		xerror("msg_controllen is 0");
+
 	iov.iov_len = sizeof(buf);
 	ret = recvmsg(fd, &msg, 0);
 	if (ret < 0)
@@ -699,9 +707,9 @@ static void process_one_client(int fd, int unixfd)
 	if (ret2 < 0)
 		die_perror("write");
 
-	do_getsockopts(&s, fd, ret, ret2);
-	if (s.mptcpi_rcv_delta && s.mptcpi_rcv_delta != (uint64_t)ret)
-		xerror("mptcpi_rcv_delta %" PRIu64 ", expect %" PRIu64, s.mptcpi_rcv_delta, ret, s.mptcpi_rcv_delta - ret);
+	do_getsockopts(&s, fd, ret + 1, ret2);
+	if (s.mptcpi_rcv_delta && s.mptcpi_rcv_delta != (uint64_t)ret + 1)
+		xerror("mptcpi_rcv_delta %" PRIu64 ", expect %" PRIu64, s.mptcpi_rcv_delta, ret + 1, s.mptcpi_rcv_delta - ret);
 
 	/* be nice when running on top of older kernel */
 	if (s.pkt_stats_avail) {
@@ -709,9 +717,9 @@ static void process_one_client(int fd, int unixfd)
 			xerror("mptcpi_bytes_sent %" PRIu64 ", expect %" PRIu64,
 			       s.last_sample.mptcpi_bytes_sent, ret2,
 			       s.last_sample.mptcpi_bytes_sent - ret2);
-		if (s.last_sample.mptcpi_bytes_received != ret)
+		if (s.last_sample.mptcpi_bytes_received != ret + 1)
 			xerror("mptcpi_bytes_received %" PRIu64 ", expect %" PRIu64,
-			       s.last_sample.mptcpi_bytes_received, ret,
+			       s.last_sample.mptcpi_bytes_received, ret + 1,
 			       s.last_sample.mptcpi_bytes_received - ret);
 		if (s.last_sample.mptcpi_bytes_acked != ret)
 			xerror("mptcpi_bytes_acked %" PRIu64 ", expect %" PRIu64,
-- 
2.48.1

[PATCH mptcp-next v3 09/14] selftests: mptcp: sockopt: add large data transfer tests

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Introduces large data transfer tests (1-17 MB) to validate MPTCP socket
options under high-volume conditions, including proper EOF handling,
preparing for TCP_INQ support in subsequent patches.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/net/mptcp/mptcp_sockopt.c       | 90 ++++++++++++++++++-
 1 file changed, 88 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index 675ed2dcee9f..fc65157a41ee 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -620,10 +620,12 @@ static void connect_one_server(int fd, int unixfd)
 	/* un-block server */
 	ret = read(unixfd, buf2, 4);
 	assert(ret == 4);
-	close(unixfd);
 
 	assert(strncmp(buf2, "xmit", 4) == 0);
 
+	ret = write(unixfd, &len, sizeof(len));
+	assert(ret == (ssize_t)sizeof(len));
+
 	ret = write(fd, buf, len);
 	if (ret < 0)
 		die_perror("write");
@@ -659,14 +661,51 @@ static void connect_one_server(int fd, int unixfd)
 		total += 1; /* sequence advances due to FIN */
 
 	assert(s.mptcpi_rcv_delta ? s.mptcpi_rcv_delta == (uint64_t)total : true);
+
+	ret = read(unixfd, buf2, 4);
+	assert(strncmp(buf2, "huge", 4) == 0);
+
+	total = rand() % (16 * 1024 * 1024);
+	total += (1 * 1024 * 1024);
+
+	ret = write(unixfd, &total, sizeof(total));
+	assert(ret == (ssize_t)sizeof(total));
+
+	while (total > 0) {
+		if (total > sizeof(buf))
+			len = sizeof(buf);
+		else
+			len = total;
+
+		ret = write(fd, buf, len);
+		if (ret < 0)
+			die_perror("write");
+		total -= ret;
+
+		/* we don't have to care about buf content, only
+		 * number of total bytes sent
+		 */
+	}
+
+	ret = read(unixfd, buf2, 4);
+	assert(ret == 4);
+	assert(strncmp(buf2, "shut", 4) == 0);
+
+	ret = write(fd, buf, 1);
+	assert(ret == 1);
 	close(fd);
+	ret = write(unixfd, "closed", 6);
+	assert(ret == 6);
+
+	close(unixfd);
 }
 
 static void process_one_client(int fd, int unixfd)
 {
-	ssize_t ret, ret2, ret3;
+	ssize_t ret, ret2, ret3, tot;
 	char msg_buf[4096];
 	struct so_state s;
+	size_t expect_len;
 	char buf[4096];
 	struct iovec iov = {
 		.iov_base = buf,
@@ -678,6 +717,7 @@ static void process_one_client(int fd, int unixfd)
 		.msg_control = msg_buf,
 		.msg_controllen = sizeof(msg_buf),
 	};
+	char tmp[16];
 
 	memset(&s, 0, sizeof(s));
 	do_getsockopts(&s, fd, 0, 0);
@@ -685,6 +725,12 @@ static void process_one_client(int fd, int unixfd)
 	ret = write(unixfd, "xmit", 4);
 	assert(ret == 4);
 
+	ret = read(unixfd, &expect_len, sizeof(expect_len));
+	assert(ret == (ssize_t)sizeof(expect_len));
+
+	if (expect_len > sizeof(buf))
+		xerror("expect len %zu exceeds buffer size", expect_len);
+
 	/* read one byte, expect cmsg to return expected - 1 */
 	ret = recvmsg(fd, &msg, 0);
 	if (ret < 0)
@@ -703,6 +749,9 @@ static void process_one_client(int fd, int unixfd)
 	if (s.tcpi_rcv_delta)
 		assert(s.tcpi_rcv_delta == (uint64_t)ret);
 
+	/* should have gotten exact remainder of all pending data */
+	assert(ret == (ssize_t)expect_len - 1);
+
 	ret2 = write(fd, buf, ret);
 	if (ret2 < 0)
 		die_perror("write");
@@ -727,6 +776,43 @@ static void process_one_client(int fd, int unixfd)
 			       s.last_sample.mptcpi_bytes_acked - ret2);
 	}
 
+	/* request a large swath of data. */
+	ret = write(unixfd, "huge", 4);
+	assert(ret == 4);
+
+	ret = read(unixfd, &expect_len, sizeof(expect_len));
+	assert(ret == (ssize_t)sizeof(expect_len));
+
+	/* peer should send us a few mb of data */
+	if (expect_len <= sizeof(buf))
+		xerror("expect len %zu too small\n", expect_len);
+
+	tot = 0;
+	do {
+		iov.iov_len = sizeof(buf);
+		ret = recvmsg(fd, &msg, 0);
+		if (ret < 0)
+			die_perror("recvmsg");
+
+		tot += ret;
+	} while ((size_t)tot < expect_len);
+
+	ret = write(unixfd, "shut", 4);
+	assert(ret == 4);
+
+	/* wait for hangup. Should have received one more byte of data. */
+	ret = read(unixfd, tmp, sizeof(tmp));
+	assert(ret == 6);
+	assert(strncmp(tmp, "closed", 6) == 0);
+
+	sleep(1);
+
+	iov.iov_len = 1;
+	ret = recvmsg(fd, &msg, 0);
+	if (ret < 0)
+		die_perror("recvmsg");
+	assert(ret == 1);
+
 	/* wait for hangup */
 	iov.iov_len = 1;
 	ret3 = recvmsg(fd, &msg, 0);
-- 
2.48.1

[PATCH mptcp-next v3 10/14] selftests: mptcp: sockopt: add wait_for_ack

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Adds wait_for_ack() helper using TIOCOUTQ/SIOCOUTQNSD ioctls to synchronize
tests by waiting for data acknowledgment before proceeding.

These codes are from mptcp_inq.c.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/net/mptcp/mptcp_sockopt.c       | 42 ++++++++++++++++++-
 1 file changed, 41 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index fc65157a41ee..b8e8fec8979f 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -21,11 +21,13 @@
 #include <sys/types.h>
 #include <sys/wait.h>
 #include <sys/random.h>
+#include <sys/ioctl.h>
 
 #include <netdb.h>
 #include <netinet/in.h>
 
 #include <linux/tcp.h>
+#include <linux/sockios.h>
 
 static int pf = AF_INET;
 static int proto_tx = IPPROTO_MPTCP;
@@ -593,10 +595,43 @@ static void do_getsockopts(struct so_state *s, int fd, size_t r, size_t w)
 		do_getsockopt_mptcp_full_info(s, fd);
 }
 
+/* wait up to timeout milliseconds */
+static void wait_for_ack(int fd, int timeout, size_t total)
+{
+	int i;
+
+	for (i = 0; i < timeout; i++) {
+		int nsd, ret, queued = -1;
+		struct timespec req;
+
+		ret = ioctl(fd, TIOCOUTQ, &queued);
+		if (ret < 0)
+			die_perror("TIOCOUTQ");
+
+		ret = ioctl(fd, SIOCOUTQNSD, &nsd);
+		if (ret < 0)
+			die_perror("SIOCOUTQNSD");
+
+		if ((size_t)queued > total)
+			xerror("TIOCOUTQ %u, but only %zu expected\n", queued, total);
+		assert(nsd <= queued);
+
+		if (queued == 0)
+			return;
+
+		/* wait for peer to ack rx of all data */
+		req.tv_sec = 0;
+		req.tv_nsec = 1 * 1000 * 1000ul; /* 1ms */
+		nanosleep(&req, NULL);
+	}
+
+	xerror("still tx data queued after %u ms\n", timeout);
+}
+
 static void connect_one_server(int fd, int unixfd)
 {
 	char buf[4096], buf2[4096];
-	size_t len, i, total;
+	size_t len, i, total, sent;
 	struct so_state s;
 	bool eof = false;
 	ssize_t ret;
@@ -667,10 +702,13 @@ static void connect_one_server(int fd, int unixfd)
 
 	total = rand() % (16 * 1024 * 1024);
 	total += (1 * 1024 * 1024);
+	sent = total;
 
 	ret = write(unixfd, &total, sizeof(total));
 	assert(ret == (ssize_t)sizeof(total));
 
+	wait_for_ack(fd, 5000, len);
+
 	while (total > 0) {
 		if (total > sizeof(buf))
 			len = sizeof(buf);
@@ -691,6 +729,8 @@ static void connect_one_server(int fd, int unixfd)
 	assert(ret == 4);
 	assert(strncmp(buf2, "shut", 4) == 0);
 
+	wait_for_ack(fd, 5000, sent);
+
 	ret = write(fd, buf, 1);
 	assert(ret == 1);
 	close(fd);
-- 
2.48.1

[PATCH mptcp-next v3 11/14] selftests: mptcp: sockopt: set FIONREAD ioctl

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Implements FIONREAD ioctl checks to validate available data size, ensuring
synchronization between write and read operations with non-blocking waits.

These codes are from mptcp_inq.c.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/net/mptcp/mptcp_sockopt.c        | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index b8e8fec8979f..7cc4626b7bf3 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -771,6 +771,24 @@ static void process_one_client(int fd, int unixfd)
 	if (expect_len > sizeof(buf))
 		xerror("expect len %zu exceeds buffer size", expect_len);
 
+	for (;;) {
+		struct timespec req;
+		unsigned int queued;
+
+		ret = ioctl(fd, FIONREAD, &queued);
+		if (ret < 0)
+			die_perror("FIONREAD");
+		if (queued > expect_len)
+			xerror("FIONREAD returned %u, but only %zu expected\n",
+			       queued, expect_len);
+		if (queued == expect_len)
+			break;
+
+		req.tv_sec = 0;
+		req.tv_nsec = 1000 * 1000ul;
+		nanosleep(&req, NULL);
+	}
+
 	/* read one byte, expect cmsg to return expected - 1 */
 	ret = recvmsg(fd, &msg, 0);
 	if (ret < 0)
-- 
2.48.1

[PATCH mptcp-next v3 12/14] selftests: mptcp: sockopt: add check_tcp_inq helper

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Adds check_tcp_inq() helper to extract and verify TCP_CM_INQ values from
control messages, ensuring correct in-queue data tracking during transfers.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../selftests/net/mptcp/mptcp_sockopt.c       | 40 +++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
index 7cc4626b7bf3..fb6f76c4dc07 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
@@ -740,6 +740,35 @@ static void connect_one_server(int fd, int unixfd)
 	close(unixfd);
 }
 
+static void get_tcp_inq(struct msghdr *msgh, unsigned int *inqv)
+{
+	struct cmsghdr *cmsg;
+
+	for (cmsg = CMSG_FIRSTHDR(msgh); cmsg ; cmsg = CMSG_NXTHDR(msgh, cmsg)) {
+		if (cmsg->cmsg_level == IPPROTO_TCP && cmsg->cmsg_type == TCP_CM_INQ) {
+			memcpy(inqv, CMSG_DATA(cmsg), sizeof(*inqv));
+			return;
+		}
+	}
+
+	xerror("could not find TCP_CM_INQ cmsg type");
+}
+
+static void check_tcp_inq(struct msghdr *msgh, unsigned int check, bool equal)
+{
+	unsigned int tcp_inq;
+
+	if (!inq)
+		return;
+
+	get_tcp_inq(msgh, &tcp_inq);
+
+	if (equal)
+		assert(tcp_inq == check);
+	else
+		assert(tcp_inq <= check);
+}
+
 static void process_one_client(int fd, int unixfd)
 {
 	ssize_t ret, ret2, ret3, tot;
@@ -797,6 +826,8 @@ static void process_one_client(int fd, int unixfd)
 	if (inq && msg.msg_controllen == 0)
 		xerror("msg_controllen is 0");
 
+	check_tcp_inq(&msg, expect_len - 1, true);
+
 	iov.iov_len = sizeof(buf);
 	ret = recvmsg(fd, &msg, 0);
 	if (ret < 0)
@@ -810,6 +841,9 @@ static void process_one_client(int fd, int unixfd)
 	/* should have gotten exact remainder of all pending data */
 	assert(ret == (ssize_t)expect_len - 1);
 
+	/* should be 0, all drained */
+	check_tcp_inq(&msg, 0, true);
+
 	ret2 = write(fd, buf, ret);
 	if (ret2 < 0)
 		die_perror("write");
@@ -853,6 +887,8 @@ static void process_one_client(int fd, int unixfd)
 			die_perror("recvmsg");
 
 		tot += ret;
+
+		check_tcp_inq(&msg, expect_len - tot, false);
 	} while ((size_t)tot < expect_len);
 
 	ret = write(unixfd, "shut", 4);
@@ -871,11 +907,15 @@ static void process_one_client(int fd, int unixfd)
 		die_perror("recvmsg");
 	assert(ret == 1);
 
+	/* tcp_inq should be 1 due to received fin. */
+	check_tcp_inq(&msg, 1, true);
+
 	/* wait for hangup */
 	iov.iov_len = 1;
 	ret3 = recvmsg(fd, &msg, 0);
 	if (ret3 != 0)
 		xerror("expected EOF, got %lu", ret3);
+	check_tcp_inq(&msg, 1, true);
 
 	close(fd);
 }
-- 
2.48.1

[PATCH mptcp-next v3 13/14] selftests: mptcp: sockopt: replace mptcp_inq with sockopt

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Update mptcp_sockopt.sh to use mptcp_sockopt binary instead of
mptcp_inq. This consolidates TCP_INQ testing into a single test binary
while maintaining equivalent test coverage.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/mptcp_sockopt.sh | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh b/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh
index 418a903c3a4d..f5150a10b064 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh
@@ -305,7 +305,7 @@ run_tests()
 do_tcpinq_test()
 {
 	print_title "TCP_INQ cmsg/ioctl $*"
-	ip netns exec "$ns_sbox" ./mptcp_inq "$@"
+	ip netns exec "$ns_sbox" ./mptcp_sockopt "$@"
 	local lret=$?
 	if [ $lret -ne 0 ];then
 		ret=$lret
@@ -331,19 +331,19 @@ do_tcpinq_tests()
 
 	local args
 	for args in "-t tcp" "-r tcp"; do
-		do_tcpinq_test $args
+		do_tcpinq_test $args -i
 		lret=$?
 		if [ $lret -ne 0 ] ; then
 			return $lret
 		fi
-		do_tcpinq_test -6 $args
+		do_tcpinq_test -6 $args -i
 		lret=$?
 		if [ $lret -ne 0 ] ; then
 			return $lret
 		fi
 	done
 
-	do_tcpinq_test -r tcp -t tcp
+	do_tcpinq_test -r tcp -t tcp -i
 
 	return $?
 }
-- 
2.48.1

[PATCH mptcp-next v3 14/14] selftests: mptcp: sockopt: remove obsolete mptcp_inq

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Delete mptcp_inq binary and references since its functionality has been
fully integrated into mptcp_sockopt. This simplifies the test suite
structure and reduces maintenance overhead.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/.gitignore  |   1 -
 tools/testing/selftests/net/mptcp/Makefile    |   2 +-
 tools/testing/selftests/net/mptcp/mptcp_inq.c | 608 ------------------
 3 files changed, 1 insertion(+), 610 deletions(-)
 delete mode 100644 tools/testing/selftests/net/mptcp/mptcp_inq.c

diff --git a/tools/testing/selftests/net/mptcp/.gitignore b/tools/testing/selftests/net/mptcp/.gitignore
index 833279fb34e2..c62e91ce385d 100644
--- a/tools/testing/selftests/net/mptcp/.gitignore
+++ b/tools/testing/selftests/net/mptcp/.gitignore
@@ -1,7 +1,6 @@
 # SPDX-License-Identifier: GPL-2.0-only
 mptcp_connect
 mptcp_diag
-mptcp_inq
 mptcp_sockopt
 pm_nl_ctl
 *.pcap
diff --git a/tools/testing/selftests/net/mptcp/Makefile b/tools/testing/selftests/net/mptcp/Makefile
index 4c7e51336ab2..9c8da49313fb 100644
--- a/tools/testing/selftests/net/mptcp/Makefile
+++ b/tools/testing/selftests/net/mptcp/Makefile
@@ -8,7 +8,7 @@ TEST_PROGS := mptcp_connect.sh mptcp_connect_mmap.sh mptcp_connect_sendfile.sh \
 	      mptcp_connect_checksum.sh pm_netlink.sh mptcp_join.sh diag.sh \
 	      simult_flows.sh mptcp_sockopt.sh userspace_pm.sh
 
-TEST_GEN_FILES = mptcp_connect pm_nl_ctl mptcp_sockopt mptcp_inq mptcp_diag
+TEST_GEN_FILES = mptcp_connect pm_nl_ctl mptcp_sockopt mptcp_diag
 
 TEST_FILES := mptcp_lib.sh settings
 
diff --git a/tools/testing/selftests/net/mptcp/mptcp_inq.c b/tools/testing/selftests/net/mptcp/mptcp_inq.c
deleted file mode 100644
index f3bcaa48df8f..000000000000
--- a/tools/testing/selftests/net/mptcp/mptcp_inq.c
+++ /dev/null
@@ -1,608 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-
-#define _GNU_SOURCE
-
-#include <assert.h>
-#include <errno.h>
-#include <fcntl.h>
-#include <limits.h>
-#include <string.h>
-#include <stdarg.h>
-#include <stdbool.h>
-#include <stdint.h>
-#include <inttypes.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <strings.h>
-#include <unistd.h>
-#include <time.h>
-
-#include <sys/ioctl.h>
-#include <sys/random.h>
-#include <sys/socket.h>
-#include <sys/types.h>
-#include <sys/wait.h>
-
-#include <netdb.h>
-#include <netinet/in.h>
-
-#include <linux/tcp.h>
-#include <linux/sockios.h>
-
-#ifndef IPPROTO_MPTCP
-#define IPPROTO_MPTCP 262
-#endif
-#ifndef SOL_MPTCP
-#define SOL_MPTCP 284
-#endif
-
-static int pf = AF_INET;
-static int proto_tx = IPPROTO_MPTCP;
-static int proto_rx = IPPROTO_MPTCP;
-
-static void die_perror(const char *msg)
-{
-	perror(msg);
-	exit(1);
-}
-
-static void die_usage(int r)
-{
-	fprintf(stderr, "Usage: mptcp_inq [-6] [ -t tcp|mptcp ] [ -r tcp|mptcp]\n");
-	exit(r);
-}
-
-static void xerror(const char *fmt, ...)
-{
-	va_list ap;
-
-	va_start(ap, fmt);
-	vfprintf(stderr, fmt, ap);
-	va_end(ap);
-	fputc('\n', stderr);
-	exit(1);
-}
-
-static const char *getxinfo_strerr(int err)
-{
-	if (err == EAI_SYSTEM)
-		return strerror(errno);
-
-	return gai_strerror(err);
-}
-
-static void xgetaddrinfo(const char *node, const char *service,
-			 struct addrinfo *hints,
-			 struct addrinfo **res)
-{
-	int err;
-
-again:
-	err = getaddrinfo(node, service, hints, res);
-	if (err) {
-		const char *errstr;
-
-		if (err == EAI_SOCKTYPE) {
-			hints->ai_protocol = IPPROTO_TCP;
-			goto again;
-		}
-
-		errstr = getxinfo_strerr(err);
-
-		fprintf(stderr, "Fatal: getaddrinfo(%s:%s): %s\n",
-			node ? node : "", service ? service : "", errstr);
-		exit(1);
-	}
-}
-
-static int sock_listen_mptcp(const char * const listenaddr,
-			     const char * const port)
-{
-	int sock = -1;
-	struct addrinfo hints = {
-		.ai_protocol = IPPROTO_MPTCP,
-		.ai_socktype = SOCK_STREAM,
-		.ai_flags = AI_PASSIVE | AI_NUMERICHOST
-	};
-
-	hints.ai_family = pf;
-
-	struct addrinfo *a, *addr;
-	int one = 1;
-
-	xgetaddrinfo(listenaddr, port, &hints, &addr);
-	hints.ai_family = pf;
-
-	for (a = addr; a; a = a->ai_next) {
-		sock = socket(a->ai_family, a->ai_socktype, proto_rx);
-		if (sock < 0)
-			continue;
-
-		if (-1 == setsockopt(sock, SOL_SOCKET, SO_REUSEADDR, &one,
-				     sizeof(one)))
-			perror("setsockopt");
-
-		if (bind(sock, a->ai_addr, a->ai_addrlen) == 0)
-			break; /* success */
-
-		perror("bind");
-		close(sock);
-		sock = -1;
-	}
-
-	freeaddrinfo(addr);
-
-	if (sock < 0)
-		xerror("could not create listen socket");
-
-	if (listen(sock, 20))
-		die_perror("listen");
-
-	return sock;
-}
-
-static int sock_connect_mptcp(const char * const remoteaddr,
-			      const char * const port, int proto)
-{
-	struct addrinfo hints = {
-		.ai_protocol = IPPROTO_MPTCP,
-		.ai_socktype = SOCK_STREAM,
-	};
-	struct addrinfo *a, *addr;
-	int sock = -1;
-
-	hints.ai_family = pf;
-
-	xgetaddrinfo(remoteaddr, port, &hints, &addr);
-	for (a = addr; a; a = a->ai_next) {
-		sock = socket(a->ai_family, a->ai_socktype, proto);
-		if (sock < 0)
-			continue;
-
-		if (connect(sock, a->ai_addr, a->ai_addrlen) == 0)
-			break; /* success */
-
-		die_perror("connect");
-	}
-
-	if (sock < 0)
-		xerror("could not create connect socket");
-
-	freeaddrinfo(addr);
-	return sock;
-}
-
-static int protostr_to_num(const char *s)
-{
-	if (strcasecmp(s, "tcp") == 0)
-		return IPPROTO_TCP;
-	if (strcasecmp(s, "mptcp") == 0)
-		return IPPROTO_MPTCP;
-
-	die_usage(1);
-	return 0;
-}
-
-static void parse_opts(int argc, char **argv)
-{
-	int c;
-
-	while ((c = getopt(argc, argv, "h6t:r:")) != -1) {
-		switch (c) {
-		case 'h':
-			die_usage(0);
-			break;
-		case '6':
-			pf = AF_INET6;
-			break;
-		case 't':
-			proto_tx = protostr_to_num(optarg);
-			break;
-		case 'r':
-			proto_rx = protostr_to_num(optarg);
-			break;
-		default:
-			die_usage(1);
-			break;
-		}
-	}
-}
-
-/* wait up to timeout milliseconds */
-static void wait_for_ack(int fd, int timeout, size_t total)
-{
-	int i;
-
-	for (i = 0; i < timeout; i++) {
-		int nsd, ret, queued = -1;
-		struct timespec req;
-
-		ret = ioctl(fd, TIOCOUTQ, &queued);
-		if (ret < 0)
-			die_perror("TIOCOUTQ");
-
-		ret = ioctl(fd, SIOCOUTQNSD, &nsd);
-		if (ret < 0)
-			die_perror("SIOCOUTQNSD");
-
-		if ((size_t)queued > total)
-			xerror("TIOCOUTQ %u, but only %zu expected\n", queued, total);
-		assert(nsd <= queued);
-
-		if (queued == 0)
-			return;
-
-		/* wait for peer to ack rx of all data */
-		req.tv_sec = 0;
-		req.tv_nsec = 1 * 1000 * 1000ul; /* 1ms */
-		nanosleep(&req, NULL);
-	}
-
-	xerror("still tx data queued after %u ms\n", timeout);
-}
-
-static void connect_one_server(int fd, int unixfd)
-{
-	size_t len, i, total, sent;
-	char buf[4096], buf2[4096];
-	ssize_t ret;
-
-	len = rand() % (sizeof(buf) - 1);
-
-	if (len < 128)
-		len = 128;
-
-	for (i = 0; i < len ; i++) {
-		buf[i] = rand() % 26;
-		buf[i] += 'A';
-	}
-
-	buf[i] = '\n';
-
-	/* un-block server */
-	ret = read(unixfd, buf2, 4);
-	assert(ret == 4);
-
-	assert(strncmp(buf2, "xmit", 4) == 0);
-
-	ret = write(unixfd, &len, sizeof(len));
-	assert(ret == (ssize_t)sizeof(len));
-
-	ret = write(fd, buf, len);
-	if (ret < 0)
-		die_perror("write");
-
-	if (ret != (ssize_t)len)
-		xerror("short write");
-
-	ret = read(unixfd, buf2, 4);
-	assert(strncmp(buf2, "huge", 4) == 0);
-
-	total = rand() % (16 * 1024 * 1024);
-	total += (1 * 1024 * 1024);
-	sent = total;
-
-	ret = write(unixfd, &total, sizeof(total));
-	assert(ret == (ssize_t)sizeof(total));
-
-	wait_for_ack(fd, 5000, len);
-
-	while (total > 0) {
-		if (total > sizeof(buf))
-			len = sizeof(buf);
-		else
-			len = total;
-
-		ret = write(fd, buf, len);
-		if (ret < 0)
-			die_perror("write");
-		total -= ret;
-
-		/* we don't have to care about buf content, only
-		 * number of total bytes sent
-		 */
-	}
-
-	ret = read(unixfd, buf2, 4);
-	assert(ret == 4);
-	assert(strncmp(buf2, "shut", 4) == 0);
-
-	wait_for_ack(fd, 5000, sent);
-
-	ret = write(fd, buf, 1);
-	assert(ret == 1);
-	close(fd);
-	ret = write(unixfd, "closed", 6);
-	assert(ret == 6);
-
-	close(unixfd);
-}
-
-static void get_tcp_inq(struct msghdr *msgh, unsigned int *inqv)
-{
-	struct cmsghdr *cmsg;
-
-	for (cmsg = CMSG_FIRSTHDR(msgh); cmsg ; cmsg = CMSG_NXTHDR(msgh, cmsg)) {
-		if (cmsg->cmsg_level == IPPROTO_TCP && cmsg->cmsg_type == TCP_CM_INQ) {
-			memcpy(inqv, CMSG_DATA(cmsg), sizeof(*inqv));
-			return;
-		}
-	}
-
-	xerror("could not find TCP_CM_INQ cmsg type");
-}
-
-static void process_one_client(int fd, int unixfd)
-{
-	unsigned int tcp_inq;
-	size_t expect_len;
-	char msg_buf[4096];
-	char buf[4096];
-	char tmp[16];
-	struct iovec iov = {
-		.iov_base = buf,
-		.iov_len = 1,
-	};
-	struct msghdr msg = {
-		.msg_iov = &iov,
-		.msg_iovlen = 1,
-		.msg_control = msg_buf,
-		.msg_controllen = sizeof(msg_buf),
-	};
-	ssize_t ret, tot;
-
-	ret = write(unixfd, "xmit", 4);
-	assert(ret == 4);
-
-	ret = read(unixfd, &expect_len, sizeof(expect_len));
-	assert(ret == (ssize_t)sizeof(expect_len));
-
-	if (expect_len > sizeof(buf))
-		xerror("expect len %zu exceeds buffer size", expect_len);
-
-	for (;;) {
-		struct timespec req;
-		unsigned int queued;
-
-		ret = ioctl(fd, FIONREAD, &queued);
-		if (ret < 0)
-			die_perror("FIONREAD");
-		if (queued > expect_len)
-			xerror("FIONREAD returned %u, but only %zu expected\n",
-			       queued, expect_len);
-		if (queued == expect_len)
-			break;
-
-		req.tv_sec = 0;
-		req.tv_nsec = 1000 * 1000ul;
-		nanosleep(&req, NULL);
-	}
-
-	/* read one byte, expect cmsg to return expected - 1 */
-	ret = recvmsg(fd, &msg, 0);
-	if (ret < 0)
-		die_perror("recvmsg");
-
-	if (msg.msg_controllen == 0)
-		xerror("msg_controllen is 0");
-
-	get_tcp_inq(&msg, &tcp_inq);
-
-	assert((size_t)tcp_inq == (expect_len - 1));
-
-	iov.iov_len = sizeof(buf);
-	ret = recvmsg(fd, &msg, 0);
-	if (ret < 0)
-		die_perror("recvmsg");
-
-	/* should have gotten exact remainder of all pending data */
-	assert(ret == (ssize_t)tcp_inq);
-
-	/* should be 0, all drained */
-	get_tcp_inq(&msg, &tcp_inq);
-	assert(tcp_inq == 0);
-
-	/* request a large swath of data. */
-	ret = write(unixfd, "huge", 4);
-	assert(ret == 4);
-
-	ret = read(unixfd, &expect_len, sizeof(expect_len));
-	assert(ret == (ssize_t)sizeof(expect_len));
-
-	/* peer should send us a few mb of data */
-	if (expect_len <= sizeof(buf))
-		xerror("expect len %zu too small\n", expect_len);
-
-	tot = 0;
-	do {
-		iov.iov_len = sizeof(buf);
-		ret = recvmsg(fd, &msg, 0);
-		if (ret < 0)
-			die_perror("recvmsg");
-
-		tot += ret;
-
-		get_tcp_inq(&msg, &tcp_inq);
-
-		if (tcp_inq > expect_len - tot)
-			xerror("inq %d, remaining %d total_len %d\n",
-			       tcp_inq, expect_len - tot, (int)expect_len);
-
-		assert(tcp_inq <= expect_len - tot);
-	} while ((size_t)tot < expect_len);
-
-	ret = write(unixfd, "shut", 4);
-	assert(ret == 4);
-
-	/* wait for hangup. Should have received one more byte of data. */
-	ret = read(unixfd, tmp, sizeof(tmp));
-	assert(ret == 6);
-	assert(strncmp(tmp, "closed", 6) == 0);
-
-	sleep(1);
-
-	iov.iov_len = 1;
-	ret = recvmsg(fd, &msg, 0);
-	if (ret < 0)
-		die_perror("recvmsg");
-	assert(ret == 1);
-
-	get_tcp_inq(&msg, &tcp_inq);
-
-	/* tcp_inq should be 1 due to received fin. */
-	assert(tcp_inq == 1);
-
-	iov.iov_len = 1;
-	ret = recvmsg(fd, &msg, 0);
-	if (ret < 0)
-		die_perror("recvmsg");
-
-	/* expect EOF */
-	assert(ret == 0);
-	get_tcp_inq(&msg, &tcp_inq);
-	assert(tcp_inq == 1);
-
-	close(fd);
-}
-
-static int xaccept(int s)
-{
-	int fd = accept(s, NULL, 0);
-
-	if (fd < 0)
-		die_perror("accept");
-
-	return fd;
-}
-
-static int server(int unixfd)
-{
-	int fd = -1, r, on = 1;
-
-	switch (pf) {
-	case AF_INET:
-		fd = sock_listen_mptcp("127.0.0.1", "15432");
-		break;
-	case AF_INET6:
-		fd = sock_listen_mptcp("::1", "15432");
-		break;
-	default:
-		xerror("Unknown pf %d\n", pf);
-		break;
-	}
-
-	r = write(unixfd, "conn", 4);
-	assert(r == 4);
-
-	alarm(15);
-	r = xaccept(fd);
-
-	if (-1 == setsockopt(r, IPPROTO_TCP, TCP_INQ, &on, sizeof(on)))
-		die_perror("setsockopt");
-
-	process_one_client(r, unixfd);
-
-	return 0;
-}
-
-static int client(int unixfd)
-{
-	int fd = -1;
-
-	alarm(15);
-
-	switch (pf) {
-	case AF_INET:
-		fd = sock_connect_mptcp("127.0.0.1", "15432", proto_tx);
-		break;
-	case AF_INET6:
-		fd = sock_connect_mptcp("::1", "15432", proto_tx);
-		break;
-	default:
-		xerror("Unknown pf %d\n", pf);
-	}
-
-	connect_one_server(fd, unixfd);
-
-	return 0;
-}
-
-static void init_rng(void)
-{
-	unsigned int foo;
-
-	if (getrandom(&foo, sizeof(foo), 0) == -1) {
-		perror("getrandom");
-		exit(1);
-	}
-
-	srand(foo);
-}
-
-static pid_t xfork(void)
-{
-	pid_t p = fork();
-
-	if (p < 0)
-		die_perror("fork");
-	else if (p == 0)
-		init_rng();
-
-	return p;
-}
-
-static int rcheck(int wstatus, const char *what)
-{
-	if (WIFEXITED(wstatus)) {
-		if (WEXITSTATUS(wstatus) == 0)
-			return 0;
-		fprintf(stderr, "%s exited, status=%d\n", what, WEXITSTATUS(wstatus));
-		return WEXITSTATUS(wstatus);
-	} else if (WIFSIGNALED(wstatus)) {
-		xerror("%s killed by signal %d\n", what, WTERMSIG(wstatus));
-	} else if (WIFSTOPPED(wstatus)) {
-		xerror("%s stopped by signal %d\n", what, WSTOPSIG(wstatus));
-	}
-
-	return 111;
-}
-
-int main(int argc, char *argv[])
-{
-	int e1, e2, wstatus;
-	pid_t s, c, ret;
-	int unixfds[2];
-
-	parse_opts(argc, argv);
-
-	e1 = socketpair(AF_UNIX, SOCK_DGRAM, 0, unixfds);
-	if (e1 < 0)
-		die_perror("pipe");
-
-	s = xfork();
-	if (s == 0)
-		return server(unixfds[1]);
-
-	close(unixfds[1]);
-
-	/* wait until server bound a socket */
-	e1 = read(unixfds[0], &e1, 4);
-	assert(e1 == 4);
-
-	c = xfork();
-	if (c == 0)
-		return client(unixfds[0]);
-
-	close(unixfds[0]);
-
-	ret = waitpid(s, &wstatus, 0);
-	if (ret == -1)
-		die_perror("waitpid");
-	e1 = rcheck(wstatus, "server");
-	ret = waitpid(c, &wstatus, 0);
-	if (ret == -1)
-		die_perror("waitpid");
-	e2 = rcheck(wstatus, "client");
-
-	return e1 ? e1 : e2;
-}
-- 
2.48.1

Re: [PATCH mptcp-next v3 00/14] selftests: consolidate TCP_INQ testing into sockopt

[MPTCP CI]

Hi Geliang,

Thank you for your modifications, that's great!

Our CI did some validations and here is its report:

- KVM Validation: normal: Unstable: 1 failed test(s): selftest_mptcp_connect 🔴
- KVM Validation: debug: Unstable: 1 failed test(s): selftest_mptcp_connect 🔴
- KVM Validation: btf-normal (only bpftest_all): Success! ✅
- KVM Validation: btf-debug (only bpftest_all): Success! ✅
- Task: https://github.com/multipath-tcp/mptcp_net-next/actions/runs/17234297108

Initiator: Patchew Applier
Commits: https://github.com/multipath-tcp/mptcp_net-next/commits/2b989c1de52d
Patchwork: https://patchwork.kernel.org/project/mptcp/list/?series=995595


If there are some issues, you can reproduce them using the same environment as
the one used by the CI thanks to a docker image, e.g.:

    $ cd [kernel source code]
    $ docker run -v "${PWD}:${PWD}:rw" -w "${PWD}" --privileged --rm -it \
        --pull always mptcp/mptcp-upstream-virtme-docker:latest \
        auto-normal

For more details:

    https://github.com/multipath-tcp/mptcp-upstream-virtme-docker


Please note that despite all the efforts that have been already done to have a
stable tests suite when executed on a public CI like here, it is possible some
reported issues are not due to your modifications. Still, do not hesitate to
help us improve that ;-)

Cheers,
MPTCP GH Action bot
Bot operated by Matthieu Baerts (NGI0 Core)