[PATCH v2 bpf-next] selftests/bpf: make tcp_rtt test more robust to failures

[PATCH v2 bpf-next] selftests/bpf: make tcp_rtt test more robust to failures

[Andrii Nakryiko]

Switch to non-blocking accept and wait for server thread to exit before
proceeding. I noticed that sometimes tcp_rtt server thread failure would
"spill over" into other tests (that would run after tcp_rtt), probably just
because server thread exits much later and tcp_rtt doesn't wait for it.

v1->v2:
  - add usleep() while waiting on initial non-blocking accept() (Stanislav);

Fixes: 8a03222f508b ("selftests/bpf: test_progs: fix client/server race in tcp_rtt")
Reviewed-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
---
 .../selftests/bpf/prog_tests/tcp_rtt.c        | 32 ++++++++++++-------
 1 file changed, 20 insertions(+), 12 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/tcp_rtt.c b/tools/testing/selftests/bpf/prog_tests/tcp_rtt.c
index f4cd60d6fba2..e08f6bb17700 100644
--- a/tools/testing/selftests/bpf/prog_tests/tcp_rtt.c
+++ b/tools/testing/selftests/bpf/prog_tests/tcp_rtt.c
@@ -188,7 +188,7 @@ static int start_server(void)
 	};
 	int fd;
 
-	fd = socket(AF_INET, SOCK_STREAM, 0);
+	fd = socket(AF_INET, SOCK_STREAM | SOCK_NONBLOCK, 0);
 	if (fd < 0) {
 		log_err("Failed to create server socket");
 		return -1;
@@ -205,6 +205,7 @@ static int start_server(void)
 
 static pthread_mutex_t server_started_mtx = PTHREAD_MUTEX_INITIALIZER;
 static pthread_cond_t server_started = PTHREAD_COND_INITIALIZER;
+static volatile bool server_done = false;
 
 static void *server_thread(void *arg)
 {
@@ -222,23 +223,24 @@ static void *server_thread(void *arg)
 
 	if (CHECK_FAIL(err < 0)) {
 		perror("Failed to listed on socket");
-		return NULL;
+		return ERR_PTR(err);
 	}
 
-	client_fd = accept(fd, (struct sockaddr *)&addr, &len);
+	while (!server_done) {
+		client_fd = accept(fd, (struct sockaddr *)&addr, &len);
+		if (client_fd == -1 && errno == EAGAIN) {
+			usleep(50);
+			continue;
+		}
+		break;
+	}
 	if (CHECK_FAIL(client_fd < 0)) {
 		perror("Failed to accept client");
-		return NULL;
+		return ERR_PTR(err);
 	}
 
-	/* Wait for the next connection (that never arrives)
-	 * to keep this thread alive to prevent calling
-	 * close() on client_fd.
-	 */
-	if (CHECK_FAIL(accept(fd, (struct sockaddr *)&addr, &len) >= 0)) {
-		perror("Unexpected success in second accept");
-		return NULL;
-	}
+	while (!server_done)
+		usleep(50);
 
 	close(client_fd);
 
@@ -249,6 +251,7 @@ void test_tcp_rtt(void)
 {
 	int server_fd, cgroup_fd;
 	pthread_t tid;
+	void *server_res;
 
 	cgroup_fd = test__join_cgroup("/tcp_rtt");
 	if (CHECK_FAIL(cgroup_fd < 0))
@@ -267,6 +270,11 @@ void test_tcp_rtt(void)
 	pthread_mutex_unlock(&server_started_mtx);
 
 	CHECK_FAIL(run_test(cgroup_fd, server_fd));
+
+	server_done = true;
+	pthread_join(tid, &server_res);
+	CHECK_FAIL(IS_ERR(server_res));
+
 close_server_fd:
 	close(server_fd);
 close_cgroup_fd:
-- 
2.17.1

Re: [PATCH v2 bpf-next] selftests/bpf: make tcp_rtt test more robust to failures

[Alexei Starovoitov]

On Wed, Mar 11, 2020 at 3:28 PM Andrii Nakryiko <andriin@fb.com> wrote:
>
> Switch to non-blocking accept and wait for server thread to exit before
> proceeding. I noticed that sometimes tcp_rtt server thread failure would
> "spill over" into other tests (that would run after tcp_rtt), probably just
> because server thread exits much later and tcp_rtt doesn't wait for it.
>
> v1->v2:
>   - add usleep() while waiting on initial non-blocking accept() (Stanislav);
>
> Fixes: 8a03222f508b ("selftests/bpf: test_progs: fix client/server race in tcp_rtt")
> Reviewed-by: Stanislav Fomichev <sdf@google.com>
> Signed-off-by: Andrii Nakryiko <andriin@fb.com>

Applied. Thanks