[LTP] [PATCH v3] fanotify22.c: handle multiple asynchronous error events

[Wei Gao]

Since the introduction of the asynchronous fserror reporting framework
(kernel commit 81d2e13a57c9), fanotify22 has encountered sporadic failures
due to the non-deterministic nature of event delivery and merging:

1) tcase3 failure: A race condition occurs when the test reads the
   notification fd between two events. uses a poll() and read() loop to wait
   until the expected.

2) tcase4 failure: The kernel may deliver errors as independent events
   instead of a single merged event, since different worker kthread can
   end up generating each event so they won't be merged. As suggested by
   Jan Kara, this patch introduces a consolidate_events() helper. It iterates
   through the event buffer, accumulates the error_count from all independent
   events, and updates the first event's count in-place.

v2->v3:
- Replaced usleep() with poll() loop
- Added secondary error check support (error2)

Reported-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/oe-lkp/202602042124.87bd00e3-lkp@intel.com
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Wei Gao <wegao@suse.com>
---
 .../kernel/syscalls/fanotify/fanotify22.c     | 87 ++++++++++++++++++-
 1 file changed, 83 insertions(+), 4 deletions(-)

diff --git a/testcases/kernel/syscalls/fanotify/fanotify22.c b/testcases/kernel/syscalls/fanotify/fanotify22.c
index e8002b160..c921cf2c0 100644
--- a/testcases/kernel/syscalls/fanotify/fanotify22.c
+++ b/testcases/kernel/syscalls/fanotify/fanotify22.c
@@ -28,6 +28,7 @@
 #include "tst_test.h"
 #include <sys/fanotify.h>
 #include <sys/types.h>
+#include <poll.h>
 
 #ifdef HAVE_SYS_FANOTIFY_H
 #include "fanotify.h"
@@ -88,7 +89,6 @@ static void trigger_bad_link_lookup(void)
 			ret, BAD_LINK, errno, EUCLEAN);
 }
 
-
 static void tcase3_trigger(void)
 {
 	trigger_bad_link_lookup();
@@ -104,6 +104,7 @@ static void tcase4_trigger(void)
 static struct test_case {
 	char *name;
 	int error;
+	int error2;
 	unsigned int error_count;
 	struct fanotify_fid_t *fid;
 	void (*trigger_error)(void);
@@ -134,10 +135,54 @@ static struct test_case {
 		.trigger_error = &tcase4_trigger,
 		.error_count = 2,
 		.error = EFSCORRUPTED,
+		.error2 = ESHUTDOWN,
 		.fid = &bad_file_fid,
 	}
 };
 
+static size_t consolidate_events(char *buf, size_t len, const struct test_case *ex)
+{
+	struct fanotify_event_metadata *metadata, *first = NULL;
+	struct fanotify_event_info_error *first_info = NULL;
+	unsigned int total_count = 0;
+	int event_num = 0;
+
+	for (metadata = (struct fanotify_event_metadata *)buf;
+			FAN_EVENT_OK(metadata, len);
+			metadata = FAN_EVENT_NEXT(metadata, len)) {
+
+		event_num++;
+		struct fanotify_event_info_error *info = get_event_info_error(metadata);
+
+		if (!info) {
+			tst_res(TFAIL, "%s: Event [%d] missing error info",
+					ex->name, event_num);
+			continue;
+		}
+
+		if (info->error != ex->error && (ex->error2 == 0 || info->error != ex->error2)) {
+			tst_res(TFAIL, "%s: Event [%d] unexpected errno (%d)",
+					ex->name, event_num, info->error);
+		}
+
+		if (info) {
+			if (!first) {
+				first = metadata;
+				first_info = info;
+			}
+			total_count += info->error_count;
+
+			tst_res(TINFO, "Event [%d]: errno=%d, error_count=%d",
+					event_num, info->error, info->error_count);
+		}
+	}
+
+	if (first_info)
+		first_info->error_count = total_count;
+
+	return (first) ? first->event_len : 0;
+}
+
 static int check_error_event_info_fid(struct fanotify_event_info_fid *fid,
 				 const struct test_case *ex)
 {
@@ -248,19 +293,53 @@ static void check_event(char *buf, size_t len, const struct test_case *ex)
 static void do_test(unsigned int i)
 {
 	const struct test_case *tcase = &testcases[i];
-	size_t read_len;
+	size_t read_len = 0;
+	struct pollfd pfd;
+	unsigned int accumulated_count = 0;
 
 	SAFE_FANOTIFY_MARK(fd_notify, FAN_MARK_ADD|FAN_MARK_FILESYSTEM,
 			   FAN_FS_ERROR, AT_FDCWD, MOUNT_PATH);
 
 	tcase->trigger_error();
 
-	read_len = SAFE_READ(0, fd_notify, event_buf, BUF_SIZE);
+	pfd.fd = fd_notify;
+	pfd.events = POLLIN;
+
+	while (accumulated_count < tcase->error_count) {
+		if (poll(&pfd, 1, 5000) <= 0) {
+			tst_res(TFAIL, "%s: Timeout waiting for events", tcase->name);
+			goto out;
+		}
+
+		char *current_pos = event_buf + read_len;
+		int ret = read(fd_notify, current_pos, BUF_SIZE - read_len);
+
+		if (ret < 0) {
+			tst_res(TFAIL, "%s: read failed: %s", tcase->name, strerror(errno));
+			goto out;
+		}
+
+		struct fanotify_event_metadata *m =
+			(struct fanotify_event_metadata *)current_pos;
+		while (FAN_EVENT_OK(m, ret)) {
+			read_len += ret;
+			struct fanotify_event_info_error *e = get_event_info_error(m);
+
+			if (e)
+				accumulated_count += e->error_count;
+			m = FAN_EVENT_NEXT(m, ret);
+		}
+	}
+
+	read_len = consolidate_events(event_buf, read_len, tcase);
+
+	check_event(event_buf, read_len, tcase);
+
+out:
 
 	SAFE_FANOTIFY_MARK(fd_notify, FAN_MARK_REMOVE|FAN_MARK_FILESYSTEM,
 			   FAN_FS_ERROR, AT_FDCWD, MOUNT_PATH);
 
-	check_event(event_buf, read_len, tcase);
 	/* Unmount and mount the filesystem to get it out of the error state */
 	SAFE_UMOUNT(MOUNT_PATH);
 	SAFE_MOUNT(tst_device->dev, MOUNT_PATH, tst_device->fs_type, 0, NULL);
-- 
2.52.0