diff mbox series

[for-6.1,2/4] io_uring: fix tw losing poll events

Message ID 00344d60f8b18907171178d7cf598de71d127b0b.1668710222.git.asml.silence@gmail.com (mailing list archive)
State New
Headers show
Series minor poll fixes | expand

Commit Message

Pavel Begunkov Nov. 17, 2022, 6:40 p.m. UTC 
We may never try to process a poll wake and its mask if there was
multiple wake ups racing for queueing up a tw. Force
io_poll_check_events() to update the mask by vfs_poll().

Cc: stable@vger.kernel.org
Fixes: aa43477b04025 ("io_uring: poll rework")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 io_uring/poll.c | 7 +++++++
 1 file changed, 7 insertions(+)
diff mbox series

Patch

diff --git a/io_uring/poll.c b/io_uring/poll.c
index 90920abf91ff..c34019b18211 100644
--- a/io_uring/poll.c
+++ b/io_uring/poll.c
@@ -228,6 +228,13 @@  static int io_poll_check_events(struct io_kiocb *req, bool *locked)
 			return IOU_POLL_DONE;
 		if (v & IO_POLL_CANCEL_FLAG)
 			return -ECANCELED;
+		/*
+		 * cqe.res contains only events of the first wake up
+		 * and all others are be lost. Redo vfs_poll() to get
+		 * up to date state.
+		 */
+		if ((v & IO_POLL_REF_MASK) != 1)
+			req->cqe.res = 0;
 
 		/* the mask was stashed in __io_poll_execute */
 		if (!req->cqe.res) {