[PATCH] io_uring/poll: fix multishot recv missing EOF on wakeup race

[Jens Axboe <axboe@kernel.dk>]

When a socket send and shutdown() happen back-to-back, both fire
wake-ups before the receiver's task_work has a chance to run. The first
wake gets poll ownership (poll_refs=1), and the second bumps it to 2.
When io_poll_check_events() runs, it calls io_poll_issue() which does a
recv that reads the data and returns IOU_RETRY. The loop then drains all
accumulated refs (atomic_sub_return(2) -> 0) and exits, even though only
the first event was consumed. Since the shutdown is a persistent state
change, no further wakeups will happen, and the multishot recv can hang
forever.

Fix this by only draining a single poll ref after io_poll_issue()
returns IOU_RETRY for the APOLL_MULTISHOT path. If additional wakes
raced in (poll_refs was > 1), the loop iterates again, vfs_poll()
discovers the remaining state.

Move the v &= IO_POLL_REF_MASK (drain all refs) into the non-APOLL
multishot poll path, since poll CQEs report the current mask state
rather than consuming individual events.

Cc: stable@vger.kernel.org
Fixes: dbc2564cfe0f ("io_uring: let fast poll support multishot")
Reported-by: francis <francis@brosseau.dev>
Link: https://github.com/axboe/liburing/issues/1549
Signed-off-by: Jens Axboe <axboe@kernel.dk>

---

diff --git a/io_uring/poll.c b/io_uring/poll.c
index b671b84657d9..0f0949d919e9 100644
--- a/io_uring/poll.c
+++ b/io_uring/poll.c
@@ -303,6 +303,7 @@ static int io_poll_check_events(struct io_kiocb *req, io_tw_token_t tw)
 				io_req_set_res(req, mask, 0);
 				return IOU_POLL_REMOVE_POLL_USE_RES;
 			}
+			v &= IO_POLL_REF_MASK;
 		} else {
 			int ret = io_poll_issue(req, tw);
 
@@ -312,6 +313,11 @@ static int io_poll_check_events(struct io_kiocb *req, io_tw_token_t tw)
 				return IOU_POLL_REQUEUE;
 			if (ret != IOU_RETRY && ret < 0)
 				return ret;
+			/*
+			 * One event consumed, but additional wakes may have
+			 * raced. Only drain a single ref.
+			 */
+			v = 1;
 		}
 
 		/* force the next iteration to vfs_poll() */
@@ -321,7 +327,6 @@ static int io_poll_check_events(struct io_kiocb *req, io_tw_token_t tw)
 		 * Release all references, retry if someone tried to restart
 		 * task_work while we were executing it.
 		 */
-		v &= IO_POLL_REF_MASK;
 	} while (atomic_sub_return(v, &req->poll_refs) & IO_POLL_REF_MASK);
 
 	io_napi_add(req);

-- 
Jens Axboe