From: Jens Axboe <axboe@kernel.dk>

[ Upstream commit b5b0ecb736f1ce1e68eb50613c0cfecff10198eb ]

The callback can only be armed, if we get -EIOCBQUEUED returned. It's
important that we clear the WAITQ bit for other cases, otherwise we can
queue for async retry and filemap will assume that we're armed and
return -EAGAIN instead of just blocking for the IO.

Cc: stable@vger.kernel.org # 5.9+
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/io_uring.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 7625b3e2db2c..06e9c2181995 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c

@@ -3501,6 +3501,7 @@ static int io_read(struct io_kiocb *req, bool force_nonblock,
 		goto out_free;
 	} else if (ret > 0 && ret < io_size) {
 		/* we got some bytes, but not all. retry. */
+		kiocb->ki_flags &= ~IOCB_WAITQ;
 		goto retry;
 	}
 done:

-- 
2.30.1