[PATCH 5.11 11/31] io_uring: clear IOCB_WAITQ for non -EIOCBQUEUED return

[Greg Kroah-Hartman]

From: Jens Axboe <axboe@xxxxxxxxx>

[ 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@xxxxxxxxxxxxxxx # 5.9+
Signed-off-by: Jens Axboe <axboe@xxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
 fs/io_uring.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index c18e4a334614..262fd4cfd3ad 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -3587,6 +3587,7 @@ 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