io_uring/rw: don't lose short results on io_setup_async_rw()

If a retry io_setup_async_rw() fails we lose result from the first
io_iter_do_read(), which is a problem mostly for streams/sockets.

Cc: stable@vger.kernel.org
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/0e8d20cebe5fc9c96ed268463c394237daabc384.1664235732.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

diff --git a/io_uring/rw.c b/io_uring/rw.c
index ed14322aadb90ee0a38169ab0971f60c1244bc36..1ae1e52ab4cbd233dcd4c559481311964326f85c 100644 (file)
--- a/io_uring/rw.c
+++ b/io_uring/rw.c
@@ -764,10 +764,12 @@ int io_read(struct io_kiocb *req, unsigned int issue_flags)
        iov_iter_restore(&s->iter, &s->iter_state);
 
        ret2 = io_setup_async_rw(req, iovec, s, true);
-       if (ret2)
-               return ret2;
-
        iovec = NULL;
+       if (ret2) {
+               ret = ret > 0 ? ret : ret2;
+               goto done;
+       }
+
        io = req->async_data;
        s = &io->s;
        /*