io_uring: halt SQO submission on ctx exit

BugLink: https://bugs.launchpad.net/bugs/1923069
[ Upstream commit f6d54255f4235448d4bbe442362d4caa62da97d5 ]

io_sq_thread_finish() is called in io_ring_ctx_free(), so SQPOLL task is
potentially running submitting new requests. It's not a disaster because
of using a "try" variant of percpu_ref_get, but is far from nice.

Remove ctx from the sqd ctx list earlier, before cancellation loop, so
SQPOLL can't find it and so won't submit new requests.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Andrea Righi <andrea.righi@canonical.com>

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 381f82ebd28286d76524d9f6c7fd61c90adc47ac..aaf9b5d49c174ab8f2f3978b582a04ba8a09045a 100644 (file)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -8800,6 +8800,14 @@ static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx)
                __io_cqring_overflow_flush(ctx, true, NULL, NULL);
        mutex_unlock(&ctx->uring_lock);
 
+       /* prevent SQPOLL from submitting new requests */
+       if (ctx->sq_data) {
+               io_sq_thread_park(ctx->sq_data);
+               list_del_init(&ctx->sqd_list);
+               io_sqd_update_thread_idle(ctx->sq_data);
+               io_sq_thread_unpark(ctx->sq_data);
+       }
+
        io_kill_timeouts(ctx, NULL, NULL);
        io_poll_remove_all(ctx, NULL, NULL);