io_uring: fix waiters missing wake ups

There are reports of mariadb hangs, which is caused by a missing
barrier in the waking code resulting in waiters losing events.

The problem was introduced in a backport
3ab9326 ("io_uring: wake up optimisations"),
and the change restores the barrier present in the original commit
3ab9326 ("io_uring: wake up optimisations")

Reported by: Xan Charbonnet <xan@charbonnet.com>
Fixes: 3ab9326 ("io_uring: wake up optimisations")
Link: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1093243#99
Reviewed-by: Li Zetao <lizetao1@huawei.com>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Author: isilence

io_uring/io_uring.c

@@ -597,8 +597,10 @@ static inline void __io_cq_unlock_post_flush(struct io_ring_ctx *ctx)
 	io_commit_cqring(ctx);
 	spin_unlock(&ctx->completion_lock);
 	io_commit_cqring_flush(ctx);
-	if (!(ctx->flags & IORING_SETUP_DEFER_TASKRUN))
+	if (!(ctx->flags & IORING_SETUP_DEFER_TASKRUN)) {
+		smp_mb();
 		__io_cqring_wake(ctx);
+	}
 }
 
 void io_cq_unlock_post(struct io_ring_ctx *ctx)