io_uring: use TWA_SIGNAL for task_work uncondtionally

An earlier commit:

b7db41c9e0 ("io_uring: fix regression with always ignoring signals in io_cqring_wait()")

ensured that we didn't get stuck waiting for eventfd reads when it's
registered with the io_uring ring for event notification, but we still
have cases where the task can be waiting on other events in the kernel and
need a bigger nudge to make forward progress. Or the task could be in the
kernel and running, but on its way to blocking.

This means that TWA_RESUME cannot reliably be used to ensure we make
progress. Use TWA_SIGNAL unconditionally.

Cc: stable@vger.kernel.org # v5.7+
Reported-by: Josef <josef.grieb@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

fs/io_uring.c

@@ -1710,22 +1710,22 @@ static int io_req_task_work_add(struct io_kiocb *req, struct callback_head *cb)
 {
 	struct task_struct *tsk = req->task;
 	struct io_ring_ctx *ctx = req->ctx;
-	int ret, notify = TWA_RESUME;
+	int ret, notify;
 
 	/*
-	 * SQPOLL kernel thread doesn't need notification, just a wakeup.
-	 * If we're not using an eventfd, then TWA_RESUME is always fine,
-	 * as we won't have dependencies between request completions for
-	 * other kernel wait conditions.
+	 * SQPOLL kernel thread doesn't need notification, just a wakeup. For
+	 * all other cases, use TWA_SIGNAL unconditionally to ensure we're
+	 * processing task_work. There's no reliable way to tell if TWA_RESUME
+	 * will do the job.
 	 */
-	if (ctx->flags & IORING_SETUP_SQPOLL)
-		notify = 0;
-	else if (ctx->cq_ev_fd)
+	notify = 0;
+	if (!(ctx->flags & IORING_SETUP_SQPOLL))
 		notify = TWA_SIGNAL;
 
 	ret = task_work_add(tsk, cb, notify);
 	if (!ret)
 		wake_up_process(tsk);
+
 	return ret;
 }