public inbox for [email protected]
 help / color / mirror / Atom feed
* [PATCH] io_uring/eventfd: move refs to refcount_t
@ 2024-09-08 22:43 Jens Axboe
  0 siblings, 0 replies; only message in thread
From: Jens Axboe @ 2024-09-08 22:43 UTC (permalink / raw)
  To: io-uring

atomic_t for the struct io_ev_fd references and there are no issues with
it. While the ref getting and putting for the eventfd code is somewhat
performance critical for cases where eventfd signaling is used (news
flash, you should not...), it probably doesn't warrant using an atomic_t
for this. Let's just move to it to refcount_t to get the added
protection of over/underflows.

Link: https://lore.kernel.org/lkml/[email protected]/
Reported-by: kernel test robot <[email protected]>
Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
Signed-off-by: Jens Axboe <[email protected]>

---

diff --git a/io_uring/eventfd.c b/io_uring/eventfd.c
index d9836d43725f..e37fddd5d9ce 100644
--- a/io_uring/eventfd.c
+++ b/io_uring/eventfd.c
@@ -15,7 +15,7 @@ struct io_ev_fd {
 	struct eventfd_ctx	*cq_ev_fd;
 	unsigned int		eventfd_async: 1;
 	struct rcu_head		rcu;
-	atomic_t		refs;
+	refcount_t		refs;
 	atomic_t		ops;
 };
 
@@ -37,7 +37,7 @@ static void io_eventfd_do_signal(struct rcu_head *rcu)
 
 	eventfd_signal_mask(ev_fd->cq_ev_fd, EPOLL_URING_WAKE);
 
-	if (atomic_dec_and_test(&ev_fd->refs))
+	if (refcount_dec_and_test(&ev_fd->refs))
 		io_eventfd_free(rcu);
 }
 
@@ -63,7 +63,7 @@ void io_eventfd_signal(struct io_ring_ctx *ctx)
 	 */
 	if (unlikely(!ev_fd))
 		return;
-	if (!atomic_inc_not_zero(&ev_fd->refs))
+	if (!refcount_inc_not_zero(&ev_fd->refs))
 		return;
 	if (ev_fd->eventfd_async && !io_wq_current_is_worker())
 		goto out;
@@ -77,7 +77,7 @@ void io_eventfd_signal(struct io_ring_ctx *ctx)
 		}
 	}
 out:
-	if (atomic_dec_and_test(&ev_fd->refs))
+	if (refcount_dec_and_test(&ev_fd->refs))
 		call_rcu(&ev_fd->rcu, io_eventfd_free);
 }
 
@@ -137,7 +137,7 @@ int io_eventfd_register(struct io_ring_ctx *ctx, void __user *arg,
 
 	ev_fd->eventfd_async = eventfd_async;
 	ctx->has_evfd = true;
-	atomic_set(&ev_fd->refs, 1);
+	refcount_set(&ev_fd->refs, 1);
 	atomic_set(&ev_fd->ops, 0);
 	rcu_assign_pointer(ctx->io_ev_fd, ev_fd);
 	return 0;
@@ -152,7 +152,7 @@ int io_eventfd_unregister(struct io_ring_ctx *ctx)
 	if (ev_fd) {
 		ctx->has_evfd = false;
 		rcu_assign_pointer(ctx->io_ev_fd, NULL);
-		if (atomic_dec_and_test(&ev_fd->refs))
+		if (refcount_dec_and_test(&ev_fd->refs))
 			call_rcu(&ev_fd->rcu, io_eventfd_free);
 		return 0;
 	}

-- 
Jens Axboe


^ permalink raw reply related	[flat|nested] only message in thread

only message in thread, other threads:[~2024-09-08 22:43 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-09-08 22:43 [PATCH] io_uring/eventfd: move refs to refcount_t Jens Axboe

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox