[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20221215184138.795576-1-dylany@meta.com>
Date: Thu, 15 Dec 2022 10:41:38 -0800
From: Dylan Yudaken <dylany@...a.com>
To: Jens Axboe <axboe@...nel.dk>,
Pavel Begunkov <asml.silence@...il.com>
CC: <io-uring@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
<kernel-team@...com>, Dylan Yudaken <dylany@...a.com>,
Joel Fernandes <joel@...lfernandes.org>,
"Paul E . McKenney" <paulmck@...nel.org>
Subject: [PATCH] io_uring: use call_rcu_hurry if signaling an eventfd
io_uring uses call_rcu in the case it needs to signal an eventfd as a
result of an eventfd signal, since recursing eventfd signals are not
allowed. This should be calling the new call_rcu_hurry API to not delay
the signal.
Signed-off-by: Dylan Yudaken <dylany@...a.com>
Cc: Joel Fernandes (Google) <joel@...lfernandes.org>
Cc: Paul E. McKenney <paulmck@...nel.org>
---
io_uring/io_uring.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index b521186efa5c..2cb8c665bab1 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -538,7 +538,7 @@ static void io_eventfd_signal(struct io_ring_ctx *ctx)
} else {
atomic_inc(&ev_fd->refs);
if (!atomic_fetch_or(BIT(IO_EVENTFD_OP_SIGNAL_BIT), &ev_fd->ops))
- call_rcu(&ev_fd->rcu, io_eventfd_ops);
+ call_rcu_hurry(&ev_fd->rcu, io_eventfd_ops);
else
atomic_dec(&ev_fd->refs);
}
base-commit: 041fae9c105ae342a4245cf1e0dc56a23fbb9d3c
--
2.30.2
Powered by blists - more mailing lists