DORMANTno replies

[PATCH net v1] rxrpc: fix io_thread race in rxrpc_wake_up_io_thread()

From: <hidden>
Date: 2026-07-08 09:36:54
Also in: lkml
Subsystem: networking [general], rxrpc sockets (af_rxrpc), the rest · Maintainers: "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, David Howells, Marc Dionne, Linus Torvalds

From: Xuanqiang Luo <redacted>

rxrpc_wake_up_io_thread() checks local->io_thread before waking it, but
then reloads the pointer for wake_up_process().

local->io_thread is cleared with WRITE_ONCE() when the I/O thread exits, so
the second load can see NULL even if the first load did not.

Take a READ_ONCE() snapshot and use it for both the NULL check and the
wake_up_process() call, as rxrpc_encap_rcv() already does.

Fixes: 5800b1cf3fd8 ("rxrpc: Allow CHALLENGEs to the passed to the app for a RESPONSE")
Signed-off-by: Xuanqiang Luo <redacted>
---
 net/rxrpc/ar-internal.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index ce946b0a03e2b..865f05fe37ab9 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -1285,9 +1285,11 @@ int rxrpc_io_thread(void *data);
 void rxrpc_post_response(struct rxrpc_connection *conn, struct sk_buff *skb);
 static inline void rxrpc_wake_up_io_thread(struct rxrpc_local *local)
 {
-	if (!local->io_thread)
+	struct task_struct *io_thread = READ_ONCE(local->io_thread);
+
+	if (!io_thread)
 		return;
-	wake_up_process(READ_ONCE(local->io_thread));
+	wake_up_process(io_thread);
 }
 
 static inline bool rxrpc_protocol_error(struct sk_buff *skb, enum rxrpc_abort_reason why)
-- 
2.43.0
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help