Files
linux/include/net
Eric Dumazet a54d51fb2d udp: fix busy polling
Generic sk_busy_loop_end() only looks at sk->sk_receive_queue
for presence of packets.

Problem is that for UDP sockets after blamed commit, some packets
could be present in another queue: udp_sk(sk)->reader_queue

In some cases, a busy poller could spin until timeout expiration,
even if some packets are available in udp_sk(sk)->reader_queue.

v3: - make sk_busy_loop_end() nicer (Willem)

v2: - add a READ_ONCE(sk->sk_family) in sk_is_inet() to avoid KCSAN splats.
    - add a sk_is_inet() check in sk_is_udp() (Willem feedback)
    - add a sk_is_inet() check in sk_is_tcp().

Fixes: 2276f58ac5 ("udp: use a separate rx queue for packet reception")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-01-21 18:09:30 +00:00
..
2023-12-29 07:46:38 +00:00
2024-01-02 12:41:16 +00:00
2022-08-09 22:14:02 -07:00
2023-10-24 13:08:14 -07:00
2023-10-25 18:04:29 -07:00
2023-11-02 09:31:02 +01:00
2021-10-15 11:33:08 +01:00
2024-01-21 18:09:30 +00:00
2023-04-22 01:39:41 +02:00
2021-10-13 09:40:46 -07:00
2023-08-22 17:31:18 -07:00
2022-05-11 12:43:10 +01:00
2023-07-14 20:39:30 -07:00
2023-12-12 14:24:14 +01:00
2024-01-21 18:09:30 +00:00
2023-10-04 11:49:20 -07:00
2024-01-02 12:41:16 +00:00
2023-10-23 15:01:53 -07:00
2023-07-28 14:07:59 -07:00
2022-12-12 15:04:39 -08:00
2023-09-14 16:16:36 +02:00
2023-12-14 16:38:59 +01:00