Re: [PATCH net-next] inet_diag: Remove _bh suffix in inet_diag_dump_reqs().
From: Eric Dumazet <hidden>
Date: 2015-06-19 06:42:13
On Thu, 2015-06-18 at 20:40 +0900, Hiroaki Shimoda wrote:
quoted hunk ↗ jump to hunk
inet_diag_dump_reqs() is called from inet_diag_dump_icsk() with BH disabled. So no need to disable BH in inet_diag_dump_reqs(). Signed-off-by: Hiroaki Shimoda <redacted> --- net/ipv4/inet_diag.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index 21985d8d41e7..4ca789ba63cb 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c@@ -746,7 +746,7 @@ static int inet_diag_dump_reqs(struct sk_buff *skb, struct sock *sk, entry.family = sk->sk_family; - spin_lock_bh(&icsk->icsk_accept_queue.syn_wait_lock); + spin_lock(&icsk->icsk_accept_queue.syn_wait_lock); lopt = icsk->icsk_accept_queue.listen_opt; if (!lopt || !listen_sock_qlen(lopt))@@ -794,7 +794,7 @@ static int inet_diag_dump_reqs(struct sk_buff *skb, struct sock *sk, } out: - spin_unlock_bh(&icsk->icsk_accept_queue.syn_wait_lock); + spin_unlock(&icsk->icsk_accept_queue.syn_wait_lock); return err; }
Sure, although this will soon be removed completely when SYN_RECV sockets will be stored in regular ehash table. Thanks