Re: [PATCH 1/1] Signed-off-by: Jan Wang

[Eric Dumazet]

On Sun, 2016-12-11 at 23:31 +0800, Jan Wang wrote:
> [net]: Missing init queue tail
> 
> Accept queue tail doesn't initialize NULL.
> Though looks like no harm, unify it with reqsk_queue_remove(
> it will initialize tail to NULL when empty).

Well, if having a NULL or not causes no harm, what about removing
code, instead of adding some defensive one ?

Also your patch submission is wrong :
Title should not include 'Signed-off-by...' 

Anyway, net-next is closed, you'll have to wait before submitting such a
change.

diff --git a/include/net/request_sock.h b/include/net/request_sock.h
index 
6ebe13eb1c4cbcd84c4b3345b051b3320d7591e6..a655c6aea7c0c8c429d5629c7fb639d6de832438
 100644
--- a/include/net/request_sock.h
+++ b/include/net/request_sock.h
@@ -195,8 +195,6 @@ static inline struct request_sock 
*reqsk_queue_remove(struct request_sock_queue
        if (req) {
                sk_acceptq_removed(parent);
                queue->rskq_accept_head = req->dl_next;
-               if (queue->rskq_accept_head == NULL)
-                       queue->rskq_accept_tail = NULL;
        }
        spin_unlock_bh(&queue->rskq_lock);
        return req;