[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZtcoblYi68X8t3Bd@hog>
Date: Tue, 3 Sep 2024 17:17:02 +0200
From: Sabrina Dubroca <sd@...asysnail.net>
To: Antonio Quartulli <antonio@...nvpn.net>
Cc: netdev@...r.kernel.org, kuba@...nel.org, pabeni@...hat.com,
ryazanov.s.a@...il.com, edumazet@...gle.com, andrew@...n.ch
Subject: Re: [PATCH net-next v6 17/25] ovpn: implement keepalive mechanism
2024-08-27, 14:07:57 +0200, Antonio Quartulli wrote:
> +static time64_t ovpn_peer_keepalive_work_mp(struct ovpn_struct *ovpn,
> + time64_t now)
> +{
> + time64_t tmp_next_run, next_run = 0;
> + struct hlist_node *tmp;
> + struct ovpn_peer *peer;
> + int bkt;
> +
> + spin_lock_bh(&ovpn->peers->lock_by_id);
> + hash_for_each_safe(ovpn->peers->by_id, bkt, tmp, peer, hash_entry_id) {
> + tmp_next_run = ovpn_peer_keepalive_work_single(peer, now);
> +
> + /* the next worker run will be scheduled based on the shortest
> + * required interval across all peers
> + */
> + if (!next_run || tmp_next_run < next_run)
I think this should exclude tmp_next_run == 0.
If we have two peers, with the first getting a non-0 value and the 2nd
getting 0, we'll end up with next_run = 0 on return.
If we have three peers and ovpn_peer_keepalive_work_single returns
12,0,42, we'll end up with 42 (after resetting to 0 on the 2nd peer),
and we could miss sending the needed keepalive for peer 1.
> + next_run = tmp_next_run;
> + }
> + spin_unlock_bh(&ovpn->peers->lock_by_id);
> +
> + return next_run;
> +}
--
Sabrina
Powered by blists - more mailing lists