[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1339104606.6001.4.camel@edumazet-glaptop>
Date: Thu, 07 Jun 2012 23:30:06 +0200
From: Eric Dumazet <eric.dumazet@...il.com>
To: Francois Romieu <romieu@...zoreil.com>
Cc: Denys Fedoryshchenko <denys@...p.net.lb>, davem@...emloft.net,
netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: Deadlock, L2TP over IP are not working, 3.4.1
On Thu, 2012-06-07 at 22:53 +0200, Francois Romieu wrote:
> Any reason why it could not be made LLTX ?
>
> /* via l2tp_session_priv() */
> @@ -87,25 +95,56 @@ static int l2tp_eth_dev_xmit(struct sk_buff *skb, struct net_device *dev)
> {
> struct l2tp_eth *priv = netdev_priv(dev);
> struct l2tp_session *session = priv->session;
> + struct l2tp_eth_stats *tstats = &priv->tstats;
>
> l2tp_xmit_skb(session, skb, session->hdr_len);
>
> - dev->stats.tx_bytes += skb->len;
> - dev->stats.tx_packets++;
> + u64_stats_update_begin(&tstats->syncp);
> + tstats->packets++;
> + tstats->bytes += skb->len;
> + u64_stats_update_end(&tstats->syncp);
>
> return 0;
> }
>
Its racy.
If LLTX is used, this means several cpus can execute this code at the
same time.
You need percpu stats, or use atomic primitives.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists