Re: [PATCH 2/2]: niu: Manage TX backlog in-driver.

[David Miller <davem@davemloft.net>]

From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Mon, 23 Jun 2008 10:04:45 +0800

> David Miller said the following on 2008-6-23 7:16:
> > +
> > +		while ((skb = __skb_dequeue(&rp->tx_backlog)) != NULL)
> > +			dev_kfree_skb(skb);
> 
> How about __skb_queue_purge(&rp->tx_backlog);

Sure, I can use that, thanks for the suggestion.

> > +static void __niu_tx_queue_backlog(struct niu *np, struct tx_ring_info *rp,
> > +				   struct sk_buff *skb)
> > +{
> > +	if (skb_queue_len(&rp->tx_backlog) < np->dev->tx_queue_len)
> > +		__skb_queue_tail(&rp->tx_backlog, skb);
> 
> Do we need lock to proteck tx_backlog?
> How about use skb_queue_tail()?

That would just grab another lock (the sk_buff_head one), and the
overall goal of these changes is to decrease the number of locks taken
to TX packets and purge the TX ring during completions.