Re: [PATCH] net_sched: Add qdisc __NET_XMIT_STOLEN flag

[Jarek Poplawski <jarkao2@gmail.com>]

On Fri, Aug 01, 2008 at 08:15:50AM +0200, Patrick McHardy wrote:
> Jarek Poplawski wrote:
>> Hi,
>>
>> Here is a complete, I hope,  version of the patch to not linger this
>> all to much, but of course any redoing is possible.
>
> This looks good to me. Just one suggestion:
>
>> +#ifdef CONFIG_NET_CLS_ACT
>> +/* additional qdisc xmit flags */
>> +enum net_xmit_qdisc_t {
>> +	__NET_XMIT_STOLEN = NET_XMIT_MASK + 1,
>> +};
>> +
>> +#define net_xmit_drop_count(e)	((e) & __NET_XMIT_STOLEN ? 0 : 1)
>> +
>> +#else
>> +#define net_xmit_drop_count(e)	(1)
>> +#endif
>> +
>>  static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
>>  {
>>  #ifdef CONFIG_NET_SCHED
>> @@ -343,7 +355,7 @@ static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
>>  static inline int qdisc_enqueue_root(struct sk_buff *skb, struct Qdisc *sch)
>>  {
>>  	qdisc_skb_cb(skb)->pkt_len = skb->len;
>> -	return qdisc_enqueue(skb, sch);
>> +	return qdisc_enqueue(skb, sch) & NET_XMIT_MASK;
>
> It would be nice to handle NET_XMIT_BYPASS similar, currently its
> mapped to NET_XMIT_SUCCESS by dev_queue_xmit(). Simply adding it
> to net_xmit_qdisc_t as
>
> __NET_XMIT_BYPASS = NET_XMIT_MASK + 0x2
>
> and removing the mapping from dev_queue_xmit() should work I think.

OK, I think, I'll prepare a separate patch for this.

Thanks,
Jarek P.