Re: [PATCH,RFC] Route match v2

[Patrick McHardy <kaber@trash.net>]

Phil Oester wrote:
> On Mon, Jul 07, 2008 at 02:00:32PM +0200, Patrick McHardy wrote:
>>> +static bool
>>> +route_mt6(const struct sk_buff *skb, const struct net_device *in,
>>> +           const struct net_device *out, const struct xt_match *match,
>>> +           const void *matchinfo, int offset, unsigned int protoff,
>>> +           bool *hotdrop)
>>> +{
>>> +	const struct xt_route_info *info = matchinfo;
>>> +	const struct ipv6hdr *iph = ipv6_hdr(skb);
>>> +	struct fib6_node *fn;
>>> +	struct flowi fl = {0};
>>> +
>>> +	switch (info->mode) {
>>> +	case XT_ROUTE_SRC_EXISTS:
>>> +		fl.nl_u.ip6_u.daddr = iph->saddr;
>>> +		fn = fib6_lookup(&dev_net(in)->ipv6.fib6_main_tbl->tb6_root, &fl.fl6_dst, NULL);
>> This is always using the main table, which is inconsistent
>> with the IPv4 support. It also shouldn't call IPv6 functions
>> directly to avoid incorrect module dependencies.
> 
> What are the alternatives?  If we need to support IPv6 in the match, then
> we need to call the functions directly, no?  Is there a helper function
> I'm missing?

There are the ->route functions is nf_afinfo, but I'm not sure
if they will work for you. You may have too add some new ones.