Re: [PATCH 05/10] netfilter: merge tcpv[4,6]_net_init into tcp_net_init

[Pablo Neira Ayuso <pablo@netfilter.org>]

On Thu, Jun 14, 2012 at 06:07:20PM +0800, Gao feng wrote:
> merge tcpv4_net_init and tcpv6_net_init into tcp_net_init to
> reduce the redundancy codes.
> 
> and use nf_proto_net.users to identify if it's the first time
> we use the nf_proto_net. when it's the first time,we will
> initialized it.
> 
> Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com>
> ---
>  net/netfilter/nf_conntrack_proto_tcp.c |   57 ++++++++------------------------
>  1 files changed, 14 insertions(+), 43 deletions(-)
> 
> diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
> index 6db9d3c..e3d5427 100644
> --- a/net/netfilter/nf_conntrack_proto_tcp.c
> +++ b/net/netfilter/nf_conntrack_proto_tcp.c
> @@ -1593,18 +1593,14 @@ static int tcp_kmemdup_compat_sysctl_table(struct nf_proto_net *pn)
>  	return 0;
>  }
>  
> -static int tcpv4_init_net(struct net *net, u_int16_t proto)
> +static int tcp_init_net(struct net *net, u_int16_t proto)
>  {
> -	int i;
>  	int ret = 0;
>  	struct nf_tcp_net *tn = tcp_pernet(net);
>  	struct nf_proto_net *pn = (struct nf_proto_net *)tn;

while at it, it would be fine if you use &tn->pn instead. I know this
cast is making the trick, but what I propose is better practise.

> -#ifdef CONFIG_SYSCTL
> -	if (!pn->ctl_table) {
> -#else
> -	if (!pn->users++) {
> -#endif
> +	if (!pn->users) {
> +		int i = 0;
>  		for (i = 0; i < TCP_CONNTRACK_TIMEOUT_MAX; i++)
>  			tn->timeouts[i] = tcp_timeouts[i];
>  
> @@ -1613,45 +1609,20 @@ static int tcpv4_init_net(struct net *net, u_int16_t proto)
>  		tn->tcp_max_retrans = nf_ct_tcp_max_retrans;
>  	}
>  
> -	ret = tcp_kmemdup_compat_sysctl_table(pn);
> +	if (proto == AF_INET) {
> +		ret = tcp_kmemdup_compat_sysctl_table(pn);
> +		if (ret < 0)
> +			return ret;
>  
> -	if (ret < 0)
> -		return ret;
> +		ret = tcp_kmemdup_sysctl_table(pn);

One thing I noticed: This kmemdup will happen twice, once for IPv4 and
once for IPv6. I think this should happen only once, as both TCP
tracker for IPv4 and IPv6 are sharing the same nf_proto_net.

So it should happen inside the if (!pn->users) thing.

AFAICS, then this should look like the following:

if (pn->users)
        return 0;

/*
 * here comes all per-net initialization
 */

> +		if (ret < 0)
> +			nf_ct_kfree_compat_sysctl_table(pn);
> +	} else
> +		ret = tcp_kmemdup_sysctl_table(pn);
>  
> -	ret = tcp_kmemdup_sysctl_table(pn);
> -
> -#ifdef CONFIG_SYSCTL
> -#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT
> -	if (ret < 0) {
> -		kfree(pn->ctl_compat_table);
> -		pn->ctl_compat_table = NULL;
> -	}
> -#endif
> -#endif
>  	return ret;
>  }
>  
> -static int tcpv6_init_net(struct net *net, u_int16_t proto)
> -{
> -	int i;
> -	struct nf_tcp_net *tn = tcp_pernet(net);
> -	struct nf_proto_net *pn = (struct nf_proto_net *)tn;
> -
> -#ifdef CONFIG_SYSCTL
> -	if (!pn->ctl_table) {
> -#else
> -	if (!pn->users++) {
> -#endif
> -		for (i = 0; i < TCP_CONNTRACK_TIMEOUT_MAX; i++)
> -			tn->timeouts[i] = tcp_timeouts[i];
> -		tn->tcp_loose = nf_ct_tcp_loose;
> -		tn->tcp_be_liberal = nf_ct_tcp_be_liberal;
> -		tn->tcp_max_retrans = nf_ct_tcp_max_retrans;
> -	}
> -
> -	return tcp_kmemdup_sysctl_table(pn);
> -}
> -
>  struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4 __read_mostly =
>  {
>  	.l3proto		= PF_INET,
> @@ -1684,7 +1655,7 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4 __read_mostly =
>  		.nla_policy	= tcp_timeout_nla_policy,
>  	},
>  #endif /* CONFIG_NF_CT_NETLINK_TIMEOUT */
> -	.init_net		= tcpv4_init_net,
> +	.init_net		= tcp_init_net,
>  };
>  EXPORT_SYMBOL_GPL(nf_conntrack_l4proto_tcp4);
>  
> @@ -1720,6 +1691,6 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp6 __read_mostly =
>  		.nla_policy	= tcp_timeout_nla_policy,
>  	},
>  #endif /* CONFIG_NF_CT_NETLINK_TIMEOUT */
> -	.init_net		= tcpv6_init_net,
> +	.init_net		= tcp_init_net,
>  };
>  EXPORT_SYMBOL_GPL(nf_conntrack_l4proto_tcp6);
> -- 
> 1.7.7.6
>