Re: [PATCH net] vrf: check forwarding on the original netdevice when generating ICMP dest unreachable

[David Ahern <dsahern@gmail.com>]

On 2/27/18 11:36 AM, Stephen Suryaputra wrote:
> When ip_error() is called the device is the l3mdev master instead of the
> original device. So the forwarding check should be on the original one.
> 
> Signed-off-by: Stephen Suryaputra <ssuryaextr@gmail.com>
> ---
>  net/ipv4/route.c | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/net/ipv4/route.c b/net/ipv4/route.c
> index a4f44d8..2274503 100644
> --- a/net/ipv4/route.c
> +++ b/net/ipv4/route.c
> @@ -933,6 +933,7 @@ static int ip_error(struct sk_buff *skb)
>  	struct in_device *in_dev = __in_dev_get_rcu(skb->dev);
>  	struct rtable *rt = skb_rtable(skb);
>  	struct inet_peer *peer;
> +	int in_dev_forward;
>  	unsigned long now;
>  	struct net *net;
>  	bool send;
> @@ -943,7 +944,13 @@ static int ip_error(struct sk_buff *skb)
>  		goto out;
>  
>  	net = dev_net(rt->dst.dev);
> -	if (!IN_DEV_FORWARD(in_dev)) {
> +	if (netif_is_l3_master(skb->dev)) {
> +		in_dev_forward = IN_DEV_FORWARD(
> +			__in_dev_get_rcu(__dev_get_by_index(net, IPCB(skb)->iif)));


I think this should be something like:

	struct net_device *dev = skb->dev;
	struct in_device *in_dev;

	if (netif_is_l3_master(skb->dev))
		dev = __dev_get_by_index(net, IPCB(skb)->iif);

	in_dev = __in_dev_get_rcu(dev);

The rest can stay; all you really need is to reset the device on which
the __in_dev_get_rcu is done.


> +	} else {
> +		in_dev_forward = IN_DEV_FORWARD(in_dev);
> +	}
> +	if (!in_dev_forward) {
>  		switch (rt->dst.error) {
>  		case EHOSTUNREACH:
>  			__IP_INC_STATS(net, IPSTATS_MIB_INADDRERRORS);
>