Re: [PATCH v2] ipv6: add `force_forwarding` sysctl to enable per-interface forwarding

[Nicolas Dichtel <nicolas.dichtel@6wind.com>]

Le 01/07/2025 à 16:04, Gabriel Goller a écrit :
> It is currently impossible to enable ipv6 forwarding on a per-interface
> basis like in ipv4. To enable forwarding on an ipv6 interface we need to
> enable it on all interfaces and disable it on the other interfaces using
> a netfilter rule. This is especially cumbersome if you have lots of
> interface and only want to enable forwarding on a few. According to the
> sysctl docs [0] the `net.ipv6.conf.all.forwarding` enables forwarding
> for all interfaces, while the interface-specific
> `net.ipv6.conf.<interface>.forwarding` configures the interface
> Host/Router configuration.
> 
> Introduce a new sysctl flag `force_forwarding`, which can be set on every
> interface. The ip6_forwarding function will then check if the global
> forwarding flag OR the force_forwarding flag is active and forward the
> packet.
> 
> To preserver backwards-compatibility reset the flag (global and on all
> interfaces) to 0 if the net.ipv6.conf.all.forwarding flag is set to 0.
> 
> [0]: https://www.kernel.org/doc/Documentation/networking/ip-sysctl.txt
> 
> Signed-off-by: Gabriel Goller <g.goller@proxmox.com>
> ---

[snip]

> @@ -896,6 +907,16 @@ static int addrconf_fixup_forwarding(const struct ctl_table *table, int *p, int
>  						     NETCONFA_IFINDEX_DEFAULT,
>  						     net->ipv6.devconf_dflt);
>  
> +		/*
> +		 * With the introduction of force_forwarding, we need to be backwards
> +		 * compatible, so that means we need to set the force_forwarding global
> +		 * flag to 0 if the global forwarding flag is set to 0. Below in
> +		 * addrconf_forward_change(), we also set the force_forwarding flag on every
> +		 * interface to 0 if the global forwarding flag is set to 0.
> +		 */
> +		if (newf == 0)
> +			WRITE_ONCE(net->ipv6.devconf_all->force_forwarding, newf);
Hmm, is this true? Configuring the default value only impacts new interfaces.
And before your patch, only the 'all' entry is took into account. In other
words, configuring the default entry today doesn't change the current behavior,
so I don't see the backward compat point.

> +
>  		addrconf_forward_change(net, newf);
>  		if ((!newf) ^ (!old))
>  			inet6_netconf_notify_devconf(net, RTM_NEWNETCONF,

[snip]

> +static int addrconf_sysctl_force_forwarding(const struct ctl_table *ctl, int write,
> +					    void *buffer, size_t *lenp, loff_t *ppos)
> +{
> +	int *valp = ctl->data;
> +	int ret;
> +	int old, new;
> +
> +	old = *valp;
> +	ret = proc_douintvec(ctl, write, buffer, lenp, ppos);
> +	new = *valp;
Maybe you can limit values to 0 and 1, like it was done in the v1.


Regards,
Nicolas