Re: [PATCH net-next] net: ipv4: allow directed broadcast routes to use dst hint

[Jakub Kicinski <kuba@kernel.org>]

On Thu, 24 Jul 2025 14:49:42 +0200 Oscar Maes wrote:
> Currently, ip_extract_route_hint uses RTN_BROADCAST to decide
> whether to use the route dst hint mechanism.
> 
> This check is too strict, as it prevents directed broadcast
> routes from using the hint, resulting in poor performance
> during bursts of directed broadcast traffic.
> 
> Fix this in ip_extract_route_hint and modify ip_route_use_hint
> to preserve the intended behaviour.

We are wrapping up our 6.17 material, I think this will need
to wait for 6.18. In the meantime, would it make sense to add
a selftest? Sounds like a relatively rare use case, easy to
regress.

> diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
> index fc323994b..1581b98bc 100644
> --- a/net/ipv4/ip_input.c
> +++ b/net/ipv4/ip_input.c
> @@ -589,8 +589,10 @@ static void ip_sublist_rcv_finish(struct list_head *head)
>  static struct sk_buff *ip_extract_route_hint(const struct net *net,
>  					     struct sk_buff *skb, int rt_type)
>  {
> -	if (fib4_has_custom_rules(net) || rt_type == RTN_BROADCAST ||
> -	    IPCB(skb)->flags & IPSKB_MULTIPATH)
> +	const struct iphdr *iph = ip_hdr(skb);
> +
> +	if (fib4_has_custom_rules(net) || ipv4_is_lbcast(iph->daddr) ||
> +	    (iph->daddr == 0 && iph->saddr == 0) || IPCB(skb)->flags & IPSKB_MULTIPATH)

nit: we still prefer to wrap lines at 80 chars in networking
-- 
pw-bot: cr