Re: [PATCH net-next RFC] netfilter: flowtable: add CT metadata action for nft flowtables

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Elad Yifee <eladwf@xxxxxxxxx> wrote:
> When offloading a flow via the default nft flowtable path,
> append a FLOW_ACTION_CT_METADATA action if the flow is associated with a conntrack entry.
> We do this in both IPv4 and IPv6 route action builders, after NAT mangles and before redirect.
> This mirrors net/sched/act_ct.c’s tcf_ct_flow_table_add_action_meta() so drivers that already
> parse FLOW_ACTION_CT_METADATA from TC offloads can reuse the same logic for nft flowtables.
> 
> Signed-off-by: Elad Yifee <eladwf@xxxxxxxxx>
> ---
>  net/netfilter/nf_flow_table_offload.c | 38 +++++++++++++++++++++++++++
>  1 file changed, 38 insertions(+)
> 
> diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
> index e06bc36f49fe..bccae4052319 100644
> --- a/net/netfilter/nf_flow_table_offload.c
> +++ b/net/netfilter/nf_flow_table_offload.c
> @@ -12,6 +12,7 @@
>  #include <net/netfilter/nf_conntrack_acct.h>
>  #include <net/netfilter/nf_conntrack_core.h>
>  #include <net/netfilter/nf_conntrack_tuple.h>
> +#include <net/netfilter/nf_conntrack_labels.h>
>  
>  static struct workqueue_struct *nf_flow_offload_add_wq;
>  static struct workqueue_struct *nf_flow_offload_del_wq;
> @@ -679,6 +680,41 @@ nf_flow_rule_route_common(struct net *net, const struct flow_offload *flow,
>  	return 0;
>  }
>  
> +static void flow_offload_add_ct_metadata(const struct flow_offload *flow,
> +					 enum flow_offload_tuple_dir dir,
> +					 struct nf_flow_rule *flow_rule)
> +{
> +	struct nf_conn *ct = flow->ct;
> +	struct flow_action_entry *entry;
> +#if IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS)
> +	u32 *dst_labels;
> +	struct nf_conn_labels *labels;
> +#endif
> +
> +	if (!ct)
> +		return;

Under what circumstances can flow->ct be NULL?

> +	entry = flow_action_entry_next(flow_rule);
> +	entry->id = FLOW_ACTION_CT_METADATA;
> +
> +#if IS_ENABLED(CONFIG_NF_CONNTRACK_MARK)
> +	entry->ct_metadata.mark = READ_ONCE(ct->mark);
> +#endif
> +
> +	entry->ct_metadata.orig_dir = (dir == FLOW_OFFLOAD_DIR_ORIGINAL);
> +
> +#if IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS)
> +	dst_labels = entry->ct_metadata.labels;
> +	labels = nf_ct_labels_find(ct);
> +	if (labels)
> +		memcpy(dst_labels, labels->bits, NF_CT_LABELS_MAX_SIZE);
> +	else
> +		memset(dst_labels, 0, NF_CT_LABELS_MAX_SIZE);
> +#else
> +	memset(entry->ct_metadata.labels, 0, NF_CT_LABELS_MAX_SIZE);
> +#endif
> +}

This looks almost identical tcf_ct_flow_table_add_action_meta().

Any chance to make it a common helper function? act_ct already depends
on nf_flow_table anyway.




[Index of Archives]     [Netfitler Users]     [Berkeley Packet Filter]     [LARTC]     [Bugtraq]     [Yosemite Forum]

  Powered by Linux