netfilter: add a xt_FLOWOFFLOAD target for NAT/routing offload support

This makes it possible to add an iptables rule that offloads routing/NAT
packet processing to a software fast path. This fast path is much
quicker than running packets through the regular tables/chains.

Requires Linux 4.14

Signed-off-by: Felix Fietkau <nbd@nbd.name>
This commit is contained in:
Felix Fietkau
2018-02-20 15:58:42 +01:00
parent 1033356442
commit 820f030998
4 changed files with 479 additions and 1 deletions

View File

@@ -143,6 +143,8 @@ $(eval $(call nf_add,IPT_IPSEC,CONFIG_IP_NF_MATCH_AH, $(P_V4)ipt_ah))
$(eval $(call nf_add,IPT_IPSEC,CONFIG_NETFILTER_XT_MATCH_ESP, $(P_XT)xt_esp))
$(eval $(call nf_add,IPT_IPSEC,CONFIG_NETFILTER_XT_MATCH_POLICY, $(P_XT)xt_policy))
# flow offload support
$(eval $(call nf_add,IPT_FLOW,CONFIG_NETFILTER_XT_TARGET_FLOWOFFLOAD, $(P_XT)xt_FLOWOFFLOAD))
# IPv6
@@ -370,6 +372,7 @@ IPT_BUILTIN += $(IPT_CONNTRACK-y)
IPT_BUILTIN += $(IPT_CONNTRACK_EXTRA-y)
IPT_BUILTIN += $(IPT_EXTRA-y)
IPT_BUILTIN += $(IPT_FILTER-y)
IPT_BUILTIN += $(IPT_FLOW-y) $(IPT_FLOW-m)
IPT_BUILTIN += $(IPT_IPOPT-y)
IPT_BUILTIN += $(IPT_IPRANGE-y)
IPT_BUILTIN += $(IPT_CLUSTER-y)