Netfilter/IPVS updates for net-next
Subsystems (4)
Pablo Neira Ayuso
maintainercover Netfilter/IPVS updates for net-next
Hi,
The following patchset contains Netfilter/IPVS updates for net-next:
1) Inspect the reply packets coming from DR/TUN and refresh connection state and timeout, from longguang yue and Julian Anastasov.
2) Series to add support for the inet ingress chain type in nf_tables.
Please, pull these changes from:
git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next.git
Thank you.
----------------------------------------------------------------
The following changes since commit bc081a693a56061f68f736c5d596134ee3c87689:
Merge branch 'Offload-tc-vlan-mangle-to-mscc_ocelot-switch' (2020-10-11 11:19:25 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next.git HEAD
for you to fetch changes up to 793d5d61242695142460ce74f124281e0681fbc7:
netfilter: flowtable: reduce calls to pskb_may_pull() (2020-10-12 01:58:10 +0200)
----------------------------------------------------------------
Pablo Neira Ayuso (5):
netfilter: add nf_static_key_{inc,dec}
netfilter: add nf_ingress_hook() helper function
netfilter: add inet ingress support
netfilter: nf_tables: add inet ingress support
netfilter: flowtable: reduce calls to pskb_may_pull()
longguang.yue (1):
ipvs: inspect reply packets from DR/TUN real servers
include/net/netfilter/nf_tables.h | 6 ++ include/net/netfilter/nf_tables_ipv4.h | 33 +++++++++ include/net/netfilter/nf_tables_ipv6.h | 46 ++++++++++++ include/uapi/linux/netfilter.h | 1 + net/netfilter/core.c | 129 ++++++++++++++++++++++++++------- net/netfilter/ipvs/ip_vs_conn.c | 18 ++++- net/netfilter/ipvs/ip_vs_core.c | 19 ++--- net/netfilter/nf_flow_table_core.c | 12 +-- net/netfilter/nf_flow_table_ip.c | 45 +++++++----- net/netfilter/nf_tables_api.c | 14 ++-- net/netfilter/nft_chain_filter.c | 35 ++++++++- 11 files changed, 282 insertions(+), 76 deletions(-)
Pablo Neira Ayuso
maintainerPATCH 1/6 ipvs: inspect reply packets from DR/TUN real servers
From: "longguang.yue" <[email protected]>
Just like for MASQ, inspect the reply packets coming from DR/TUN real servers and alter the connection's state and timeout according to the protocol.
It's ipvs's duty to do traffic statistic if packets get hit, no matter what mode it is.
--- net/netfilter/ipvs/ip_vs_conn.c | 18 +++++++++++++++--- net/netfilter/ipvs/ip_vs_core.c | 19 +++++++------------ 2 files changed, 22 insertions(+), 15 deletions(-)
net/netfilter/ipvs/ip_vs_conn.c +15 -3
@@ -402,6 +402,8 @@ struct ip_vs_conn *ip_vs_conn_out_get(const struct ip_vs_conn_param *p) {unsigned int hash;struct ip_vs_conn *cp, *ret=NULL ; +const union nf_inet_addr *saddr; +__be16 sport;/* * Check for "full" addressed entries
@@ -411,10 +413,20 @@ struct ip_vs_conn *ip_vs_conn_out_get(const struct ip_vs_conn_param *p)rcu_read_lock ();hlist_for_each_entry_rcu (cp, &ip_vs_conn_tab[hash], c_list) { -if (p->vport == cp->cport && p->cport == cp->dport && - cp->af == p->af && +if (p->vport != cp->cport) +continue ; + +if (IP_VS_FWD_METHOD (cp) !=IP_VS_CONN_F_MASQ ) { + sport = cp->vport; + saddr = &cp->vaddr; + }else { + sport = cp->dport; + saddr = &cp->daddr; + } + +if (p->cport == sport && cp->af == p->af &&ip_vs_addr_equal (p->af, p->vaddr, &cp->caddr) && -ip_vs_addr_equal (p->af, p->caddr, &cp->daddr) && +ip_vs_addr_equal (p->af, p->caddr, saddr) && p->protocol == cp->protocol && cp->ipvs == p->ipvs) {if (!__ip_vs_conn_get (cp))
net/netfilter/ipvs/ip_vs_core.c +7 -12
@@ -875,7 +875,7 @@ static int handle_response_icmp(int af, struct sk_buff *skb,unsigned int verdict =NF_DROP ;if (IP_VS_FWD_METHOD (cp) !=IP_VS_CONN_F_MASQ ) - goto ignore_cp; + goto after_nat;/* Ensure the checksum is correct */ if (!skb_csum_unnecessary (skb) &&ip_vs_checksum_complete (skb, ihl)) {
@@ -901,6 +901,7 @@ static int handle_response_icmp(int af, struct sk_buff *skb,if (ip_vs_route_me_harder (cp->ipvs, af, skb, hooknum)) goto out; +after_nat:/* do the statistics and put it back */ ip_vs_out_stats (cp, skb);
@@ -909,8 +910,6 @@ static int handle_response_icmp(int af, struct sk_buff *skb,ip_vs_notrack (skb); else ip_vs_update_conntrack(skb, cp, 0); - -ignore_cp: verdict = NF_ACCEPT; out:
@@ -1276,6 +1275,9 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, {struct ip_vs_protocol *pp = pd->pp; +if (IP_VS_FWD_METHOD (cp) !=IP_VS_CONN_F_MASQ ) + goto after_nat; +IP_VS_DBG_PKT (11 , af, pp, skb, iph->off,"Outgoing packet" );if (skb_ensure_writable (skb, iph->len))
@@ -1316,6 +1318,7 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd,IP_VS_DBG_PKT (10 , af, pp, skb, iph->off,"After SNAT" ); +after_nat:ip_vs_out_stats (cp, skb);ip_vs_set_state (cp,IP_VS_DIR_OUTPUT , skb, pd); skb->ipvs_property =1 ;
@@ -1412,11 +1415,8 @@ ip_vs_out(struct netns_ipvs *ipvs, unsigned int hooknum, struct sk_buff *skb, in cp =INDIRECT_CALL_1 (pp->conn_out_get, ip_vs_conn_out_get_proto, ipvs, af, skb, &iph); -if (likely (cp)) { -if (IP_VS_FWD_METHOD (cp) !=IP_VS_CONN_F_MASQ ) - goto ignore_cp; +if (likely (cp))return handle_response (af, skb, pd, cp, &iph, hooknum); - }/* Check for real-server-started requests */ if (atomic_read (&ipvs->conn_out_counter)) {
@@ -1475,14 +1475,9 @@ ip_vs_out(struct netns_ipvs *ipvs, unsigned int hooknum, struct sk_buff *skb, in } } -out:IP_VS_DBG_PKT (12 , af, pp, skb, iph.off,"ip_vs_out: packet continues traversal as normal" );return NF_ACCEPT ; - -ignore_cp: -__ip_vs_conn_put (cp); - goto out; }/*
Signature · 1 line
2.20.1
Pablo Neira Ayuso
maintainerPATCH 2/6 netfilter: add nf_static_key_{inc,dec}
In reply to Pablo Neira AyusoAdd helper functions increment and decrement the hook static keys.
--- net/netfilter/core.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-)
net/netfilter/core.c +17 -6
@@ -311,6 +311,20 @@ nf_hook_entry_head(struct net *net, int pf, unsigned int hooknum,return NULL ; } +static void nf_static_key_inc (const struct nf_hook_ops *reg,int pf) +{ +#ifdef CONFIG_JUMP_LABEL +static_key_slow_inc (&nf_hooks_needed[pf][reg->hooknum]); +#endif +} + +static void nf_static_key_dec (const struct nf_hook_ops *reg,int pf) +{ +#ifdef CONFIG_JUMP_LABEL +static_key_slow_dec (&nf_hooks_needed[pf][reg->hooknum]); +#endif +} +static int __nf_register_net_hook (struct net *net,int pf,const struct nf_hook_ops *reg) {
@@ -348,9 +362,8 @@ static int __nf_register_net_hook(struct net *net, int pf,if (pf ==NFPROTO_NETDEV && reg->hooknum ==NF_NETDEV_INGRESS )net_inc_ingress_queue ();#endif -#ifdef CONFIG_JUMP_LABEL -static_key_slow_inc (&nf_hooks_needed[pf][reg->hooknum]); -#endif +nf_static_key_inc (reg, pf); +BUG_ON (p == new_hooks);nf_hook_entries_free (p);return 0 ;
@@ -406,9 +419,7 @@ static void __nf_unregister_net_hook(struct net *net, int pf,if (pf ==NFPROTO_NETDEV && reg->hooknum ==NF_NETDEV_INGRESS )net_dec_ingress_queue ();#endif -#ifdef CONFIG_JUMP_LABEL -static_key_slow_dec (&nf_hooks_needed[pf][reg->hooknum]); -#endif +nf_static_key_dec (reg, pf); }else {WARN_ONCE (1 ,"hook not found, pf %d num %d" , pf, reg->hooknum); }
Signature · 1 line
2.20.1
Pablo Neira Ayuso
maintainerPATCH 3/6 netfilter: add nf_ingress_hook() helper function
In reply to Pablo Neira AyusoAdd helper function to check if this is an ingress hook.
--- net/netfilter/core.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
net/netfilter/core.c +7 -2
@@ -311,6 +311,11 @@ nf_hook_entry_head(struct net *net, int pf, unsigned int hooknum,return NULL ; } +static inline bool nf_ingress_hook (const struct nf_hook_ops *reg,int pf) +{ +return pf ==NFPROTO_NETDEV && reg->hooknum ==NF_NETDEV_INGRESS ; +} +static void nf_static_key_inc (const struct nf_hook_ops *reg,int pf) {#ifdef CONFIG_JUMP_LABEL
@@ -359,7 +364,7 @@ static int __nf_register_net_hook(struct net *net, int pf,hooks_validate (new_hooks);#ifdef CONFIG_NETFILTER_INGRESS -if (pf ==NFPROTO_NETDEV && reg->hooknum ==NF_NETDEV_INGRESS ) +if (nf_ingress_hook (reg, pf))net_inc_ingress_queue ();#endif nf_static_key_inc (reg, pf);
@@ -416,7 +421,7 @@ static void __nf_unregister_net_hook(struct net *net, int pf,if (nf_remove_net_hook (p, reg)) {#ifdef CONFIG_NETFILTER_INGRESS -if (pf ==NFPROTO_NETDEV && reg->hooknum ==NF_NETDEV_INGRESS ) +if (nf_ingress_hook (reg, pf))net_dec_ingress_queue ();#endif nf_static_key_dec (reg, pf);
Signature · 1 line
2.20.1
Pablo Neira Ayuso
maintainerPATCH 4/6 netfilter: add inet ingress support
In reply to Pablo Neira AyusoThis patch adds the NF_INET_INGRESS pseudohook for the NFPROTO_INET family. This is a mapping this new hook to the existing NFPROTO_NETDEV and NF_NETDEV_INGRESS hook. The hook does not guarantee that packets are inet only, users must filter out non-ip traffic explicitly.
This infrastructure makes it easier to support this new hook in nf_tables.
--- include/uapi/linux/netfilter.h | 1 + net/netfilter/core.c | 103 ++++++++++++++++++++++++++------- 2 files changed, 83 insertions(+), 21 deletions(-)
include/uapi/linux/netfilter.h +1 -0
@@ -45,6 +45,7 @@ enum nf_inet_hooks {NF_INET_FORWARD ,NF_INET_LOCAL_OUT ,NF_INET_POST_ROUTING , +NF_INET_INGRESS , NF_INET_NUMHOOKS };
net/netfilter/core.c +82 -21
147 lines not on this page · raw message
Signature · 1 line
2.20.1
Pablo Neira Ayuso
maintainerPATCH 5/6 netfilter: nf_tables: add inet ingress support
In reply to Pablo Neira AyusoThis patch adds a new ingress hook for the inet family. The inet ingress hook emulates the IP receive path code, therefore, unclean packets are drop before walking over the ruleset in this basechain.
This patch also introduces the nft_base_chain_netdev() helper function to check if this hook is bound to one or more devices (through the hook list infrastructure). This check allows to perform the same handling for the inet ingress as it would be a netdev ingress chain from the control plane.
--- include/net/netfilter/nf_tables.h | 6 ++++ include/net/netfilter/nf_tables_ipv4.h | 33 ++++++++++++++++++ include/net/netfilter/nf_tables_ipv6.h | 46 ++++++++++++++++++++++++++ net/netfilter/nf_tables_api.c | 14 ++++---- net/netfilter/nft_chain_filter.c | 35 +++++++++++++++++++- 5 files changed, 126 insertions(+), 8 deletions(-)
include/net/netfilter/nf_tables.h +6 -0
@@ -1081,6 +1081,12 @@ struct nft_table {u8 *udata; }; +static inline bool nft_base_chain_netdev (int family,u32 hooknum) +{ +return family ==NFPROTO_NETDEV || + (family ==NFPROTO_INET && hooknum ==NF_INET_INGRESS ); +} +void nft_register_chain_type (const struct nft_chain_type *);void nft_unregister_chain_type (const struct nft_chain_type *);
include/net/netfilter/nf_tables_ipv4.h +33 -0
@@ -53,4 +53,37 @@ static inline void nft_set_pktinfo_ipv4_validate(struct nft_pktinfo *pkt,nft_set_pktinfo_unspec (pkt, skb); } +static inline int nft_set_pktinfo_ipv4_ingress (struct nft_pktinfo *pkt, +struct sk_buff *skb) +{ +struct iphdr *iph; +u32 len, thoff; + +if (!pskb_may_pull (skb,sizeof (*iph))) +return -1 ; + + iph =ip_hdr (skb); +if (iph->ihl <5 || iph->version !=4 ) + goto inhdr_error; + + len =ntohs (iph->tot_len); + thoff = iph->ihl *4 ; +if (skb->len < len) { +__IP_INC_STATS (nft_net (pkt),IPSTATS_MIB_INTRUNCATEDPKTS ); +return -1 ; + }else if (len < thoff) { + goto inhdr_error; + } + + pkt->tprot_set = true; + pkt->tprot = iph->protocol; + pkt->xt.thoff = thoff; + pkt->xt.fragoff =ntohs (iph->frag_off) &IP_OFFSET ; + +return 0 ; + +inhdr_error: +__IP_INC_STATS (nft_net (pkt),IPSTATS_MIB_INHDRERRORS ); +return -1 ; +}#endif
include/net/netfilter/nf_tables_ipv6.h +46 -0
50 lines not on this page · raw message
net/netfilter/nf_tables_api.c +7 -7
@@ -206,7 +206,7 @@ static int nf_tables_register_hook(struct net *net,if (basechain->type->ops_register)return basechain->type->ops_register (net, ops); -if (table->family ==NFPROTO_NETDEV ) +if (nft_base_chain_netdev (table->family, basechain->ops.hooknum))return nft_netdev_register_hooks (net, &basechain->hook_list);return nf_register_net_hook (net, &basechain->ops);
@@ -228,7 +228,7 @@ static void nf_tables_unregister_hook(struct net *net,if (basechain->type->ops_unregister)return basechain->type->ops_unregister (net, ops); -if (table->family ==NFPROTO_NETDEV ) +if (nft_base_chain_netdev (table->family, basechain->ops.hooknum))nft_netdev_unregister_hooks (net, &basechain->hook_list);else nf_unregister_net_hook (net, &basechain->ops);
@@ -1381,7 +1381,7 @@ static int nft_dump_basechain_hook(struct sk_buff *skb, int family,if (nla_put_be32 (skb,NFTA_HOOK_PRIORITY ,htonl (ops->priority))) goto nla_put_failure; -if (family ==NFPROTO_NETDEV ) { +if (nft_base_chain_netdev (family, ops->hooknum)) { nest_devs =nla_nest_start_noflag (skb,NFTA_HOOK_DEVS );list_for_each_entry (hook, &basechain->hook_list, list) {if (!first)
@@ -1685,7 +1685,7 @@ void nf_tables_chain_destroy(struct nft_ctx *ctx)if (nft_is_base_chain (chain)) {struct nft_base_chain *basechain =nft_base_chain (chain); -if (ctx->family ==NFPROTO_NETDEV ) { +if (nft_base_chain_netdev (ctx->family, basechain->ops.hooknum)) {list_for_each_entry_safe (hook, next, &basechain->hook_list, list) {list_del_rcu (&hook->list);
@@ -1877,7 +1877,7 @@ static int nft_chain_parse_hook(struct net *net, hook->type = type;INIT_LIST_HEAD (&hook->list); -if (family ==NFPROTO_NETDEV ) { +if (nft_base_chain_netdev (family, hook->num)) { err =nft_chain_parse_netdev (net, ha, &hook->list);if (err <0 ) {module_put (type->owner);
@@ -1944,7 +1944,7 @@ static int nft_basechain_init(struct nft_base_chain *basechain, u8 family,INIT_LIST_HEAD (&basechain->hook_list); chain = &basechain->chain; -if (family ==NFPROTO_NETDEV ) { +if (nft_base_chain_netdev (family, hook->num)) {list_splice_init (&hook->list, &basechain->hook_list);list_for_each_entry (h, &basechain->hook_list, list)nft_basechain_hook_init (&h->ops, family, hook, chain);
@@ -2168,7 +2168,7 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy,return -EEXIST ; } -if (ctx->family ==NFPROTO_NETDEV ) { +if (nft_base_chain_netdev (ctx->family, hook.num)) {if (!nft_hook_list_equal (&basechain->hook_list, &hook.list)) {nft_chain_release_hook (&hook);
net/netfilter/nft_chain_filter.c +34 -1
@@ -161,16 +161,49 @@ static unsigned int nft_do_chain_inet(void *priv, struct sk_buff *skb,return nft_do_chain (&pkt, priv); } +static unsigned int nft_do_chain_inet_ingress (void *priv,struct sk_buff *skb, +const struct nf_hook_state *state) +{ +struct nf_hook_state ingress_state = *state; +struct nft_pktinfo pkt; + +switch (skb->protocol) { +case htons (ETH_P_IP ): +/* Original hook is NFPROTO_NETDEV and NF_NETDEV_INGRESS. */ + ingress_state.pf =NFPROTO_IPV4 ; + ingress_state.hook =NF_INET_INGRESS ; +nft_set_pktinfo (&pkt, skb, &ingress_state); + +if (nft_set_pktinfo_ipv4_ingress (&pkt, skb) <0 ) +return NF_DROP ; +break ; +case htons (ETH_P_IPV6 ): + ingress_state.pf =NFPROTO_IPV6 ; + ingress_state.hook =NF_INET_INGRESS ; +nft_set_pktinfo (&pkt, skb, &ingress_state); + +if (nft_set_pktinfo_ipv6_ingress (&pkt, skb) <0 ) +return NF_DROP ; +break ; +default : +return NF_ACCEPT ; + } + +return nft_do_chain (&pkt, priv); +} +static const struct nft_chain_type nft_chain_filter_inet = { .name ="filter" , .type =NFT_CHAIN_T_DEFAULT , .family =NFPROTO_INET , - .hook_mask = (1 <<NF_INET_LOCAL_IN ) | + .hook_mask = (1 <<NF_INET_INGRESS ) | + (1 <<NF_INET_LOCAL_IN ) | (1 <<NF_INET_LOCAL_OUT ) | (1 <<NF_INET_FORWARD ) | (1 <<NF_INET_PRE_ROUTING ) | (1 <<NF_INET_POST_ROUTING ), .hooks = { + [NF_INET_INGRESS ] = nft_do_chain_inet_ingress, [NF_INET_LOCAL_IN ] = nft_do_chain_inet, [NF_INET_LOCAL_OUT ] = nft_do_chain_inet, [NF_INET_FORWARD ] = nft_do_chain_inet,
Signature · 1 line
2.20.1
Pablo Neira Ayuso
maintainerPATCH 6/6 netfilter: flowtable: reduce calls to pskb_may_pull()
In reply to Pablo Neira AyusoMake two unfront calls to pskb_may_pull() to linearize the network and transport header.
--- net/netfilter/nf_flow_table_core.c | 12 +++----- net/netfilter/nf_flow_table_ip.c | 45 +++++++++++++++++------------- 2 files changed, 30 insertions(+), 27 deletions(-)
net/netfilter/nf_flow_table_core.c +4 -8
@@ -395,8 +395,7 @@ static int nf_flow_nat_port_tcp(struct sk_buff *skb, unsigned int thoff, {struct tcphdr *tcph; -if (!pskb_may_pull (skb, thoff +sizeof (*tcph)) || -skb_try_make_writable (skb, thoff +sizeof (*tcph))) +if (skb_try_make_writable (skb, thoff +sizeof (*tcph)))return -1 ; tcph = (void *)(skb_network_header (skb) + thoff);
@@ -410,8 +409,7 @@ static int nf_flow_nat_port_udp(struct sk_buff *skb, unsigned int thoff, {struct udphdr *udph; -if (!pskb_may_pull (skb, thoff +sizeof (*udph)) || -skb_try_make_writable (skb, thoff +sizeof (*udph))) +if (skb_try_make_writable (skb, thoff +sizeof (*udph)))return -1 ; udph = (void *)(skb_network_header (skb) + thoff);
@@ -449,8 +447,7 @@ int nf_flow_snat_port(const struct flow_offload *flow,struct flow_ports *hdr;__be16 port, new_port; -if (!pskb_may_pull (skb, thoff +sizeof (*hdr)) || -skb_try_make_writable (skb, thoff +sizeof (*hdr))) +if (skb_try_make_writable (skb, thoff +sizeof (*hdr)))return -1 ; hdr = (void *)(skb_network_header (skb) + thoff);
@@ -481,8 +478,7 @@ int nf_flow_dnat_port(const struct flow_offload *flow,struct flow_ports *hdr;__be16 port, new_port; -if (!pskb_may_pull (skb, thoff +sizeof (*hdr)) || -skb_try_make_writable (skb, thoff +sizeof (*hdr))) +if (skb_try_make_writable (skb, thoff +sizeof (*hdr)))return -1 ; hdr = (void *)(skb_network_header (skb) + thoff);
net/netfilter/nf_flow_table_ip.c +26 -19
114 lines not on this page · raw message
Signature · 1 line
2.20.1
Jakub Kicinski
maintainerQuoted 8 lines · On Mon, 12 Oct 2020 03:38:13 +0200 Pablo Neira Ayuso wrote:
Hi,
The following patchset contains Netfilter/IPVS updates for net-next:
1) Inspect the reply packets coming from DR/TUN and refresh connection state and timeout, from longguang yue and Julian Anastasov.
2) Series to add support for the inet ingress chain type in
nf_tables.
Pulled, thanks!