Skip to contentSkip to conversation

Passing net through the netfilter hooks

PATCH In review Eric W. Biederman 9 participants 333 messages Sep 16, 2015 – Oct 13, 2015 Page 1 of 3

Subsystems (15)
lore
Share & tools
lore
#0 lore raw

cover Passing net through the netfilter hooks

My primary goal with this patchset and it's follow ups is to cleanup the network routing paths so that we do not look at the output device to derive the network namespace. My plan is to pass the network namespace of the transmitting socket through the output path, to replace code that looks at the output network device today. Once that is done we can have routes with output devices outside of the current network namespace. Which should allow reception and transmission of packets in network namespaces to be as fast as normal packet reception and transmission with early demux disabled, because it will same code path.

Once skb_dst(skb)->dev is a little better under control I think it will also be possible to use rcu to cleanup the ancient hack that sets dst->dev to loopback_dev when a network device is removed.

The work to get there is a series of code cleanups. I am starting with passing net into the netfilter hooks and into the functions that are called after the netfilter hooks. This removes from netfilter the need to guess which network namespace it is working on.

To get there I perform a series of minor prep patches so the big changes at the end are possible to audit without getting lost in the noise. In particular I have a lot of patches computing net into a local variable and then using it through out the function.

So this patchset encompases removing dead code, sorting out the _sk functions that were added last time someone pushed a prototype change through the post netfilter functions. Cleaning up individual functions use of the network namespace. Passing net into the netfilter hooks. Passing net into the post netfilter functions. Using state->net in the netfilter code where it is available and trivially usable.

Pablo, Dave I don't know whose tree this makes more sense to go through. I am assuming at least initially Pablos as netfilter is involved. From what I have seen there will be a lot of back and forth between the netfilter code paths and the routing code paths.

The patches are also available (against 4.3-rc1) at:
git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/net-next.git master

Eric W. Biederman (30):
netfilter: Remove !CONFIG_NETFITLER definition of nf_hook_thresh
netfilter: Store net in nf_hook_state
netfilter: Pass net to nf_hook_thresh
xfrm: Remove unused afinfo method init_dst
net: Merge dst_output and dst_output_sk
ipv4: Compute net once in ip_forward
ipv4: Compute net once in ip_forward_finish
ipv4: Compute net once in ip_rcv
ipv4: Remember the net in ip_output and ip_mc_output
ipv4: Don't recompute net in ipmr_queue_xmit
ipv4: Only compute net once in ip_do_fragment
ipv4: Explicitly compute net in ip_fragment
ipv4: Only compute net once in ip_finish_output2
ipv4: Only compute net once in ip_rcv_finish
ipv4: Only compute net once in ipmr_forward_finish
ipv6: Only compute net once in ip6mr_forward2_finish
arp: Introduce arp_xmit_finish
bridge: Introduce br_send_bpdu_finish
net: Remove dev_queue_xmit_sk
ipv6: Don't recompute net in ip6_rcv
ipv6: Only compute net once in ip6_finish_output2
ipv6: Cache net in ip6_output
ipv6: Compute net once in raw6_send_hdrinc
bridge: Pass net into br_nf_ip_fragment
bridge: Pass net into br_nf_push_frag_xmit
bridge: Cache net in br_nf_pre_routing_finish
bridge: Add br_netif_receive_skb remove netif_receive_skb_sk
netfilter: Pass struct net into the netfilter hooks
netfilter: Use nf_hook_state.net
netfilter: Pass net into okfn

 drivers/net/vrf.c                              |  9 ++--
 include/linux/netdevice.h                      | 14 ++----
 include/linux/netfilter.h                      | 68 ++++++++++++--------------
 include/linux/netfilter_bridge.h               |  2 +-
 include/linux/netfilter_ingress.h              |  2 +-
 include/net/dn_neigh.h                         |  6 +--
 include/net/dst.h                              |  6 +--
 include/net/ipv6.h                             |  2 +-
 include/net/netfilter/br_netfilter.h           |  2 +-
 include/net/xfrm.h                             |  2 -
 net/bridge/br_forward.c                        | 16 +++---
 net/bridge/br_input.c                          | 25 ++++++----
 net/bridge/br_multicast.c                      |  4 +-
 net/bridge/br_netfilter_hooks.c                | 54 ++++++++++----------
 net/bridge/br_netfilter_ipv6.c                 |  8 +--
 net/bridge/br_private.h                        |  6 +--
 net/bridge/br_stp_bpdu.c                       | 12 +++--
 net/bridge/netfilter/ebtable_filter.c          |  4 +-
 net/bridge/netfilter/ebtable_nat.c             |  4 +-
 net/core/dev.c                                 | 12 +++--
 net/decnet/dn_neigh.c                          | 23 +++++----
 net/decnet/dn_nsp_in.c                         |  7 +--
 net/decnet/dn_nsp_out.c                        |  4 +-
 net/decnet/dn_route.c                          | 30 ++++++------
 net/ipv4/arp.c                                 | 20 +++++---
 net/ipv4/ip_forward.c                          | 19 ++++---
 net/ipv4/ip_input.c                            | 39 +++++++--------
 net/ipv4/ip_output.c                           | 58 ++++++++++++----------
 net/ipv4/ip_vti.c                              |  2 +-
 net/ipv4/ipmr.c                                | 15 +++---
 net/ipv4/netfilter/arptable_filter.c           |  4 +-
 net/ipv4/netfilter/ip_tables.c                 |  8 +--
 net/ipv4/netfilter/ipt_CLUSTERIP.c             |  2 +-
 net/ipv4/netfilter/ipt_SYNPROXY.c              |  2 +-
 net/ipv4/netfilter/iptable_filter.c            |  6 +--
 net/ipv4/netfilter/iptable_mangle.c            |  7 ++-
 net/ipv4/netfilter/iptable_nat.c               |  5 +-
 net/ipv4/netfilter/iptable_raw.c               |  6 +--
 net/ipv4/netfilter/iptable_security.c          |  5 +-
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c |  4 +-
 net/ipv4/raw.c                                 |  5 +-
 net/ipv4/xfrm4_input.c                         |  7 +--
 net/ipv4/xfrm4_output.c                        | 10 ++--
 net/ipv6/ip6_input.c                           | 15 +++---
 net/ipv6/ip6_output.c                          | 37 +++++++-------
 net/ipv6/ip6_vti.c                             |  2 +-
 net/ipv6/ip6mr.c                               | 12 ++---
 net/ipv6/mcast.c                               |  9 ++--
 net/ipv6/ndisc.c                               |  6 +--
 net/ipv6/netfilter/ip6_tables.c                |  8 +--
 net/ipv6/netfilter/ip6t_SYNPROXY.c             |  2 +-
 net/ipv6/netfilter/ip6table_filter.c           |  5 +-
 net/ipv6/netfilter/ip6table_mangle.c           |  6 +--
 net/ipv6/netfilter/ip6table_nat.c              |  5 +-
 net/ipv6/netfilter/ip6table_raw.c              |  5 +-
 net/ipv6/netfilter/ip6table_security.c         |  4 +-
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c |  4 +-
 net/ipv6/netfilter/nf_defrag_ipv6_hooks.c      |  2 +-
 net/ipv6/output_core.c                         |  8 +--
 net/ipv6/raw.c                                 |  9 ++--
 net/ipv6/xfrm6_input.c                         |  4 +-
 net/ipv6/xfrm6_output.c                        | 10 ++--
 net/netfilter/ipvs/ip_vs_xmit.c                |  8 +--
 net/netfilter/nf_queue.c                       |  2 +-
 net/netfilter/nfnetlink_queue_core.c           |  3 +-
 net/xfrm/xfrm_output.c                         | 17 ++++---
 net/xfrm/xfrm_policy.c                         |  4 +-
 67 files changed, 375 insertions(+), 358 deletions(-)
#1 lore raw

PATCH 1/30 netfilter: Remove !CONFIG_NETFITLER definition of nf_hook_thresh

The !CONFIG_NETFILTER definition of nf_hook_thresh calls okfn when the CONFIG_NETFITLER defintion does not, making it buggy.

As the !CONFIG_NETFILTER defintion of nf_hook_thresh is not used remove it rather than fix it.

---
 include/linux/netfilter.h | 9 ---------
 1 file changed, 9 deletions(-)
include/linux/netfilter.h +0 -9
index 36a652531791..1abac85ec907 100644
@@ -344,15 +344,6 @@ nf_nat_decode_session(struct sk_buff *skb, struct flowi *fl, u_int8_t family)
 #else /* !CONFIG_NETFILTER */
 #define NF_HOOK(pf, hook, sk, skb, indev, outdev, okfn) (okfn)(sk, skb)
 #define NF_HOOK_COND(pf, hook, sk, skb, indev, outdev, okfn, cond) (okfn)(sk, skb)
-static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook,
-				 struct sock *sk,
-				 struct sk_buff *skb,
-				 struct net_device *indev,
-				 struct net_device *outdev,
-				 int (*okfn)(struct sock *sk, struct sk_buff *), int thresh)
-{
-	return okfn(sk, skb);
-}
 static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sock *sk,
 			  struct sk_buff *skb, struct net_device *indev,
 			  struct net_device *outdev,
Signature · 1 line
2.2.1
#2 lore raw

PATCH 2/30 netfilter: Store net in nf_hook_state

In reply to Eric W. Biederman
---
 include/linux/netfilter.h         | 5 ++++-
 include/linux/netfilter_ingress.h | 2 +-
 2 files changed, 5 insertions(+), 2 deletions(-)
include/linux/netfilter.h +4 -1
index 1abac85ec907..889ac0e11f01 100644
@@ -54,6 +54,7 @@ struct nf_hook_state {
 	struct net_device *in;
 	struct net_device *out;
 	struct sock *sk;
+	struct net *net;
 	struct list_head *hook_list;
 	int (*okfn)(struct sock *, struct sk_buff *);
 };
@@ -65,6 +66,7 @@ static inline void nf_hook_state_init(struct nf_hook_state *p,
 				      struct net_device *indev,
 				      struct net_device *outdev,
 				      struct sock *sk,
+				      struct net *net,
 				      int (*okfn)(struct sock *, struct sk_buff *))
 {
 	p->hook = hook;
@@ -73,6 +75,7 @@ static inline void nf_hook_state_init(struct nf_hook_state *p,
 	p->in = indev;
 	p->out = outdev;
 	p->sk = sk;
+	p->net = net;
 	p->hook_list = hook_list;
 	p->okfn = okfn;
 }
@@ -181,7 +184,7 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook,
 		struct nf_hook_state state;
 
 		nf_hook_state_init(&state, hook_list, hook, thresh,
-				   pf, indev, outdev, sk, okfn);
+				   pf, indev, outdev, sk, net, okfn);
 		return nf_hook_slow(skb, &state);
 	}
 	return 1;
include/linux/netfilter_ingress.h +1 -1
index cb0727fe2b3d..187feabe557c 100644
@@ -17,7 +17,7 @@ static inline int nf_hook_ingress(struct sk_buff *skb)
 
 	nf_hook_state_init(&state, &skb->dev->nf_hooks_ingress,
 			   NF_NETDEV_INGRESS, INT_MIN, NFPROTO_NETDEV, NULL,
-			   skb->dev, NULL, NULL);
+			   skb->dev, NULL, dev_net(skb->dev), NULL);
 	return nf_hook_slow(skb, &state);
 }
 
Signature · 1 line
2.2.1
#3 lore raw

PATCH 3/30 netfilter: Pass net to nf_hook_thresh

In reply to Eric W. Biederman
---
 include/linux/netfilter.h | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
include/linux/netfilter.h +7 -4
index 889ac0e11f01..042148dc1e22 100644
@@ -170,6 +170,7 @@ int nf_hook_slow(struct sk_buff *skb, struct nf_hook_state *state);
  *	value indicates the packet has been consumed by the hook.
  */
 static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook,
+				 struct net *net,
 				 struct sock *sk,
 				 struct sk_buff *skb,
 				 struct net_device *indev,
@@ -177,7 +178,6 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook,
 				 int (*okfn)(struct sock *, struct sk_buff *),
 				 int thresh)
 {
-	struct net *net = dev_net(indev ? indev : outdev);
 	struct list_head *hook_list = &net->nf.hooks[pf][hook];
 
 	if (nf_hook_list_active(hook_list, pf, hook)) {
@@ -195,7 +195,8 @@ static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sock *sk,
 			  struct net_device *outdev,
 			  int (*okfn)(struct sock *, struct sk_buff *))
 {
-	return nf_hook_thresh(pf, hook, sk, skb, indev, outdev, okfn, INT_MIN);
+	struct net *net = dev_net(indev ? indev : outdev);
+	return nf_hook_thresh(pf, hook, net, sk, skb, indev, outdev, okfn, INT_MIN);
 }
                    
 /* Activate hook; either okfn or kfree_skb called, unless a hook
@@ -221,7 +222,8 @@ NF_HOOK_THRESH(uint8_t pf, unsigned int hook, struct sock *sk,
 	       struct net_device *out,
 	       int (*okfn)(struct sock *, struct sk_buff *), int thresh)
 {
-	int ret = nf_hook_thresh(pf, hook, sk, skb, in, out, okfn, thresh);
+	struct net *net = dev_net(in ? in : out);
+	int ret = nf_hook_thresh(pf, hook, net, sk, skb, in, out, okfn, thresh);
 	if (ret == 1)
 		ret = okfn(sk, skb);
 	return ret;
@@ -232,10 +234,11 @@ NF_HOOK_COND(uint8_t pf, unsigned int hook, struct sock *sk,
 	     struct sk_buff *skb, struct net_device *in, struct net_device *out,
 	     int (*okfn)(struct sock *, struct sk_buff *), bool cond)
 {
+	struct net *net = dev_net(in ? in : out);
 	int ret;
 
 	if (!cond ||
-	    ((ret = nf_hook_thresh(pf, hook, sk, skb, in, out, okfn, INT_MIN)) == 1))
+	    ((ret = nf_hook_thresh(pf, hook, net, sk, skb, in, out, okfn, INT_MIN)) == 1))
 		ret = okfn(sk, skb);
 	return ret;
 }
Signature · 1 line
2.2.1
#4 lore raw

PATCH 4/30 xfrm: Remove unused afinfo method init_dst

In reply to Eric W. Biederman
---
 include/net/xfrm.h     | 2 --
 net/xfrm/xfrm_policy.c | 2 --
 2 files changed, 4 deletions(-)
include/net/xfrm.h +0 -2
index 312e3fee9ccf..fd176106909a 100644
@@ -296,8 +296,6 @@ struct xfrm_policy_afinfo {
 						  struct flowi *fl,
 						  int reverse);
 	int			(*get_tos)(const struct flowi *fl);
-	void			(*init_dst)(struct net *net,
-					    struct xfrm_dst *dst);
 	int			(*init_path)(struct xfrm_dst *path,
 					     struct dst_entry *dst,
 					     int nfheader_len);
net/xfrm/xfrm_policy.c +0 -2
index 94af3d065785..6b5d6e2b9a49 100644
@@ -1583,8 +1583,6 @@ static inline struct xfrm_dst *xfrm_alloc_dst(struct net *net, int family)
 
 		memset(dst + 1, 0, sizeof(*xdst) - sizeof(*dst));
 		xdst->flo.ops = &xfrm_bundle_fc_ops;
-		if (afinfo->init_dst)
-			afinfo->init_dst(net, xdst);
 	} else
 		xdst = ERR_PTR(-ENOBUFS);
 
Signature · 1 line
2.2.1
#5 lore raw

PATCH 5/30 net: Merge dst_output and dst_output_sk

In reply to Eric W. Biederman

Add a sock paramter to dst_output making dst_output_sk superfluous. Add a skb->sk parameter to all of the callers of dst_output Have the callers of dst_output_sk call dst_output.

---
 include/net/dst.h               | 6 +-----
 net/decnet/dn_nsp_out.c         | 4 ++--
 net/ipv4/ip_forward.c           | 2 +-
 net/ipv4/ip_output.c            | 6 +++---
 net/ipv4/ip_vti.c               | 2 +-
 net/ipv4/ipmr.c                 | 2 +-
 net/ipv4/raw.c                  | 2 +-
 net/ipv4/xfrm4_output.c         | 2 +-
 net/ipv6/ip6_output.c           | 4 ++--
 net/ipv6/ip6_vti.c              | 2 +-
 net/ipv6/ip6mr.c                | 2 +-
 net/ipv6/mcast.c                | 4 ++--
 net/ipv6/ndisc.c                | 2 +-
 net/ipv6/output_core.c          | 4 ++--
 net/ipv6/raw.c                  | 2 +-
 net/ipv6/xfrm6_output.c         | 2 +-
 net/netfilter/ipvs/ip_vs_xmit.c | 4 ++--
 net/xfrm/xfrm_output.c          | 2 +-
 net/xfrm/xfrm_policy.c          | 2 +-
 19 files changed, 26 insertions(+), 30 deletions(-)
include/net/dst.h +1 -5
index 9261d928303d..c72e58474e52 100644
@@ -454,14 +454,10 @@ static inline void dst_set_expires(struct dst_entry *dst, int timeout)
 }
 
 /* Output packet to network from transport.  */
-static inline int dst_output_sk(struct sock *sk, struct sk_buff *skb)
+static inline int dst_output(struct sock *sk, struct sk_buff *skb)
 {
 	return skb_dst(skb)->output(sk, skb);
 }
-static inline int dst_output(struct sk_buff *skb)
-{
-	return dst_output_sk(skb->sk, skb);
-}
 
 /* Input packet from network to transport.  */
 static inline int dst_input(struct sk_buff *skb)
net/decnet/dn_nsp_out.c +2 -2
index 1aaa51ebbda6..4b02dd300f50 100644
@@ -85,7 +85,7 @@ static void dn_nsp_send(struct sk_buff *skb)
 	if (dst) {
 try_again:
 		skb_dst_set(skb, dst);
-		dst_output(skb);
+		dst_output(skb->sk, skb);
 		return;
 	}
 
@@ -582,7 +582,7 @@ static __inline__ void dn_nsp_do_disc(struct sock *sk, unsigned char msgflg,
 	 * associations.
 	 */
 	skb_dst_set(skb, dst_clone(dst));
-	dst_output(skb);
+	dst_output(skb->sk, skb);
 }
 
 
net/ipv4/ip_forward.c +1 -1
index 2d3aa408fbdc..28fb90108f56 100644
@@ -72,7 +72,7 @@ static int ip_forward_finish(struct sock *sk, struct sk_buff *skb)
 		ip_forward_options(skb);
 
 	skb_sender_cpu_clear(skb);
-	return dst_output_sk(sk, skb);
+	return dst_output(sk, skb);
 }
 
 int ip_forward(struct sk_buff *skb)
net/ipv4/ip_output.c +3 -3
index 0138fada0951..f076f11aa94a 100644
@@ -102,7 +102,7 @@ static int __ip_local_out_sk(struct sock *sk, struct sk_buff *skb)
 	iph->tot_len = htons(skb->len);
 	ip_send_check(iph);
 	return nf_hook(NFPROTO_IPV4, NF_INET_LOCAL_OUT, sk, skb, NULL,
-		       skb_dst(skb)->dev, dst_output_sk);
+		       skb_dst(skb)->dev, dst_output);
 }
 
 int __ip_local_out(struct sk_buff *skb)
@@ -116,7 +116,7 @@ int ip_local_out_sk(struct sock *sk, struct sk_buff *skb)
 
 	err = __ip_local_out(skb);
 	if (likely(err == 1))
-		err = dst_output_sk(sk, skb);
+		err = dst_output(sk, skb);
 
 	return err;
 }
@@ -271,7 +271,7 @@ static int ip_finish_output(struct sock *sk, struct sk_buff *skb)
 	/* Policy lookup after SNAT yielded a new policy */
 	if (skb_dst(skb)->xfrm) {
 		IPCB(skb)->flags |= IPSKB_REROUTED;
-		return dst_output_sk(sk, skb);
+		return dst_output(sk, skb);
 	}
 #endif
 	mtu = ip_skb_dst_mtu(skb);
net/ipv4/ip_vti.c +1 -1
index 0c152087ca15..3b87ec5178f9 100644
@@ -197,7 +197,7 @@ static netdev_tx_t vti_xmit(struct sk_buff *skb, struct net_device *dev,
 	skb_dst_set(skb, dst);
 	skb->dev = skb_dst(skb)->dev;
 
-	err = dst_output(skb);
+	err = dst_output(skb->sk, skb);
 	if (net_xmit_eval(err) == 0)
 		err = skb->len;
 	iptunnel_xmit_stats(err, &dev->stats, dev->tstats);
net/ipv4/ipmr.c +1 -1
index 866ee89f5254..a0a5def920fc 100644
@@ -1688,7 +1688,7 @@ static inline int ipmr_forward_finish(struct sock *sk, struct sk_buff *skb)
 	if (unlikely(opt->optlen))
 		ip_forward_options(skb);
 
-	return dst_output_sk(sk, skb);
+	return dst_output(sk, skb);
 }
 
 /*
net/ipv4/raw.c +1 -1
index 561cd4b8fc6e..09ab5bb6913a 100644
@@ -412,7 +412,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
 			skb_transport_header(skb))->type);
 
 	err = NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_OUT, sk, skb,
-		      NULL, rt->dst.dev, dst_output_sk);
+		      NULL, rt->dst.dev, dst_output);
 	if (err > 0)
 		err = net_xmit_errno(err);
 	if (err)
net/ipv4/xfrm4_output.c +1 -1
index 2878dbfffeb7..737f5e395a6a 100644
@@ -87,7 +87,7 @@ static int __xfrm4_output(struct sock *sk, struct sk_buff *skb)
 #ifdef CONFIG_NETFILTER
 	if (!x) {
 		IPCB(skb)->flags |= IPSKB_REROUTED;
-		return dst_output_sk(sk, skb);
+		return dst_output(sk, skb);
 	}
 #endif
 
net/ipv6/ip6_output.c +2 -2
index 26ea47930740..a80502c64523 100644
@@ -225,7 +225,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6,
 		IP6_UPD_PO_STATS(net, ip6_dst_idev(skb_dst(skb)),
 			      IPSTATS_MIB_OUT, skb->len);
 		return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
-			       NULL, dst->dev, dst_output_sk);
+			       NULL, dst->dev, dst_output);
 	}
 
 	skb->dev = dst->dev;
@@ -320,7 +320,7 @@ static int ip6_forward_proxy_check(struct sk_buff *skb)
 static inline int ip6_forward_finish(struct sock *sk, struct sk_buff *skb)
 {
 	skb_sender_cpu_clear(skb);
-	return dst_output_sk(sk, skb);
+	return dst_output(sk, skb);
 }
 
 static unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst)
net/ipv6/ip6_vti.c +1 -1
index 0224c032dca5..f96f1c19b4a8 100644
@@ -482,7 +482,7 @@ vti6_xmit(struct sk_buff *skb, struct net_device *dev, struct flowi *fl)
 		return -EMSGSIZE;
 	}
 
-	err = dst_output(skb);
+	err = dst_output(skb->sk, skb);
 	if (net_xmit_eval(err) == 0) {
 		struct pcpu_sw_netstats *tstats = this_cpu_ptr(dev->tstats);
 
net/ipv6/ip6mr.c +1 -1
index 0e004cc42a22..e95f6b6281de 100644
@@ -1991,7 +1991,7 @@ static inline int ip6mr_forward2_finish(struct sock *sk, struct sk_buff *skb)
 			 IPSTATS_MIB_OUTFORWDATAGRAMS);
 	IP6_ADD_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)),
 			 IPSTATS_MIB_OUTOCTETS, skb->len);
-	return dst_output_sk(sk, skb);
+	return dst_output(sk, skb);
 }
 
 /*
net/ipv6/mcast.c +2 -2
index 083b2927fc67..5b3f2841acf6 100644
@@ -1646,7 +1646,7 @@ static void mld_sendpack(struct sk_buff *skb)
 
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
 		      net->ipv6.igmp_sk, skb, NULL, skb->dev,
-		      dst_output_sk);
+		      dst_output);
 out:
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(net, idev, ICMPV6_MLD2_REPORT);
@@ -2009,7 +2009,7 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
 
 	skb_dst_set(skb, dst);
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
-		      NULL, skb->dev, dst_output_sk);
+		      NULL, skb->dev, dst_output);
 out:
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(net, idev, type);
net/ipv6/ndisc.c +1 -1
index 64a71354b069..349ac1b022b6 100644
@@ -465,7 +465,7 @@ static void ndisc_send_skb(struct sk_buff *skb,
 
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
 		      NULL, dst->dev,
-		      dst_output_sk);
+		      dst_output);
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(net, idev, type);
 		ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTMSGS);
net/ipv6/output_core.c +2 -2
index 928a0fb0b744..8178f72fe90d 100644
@@ -149,7 +149,7 @@ static int __ip6_local_out_sk(struct sock *sk, struct sk_buff *skb)
 	IP6CB(skb)->nhoff = offsetof(struct ipv6hdr, nexthdr);
 
 	return nf_hook(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
-		       NULL, skb_dst(skb)->dev, dst_output_sk);
+		       NULL, skb_dst(skb)->dev, dst_output);
 }
 
 int __ip6_local_out(struct sk_buff *skb)
@@ -164,7 +164,7 @@ int ip6_local_out_sk(struct sock *sk, struct sk_buff *skb)
 
 	err = __ip6_local_out_sk(sk, skb);
 	if (likely(err == 1))
-		err = dst_output_sk(sk, skb);
+		err = dst_output(sk, skb);
 
 	return err;
 }
net/ipv6/raw.c +1 -1
index fdbada1569a3..1636537705f5 100644
@@ -654,7 +654,7 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length,
 
 	IP6_UPD_PO_STATS(sock_net(sk), rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len);
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
-		      NULL, rt->dst.dev, dst_output_sk);
+		      NULL, rt->dst.dev, dst_output);
 	if (err > 0)
 		err = net_xmit_errno(err);
 	if (err)
net/ipv6/xfrm6_output.c +1 -1
index 09c76a7b474d..b0fcd6c09837 100644
@@ -140,7 +140,7 @@ static int __xfrm6_output(struct sock *sk, struct sk_buff *skb)
 #ifdef CONFIG_NETFILTER
 	if (!x) {
 		IP6CB(skb)->flags |= IP6SKB_REROUTED;
-		return dst_output_sk(sk, skb);
+		return dst_output(sk, skb);
 	}
 #endif
 
net/netfilter/ipvs/ip_vs_xmit.c +2 -2
index 258a0b0e82a2..b8e5544af87f 100644
@@ -574,7 +574,7 @@ static inline int ip_vs_nat_send_or_cont(int pf, struct sk_buff *skb,
 		if (!skb->sk)
 			skb_sender_cpu_clear(skb);
 		NF_HOOK(pf, NF_INET_LOCAL_OUT, NULL, skb,
-			NULL, skb_dst(skb)->dev, dst_output_sk);
+			NULL, skb_dst(skb)->dev, dst_output);
 	} else
 		ret = NF_ACCEPT;
 
@@ -596,7 +596,7 @@ static inline int ip_vs_send_or_cont(int pf, struct sk_buff *skb,
 		if (!skb->sk)
 			skb_sender_cpu_clear(skb);
 		NF_HOOK(pf, NF_INET_LOCAL_OUT, NULL, skb,
-			NULL, skb_dst(skb)->dev, dst_output_sk);
+			NULL, skb_dst(skb)->dev, dst_output);
 	} else
 		ret = NF_ACCEPT;
 	return ret;
net/xfrm/xfrm_output.c +1 -1
index 68ada2ca4b60..57a50f6ce28c 100644
@@ -139,7 +139,7 @@ int xfrm_output_resume(struct sk_buff *skb, int err)
 			goto out;
 
 		if (!skb_dst(skb)->xfrm)
-			return dst_output(skb);
+			return dst_output(skb->sk, skb);
 
 		err = nf_hook(skb_dst(skb)->ops->family,
 			      NF_INET_POST_ROUTING, skb->sk, skb,
net/xfrm/xfrm_policy.c +1 -1
index 6b5d6e2b9a49..e7f64bcb78a8 100644
@@ -1945,7 +1945,7 @@ static void xfrm_policy_queue_process(unsigned long arg)
 		skb_dst_drop(skb);
 		skb_dst_set(skb, dst);
 
-		dst_output(skb);
+		dst_output(skb->sk, skb);
 	}
 
 out:
Signature · 1 line
2.2.1
#6 lore raw

PATCH 6/30 ipv4: Compute net once in ip_forward

In reply to Eric W. Biederman

Compute struct net from the input device in ip_forward before it is used.

---
 net/ipv4/ip_forward.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
net/ipv4/ip_forward.c +4 -2
index 28fb90108f56..ba2f66b3b3f6 100644
@@ -81,6 +81,7 @@ int ip_forward(struct sk_buff *skb)
 	struct iphdr *iph;	/* Our header */
 	struct rtable *rt;	/* Route we use */
 	struct ip_options *opt	= &(IPCB(skb)->opt);
+	struct net *net;
 
 	/* that should never happen */
 	if (skb->pkt_type != PACKET_HOST)
@@ -99,6 +100,7 @@ int ip_forward(struct sk_buff *skb)
 		return NET_RX_SUCCESS;
 
 	skb_forward_csum(skb);
+	net = dev_net(skb->dev);
 
 	/*
 	 *	According to the RFC, we must first decrease the TTL field. If
@@ -119,7 +121,7 @@ int ip_forward(struct sk_buff *skb)
 	IPCB(skb)->flags |= IPSKB_FORWARDED;
 	mtu = ip_dst_mtu_maybe_forward(&rt->dst, true);
 	if (ip_exceeds_mtu(skb, mtu)) {
-		IP_INC_STATS(dev_net(rt->dst.dev), IPSTATS_MIB_FRAGFAILS);
+		IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS);
 		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
 			  htonl(mtu));
 		goto drop;
@@ -155,7 +157,7 @@ sr_failed:
 
 too_many_hops:
 	/* Tell the sender its packet died... */
-	IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_INHDRERRORS);
+	IP_INC_STATS_BH(net, IPSTATS_MIB_INHDRERRORS);
 	icmp_send(skb, ICMP_TIME_EXCEEDED, ICMP_EXC_TTL, 0);
 drop:
 	kfree_skb(skb);
Signature · 1 line
2.2.1
#7 lore raw

PATCH 7/30 ipv4: Compute net once in ip_forward_finish

In reply to Eric W. Biederman
---
 net/ipv4/ip_forward.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/ipv4/ip_forward.c +3 -2
index ba2f66b3b3f6..95235c813f18 100644
@@ -63,10 +63,11 @@ static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu)
 
 static int ip_forward_finish(struct sock *sk, struct sk_buff *skb)
 {
+	struct net *net = dev_net(skb_dst(skb)->dev);
 	struct ip_options *opt	= &(IPCB(skb)->opt);
 
-	IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTFORWDATAGRAMS);
-	IP_ADD_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTOCTETS, skb->len);
+	IP_INC_STATS_BH(net, IPSTATS_MIB_OUTFORWDATAGRAMS);
+	IP_ADD_STATS_BH(net, IPSTATS_MIB_OUTOCTETS, skb->len);
 
 	if (unlikely(opt->optlen))
 		ip_forward_options(skb);
Signature · 1 line
2.2.1
#8 lore raw

PATCH 8/30 ipv4: Compute net once in ip_rcv

In reply to Eric W. Biederman
---
 net/ipv4/ip_input.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)
net/ipv4/ip_input.c +9 -7
index f4fc8a77aaa7..ff908863f22e 100644
@@ -378,6 +378,7 @@ drop:
 int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev)
 {
 	const struct iphdr *iph;
+	struct net *net;
 	u32 len;
 
 	/* When the interface is in promisc. mode, drop all the crap
@@ -387,11 +388,12 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 		goto drop;
 
 
-	IP_UPD_PO_STATS_BH(dev_net(dev), IPSTATS_MIB_IN, skb->len);
+	net = dev_net(dev);
+	IP_UPD_PO_STATS_BH(net, IPSTATS_MIB_IN, skb->len);
 
 	skb = skb_share_check(skb, GFP_ATOMIC);
 	if (!skb) {
-		IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INDISCARDS);
+		IP_INC_STATS_BH(net, IPSTATS_MIB_INDISCARDS);
 		goto out;
 	}
 
@@ -417,7 +419,7 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 	BUILD_BUG_ON(IPSTATS_MIB_ECT1PKTS != IPSTATS_MIB_NOECTPKTS + INET_ECN_ECT_1);
 	BUILD_BUG_ON(IPSTATS_MIB_ECT0PKTS != IPSTATS_MIB_NOECTPKTS + INET_ECN_ECT_0);
 	BUILD_BUG_ON(IPSTATS_MIB_CEPKTS != IPSTATS_MIB_NOECTPKTS + INET_ECN_CE);
-	IP_ADD_STATS_BH(dev_net(dev),
+	IP_ADD_STATS_BH(net,
 			IPSTATS_MIB_NOECTPKTS + (iph->tos & INET_ECN_MASK),
 			max_t(unsigned short, 1, skb_shinfo(skb)->gso_segs));
 
@@ -431,7 +433,7 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 
 	len = ntohs(iph->tot_len);
 	if (skb->len < len) {
-		IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INTRUNCATEDPKTS);
+		IP_INC_STATS_BH(net, IPSTATS_MIB_INTRUNCATEDPKTS);
 		goto drop;
 	} else if (len < (iph->ihl*4))
 		goto inhdr_error;
@@ -441,7 +443,7 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 	 * Note this now means skb->len holds ntohs(iph->tot_len).
 	 */
 	if (pskb_trim_rcsum(skb, len)) {
-		IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INDISCARDS);
+		IP_INC_STATS_BH(net, IPSTATS_MIB_INDISCARDS);
 		goto drop;
 	}
 
@@ -458,9 +460,9 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 		       ip_rcv_finish);
 
 csum_error:
-	IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_CSUMERRORS);
+	IP_INC_STATS_BH(net, IPSTATS_MIB_CSUMERRORS);
 inhdr_error:
-	IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INHDRERRORS);
+	IP_INC_STATS_BH(net, IPSTATS_MIB_INHDRERRORS);
 drop:
 	kfree_skb(skb);
 out:
Signature · 1 line
2.2.1
#9 lore raw

PATCH 9/30 ipv4: Remember the net in ip_output and ip_mc_output

In reply to Eric W. Biederman

This is a prepatory patch to passing net int the netfilter hooks, where net will be used again.

---
 net/ipv4/ip_output.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
net/ipv4/ip_output.c +4 -2
index f076f11aa94a..9ee622ad8dfa 100644
@@ -288,11 +288,12 @@ int ip_mc_output(struct sock *sk, struct sk_buff *skb)
 {
 	struct rtable *rt = skb_rtable(skb);
 	struct net_device *dev = rt->dst.dev;
+	struct net *net = dev_net(dev);
 
 	/*
 	 *	If the indicated interface is up and running, send the packet.
 	 */
-	IP_UPD_PO_STATS(dev_net(dev), IPSTATS_MIB_OUT, skb->len);
+	IP_UPD_PO_STATS(net, IPSTATS_MIB_OUT, skb->len);
 
 	skb->dev = dev;
 	skb->protocol = htons(ETH_P_IP);
@@ -347,8 +348,9 @@ int ip_mc_output(struct sock *sk, struct sk_buff *skb)
 int ip_output(struct sock *sk, struct sk_buff *skb)
 {
 	struct net_device *dev = skb_dst(skb)->dev;
+	struct net *net = dev_net(dev);
 
-	IP_UPD_PO_STATS(dev_net(dev), IPSTATS_MIB_OUT, skb->len);
+	IP_UPD_PO_STATS(net, IPSTATS_MIB_OUT, skb->len);
 
 	skb->dev = dev;
 	skb->protocol = htons(ETH_P_IP);
Signature · 1 line
2.2.1
#10 lore raw

PATCH 10/30 ipv4: Don't recompute net in ipmr_queue_xmit

In reply to Eric W. Biederman

Calling dev_net(dev) for is just silly.

---
 net/ipv4/ipmr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv4/ipmr.c +1 -1
index a0a5def920fc..075bc695ae34 100644
@@ -1745,7 +1745,7 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt,
 		 * to blackhole.
 		 */
 
-		IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
+		IP_INC_STATS_BH(net, IPSTATS_MIB_FRAGFAILS);
 		ip_rt_put(rt);
 		goto out_free;
 	}
Signature · 1 line
2.2.1
#11 lore raw

PATCH 11/30 ipv4: Only compute net once in ip_do_fragment

In reply to Eric W. Biederman
---
 net/ipv4/ip_output.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)
net/ipv4/ip_output.c +8 -6
index 9ee622ad8dfa..85b72d450184 100644
@@ -531,9 +531,11 @@ int ip_do_fragment(struct sock *sk, struct sk_buff *skb,
 	int offset;
 	__be16 not_last_frag;
 	struct rtable *rt = skb_rtable(skb);
+	struct net *net;
 	int err = 0;
 
 	dev = rt->dst.dev;
+	net = dev_net(dev);
 
 	/*
 	 *	Point into the IP datagram header.
@@ -626,7 +628,7 @@ int ip_do_fragment(struct sock *sk, struct sk_buff *skb,
 			err = output(sk, skb);
 
 			if (!err)
-				IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGCREATES);
+				IP_INC_STATS(net, IPSTATS_MIB_FRAGCREATES);
 			if (err || !frag)
 				break;
 
@@ -636,7 +638,7 @@ int ip_do_fragment(struct sock *sk, struct sk_buff *skb,
 		}
 
 		if (err == 0) {
-			IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGOKS);
+			IP_INC_STATS(net, IPSTATS_MIB_FRAGOKS);
 			return 0;
 		}
 
@@ -645,7 +647,7 @@ int ip_do_fragment(struct sock *sk, struct sk_buff *skb,
 			kfree_skb(frag);
 			frag = skb;
 		}
-		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
+		IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS);
 		return err;
 
 slow_path_clean:
@@ -767,15 +769,15 @@ slow_path:
 		if (err)
 			goto fail;
 
-		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGCREATES);
+		IP_INC_STATS(net, IPSTATS_MIB_FRAGCREATES);
 	}
 	consume_skb(skb);
-	IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGOKS);
+	IP_INC_STATS(net, IPSTATS_MIB_FRAGOKS);
 	return err;
 
 fail:
 	kfree_skb(skb);
-	IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
+	IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS);
 	return err;
 }
 EXPORT_SYMBOL(ip_do_fragment);
Signature · 1 line
2.2.1
#12 lore raw

PATCH 12/30 ipv4: Explicitly compute net in ip_fragment

In reply to Eric W. Biederman
---
 net/ipv4/ip_output.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
net/ipv4/ip_output.c +2 -3
index 85b72d450184..095754c99061 100644
@@ -500,10 +500,9 @@ static int ip_fragment(struct sock *sk, struct sk_buff *skb,
 	if (unlikely(!skb->ignore_df ||
 		     (IPCB(skb)->frag_max_size &&
 		      IPCB(skb)->frag_max_size > mtu))) {
-		struct rtable *rt = skb_rtable(skb);
-		struct net_device *dev = rt->dst.dev;
+		struct net *net = dev_net(skb_rtable(skb)->dst.dev);
 
-		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
+		IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS);
 		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
 			  htonl(mtu));
 		kfree_skb(skb);
Signature · 1 line
2.2.1
#13 lore raw

PATCH 13/30 ipv4: Only compute net once in ip_finish_output2

In reply to Eric W. Biederman
---
 net/ipv4/ip_output.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/ipv4/ip_output.c +3 -2
index 095754c99061..fc550e97daac 100644
@@ -177,14 +177,15 @@ static int ip_finish_output2(struct sock *sk, struct sk_buff *skb)
 	struct dst_entry *dst = skb_dst(skb);
 	struct rtable *rt = (struct rtable *)dst;
 	struct net_device *dev = dst->dev;
+	struct net *net = dev_net(dev);
 	unsigned int hh_len = LL_RESERVED_SPACE(dev);
 	struct neighbour *neigh;
 	u32 nexthop;
 
 	if (rt->rt_type == RTN_MULTICAST) {
-		IP_UPD_PO_STATS(dev_net(dev), IPSTATS_MIB_OUTMCAST, skb->len);
+		IP_UPD_PO_STATS(net, IPSTATS_MIB_OUTMCAST, skb->len);
 	} else if (rt->rt_type == RTN_BROADCAST)
-		IP_UPD_PO_STATS(dev_net(dev), IPSTATS_MIB_OUTBCAST, skb->len);
+		IP_UPD_PO_STATS(net, IPSTATS_MIB_OUTBCAST, skb->len);
 
 	/* Be paranoid, rather than too clever. */
 	if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) {
Signature · 1 line
2.2.1
#14 lore raw

PATCH 14/30 ipv4: Only compute net once in ip_rcv_finish

In reply to Eric W. Biederman
---
 net/ipv4/ip_input.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)
net/ipv4/ip_input.c +4 -6
index ff908863f22e..cc242b9501d9 100644
@@ -314,6 +314,7 @@ EXPORT_SYMBOL(sysctl_ip_early_demux);
 static int ip_rcv_finish(struct sock *sk, struct sk_buff *skb)
 {
 	const struct iphdr *iph = ip_hdr(skb);
+	struct net *net = dev_net(skb->dev);
 	struct rtable *rt;
 
 	if (sysctl_ip_early_demux && !skb_dst(skb) && !skb->sk) {
@@ -337,8 +338,7 @@ static int ip_rcv_finish(struct sock *sk, struct sk_buff *skb)
 					       iph->tos, skb->dev);
 		if (unlikely(err)) {
 			if (err == -EXDEV)
-				NET_INC_STATS_BH(dev_net(skb->dev),
-						 LINUX_MIB_IPRPFILTER);
+				NET_INC_STATS_BH(net, LINUX_MIB_IPRPFILTER);
 			goto drop;
 		}
 	}
@@ -359,11 +359,9 @@ static int ip_rcv_finish(struct sock *sk, struct sk_buff *skb)
 
 	rt = skb_rtable(skb);
 	if (rt->rt_type == RTN_MULTICAST) {
-		IP_UPD_PO_STATS_BH(dev_net(rt->dst.dev), IPSTATS_MIB_INMCAST,
-				skb->len);
+		IP_UPD_PO_STATS_BH(net, IPSTATS_MIB_INMCAST, skb->len);
 	} else if (rt->rt_type == RTN_BROADCAST)
-		IP_UPD_PO_STATS_BH(dev_net(rt->dst.dev), IPSTATS_MIB_INBCAST,
-				skb->len);
+		IP_UPD_PO_STATS_BH(net, IPSTATS_MIB_INBCAST, skb->len);
 
 	return dst_input(skb);
 
Signature · 1 line
2.2.1
#15 lore raw

PATCH 15/30 ipv4: Only compute net once in ipmr_forward_finish

In reply to Eric W. Biederman
---
 net/ipv4/ipmr.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/ipv4/ipmr.c +3 -2
index 075bc695ae34..dfe4e8ec6c3a 100644
@@ -1681,9 +1681,10 @@ static void ip_encap(struct net *net, struct sk_buff *skb,
 static inline int ipmr_forward_finish(struct sock *sk, struct sk_buff *skb)
 {
 	struct ip_options *opt = &(IPCB(skb)->opt);
+	struct net *net = dev_net(skb_dst(skb)->dev);
 
-	IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTFORWDATAGRAMS);
-	IP_ADD_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTOCTETS, skb->len);
+	IP_INC_STATS_BH(net, IPSTATS_MIB_OUTFORWDATAGRAMS);
+	IP_ADD_STATS_BH(net, IPSTATS_MIB_OUTOCTETS, skb->len);
 
 	if (unlikely(opt->optlen))
 		ip_forward_options(skb);
Signature · 1 line
2.2.1
#16 lore raw

PATCH 16/30 ipv6: Only compute net once in ip6mr_forward2_finish

In reply to Eric W. Biederman
---
 net/ipv6/ip6mr.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/ipv6/ip6mr.c +3 -2
index e95f6b6281de..3e3085b37a91 100644
@@ -1987,9 +1987,10 @@ int ip6mr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg)
 
 static inline int ip6mr_forward2_finish(struct sock *sk, struct sk_buff *skb)
 {
-	IP6_INC_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)),
+	struct net *net = dev_net(skb_dst(skb)->dev);
+	IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 			 IPSTATS_MIB_OUTFORWDATAGRAMS);
-	IP6_ADD_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)),
+	IP6_ADD_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 			 IPSTATS_MIB_OUTOCTETS, skb->len);
 	return dst_output(sk, skb);
 }
Signature · 1 line
2.2.1
#17 lore raw

Re: [PATCH next 16/30] ipv6: Only compute net once in ip6mr_forward2_finish

Quoted 10 lines · Eric W. Biederman
---
  net/ipv6/ip6mr.c | 5 +++--
  1 file changed, 3 insertions(+), 2 deletions(-)
net/ipv6/ip6mr.c +0 -0
index e95f6b6281de..3e3085b37a91 100644

net/ipv6/ip6mr.c int ip6mr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg) · line 1990

Quoted 5 lines
net/ipv6/ip6mr.c +1 -1
@@ -1987,9 +1987,10 @@ int ip6mr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg)
 
  static inline int ip6mr_forward2_finish(struct sock *sk, struct sk_buff *skb)
  {
-	IP6_INC_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)),
+	struct net *net = dev_net(skb_dst(skb)->dev);

nit: a blank line is needed after this declaration.

Quoted 8 lines
net/ipv6/ip6mr.c +2 -1
+	IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
  			 IPSTATS_MIB_OUTFORWDATAGRAMS);
-	IP6_ADD_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)),
+	IP6_ADD_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
  			 IPSTATS_MIB_OUTOCTETS, skb->len);
  	return dst_output(sk, skb);
  }
#18 lore raw

PATCH 17/30 arp: Introduce arp_xmit_finish

In reply to Eric W. Biederman

The function dev_queue_xmit_skb_sk is unncessary and very confusing. Introduce arp_xmit_finish to remove the need for dev_queue_xmit_skb_sk, and have arp_xmit_finish call dev_queue_xmit.

---
 net/ipv4/arp.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
net/ipv4/arp.c +6 -1
index 30409b75e925..3632e98eb0f9 100644
@@ -621,6 +621,11 @@ out:
 }
 EXPORT_SYMBOL(arp_create);
 
+static int arp_xmit_finish(struct sock *sk, struct sk_buff *skb)
+{
+	return dev_queue_xmit(skb);
+}
+
 /*
  *	Send an arp packet.
  */
@@ -628,7 +633,7 @@ void arp_xmit(struct sk_buff *skb)
 {
 	/* Send it off, maybe filter it using firewalling first.  */
 	NF_HOOK(NFPROTO_ARP, NF_ARP_OUT, NULL, skb,
-		NULL, skb->dev, dev_queue_xmit_sk);
+		NULL, skb->dev, arp_xmit_finish);
 }
 EXPORT_SYMBOL(arp_xmit);
 
Signature · 1 line
2.2.1
#19 lore raw

PATCH 18/30 bridge: Introduce br_send_bpdu_finish

In reply to Eric W. Biederman

The function dev_queue_xmit_skb_sk is unncessary and very confusing. Introduce br_send_bpdu_finish to remove the need for dev_queue_xmit_skb_sk, and have br_send_bpdu_finish call dev_queue_xmit.

---
 net/bridge/br_stp_bpdu.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
net/bridge/br_stp_bpdu.c +6 -1
index 534fc4cd263e..3017a396cdef 100644
@@ -30,6 +30,11 @@
 
 #define LLC_RESERVE sizeof(struct llc_pdu_un)
 
+static int br_send_bpdu_finish(struct sock *sk, struct sk_buff *skb)
+{
+	return dev_queue_xmit(skb);
+}
+
 static void br_send_bpdu(struct net_bridge_port *p,
 			 const unsigned char *data, int length)
 {
@@ -56,7 +61,7 @@ static void br_send_bpdu(struct net_bridge_port *p,
 
 	NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, NULL, skb,
 		NULL, skb->dev,
-		dev_queue_xmit_sk);
+		br_send_bpdu_finish);
 }
 
 static inline void br_set_ticks(unsigned char *dest, int j)
Signature · 1 line
2.2.1
#20 lore raw

PATCH 19/30 net: Remove dev_queue_xmit_sk

In reply to Eric W. Biederman

A function with weird arguments that it will never use to accomdate a netfilter callback prototype is absolutely in the core of the networking stack. Frankly it does not make sense and it causes a lot of confusion as to why arguments that are never used are being passed to the function.

As I am preparing to make a second change to arguments to the okfn even the names stops making sense.

As I have removed the two callers of this function remove this confusion from the networking stack.

---
 include/linux/netdevice.h | 6 +-----
 net/core/dev.c            | 4 ++--
 2 files changed, 3 insertions(+), 7 deletions(-)
include/linux/netdevice.h +1 -5
index 88a00694eda5..e664f87c8e4c 100644
@@ -2213,11 +2213,7 @@ int dev_close(struct net_device *dev);
 int dev_close_many(struct list_head *head, bool unlink);
 void dev_disable_lro(struct net_device *dev);
 int dev_loopback_xmit(struct sock *sk, struct sk_buff *newskb);
-int dev_queue_xmit_sk(struct sock *sk, struct sk_buff *skb);
-static inline int dev_queue_xmit(struct sk_buff *skb)
-{
-	return dev_queue_xmit_sk(skb->sk, skb);
-}
+int dev_queue_xmit(struct sk_buff *skb);
 int dev_queue_xmit_accel(struct sk_buff *skb, void *accel_priv);
 int register_netdevice(struct net_device *dev);
 void unregister_netdevice_queue(struct net_device *dev, struct list_head *head);
net/core/dev.c +2 -2
index 877c84834d81..dcf9ff913925 100644
@@ -3143,11 +3143,11 @@ out:
 	return rc;
 }
 
-int dev_queue_xmit_sk(struct sock *sk, struct sk_buff *skb)
+int dev_queue_xmit(struct sk_buff *skb)
 {
 	return __dev_queue_xmit(skb, NULL);
 }
-EXPORT_SYMBOL(dev_queue_xmit_sk);
+EXPORT_SYMBOL(dev_queue_xmit);
 
 int dev_queue_xmit_accel(struct sk_buff *skb, void *accel_priv)
 {
Signature · 1 line
2.2.1
#21 lore raw

PATCH 20/30 ipv6: Don't recompute net in ip6_rcv

In reply to Eric W. Biederman

Avoid silly redundant code

---
 net/ipv6/ip6_input.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/ip6_input.c +1 -1
index adba03ac7ce9..c628dba477d4 100644
@@ -109,7 +109,7 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt
 	if (hdr->version != 6)
 		goto err;
 
-	IP6_ADD_STATS_BH(dev_net(dev), idev,
+	IP6_ADD_STATS_BH(net, idev,
 			 IPSTATS_MIB_NOECTPKTS +
 				(ipv6_get_dsfield(hdr) & INET_ECN_MASK),
 			 max_t(unsigned short, 1, skb_shinfo(skb)->gso_segs));
Signature · 1 line
2.2.1
#22 lore raw

PATCH 21/30 ipv6: Only compute net once in ip6_finish_output2

In reply to Eric W. Biederman
---
 net/ipv6/ip6_output.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
net/ipv6/ip6_output.c +5 -6
index a80502c64523..12d0166a64cd 100644
@@ -60,6 +60,7 @@ static int ip6_finish_output2(struct sock *sk, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb_dst(skb);
 	struct net_device *dev = dst->dev;
+	struct net *net = dev_net(dev);
 	struct neighbour *neigh;
 	struct in6_addr *nexthop;
 	int ret;
@@ -71,7 +72,7 @@ static int ip6_finish_output2(struct sock *sk, struct sk_buff *skb)
 		struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb));
 
 		if (!(dev->flags & IFF_LOOPBACK) && sk_mc_loop(sk) &&
-		    ((mroute6_socket(dev_net(dev), skb) &&
+		    ((mroute6_socket(net, skb) &&
 		     !(IP6CB(skb)->flags & IP6SKB_FORWARDED)) ||
 		     ipv6_chk_mcast_addr(dev, &ipv6_hdr(skb)->daddr,
 					 &ipv6_hdr(skb)->saddr))) {
@@ -86,15 +87,14 @@ static int ip6_finish_output2(struct sock *sk, struct sk_buff *skb)
 					dev_loopback_xmit);
 
 			if (ipv6_hdr(skb)->hop_limit == 0) {
-				IP6_INC_STATS(dev_net(dev), idev,
+				IP6_INC_STATS(net, idev,
 					      IPSTATS_MIB_OUTDISCARDS);
 				kfree_skb(skb);
 				return 0;
 			}
 		}
 
-		IP6_UPD_PO_STATS(dev_net(dev), idev, IPSTATS_MIB_OUTMCAST,
-				skb->len);
+		IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUTMCAST, skb->len);
 
 		if (IPV6_ADDR_MC_SCOPE(&ipv6_hdr(skb)->daddr) <=
 		    IPV6_ADDR_SCOPE_NODELOCAL &&
@@ -116,8 +116,7 @@ static int ip6_finish_output2(struct sock *sk, struct sk_buff *skb)
 	}
 	rcu_read_unlock_bh();
 
-	IP6_INC_STATS(dev_net(dst->dev),
-		      ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES);
+	IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES);
 	kfree_skb(skb);
 	return -EINVAL;
 }
Signature · 1 line
2.2.1
#23 lore raw

PATCH 22/30 ipv6: Cache net in ip6_output

In reply to Eric W. Biederman

Keep net in a local variable so I can use it in NF_HOOK_COND when I pass struct net to all of the netfilter hooks.

---
 net/ipv6/ip6_output.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/ipv6/ip6_output.c +2 -2
index 12d0166a64cd..8cab909b181e 100644
@@ -135,9 +135,9 @@ int ip6_output(struct sock *sk, struct sk_buff *skb)
 {
 	struct net_device *dev = skb_dst(skb)->dev;
 	struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb));
+	struct net *net = dev_net(dev);
 	if (unlikely(idev->cnf.disable_ipv6)) {
-		IP6_INC_STATS(dev_net(dev), idev,
-			      IPSTATS_MIB_OUTDISCARDS);
+		IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS);
 		kfree_skb(skb);
 		return 0;
 	}
Signature · 1 line
2.2.1
#24 lore raw

Re: [PATCH next 22/30] ipv6: Cache net in ip6_output

Quoted 13 lines · Eric W. Biederman

Keep net in a local variable so I can use it in NF_HOOK_COND when I pass struct net to all of the netfilter hooks.

---
  net/ipv6/ip6_output.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)
net/ipv6/ip6_output.c +0 -0
index 12d0166a64cd..8cab909b181e 100644

net/ipv6/ip6_output.c int ip6_output(struct sock *sk, struct sk_buff *skb) · line 138

Quoted 4 lines
net/ipv6/ip6_output.c +1 -0
@@ -135,9 +135,9 @@ int ip6_output(struct sock *sk, struct sk_buff *skb)
  {
  	struct net_device *dev = skb_dst(skb)->dev;
  	struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb));
+	struct net *net = dev_net(dev);

nit: same here for the blank line.

net/ipv6/ip6_output.c int ip6_output(struct sock *sk, struct sk_buff *skb) · line 143

Quoted 8 lines
net/ipv6/ip6_output.c +1 -2
  	if (unlikely(idev->cnf.disable_ipv6)) {
-		IP6_INC_STATS(dev_net(dev), idev,
-			      IPSTATS_MIB_OUTDISCARDS);
+		IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS);
  		kfree_skb(skb);
  		return 0;
  	}

--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#25 lore raw

PATCH 23/30 ipv6: Compute net once in raw6_send_hdrinc

In reply to Eric W. Biederman
---
 net/ipv6/raw.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/ipv6/raw.c +3 -2
index 1636537705f5..5aa461302716 100644
@@ -614,6 +614,7 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length,
 			unsigned int flags)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
+	struct net *net = sock_net(sk);
 	struct ipv6hdr *iph;
 	struct sk_buff *skb;
 	int err;
@@ -652,7 +653,7 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length,
 	if (err)
 		goto error_fault;
 
-	IP6_UPD_PO_STATS(sock_net(sk), rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len);
+	IP6_UPD_PO_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len);
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
 		      NULL, rt->dst.dev, dst_output);
 	if (err > 0)
@@ -666,7 +667,7 @@ error_fault:
 	err = -EFAULT;
 	kfree_skb(skb);
 error:
-	IP6_INC_STATS(sock_net(sk), rt->rt6i_idev, IPSTATS_MIB_OUTDISCARDS);
+	IP6_INC_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUTDISCARDS);
 	if (err == -ENOBUFS && !np->recverr)
 		err = 0;
 	return err;
Signature · 1 line
2.2.1
#26 lore raw

PATCH 24/30 bridge: Pass net into br_nf_ip_fragment

In reply to Eric W. Biederman

This is a prep work for passing struct net through ip_do_fragment and
later the netfilter okfn. Doing this independently makes the later
code changes clearer.

---
 net/bridge/br_netfilter_hooks.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
net/bridge/br_netfilter_hooks.c +6 -6
index 0a6f095bb0c9..971d45d24c64 100644
@@ -695,18 +695,17 @@ static int br_nf_push_frag_xmit(struct sock *sk, struct sk_buff *skb)
 #endif
 
 #if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4)
-static int br_nf_ip_fragment(struct sock *sk, struct sk_buff *skb,
-			     int (*output)(struct sock *, struct sk_buff *))
+static int
+br_nf_ip_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
+		  int (*output)(struct sock *, struct sk_buff *))
 {
 	unsigned int mtu = ip_skb_dst_mtu(skb);
 	struct iphdr *iph = ip_hdr(skb);
-	struct rtable *rt = skb_rtable(skb);
-	struct net_device *dev = rt->dst.dev;
 
 	if (unlikely(((iph->frag_off & htons(IP_DF)) && !skb->ignore_df) ||
 		     (IPCB(skb)->frag_max_size &&
 		      IPCB(skb)->frag_max_size > mtu))) {
-		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
+		IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS);
 		kfree_skb(skb);
 		return -EMSGSIZE;
 	}
@@ -726,6 +725,7 @@ static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb)
 {
 	struct nf_bridge_info *nf_bridge;
 	unsigned int mtu_reserved;
+	struct net *net = dev_net(skb_dst(skb)->dev);
 
 	mtu_reserved = nf_bridge_mtu_reduction(skb);
 
@@ -760,7 +760,7 @@ static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb)
 		skb_copy_from_linear_data_offset(skb, -data->size, data->mac,
 						 data->size);
 
-		return br_nf_ip_fragment(sk, skb, br_nf_push_frag_xmit);
+		return br_nf_ip_fragment(net, sk, skb, br_nf_push_frag_xmit);
 	}
 #endif
 #if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
Signature · 1 line
2.2.1
#27 lore raw

PATCH 25/30 bridge: Pass net into br_nf_push_frag_xmit

In reply to Eric W. Biederman

When struct net starts being passed through the ipv4 and ipv6 fragment routines br_nf_push_frag_xmit will need to take a net parameter. Prepare br_nf_push_frag_xmit before that is needed and introduce br_nf_push_frag_xmit_sk for the call sites that still need the old calling conventions.

---
 net/bridge/br_netfilter_hooks.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)
net/bridge/br_netfilter_hooks.c +8 -3
index 971d45d24c64..e6910b71af6e 100644
@@ -668,7 +668,7 @@ static unsigned int br_nf_forward_arp(const struct nf_hook_ops *ops,
 }
 
 #if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4) || IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
-static int br_nf_push_frag_xmit(struct sock *sk, struct sk_buff *skb)
+static int br_nf_push_frag_xmit(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct brnf_frag_data *data;
 	int err;
@@ -692,6 +692,11 @@ static int br_nf_push_frag_xmit(struct sock *sk, struct sk_buff *skb)
 	nf_bridge_info_free(skb);
 	return br_dev_queue_push_xmit(sk, skb);
 }
+static int br_nf_push_frag_xmit_sk(struct sock *sk, struct sk_buff *skb)
+{
+	struct net *net = dev_net(skb_dst(skb)->dev);
+	return br_nf_push_frag_xmit(net, sk, skb);
+}
 #endif
 
 #if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4)
@@ -760,7 +765,7 @@ static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb)
 		skb_copy_from_linear_data_offset(skb, -data->size, data->mac,
 						 data->size);
 
-		return br_nf_ip_fragment(net, sk, skb, br_nf_push_frag_xmit);
+		return br_nf_ip_fragment(net, sk, skb, br_nf_push_frag_xmit_sk);
 	}
 #endif
 #if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
@@ -783,7 +788,7 @@ static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb)
 						 data->size);
 
 		if (v6ops)
-			return v6ops->fragment(sk, skb, br_nf_push_frag_xmit);
+			return v6ops->fragment(sk, skb, br_nf_push_frag_xmit_sk);
 
 		kfree_skb(skb);
 		return -EMSGSIZE;
Signature · 1 line
2.2.1
#28 lore raw

PATCH 26/30 bridge: Cache net in br_nf_pre_routing_finish

In reply to Eric W. Biederman

This is prep work for passing net to the netfilter hooks.

---
 net/bridge/br_netfilter_hooks.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
net/bridge/br_netfilter_hooks.c +2 -1
index e6910b71af6e..c1127908e23a 100644
@@ -346,6 +346,7 @@ static int br_nf_pre_routing_finish(struct sock *sk, struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
 	struct iphdr *iph = ip_hdr(skb);
+	struct net *net = dev_net(dev);
 	struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb);
 	struct rtable *rt;
 	int err;
@@ -371,7 +372,7 @@ static int br_nf_pre_routing_finish(struct sock *sk, struct sk_buff *skb)
 			if (err != -EHOSTUNREACH || !in_dev || IN_DEV_FORWARD(in_dev))
 				goto free_skb;
 
-			rt = ip_route_output(dev_net(dev), iph->daddr, 0,
+			rt = ip_route_output(net, iph->daddr, 0,
 					     RT_TOS(iph->tos), 0);
 			if (!IS_ERR(rt)) {
 				/* - Bridged-and-DNAT'ed traffic doesn't
Signature · 1 line
2.2.1
#29 lore raw

PATCH 27/30 bridge: Add br_netif_receive_skb remove netif_receive_skb_sk

In reply to Eric W. Biederman

netif_receive_skb_sk is only called once in the bridge code, replace it with a bridge specific function that calls netif_receive_skb.

---
 include/linux/netdevice.h | 6 +-----
 net/bridge/br_input.c     | 7 ++++++-
 net/core/dev.c            | 4 ++--
 3 files changed, 9 insertions(+), 8 deletions(-)
include/linux/netdevice.h +1 -5
index e664f87c8e4c..97ab5c9a7069 100644
@@ -2985,11 +2985,7 @@ static inline void dev_consume_skb_any(struct sk_buff *skb)
 
 int netif_rx(struct sk_buff *skb);
 int netif_rx_ni(struct sk_buff *skb);
-int netif_receive_skb_sk(struct sock *sk, struct sk_buff *skb);
-static inline int netif_receive_skb(struct sk_buff *skb)
-{
-	return netif_receive_skb_sk(skb->sk, skb);
-}
+int netif_receive_skb(struct sk_buff *skb);
 gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb);
 void napi_gro_flush(struct napi_struct *napi, bool flush_old);
 struct sk_buff *napi_get_frags(struct napi_struct *napi);
net/bridge/br_input.c +6 -1
index f921a5dce22d..2359c041e27c 100644
@@ -26,6 +26,11 @@
 br_should_route_hook_t __rcu *br_should_route_hook __read_mostly;
 EXPORT_SYMBOL(br_should_route_hook);
 
+static int br_netif_receive_skb(struct sock *sk, struct sk_buff *skb)
+{
+	return netif_receive_skb(skb);
+}
+
 static int br_pass_frame_up(struct sk_buff *skb)
 {
 	struct net_device *indev, *brdev = BR_INPUT_SKB_CB(skb)->brdev;
@@ -57,7 +62,7 @@ static int br_pass_frame_up(struct sk_buff *skb)
 
 	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, NULL, skb,
 		       indev, NULL,
-		       netif_receive_skb_sk);
+		       br_netif_receive_skb);
 }
 
 static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br,
net/core/dev.c +2 -2
index dcf9ff913925..7db9b012dfb7 100644
@@ -3982,13 +3982,13 @@ static int netif_receive_skb_internal(struct sk_buff *skb)
  *	NET_RX_SUCCESS: no congestion
  *	NET_RX_DROP: packet was dropped
  */
-int netif_receive_skb_sk(struct sock *sk, struct sk_buff *skb)
+int netif_receive_skb(struct sk_buff *skb)
 {
 	trace_netif_receive_skb_entry(skb);
 
 	return netif_receive_skb_internal(skb);
 }
-EXPORT_SYMBOL(netif_receive_skb_sk);
+EXPORT_SYMBOL(netif_receive_skb);
 
 /* Network device is going away, flush any packets still pending
  * Called with irqs disabled.
Signature · 1 line
2.2.1
#30 lore raw

PATCH 28/30 netfilter: Pass struct net into the netfilter hooks

In reply to Eric W. Biederman

Pass a network namespace parameter into the netfilter hooks. At the call site of the netfilter hooks the path a packet is taking through the network stack is well known which allows the network namespace to be easily and reliabily.

This allows the replacement of magic code like "dev_net(state->in?:state->out)" that appears at the start of most netfilter hooks with "state->net".

In almost all cases the network namespace passed in is derived from the first network device passed in, guaranteeing those paths will not see any changes in practice.

The exceptions are:

xfrm/xfrm_output.c:xfrm_output_resume()         xs_net(skb_dst(skb)->xfrm)
ipvs/ip_vs_xmit.c:ip_vs_nat_send_or_cont()      ip_vs_conn_net(cp)
ipvs/ip_vs_xmit.c:ip_vs_send_or_cont()          ip_vs_conn_net(cp)
ipv4/raw.c:raw_send_hdrinc()                    sock_net(sk)

ipv6/ip6_output.c:ip6_xmit() sock_net(sk)
ipv6/ndisc.c:ndisc_send_skb() dev_net(skb->dev) not dev_net(dst->dev)
ipv6/raw.c:raw6_send_hdrinc() sock_net(sk)
br_netfilter_hooks.c:br_nf_pre_routing_finish() dev_net(skb->dev) before skb->dev is set to nf_bridge->physindev

In all cases these exceptions seem to be a better expression for the network namespace the packet is being processed in then the historic "dev_net(in?in:out)". I am documenting them in case something odd pops up and someone starts trying to track down what happened.

---
 drivers/net/vrf.c                         |  7 ++++---
 include/linux/netfilter.h                 | 27 ++++++++++++---------------
 net/bridge/br_forward.c                   | 13 +++++++------
 net/bridge/br_input.c                     | 13 +++++++------
 net/bridge/br_multicast.c                 |  4 ++--
 net/bridge/br_netfilter_hooks.c           | 15 ++++++++-------
 net/bridge/br_netfilter_ipv6.c            |  7 ++++---
 net/bridge/br_stp_bpdu.c                  |  4 ++--
 net/decnet/dn_neigh.c                     | 15 +++++++++------
 net/decnet/dn_nsp_in.c                    |  4 ++--
 net/decnet/dn_route.c                     | 24 ++++++++++++------------
 net/ipv4/arp.c                            | 10 ++++++----
 net/ipv4/ip_forward.c                     |  5 +++--
 net/ipv4/ip_input.c                       |  8 ++++----
 net/ipv4/ip_output.c                      | 22 +++++++++++++---------
 net/ipv4/ipmr.c                           |  4 ++--
 net/ipv4/raw.c                            |  5 +++--
 net/ipv4/xfrm4_input.c                    |  4 ++--
 net/ipv4/xfrm4_output.c                   |  6 ++++--
 net/ipv6/ip6_input.c                      |  8 ++++----
 net/ipv6/ip6_output.c                     | 15 ++++++++-------
 net/ipv6/ip6mr.c                          |  4 ++--
 net/ipv6/mcast.c                          |  7 ++++---
 net/ipv6/ndisc.c                          |  4 ++--
 net/ipv6/netfilter/nf_defrag_ipv6_hooks.c |  2 +-
 net/ipv6/output_core.c                    |  6 ++++--
 net/ipv6/raw.c                            |  2 +-
 net/ipv6/xfrm6_input.c                    |  4 ++--
 net/ipv6/xfrm6_output.c                   |  6 ++++--
 net/netfilter/ipvs/ip_vs_xmit.c           |  4 ++--
 net/xfrm/xfrm_output.c                    |  3 ++-
 31 files changed, 142 insertions(+), 120 deletions(-)
drivers/net/vrf.c +4 -3
index e7094fbd7568..c82260341b72 100644
@@ -298,14 +298,15 @@ err:
 static int vrf_output(struct sock *sk, struct sk_buff *skb)
 {
 	struct net_device *dev = skb_dst(skb)->dev;
+	struct net *net = dev_net(dev);
 
-	IP_UPD_PO_STATS(dev_net(dev), IPSTATS_MIB_OUT, skb->len);
+	IP_UPD_PO_STATS(net, IPSTATS_MIB_OUT, skb->len);
 
 	skb->dev = dev;
 	skb->protocol = htons(ETH_P_IP);
 
-	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, skb,
-			    NULL, dev,
+	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING,
+			    net, sk, skb, NULL, dev,
 			    vrf_finish_output,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
include/linux/netfilter.h +12 -15
index 042148dc1e22..295f2650b5dc 100644
@@ -190,12 +190,11 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook,
 	return 1;
 }
 
-static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sock *sk,
-			  struct sk_buff *skb, struct net_device *indev,
-			  struct net_device *outdev,
+static inline int nf_hook(u_int8_t pf, unsigned int hook, struct net *net,
+			  struct sock *sk, struct sk_buff *skb,
+			  struct net_device *indev, struct net_device *outdev,
 			  int (*okfn)(struct sock *, struct sk_buff *))
 {
-	struct net *net = dev_net(indev ? indev : outdev);
 	return nf_hook_thresh(pf, hook, net, sk, skb, indev, outdev, okfn, INT_MIN);
 }
                    
@@ -217,12 +216,11 @@ static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sock *sk,
 */
 
 static inline int
-NF_HOOK_THRESH(uint8_t pf, unsigned int hook, struct sock *sk,
+NF_HOOK_THRESH(uint8_t pf, unsigned int hook, struct net *net, struct sock *sk,
 	       struct sk_buff *skb, struct net_device *in,
 	       struct net_device *out,
 	       int (*okfn)(struct sock *, struct sk_buff *), int thresh)
 {
-	struct net *net = dev_net(in ? in : out);
 	int ret = nf_hook_thresh(pf, hook, net, sk, skb, in, out, okfn, thresh);
 	if (ret == 1)
 		ret = okfn(sk, skb);
@@ -230,11 +228,10 @@ NF_HOOK_THRESH(uint8_t pf, unsigned int hook, struct sock *sk,
 }
 
 static inline int
-NF_HOOK_COND(uint8_t pf, unsigned int hook, struct sock *sk,
+NF_HOOK_COND(uint8_t pf, unsigned int hook, struct net *net, struct sock *sk,
 	     struct sk_buff *skb, struct net_device *in, struct net_device *out,
 	     int (*okfn)(struct sock *, struct sk_buff *), bool cond)
 {
-	struct net *net = dev_net(in ? in : out);
 	int ret;
 
 	if (!cond ||
@@ -244,11 +241,11 @@ NF_HOOK_COND(uint8_t pf, unsigned int hook, struct sock *sk,
 }
 
 static inline int
-NF_HOOK(uint8_t pf, unsigned int hook, struct sock *sk, struct sk_buff *skb,
+NF_HOOK(uint8_t pf, unsigned int hook, struct net *net, struct sock *sk, struct sk_buff *skb,
 	struct net_device *in, struct net_device *out,
 	int (*okfn)(struct sock *, struct sk_buff *))
 {
-	return NF_HOOK_THRESH(pf, hook, sk, skb, in, out, okfn, INT_MIN);
+	return NF_HOOK_THRESH(pf, hook, net, sk, skb, in, out, okfn, INT_MIN);
 }
 
 /* Call setsockopt() */
@@ -348,11 +345,11 @@ nf_nat_decode_session(struct sk_buff *skb, struct flowi *fl, u_int8_t family)
 }
 
 #else /* !CONFIG_NETFILTER */
-#define NF_HOOK(pf, hook, sk, skb, indev, outdev, okfn) (okfn)(sk, skb)
-#define NF_HOOK_COND(pf, hook, sk, skb, indev, outdev, okfn, cond) (okfn)(sk, skb)
-static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sock *sk,
-			  struct sk_buff *skb, struct net_device *indev,
-			  struct net_device *outdev,
+#define NF_HOOK(pf, hook, net, sk, skb, indev, outdev, okfn) (okfn)(sk, skb)
+#define NF_HOOK_COND(pf, hook, net, sk, skb, indev, outdev, okfn, cond) (okfn)(sk, skb)
+static inline int nf_hook(u_int8_t pf, unsigned int hook, struct net *net,
+			  struct sock *sk, struct sk_buff *skb,
+			  struct net_device *indev, struct net_device *outdev,
 			  int (*okfn)(struct sock *, struct sk_buff *))
 {
 	return 1;
net/bridge/br_forward.c +7 -6
index fa7bfced888e..2dd2a23ce707 100644
@@ -67,8 +67,9 @@ EXPORT_SYMBOL_GPL(br_dev_queue_push_xmit);
 
 int br_forward_finish(struct sock *sk, struct sk_buff *skb)
 {
-	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING, sk, skb,
-		       NULL, skb->dev,
+	struct net *net = dev_net(skb->dev);
+	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING,
+		       net, sk, skb, NULL, skb->dev,
 		       br_dev_queue_push_xmit);
 
 }
@@ -92,8 +93,8 @@ static void __br_deliver(const struct net_bridge_port *to, struct sk_buff *skb)
 		return;
 	}
 
-	NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, NULL, skb,
-		NULL, skb->dev,
+	NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT,
+		dev_net(skb->dev), NULL, skb,NULL, skb->dev,
 		br_forward_finish);
 }
 
@@ -114,8 +115,8 @@ static void __br_forward(const struct net_bridge_port *to, struct sk_buff *skb)
 	skb->dev = to->dev;
 	skb_forward_csum(skb);
 
-	NF_HOOK(NFPROTO_BRIDGE, NF_BR_FORWARD, NULL, skb,
-		indev, skb->dev,
+	NF_HOOK(NFPROTO_BRIDGE, NF_BR_FORWARD,
+		dev_net(indev), NULL, skb, indev, skb->dev,
 		br_forward_finish);
 }
 
net/bridge/br_input.c +7 -6
index 2359c041e27c..78fa7acd836e 100644
@@ -60,8 +60,8 @@ static int br_pass_frame_up(struct sk_buff *skb)
 	if (!skb)
 		return NET_RX_DROP;
 
-	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, NULL, skb,
-		       indev, NULL,
+	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN,
+		       dev_net(indev), NULL, skb, indev, NULL,
 		       br_netif_receive_skb);
 }
 
@@ -283,8 +283,9 @@ rx_handler_result_t br_handle_frame(struct sk_buff **pskb)
 		}
 
 		/* Deliver packet to local host only */
-		if (NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, NULL, skb,
-			    skb->dev, NULL, br_handle_local_finish)) {
+		if (NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN,
+			    dev_net(skb->dev), NULL, skb, skb->dev, NULL,
+			    br_handle_local_finish)) {
 			return RX_HANDLER_CONSUMED; /* consumed by filter */
 		} else {
 			*pskb = skb;
@@ -308,8 +309,8 @@ forward:
 		if (ether_addr_equal(p->br->dev->dev_addr, dest))
 			skb->pkt_type = PACKET_HOST;
 
-		NF_HOOK(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, NULL, skb,
-			skb->dev, NULL,
+		NF_HOOK(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING,
+			dev_net(skb->dev), NULL, skb, skb->dev, NULL,
 			br_handle_frame_finish);
 		break;
 	default:
net/bridge/br_multicast.c +2 -2
index 66efdc21f548..b4d858a18eb6 100644
@@ -829,8 +829,8 @@ static void __br_multicast_send_query(struct net_bridge *br,
 
 	if (port) {
 		skb->dev = port->dev;
-		NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, NULL, skb,
-			NULL, skb->dev,
+		NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT,
+			dev_net(port->dev), NULL, skb, NULL, skb->dev,
 			br_dev_queue_push_xmit);
 	} else {
 		br_multicast_select_own_querier(br, ip, skb);
net/bridge/br_netfilter_hooks.c +8 -7
index c1127908e23a..7886c9d7e23d 100644
@@ -394,7 +394,7 @@ bridged_dnat:
 				nf_bridge_push_encap_header(skb);
 				NF_HOOK_THRESH(NFPROTO_BRIDGE,
 					       NF_BR_PRE_ROUTING,
-					       sk, skb, skb->dev, NULL,
+					       net, sk, skb, skb->dev, NULL,
 					       br_nf_pre_routing_finish_bridge,
 					       1);
 				return 0;
@@ -414,7 +414,7 @@ bridged_dnat:
 	skb->dev = nf_bridge->physindev;
 	nf_bridge_update_protocol(skb);
 	nf_bridge_push_encap_header(skb);
-	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, sk, skb,
+	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, net, sk, skb,
 		       skb->dev, NULL,
 		       br_handle_frame_finish, 1);
 
@@ -512,7 +512,7 @@ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops,
 
 	skb->protocol = htons(ETH_P_IP);
 
-	NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, state->sk, skb,
+	NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, state->net, state->sk, skb,
 		skb->dev, NULL,
 		br_nf_pre_routing_finish);
 
@@ -539,6 +539,7 @@ static unsigned int br_nf_local_in(const struct nf_hook_ops *ops,
 static int br_nf_forward_finish(struct sock *sk, struct sk_buff *skb)
 {
 	struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb);
+	struct net *net = dev_net(skb->dev);
 	struct net_device *in;
 
 	if (!IS_ARP(skb) && !IS_VLAN_ARP(skb)) {
@@ -560,7 +561,7 @@ static int br_nf_forward_finish(struct sock *sk, struct sk_buff *skb)
 	}
 	nf_bridge_push_encap_header(skb);
 
-	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_FORWARD, sk, skb,
+	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_FORWARD, net, sk, skb,
 		       in, skb->dev, br_forward_finish, 1);
 	return 0;
 }
@@ -627,7 +628,7 @@ static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops,
 	else
 		skb->protocol = htons(ETH_P_IPV6);
 
-	NF_HOOK(pf, NF_INET_FORWARD, NULL, skb,
+	NF_HOOK(pf, NF_INET_FORWARD, state->net, NULL, skb,
 		brnf_get_logical_dev(skb, state->in),
 		parent,	br_nf_forward_finish);
 
@@ -662,7 +663,7 @@ static unsigned int br_nf_forward_arp(const struct nf_hook_ops *ops,
 		return NF_ACCEPT;
 	}
 	*d = state->in;
-	NF_HOOK(NFPROTO_ARP, NF_ARP_FORWARD, state->sk, skb,
+	NF_HOOK(NFPROTO_ARP, NF_ARP_FORWARD, state->net, state->sk, skb,
 		state->in, state->out, br_nf_forward_finish);
 
 	return NF_STOLEN;
@@ -842,7 +843,7 @@ static unsigned int br_nf_post_routing(const struct nf_hook_ops *ops,
 	else
 		skb->protocol = htons(ETH_P_IPV6);
 
-	NF_HOOK(pf, NF_INET_POST_ROUTING, state->sk, skb,
+	NF_HOOK(pf, NF_INET_POST_ROUTING, state->net, state->sk, skb,
 		NULL, realoutdev,
 		br_nf_dev_queue_xmit);
 
net/bridge/br_netfilter_ipv6.c +4 -3
index 77383bfe7ea3..5d19361ad5d3 100644
@@ -166,6 +166,7 @@ static int br_nf_pre_routing_finish_ipv6(struct sock *sk, struct sk_buff *skb)
 	struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb);
 	struct rtable *rt;
 	struct net_device *dev = skb->dev;
+	struct net *net = dev_net(dev);
 	const struct nf_ipv6_ops *v6ops = nf_get_ipv6_ops();
 
 	nf_bridge->frag_max_size = IP6CB(skb)->frag_max_size;
@@ -189,7 +190,7 @@ static int br_nf_pre_routing_finish_ipv6(struct sock *sk, struct sk_buff *skb)
 			nf_bridge_update_protocol(skb);
 			nf_bridge_push_encap_header(skb);
 			NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING,
-				       sk, skb, skb->dev, NULL,
+				       net, sk, skb, skb->dev, NULL,
 				       br_nf_pre_routing_finish_bridge,
 				       1);
 			return 0;
@@ -208,7 +209,7 @@ static int br_nf_pre_routing_finish_ipv6(struct sock *sk, struct sk_buff *skb)
 	skb->dev = nf_bridge->physindev;
 	nf_bridge_update_protocol(skb);
 	nf_bridge_push_encap_header(skb);
-	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, sk, skb,
+	NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, net, sk, skb,
 		       skb->dev, NULL,
 		       br_handle_frame_finish, 1);
 
@@ -237,7 +238,7 @@ unsigned int br_nf_pre_routing_ipv6(const struct nf_hook_ops *ops,
 	nf_bridge->ipv6_daddr = ipv6_hdr(skb)->daddr;
 
 	skb->protocol = htons(ETH_P_IPV6);
-	NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, state->sk, skb,
+	NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, state->net, state->sk, skb,
 		skb->dev, NULL,
 		br_nf_pre_routing_finish_ipv6);
 
net/bridge/br_stp_bpdu.c +2 -2
index 3017a396cdef..8e2e8c352198 100644
@@ -59,8 +59,8 @@ static void br_send_bpdu(struct net_bridge_port *p,
 
 	skb_reset_mac_header(skb);
 
-	NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, NULL, skb,
-		NULL, skb->dev,
+	NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT,
+		dev_net(p->dev), NULL, skb, NULL, skb->dev,
 		br_send_bpdu_finish);
 }
 
net/decnet/dn_neigh.c +9 -6
index 4507b188fc51..305ab2fe25cd 100644
@@ -246,8 +246,9 @@ static int dn_long_output(struct neighbour *neigh, struct sock *sk,
 
 	skb_reset_network_header(skb);
 
-	return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING, sk, skb,
-		       NULL, neigh->dev, dn_neigh_output_packet);
+	return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING,
+		       &init_net, sk, skb, NULL, neigh->dev,
+		       dn_neigh_output_packet);
 }
 
 /*
@@ -286,8 +287,9 @@ static int dn_short_output(struct neighbour *neigh, struct sock *sk,
 
 	skb_reset_network_header(skb);
 
-	return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING, sk, skb,
-		       NULL, neigh->dev, dn_neigh_output_packet);
+	return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING,
+		       &init_net, sk, skb, NULL, neigh->dev,
+		       dn_neigh_output_packet);
 }
 
 /*
@@ -327,8 +329,9 @@ static int dn_phase3_output(struct neighbour *neigh, struct sock *sk,
 
 	skb_reset_network_header(skb);
 
-	return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING, sk, skb,
-		       NULL, neigh->dev, dn_neigh_output_packet);
+	return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING,
+		       &init_net, sk, skb, NULL, neigh->dev,
+		       dn_neigh_output_packet);
 }
 
 int dn_to_neigh_output(struct sock *sk, struct sk_buff *skb)
net/decnet/dn_nsp_in.c +2 -2
index a321eac9fd0c..e7b0605ca34a 100644
@@ -814,8 +814,8 @@ free_out:
 
 int dn_nsp_rx(struct sk_buff *skb)
 {
-	return NF_HOOK(NFPROTO_DECNET, NF_DN_LOCAL_IN, NULL, skb,
-		       skb->dev, NULL,
+	return NF_HOOK(NFPROTO_DECNET, NF_DN_LOCAL_IN,
+		       &init_net, NULL, skb, skb->dev, NULL,
 		       dn_nsp_rx_packet);
 }
 
net/decnet/dn_route.c +12 -12
index 03227ffd19ce..fefcd2e85ef9 100644
@@ -573,8 +573,8 @@ static int dn_route_rx_long(struct sk_buff *skb)
 	ptr++;
 	cb->hops = *ptr++; /* Visit Count */
 
-	return NF_HOOK(NFPROTO_DECNET, NF_DN_PRE_ROUTING, NULL, skb,
-		       skb->dev, NULL,
+	return NF_HOOK(NFPROTO_DECNET, NF_DN_PRE_ROUTING,
+		       &init_net, NULL, skb, skb->dev, NULL,
 		       dn_route_rx_packet);
 
 drop_it:
@@ -601,8 +601,8 @@ static int dn_route_rx_short(struct sk_buff *skb)
 	ptr += 2;
 	cb->hops = *ptr & 0x3f;
 
-	return NF_HOOK(NFPROTO_DECNET, NF_DN_PRE_ROUTING, NULL, skb,
-		       skb->dev, NULL,
+	return NF_HOOK(NFPROTO_DECNET, NF_DN_PRE_ROUTING,
+		       &init_net, NULL, skb, skb->dev, NULL,
 		       dn_route_rx_packet);
 
 drop_it:
@@ -706,22 +706,22 @@ int dn_route_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type
 		switch (flags & DN_RT_CNTL_MSK) {
 		case DN_RT_PKT_HELO:
 			return NF_HOOK(NFPROTO_DECNET, NF_DN_HELLO,
-				       NULL, skb, skb->dev, NULL,
+				       &init_net, NULL, skb, skb->dev, NULL,
 				       dn_route_ptp_hello);
 
 		case DN_RT_PKT_L1RT:
 		case DN_RT_PKT_L2RT:
 			return NF_HOOK(NFPROTO_DECNET, NF_DN_ROUTE,
-				       NULL, skb, skb->dev, NULL,
+				       &init_net, NULL, skb, skb->dev, NULL,
 				       dn_route_discard);
 		case DN_RT_PKT_ERTH:
 			return NF_HOOK(NFPROTO_DECNET, NF_DN_HELLO,
-				       NULL, skb, skb->dev, NULL,
+				       &init_net, NULL, skb, skb->dev, NULL,
 				       dn_neigh_router_hello);
 
 		case DN_RT_PKT_EEDH:
 			return NF_HOOK(NFPROTO_DECNET, NF_DN_HELLO,
-				       NULL, skb, skb->dev, NULL,
+				       &init_net, NULL, skb, skb->dev, NULL,
 				       dn_neigh_endnode_hello);
 		}
 	} else {
@@ -770,8 +770,8 @@ static int dn_output(struct sock *sk, struct sk_buff *skb)
 	cb->rt_flags |= DN_RT_F_IE;
 	cb->hops = 0;
 
-	return NF_HOOK(NFPROTO_DECNET, NF_DN_LOCAL_OUT, sk, skb,
-		       NULL, dev,
+	return NF_HOOK(NFPROTO_DECNET, NF_DN_LOCAL_OUT,
+		       &init_net, sk, skb, NULL, dev,
 		       dn_to_neigh_output);
 
 error:
@@ -819,8 +819,8 @@ static int dn_forward(struct sk_buff *skb)
 	if (rt->rt_flags & RTCF_DOREDIRECT)
 		cb->rt_flags |= DN_RT_F_IE;
 
-	return NF_HOOK(NFPROTO_DECNET, NF_DN_FORWARD, NULL, skb,
-		       dev, skb->dev,
+	return NF_HOOK(NFPROTO_DECNET, NF_DN_FORWARD,
+		       &init_net, NULL, skb, dev, skb->dev,
 		       dn_to_neigh_output);
 
 drop:
net/ipv4/arp.c +6 -4
index 3632e98eb0f9..ae71e9ade5f9 100644
@@ -632,8 +632,9 @@ static int arp_xmit_finish(struct sock *sk, struct sk_buff *skb)
 void arp_xmit(struct sk_buff *skb)
 {
 	/* Send it off, maybe filter it using firewalling first.  */
-	NF_HOOK(NFPROTO_ARP, NF_ARP_OUT, NULL, skb,
-		NULL, skb->dev, arp_xmit_finish);
+	NF_HOOK(NFPROTO_ARP, NF_ARP_OUT,
+		dev_net(skb->dev), NULL, skb, NULL, skb->dev,
+		arp_xmit_finish);
 }
 EXPORT_SYMBOL(arp_xmit);
 
@@ -897,8 +898,9 @@ static int arp_rcv(struct sk_buff *skb, struct net_device *dev,
 
 	memset(NEIGH_CB(skb), 0, sizeof(struct neighbour_cb));
 
-	return NF_HOOK(NFPROTO_ARP, NF_ARP_IN, NULL, skb,
-		       dev, NULL, arp_process);
+	return NF_HOOK(NFPROTO_ARP, NF_ARP_IN,
+		       dev_net(dev), NULL, skb, dev, NULL,
+		       arp_process);
 
 consumeskb:
 	consume_skb(skb);
net/ipv4/ip_forward.c +3 -2
index 95235c813f18..0a3c45a2e757 100644
@@ -146,8 +146,9 @@ int ip_forward(struct sk_buff *skb)
 
 	skb->priority = rt_tos2priority(iph->tos);
 
-	return NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD, NULL, skb,
-		       skb->dev, rt->dst.dev, ip_forward_finish);
+	return NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD,
+		       net, NULL, skb, skb->dev, rt->dst.dev,
+		       ip_forward_finish);
 
 sr_failed:
 	/*
net/ipv4/ip_input.c +4 -4
index cc242b9501d9..991d082c7312 100644
@@ -254,8 +254,8 @@ int ip_local_deliver(struct sk_buff *skb)
 			return 0;
 	}
 
-	return NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_IN, NULL, skb,
-		       skb->dev, NULL,
+	return NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_IN,
+		       dev_net(skb->dev), NULL, skb, skb->dev, NULL,
 		       ip_local_deliver_finish);
 }
 
@@ -453,8 +453,8 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 	/* Must drop socket now because of tproxy. */
 	skb_orphan(skb);
 
-	return NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, NULL, skb,
-		       dev, NULL,
+	return NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING,
+		       net, NULL, skb, dev, NULL,
 		       ip_rcv_finish);
 
 csum_error:
net/ipv4/ip_output.c +13 -9
index fc550e97daac..4c9532259a7f 100644
@@ -97,12 +97,14 @@ EXPORT_SYMBOL(ip_send_check);
 
 static int __ip_local_out_sk(struct sock *sk, struct sk_buff *skb)
 {
+	struct net *net = dev_net(skb_dst(skb)->dev);
 	struct iphdr *iph = ip_hdr(skb);
 
 	iph->tot_len = htons(skb->len);
 	ip_send_check(iph);
-	return nf_hook(NFPROTO_IPV4, NF_INET_LOCAL_OUT, sk, skb, NULL,
-		       skb_dst(skb)->dev, dst_output);
+	return nf_hook(NFPROTO_IPV4, NF_INET_LOCAL_OUT,
+		       net, sk, skb, NULL, skb_dst(skb)->dev,
+		       dst_output);
 }
 
 int __ip_local_out(struct sk_buff *skb)
@@ -322,7 +324,7 @@ int ip_mc_output(struct sock *sk, struct sk_buff *skb)
 			struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC);
 			if (newskb)
 				NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING,
-					sk, newskb, NULL, newskb->dev,
+					net, sk, newskb, NULL, newskb->dev,
 					dev_loopback_xmit);
 		}
 
@@ -337,12 +339,14 @@ int ip_mc_output(struct sock *sk, struct sk_buff *skb)
 	if (rt->rt_flags&RTCF_BROADCAST) {
 		struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC);
 		if (newskb)
-			NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, newskb,
-				NULL, newskb->dev, dev_loopback_xmit);
+			NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING,
+				net, sk, newskb, NULL, newskb->dev,
+				dev_loopback_xmit);
 	}
 
-	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, skb, NULL,
-			    skb->dev, ip_finish_output,
+	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING,
+			    net, sk, skb, NULL, skb->dev,
+			    ip_finish_output,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
 
@@ -356,8 +360,8 @@ int ip_output(struct sock *sk, struct sk_buff *skb)
 	skb->dev = dev;
 	skb->protocol = htons(ETH_P_IP);
 
-	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, skb,
-			    NULL, dev,
+	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING,
+			    net, sk, skb, NULL, dev,
 			    ip_finish_output,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
net/ipv4/ipmr.c +2 -2
index dfe4e8ec6c3a..a88c0c5374ff 100644
@@ -1788,8 +1788,8 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt,
 	 * not mrouter) cannot join to more than one interface - it will
 	 * result in receiving multiple packets.
 	 */
-	NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD, NULL, skb,
-		skb->dev, dev,
+	NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD,
+		net, NULL, skb, skb->dev, dev,
 		ipmr_forward_finish);
 	return;
 
net/ipv4/raw.c +3 -2
index 09ab5bb6913a..2045b1aaa6ef 100644
@@ -411,8 +411,9 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
 		icmp_out_count(net, ((struct icmphdr *)
 			skb_transport_header(skb))->type);
 
-	err = NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_OUT, sk, skb,
-		      NULL, rt->dst.dev, dst_output);
+	err = NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_OUT,
+		      net, sk, skb, NULL, rt->dst.dev,
+		      dst_output);
 	if (err > 0)
 		err = net_xmit_errno(err);
 	if (err)
net/ipv4/xfrm4_input.c +2 -2
index 60b032f58ccc..5093000d3d5e 100644
@@ -52,8 +52,8 @@ int xfrm4_transport_finish(struct sk_buff *skb, int async)
 	iph->tot_len = htons(skb->len);
 	ip_send_check(iph);
 
-	NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, NULL, skb,
-		skb->dev, NULL,
+	NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING,
+		dev_net(skb->dev), NULL, skb, skb->dev, NULL,
 		xfrm4_rcv_encap_finish);
 	return 0;
 }
net/ipv4/xfrm4_output.c +4 -2
index 737f5e395a6a..e4a85199e015 100644
@@ -96,8 +96,10 @@ static int __xfrm4_output(struct sock *sk, struct sk_buff *skb)
 
 int xfrm4_output(struct sock *sk, struct sk_buff *skb)
 {
-	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, skb,
-			    NULL, skb_dst(skb)->dev, __xfrm4_output,
+	struct net *net = dev_net(skb_dst(skb)->dev);
+	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING,
+			    net, sk, skb, NULL, skb_dst(skb)->dev,
+			    __xfrm4_output,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
 
net/ipv6/ip6_input.c +4 -4
index c628dba477d4..583cf959c23d 100644
@@ -183,8 +183,8 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt
 	/* Must drop socket now because of tproxy. */
 	skb_orphan(skb);
 
-	return NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, NULL, skb,
-		       dev, NULL,
+	return NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING,
+		       net, NULL, skb, dev, NULL,
 		       ip6_rcv_finish);
 err:
 	IP6_INC_STATS_BH(net, idev, IPSTATS_MIB_INHDRERRORS);
@@ -278,8 +278,8 @@ discard:
 
 int ip6_input(struct sk_buff *skb)
 {
-	return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_IN, NULL, skb,
-		       skb->dev, NULL,
+	return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_IN,
+		       dev_net(skb->dev), NULL, skb, skb->dev, NULL,
 		       ip6_input_finish);
 }
 
net/ipv6/ip6_output.c +8 -7
index 8cab909b181e..96e76ddd4a44 100644
@@ -83,7 +83,7 @@ static int ip6_finish_output2(struct sock *sk, struct sk_buff *skb)
 			 */
 			if (newskb)
 				NF_HOOK(NFPROTO_IPV6, NF_INET_POST_ROUTING,
-					sk, newskb, NULL, newskb->dev,
+					net, sk, newskb, NULL, newskb->dev,
 					dev_loopback_xmit);
 
 			if (ipv6_hdr(skb)->hop_limit == 0) {
@@ -142,8 +142,8 @@ int ip6_output(struct sock *sk, struct sk_buff *skb)
 		return 0;
 	}
 
-	return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, sk, skb,
-			    NULL, dev,
+	return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING,
+			    net, sk, skb, NULL, dev,
 			    ip6_finish_output,
 			    !(IP6CB(skb)->flags & IP6SKB_REROUTED));
 }
@@ -223,8 +223,9 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6,
 	if ((skb->len <= mtu) || skb->ignore_df || skb_is_gso(skb)) {
 		IP6_UPD_PO_STATS(net, ip6_dst_idev(skb_dst(skb)),
 			      IPSTATS_MIB_OUT, skb->len);
-		return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
-			       NULL, dst->dev, dst_output);
+		return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
+			       net, sk, skb, NULL, dst->dev,
+			       dst_output);
 	}
 
 	skb->dev = dst->dev;
@@ -511,8 +512,8 @@ int ip6_forward(struct sk_buff *skb)
 
 	IP6_INC_STATS_BH(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTFORWDATAGRAMS);
 	IP6_ADD_STATS_BH(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTOCTETS, skb->len);
-	return NF_HOOK(NFPROTO_IPV6, NF_INET_FORWARD, NULL, skb,
-		       skb->dev, dst->dev,
+	return NF_HOOK(NFPROTO_IPV6, NF_INET_FORWARD,
+		       net, NULL, skb, skb->dev, dst->dev,
 		       ip6_forward_finish);
 
 error:
net/ipv6/ip6mr.c +2 -2
index 3e3085b37a91..e830942b2090 100644
@@ -2064,8 +2064,8 @@ static int ip6mr_forward2(struct net *net, struct mr6_table *mrt,
 
 	IP6CB(skb)->flags |= IP6SKB_FORWARDED;
 
-	return NF_HOOK(NFPROTO_IPV6, NF_INET_FORWARD, NULL, skb,
-		       skb->dev, dev,
+	return NF_HOOK(NFPROTO_IPV6, NF_INET_FORWARD,
+		       net, NULL, skb, skb->dev, dev,
 		       ip6mr_forward2_finish);
 
 out_free:
net/ipv6/mcast.c +4 -3
index 5b3f2841acf6..124338a39e29 100644
@@ -1645,7 +1645,7 @@ static void mld_sendpack(struct sk_buff *skb)
 	payload_len = skb->len;
 
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
-		      net->ipv6.igmp_sk, skb, NULL, skb->dev,
+		      net, net->ipv6.igmp_sk, skb, NULL, skb->dev,
 		      dst_output);
 out:
 	if (!err) {
@@ -2008,8 +2008,9 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
 	}
 
 	skb_dst_set(skb, dst);
-	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
-		      NULL, skb->dev, dst_output);
+	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
+		      net, sk, skb, NULL, skb->dev,
+		      dst_output);
 out:
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(net, idev, type);
net/ipv6/ndisc.c +2 -2
index 349ac1b022b6..dd2b08d7c8d1 100644
@@ -463,8 +463,8 @@ static void ndisc_send_skb(struct sk_buff *skb,
 	idev = __in6_dev_get(dst->dev);
 	IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len);
 
-	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
-		      NULL, dst->dev,
+	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
+		      net, sk, skb, NULL, dst->dev,
 		      dst_output);
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(net, idev, type);
net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +1 -1
index 6d9c0b3d5b8c..6b576be3c83e 100644
@@ -74,7 +74,7 @@ static unsigned int ipv6_defrag(const struct nf_hook_ops *ops,
 
 	nf_ct_frag6_consume_orig(reasm);
 
-	NF_HOOK_THRESH(NFPROTO_IPV6, ops->hooknum, state->sk, reasm,
+	NF_HOOK_THRESH(NFPROTO_IPV6, ops->hooknum, state->net, state->sk, reasm,
 		       state->in, state->out,
 		       state->okfn, NF_IP6_PRI_CONNTRACK_DEFRAG + 1);
 
net/ipv6/output_core.c +4 -2
index 8178f72fe90d..9cc9127fb5e7 100644
@@ -140,6 +140,7 @@ EXPORT_SYMBOL(ip6_dst_hoplimit);
 
 static int __ip6_local_out_sk(struct sock *sk, struct sk_buff *skb)
 {
+	struct net *net = dev_net(skb_dst(skb)->dev);
 	int len;
 
 	len = skb->len - sizeof(struct ipv6hdr);
@@ -148,8 +149,9 @@ static int __ip6_local_out_sk(struct sock *sk, struct sk_buff *skb)
 	ipv6_hdr(skb)->payload_len = htons(len);
 	IP6CB(skb)->nhoff = offsetof(struct ipv6hdr, nexthdr);
 
-	return nf_hook(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
-		       NULL, skb_dst(skb)->dev, dst_output);
+	return nf_hook(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
+		       net, sk, skb, NULL, skb_dst(skb)->dev,
+		       dst_output);
 }
 
 int __ip6_local_out(struct sk_buff *skb)
net/ipv6/raw.c +1 -1
index 5aa461302716..dc65ec198f7c 100644
@@ -654,7 +654,7 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length,
 		goto error_fault;
 
 	IP6_UPD_PO_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len);
-	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb,
+	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, net, sk, skb,
 		      NULL, rt->dst.dev, dst_output);
 	if (err > 0)
 		err = net_xmit_errno(err);
net/ipv6/xfrm6_input.c +2 -2
index 74bd17882a2f..0eaab1fa6be5 100644
@@ -42,8 +42,8 @@ int xfrm6_transport_finish(struct sk_buff *skb, int async)
 	ipv6_hdr(skb)->payload_len = htons(skb->len);
 	__skb_push(skb, skb->data - skb_network_header(skb));
 
-	NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, NULL, skb,
-		skb->dev, NULL,
+	NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING,
+		dev_net(skb->dev), NULL, skb, skb->dev, NULL,
 		ip6_rcv_finish);
 	return -1;
 }
net/ipv6/xfrm6_output.c +4 -2
index b0fcd6c09837..431ae2c22234 100644
@@ -168,7 +168,9 @@ static int __xfrm6_output(struct sock *sk, struct sk_buff *skb)
 
 int xfrm6_output(struct sock *sk, struct sk_buff *skb)
 {
-	return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, sk, skb,
-			    NULL, skb_dst(skb)->dev, __xfrm6_output,
+	struct net *net = dev_net(skb_dst(skb)->dev);
+	return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING,
+			    net, sk, skb,  NULL, skb_dst(skb)->dev,
+			    __xfrm6_output,
 			    !(IP6CB(skb)->flags & IP6SKB_REROUTED));
 }
net/netfilter/ipvs/ip_vs_xmit.c +2 -2
index b8e5544af87f..65c996c14bca 100644
@@ -573,7 +573,7 @@ static inline int ip_vs_nat_send_or_cont(int pf, struct sk_buff *skb,
 		skb_forward_csum(skb);
 		if (!skb->sk)
 			skb_sender_cpu_clear(skb);
-		NF_HOOK(pf, NF_INET_LOCAL_OUT, NULL, skb,
+		NF_HOOK(pf, NF_INET_LOCAL_OUT, ip_vs_conn_net(cp), NULL, skb,
 			NULL, skb_dst(skb)->dev, dst_output);
 	} else
 		ret = NF_ACCEPT;
@@ -595,7 +595,7 @@ static inline int ip_vs_send_or_cont(int pf, struct sk_buff *skb,
 		skb_forward_csum(skb);
 		if (!skb->sk)
 			skb_sender_cpu_clear(skb);
-		NF_HOOK(pf, NF_INET_LOCAL_OUT, NULL, skb,
+		NF_HOOK(pf, NF_INET_LOCAL_OUT, ip_vs_conn_net(cp), NULL, skb,
 			NULL, skb_dst(skb)->dev, dst_output);
 	} else
 		ret = NF_ACCEPT;
net/xfrm/xfrm_output.c +2 -1
index 57a50f6ce28c..c21f1a02ce13 100644
@@ -131,6 +131,7 @@ out:
 
 int xfrm_output_resume(struct sk_buff *skb, int err)
 {
+	struct net *net = xs_net(skb_dst(skb)->xfrm);
 	while (likely((err = xfrm_output_one(skb, err)) == 0)) {
 		nf_reset(skb);
 
@@ -142,7 +143,7 @@ int xfrm_output_resume(struct sk_buff *skb, int err)
 			return dst_output(skb->sk, skb);
 
 		err = nf_hook(skb_dst(skb)->ops->family,
-			      NF_INET_POST_ROUTING, skb->sk, skb,
+			      NF_INET_POST_ROUTING, net, skb->sk, skb,
 			      NULL, skb_dst(skb)->dev, xfrm_output2);
 		if (unlikely(err != 1))
 			goto out;
Signature · 1 line
2.2.1
#31 lore raw

Re: [PATCH next 28/30] netfilter: Pass struct net into the netfilter hooks

Quoted 30 lines · Eric W. Biederman

Pass a network namespace parameter into the netfilter hooks. At the call site of the netfilter hooks the path a packet is taking through the network stack is well known which allows the network namespace to be easily and reliabily.

This allows the replacement of magic code like "dev_net(state->in?:state->out)" that appears at the start of most netfilter hooks with "state->net".

In almost all cases the network namespace passed in is derived from the first network device passed in, guaranteeing those paths will not see any changes in practice.

The exceptions are:

xfrm/xfrm_output.c:xfrm_output_resume()         xs_net(skb_dst(skb)->xfrm)
ipvs/ip_vs_xmit.c:ip_vs_nat_send_or_cont()      ip_vs_conn_net(cp)
ipvs/ip_vs_xmit.c:ip_vs_send_or_cont()          ip_vs_conn_net(cp)
ipv4/raw.c:raw_send_hdrinc()                    sock_net(sk)

ipv6/ip6_output.c:ip6_xmit() sock_net(sk)
ipv6/ndisc.c:ndisc_send_skb() dev_net(skb->dev) not dev_net(dst->dev)
ipv6/raw.c:raw6_send_hdrinc() sock_net(sk)
br_netfilter_hooks.c:br_nf_pre_routing_finish() dev_net(skb->dev) before skb->dev is set to nf_bridge->physindev

In all cases these exceptions seem to be a better expression for the network namespace the packet is being processed in then the historic "dev_net(in?in:out)". I am documenting them in case something odd pops up and someone starts trying to track down what happened.

---

[snip]

net/bridge/br_forward.c EXPORT_SYMBOL_GPL(br_dev_queue_push_xmit); · line 70

Quoted 5 lines
  int br_forward_finish(struct sock *sk, struct sk_buff *skb)
  {
net/bridge/br_forward.c +1 -2
-	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING, sk, skb,
-		       NULL, skb->dev,
+	struct net *net = dev_net(skb->dev);

nit: blank line after the declaration

Quoted 5 lines
net/bridge/br_forward.c +2 -0
+	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING,
+		       net, sk, skb, NULL, skb->dev,
  		       br_dev_queue_push_xmit);
 
  }

[snip]

net/ipv4/xfrm4_output.c static int __xfrm4_output(struct sock *sk, struct sk_buff *skb) · line 99

Quoted 5 lines
  int xfrm4_output(struct sock *sk, struct sk_buff *skb)
  {
net/ipv4/xfrm4_output.c +1 -2
-	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, skb,
-			    NULL, skb_dst(skb)->dev, __xfrm4_output,
+	struct net *net = dev_net(skb_dst(skb)->dev);

nit: same here

Quoted 5 lines
net/ipv4/xfrm4_output.c +3 -0
+	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING,
+			    net, sk, skb, NULL, skb_dst(skb)->dev,
+			    __xfrm4_output,
  			    !(IPCB(skb)->flags & IPSKB_REROUTED));
  }

[snip]

net/ipv6/xfrm6_output.c static int __xfrm6_output(struct sock *sk, struct sk_buff *skb) · line 171

Quoted 5 lines
  int xfrm6_output(struct sock *sk, struct sk_buff *skb)
  {
net/ipv6/xfrm6_output.c +1 -2
-	return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, sk, skb,
-			    NULL, skb_dst(skb)->dev, __xfrm6_output,
+	struct net *net = dev_net(skb_dst(skb)->dev);

nit: same here

Quoted 5 lines
net/ipv6/xfrm6_output.c +3 -0
+	return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING,
+			    net, sk, skb,  NULL, skb_dst(skb)->dev,
+			    __xfrm6_output,
  			    !(IP6CB(skb)->flags & IP6SKB_REROUTED));
  }

[snip]

net/xfrm/xfrm_output.c out: · line 134

Quoted 3 lines
  int xfrm_output_resume(struct sk_buff *skb, int err)
  {
net/xfrm/xfrm_output.c +1 -0
+	struct net *net = xs_net(skb_dst(skb)->xfrm);

nit: same here

Quoted 2 lines
  	while (likely((err = xfrm_output_one(skb, err)) == 0)) {
  		nf_reset(skb);
#32 lore raw

PATCH 29/30 netfilter: Use nf_hook_state.net

In reply to Eric W. Biederman

Instead of saying "net = dev_net(state->in?state->in:state->out)" just say "state->net". As that information is now availabe, much less confusing and much less error prone.

---
 net/bridge/netfilter/ebtable_filter.c          | 4 ++--
 net/bridge/netfilter/ebtable_nat.c             | 4 ++--
 net/ipv4/netfilter/arptable_filter.c           | 4 +---
 net/ipv4/netfilter/ip_tables.c                 | 8 ++++----
 net/ipv4/netfilter/ipt_CLUSTERIP.c             | 2 +-
 net/ipv4/netfilter/ipt_SYNPROXY.c              | 2 +-
 net/ipv4/netfilter/iptable_filter.c            | 6 ++----
 net/ipv4/netfilter/iptable_mangle.c            | 7 +++----
 net/ipv4/netfilter/iptable_nat.c               | 5 ++---
 net/ipv4/netfilter/iptable_raw.c               | 6 ++----
 net/ipv4/netfilter/iptable_security.c          | 5 +----
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 4 ++--
 net/ipv6/netfilter/ip6_tables.c                | 8 ++++----
 net/ipv6/netfilter/ip6t_SYNPROXY.c             | 2 +-
 net/ipv6/netfilter/ip6table_filter.c           | 5 ++---
 net/ipv6/netfilter/ip6table_mangle.c           | 6 +++---
 net/ipv6/netfilter/ip6table_nat.c              | 5 ++---
 net/ipv6/netfilter/ip6table_raw.c              | 5 ++---
 net/ipv6/netfilter/ip6table_security.c         | 4 +---
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 4 ++--
 net/netfilter/nfnetlink_queue_core.c           | 3 +--
 21 files changed, 41 insertions(+), 58 deletions(-)
net/bridge/netfilter/ebtable_filter.c +2 -2
index 8a3f63b2e807..ab20d6ed6e2f 100644
@@ -61,7 +61,7 @@ ebt_in_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	    const struct nf_hook_state *state)
 {
 	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    dev_net(state->in)->xt.frame_filter);
+			    state->net->xt.frame_filter);
 }
 
 static unsigned int
@@ -69,7 +69,7 @@ ebt_out_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
 	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    dev_net(state->out)->xt.frame_filter);
+			    state->net->xt.frame_filter);
 }
 
 static struct nf_hook_ops ebt_ops_filter[] __read_mostly = {
net/bridge/netfilter/ebtable_nat.c +2 -2
index c5ef5b1ab678..ad81a5a65644 100644
@@ -61,7 +61,7 @@ ebt_nat_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	   const struct nf_hook_state *state)
 {
 	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    dev_net(state->in)->xt.frame_nat);
+			    state->net->xt.frame_nat);
 }
 
 static unsigned int
@@ -69,7 +69,7 @@ ebt_nat_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	    const struct nf_hook_state *state)
 {
 	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    dev_net(state->out)->xt.frame_nat);
+			    state->net->xt.frame_nat);
 }
 
 static struct nf_hook_ops ebt_ops_nat[] __read_mostly = {
net/ipv4/netfilter/arptable_filter.c +1 -3
index 93876d03120c..d217e4c19645 100644
@@ -30,10 +30,8 @@ static unsigned int
 arptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	const struct net *net = dev_net(state->in ? state->in : state->out);
-
 	return arpt_do_table(skb, ops->hooknum, state,
-			     net->ipv4.arptable_filter);
+			     state->net->ipv4.arptable_filter);
 }
 
 static struct nf_hook_ops *arpfilter_ops __read_mostly;
net/ipv4/netfilter/ip_tables.c +4 -4
index b0a86e73451c..5d514eac4c31 100644
@@ -246,7 +246,8 @@ get_chainname_rulenum(const struct ipt_entry *s, const struct ipt_entry *e,
 	return 0;
 }
 
-static void trace_packet(const struct sk_buff *skb,
+static void trace_packet(struct net *net,
+			 const struct sk_buff *skb,
 			 unsigned int hook,
 			 const struct net_device *in,
 			 const struct net_device *out,
@@ -258,7 +259,6 @@ static void trace_packet(const struct sk_buff *skb,
 	const char *hookname, *chainname, *comment;
 	const struct ipt_entry *iter;
 	unsigned int rulenum = 0;
-	struct net *net = dev_net(in ? in : out);
 
 	root = get_entry(private->entries, private->hook_entry[hook]);
 
@@ -378,8 +378,8 @@ ipt_do_table(struct sk_buff *skb,
 #if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE)
 		/* The packet is traced: log it */
 		if (unlikely(skb->nf_trace))
-			trace_packet(skb, hook, state->in, state->out,
-				     table->name, private, e);
+			trace_packet(state->net, skb, hook, state->in,
+				     state->out, table->name, private, e);
 #endif
 		/* Standard target? */
 		if (!t->u.kernel.target->target) {
net/ipv4/netfilter/ipt_CLUSTERIP.c +1 -1
index 45cb16a6a4a3..69157d8eba95 100644
@@ -514,7 +514,7 @@ arp_mangle(const struct nf_hook_ops *ops,
 	struct arphdr *arp = arp_hdr(skb);
 	struct arp_payload *payload;
 	struct clusterip_config *c;
-	struct net *net = dev_net(state->in ? state->in : state->out);
+	struct net *net = state->net;
 
 	/* we don't care about non-ethernet and non-ipv4 ARP */
 	if (arp->ar_hrd != htons(ARPHRD_ETHER) ||
net/ipv4/netfilter/ipt_SYNPROXY.c +1 -1
index 95ea633e8356..f471a0628c75 100644
@@ -303,7 +303,7 @@ static unsigned int ipv4_synproxy_hook(const struct nf_hook_ops *ops,
 				       struct sk_buff *skb,
 				       const struct nf_hook_state *nhs)
 {
-	struct synproxy_net *snet = synproxy_pernet(dev_net(nhs->in ? : nhs->out));
+	struct synproxy_net *snet = synproxy_pernet(nhs->net);
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct;
 	struct nf_conn_synproxy *synproxy;
net/ipv4/netfilter/iptable_filter.c +2 -4
index a0f3beca52d2..32feff32b116 100644
@@ -36,16 +36,14 @@ static unsigned int
 iptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		    const struct nf_hook_state *state)
 {
-	const struct net *net;
-
 	if (ops->hooknum == NF_INET_LOCAL_OUT &&
 	    (skb->len < sizeof(struct iphdr) ||
 	     ip_hdrlen(skb) < sizeof(struct iphdr)))
 		/* root is playing with raw sockets. */
 		return NF_ACCEPT;
 
-	net = dev_net(state->in ? state->in : state->out);
-	return ipt_do_table(skb, ops->hooknum, state, net->ipv4.iptable_filter);
+	return ipt_do_table(skb, ops->hooknum, state,
+			    state->net->ipv4.iptable_filter);
 }
 
 static struct nf_hook_ops *filter_ops __read_mostly;
net/ipv4/netfilter/iptable_mangle.c +3 -4
index 62cbb8c5f4a8..4a5150fc9510 100644
@@ -39,7 +39,6 @@ static const struct xt_table packet_mangler = {
 static unsigned int
 ipt_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state)
 {
-	struct net_device *out = state->out;
 	unsigned int ret;
 	const struct iphdr *iph;
 	u_int8_t tos;
@@ -60,7 +59,7 @@ ipt_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state)
 	tos = iph->tos;
 
 	ret = ipt_do_table(skb, NF_INET_LOCAL_OUT, state,
-			   dev_net(out)->ipv4.iptable_mangle);
+			   state->net->ipv4.iptable_mangle);
 	/* Reroute for ANY change. */
 	if (ret != NF_DROP && ret != NF_STOLEN) {
 		iph = ip_hdr(skb);
@@ -88,10 +87,10 @@ iptable_mangle_hook(const struct nf_hook_ops *ops,
 		return ipt_mangle_out(skb, state);
 	if (ops->hooknum == NF_INET_POST_ROUTING)
 		return ipt_do_table(skb, ops->hooknum, state,
-				    dev_net(state->out)->ipv4.iptable_mangle);
+				    state->net->ipv4.iptable_mangle);
 	/* PREROUTING/INPUT/FORWARD: */
 	return ipt_do_table(skb, ops->hooknum, state,
-			    dev_net(state->in)->ipv4.iptable_mangle);
+			    state->net->ipv4.iptable_mangle);
 }
 
 static struct nf_hook_ops *mangle_ops __read_mostly;
net/ipv4/netfilter/iptable_nat.c +2 -3
index 0d4d9cdf98a4..4f4c64f81169 100644
@@ -33,9 +33,8 @@ static unsigned int iptable_nat_do_chain(const struct nf_hook_ops *ops,
 					 const struct nf_hook_state *state,
 					 struct nf_conn *ct)
 {
-	struct net *net = nf_ct_net(ct);
-
-	return ipt_do_table(skb, ops->hooknum, state, net->ipv4.nat_table);
+	return ipt_do_table(skb, ops->hooknum, state,
+			    state->net->ipv4.nat_table);
 }
 
 static unsigned int iptable_nat_ipv4_fn(const struct nf_hook_ops *ops,
net/ipv4/netfilter/iptable_raw.c +2 -4
index 0356e6da4bb7..20126e469ffb 100644
@@ -23,16 +23,14 @@ static unsigned int
 iptable_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		 const struct nf_hook_state *state)
 {
-	const struct net *net;
-
 	if (ops->hooknum == NF_INET_LOCAL_OUT &&
 	    (skb->len < sizeof(struct iphdr) ||
 	     ip_hdrlen(skb) < sizeof(struct iphdr)))
 		/* root is playing with raw sockets. */
 		return NF_ACCEPT;
 
-	net = dev_net(state->in ? state->in : state->out);
-	return ipt_do_table(skb, ops->hooknum, state, net->ipv4.iptable_raw);
+	return ipt_do_table(skb, ops->hooknum, state,
+			    state->net->ipv4.iptable_raw);
 }
 
 static struct nf_hook_ops *rawtable_ops __read_mostly;
net/ipv4/netfilter/iptable_security.c +1 -4
index 4bce3980ccd9..82fefd609b85 100644
@@ -40,17 +40,14 @@ static unsigned int
 iptable_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
-	const struct net *net;
-
 	if (ops->hooknum == NF_INET_LOCAL_OUT &&
 	    (skb->len < sizeof(struct iphdr) ||
 	     ip_hdrlen(skb) < sizeof(struct iphdr)))
 		/* Somebody is playing with raw sockets. */
 		return NF_ACCEPT;
 
-	net = dev_net(state->in ? state->in : state->out);
 	return ipt_do_table(skb, ops->hooknum, state,
-			    net->ipv4.iptable_security);
+			    state->net->ipv4.iptable_security);
 }
 
 static struct nf_hook_ops *sectbl_ops __read_mostly;
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +2 -2
index 8a2caaf3940b..9564684876c9 100644
@@ -147,7 +147,7 @@ static unsigned int ipv4_conntrack_in(const struct nf_hook_ops *ops,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
-	return nf_conntrack_in(dev_net(state->in), PF_INET, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET, ops->hooknum, skb);
 }
 
 static unsigned int ipv4_conntrack_local(const struct nf_hook_ops *ops,
@@ -158,7 +158,7 @@ static unsigned int ipv4_conntrack_local(const struct nf_hook_ops *ops,
 	if (skb->len < sizeof(struct iphdr) ||
 	    ip_hdrlen(skb) < sizeof(struct iphdr))
 		return NF_ACCEPT;
-	return nf_conntrack_in(dev_net(state->out), PF_INET, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET, ops->hooknum, skb);
 }
 
 /* Connection tracking may drop packets, but never alters them, so
net/ipv6/netfilter/ip6_tables.c +4 -4
index 0771991ed812..cd9b401231d3 100644
@@ -275,7 +275,8 @@ get_chainname_rulenum(const struct ip6t_entry *s, const struct ip6t_entry *e,
 	return 0;
 }
 
-static void trace_packet(const struct sk_buff *skb,
+static void trace_packet(struct net *net,
+			 const struct sk_buff *skb,
 			 unsigned int hook,
 			 const struct net_device *in,
 			 const struct net_device *out,
@@ -287,7 +288,6 @@ static void trace_packet(const struct sk_buff *skb,
 	const char *hookname, *chainname, *comment;
 	const struct ip6t_entry *iter;
 	unsigned int rulenum = 0;
-	struct net *net = dev_net(in ? in : out);
 
 	root = get_entry(private->entries, private->hook_entry[hook]);
 
@@ -401,8 +401,8 @@ ip6t_do_table(struct sk_buff *skb,
 #if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE)
 		/* The packet is traced: log it */
 		if (unlikely(skb->nf_trace))
-			trace_packet(skb, hook, state->in, state->out,
-				     table->name, private, e);
+			trace_packet(state->net, skb, hook, state->in,
+				     state->out, table->name, private, e);
 #endif
 		/* Standard target? */
 		if (!t->u.kernel.target->target) {
net/ipv6/netfilter/ip6t_SYNPROXY.c +1 -1
index 1e4bf99ed16e..4c9f3e79d75f 100644
@@ -320,7 +320,7 @@ static unsigned int ipv6_synproxy_hook(const struct nf_hook_ops *ops,
 				       struct sk_buff *skb,
 				       const struct nf_hook_state *nhs)
 {
-	struct synproxy_net *snet = synproxy_pernet(dev_net(nhs->in ? : nhs->out));
+	struct synproxy_net *snet = synproxy_pernet(nhs->net);
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct;
 	struct nf_conn_synproxy *synproxy;
net/ipv6/netfilter/ip6table_filter.c +2 -3
index 5c33d8abc077..2449005fb5dc 100644
@@ -35,9 +35,8 @@ static unsigned int
 ip6table_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	const struct net *net = dev_net(state->in ? state->in : state->out);
-
-	return ip6t_do_table(skb, ops->hooknum, state, net->ipv6.ip6table_filter);
+	return ip6t_do_table(skb, ops->hooknum, state,
+			     state->net->ipv6.ip6table_filter);
 }
 
 static struct nf_hook_ops *filter_ops __read_mostly;
net/ipv6/netfilter/ip6table_mangle.c +3 -3
index b551f5b79fe2..a46dbf097d29 100644
@@ -58,7 +58,7 @@ ip6t_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state)
 	flowlabel = *((u_int32_t *)ipv6_hdr(skb));
 
 	ret = ip6t_do_table(skb, NF_INET_LOCAL_OUT, state,
-			    dev_net(state->out)->ipv6.ip6table_mangle);
+			    state->net->ipv6.ip6table_mangle);
 
 	if (ret != NF_DROP && ret != NF_STOLEN &&
 	    (!ipv6_addr_equal(&ipv6_hdr(skb)->saddr, &saddr) ||
@@ -83,10 +83,10 @@ ip6table_mangle_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		return ip6t_mangle_out(skb, state);
 	if (ops->hooknum == NF_INET_POST_ROUTING)
 		return ip6t_do_table(skb, ops->hooknum, state,
-				     dev_net(state->out)->ipv6.ip6table_mangle);
+				     state->net->ipv6.ip6table_mangle);
 	/* INPUT/FORWARD */
 	return ip6t_do_table(skb, ops->hooknum, state,
-			     dev_net(state->in)->ipv6.ip6table_mangle);
+			     state->net->ipv6.ip6table_mangle);
 }
 
 static struct nf_hook_ops *mangle_ops __read_mostly;
net/ipv6/netfilter/ip6table_nat.c +2 -3
index c3a7f7af0ed4..a56451de127f 100644
@@ -35,9 +35,8 @@ static unsigned int ip6table_nat_do_chain(const struct nf_hook_ops *ops,
 					  const struct nf_hook_state *state,
 					  struct nf_conn *ct)
 {
-	struct net *net = nf_ct_net(ct);
-
-	return ip6t_do_table(skb, ops->hooknum, state, net->ipv6.ip6table_nat);
+	return ip6t_do_table(skb, ops->hooknum, state,
+			     state->net->ipv6.ip6table_nat);
 }
 
 static unsigned int ip6table_nat_fn(const struct nf_hook_ops *ops,
net/ipv6/netfilter/ip6table_raw.c +2 -3
index 0b33caad2b69..18e831e35782 100644
@@ -22,9 +22,8 @@ static unsigned int
 ip6table_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		  const struct nf_hook_state *state)
 {
-	const struct net *net = dev_net(state->in ? state->in : state->out);
-
-	return ip6t_do_table(skb, ops->hooknum, state, net->ipv6.ip6table_raw);
+	return ip6t_do_table(skb, ops->hooknum, state,
+			     state->net->ipv6.ip6table_raw);
 }
 
 static struct nf_hook_ops *rawtable_ops __read_mostly;
net/ipv6/netfilter/ip6table_security.c +1 -3
index fcef83c25f7b..83bc96ae5d73 100644
@@ -39,10 +39,8 @@ static unsigned int
 ip6table_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		       const struct nf_hook_state *state)
 {
-	const struct net *net = dev_net(state->in ? state->in : state->out);
-
 	return ip6t_do_table(skb, ops->hooknum, state,
-			     net->ipv6.ip6table_security);
+			     state->net->ipv6.ip6table_security);
 }
 
 static struct nf_hook_ops *sectbl_ops __read_mostly;
net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +2 -2
index 7302900c321a..1ef1b79def56 100644
@@ -169,7 +169,7 @@ static unsigned int ipv6_conntrack_in(const struct nf_hook_ops *ops,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
-	return nf_conntrack_in(dev_net(state->in), PF_INET6, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET6, ops->hooknum, skb);
 }
 
 static unsigned int ipv6_conntrack_local(const struct nf_hook_ops *ops,
@@ -181,7 +181,7 @@ static unsigned int ipv6_conntrack_local(const struct nf_hook_ops *ops,
 		net_notice_ratelimited("ipv6_conntrack_local: packet too short\n");
 		return NF_ACCEPT;
 	}
-	return nf_conntrack_in(dev_net(state->out), PF_INET6, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET6, ops->hooknum, skb);
 }
 
 static struct nf_hook_ops ipv6_conntrack_ops[] __read_mostly = {
net/netfilter/nfnetlink_queue_core.c +1 -2
index a5cd6d90b78b..41583e30051b 100644
@@ -670,8 +670,7 @@ nfqnl_enqueue_packet(struct nf_queue_entry *entry, unsigned int queuenum)
 	struct nfqnl_instance *queue;
 	struct sk_buff *skb, *segs;
 	int err = -ENOBUFS;
-	struct net *net = dev_net(entry->state.in ?
-				  entry->state.in : entry->state.out);
+	struct net *net = entry->state.net;
 	struct nfnl_queue_net *q = nfnl_queue_pernet(net);
 
 	/* rcu_read_lock()ed by nf_hook_slow() */
Signature · 1 line
2.2.1
#33 lore raw

PATCH 30/30 netfilter: Pass net into okfn

In reply to Eric W. Biederman

This is immediately motivated by the bridge code that chains functions that call into netfilter. Without passing net into the okfns the bridge code would need to guess about the best expression for the network namespace to process packets in.

As net is frequently one of the first things computed in continuation functions after netfilter has done it's job passing in the desired network namespace is in many cases a code simplification.

To support this change the function dst_output_okfn is introduced to simplify passing dst_output as an okfn. For the moment dst_output_okfn just silently drops the struct net.

---
 drivers/net/vrf.c                    |  2 +-
 include/linux/netdevice.h            |  2 +-
 include/linux/netfilter.h            | 26 ++++++++++++++------------
 include/linux/netfilter_bridge.h     |  2 +-
 include/net/dn_neigh.h               |  6 +++---
 include/net/dst.h                    |  4 ++++
 include/net/ipv6.h                   |  2 +-
 include/net/netfilter/br_netfilter.h |  2 +-
 net/bridge/br_forward.c              |  5 ++---
 net/bridge/br_input.c                |  7 ++++---
 net/bridge/br_netfilter_hooks.c      | 21 +++++++++------------
 net/bridge/br_netfilter_ipv6.c       |  3 +--
 net/bridge/br_private.h              |  6 +++---
 net/bridge/br_stp_bpdu.c             |  3 ++-
 net/core/dev.c                       |  4 +++-
 net/decnet/dn_neigh.c                |  8 ++++----
 net/decnet/dn_nsp_in.c               |  3 ++-
 net/decnet/dn_route.c                |  6 +++---
 net/ipv4/arp.c                       |  7 +++----
 net/ipv4/ip_forward.c                |  3 +--
 net/ipv4/ip_input.c                  |  7 ++-----
 net/ipv4/ip_output.c                 |  4 ++--
 net/ipv4/ipmr.c                      |  4 ++--
 net/ipv4/raw.c                       |  2 +-
 net/ipv4/xfrm4_input.c               |  3 ++-
 net/ipv4/xfrm4_output.c              |  2 +-
 net/ipv6/ip6_input.c                 |  5 ++---
 net/ipv6/ip6_output.c                |  7 ++++---
 net/ipv6/ip6mr.c                     |  3 +--
 net/ipv6/mcast.c                     |  4 ++--
 net/ipv6/ndisc.c                     |  2 +-
 net/ipv6/output_core.c               |  2 +-
 net/ipv6/raw.c                       |  2 +-
 net/ipv6/xfrm6_output.c              |  2 +-
 net/netfilter/ipvs/ip_vs_xmit.c      |  4 ++--
 net/netfilter/nf_queue.c             |  2 +-
 net/xfrm/xfrm_output.c               | 12 ++++++------
 37 files changed, 95 insertions(+), 94 deletions(-)
drivers/net/vrf.c +1 -1
index c82260341b72..4dd701d7b8e6 100644
@@ -253,7 +253,7 @@ static netdev_tx_t vrf_xmit(struct sk_buff *skb, struct net_device *dev)
 }
 
 /* modelled after ip_finish_output2 */
-static int vrf_finish_output(struct sock *sk, struct sk_buff *skb)
+static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb_dst(skb);
 	struct rtable *rt = (struct rtable *)dst;
include/linux/netdevice.h +1 -1
index 97ab5c9a7069..b791405958b4 100644
@@ -2212,7 +2212,7 @@ int dev_open(struct net_device *dev);
 int dev_close(struct net_device *dev);
 int dev_close_many(struct list_head *head, bool unlink);
 void dev_disable_lro(struct net_device *dev);
-int dev_loopback_xmit(struct sock *sk, struct sk_buff *newskb);
+int dev_loopback_xmit(struct net *net, struct sock *sk, struct sk_buff *newskb);
 int dev_queue_xmit(struct sk_buff *skb);
 int dev_queue_xmit_accel(struct sk_buff *skb, void *accel_priv);
 int register_netdevice(struct net_device *dev);
include/linux/netfilter.h +14 -12
index 295f2650b5dc..0b4d4560f33d 100644
@@ -56,7 +56,7 @@ struct nf_hook_state {
 	struct sock *sk;
 	struct net *net;
 	struct list_head *hook_list;
-	int (*okfn)(struct sock *, struct sk_buff *);
+	int (*okfn)(struct net *, struct sock *, struct sk_buff *);
 };
 
 static inline void nf_hook_state_init(struct nf_hook_state *p,
@@ -67,7 +67,7 @@ static inline void nf_hook_state_init(struct nf_hook_state *p,
 				      struct net_device *outdev,
 				      struct sock *sk,
 				      struct net *net,
-				      int (*okfn)(struct sock *, struct sk_buff *))
+				      int (*okfn)(struct net *, struct sock *, struct sk_buff *))
 {
 	p->hook = hook;
 	p->thresh = thresh;
@@ -175,7 +175,7 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook,
 				 struct sk_buff *skb,
 				 struct net_device *indev,
 				 struct net_device *outdev,
-				 int (*okfn)(struct sock *, struct sk_buff *),
+				 int (*okfn)(struct net *, struct sock *, struct sk_buff *),
 				 int thresh)
 {
 	struct list_head *hook_list = &net->nf.hooks[pf][hook];
@@ -193,7 +193,7 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook,
 static inline int nf_hook(u_int8_t pf, unsigned int hook, struct net *net,
 			  struct sock *sk, struct sk_buff *skb,
 			  struct net_device *indev, struct net_device *outdev,
-			  int (*okfn)(struct sock *, struct sk_buff *))
+			  int (*okfn)(struct net *, struct sock *, struct sk_buff *))
 {
 	return nf_hook_thresh(pf, hook, net, sk, skb, indev, outdev, okfn, INT_MIN);
 }
@@ -219,31 +219,33 @@ static inline int
 NF_HOOK_THRESH(uint8_t pf, unsigned int hook, struct net *net, struct sock *sk,
 	       struct sk_buff *skb, struct net_device *in,
 	       struct net_device *out,
-	       int (*okfn)(struct sock *, struct sk_buff *), int thresh)
+	       int (*okfn)(struct net *, struct sock *, struct sk_buff *),
+	       int thresh)
 {
 	int ret = nf_hook_thresh(pf, hook, net, sk, skb, in, out, okfn, thresh);
 	if (ret == 1)
-		ret = okfn(sk, skb);
+		ret = okfn(net, sk, skb);
 	return ret;
 }
 
 static inline int
 NF_HOOK_COND(uint8_t pf, unsigned int hook, struct net *net, struct sock *sk,
 	     struct sk_buff *skb, struct net_device *in, struct net_device *out,
-	     int (*okfn)(struct sock *, struct sk_buff *), bool cond)
+	     int (*okfn)(struct net *, struct sock *, struct sk_buff *),
+	     bool cond)
 {
 	int ret;
 
 	if (!cond ||
 	    ((ret = nf_hook_thresh(pf, hook, net, sk, skb, in, out, okfn, INT_MIN)) == 1))
-		ret = okfn(sk, skb);
+		ret = okfn(net, sk, skb);
 	return ret;
 }
 
 static inline int
 NF_HOOK(uint8_t pf, unsigned int hook, struct net *net, struct sock *sk, struct sk_buff *skb,
 	struct net_device *in, struct net_device *out,
-	int (*okfn)(struct sock *, struct sk_buff *))
+	int (*okfn)(struct net *, struct sock *, struct sk_buff *))
 {
 	return NF_HOOK_THRESH(pf, hook, net, sk, skb, in, out, okfn, INT_MIN);
 }
@@ -345,12 +347,12 @@ nf_nat_decode_session(struct sk_buff *skb, struct flowi *fl, u_int8_t family)
 }
 
 #else /* !CONFIG_NETFILTER */
-#define NF_HOOK(pf, hook, net, sk, skb, indev, outdev, okfn) (okfn)(sk, skb)
-#define NF_HOOK_COND(pf, hook, net, sk, skb, indev, outdev, okfn, cond) (okfn)(sk, skb)
+#define NF_HOOK(pf, hook, net, sk, skb, indev, outdev, okfn) (okfn)(net, sk, skb)
+#define NF_HOOK_COND(pf, hook, net, sk, skb, indev, outdev, okfn, cond) (okfn)(net, sk, skb)
 static inline int nf_hook(u_int8_t pf, unsigned int hook, struct net *net,
 			  struct sock *sk, struct sk_buff *skb,
 			  struct net_device *indev, struct net_device *outdev,
-			  int (*okfn)(struct sock *, struct sk_buff *))
+			  int (*okfn)(struct net *, struct sock *, struct sk_buff *))
 {
 	return 1;
 }
include/linux/netfilter_bridge.h +1 -1
index 2437b8a5d7a9..2ed40c402b5e 100644
@@ -17,7 +17,7 @@ enum nf_br_hook_priorities {
 
 #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 
-int br_handle_frame_finish(struct sock *sk, struct sk_buff *skb);
+int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb);
 
 static inline void br_drop_fake_rtable(struct sk_buff *skb)
 {
include/net/dn_neigh.h +3 -3
index d0424269313f..5e902fc3f4eb 100644
@@ -18,11 +18,11 @@ struct dn_neigh {
 
 void dn_neigh_init(void);
 void dn_neigh_cleanup(void);
-int dn_neigh_router_hello(struct sock *sk, struct sk_buff *skb);
-int dn_neigh_endnode_hello(struct sock *sk, struct sk_buff *skb);
+int dn_neigh_router_hello(struct net *net, struct sock *sk, struct sk_buff *skb);
+int dn_neigh_endnode_hello(struct net *net, struct sock *sk, struct sk_buff *skb);
 void dn_neigh_pointopoint_hello(struct sk_buff *skb);
 int dn_neigh_elist(struct net_device *dev, unsigned char *ptr, int n);
-int dn_to_neigh_output(struct sock *sk, struct sk_buff *skb);
+int dn_to_neigh_output(struct net *net, struct sock *sk, struct sk_buff *skb);
 
 extern struct neigh_table dn_neigh_table;
 
include/net/dst.h +4 -0
index c72e58474e52..df0481a07029 100644
@@ -458,6 +458,10 @@ static inline int dst_output(struct sock *sk, struct sk_buff *skb)
 {
 	return skb_dst(skb)->output(sk, skb);
 }
+static inline int dst_output_okfn(struct net *net, struct sock *sk, struct sk_buff *skb)
+{
+	return dst_output(sk, skb);
+}
 
 /* Input packet from network to transport.  */
 static inline int dst_input(struct sk_buff *skb)
include/net/ipv6.h +1 -1
index 711cca428cc8..384a93cf07d6 100644
@@ -807,7 +807,7 @@ static inline u8 ip6_tclass(__be32 flowinfo)
 int ipv6_rcv(struct sk_buff *skb, struct net_device *dev,
 	     struct packet_type *pt, struct net_device *orig_dev);
 
-int ip6_rcv_finish(struct sock *sk, struct sk_buff *skb);
+int ip6_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb);
 
 /*
  *	upper-layer output functions
include/net/netfilter/br_netfilter.h +1 -1
index d4c6b5f30acd..8fe266504900 100644
@@ -31,7 +31,7 @@ static inline void nf_bridge_push_encap_header(struct sk_buff *skb)
 	skb->network_header -= len;
 }
 
-int br_nf_pre_routing_finish_bridge(struct sock *sk, struct sk_buff *skb);
+int br_nf_pre_routing_finish_bridge(struct net *net, struct sock *sk, struct sk_buff *skb);
 
 static inline struct rtable *bridge_parent_rtable(const struct net_device *dev)
 {
net/bridge/br_forward.c +2 -3
index 2dd2a23ce707..48afca729ed7 100644
@@ -35,7 +35,7 @@ static inline int should_deliver(const struct net_bridge_port *p,
 		p->state == BR_STATE_FORWARDING;
 }
 
-int br_dev_queue_push_xmit(struct sock *sk, struct sk_buff *skb)
+int br_dev_queue_push_xmit(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	if (!is_skb_forwardable(skb->dev, skb))
 		goto drop;
@@ -65,9 +65,8 @@ drop:
 }
 EXPORT_SYMBOL_GPL(br_dev_queue_push_xmit);
 
-int br_forward_finish(struct sock *sk, struct sk_buff *skb)
+int br_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	struct net *net = dev_net(skb->dev);
 	return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING,
 		       net, sk, skb, NULL, skb->dev,
 		       br_dev_queue_push_xmit);
net/bridge/br_input.c +4 -3
index 78fa7acd836e..223f4040d9df 100644
@@ -26,7 +26,8 @@
 br_should_route_hook_t __rcu *br_should_route_hook __read_mostly;
 EXPORT_SYMBOL(br_should_route_hook);
 
-static int br_netif_receive_skb(struct sock *sk, struct sk_buff *skb)
+static int
+br_netif_receive_skb(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	return netif_receive_skb(skb);
 }
@@ -125,7 +126,7 @@ static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br,
 }
 
 /* note: already called with rcu_read_lock */
-int br_handle_frame_finish(struct sock *sk, struct sk_buff *skb)
+int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	const unsigned char *dest = eth_hdr(skb)->h_dest;
 	struct net_bridge_port *p = br_port_get_rcu(skb->dev);
@@ -213,7 +214,7 @@ drop:
 EXPORT_SYMBOL_GPL(br_handle_frame_finish);
 
 /* note: already called with rcu_read_lock */
-static int br_handle_local_finish(struct sock *sk, struct sk_buff *skb)
+static int br_handle_local_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct net_bridge_port *p = br_port_get_rcu(skb->dev);
 	u16 vid = 0;
net/bridge/br_netfilter_hooks.c +9 -12
index 7886c9d7e23d..e6e76bbdc82f 100644
@@ -256,7 +256,7 @@ void nf_bridge_update_protocol(struct sk_buff *skb)
  * don't, we use the neighbour framework to find out. In both cases, we make
  * sure that br_handle_frame_finish() is called afterwards.
  */
-int br_nf_pre_routing_finish_bridge(struct sock *sk, struct sk_buff *skb)
+int br_nf_pre_routing_finish_bridge(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct neighbour *neigh;
 	struct dst_entry *dst;
@@ -273,7 +273,7 @@ int br_nf_pre_routing_finish_bridge(struct sock *sk, struct sk_buff *skb)
 		if (neigh->hh.hh_len) {
 			neigh_hh_bridge(&neigh->hh, skb);
 			skb->dev = nf_bridge->physindev;
-			ret = br_handle_frame_finish(sk, skb);
+			ret = br_handle_frame_finish(net, sk, skb);
 		} else {
 			/* the neighbour function below overwrites the complete
 			 * MAC header, so we save the Ethernet source address and
@@ -342,11 +342,10 @@ br_nf_ipv4_daddr_was_changed(const struct sk_buff *skb,
  * device, we proceed as if ip_route_input() succeeded. If it differs from the
  * logical bridge port or if ip_route_output_key() fails we drop the packet.
  */
-static int br_nf_pre_routing_finish(struct sock *sk, struct sk_buff *skb)
+static int br_nf_pre_routing_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
 	struct iphdr *iph = ip_hdr(skb);
-	struct net *net = dev_net(dev);
 	struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb);
 	struct rtable *rt;
 	int err;
@@ -536,10 +535,9 @@ static unsigned int br_nf_local_in(const struct nf_hook_ops *ops,
 }
 
 /* PF_BRIDGE/FORWARD *************************************************/
-static int br_nf_forward_finish(struct sock *sk, struct sk_buff *skb)
+static int br_nf_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb);
-	struct net *net = dev_net(skb->dev);
 	struct net_device *in;
 
 	if (!IS_ARP(skb) && !IS_VLAN_ARP(skb)) {
@@ -692,7 +690,7 @@ static int br_nf_push_frag_xmit(struct net *net, struct sock *sk, struct sk_buff
 	__skb_push(skb, data->encap_size);
 
 	nf_bridge_info_free(skb);
-	return br_dev_queue_push_xmit(sk, skb);
+	return br_dev_queue_push_xmit(net, sk, skb);
 }
 static int br_nf_push_frag_xmit_sk(struct sock *sk, struct sk_buff *skb)
 {
@@ -728,17 +726,16 @@ static unsigned int nf_bridge_mtu_reduction(const struct sk_buff *skb)
 	return 0;
 }
 
-static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb)
+static int br_nf_dev_queue_xmit(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct nf_bridge_info *nf_bridge;
 	unsigned int mtu_reserved;
-	struct net *net = dev_net(skb_dst(skb)->dev);
 
 	mtu_reserved = nf_bridge_mtu_reduction(skb);
 
 	if (skb_is_gso(skb) || skb->len + mtu_reserved <= skb->dev->mtu) {
 		nf_bridge_info_free(skb);
-		return br_dev_queue_push_xmit(sk, skb);
+		return br_dev_queue_push_xmit(net, sk, skb);
 	}
 
 	nf_bridge = nf_bridge_info_get(skb);
@@ -797,7 +794,7 @@ static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb)
 	}
 #endif
 	nf_bridge_info_free(skb);
-	return br_dev_queue_push_xmit(sk, skb);
+	return br_dev_queue_push_xmit(net, sk, skb);
  drop:
 	kfree_skb(skb);
 	return 0;
@@ -887,7 +884,7 @@ static void br_nf_pre_routing_finish_bridge_slow(struct sk_buff *skb)
 	skb->dev = nf_bridge->physindev;
 
 	nf_bridge->physoutdev = NULL;
-	br_handle_frame_finish(NULL, skb);
+	br_handle_frame_finish(dev_net(skb->dev), NULL, skb);
 }
 
 static int br_nf_dev_xmit(struct sk_buff *skb)
net/bridge/br_netfilter_ipv6.c +1 -2
index 5d19361ad5d3..e4dbbe44c724 100644
@@ -161,12 +161,11 @@ br_nf_ipv6_daddr_was_changed(const struct sk_buff *skb,
  * for br_nf_pre_routing_finish(), same logic is used here but
  * equivalent IPv6 function ip6_route_input() called indirectly.
  */
-static int br_nf_pre_routing_finish_ipv6(struct sock *sk, struct sk_buff *skb)
+static int br_nf_pre_routing_finish_ipv6(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb);
 	struct rtable *rt;
 	struct net_device *dev = skb->dev;
-	struct net *net = dev_net(dev);
 	const struct nf_ipv6_ops *v6ops = nf_get_ipv6_ops();
 
 	nf_bridge->frag_max_size = IP6CB(skb)->frag_max_size;
net/bridge/br_private.h +3 -3
index 213baf7aaa93..74e99c75c8e4 100644
@@ -413,10 +413,10 @@ int br_fdb_external_learn_del(struct net_bridge *br, struct net_bridge_port *p,
 
 /* br_forward.c */
 void br_deliver(const struct net_bridge_port *to, struct sk_buff *skb);
-int br_dev_queue_push_xmit(struct sock *sk, struct sk_buff *skb);
+int br_dev_queue_push_xmit(struct net *net, struct sock *sk, struct sk_buff *skb);
 void br_forward(const struct net_bridge_port *to,
 		struct sk_buff *skb, struct sk_buff *skb0);
-int br_forward_finish(struct sock *sk, struct sk_buff *skb);
+int br_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb);
 void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb, bool unicast);
 void br_flood_forward(struct net_bridge *br, struct sk_buff *skb,
 		      struct sk_buff *skb2, bool unicast);
@@ -434,7 +434,7 @@ void br_port_flags_change(struct net_bridge_port *port, unsigned long mask);
 void br_manage_promisc(struct net_bridge *br);
 
 /* br_input.c */
-int br_handle_frame_finish(struct sock *sk, struct sk_buff *skb);
+int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb);
 rx_handler_result_t br_handle_frame(struct sk_buff **pskb);
 
 static inline bool br_rx_handler_check_rcu(const struct net_device *dev)
net/bridge/br_stp_bpdu.c +2 -1
index 8e2e8c352198..5881fbc114a9 100644
@@ -30,7 +30,8 @@
 
 #define LLC_RESERVE sizeof(struct llc_pdu_un)
 
-static int br_send_bpdu_finish(struct sock *sk, struct sk_buff *skb)
+static int br_send_bpdu_finish(struct net *net, struct sock *sk,
+			       struct sk_buff *skb)
 {
 	return dev_queue_xmit(skb);
 }
net/core/dev.c +3 -1
index 7db9b012dfb7..00dccfac8939 100644
@@ -2915,9 +2915,11 @@ EXPORT_SYMBOL(xmit_recursion);
 
 /**
  *	dev_loopback_xmit - loop back @skb
+ *	@net: network namespace this loopback is happening in
+ *	@sk:  sk needed to be a netfilter okfn
  *	@skb: buffer to transmit
  */
-int dev_loopback_xmit(struct sock *sk, struct sk_buff *skb)
+int dev_loopback_xmit(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	skb_reset_mac_header(skb);
 	__skb_pull(skb, skb_network_offset(skb));
net/decnet/dn_neigh.c +4 -4
index 305ab2fe25cd..482730cd8a56 100644
@@ -194,7 +194,7 @@ static int dn_neigh_output(struct neighbour *neigh, struct sk_buff *skb)
 	return err;
 }
 
-static int dn_neigh_output_packet(struct sock *sk, struct sk_buff *skb)
+static int dn_neigh_output_packet(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb_dst(skb);
 	struct dn_route *rt = (struct dn_route *)dst;
@@ -334,7 +334,7 @@ static int dn_phase3_output(struct neighbour *neigh, struct sock *sk,
 		       dn_neigh_output_packet);
 }
 
-int dn_to_neigh_output(struct sock *sk, struct sk_buff *skb)
+int dn_to_neigh_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb_dst(skb);
 	struct dn_route *rt = (struct dn_route *) dst;
@@ -378,7 +378,7 @@ void dn_neigh_pointopoint_hello(struct sk_buff *skb)
 /*
  * Ethernet router hello message received
  */
-int dn_neigh_router_hello(struct sock *sk, struct sk_buff *skb)
+int dn_neigh_router_hello(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct rtnode_hello_message *msg = (struct rtnode_hello_message *)skb->data;
 
@@ -440,7 +440,7 @@ int dn_neigh_router_hello(struct sock *sk, struct sk_buff *skb)
 /*
  * Endnode hello message received
  */
-int dn_neigh_endnode_hello(struct sock *sk, struct sk_buff *skb)
+int dn_neigh_endnode_hello(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct endnode_hello_message *msg = (struct endnode_hello_message *)skb->data;
 	struct neighbour *neigh;
net/decnet/dn_nsp_in.c +2 -1
index e7b0605ca34a..7ac086d5c0c0 100644
@@ -714,7 +714,8 @@ out:
 	return ret;
 }
 
-static int dn_nsp_rx_packet(struct sock *sk2, struct sk_buff *skb)
+static int dn_nsp_rx_packet(struct net *net, struct sock *sk2,
+			    struct sk_buff *skb)
 {
 	struct dn_skb_cb *cb = DN_SKB_CB(skb);
 	struct sock *sk = NULL;
net/decnet/dn_route.c +3 -3
index fefcd2e85ef9..e930321e2c1d 100644
@@ -512,7 +512,7 @@ static int dn_return_long(struct sk_buff *skb)
  *
  * Returns: result of input function if route is found, error code otherwise
  */
-static int dn_route_rx_packet(struct sock *sk, struct sk_buff *skb)
+static int dn_route_rx_packet(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct dn_skb_cb *cb;
 	int err;
@@ -610,7 +610,7 @@ drop_it:
 	return NET_RX_DROP;
 }
 
-static int dn_route_discard(struct sock *sk, struct sk_buff *skb)
+static int dn_route_discard(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	/*
 	 * I know we drop the packet here, but thats considered success in
@@ -620,7 +620,7 @@ static int dn_route_discard(struct sock *sk, struct sk_buff *skb)
 	return NET_RX_SUCCESS;
 }
 
-static int dn_route_ptp_hello(struct sock *sk, struct sk_buff *skb)
+static int dn_route_ptp_hello(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	dn_dev_hello(skb);
 	dn_neigh_pointopoint_hello(skb);
net/ipv4/arp.c +3 -4
index ae71e9ade5f9..61ff5ea31283 100644
@@ -621,7 +621,7 @@ out:
 }
 EXPORT_SYMBOL(arp_create);
 
-static int arp_xmit_finish(struct sock *sk, struct sk_buff *skb)
+static int arp_xmit_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	return dev_queue_xmit(skb);
 }
@@ -642,7 +642,7 @@ EXPORT_SYMBOL(arp_xmit);
  *	Process an arp request.
  */
 
-static int arp_process(struct sock *sk, struct sk_buff *skb)
+static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
 	struct in_device *in_dev = __in_dev_get_rcu(dev);
@@ -654,7 +654,6 @@ static int arp_process(struct sock *sk, struct sk_buff *skb)
 	u16 dev_type = dev->type;
 	int addr_type;
 	struct neighbour *n;
-	struct net *net = dev_net(dev);
 	bool is_garp = false;
 
 	/* arp_rcv below verifies the ARP header and verifies the device
@@ -865,7 +864,7 @@ out:
 
 static void parp_redo(struct sk_buff *skb)
 {
-	arp_process(NULL, skb);
+	arp_process(dev_net(skb->dev), NULL, skb);
 }
 
 
net/ipv4/ip_forward.c +1 -2
index 0a3c45a2e757..d66cfb35ba74 100644
@@ -61,9 +61,8 @@ static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu)
 }
 
 
-static int ip_forward_finish(struct sock *sk, struct sk_buff *skb)
+static int ip_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	struct net *net = dev_net(skb_dst(skb)->dev);
 	struct ip_options *opt	= &(IPCB(skb)->opt);
 
 	IP_INC_STATS_BH(net, IPSTATS_MIB_OUTFORWDATAGRAMS);
net/ipv4/ip_input.c +2 -5
index 991d082c7312..7cc9f7bb7fb7 100644
@@ -188,10 +188,8 @@ bool ip_call_ra_chain(struct sk_buff *skb)
 	return false;
 }
 
-static int ip_local_deliver_finish(struct sock *sk, struct sk_buff *skb)
+static int ip_local_deliver_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	struct net *net = dev_net(skb->dev);
-
 	__skb_pull(skb, skb_network_header_len(skb));
 
 	rcu_read_lock();
@@ -311,10 +309,9 @@ drop:
 int sysctl_ip_early_demux __read_mostly = 1;
 EXPORT_SYMBOL(sysctl_ip_early_demux);
 
-static int ip_rcv_finish(struct sock *sk, struct sk_buff *skb)
+static int ip_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	const struct iphdr *iph = ip_hdr(skb);
-	struct net *net = dev_net(skb->dev);
 	struct rtable *rt;
 
 	if (sysctl_ip_early_demux && !skb_dst(skb) && !skb->sk) {
net/ipv4/ip_output.c +2 -2
index 4c9532259a7f..09a6b7bb7ea3 100644
@@ -104,7 +104,7 @@ static int __ip_local_out_sk(struct sock *sk, struct sk_buff *skb)
 	ip_send_check(iph);
 	return nf_hook(NFPROTO_IPV4, NF_INET_LOCAL_OUT,
 		       net, sk, skb, NULL, skb_dst(skb)->dev,
-		       dst_output);
+		       dst_output_okfn);
 }
 
 int __ip_local_out(struct sk_buff *skb)
@@ -266,7 +266,7 @@ static int ip_finish_output_gso(struct sock *sk, struct sk_buff *skb,
 	return ret;
 }
 
-static int ip_finish_output(struct sock *sk, struct sk_buff *skb)
+static int ip_finish_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	unsigned int mtu;
 
net/ipv4/ipmr.c +2 -2
index a88c0c5374ff..cfcb996ec51b 100644
@@ -1678,10 +1678,10 @@ static void ip_encap(struct net *net, struct sk_buff *skb,
 	nf_reset(skb);
 }
 
-static inline int ipmr_forward_finish(struct sock *sk, struct sk_buff *skb)
+static inline int ipmr_forward_finish(struct net *net, struct sock *sk,
+				      struct sk_buff *skb)
 {
 	struct ip_options *opt = &(IPCB(skb)->opt);
-	struct net *net = dev_net(skb_dst(skb)->dev);
 
 	IP_INC_STATS_BH(net, IPSTATS_MIB_OUTFORWDATAGRAMS);
 	IP_ADD_STATS_BH(net, IPSTATS_MIB_OUTOCTETS, skb->len);
net/ipv4/raw.c +1 -1
index 2045b1aaa6ef..28ef8a913130 100644
@@ -413,7 +413,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
 
 	err = NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_OUT,
 		      net, sk, skb, NULL, rt->dst.dev,
-		      dst_output);
+		      dst_output_okfn);
 	if (err > 0)
 		err = net_xmit_errno(err);
 	if (err)
net/ipv4/xfrm4_input.c +2 -1
index 5093000d3d5e..62e1e72db461 100644
@@ -22,7 +22,8 @@ int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb)
 	return xfrm4_extract_header(skb);
 }
 
-static inline int xfrm4_rcv_encap_finish(struct sock *sk, struct sk_buff *skb)
+static inline int xfrm4_rcv_encap_finish(struct net *net, struct sock *sk,
+					 struct sk_buff *skb)
 {
 	if (!skb_dst(skb)) {
 		const struct iphdr *iph = ip_hdr(skb);
net/ipv4/xfrm4_output.c +1 -1
index e4a85199e015..28ae2048b93a 100644
@@ -80,7 +80,7 @@ int xfrm4_output_finish(struct sock *sk, struct sk_buff *skb)
 	return xfrm_output(sk, skb);
 }
 
-static int __xfrm4_output(struct sock *sk, struct sk_buff *skb)
+static int __xfrm4_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct xfrm_state *x = skb_dst(skb)->xfrm;
 
net/ipv6/ip6_input.c +2 -3
index 583cf959c23d..9075acf081dd 100644
@@ -47,7 +47,7 @@
 #include <net/inet_ecn.h>
 #include <net/dst_metadata.h>
 
-int ip6_rcv_finish(struct sock *sk, struct sk_buff *skb)
+int ip6_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	if (sysctl_ip_early_demux && !skb_dst(skb) && skb->sk == NULL) {
 		const struct inet6_protocol *ipprot;
@@ -199,9 +199,8 @@ drop:
  */
 
 
-static int ip6_input_finish(struct sock *sk, struct sk_buff *skb)
+static int ip6_input_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	struct net *net = dev_net(skb_dst(skb)->dev);
 	const struct inet6_protocol *ipprot;
 	struct inet6_dev *idev;
 	unsigned int nhoff;
net/ipv6/ip6_output.c +4 -3
index 96e76ddd4a44..d8d68e81d123 100644
@@ -121,7 +121,7 @@ static int ip6_finish_output2(struct sock *sk, struct sk_buff *skb)
 	return -EINVAL;
 }
 
-static int ip6_finish_output(struct sock *sk, struct sk_buff *skb)
+static int ip6_finish_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	if ((skb->len > ip6_skb_dst_mtu(skb) && !skb_is_gso(skb)) ||
 	    dst_allfrag(skb_dst(skb)) ||
@@ -225,7 +225,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6,
 			      IPSTATS_MIB_OUT, skb->len);
 		return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
 			       net, sk, skb, NULL, dst->dev,
-			       dst_output);
+			       dst_output_okfn);
 	}
 
 	skb->dev = dst->dev;
@@ -317,7 +317,8 @@ static int ip6_forward_proxy_check(struct sk_buff *skb)
 	return 0;
 }
 
-static inline int ip6_forward_finish(struct sock *sk, struct sk_buff *skb)
+static inline int ip6_forward_finish(struct net *net, struct sock *sk,
+				     struct sk_buff *skb)
 {
 	skb_sender_cpu_clear(skb);
 	return dst_output(sk, skb);
net/ipv6/ip6mr.c +1 -2
index e830942b2090..5e5d16e7ce85 100644
@@ -1985,9 +1985,8 @@ int ip6mr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg)
 }
 #endif
 
-static inline int ip6mr_forward2_finish(struct sock *sk, struct sk_buff *skb)
+static inline int ip6mr_forward2_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	struct net *net = dev_net(skb_dst(skb)->dev);
 	IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 			 IPSTATS_MIB_OUTFORWDATAGRAMS);
 	IP6_ADD_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
net/ipv6/mcast.c +2 -2
index 124338a39e29..a8bf57ca74d3 100644
@@ -1646,7 +1646,7 @@ static void mld_sendpack(struct sk_buff *skb)
 
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
 		      net, net->ipv6.igmp_sk, skb, NULL, skb->dev,
-		      dst_output);
+		      dst_output_okfn);
 out:
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(net, idev, ICMPV6_MLD2_REPORT);
@@ -2010,7 +2010,7 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
 	skb_dst_set(skb, dst);
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
 		      net, sk, skb, NULL, skb->dev,
-		      dst_output);
+		      dst_output_okfn);
 out:
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(net, idev, type);
net/ipv6/ndisc.c +1 -1
index dd2b08d7c8d1..dde5a1e5875a 100644
@@ -465,7 +465,7 @@ static void ndisc_send_skb(struct sk_buff *skb,
 
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
 		      net, sk, skb, NULL, dst->dev,
-		      dst_output);
+		      dst_output_okfn);
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(net, idev, type);
 		ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTMSGS);
net/ipv6/output_core.c +1 -1
index 9cc9127fb5e7..e77102c4f804 100644
@@ -151,7 +151,7 @@ static int __ip6_local_out_sk(struct sock *sk, struct sk_buff *skb)
 
 	return nf_hook(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
 		       net, sk, skb, NULL, skb_dst(skb)->dev,
-		       dst_output);
+		       dst_output_okfn);
 }
 
 int __ip6_local_out(struct sk_buff *skb)
net/ipv6/raw.c +1 -1
index dc65ec198f7c..fec0151522a2 100644
@@ -655,7 +655,7 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length,
 
 	IP6_UPD_PO_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len);
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, net, sk, skb,
-		      NULL, rt->dst.dev, dst_output);
+		      NULL, rt->dst.dev, dst_output_okfn);
 	if (err > 0)
 		err = net_xmit_errno(err);
 	if (err)
net/ipv6/xfrm6_output.c +1 -1
index 431ae2c22234..68a996f8a044 100644
@@ -131,7 +131,7 @@ int xfrm6_output_finish(struct sock *sk, struct sk_buff *skb)
 	return xfrm_output(sk, skb);
 }
 
-static int __xfrm6_output(struct sock *sk, struct sk_buff *skb)
+static int __xfrm6_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb_dst(skb);
 	struct xfrm_state *x = dst->xfrm;
net/netfilter/ipvs/ip_vs_xmit.c +2 -2
index 65c996c14bca..cc7299033af8 100644
@@ -574,7 +574,7 @@ static inline int ip_vs_nat_send_or_cont(int pf, struct sk_buff *skb,
 		if (!skb->sk)
 			skb_sender_cpu_clear(skb);
 		NF_HOOK(pf, NF_INET_LOCAL_OUT, ip_vs_conn_net(cp), NULL, skb,
-			NULL, skb_dst(skb)->dev, dst_output);
+			NULL, skb_dst(skb)->dev, dst_output_okfn);
 	} else
 		ret = NF_ACCEPT;
 
@@ -596,7 +596,7 @@ static inline int ip_vs_send_or_cont(int pf, struct sk_buff *skb,
 		if (!skb->sk)
 			skb_sender_cpu_clear(skb);
 		NF_HOOK(pf, NF_INET_LOCAL_OUT, ip_vs_conn_net(cp), NULL, skb,
-			NULL, skb_dst(skb)->dev, dst_output);
+			NULL, skb_dst(skb)->dev, dst_output_okfn);
 	} else
 		ret = NF_ACCEPT;
 	return ret;
net/netfilter/nf_queue.c +1 -1
index 96777f9a9350..9f3c3c25fa73 100644
@@ -215,7 +215,7 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict)
 	case NF_ACCEPT:
 	case NF_STOP:
 		local_bh_disable();
-		entry->state.okfn(entry->state.sk, skb);
+		entry->state.okfn(entry->state.net, entry->state.sk, skb);
 		local_bh_enable();
 		break;
 	case NF_QUEUE:
net/xfrm/xfrm_output.c +6 -6
index c21f1a02ce13..61ba99f61dc8 100644
@@ -19,7 +19,7 @@
 #include <net/dst.h>
 #include <net/xfrm.h>
 
-static int xfrm_output2(struct sock *sk, struct sk_buff *skb);
+static int xfrm_output2(struct net *net, struct sock *sk, struct sk_buff *skb);
 
 static int xfrm_skb_check_space(struct sk_buff *skb)
 {
@@ -157,12 +157,12 @@ out:
 }
 EXPORT_SYMBOL_GPL(xfrm_output_resume);
 
-static int xfrm_output2(struct sock *sk, struct sk_buff *skb)
+static int xfrm_output2(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	return xfrm_output_resume(skb, 1);
 }
 
-static int xfrm_output_gso(struct sock *sk, struct sk_buff *skb)
+static int xfrm_output_gso(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct sk_buff *segs;
 
@@ -178,7 +178,7 @@ static int xfrm_output_gso(struct sock *sk, struct sk_buff *skb)
 		int err;
 
 		segs->next = NULL;
-		err = xfrm_output2(sk, segs);
+		err = xfrm_output2(net, sk, segs);
 
 		if (unlikely(err)) {
 			kfree_skb_list(nskb);
@@ -197,7 +197,7 @@ int xfrm_output(struct sock *sk, struct sk_buff *skb)
 	int err;
 
 	if (skb_is_gso(skb))
-		return xfrm_output_gso(sk, skb);
+		return xfrm_output_gso(net, sk, skb);
 
 	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		err = skb_checksum_help(skb);
@@ -208,7 +208,7 @@ int xfrm_output(struct sock *sk, struct sk_buff *skb)
 		}
 	}
 
-	return xfrm_output2(sk, skb);
+	return xfrm_output2(net, sk, skb);
 }
 EXPORT_SYMBOL_GPL(xfrm_output);
 
Signature · 1 line
2.2.1
Acked-by on the v1 series
#34 lore raw
In reply to Eric W. Biederman
Quoted 25 lines · Eric W. Biederman

My primary goal with this patchset and it's follow ups is to cleanup the network routing paths so that we do not look at the output device to derive the network namespace. My plan is to pass the network namespace of the transmitting socket through the output path, to replace code that looks at the output network device today. Once that is done we can have routes with output devices outside of the current network namespace. Which should allow reception and transmission of packets in network namespaces to be as fast as normal packet reception and transmission with early demux disabled, because it will same code path.

Once skb_dst(skb)->dev is a little better under control I think it will also be possible to use rcu to cleanup the ancient hack that sets dst->dev to loopback_dev when a network device is removed.

The work to get there is a series of code cleanups. I am starting with passing net into the netfilter hooks and into the functions that are called after the netfilter hooks. This removes from netfilter the need to guess which network namespace it is working on.

To get there I perform a series of minor prep patches so the big changes at the end are possible to audit without getting lost in the noise. In particular I have a lot of patches computing net into a local variable and then using it through out the function.

Quoted 6 lines

So this patchset encompases removing dead code, sorting out the _sk functions that were added last time someone pushed a prototype change through the post netfilter functions. Cleaning up individual functions use of the network namespace. Passing net into the netfilter hooks. Passing net into the post netfilter functions. Using state->net in the netfilter code where it is available and trivially usable.

LGTM (except some minor comments).

--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#35 lore raw
Quoted 32 lines · Nicolas Dichtel
Quoted 25 lines · Eric W. Biederman

My primary goal with this patchset and it's follow ups is to cleanup the network routing paths so that we do not look at the output device to derive the network namespace. My plan is to pass the network namespace of the transmitting socket through the output path, to replace code that looks at the output network device today. Once that is done we can have routes with output devices outside of the current network namespace. Which should allow reception and transmission of packets in network namespaces to be as fast as normal packet reception and transmission with early demux disabled, because it will same code path.

Once skb_dst(skb)->dev is a little better under control I think it will also be possible to use rcu to cleanup the ancient hack that sets dst->dev to loopback_dev when a network device is removed.

The work to get there is a series of code cleanups. I am starting with passing net into the netfilter hooks and into the functions that are called after the netfilter hooks. This removes from netfilter the need to guess which network namespace it is working on.

To get there I perform a series of minor prep patches so the big changes at the end are possible to audit without getting lost in the noise. In particular I have a lot of patches computing net into a local variable and then using it through out the function.

Quoted 6 lines

So this patchset encompases removing dead code, sorting out the _sk functions that were added last time someone pushed a prototype change through the post netfilter functions. Cleaning up individual functions use of the network namespace. Passing net into the netfilter hooks. Passing net into the post netfilter functions. Using state->net in the netfilter code where it is available and trivially usable.

Quoted 3 lines

LGTM (except some minor comments).

Thanks for review. I have added an extra patch for the missing blank lines that are still missing after the entire series. As they affect neither code correctness nor bisectability I don't think there is any point respinning the indivdual patches.

Eric
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#36 lore raw

Le 18/09/2015 00:24, Eric W. Biederman a écrit :
[snip]

Quoted 4 lines

Thanks for review. I have added an extra patch for the missing blank lines that are still missing after the entire series. As they affect neither code correctness nor bisectability I don't think there is any point respinning the indivdual patches.

It's fine for me, thanks.
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#37 lore raw

[PATCH next 31/30] netfilter: Add blank lines in callers of netfilter hooks

In reply to Eric W. Biederman

In code review it was noticed that I had failed to add some blank lines in places where they are customarily used. Taking a second look at the code I have to agree blank lines would be nice so I have added them here.

---
 net/ipv4/xfrm4_output.c | 1 +
 net/ipv6/ip6_output.c   | 1 +
 net/ipv6/xfrm6_output.c | 1 +
 net/xfrm/xfrm_output.c  | 1 +
 4 files changed, 4 insertions(+)
net/ipv4/xfrm4_output.c +1 -0
index 28ae2048b93a..cd6be736e19f 100644
@@ -97,6 +97,7 @@ static int __xfrm4_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 int xfrm4_output(struct sock *sk, struct sk_buff *skb)
 {
 	struct net *net = dev_net(skb_dst(skb)->dev);
+
 	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING,
 			    net, sk, skb, NULL, skb_dst(skb)->dev,
 			    __xfrm4_output,
net/ipv6/ip6_output.c +1 -0
index d8d68e81d123..291a07be5dfb 100644
@@ -136,6 +136,7 @@ int ip6_output(struct sock *sk, struct sk_buff *skb)
 	struct net_device *dev = skb_dst(skb)->dev;
 	struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb));
 	struct net *net = dev_net(dev);
+
 	if (unlikely(idev->cnf.disable_ipv6)) {
 		IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS);
 		kfree_skb(skb);
net/ipv6/xfrm6_output.c +1 -0
index 68a996f8a044..0c3e9ffcf231 100644
@@ -169,6 +169,7 @@ static int __xfrm6_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 int xfrm6_output(struct sock *sk, struct sk_buff *skb)
 {
 	struct net *net = dev_net(skb_dst(skb)->dev);
+
 	return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING,
 			    net, sk, skb,  NULL, skb_dst(skb)->dev,
 			    __xfrm6_output,
net/xfrm/xfrm_output.c +1 -0
index 61ba99f61dc8..c48a4b8582bb 100644
@@ -132,6 +132,7 @@ out:
 int xfrm_output_resume(struct sk_buff *skb, int err)
 {
 	struct net *net = xs_net(skb_dst(skb)->xfrm);
+
 	while (likely((err = xfrm_output_one(skb, err)) == 0)) {
 		nf_reset(skb);
 
Signature · 1 line
2.2.1
maintainer
#38 lore raw
In reply to Eric W. Biederman

From: [email protected] (Eric W. Biederman)
Date: Tue, 15 Sep 2015 19:59:49 -0500

Quoted 4 lines

Pablo, Dave I don't know whose tree this makes more sense to go through. I am assuming at least initially Pablos as netfilter is involved. From what I have seen there will be a lot of back and forth between the netfilter code paths and the routing code paths.

I think it might reduce conflicts actually if it went via my net-next tree.

Pablo, any objections?

maintainer
#39 lore raw

From: David Miller <[email protected]>
Date: Thu, 17 Sep 2015 17:19:04 -0700 (PDT)

Quoted 8 lines

From: [email protected] (Eric W. Biederman)
Date: Tue, 15 Sep 2015 19:59:49 -0500

Quoted 4 lines

Pablo, Dave I don't know whose tree this makes more sense to go through. I am assuming at least initially Pablos as netfilter is involved. From what I have seen there will be a lot of back and forth between the netfilter code paths and the routing code paths.

Quoted 4 lines

I think it might reduce conflicts actually if it went via my net-next tree.

Pablo, any objections?

I actually decided to just push it out to my tree, if there are any problems with that I will revert.

Thanks.

#40 lore raw
In reply to David Miller
Quoted 8 lines · David Miller

From: [email protected] (Eric W. Biederman)
Date: Tue, 15 Sep 2015 19:59:49 -0500

Quoted 4 lines

Pablo, Dave I don't know whose tree this makes more sense to go through. I am assuming at least initially Pablos as netfilter is involved. From what I have seen there will be a lot of back and forth between the netfilter code paths and the routing code paths.

Quoted 4 lines

I think it might reduce conflicts actually if it went via my net-next tree.

Pablo, any objections?

This looks good, thanks for taking these David.

#41 lore raw

cover netfilter: Stop guessing net

In reply to Eric W. Biederman

This is the next installment of my work to pass struct net through the output path so the code does not need to guess how to figure out which network namespace it is in, and ultimately routes can have output devices in another network namespace.

This round the changes focus on using nf_hook_state->net in the netfilter functions, and not passing nf_hook_ops to the functions on the netfilter chains.

Given the way per network namespace operations are implemented callers can not rely on addresses or games with container_of. None of the netfilter hooks do but we should remove the temptation. Plus removing the extra parameter and the extra jump to get at values should reduce the number of cache line misses.

At a practical matter this is the bulk of the work that other pieces of code in the netfilter stack will depend on.

The changes are also available against net-next at:
git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/net-next.git master

My entire pending set of changes for those who want to look ahead is at:
git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/net-next.git for-testing

Eric

Eric W. Biederman (14):
ebtables: Simplify the arguments to ebt_do_table
inet netfilter: Remove hook from ip6t_do_table, arp_do_table, ipt_do_table inet netfilter: Prefer state->hook to ops->hooknum
nftables: kill nft_pktinfo.ops
x_tables: Pass struct net in xt_action_param
x_tables: Use par->net instead of computing from the passed net devices
nftables: Pass struct net in nft_pktinfo
nftables: Use pkt->net instead of computing net from the passed net_devices
net: Pass net to nf_dup_ipv4 and nf_dup_ipv6
nf_conntrack: Add a struct net parameter to l4_pkt_to_tuple
ipvs: Read hooknum from state rather than ops->hooknum
netfilter: Pass priv instead of nf_hook_ops to netfilter hooks
netfilter: Pass net into nf_xfrm_me_harder
netfilter: Use nf_ct_net instead of dev_net(out) in nf_nat_masquerade_ipv6

 include/linux/netfilter.h                      |  2 +-
 include/linux/netfilter/x_tables.h             |  3 +-
 include/linux/netfilter_arp/arp_tables.h       |  1 -
 include/linux/netfilter_bridge/ebtables.h      |  6 ++--
 include/linux/netfilter_ipv4/ip_tables.h       |  1 -
 include/linux/netfilter_ipv6/ip6_tables.h      |  1 -
 include/net/netfilter/br_netfilter.h           |  2 +-
 include/net/netfilter/ipv4/nf_dup_ipv4.h       |  2 +-
 include/net/netfilter/ipv6/nf_dup_ipv6.h       |  2 +-
 include/net/netfilter/nf_conntrack.h           |  3 +-
 include/net/netfilter/nf_conntrack_core.h      |  1 +
 include/net/netfilter/nf_conntrack_l4proto.h   |  2 +-
 include/net/netfilter/nf_nat_core.h            |  2 +-
 include/net/netfilter/nf_nat_l3proto.h         | 32 ++++++++++----------
 include/net/netfilter/nf_tables.h              | 14 ++++-----
 include/net/netfilter/nf_tables_ipv4.h         |  3 +-
 include/net/netfilter/nf_tables_ipv6.h         |  3 +-
 net/bridge/br_netfilter_hooks.c                | 14 ++++-----
 net/bridge/br_netfilter_ipv6.c                 |  2 +-
 net/bridge/netfilter/ebt_log.c                 |  2 +-
 net/bridge/netfilter/ebt_nflog.c               |  2 +-
 net/bridge/netfilter/ebtable_broute.c          |  8 +++--
 net/bridge/netfilter/ebtable_filter.c          | 10 +++---
 net/bridge/netfilter/ebtable_nat.c             | 10 +++---
 net/bridge/netfilter/ebtables.c                | 14 +++++----
 net/bridge/netfilter/nf_tables_bridge.c        | 20 ++++++------
 net/bridge/netfilter/nft_reject_bridge.c       | 19 ++++++------
 net/decnet/netfilter/dn_rtmsg.c                |  2 +-
 net/ipv4/netfilter/arp_tables.c                |  3 +-
 net/ipv4/netfilter/arptable_filter.c           |  5 ++-
 net/ipv4/netfilter/ip_tables.c                 |  3 +-
 net/ipv4/netfilter/ipt_CLUSTERIP.c             |  2 +-
 net/ipv4/netfilter/ipt_SYNPROXY.c              |  4 +--
 net/ipv4/netfilter/ipt_rpfilter.c              |  5 ++-
 net/ipv4/netfilter/iptable_filter.c            |  7 ++---
 net/ipv4/netfilter/iptable_mangle.c            | 14 ++++-----
 net/ipv4/netfilter/iptable_nat.c               | 21 ++++++-------
 net/ipv4/netfilter/iptable_raw.c               |  7 ++---
 net/ipv4/netfilter/iptable_security.c          |  7 ++---
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 12 ++++----
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   |  4 +--
 net/ipv4/netfilter/nf_defrag_ipv4.c            |  4 +--
 net/ipv4/netfilter/nf_dup_ipv4.c               | 23 +++-----------
 net/ipv4/netfilter/nf_nat_l3proto_ipv4.c       | 42 +++++++++++++-------------
 net/ipv4/netfilter/nf_tables_arp.c             |  6 ++--
 net/ipv4/netfilter/nf_tables_ipv4.c            | 10 +++---
 net/ipv4/netfilter/nft_chain_nat_ipv4.c        | 22 +++++++-------
 net/ipv4/netfilter/nft_chain_route_ipv4.c      |  6 ++--
 net/ipv4/netfilter/nft_dup_ipv4.c              |  2 +-
 net/ipv4/netfilter/nft_masq_ipv4.c             |  2 +-
 net/ipv4/netfilter/nft_redir_ipv4.c            |  2 +-
 net/ipv4/netfilter/nft_reject_ipv4.c           |  5 ++-
 net/ipv6/netfilter/ip6_tables.c                |  3 +-
 net/ipv6/netfilter/ip6t_REJECT.c               |  2 +-
 net/ipv6/netfilter/ip6t_SYNPROXY.c             |  4 +--
 net/ipv6/netfilter/ip6t_rpfilter.c             |  6 ++--
 net/ipv6/netfilter/ip6table_filter.c           |  5 ++-
 net/ipv6/netfilter/ip6table_mangle.c           | 14 ++++-----
 net/ipv6/netfilter/ip6table_nat.c              | 21 ++++++-------
 net/ipv6/netfilter/ip6table_raw.c              |  5 ++-
 net/ipv6/netfilter/ip6table_security.c         |  5 ++-
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 12 ++++----
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c |  3 +-
 net/ipv6/netfilter/nf_defrag_ipv6_hooks.c      |  6 ++--
 net/ipv6/netfilter/nf_dup_ipv6.c               | 23 +++-----------
 net/ipv6/netfilter/nf_nat_l3proto_ipv6.c       | 42 +++++++++++++-------------
 net/ipv6/netfilter/nf_nat_masquerade_ipv6.c    |  2 +-
 net/ipv6/netfilter/nf_tables_ipv6.c            | 10 +++---
 net/ipv6/netfilter/nft_chain_nat_ipv6.c        | 22 +++++++-------
 net/ipv6/netfilter/nft_chain_route_ipv6.c      |  6 ++--
 net/ipv6/netfilter/nft_dup_ipv6.c              |  2 +-
 net/ipv6/netfilter/nft_redir_ipv6.c            |  3 +-
 net/ipv6/netfilter/nft_reject_ipv6.c           |  7 ++---
 net/netfilter/core.c                           |  2 +-
 net/netfilter/ipset/ip_set_core.c              |  9 ++----
 net/netfilter/ipvs/ip_vs_core.c                | 40 ++++++++++++------------
 net/netfilter/nf_conntrack_core.c              | 10 +++---
 net/netfilter/nf_conntrack_proto_dccp.c        |  2 +-
 net/netfilter/nf_conntrack_proto_generic.c     |  2 +-
 net/netfilter/nf_conntrack_proto_gre.c         |  3 +-
 net/netfilter/nf_conntrack_proto_sctp.c        |  2 +-
 net/netfilter/nf_conntrack_proto_tcp.c         |  2 +-
 net/netfilter/nf_conntrack_proto_udp.c         |  1 +
 net/netfilter/nf_conntrack_proto_udplite.c     |  1 +
 net/netfilter/nf_nat_core.c                    |  4 +--
 net/netfilter/nf_tables_core.c                 | 10 +++---
 net/netfilter/nf_tables_netdev.c               | 20 ++++++------
 net/netfilter/nft_log.c                        |  3 +-
 net/netfilter/nft_meta.c                       |  4 +--
 net/netfilter/nft_queue.c                      |  2 +-
 net/netfilter/nft_reject_inet.c                | 19 ++++++------
 net/netfilter/xt_LOG.c                         |  2 +-
 net/netfilter/xt_NFLOG.c                       |  2 +-
 net/netfilter/xt_TCPMSS.c                      |  2 +-
 net/netfilter/xt_TEE.c                         |  4 +--
 net/netfilter/xt_TPROXY.c                      | 24 +++++++--------
 net/netfilter/xt_addrtype.c                    |  4 +--
 net/netfilter/xt_connlimit.c                   |  4 +--
 net/netfilter/xt_osf.c                         |  2 +-
 net/netfilter/xt_recent.c                      |  2 +-
 net/netfilter/xt_socket.c                      | 14 +++++----
 net/openvswitch/conntrack.c                    |  2 +-
 net/sched/act_connmark.c                       |  2 +-
 net/sched/act_ipt.c                            |  1 +
 net/sched/em_ipset.c                           |  1 +
 security/selinux/hooks.c                       | 10 +++---
 security/smack/smack_netfilter.c               |  4 +--
 107 files changed, 386 insertions(+), 434 deletions(-)
#42 lore raw

Re: [PATCH next 0/14] netfilter: Stop guessing net

Quoted 10 lines · Eric W. Biederman

This is the next installment of my work to pass struct net through the output path so the code does not need to guess how to figure out which network namespace it is in, and ultimately routes can have output devices in another network namespace.

This round the changes focus on using nf_hook_state->net in the netfilter functions, and not passing nf_hook_ops to the functions on the netfilter chains.

Quoted 8 lines

Given the way per network namespace operations are implemented callers can not rely on addresses or games with container_of. None of the netfilter hooks do but we should remove the temptation. Plus removing the extra parameter and the extra jump to get at values should reduce the number of cache line misses.

At a practical matter this is the bulk of the work that other pieces of code in the netfilter stack will depend on.

I only see the cover letter. Patches are also not visible on the patchwork. --
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#43 lore raw

Re: [PATCH next 0/14] netfilter: Stop guessing net

Quoted 17 lines · Nicolas Dichtel
Quoted 16 lines · Eric W. Biederman

This is the next installment of my work to pass struct net through the output path so the code does not need to guess how to figure out which network namespace it is in, and ultimately routes can have output devices in another network namespace.

This round the changes focus on using nf_hook_state->net in the netfilter functions, and not passing nf_hook_ops to the functions on the netfilter chains.

Given the way per network namespace operations are implemented callers can not rely on addresses or games with container_of. None of the netfilter hooks do but we should remove the temptation. Plus removing the extra parameter and the extra jump to get at values should reduce the number of cache line misses.

Quoted 4 lines
Quoted 2 lines

At a practical matter this is the bulk of the work that other pieces of code in the netfilter stack will depend on.

I only see the cover letter. Patches are also not visible on the patchwork.

It appears I forgot to hit send. The patches should show up any time now.

Eric

#44 lore raw

Re: [PATCH next 0/14] netfilter: Stop guessing net

Le 18/09/2015 18:13, Eric W. Biederman a écrit :
[snip]

Quoted 1 line

It appears I forgot to hit send. The patches should show up any time now.

Yes, thank you.

Nicolas

#45 lore raw

PATCH 1/14 ebtables: Simplify the arguments to ebt_do_table

In reply to Eric W. Biederman

Nearly everything thing of interest to ebt_do_table is already present in nf_hook_state. Simplify ebt_do_table by just passing in the skb, nf_hook_state, and the table. This make the code easier to read and maintenance easier.

To support this create an nf_hook_state on the stack in ebt_broute (the only caller without a nf_hook_state already available). This new nf_hook_state adds no new computations to ebt_broute, but does use a few more bytes of stack.

---
 include/linux/netfilter_bridge/ebtables.h |  6 +++---
 net/bridge/netfilter/ebtable_broute.c     |  8 ++++++--
 net/bridge/netfilter/ebtable_filter.c     |  6 ++----
 net/bridge/netfilter/ebtable_nat.c        |  6 ++----
 net/bridge/netfilter/ebtables.c           | 13 +++++++------
 5 files changed, 20 insertions(+), 19 deletions(-)
include/linux/netfilter_bridge/ebtables.h +3 -3
index 8ca6d6464ea3..2ea517c7c6b9 100644
@@ -111,9 +111,9 @@ struct ebt_table {
 extern struct ebt_table *ebt_register_table(struct net *net,
 					    const struct ebt_table *table);
 extern void ebt_unregister_table(struct net *net, struct ebt_table *table);
-extern unsigned int ebt_do_table(unsigned int hook, struct sk_buff *skb,
-   const struct net_device *in, const struct net_device *out,
-   struct ebt_table *table);
+extern unsigned int ebt_do_table(struct sk_buff *skb,
+				 const struct nf_hook_state *state,
+				 struct ebt_table *table);
 
 /* Used in the kernel match() functions */
 #define FWINV(bool,invflg) ((bool) ^ !!(info->invflags & invflg))
net/bridge/netfilter/ebtable_broute.c +6 -2
index d2cdf5d6e98c..ec94c6f1ae88 100644
@@ -50,10 +50,14 @@ static const struct ebt_table broute_table = {
 
 static int ebt_broute(struct sk_buff *skb)
 {
+	struct nf_hook_state state;
 	int ret;
 
-	ret = ebt_do_table(NF_BR_BROUTING, skb, skb->dev, NULL,
-			   dev_net(skb->dev)->xt.broute_table);
+	nf_hook_state_init(&state, NULL, NF_BR_BROUTING, INT_MIN,
+			   NFPROTO_BRIDGE, skb->dev, NULL, NULL,
+			   dev_net(skb->dev), NULL);
+
+	ret = ebt_do_table(skb, &state, state.net->xt.broute_table);
 	if (ret == NF_DROP)
 		return 1; /* route it */
 	return 0; /* bridge it */
net/bridge/netfilter/ebtable_filter.c +2 -4
index ab20d6ed6e2f..118ce40ac181 100644
@@ -60,16 +60,14 @@ static unsigned int
 ebt_in_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	    const struct nf_hook_state *state)
 {
-	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    state->net->xt.frame_filter);
+	return ebt_do_table(skb, state, state->net->xt.frame_filter);
 }
 
 static unsigned int
 ebt_out_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
-	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    state->net->xt.frame_filter);
+	return ebt_do_table(skb, state, state->net->xt.frame_filter);
 }
 
 static struct nf_hook_ops ebt_ops_filter[] __read_mostly = {
net/bridge/netfilter/ebtable_nat.c +2 -4
index ad81a5a65644..56c3329d6c37 100644
@@ -60,16 +60,14 @@ static unsigned int
 ebt_nat_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	   const struct nf_hook_state *state)
 {
-	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    state->net->xt.frame_nat);
+	return ebt_do_table(skb, state, state->net->xt.frame_nat);
 }
 
 static unsigned int
 ebt_nat_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	    const struct nf_hook_state *state)
 {
-	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    state->net->xt.frame_nat);
+	return ebt_do_table(skb, state, state->net->xt.frame_nat);
 }
 
 static struct nf_hook_ops ebt_ops_nat[] __read_mostly = {
net/bridge/netfilter/ebtables.c +7 -6
index 48b6b01295de..8d5a3975b963 100644
@@ -183,10 +183,11 @@ struct ebt_entry *ebt_next_entry(const struct ebt_entry *entry)
 }
 
 /* Do some firewalling */
-unsigned int ebt_do_table (unsigned int hook, struct sk_buff *skb,
-   const struct net_device *in, const struct net_device *out,
-   struct ebt_table *table)
+unsigned int ebt_do_table(struct sk_buff *skb,
+			  const struct nf_hook_state *state,
+			  struct ebt_table *table)
 {
+	unsigned int hook = state->hook;
 	int i, nentries;
 	struct ebt_entry *point;
 	struct ebt_counter *counter_base, *cb_base;
@@ -199,8 +200,8 @@ unsigned int ebt_do_table (unsigned int hook, struct sk_buff *skb,
 	struct xt_action_param acpar;
 
 	acpar.family  = NFPROTO_BRIDGE;
-	acpar.in      = in;
-	acpar.out     = out;
+	acpar.in      = state->in;
+	acpar.out     = state->out;
 	acpar.hotdrop = false;
 	acpar.hooknum = hook;
 
@@ -220,7 +221,7 @@ unsigned int ebt_do_table (unsigned int hook, struct sk_buff *skb,
 	base = private->entries;
 	i = 0;
 	while (i < nentries) {
-		if (ebt_basic_match(point, skb, in, out))
+		if (ebt_basic_match(point, skb, state->in, state->out))
 			goto letscontinue;
 
 		if (EBT_MATCH_ITERATE(point, ebt_do_match, skb, &acpar) != 0)
Signature · 1 line
2.2.1
#46 lore raw

PATCH 2/14 inet netfilter: Remove hook from ip6t_do_table, arp_do_table, ipt_do_table

In reply to Eric W. Biederman

The values of ops->hooknum and state->hook are guaraneted to be equal making the hook argument to ip6t_do_table, arp_do_table, and ipt_do_table is unnecessary. Remove the unnecessary hook argument.

In the callers use state->hook instead of ops->hooknum for clarity and to reduce the number of cachelines the callers touch.

---
 include/linux/netfilter_arp/arp_tables.h  |  1 -
 include/linux/netfilter_ipv4/ip_tables.h  |  1 -
 include/linux/netfilter_ipv6/ip6_tables.h |  1 -
 net/ipv4/netfilter/arp_tables.c           |  2 +-
 net/ipv4/netfilter/arptable_filter.c      |  3 +--
 net/ipv4/netfilter/ip_tables.c            |  2 +-
 net/ipv4/netfilter/iptable_filter.c       |  5 ++---
 net/ipv4/netfilter/iptable_mangle.c       | 12 +++++-------
 net/ipv4/netfilter/iptable_nat.c          |  3 +--
 net/ipv4/netfilter/iptable_raw.c          |  5 ++---
 net/ipv4/netfilter/iptable_security.c     |  5 ++---
 net/ipv6/netfilter/ip6_tables.c           |  2 +-
 net/ipv6/netfilter/ip6table_filter.c      |  3 +--
 net/ipv6/netfilter/ip6table_mangle.c      | 12 +++++-------
 net/ipv6/netfilter/ip6table_nat.c         |  3 +--
 net/ipv6/netfilter/ip6table_raw.c         |  3 +--
 net/ipv6/netfilter/ip6table_security.c    |  3 +--
 17 files changed, 25 insertions(+), 41 deletions(-)
include/linux/netfilter_arp/arp_tables.h +0 -1
index c22a7fb8d0df..6f074db2f23d 100644
@@ -53,7 +53,6 @@ extern struct xt_table *arpt_register_table(struct net *net,
 					    const struct arpt_replace *repl);
 extern void arpt_unregister_table(struct xt_table *table);
 extern unsigned int arpt_do_table(struct sk_buff *skb,
-				  unsigned int hook,
 				  const struct nf_hook_state *state,
 				  struct xt_table *table);
 
include/linux/netfilter_ipv4/ip_tables.h +0 -1
index 4073510da485..aa598f942c01 100644
@@ -64,7 +64,6 @@ struct ipt_error {
 
 extern void *ipt_alloc_initial_table(const struct xt_table *);
 extern unsigned int ipt_do_table(struct sk_buff *skb,
-				 unsigned int hook,
 				 const struct nf_hook_state *state,
 				 struct xt_table *table);
 
include/linux/netfilter_ipv6/ip6_tables.h +0 -1
index b40d2b635778..0f76e5c674f9 100644
@@ -30,7 +30,6 @@ extern struct xt_table *ip6t_register_table(struct net *net,
 					    const struct ip6t_replace *repl);
 extern void ip6t_unregister_table(struct net *net, struct xt_table *table);
 extern unsigned int ip6t_do_table(struct sk_buff *skb,
-				  unsigned int hook,
 				  const struct nf_hook_state *state,
 				  struct xt_table *table);
 
net/ipv4/netfilter/arp_tables.c +1 -1
index 8f87fc38ccde..10eb2b297450 100644
@@ -247,10 +247,10 @@ struct arpt_entry *arpt_next_entry(const struct arpt_entry *entry)
 }
 
 unsigned int arpt_do_table(struct sk_buff *skb,
-			   unsigned int hook,
 			   const struct nf_hook_state *state,
 			   struct xt_table *table)
 {
+	unsigned int hook = state->hook;
 	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
 	unsigned int verdict = NF_DROP;
 	const struct arphdr *arp;
net/ipv4/netfilter/arptable_filter.c +1 -2
index d217e4c19645..1352e12d4068 100644
@@ -30,8 +30,7 @@ static unsigned int
 arptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	return arpt_do_table(skb, ops->hooknum, state,
-			     state->net->ipv4.arptable_filter);
+	return arpt_do_table(skb, state, state->net->ipv4.arptable_filter);
 }
 
 static struct nf_hook_ops *arpfilter_ops __read_mostly;
net/ipv4/netfilter/ip_tables.c +1 -1
index 5d514eac4c31..2b049e135de8 100644
@@ -285,10 +285,10 @@ struct ipt_entry *ipt_next_entry(const struct ipt_entry *entry)
 /* Returns one of the generic firewall policies, like NF_ACCEPT. */
 unsigned int
 ipt_do_table(struct sk_buff *skb,
-	     unsigned int hook,
 	     const struct nf_hook_state *state,
 	     struct xt_table *table)
 {
+	unsigned int hook = state->hook;
 	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
 	const struct iphdr *ip;
 	/* Initializing verdict to NF_DROP keeps gcc happy. */
net/ipv4/netfilter/iptable_filter.c +2 -3
index 32feff32b116..02d4c5395d6e 100644
@@ -36,14 +36,13 @@ static unsigned int
 iptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		    const struct nf_hook_state *state)
 {
-	if (ops->hooknum == NF_INET_LOCAL_OUT &&
+	if (state->hook == NF_INET_LOCAL_OUT &&
 	    (skb->len < sizeof(struct iphdr) ||
 	     ip_hdrlen(skb) < sizeof(struct iphdr)))
 		/* root is playing with raw sockets. */
 		return NF_ACCEPT;
 
-	return ipt_do_table(skb, ops->hooknum, state,
-			    state->net->ipv4.iptable_filter);
+	return ipt_do_table(skb, state, state->net->ipv4.iptable_filter);
 }
 
 static struct nf_hook_ops *filter_ops __read_mostly;
net/ipv4/netfilter/iptable_mangle.c +5 -7
index 4a5150fc9510..dc2ff6884999 100644
@@ -58,8 +58,7 @@ ipt_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state)
 	daddr = iph->daddr;
 	tos = iph->tos;
 
-	ret = ipt_do_table(skb, NF_INET_LOCAL_OUT, state,
-			   state->net->ipv4.iptable_mangle);
+	ret = ipt_do_table(skb, state, state->net->ipv4.iptable_mangle);
 	/* Reroute for ANY change. */
 	if (ret != NF_DROP && ret != NF_STOLEN) {
 		iph = ip_hdr(skb);
@@ -83,14 +82,13 @@ iptable_mangle_hook(const struct nf_hook_ops *ops,
 		     struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	if (ops->hooknum == NF_INET_LOCAL_OUT)
+	if (state->hook == NF_INET_LOCAL_OUT)
 		return ipt_mangle_out(skb, state);
-	if (ops->hooknum == NF_INET_POST_ROUTING)
-		return ipt_do_table(skb, ops->hooknum, state,
+	if (state->hook == NF_INET_POST_ROUTING)
+		return ipt_do_table(skb, state,
 				    state->net->ipv4.iptable_mangle);
 	/* PREROUTING/INPUT/FORWARD: */
-	return ipt_do_table(skb, ops->hooknum, state,
-			    state->net->ipv4.iptable_mangle);
+	return ipt_do_table(skb, state, state->net->ipv4.iptable_mangle);
 }
 
 static struct nf_hook_ops *mangle_ops __read_mostly;
net/ipv4/netfilter/iptable_nat.c +1 -2
index 4f4c64f81169..8ff63ac1f0d6 100644
@@ -33,8 +33,7 @@ static unsigned int iptable_nat_do_chain(const struct nf_hook_ops *ops,
 					 const struct nf_hook_state *state,
 					 struct nf_conn *ct)
 {
-	return ipt_do_table(skb, ops->hooknum, state,
-			    state->net->ipv4.nat_table);
+	return ipt_do_table(skb, state, state->net->ipv4.nat_table);
 }
 
 static unsigned int iptable_nat_ipv4_fn(const struct nf_hook_ops *ops,
net/ipv4/netfilter/iptable_raw.c +2 -3
index 20126e469ffb..bbb0523d87de 100644
@@ -23,14 +23,13 @@ static unsigned int
 iptable_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		 const struct nf_hook_state *state)
 {
-	if (ops->hooknum == NF_INET_LOCAL_OUT &&
+	if (state->hook == NF_INET_LOCAL_OUT &&
 	    (skb->len < sizeof(struct iphdr) ||
 	     ip_hdrlen(skb) < sizeof(struct iphdr)))
 		/* root is playing with raw sockets. */
 		return NF_ACCEPT;
 
-	return ipt_do_table(skb, ops->hooknum, state,
-			    state->net->ipv4.iptable_raw);
+	return ipt_do_table(skb, state, state->net->ipv4.iptable_raw);
 }
 
 static struct nf_hook_ops *rawtable_ops __read_mostly;
net/ipv4/netfilter/iptable_security.c +2 -3
index 82fefd609b85..b92417038705 100644
@@ -40,14 +40,13 @@ static unsigned int
 iptable_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
-	if (ops->hooknum == NF_INET_LOCAL_OUT &&
+	if (state->hook == NF_INET_LOCAL_OUT &&
 	    (skb->len < sizeof(struct iphdr) ||
 	     ip_hdrlen(skb) < sizeof(struct iphdr)))
 		/* Somebody is playing with raw sockets. */
 		return NF_ACCEPT;
 
-	return ipt_do_table(skb, ops->hooknum, state,
-			    state->net->ipv4.iptable_security);
+	return ipt_do_table(skb, state, state->net->ipv4.iptable_security);
 }
 
 static struct nf_hook_ops *sectbl_ops __read_mostly;
net/ipv6/netfilter/ip6_tables.c +1 -1
index cd9b401231d3..da6446b6e3f9 100644
@@ -314,10 +314,10 @@ ip6t_next_entry(const struct ip6t_entry *entry)
 /* Returns one of the generic firewall policies, like NF_ACCEPT. */
 unsigned int
 ip6t_do_table(struct sk_buff *skb,
-	      unsigned int hook,
 	      const struct nf_hook_state *state,
 	      struct xt_table *table)
 {
+	unsigned int hook = state->hook;
 	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
 	/* Initializing verdict to NF_DROP keeps gcc happy. */
 	unsigned int verdict = NF_DROP;
net/ipv6/netfilter/ip6table_filter.c +1 -2
index 2449005fb5dc..a7327f61b90c 100644
@@ -35,8 +35,7 @@ static unsigned int
 ip6table_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	return ip6t_do_table(skb, ops->hooknum, state,
-			     state->net->ipv6.ip6table_filter);
+	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_filter);
 }
 
 static struct nf_hook_ops *filter_ops __read_mostly;
net/ipv6/netfilter/ip6table_mangle.c +5 -7
index a46dbf097d29..c2e061dcedf3 100644
@@ -57,8 +57,7 @@ ip6t_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state)
 	/* flowlabel and prio (includes version, which shouldn't change either */
 	flowlabel = *((u_int32_t *)ipv6_hdr(skb));
 
-	ret = ip6t_do_table(skb, NF_INET_LOCAL_OUT, state,
-			    state->net->ipv6.ip6table_mangle);
+	ret = ip6t_do_table(skb, state, state->net->ipv6.ip6table_mangle);
 
 	if (ret != NF_DROP && ret != NF_STOLEN &&
 	    (!ipv6_addr_equal(&ipv6_hdr(skb)->saddr, &saddr) ||
@@ -79,14 +78,13 @@ static unsigned int
 ip6table_mangle_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	if (ops->hooknum == NF_INET_LOCAL_OUT)
+	if (state->hook == NF_INET_LOCAL_OUT)
 		return ip6t_mangle_out(skb, state);
-	if (ops->hooknum == NF_INET_POST_ROUTING)
-		return ip6t_do_table(skb, ops->hooknum, state,
+	if (state->hook == NF_INET_POST_ROUTING)
+		return ip6t_do_table(skb, state,
 				     state->net->ipv6.ip6table_mangle);
 	/* INPUT/FORWARD */
-	return ip6t_do_table(skb, ops->hooknum, state,
-			     state->net->ipv6.ip6table_mangle);
+	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_mangle);
 }
 
 static struct nf_hook_ops *mangle_ops __read_mostly;
net/ipv6/netfilter/ip6table_nat.c +1 -2
index a56451de127f..efa6754c4d06 100644
@@ -35,8 +35,7 @@ static unsigned int ip6table_nat_do_chain(const struct nf_hook_ops *ops,
 					  const struct nf_hook_state *state,
 					  struct nf_conn *ct)
 {
-	return ip6t_do_table(skb, ops->hooknum, state,
-			     state->net->ipv6.ip6table_nat);
+	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_nat);
 }
 
 static unsigned int ip6table_nat_fn(const struct nf_hook_ops *ops,
net/ipv6/netfilter/ip6table_raw.c +1 -2
index 18e831e35782..fac6ad7c0a7c 100644
@@ -22,8 +22,7 @@ static unsigned int
 ip6table_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		  const struct nf_hook_state *state)
 {
-	return ip6t_do_table(skb, ops->hooknum, state,
-			     state->net->ipv6.ip6table_raw);
+	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_raw);
 }
 
 static struct nf_hook_ops *rawtable_ops __read_mostly;
net/ipv6/netfilter/ip6table_security.c +1 -2
index 83bc96ae5d73..96c94fc240c8 100644
@@ -39,8 +39,7 @@ static unsigned int
 ip6table_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		       const struct nf_hook_state *state)
 {
-	return ip6t_do_table(skb, ops->hooknum, state,
-			     state->net->ipv6.ip6table_security);
+	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_security);
 }
 
 static struct nf_hook_ops *sectbl_ops __read_mostly;
Signature · 1 line
2.2.1
#47 lore raw

PATCH 3/14 inet netfilter: Prefer state->hook to ops->hooknum

In reply to Eric W. Biederman

The values of nf_hook_state.hook and nf_hook_ops.hooknum must be the same by definition.

We are more likely to access the fields in nf_hook_state over the fields in nf_hook_ops so with a little luck this results in fewer cache line misses, and slightly more consistent code.

---
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c |  4 ++--
 net/ipv4/netfilter/nf_defrag_ipv4.c            |  2 +-
 net/ipv4/netfilter/nf_nat_l3proto_ipv4.c       | 14 +++++++-------
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c |  4 ++--
 net/ipv6/netfilter/nf_defrag_ipv6_hooks.c      |  4 ++--
 net/ipv6/netfilter/nf_nat_l3proto_ipv6.c       | 14 +++++++-------
 6 files changed, 21 insertions(+), 21 deletions(-)
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +2 -2
index 9564684876c9..15749cc5cf2b 100644
@@ -147,7 +147,7 @@ static unsigned int ipv4_conntrack_in(const struct nf_hook_ops *ops,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
-	return nf_conntrack_in(state->net, PF_INET, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET, state->hook, skb);
 }
 
 static unsigned int ipv4_conntrack_local(const struct nf_hook_ops *ops,
@@ -158,7 +158,7 @@ static unsigned int ipv4_conntrack_local(const struct nf_hook_ops *ops,
 	if (skb->len < sizeof(struct iphdr) ||
 	    ip_hdrlen(skb) < sizeof(struct iphdr))
 		return NF_ACCEPT;
-	return nf_conntrack_in(state->net, PF_INET, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET, state->hook, skb);
 }
 
 /* Connection tracking may drop packets, but never alters them, so
net/ipv4/netfilter/nf_defrag_ipv4.c +1 -1
index 9306ec4fab41..8aea536d2e83 100644
@@ -83,7 +83,7 @@ static unsigned int ipv4_conntrack_defrag(const struct nf_hook_ops *ops,
 	/* Gather fragments. */
 	if (ip_is_fragment(ip_hdr(skb))) {
 		enum ip_defrag_users user =
-			nf_ct_defrag_user(ops->hooknum, skb);
+			nf_ct_defrag_user(state->hook, skb);
 
 		if (nf_ct_ipv4_gather_frags(skb, user))
 			return NF_STOLEN;
net/ipv4/netfilter/nf_nat_l3proto_ipv4.c +7 -7
index 22f4579b0c2a..16da45a76dac 100644
@@ -266,7 +266,7 @@ nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn_nat *nat;
 	/* maniptype == SRC for postrouting. */
-	enum nf_nat_manip_type maniptype = HOOK2MANIP(ops->hooknum);
+	enum nf_nat_manip_type maniptype = HOOK2MANIP(state->hook);
 
 	/* We never see fragments: conntrack defrags on pre-routing
 	 * and local-out, and nf_nat_out protects post-routing.
@@ -295,7 +295,7 @@ nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	case IP_CT_RELATED_REPLY:
 		if (ip_hdr(skb)->protocol == IPPROTO_ICMP) {
 			if (!nf_nat_icmp_reply_translation(skb, ct, ctinfo,
-							   ops->hooknum))
+							   state->hook))
 				return NF_DROP;
 			else
 				return NF_ACCEPT;
@@ -312,17 +312,17 @@ nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 			if (ret != NF_ACCEPT)
 				return ret;
 
-			if (nf_nat_initialized(ct, HOOK2MANIP(ops->hooknum)))
+			if (nf_nat_initialized(ct, HOOK2MANIP(state->hook)))
 				break;
 
-			ret = nf_nat_alloc_null_binding(ct, ops->hooknum);
+			ret = nf_nat_alloc_null_binding(ct, state->hook);
 			if (ret != NF_ACCEPT)
 				return ret;
 		} else {
 			pr_debug("Already setup manip %s for ct %p\n",
 				 maniptype == NF_NAT_MANIP_SRC ? "SRC" : "DST",
 				 ct);
-			if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat,
+			if (nf_nat_oif_changed(state->hook, ctinfo, nat,
 					       state->out))
 				goto oif_changed;
 		}
@@ -332,11 +332,11 @@ nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		/* ESTABLISHED */
 		NF_CT_ASSERT(ctinfo == IP_CT_ESTABLISHED ||
 			     ctinfo == IP_CT_ESTABLISHED_REPLY);
-		if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat, state->out))
+		if (nf_nat_oif_changed(state->hook, ctinfo, nat, state->out))
 			goto oif_changed;
 	}
 
-	return nf_nat_packet(ct, ctinfo, ops->hooknum, skb);
+	return nf_nat_packet(ct, ctinfo, state->hook, skb);
 
 oif_changed:
 	nf_ct_kill_acct(ct, ctinfo, skb);
net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +2 -2
index 1ef1b79def56..339be1d59afc 100644
@@ -169,7 +169,7 @@ static unsigned int ipv6_conntrack_in(const struct nf_hook_ops *ops,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
-	return nf_conntrack_in(state->net, PF_INET6, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET6, state->hook, skb);
 }
 
 static unsigned int ipv6_conntrack_local(const struct nf_hook_ops *ops,
@@ -181,7 +181,7 @@ static unsigned int ipv6_conntrack_local(const struct nf_hook_ops *ops,
 		net_notice_ratelimited("ipv6_conntrack_local: packet too short\n");
 		return NF_ACCEPT;
 	}
-	return nf_conntrack_in(state->net, PF_INET6, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET6, state->hook, skb);
 }
 
 static struct nf_hook_ops ipv6_conntrack_ops[] __read_mostly = {
net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +2 -2
index 6b576be3c83e..a9c08520596b 100644
@@ -63,7 +63,7 @@ static unsigned int ipv6_defrag(const struct nf_hook_ops *ops,
 		return NF_ACCEPT;
 #endif
 
-	reasm = nf_ct_frag6_gather(skb, nf_ct6_defrag_user(ops->hooknum, skb));
+	reasm = nf_ct_frag6_gather(skb, nf_ct6_defrag_user(state->hook, skb));
 	/* queued */
 	if (reasm == NULL)
 		return NF_STOLEN;
@@ -74,7 +74,7 @@ static unsigned int ipv6_defrag(const struct nf_hook_ops *ops,
 
 	nf_ct_frag6_consume_orig(reasm);
 
-	NF_HOOK_THRESH(NFPROTO_IPV6, ops->hooknum, state->net, state->sk, reasm,
+	NF_HOOK_THRESH(NFPROTO_IPV6, state->hook, state->net, state->sk, reasm,
 		       state->in, state->out,
 		       state->okfn, NF_IP6_PRI_CONNTRACK_DEFRAG + 1);
 
net/ipv6/netfilter/nf_nat_l3proto_ipv6.c +7 -7
index 70fbaed49edb..8bc94907dbd9 100644
@@ -272,7 +272,7 @@ nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	struct nf_conn *ct;
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn_nat *nat;
-	enum nf_nat_manip_type maniptype = HOOK2MANIP(ops->hooknum);
+	enum nf_nat_manip_type maniptype = HOOK2MANIP(state->hook);
 	__be16 frag_off;
 	int hdrlen;
 	u8 nexthdr;
@@ -303,7 +303,7 @@ nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 
 		if (hdrlen >= 0 && nexthdr == IPPROTO_ICMPV6) {
 			if (!nf_nat_icmpv6_reply_translation(skb, ct, ctinfo,
-							     ops->hooknum,
+							     state->hook,
 							     hdrlen))
 				return NF_DROP;
 			else
@@ -321,17 +321,17 @@ nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 			if (ret != NF_ACCEPT)
 				return ret;
 
-			if (nf_nat_initialized(ct, HOOK2MANIP(ops->hooknum)))
+			if (nf_nat_initialized(ct, HOOK2MANIP(state->hook)))
 				break;
 
-			ret = nf_nat_alloc_null_binding(ct, ops->hooknum);
+			ret = nf_nat_alloc_null_binding(ct, state->hook);
 			if (ret != NF_ACCEPT)
 				return ret;
 		} else {
 			pr_debug("Already setup manip %s for ct %p\n",
 				 maniptype == NF_NAT_MANIP_SRC ? "SRC" : "DST",
 				 ct);
-			if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat, state->out))
+			if (nf_nat_oif_changed(state->hook, ctinfo, nat, state->out))
 				goto oif_changed;
 		}
 		break;
@@ -340,11 +340,11 @@ nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		/* ESTABLISHED */
 		NF_CT_ASSERT(ctinfo == IP_CT_ESTABLISHED ||
 			     ctinfo == IP_CT_ESTABLISHED_REPLY);
-		if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat, state->out))
+		if (nf_nat_oif_changed(state->hook, ctinfo, nat, state->out))
 			goto oif_changed;
 	}
 
-	return nf_nat_packet(ct, ctinfo, ops->hooknum, skb);
+	return nf_nat_packet(ct, ctinfo, state->hook, skb);
 
 oif_changed:
 	nf_ct_kill_acct(ct, ctinfo, skb);
Signature · 1 line
2.2.1
#48 lore raw

PATCH 4/14 nftables: kill nft_pktinfo.ops

In reply to Eric W. Biederman

- Add nft_pktinfo.pf to replace ops->pf
- Add nft_pktinfo.hook to replace ops->hooknum

This simplifies the code, makes it more readable, and likely reduces cache line misses. Maintainability is enhanced as the details of nft_hook_ops are of no concern to the recpients of nft_pktinfo.

---
 include/net/netfilter/nf_tables.h         |  9 ++++-----
 include/net/netfilter/nf_tables_ipv4.h    |  3 +--
 include/net/netfilter/nf_tables_ipv6.h    |  3 +--
 net/bridge/netfilter/nf_tables_bridge.c   | 16 +++++++---------
 net/bridge/netfilter/nft_reject_bridge.c  | 12 ++++++------
 net/ipv4/netfilter/nf_tables_arp.c        |  2 +-
 net/ipv4/netfilter/nf_tables_ipv4.c       |  2 +-
 net/ipv4/netfilter/nft_chain_nat_ipv4.c   |  2 +-
 net/ipv4/netfilter/nft_chain_route_ipv4.c |  2 +-
 net/ipv4/netfilter/nft_dup_ipv4.c         |  2 +-
 net/ipv4/netfilter/nft_masq_ipv4.c        |  2 +-
 net/ipv4/netfilter/nft_redir_ipv4.c       |  2 +-
 net/ipv4/netfilter/nft_reject_ipv4.c      |  5 ++---
 net/ipv6/netfilter/nf_tables_ipv6.c       |  2 +-
 net/ipv6/netfilter/nft_chain_nat_ipv6.c   |  2 +-
 net/ipv6/netfilter/nft_chain_route_ipv6.c |  2 +-
 net/ipv6/netfilter/nft_dup_ipv6.c         |  2 +-
 net/ipv6/netfilter/nft_redir_ipv6.c       |  3 +--
 net/ipv6/netfilter/nft_reject_ipv6.c      |  5 ++---
 net/netfilter/nf_tables_core.c            |  2 +-
 net/netfilter/nf_tables_netdev.c          | 16 +++++++---------
 net/netfilter/nft_log.c                   |  2 +-
 net/netfilter/nft_meta.c                  |  4 ++--
 net/netfilter/nft_queue.c                 |  2 +-
 net/netfilter/nft_reject_inet.c           | 14 +++++++-------
 25 files changed, 54 insertions(+), 64 deletions(-)
include/net/netfilter/nf_tables.h +4 -5
index aa8bee72c9d3..c0899f97ff8d 100644
@@ -16,7 +16,8 @@ struct nft_pktinfo {
 	struct sk_buff			*skb;
 	const struct net_device		*in;
 	const struct net_device		*out;
-	const struct nf_hook_ops	*ops;
+	u8				pf;
+	u8				hook;
 	u8				nhoff;
 	u8				thoff;
 	u8				tprot;
@@ -25,16 +26,14 @@ struct nft_pktinfo {
 };
 
 static inline void nft_set_pktinfo(struct nft_pktinfo *pkt,
-				   const struct nf_hook_ops *ops,
 				   struct sk_buff *skb,
 				   const struct nf_hook_state *state)
 {
 	pkt->skb = skb;
 	pkt->in = pkt->xt.in = state->in;
 	pkt->out = pkt->xt.out = state->out;
-	pkt->ops = ops;
-	pkt->xt.hooknum = ops->hooknum;
-	pkt->xt.family = ops->pf;
+	pkt->hook = pkt->xt.hooknum = state->hook;
+	pkt->pf = pkt->xt.family = state->pf;
 }
 
 /**
include/net/netfilter/nf_tables_ipv4.h +1 -2
index 2df7f96902ee..ca6ef6bf775e 100644
@@ -6,13 +6,12 @@
 
 static inline void
 nft_set_pktinfo_ipv4(struct nft_pktinfo *pkt,
-		     const struct nf_hook_ops *ops,
 		     struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
 	struct iphdr *ip;
 
-	nft_set_pktinfo(pkt, ops, skb, state);
+	nft_set_pktinfo(pkt, skb, state);
 
 	ip = ip_hdr(pkt->skb);
 	pkt->tprot = ip->protocol;
include/net/netfilter/nf_tables_ipv6.h +1 -2
index 97db2e3a5e65..8ad39a6a5fe1 100644
@@ -6,14 +6,13 @@
 
 static inline int
 nft_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
-		     const struct nf_hook_ops *ops,
 		     struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
 	int protohdr, thoff = 0;
 	unsigned short frag_off;
 
-	nft_set_pktinfo(pkt, ops, skb, state);
+	nft_set_pktinfo(pkt, skb, state);
 
 	protohdr = ipv6_find_hdr(pkt->skb, &thoff, -1, &frag_off, NULL);
 	/* If malformed, drop it */
net/bridge/netfilter/nf_tables_bridge.c +7 -9
index a343e62442b1..318d825e4207 100644
@@ -65,27 +65,25 @@ int nft_bridge_ip6hdr_validate(struct sk_buff *skb)
 EXPORT_SYMBOL_GPL(nft_bridge_ip6hdr_validate);
 
 static inline void nft_bridge_set_pktinfo_ipv4(struct nft_pktinfo *pkt,
-					       const struct nf_hook_ops *ops,
 					       struct sk_buff *skb,
 					       const struct nf_hook_state *state)
 {
 	if (nft_bridge_iphdr_validate(skb))
-		nft_set_pktinfo_ipv4(pkt, ops, skb, state);
+		nft_set_pktinfo_ipv4(pkt, skb, state);
 	else
-		nft_set_pktinfo(pkt, ops, skb, state);
+		nft_set_pktinfo(pkt, skb, state);
 }
 
 static inline void nft_bridge_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
-					       const struct nf_hook_ops *ops,
 					       struct sk_buff *skb,
 					       const struct nf_hook_state *state)
 {
 #if IS_ENABLED(CONFIG_IPV6)
 	if (nft_bridge_ip6hdr_validate(skb) &&
-	    nft_set_pktinfo_ipv6(pkt, ops, skb, state) == 0)
+	    nft_set_pktinfo_ipv6(pkt, skb, state) == 0)
 		return;
 #endif
-	nft_set_pktinfo(pkt, ops, skb, state);
+	nft_set_pktinfo(pkt, skb, state);
 }
 
 static unsigned int
@@ -97,13 +95,13 @@ nft_do_chain_bridge(const struct nf_hook_ops *ops,
 
 	switch (eth_hdr(skb)->h_proto) {
 	case htons(ETH_P_IP):
-		nft_bridge_set_pktinfo_ipv4(&pkt, ops, skb, state);
+		nft_bridge_set_pktinfo_ipv4(&pkt, skb, state);
 		break;
 	case htons(ETH_P_IPV6):
-		nft_bridge_set_pktinfo_ipv6(&pkt, ops, skb, state);
+		nft_bridge_set_pktinfo_ipv6(&pkt, skb, state);
 		break;
 	default:
-		nft_set_pktinfo(&pkt, ops, skb, state);
+		nft_set_pktinfo(&pkt, skb, state);
 		break;
 	}
 
net/bridge/netfilter/nft_reject_bridge.c +6 -6
index 858d848564ee..cee92612b2cc 100644
@@ -273,16 +273,16 @@ static void nft_reject_bridge_eval(const struct nft_expr *expr,
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
 			nft_reject_br_send_v4_unreach(pkt->skb, pkt->in,
-						      pkt->ops->hooknum,
+						      pkt->hook,
 						      priv->icmp_code);
 			break;
 		case NFT_REJECT_TCP_RST:
 			nft_reject_br_send_v4_tcp_reset(pkt->skb, pkt->in,
-							pkt->ops->hooknum);
+							pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
 			nft_reject_br_send_v4_unreach(pkt->skb, pkt->in,
-						      pkt->ops->hooknum,
+						      pkt->hook,
 						      nft_reject_icmp_code(priv->icmp_code));
 			break;
 		}
@@ -291,16 +291,16 @@ static void nft_reject_bridge_eval(const struct nft_expr *expr,
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
 			nft_reject_br_send_v6_unreach(net, pkt->skb, pkt->in,
-						      pkt->ops->hooknum,
+						      pkt->hook,
 						      priv->icmp_code);
 			break;
 		case NFT_REJECT_TCP_RST:
 			nft_reject_br_send_v6_tcp_reset(net, pkt->skb, pkt->in,
-							pkt->ops->hooknum);
+							pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
 			nft_reject_br_send_v6_unreach(net, pkt->skb, pkt->in,
-						      pkt->ops->hooknum,
+						      pkt->hook,
 						      nft_reject_icmpv6_code(priv->icmp_code));
 			break;
 		}
net/ipv4/netfilter/nf_tables_arp.c +1 -1
index 8412268bbad1..883bbf83fe09 100644
@@ -21,7 +21,7 @@ nft_do_chain_arp(const struct nf_hook_ops *ops,
 {
 	struct nft_pktinfo pkt;
 
-	nft_set_pktinfo(&pkt, ops, skb, state);
+	nft_set_pktinfo(&pkt, skb, state);
 
 	return nft_do_chain(&pkt, ops);
 }
net/ipv4/netfilter/nf_tables_ipv4.c +1 -1
index aa180d3a69a5..805be5c9fcc3 100644
@@ -24,7 +24,7 @@ static unsigned int nft_do_chain_ipv4(const struct nf_hook_ops *ops,
 {
 	struct nft_pktinfo pkt;
 
-	nft_set_pktinfo_ipv4(&pkt, ops, skb, state);
+	nft_set_pktinfo_ipv4(&pkt, skb, state);
 
 	return nft_do_chain(&pkt, ops);
 }
net/ipv4/netfilter/nft_chain_nat_ipv4.c +1 -1
index bf5c30ae14e4..c3ffecf28d38 100644
@@ -33,7 +33,7 @@ static unsigned int nft_nat_do_chain(const struct nf_hook_ops *ops,
 {
 	struct nft_pktinfo pkt;
 
-	nft_set_pktinfo_ipv4(&pkt, ops, skb, state);
+	nft_set_pktinfo_ipv4(&pkt, skb, state);
 
 	return nft_do_chain(&pkt, ops);
 }
net/ipv4/netfilter/nft_chain_route_ipv4.c +1 -1
index e335b0afdaf3..2a1e3d8a3e43 100644
@@ -37,7 +37,7 @@ static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops,
 	    ip_hdrlen(skb) < sizeof(struct iphdr))
 		return NF_ACCEPT;
 
-	nft_set_pktinfo_ipv4(&pkt, ops, skb, state);
+	nft_set_pktinfo_ipv4(&pkt, skb, state);
 
 	mark = skb->mark;
 	iph = ip_hdr(skb);
net/ipv4/netfilter/nft_dup_ipv4.c +1 -1
index b45932d43b69..30bcf820e8bd 100644
@@ -30,7 +30,7 @@ static void nft_dup_ipv4_eval(const struct nft_expr *expr,
 	};
 	int oif = regs->data[priv->sreg_dev];
 
-	nf_dup_ipv4(pkt->skb, pkt->ops->hooknum, &gw, oif);
+	nf_dup_ipv4(pkt->skb, pkt->hook, &gw, oif);
 }
 
 static int nft_dup_ipv4_init(const struct nft_ctx *ctx,
net/ipv4/netfilter/nft_masq_ipv4.c +1 -1
index 40e414c4ca56..b72ffc58e255 100644
@@ -26,7 +26,7 @@ static void nft_masq_ipv4_eval(const struct nft_expr *expr,
 	memset(&range, 0, sizeof(range));
 	range.flags = priv->flags;
 
-	regs->verdict.code = nf_nat_masquerade_ipv4(pkt->skb, pkt->ops->hooknum,
+	regs->verdict.code = nf_nat_masquerade_ipv4(pkt->skb, pkt->hook,
 						    &range, pkt->out);
 }
 
net/ipv4/netfilter/nft_redir_ipv4.c +1 -1
index d8d795df9c13..c09d4381427e 100644
@@ -36,7 +36,7 @@ static void nft_redir_ipv4_eval(const struct nft_expr *expr,
 	mr.range[0].flags |= priv->flags;
 
 	regs->verdict.code = nf_nat_redirect_ipv4(pkt->skb, &mr,
-						  pkt->ops->hooknum);
+						  pkt->hook);
 }
 
 static struct nft_expr_type nft_redir_ipv4_type;
net/ipv4/netfilter/nft_reject_ipv4.c +2 -3
index b07e58b51158..c1582e03b628 100644
@@ -27,11 +27,10 @@ static void nft_reject_ipv4_eval(const struct nft_expr *expr,
 
 	switch (priv->type) {
 	case NFT_REJECT_ICMP_UNREACH:
-		nf_send_unreach(pkt->skb, priv->icmp_code,
-				pkt->ops->hooknum);
+		nf_send_unreach(pkt->skb, priv->icmp_code, pkt->hook);
 		break;
 	case NFT_REJECT_TCP_RST:
-		nf_send_reset(pkt->skb, pkt->ops->hooknum);
+		nf_send_reset(pkt->skb, pkt->hook);
 		break;
 	default:
 		break;
net/ipv6/netfilter/nf_tables_ipv6.c +1 -1
index c8148ba76d1a..41340b794f9b 100644
@@ -23,7 +23,7 @@ static unsigned int nft_do_chain_ipv6(const struct nf_hook_ops *ops,
 	struct nft_pktinfo pkt;
 
 	/* malformed packet, drop it */
-	if (nft_set_pktinfo_ipv6(&pkt, ops, skb, state) < 0)
+	if (nft_set_pktinfo_ipv6(&pkt, skb, state) < 0)
 		return NF_DROP;
 
 	return nft_do_chain(&pkt, ops);
net/ipv6/netfilter/nft_chain_nat_ipv6.c +1 -1
index 951bb458b7bd..e96feaefeb14 100644
@@ -31,7 +31,7 @@ static unsigned int nft_nat_do_chain(const struct nf_hook_ops *ops,
 {
 	struct nft_pktinfo pkt;
 
-	nft_set_pktinfo_ipv6(&pkt, ops, skb, state);
+	nft_set_pktinfo_ipv6(&pkt, skb, state);
 
 	return nft_do_chain(&pkt, ops);
 }
net/ipv6/netfilter/nft_chain_route_ipv6.c +1 -1
index 0dafdaac5e17..d1bcd2ed7bcc 100644
@@ -33,7 +33,7 @@ static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops,
 	u32 mark, flowlabel;
 
 	/* malformed packet, drop it */
-	if (nft_set_pktinfo_ipv6(&pkt, ops, skb, state) < 0)
+	if (nft_set_pktinfo_ipv6(&pkt, skb, state) < 0)
 		return NF_DROP;
 
 	/* save source/dest address, mark, hoplimit, flowlabel, priority */
net/ipv6/netfilter/nft_dup_ipv6.c +1 -1
index 0eaa4f65fdea..c81204faf15d 100644
@@ -28,7 +28,7 @@ static void nft_dup_ipv6_eval(const struct nft_expr *expr,
 	struct in6_addr *gw = (struct in6_addr *)&regs->data[priv->sreg_addr];
 	int oif = regs->data[priv->sreg_dev];
 
-	nf_dup_ipv6(pkt->skb, pkt->ops->hooknum, gw, oif);
+	nf_dup_ipv6(pkt->skb, pkt->hook, gw, oif);
 }
 
 static int nft_dup_ipv6_init(const struct nft_ctx *ctx,
net/ipv6/netfilter/nft_redir_ipv6.c +1 -2
index effd393bd517..aca44e89a881 100644
@@ -35,8 +35,7 @@ static void nft_redir_ipv6_eval(const struct nft_expr *expr,
 
 	range.flags |= priv->flags;
 
-	regs->verdict.code = nf_nat_redirect_ipv6(pkt->skb, &range,
-						  pkt->ops->hooknum);
+	regs->verdict.code = nf_nat_redirect_ipv6(pkt->skb, &range, pkt->hook);
 }
 
 static struct nft_expr_type nft_redir_ipv6_type;
net/ipv6/netfilter/nft_reject_ipv6.c +2 -3
index d0d1540ecf87..ffcac7d5da43 100644
@@ -28,11 +28,10 @@ static void nft_reject_ipv6_eval(const struct nft_expr *expr,
 
 	switch (priv->type) {
 	case NFT_REJECT_ICMP_UNREACH:
-		nf_send_unreach6(net, pkt->skb, priv->icmp_code,
-				 pkt->ops->hooknum);
+		nf_send_unreach6(net, pkt->skb, priv->icmp_code, pkt->hook);
 		break;
 	case NFT_REJECT_TCP_RST:
-		nf_send_reset6(net, pkt->skb, pkt->ops->hooknum);
+		nf_send_reset6(net, pkt->skb, pkt->hook);
 		break;
 	default:
 		break;
net/netfilter/nf_tables_core.c +1 -1
index 05d0b03530f6..539083099c0d 100644
@@ -50,7 +50,7 @@ static void __nft_trace_packet(const struct nft_pktinfo *pkt,
 {
 	struct net *net = dev_net(pkt->in ? pkt->in : pkt->out);
 
-	nf_log_trace(net, pkt->xt.family, pkt->ops->hooknum, pkt->skb, pkt->in,
+	nf_log_trace(net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
 		     pkt->out, &trace_loginfo, "TRACE: %s:%s:%s:%u ",
 		     chain->table->name, chain->name, comments[type],
 		     rulenum);
net/netfilter/nf_tables_netdev.c +7 -9
index 2cae4d4a03b7..db416a3396e9 100644
@@ -17,13 +17,13 @@
 
 static inline void
 nft_netdev_set_pktinfo_ipv4(struct nft_pktinfo *pkt,
-			    const struct nf_hook_ops *ops, struct sk_buff *skb,
+			    struct sk_buff *skb,
 			    const struct nf_hook_state *state)
 {
 	struct iphdr *iph, _iph;
 	u32 len, thoff;
 
-	nft_set_pktinfo(pkt, ops, skb, state);
+	nft_set_pktinfo(pkt, skb, state);
 
 	iph = skb_header_pointer(skb, skb_network_offset(skb), sizeof(*iph),
 				 &_iph);
@@ -48,7 +48,6 @@ nft_netdev_set_pktinfo_ipv4(struct nft_pktinfo *pkt,
 
 static inline void
 __nft_netdev_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
-			      const struct nf_hook_ops *ops,
 			      struct sk_buff *skb,
 			      const struct nf_hook_state *state)
 {
@@ -82,12 +81,11 @@ __nft_netdev_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
 }
 
 static inline void nft_netdev_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
-					       const struct nf_hook_ops *ops,
 					       struct sk_buff *skb,
 					       const struct nf_hook_state *state)
 {
-	nft_set_pktinfo(pkt, ops, skb, state);
-	__nft_netdev_set_pktinfo_ipv6(pkt, ops, skb, state);
+	nft_set_pktinfo(pkt, skb, state);
+	__nft_netdev_set_pktinfo_ipv6(pkt, skb, state);
 }
 
 static unsigned int
@@ -98,13 +96,13 @@ nft_do_chain_netdev(const struct nf_hook_ops *ops, struct sk_buff *skb,
 
 	switch (eth_hdr(skb)->h_proto) {
 	case htons(ETH_P_IP):
-		nft_netdev_set_pktinfo_ipv4(&pkt, ops, skb, state);
+		nft_netdev_set_pktinfo_ipv4(&pkt, skb, state);
 		break;
 	case htons(ETH_P_IPV6):
-		nft_netdev_set_pktinfo_ipv6(&pkt, ops, skb, state);
+		nft_netdev_set_pktinfo_ipv6(&pkt, skb, state);
 		break;
 	default:
-		nft_set_pktinfo(&pkt, ops, skb, state);
+		nft_set_pktinfo(&pkt, skb, state);
 		break;
 	}
 
net/netfilter/nft_log.c +1 -1
index a13d6a386d63..c7c7df85f0b7 100644
@@ -33,7 +33,7 @@ static void nft_log_eval(const struct nft_expr *expr,
 	const struct nft_log *priv = nft_expr_priv(expr);
 	struct net *net = dev_net(pkt->in ? pkt->in : pkt->out);
 
-	nf_log_packet(net, pkt->ops->pf, pkt->ops->hooknum, pkt->skb, pkt->in,
+	nf_log_packet(net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
 		      pkt->out, &priv->loginfo, "%s", priv->prefix);
 }
 
net/netfilter/nft_meta.c +2 -2
index cb2f13ebb5a6..e4ad2c24bc41 100644
@@ -42,7 +42,7 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 		*(__be16 *)dest = skb->protocol;
 		break;
 	case NFT_META_NFPROTO:
-		*dest = pkt->ops->pf;
+		*dest = pkt->pf;
 		break;
 	case NFT_META_L4PROTO:
 		*dest = pkt->tprot;
@@ -135,7 +135,7 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 			break;
 		}
 
-		switch (pkt->ops->pf) {
+		switch (pkt->pf) {
 		case NFPROTO_IPV4:
 			if (ipv4_is_multicast(ip_hdr(skb)->daddr))
 				*dest = PACKET_MULTICAST;
net/netfilter/nft_queue.c +1 -1
index 96805d21d618..61d216eb7917 100644
@@ -42,7 +42,7 @@ static void nft_queue_eval(const struct nft_expr *expr,
 			queue = priv->queuenum + cpu % priv->queues_total;
 		} else {
 			queue = nfqueue_hash(pkt->skb, queue,
-					     priv->queues_total, pkt->ops->pf,
+					     priv->queues_total, pkt->pf,
 					     jhash_initval);
 		}
 	}
net/netfilter/nft_reject_inet.c +7 -7
index 635dbba93d01..dea6750af6ff 100644
@@ -24,20 +24,20 @@ static void nft_reject_inet_eval(const struct nft_expr *expr,
 	struct nft_reject *priv = nft_expr_priv(expr);
 	struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out);
 
-	switch (pkt->ops->pf) {
+	switch (pkt->pf) {
 	case NFPROTO_IPV4:
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
 			nf_send_unreach(pkt->skb, priv->icmp_code,
-					pkt->ops->hooknum);
+					pkt->hook);
 			break;
 		case NFT_REJECT_TCP_RST:
-			nf_send_reset(pkt->skb, pkt->ops->hooknum);
+			nf_send_reset(pkt->skb, pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
 			nf_send_unreach(pkt->skb,
 					nft_reject_icmp_code(priv->icmp_code),
-					pkt->ops->hooknum);
+					pkt->hook);
 			break;
 		}
 		break;
@@ -45,15 +45,15 @@ static void nft_reject_inet_eval(const struct nft_expr *expr,
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
 			nf_send_unreach6(net, pkt->skb, priv->icmp_code,
-					 pkt->ops->hooknum);
+					 pkt->hook);
 			break;
 		case NFT_REJECT_TCP_RST:
-			nf_send_reset6(net, pkt->skb, pkt->ops->hooknum);
+			nf_send_reset6(net, pkt->skb, pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
 			nf_send_unreach6(net, pkt->skb,
 					 nft_reject_icmpv6_code(priv->icmp_code),
-					 pkt->ops->hooknum);
+					 pkt->hook);
 			break;
 		}
 		break;
Signature · 1 line
2.2.1
#49 lore raw

PATCH 5/14 x_tables: Pass struct net in xt_action_param

In reply to Eric W. Biederman

As xt_action_param lives on the stack this does not bloat any persistent data structures.

This is a first step in making netfilter code that needs to know which network namespace it is executing in simpler.

---
 include/linux/netfilter/x_tables.h | 3 ++-
 include/net/netfilter/nf_tables.h  | 1 +
 net/bridge/netfilter/ebtables.c    | 1 +
 net/ipv4/netfilter/arp_tables.c    | 1 +
 net/ipv4/netfilter/ip_tables.c     | 1 +
 net/ipv6/netfilter/ip6_tables.c    | 1 +
 net/sched/act_ipt.c                | 1 +
 net/sched/em_ipset.c               | 1 +
 8 files changed, 9 insertions(+), 1 deletion(-)
include/linux/netfilter/x_tables.h +2 -1
index b006b719183f..c5577410c25d 100644
@@ -13,6 +13,7 @@
  * @target:	the target extension
  * @matchinfo:	per-match data
  * @targetinfo:	per-target data
+ * @net		network namespace through which the action was invoked
  * @in:		input netdevice
  * @out:	output netdevice
  * @fragoff:	packet is a fragment, this is the data offset
@@ -24,7 +25,6 @@
  * Fields written to by extensions:
  *
  * @hotdrop:	drop packet if we had inspection problems
- * Network namespace obtainable using dev_net(in/out)
  */
 struct xt_action_param {
 	union {
@@ -34,6 +34,7 @@ struct xt_action_param {
 	union {
 		const void *matchinfo, *targinfo;
 	};
+	struct net *net;
 	const struct net_device *in, *out;
 	int fragoff;
 	unsigned int thoff;
include/net/netfilter/nf_tables.h +1 -0
index c0899f97ff8d..c0516529e8a0 100644
@@ -30,6 +30,7 @@ static inline void nft_set_pktinfo(struct nft_pktinfo *pkt,
 				   const struct nf_hook_state *state)
 {
 	pkt->skb = skb;
+	pkt->xt.net = state->net;
 	pkt->in = pkt->xt.in = state->in;
 	pkt->out = pkt->xt.out = state->out;
 	pkt->hook = pkt->xt.hooknum = state->hook;
net/bridge/netfilter/ebtables.c +1 -0
index 8d5a3975b963..f46ca417bf2d 100644
@@ -200,6 +200,7 @@ unsigned int ebt_do_table(struct sk_buff *skb,
 	struct xt_action_param acpar;
 
 	acpar.family  = NFPROTO_BRIDGE;
+	acpar.net     = state->net;
 	acpar.in      = state->in;
 	acpar.out     = state->out;
 	acpar.hotdrop = false;
net/ipv4/netfilter/arp_tables.c +1 -0
index 10eb2b297450..2dad3e1c5f11 100644
@@ -285,6 +285,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 	 */
 	e = get_entry(table_base, private->hook_entry[hook]);
 
+	acpar.net     = state->net;
 	acpar.in      = state->in;
 	acpar.out     = state->out;
 	acpar.hooknum = hook;
net/ipv4/netfilter/ip_tables.c +1 -0
index 2b049e135de8..42d0946956db 100644
@@ -315,6 +315,7 @@ ipt_do_table(struct sk_buff *skb,
 	acpar.fragoff = ntohs(ip->frag_off) & IP_OFFSET;
 	acpar.thoff   = ip_hdrlen(skb);
 	acpar.hotdrop = false;
+	acpar.net     = state->net;
 	acpar.in      = state->in;
 	acpar.out     = state->out;
 	acpar.family  = NFPROTO_IPV4;
net/ipv6/netfilter/ip6_tables.c +1 -0
index da6446b6e3f9..80e3bd72b715 100644
@@ -340,6 +340,7 @@ ip6t_do_table(struct sk_buff *skb,
 	 * rule is also a fragment-specific rule, non-fragments won't
 	 * match it. */
 	acpar.hotdrop = false;
+	acpar.net     = state->net;
 	acpar.in      = state->in;
 	acpar.out     = state->out;
 	acpar.family  = NFPROTO_IPV6;
net/sched/act_ipt.c +1 -0
index 99c9cc1c7af9..d05869646515 100644
@@ -189,6 +189,7 @@ static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a,
 	 * worry later - danger - this API seems to have changed
 	 * from earlier kernels
 	 */
+	par.net	     = dev_net(skb->dev);
 	par.in       = skb->dev;
 	par.out      = NULL;
 	par.hooknum  = ipt->tcfi_hook;
net/sched/em_ipset.c +1 -0
index df0328ba6a48..c66ca9400ab4 100644
@@ -95,6 +95,7 @@ static int em_ipset_match(struct sk_buff *skb, struct tcf_ematch *em,
 	if (skb->skb_iif)
 		indev = dev_get_by_index_rcu(em->net, skb->skb_iif);
 
+	acpar.net     = em->net;
 	acpar.in      = indev ? indev : dev;
 	acpar.out     = dev;
 
Signature · 1 line
2.2.1
#50 lore raw

PATCH 6/14 x_tables: Use par->net instead of computing from the passed net devices

In reply to Eric W. Biederman
---
 net/bridge/netfilter/ebt_log.c     |  2 +-
 net/bridge/netfilter/ebt_nflog.c   |  2 +-
 net/ipv4/netfilter/ipt_SYNPROXY.c  |  2 +-
 net/ipv4/netfilter/ipt_rpfilter.c  |  5 ++---
 net/ipv6/netfilter/ip6t_REJECT.c   |  2 +-
 net/ipv6/netfilter/ip6t_SYNPROXY.c |  2 +-
 net/ipv6/netfilter/ip6t_rpfilter.c |  6 +++---
 net/netfilter/ipset/ip_set_core.c  |  9 +++------
 net/netfilter/xt_LOG.c             |  2 +-
 net/netfilter/xt_NFLOG.c           |  2 +-
 net/netfilter/xt_TCPMSS.c          |  2 +-
 net/netfilter/xt_TPROXY.c          | 24 ++++++++++++------------
 net/netfilter/xt_addrtype.c        |  4 ++--
 net/netfilter/xt_connlimit.c       |  2 +-
 net/netfilter/xt_osf.c             |  2 +-
 net/netfilter/xt_recent.c          |  2 +-
 net/netfilter/xt_socket.c          | 14 ++++++++------
 17 files changed, 41 insertions(+), 43 deletions(-)
net/bridge/netfilter/ebt_log.c +1 -1
index 17f2e4bc2a29..0ad639a96142 100644
@@ -180,7 +180,7 @@ ebt_log_tg(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct ebt_log_info *info = par->targinfo;
 	struct nf_loginfo li;
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 
 	li.type = NF_LOG_TYPE_LOG;
 	li.u.log.level = info->loglevel;
net/bridge/netfilter/ebt_nflog.c +1 -1
index 59ac7952010d..54816150608e 100644
@@ -24,7 +24,7 @@ ebt_nflog_tg(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct ebt_nflog_info *info = par->targinfo;
 	struct nf_loginfo li;
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 
 	li.type = NF_LOG_TYPE_ULOG;
 	li.u.ulog.copy_len = info->len;
net/ipv4/netfilter/ipt_SYNPROXY.c +1 -1
index f471a0628c75..dfab314981e9 100644
@@ -258,7 +258,7 @@ static unsigned int
 synproxy_tg4(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_synproxy_info *info = par->targinfo;
-	struct synproxy_net *snet = synproxy_pernet(dev_net(par->in));
+	struct synproxy_net *snet = synproxy_pernet(par->net);
 	struct synproxy_options opts = {};
 	struct tcphdr *th, _th;
 
net/ipv4/netfilter/ipt_rpfilter.c +2 -3
index 8618fd150c96..74dd6671b66d 100644
@@ -32,12 +32,11 @@ static __be32 rpfilter_get_saddr(__be32 addr)
 	return addr;
 }
 
-static bool rpfilter_lookup_reverse(struct flowi4 *fl4,
+static bool rpfilter_lookup_reverse(struct net *net, struct flowi4 *fl4,
 				const struct net_device *dev, u8 flags)
 {
 	struct fib_result res;
 	bool dev_match;
-	struct net *net = dev_net(dev);
 	int ret __maybe_unused;
 
 	if (fib_lookup(net, fl4, &res, FIB_LOOKUP_IGNORE_LINKSTATE))
@@ -98,7 +97,7 @@ static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par)
 	flow.flowi4_tos = RT_TOS(iph->tos);
 	flow.flowi4_scope = RT_SCOPE_UNIVERSE;
 
-	return rpfilter_lookup_reverse(&flow, par->in, info->flags) ^ invert;
+	return rpfilter_lookup_reverse(par->net, &flow, par->in, info->flags) ^ invert;
 }
 
 static int rpfilter_check(const struct xt_mtchk_param *par)
net/ipv6/netfilter/ip6t_REJECT.c +1 -1
index 0ed841a3fa33..db29bbf41b59 100644
@@ -39,7 +39,7 @@ static unsigned int
 reject_tg6(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct ip6t_reject_info *reject = par->targinfo;
-	struct net *net = dev_net((par->in != NULL) ? par->in : par->out);
+	struct net *net = par->net;
 
 	switch (reject->with) {
 	case IP6T_ICMP6_NO_ROUTE:
net/ipv6/netfilter/ip6t_SYNPROXY.c +1 -1
index 4c9f3e79d75f..41451809b37c 100644
@@ -275,7 +275,7 @@ static unsigned int
 synproxy_tg6(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_synproxy_info *info = par->targinfo;
-	struct synproxy_net *snet = synproxy_pernet(dev_net(par->in));
+	struct synproxy_net *snet = synproxy_pernet(par->net);
 	struct synproxy_options opts = {};
 	struct tcphdr *th, _th;
 
net/ipv6/netfilter/ip6t_rpfilter.c +3 -3
index 790e0c6b19e1..1ee1b25df096 100644
@@ -26,7 +26,7 @@ static bool rpfilter_addr_unicast(const struct in6_addr *addr)
 	return addr_type & IPV6_ADDR_UNICAST;
 }
 
-static bool rpfilter_lookup_reverse6(const struct sk_buff *skb,
+static bool rpfilter_lookup_reverse6(struct net *net, const struct sk_buff *skb,
 				     const struct net_device *dev, u8 flags)
 {
 	struct rt6_info *rt;
@@ -53,7 +53,7 @@ static bool rpfilter_lookup_reverse6(const struct sk_buff *skb,
 		lookup_flags |= RT6_LOOKUP_F_IFACE;
 	}
 
-	rt = (void *) ip6_route_lookup(dev_net(dev), &fl6, lookup_flags);
+	rt = (void *) ip6_route_lookup(net, &fl6, lookup_flags);
 	if (rt->dst.error)
 		goto out;
 
@@ -93,7 +93,7 @@ static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par)
 	if (unlikely(saddrtype == IPV6_ADDR_ANY))
 		return true ^ invert; /* not routable: forward path will drop it */
 
-	return rpfilter_lookup_reverse6(skb, par->in, info->flags) ^ invert;
+	return rpfilter_lookup_reverse6(par->net, skb, par->in, info->flags) ^ invert;
 }
 
 static int rpfilter_check(const struct xt_mtchk_param *par)
net/netfilter/ipset/ip_set_core.c +3 -6
index 338b4047776f..69ab9c2634e1 100644
@@ -519,8 +519,7 @@ int
 ip_set_test(ip_set_id_t index, const struct sk_buff *skb,
 	    const struct xt_action_param *par, struct ip_set_adt_opt *opt)
 {
-	struct ip_set *set = ip_set_rcu_get(
-			dev_net(par->in ? par->in : par->out), index);
+	struct ip_set *set = ip_set_rcu_get(par->net, index);
 	int ret = 0;
 
 	BUG_ON(!set);
@@ -558,8 +557,7 @@ int
 ip_set_add(ip_set_id_t index, const struct sk_buff *skb,
 	   const struct xt_action_param *par, struct ip_set_adt_opt *opt)
 {
-	struct ip_set *set = ip_set_rcu_get(
-			dev_net(par->in ? par->in : par->out), index);
+	struct ip_set *set = ip_set_rcu_get(par->net, index);
 	int ret;
 
 	BUG_ON(!set);
@@ -581,8 +579,7 @@ int
 ip_set_del(ip_set_id_t index, const struct sk_buff *skb,
 	   const struct xt_action_param *par, struct ip_set_adt_opt *opt)
 {
-	struct ip_set *set = ip_set_rcu_get(
-			dev_net(par->in ? par->in : par->out), index);
+	struct ip_set *set = ip_set_rcu_get(par->net, index);
 	int ret = 0;
 
 	BUG_ON(!set);
net/netfilter/xt_LOG.c +1 -1
index c13b79440ede..1763ab82bcd7 100644
@@ -33,7 +33,7 @@ log_tg(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_log_info *loginfo = par->targinfo;
 	struct nf_loginfo li;
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 
 	li.type = NF_LOG_TYPE_LOG;
 	li.u.log.level = loginfo->level;
net/netfilter/xt_NFLOG.c +1 -1
index fb7497c928a0..a1fa2c800cb9 100644
@@ -26,7 +26,7 @@ nflog_tg(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_nflog_info *info = par->targinfo;
 	struct nf_loginfo li;
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 
 	li.type		     = NF_LOG_TYPE_ULOG;
 	li.u.ulog.copy_len   = info->len;
net/netfilter/xt_TCPMSS.c +1 -1
index 8c02501a530f..b7c43def0dc6 100644
@@ -108,7 +108,7 @@ tcpmss_mangle_packet(struct sk_buff *skb,
 		return -1;
 
 	if (info->mss == XT_TCPMSS_CLAMP_PMTU) {
-		struct net *net = dev_net(par->in ? par->in : par->out);
+		struct net *net = par->net;
 		unsigned int in_mtu = tcpmss_reverse_mtu(net, skb, family);
 
 		if (dst_mtu(skb_dst(skb)) <= minlen) {
net/netfilter/xt_TPROXY.c +12 -12
index d0c96c5ae29a..3ab591e73ec0 100644
@@ -250,8 +250,8 @@ nf_tproxy_get_sock_v6(struct net *net, const u8 protocol,
  * no such listener is found, or NULL if the TCP header is incomplete.
  */
 static struct sock *
-tproxy_handle_time_wait4(struct sk_buff *skb, __be32 laddr, __be16 lport,
-			struct sock *sk)
+tproxy_handle_time_wait4(struct net *net, struct sk_buff *skb,
+			 __be32 laddr, __be16 lport, struct sock *sk)
 {
 	const struct iphdr *iph = ip_hdr(skb);
 	struct tcphdr _hdr, *hp;
@@ -267,7 +267,7 @@ tproxy_handle_time_wait4(struct sk_buff *skb, __be32 laddr, __be16 lport,
 		 * to a listener socket if there's one */
 		struct sock *sk2;
 
-		sk2 = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol,
+		sk2 = nf_tproxy_get_sock_v4(net, iph->protocol,
 					    iph->saddr, laddr ? laddr : iph->daddr,
 					    hp->source, lport ? lport : hp->dest,
 					    skb->dev, NFT_LOOKUP_LISTENER);
@@ -290,7 +290,7 @@ nf_tproxy_assign_sock(struct sk_buff *skb, struct sock *sk)
 }
 
 static unsigned int
-tproxy_tg4(struct sk_buff *skb, __be32 laddr, __be16 lport,
+tproxy_tg4(struct net *net, struct sk_buff *skb, __be32 laddr, __be16 lport,
 	   u_int32_t mark_mask, u_int32_t mark_value)
 {
 	const struct iphdr *iph = ip_hdr(skb);
@@ -305,7 +305,7 @@ tproxy_tg4(struct sk_buff *skb, __be32 laddr, __be16 lport,
 	 * addresses, this happens if the redirect already happened
 	 * and the current packet belongs to an already established
 	 * connection */
-	sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol,
+	sk = nf_tproxy_get_sock_v4(net, iph->protocol,
 				   iph->saddr, iph->daddr,
 				   hp->source, hp->dest,
 				   skb->dev, NFT_LOOKUP_ESTABLISHED);
@@ -317,11 +317,11 @@ tproxy_tg4(struct sk_buff *skb, __be32 laddr, __be16 lport,
 	/* UDP has no TCP_TIME_WAIT state, so we never enter here */
 	if (sk && sk->sk_state == TCP_TIME_WAIT)
 		/* reopening a TIME_WAIT connection needs special handling */
-		sk = tproxy_handle_time_wait4(skb, laddr, lport, sk);
+		sk = tproxy_handle_time_wait4(net, skb, laddr, lport, sk);
 	else if (!sk)
 		/* no, there's no established connection, check if
 		 * there's a listener on the redirected addr/port */
-		sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol,
+		sk = nf_tproxy_get_sock_v4(net, iph->protocol,
 					   iph->saddr, laddr,
 					   hp->source, lport,
 					   skb->dev, NFT_LOOKUP_LISTENER);
@@ -351,7 +351,7 @@ tproxy_tg4_v0(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_tproxy_target_info *tgi = par->targinfo;
 
-	return tproxy_tg4(skb, tgi->laddr, tgi->lport, tgi->mark_mask, tgi->mark_value);
+	return tproxy_tg4(par->net, skb, tgi->laddr, tgi->lport, tgi->mark_mask, tgi->mark_value);
 }
 
 static unsigned int
@@ -359,7 +359,7 @@ tproxy_tg4_v1(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_tproxy_target_info_v1 *tgi = par->targinfo;
 
-	return tproxy_tg4(skb, tgi->laddr.ip, tgi->lport, tgi->mark_mask, tgi->mark_value);
+	return tproxy_tg4(par->net, skb, tgi->laddr.ip, tgi->lport, tgi->mark_mask, tgi->mark_value);
 }
 
 #ifdef XT_TPROXY_HAVE_IPV6
@@ -429,7 +429,7 @@ tproxy_handle_time_wait6(struct sk_buff *skb, int tproto, int thoff,
 		 * to a listener socket if there's one */
 		struct sock *sk2;
 
-		sk2 = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto,
+		sk2 = nf_tproxy_get_sock_v6(par->net, tproto,
 					    &iph->saddr,
 					    tproxy_laddr6(skb, &tgi->laddr.in6, &iph->daddr),
 					    hp->source,
@@ -472,7 +472,7 @@ tproxy_tg6_v1(struct sk_buff *skb, const struct xt_action_param *par)
 	 * addresses, this happens if the redirect already happened
 	 * and the current packet belongs to an already established
 	 * connection */
-	sk = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto,
+	sk = nf_tproxy_get_sock_v6(par->net, tproto,
 				   &iph->saddr, &iph->daddr,
 				   hp->source, hp->dest,
 				   par->in, NFT_LOOKUP_ESTABLISHED);
@@ -487,7 +487,7 @@ tproxy_tg6_v1(struct sk_buff *skb, const struct xt_action_param *par)
 	else if (!sk)
 		/* no there's no established connection, check if
 		 * there's a listener on the redirected addr/port */
-		sk = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto,
+		sk = nf_tproxy_get_sock_v6(par->net, tproto,
 					   &iph->saddr, laddr,
 					   hp->source, lport,
 					   par->in, NFT_LOOKUP_LISTENER);
net/netfilter/xt_addrtype.c +2 -2
index 5b4743cc0436..11d6091991a4 100644
@@ -125,7 +125,7 @@ static inline bool match_type(struct net *net, const struct net_device *dev,
 static bool
 addrtype_mt_v0(const struct sk_buff *skb, struct xt_action_param *par)
 {
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 	const struct xt_addrtype_info *info = par->matchinfo;
 	const struct iphdr *iph = ip_hdr(skb);
 	bool ret = true;
@@ -143,7 +143,7 @@ addrtype_mt_v0(const struct sk_buff *skb, struct xt_action_param *par)
 static bool
 addrtype_mt_v1(const struct sk_buff *skb, struct xt_action_param *par)
 {
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 	const struct xt_addrtype_info_v1 *info = par->matchinfo;
 	const struct iphdr *iph;
 	const struct net_device *dev = NULL;
net/netfilter/xt_connlimit.c +1 -1
index 075d89d94d28..213db252e5be 100644
@@ -317,7 +317,7 @@ static int count_them(struct net *net,
 static bool
 connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par)
 {
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 	const struct xt_connlimit_info *info = par->matchinfo;
 	union nf_inet_addr addr;
 	struct nf_conntrack_tuple tuple;
net/netfilter/xt_osf.c +1 -1
index 0778855ea5e7..df8801e02a32 100644
@@ -200,7 +200,7 @@ xt_osf_match_packet(const struct sk_buff *skb, struct xt_action_param *p)
 	unsigned char opts[MAX_IPOPTLEN];
 	const struct xt_osf_finger *kf;
 	const struct xt_osf_user_finger *f;
-	struct net *net = dev_net(p->in ? p->in : p->out);
+	struct net *net = p->net;
 
 	if (!info)
 		return false;
net/netfilter/xt_recent.c +1 -1
index 45e1b30e4fb2..d725a27743a1 100644
@@ -237,7 +237,7 @@ static void recent_table_flush(struct recent_table *t)
 static bool
 recent_mt(const struct sk_buff *skb, struct xt_action_param *par)
 {
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 	struct recent_net *recent_net = recent_pernet(net);
 	const struct xt_recent_mtinfo_v1 *info = par->matchinfo;
 	struct recent_table *t;
net/netfilter/xt_socket.c +8 -6
index 43e26c881100..2ec08f04b816 100644
@@ -143,7 +143,8 @@ static bool xt_socket_sk_is_transparent(struct sock *sk)
 	}
 }
 
-static struct sock *xt_socket_lookup_slow_v4(const struct sk_buff *skb,
+static struct sock *xt_socket_lookup_slow_v4(struct net *net,
+					     const struct sk_buff *skb,
 					     const struct net_device *indev)
 {
 	const struct iphdr *iph = ip_hdr(skb);
@@ -197,7 +198,7 @@ static struct sock *xt_socket_lookup_slow_v4(const struct sk_buff *skb,
 	}
 #endif
 
-	return xt_socket_get_sock_v4(dev_net(skb->dev), protocol, saddr, daddr,
+	return xt_socket_get_sock_v4(net, protocol, saddr, daddr,
 				     sport, dport, indev);
 }
 
@@ -209,7 +210,7 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par,
 	struct sock *sk = skb->sk;
 
 	if (!sk)
-		sk = xt_socket_lookup_slow_v4(skb, par->in);
+		sk = xt_socket_lookup_slow_v4(par->net, skb, par->in);
 	if (sk) {
 		bool wildcard;
 		bool transparent = true;
@@ -335,7 +336,8 @@ xt_socket_get_sock_v6(struct net *net, const u8 protocol,
 	return NULL;
 }
 
-static struct sock *xt_socket_lookup_slow_v6(const struct sk_buff *skb,
+static struct sock *xt_socket_lookup_slow_v6(struct net *net,
+					     const struct sk_buff *skb,
 					     const struct net_device *indev)
 {
 	__be16 uninitialized_var(dport), uninitialized_var(sport);
@@ -371,7 +373,7 @@ static struct sock *xt_socket_lookup_slow_v6(const struct sk_buff *skb,
 		return NULL;
 	}
 
-	return xt_socket_get_sock_v6(dev_net(skb->dev), tproto, saddr, daddr,
+	return xt_socket_get_sock_v6(net, tproto, saddr, daddr,
 				     sport, dport, indev);
 }
 
@@ -383,7 +385,7 @@ socket_mt6_v1_v2_v3(const struct sk_buff *skb, struct xt_action_param *par)
 	struct sock *sk = skb->sk;
 
 	if (!sk)
-		sk = xt_socket_lookup_slow_v6(skb, par->in);
+		sk = xt_socket_lookup_slow_v6(par->net, skb, par->in);
 	if (sk) {
 		bool wildcard;
 		bool transparent = true;
Signature · 1 line
2.2.1
#51 lore raw

PATCH 7/14 nftables: Pass struct net in nft_pktinfo

In reply to Eric W. Biederman

nft_pktinfo is passed on the stack so this does not bloat any in core data structures.

By centrally computing this information this makes maintence of the code simpler, and understading of the code easier.

---
 include/net/netfilter/nf_tables.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
include/net/netfilter/nf_tables.h +2 -1
index c0516529e8a0..42e239e55aa3 100644
@@ -14,6 +14,7 @@
 
 struct nft_pktinfo {
 	struct sk_buff			*skb;
+	struct net			*net;
 	const struct net_device		*in;
 	const struct net_device		*out;
 	u8				pf;
@@ -30,7 +31,7 @@ static inline void nft_set_pktinfo(struct nft_pktinfo *pkt,
 				   const struct nf_hook_state *state)
 {
 	pkt->skb = skb;
-	pkt->xt.net = state->net;
+	pkt->net = pkt->xt.net = state->net;
 	pkt->in = pkt->xt.in = state->in;
 	pkt->out = pkt->xt.out = state->out;
 	pkt->hook = pkt->xt.hooknum = state->hook;
Signature · 1 line
2.2.1
#52 lore raw

PATCH 8/14 nftables: Use pkt->net instead of computing net from the passed net_devices

In reply to Eric W. Biederman
---
 net/bridge/netfilter/nft_reject_bridge.c | 13 ++++++-------
 net/ipv6/netfilter/nft_reject_ipv6.c     |  6 +++---
 net/netfilter/nf_tables_core.c           |  6 ++----
 net/netfilter/nft_log.c                  |  3 +--
 net/netfilter/nft_reject_inet.c          |  7 +++----
 5 files changed, 15 insertions(+), 20 deletions(-)
net/bridge/netfilter/nft_reject_bridge.c +6 -7
index cee92612b2cc..fdba3d9fbff3 100644
@@ -261,7 +261,6 @@ static void nft_reject_bridge_eval(const struct nft_expr *expr,
 				   const struct nft_pktinfo *pkt)
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
-	struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out);
 	const unsigned char *dest = eth_hdr(pkt->skb)->h_dest;
 
 	if (is_broadcast_ether_addr(dest) ||
@@ -290,17 +289,17 @@ static void nft_reject_bridge_eval(const struct nft_expr *expr,
 	case htons(ETH_P_IPV6):
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
-			nft_reject_br_send_v6_unreach(net, pkt->skb, pkt->in,
-						      pkt->hook,
+			nft_reject_br_send_v6_unreach(pkt->net, pkt->skb,
+						      pkt->in, pkt->hook,
 						      priv->icmp_code);
 			break;
 		case NFT_REJECT_TCP_RST:
-			nft_reject_br_send_v6_tcp_reset(net, pkt->skb, pkt->in,
-							pkt->hook);
+			nft_reject_br_send_v6_tcp_reset(pkt->net, pkt->skb,
+							pkt->in, pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
-			nft_reject_br_send_v6_unreach(net, pkt->skb, pkt->in,
-						      pkt->hook,
+			nft_reject_br_send_v6_unreach(pkt->net, pkt->skb,
+						      pkt->in, pkt->hook,
 						      nft_reject_icmpv6_code(priv->icmp_code));
 			break;
 		}
net/ipv6/netfilter/nft_reject_ipv6.c +3 -3
index ffcac7d5da43..533cd5719c59 100644
@@ -24,14 +24,14 @@ static void nft_reject_ipv6_eval(const struct nft_expr *expr,
 				 const struct nft_pktinfo *pkt)
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
-	struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out);
 
 	switch (priv->type) {
 	case NFT_REJECT_ICMP_UNREACH:
-		nf_send_unreach6(net, pkt->skb, priv->icmp_code, pkt->hook);
+		nf_send_unreach6(pkt->net, pkt->skb, priv->icmp_code,
+				 pkt->hook);
 		break;
 	case NFT_REJECT_TCP_RST:
-		nf_send_reset6(net, pkt->skb, pkt->hook);
+		nf_send_reset6(pkt->net, pkt->skb, pkt->hook);
 		break;
 	default:
 		break;
net/netfilter/nf_tables_core.c +2 -4
index 539083099c0d..e5c1f332e45e 100644
@@ -48,9 +48,7 @@ static void __nft_trace_packet(const struct nft_pktinfo *pkt,
 			       const struct nft_chain *chain,
 			       int rulenum, enum nft_trace type)
 {
-	struct net *net = dev_net(pkt->in ? pkt->in : pkt->out);
-
-	nf_log_trace(net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
+	nf_log_trace(pkt->net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
 		     pkt->out, &trace_loginfo, "TRACE: %s:%s:%s:%u ",
 		     chain->table->name, chain->name, comments[type],
 		     rulenum);
@@ -114,7 +112,7 @@ unsigned int
 nft_do_chain(struct nft_pktinfo *pkt, const struct nf_hook_ops *ops)
 {
 	const struct nft_chain *chain = ops->priv, *basechain = chain;
-	const struct net *net = dev_net(pkt->in ? pkt->in : pkt->out);
+	const struct net *net = pkt->net;
 	const struct nft_rule *rule;
 	const struct nft_expr *expr, *last;
 	struct nft_regs regs;
net/netfilter/nft_log.c +1 -2
index c7c7df85f0b7..319c22b4bca2 100644
@@ -31,9 +31,8 @@ static void nft_log_eval(const struct nft_expr *expr,
 			 const struct nft_pktinfo *pkt)
 {
 	const struct nft_log *priv = nft_expr_priv(expr);
-	struct net *net = dev_net(pkt->in ? pkt->in : pkt->out);
 
-	nf_log_packet(net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
+	nf_log_packet(pkt->net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
 		      pkt->out, &priv->loginfo, "%s", priv->prefix);
 }
 
net/netfilter/nft_reject_inet.c +3 -4
index dea6750af6ff..0bc19f97e238 100644
@@ -22,7 +22,6 @@ static void nft_reject_inet_eval(const struct nft_expr *expr,
 				 const struct nft_pktinfo *pkt)
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
-	struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out);
 
 	switch (pkt->pf) {
 	case NFPROTO_IPV4:
@@ -44,14 +43,14 @@ static void nft_reject_inet_eval(const struct nft_expr *expr,
 	case NFPROTO_IPV6:
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
-			nf_send_unreach6(net, pkt->skb, priv->icmp_code,
+			nf_send_unreach6(pkt->net, pkt->skb, priv->icmp_code,
 					 pkt->hook);
 			break;
 		case NFT_REJECT_TCP_RST:
-			nf_send_reset6(net, pkt->skb, pkt->hook);
+			nf_send_reset6(pkt->net, pkt->skb, pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
-			nf_send_unreach6(net, pkt->skb,
+			nf_send_unreach6(pkt->net, pkt->skb,
 					 nft_reject_icmpv6_code(priv->icmp_code),
 					 pkt->hook);
 			break;
Signature · 1 line
2.2.1
#53 lore raw

PATCH 9/14 net: Pass net to nf_dup_ipv4 and nf_dup_ipv6

In reply to Eric W. Biederman
---
 include/net/netfilter/ipv4/nf_dup_ipv4.h |  2 +-
 include/net/netfilter/ipv6/nf_dup_ipv6.h |  2 +-
 net/ipv4/netfilter/nf_dup_ipv4.c         | 23 ++++-------------------
 net/ipv4/netfilter/nft_dup_ipv4.c        |  2 +-
 net/ipv6/netfilter/nf_dup_ipv6.c         | 23 ++++-------------------
 net/ipv6/netfilter/nft_dup_ipv6.c        |  2 +-
 net/netfilter/xt_TEE.c                   |  4 ++--
 7 files changed, 14 insertions(+), 44 deletions(-)
include/net/netfilter/ipv4/nf_dup_ipv4.h +1 -1
index 42008f10dfc4..0a14733e8b82 100644
@@ -1,7 +1,7 @@
 #ifndef _NF_DUP_IPV4_H_
 #define _NF_DUP_IPV4_H_
 
-void nf_dup_ipv4(struct sk_buff *skb, unsigned int hooknum,
+void nf_dup_ipv4(struct net *net, struct sk_buff *skb, unsigned int hooknum,
 		 const struct in_addr *gw, int oif);
 
 #endif /* _NF_DUP_IPV4_H_ */
include/net/netfilter/ipv6/nf_dup_ipv6.h +1 -1
index ed6bd66fa5a0..fa6237b382a3 100644
@@ -1,7 +1,7 @@
 #ifndef _NF_DUP_IPV6_H_
 #define _NF_DUP_IPV6_H_
 
-void nf_dup_ipv6(struct sk_buff *skb, unsigned int hooknum,
+void nf_dup_ipv6(struct net *net, struct sk_buff *skb, unsigned int hooknum,
 		 const struct in6_addr *gw, int oif);
 
 #endif /* _NF_DUP_IPV6_H_ */
net/ipv4/netfilter/nf_dup_ipv4.c +4 -19
index 2d79e6e8d934..ce2a59e5c665 100644
@@ -23,25 +23,10 @@
 #include <net/netfilter/nf_conntrack.h>
 #endif
 
-static struct net *pick_net(struct sk_buff *skb)
-{
-#ifdef CONFIG_NET_NS
-	const struct dst_entry *dst;
-
-	if (skb->dev != NULL)
-		return dev_net(skb->dev);
-	dst = skb_dst(skb);
-	if (dst != NULL && dst->dev != NULL)
-		return dev_net(dst->dev);
-#endif
-	return &init_net;
-}
-
-static bool nf_dup_ipv4_route(struct sk_buff *skb, const struct in_addr *gw,
-			      int oif)
+static bool nf_dup_ipv4_route(struct net *net, struct sk_buff *skb,
+			      const struct in_addr *gw, int oif)
 {
 	const struct iphdr *iph = ip_hdr(skb);
-	struct net *net = pick_net(skb);
 	struct rtable *rt;
 	struct flowi4 fl4;
 
@@ -65,7 +50,7 @@ static bool nf_dup_ipv4_route(struct sk_buff *skb, const struct in_addr *gw,
 	return true;
 }
 
-void nf_dup_ipv4(struct sk_buff *skb, unsigned int hooknum,
+void nf_dup_ipv4(struct net *net, struct sk_buff *skb, unsigned int hooknum,
 		 const struct in_addr *gw, int oif)
 {
 	struct iphdr *iph;
@@ -105,7 +90,7 @@ void nf_dup_ipv4(struct sk_buff *skb, unsigned int hooknum,
 		--iph->ttl;
 	ip_send_check(iph);
 
-	if (nf_dup_ipv4_route(skb, gw, oif)) {
+	if (nf_dup_ipv4_route(net, skb, gw, oif)) {
 		__this_cpu_write(nf_skb_duplicated, true);
 		ip_local_out(skb);
 		__this_cpu_write(nf_skb_duplicated, false);
net/ipv4/netfilter/nft_dup_ipv4.c +1 -1
index 30bcf820e8bd..bf855e64fc45 100644
@@ -30,7 +30,7 @@ static void nft_dup_ipv4_eval(const struct nft_expr *expr,
 	};
 	int oif = regs->data[priv->sreg_dev];
 
-	nf_dup_ipv4(pkt->skb, pkt->hook, &gw, oif);
+	nf_dup_ipv4(pkt->net, pkt->skb, pkt->hook, &gw, oif);
 }
 
 static int nft_dup_ipv4_init(const struct nft_ctx *ctx,
net/ipv6/netfilter/nf_dup_ipv6.c +4 -19
index c8ab626556a0..ee0d9a5b16c3 100644
@@ -19,25 +19,10 @@
 #include <net/netfilter/nf_conntrack.h>
 #endif
 
-static struct net *pick_net(struct sk_buff *skb)
-{
-#ifdef CONFIG_NET_NS
-	const struct dst_entry *dst;
-
-	if (skb->dev != NULL)
-		return dev_net(skb->dev);
-	dst = skb_dst(skb);
-	if (dst != NULL && dst->dev != NULL)
-		return dev_net(dst->dev);
-#endif
-	return &init_net;
-}
-
-static bool nf_dup_ipv6_route(struct sk_buff *skb, const struct in6_addr *gw,
-			      int oif)
+static bool nf_dup_ipv6_route(struct net *net, struct sk_buff *skb,
+			      const struct in6_addr *gw, int oif)
 {
 	const struct ipv6hdr *iph = ipv6_hdr(skb);
-	struct net *net = pick_net(skb);
 	struct dst_entry *dst;
 	struct flowi6 fl6;
 
@@ -61,7 +46,7 @@ static bool nf_dup_ipv6_route(struct sk_buff *skb, const struct in6_addr *gw,
 	return true;
 }
 
-void nf_dup_ipv6(struct sk_buff *skb, unsigned int hooknum,
+void nf_dup_ipv6(struct net *net, struct sk_buff *skb, unsigned int hooknum,
 		 const struct in6_addr *gw, int oif)
 {
 	if (this_cpu_read(nf_skb_duplicated))
@@ -81,7 +66,7 @@ void nf_dup_ipv6(struct sk_buff *skb, unsigned int hooknum,
 		struct ipv6hdr *iph = ipv6_hdr(skb);
 		--iph->hop_limit;
 	}
-	if (nf_dup_ipv6_route(skb, gw, oif)) {
+	if (nf_dup_ipv6_route(net, skb, gw, oif)) {
 		__this_cpu_write(nf_skb_duplicated, true);
 		ip6_local_out(skb);
 		__this_cpu_write(nf_skb_duplicated, false);
net/ipv6/netfilter/nft_dup_ipv6.c +1 -1
index c81204faf15d..8bfd470cbe72 100644
@@ -28,7 +28,7 @@ static void nft_dup_ipv6_eval(const struct nft_expr *expr,
 	struct in6_addr *gw = (struct in6_addr *)&regs->data[priv->sreg_addr];
 	int oif = regs->data[priv->sreg_dev];
 
-	nf_dup_ipv6(pkt->skb, pkt->hook, gw, oif);
+	nf_dup_ipv6(pkt->net, pkt->skb, pkt->hook, gw, oif);
 }
 
 static int nft_dup_ipv6_init(const struct nft_ctx *ctx,
net/netfilter/xt_TEE.c +2 -2
index fd980aa7715d..899b06115fc5 100644
@@ -32,7 +32,7 @@ tee_tg4(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_tee_tginfo *info = par->targinfo;
 
-	nf_dup_ipv4(skb, par->hooknum, &info->gw.in, info->priv->oif);
+	nf_dup_ipv4(par->net, skb, par->hooknum, &info->gw.in, info->priv->oif);
 
 	return XT_CONTINUE;
 }
@@ -43,7 +43,7 @@ tee_tg6(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_tee_tginfo *info = par->targinfo;
 
-	nf_dup_ipv6(skb, par->hooknum, &info->gw.in6, info->priv->oif);
+	nf_dup_ipv6(par->net, skb, par->hooknum, &info->gw.in6, info->priv->oif);
 
 	return XT_CONTINUE;
 }
Signature · 1 line
2.2.1
#54 lore raw

PATCH 10/14 nf_conntrack: Add a struct net parameter to l4_pkt_to_tuple

In reply to Eric W. Biederman

As gre does not have the srckey in the packet gre_pkt_to_tuple needs to perform a lookup in it's per network namespace tables.

Pass in the proper network namespace to all pkt_to_tuple implementations to ensure gre (and any similar protocols) can get this right.

---
 include/net/netfilter/nf_conntrack.h           |  3 ++-
 include/net/netfilter/nf_conntrack_core.h      |  1 +
 include/net/netfilter/nf_conntrack_l4proto.h   |  2 +-
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   |  4 ++--
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c |  3 ++-
 net/netfilter/nf_conntrack_core.c              | 10 ++++++----
 net/netfilter/nf_conntrack_proto_dccp.c        |  2 +-
 net/netfilter/nf_conntrack_proto_generic.c     |  2 +-
 net/netfilter/nf_conntrack_proto_gre.c         |  3 +--
 net/netfilter/nf_conntrack_proto_sctp.c        |  2 +-
 net/netfilter/nf_conntrack_proto_tcp.c         |  2 +-
 net/netfilter/nf_conntrack_proto_udp.c         |  1 +
 net/netfilter/nf_conntrack_proto_udplite.c     |  1 +
 net/netfilter/xt_connlimit.c                   |  2 +-
 net/openvswitch/conntrack.c                    |  2 +-
 net/sched/act_connmark.c                       |  2 +-
 16 files changed, 24 insertions(+), 18 deletions(-)
include/net/netfilter/nf_conntrack.h +2 -1
index e8ad46834df8..d642f68a7c73 100644
@@ -191,7 +191,8 @@ int nf_conntrack_hash_check_insert(struct nf_conn *ct);
 bool nf_ct_delete(struct nf_conn *ct, u32 pid, int report);
 
 bool nf_ct_get_tuplepr(const struct sk_buff *skb, unsigned int nhoff,
-		       u_int16_t l3num, struct nf_conntrack_tuple *tuple);
+		       u_int16_t l3num, struct net *net,
+		       struct nf_conntrack_tuple *tuple);
 bool nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse,
 			  const struct nf_conntrack_tuple *orig);
 
include/net/netfilter/nf_conntrack_core.h +1 -0
index c03f9c42b3cd..788ef58a66b9 100644
@@ -41,6 +41,7 @@ void nf_conntrack_cleanup_end(void);
 
 bool nf_ct_get_tuple(const struct sk_buff *skb, unsigned int nhoff,
 		     unsigned int dataoff, u_int16_t l3num, u_int8_t protonum,
+		     struct net *net,
 		     struct nf_conntrack_tuple *tuple,
 		     const struct nf_conntrack_l3proto *l3proto,
 		     const struct nf_conntrack_l4proto *l4proto);
include/net/netfilter/nf_conntrack_l4proto.h +1 -1
index 1f7061313d54..956d8a6ac069 100644
@@ -26,7 +26,7 @@ struct nf_conntrack_l4proto {
 	/* Try to fill in the third arg: dataoff is offset past network protocol
            hdr.  Return true if possible. */
 	bool (*pkt_to_tuple)(const struct sk_buff *skb, unsigned int dataoff,
-			     struct nf_conntrack_tuple *tuple);
+			     struct net *net, struct nf_conntrack_tuple *tuple);
 
 	/* Invert the per-proto part of the tuple: ie. turn xmit into reply.
 	 * Some packets can't be inverted: return 0 in that case.
net/ipv4/netfilter/nf_conntrack_proto_icmp.c +2 -2
index cdde3ec496e9..c567e1b5d799 100644
@@ -30,7 +30,7 @@ static inline struct nf_icmp_net *icmp_pernet(struct net *net)
 }
 
 static bool icmp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			      struct nf_conntrack_tuple *tuple)
+			      struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	const struct icmphdr *hp;
 	struct icmphdr _hdr;
@@ -144,7 +144,7 @@ icmp_error_message(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb,
 	if (!nf_ct_get_tuplepr(skb,
 			       skb_network_offset(skb) + ip_hdrlen(skb)
 						       + sizeof(struct icmphdr),
-			       PF_INET, &origtuple)) {
+			       PF_INET, net, &origtuple)) {
 		pr_debug("icmp_error_message: failed to get tuple\n");
 		return -NF_ACCEPT;
 	}
net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +2 -1
index 0e6fae103d33..d3b797446cea 100644
@@ -36,6 +36,7 @@ static inline struct nf_icmp_net *icmpv6_pernet(struct net *net)
 
 static bool icmpv6_pkt_to_tuple(const struct sk_buff *skb,
 				unsigned int dataoff,
+				struct net *net,
 				struct nf_conntrack_tuple *tuple)
 {
 	const struct icmp6hdr *hp;
@@ -159,7 +160,7 @@ icmpv6_error_message(struct net *net, struct nf_conn *tmpl,
 			       skb_network_offset(skb)
 				+ sizeof(struct ipv6hdr)
 				+ sizeof(struct icmp6hdr),
-			       PF_INET6, &origtuple)) {
+			       PF_INET6, net, &origtuple)) {
 		pr_debug("icmpv6_error: Can't get tuple\n");
 		return -NF_ACCEPT;
 	}
net/netfilter/nf_conntrack_core.c +6 -4
index c09d6c7198f6..09d1d19b2ab9 100644
@@ -168,6 +168,7 @@ nf_ct_get_tuple(const struct sk_buff *skb,
 		unsigned int dataoff,
 		u_int16_t l3num,
 		u_int8_t protonum,
+		struct net *net,
 		struct nf_conntrack_tuple *tuple,
 		const struct nf_conntrack_l3proto *l3proto,
 		const struct nf_conntrack_l4proto *l4proto)
@@ -181,12 +182,13 @@ nf_ct_get_tuple(const struct sk_buff *skb,
 	tuple->dst.protonum = protonum;
 	tuple->dst.dir = IP_CT_DIR_ORIGINAL;
 
-	return l4proto->pkt_to_tuple(skb, dataoff, tuple);
+	return l4proto->pkt_to_tuple(skb, dataoff, net, tuple);
 }
 EXPORT_SYMBOL_GPL(nf_ct_get_tuple);
 
 bool nf_ct_get_tuplepr(const struct sk_buff *skb, unsigned int nhoff,
-		       u_int16_t l3num, struct nf_conntrack_tuple *tuple)
+		       u_int16_t l3num,
+		       struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	struct nf_conntrack_l3proto *l3proto;
 	struct nf_conntrack_l4proto *l4proto;
@@ -205,7 +207,7 @@ bool nf_ct_get_tuplepr(const struct sk_buff *skb, unsigned int nhoff,
 
 	l4proto = __nf_ct_l4proto_find(l3num, protonum);
 
-	ret = nf_ct_get_tuple(skb, nhoff, protoff, l3num, protonum, tuple,
+	ret = nf_ct_get_tuple(skb, nhoff, protoff, l3num, protonum, net, tuple,
 			      l3proto, l4proto);
 
 	rcu_read_unlock();
@@ -1029,7 +1031,7 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl,
 	u32 hash;
 
 	if (!nf_ct_get_tuple(skb, skb_network_offset(skb),
-			     dataoff, l3num, protonum, &tuple, l3proto,
+			     dataoff, l3num, protonum, net, &tuple, l3proto,
 			     l4proto)) {
 		pr_debug("resolve_normal_ct: Can't get tuple\n");
 		return NULL;
net/netfilter/nf_conntrack_proto_dccp.c +1 -1
index 6dd995c7c72b..fce1b1cca32d 100644
@@ -398,7 +398,7 @@ static inline struct dccp_net *dccp_pernet(struct net *net)
 }
 
 static bool dccp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			      struct nf_conntrack_tuple *tuple)
+			      struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	struct dccp_hdr _hdr, *dh;
 
net/netfilter/nf_conntrack_proto_generic.c +1 -1
index 2281be419a74..86dc752e5349 100644
@@ -45,7 +45,7 @@ static inline struct nf_generic_net *generic_pernet(struct net *net)
 
 static bool generic_pkt_to_tuple(const struct sk_buff *skb,
 				 unsigned int dataoff,
-				 struct nf_conntrack_tuple *tuple)
+				 struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	tuple->src.u.all = 0;
 	tuple->dst.u.all = 0;
net/netfilter/nf_conntrack_proto_gre.c +1 -2
index 7648674f29c3..a96451a7af20 100644
@@ -190,9 +190,8 @@ static bool gre_invert_tuple(struct nf_conntrack_tuple *tuple,
 
 /* gre hdr info to tuple */
 static bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			     struct nf_conntrack_tuple *tuple)
+			     struct net *net, struct nf_conntrack_tuple *tuple)
 {
-	struct net *net = dev_net(skb->dev ? skb->dev : skb_dst(skb)->dev);
 	const struct gre_hdr_pptp *pgrehdr;
 	struct gre_hdr_pptp _pgrehdr;
 	__be16 srckey;
net/netfilter/nf_conntrack_proto_sctp.c +1 -1
index 67197731eb68..9578a7c371ef 100644
@@ -156,7 +156,7 @@ static inline struct sctp_net *sctp_pernet(struct net *net)
 }
 
 static bool sctp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			      struct nf_conntrack_tuple *tuple)
+			      struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	const struct sctphdr *hp;
 	struct sctphdr _hdr;
net/netfilter/nf_conntrack_proto_tcp.c +1 -1
index 70383de72054..278f3b9356ef 100644
@@ -277,7 +277,7 @@ static inline struct nf_tcp_net *tcp_pernet(struct net *net)
 }
 
 static bool tcp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			     struct nf_conntrack_tuple *tuple)
+			     struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	const struct tcphdr *hp;
 	struct tcphdr _hdr;
net/netfilter/nf_conntrack_proto_udp.c +1 -0
index 6957281ffee5..478f92f834b6 100644
@@ -38,6 +38,7 @@ static inline struct nf_udp_net *udp_pernet(struct net *net)
 
 static bool udp_pkt_to_tuple(const struct sk_buff *skb,
 			     unsigned int dataoff,
+			     struct net *net,
 			     struct nf_conntrack_tuple *tuple)
 {
 	const struct udphdr *hp;
net/netfilter/nf_conntrack_proto_udplite.c +1 -0
index c5903d1649f9..1ac8ee13a873 100644
@@ -48,6 +48,7 @@ static inline struct udplite_net *udplite_pernet(struct net *net)
 
 static bool udplite_pkt_to_tuple(const struct sk_buff *skb,
 				 unsigned int dataoff,
+				 struct net *net,
 				 struct nf_conntrack_tuple *tuple)
 {
 	const struct udphdr *hp;
net/netfilter/xt_connlimit.c +1 -1
index 213db252e5be..99bbc829868d 100644
@@ -332,7 +332,7 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par)
 		tuple_ptr = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
 		zone = nf_ct_zone(ct);
 	} else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb),
-				    par->family, &tuple)) {
+				      par->family, net, &tuple)) {
 		goto hotdrop;
 	}
 
net/openvswitch/conntrack.c +1 -1
index e8e524ad8a01..aaf5cbd6d9ae 100644
@@ -345,7 +345,7 @@ ovs_ct_expect_find(struct net *net, const struct nf_conntrack_zone *zone,
 {
 	struct nf_conntrack_tuple tuple;
 
-	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), proto, &tuple))
+	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), proto, net, &tuple))
 		return NULL;
 	return __nf_ct_expect_find(net, zone, &tuple);
 }
net/sched/act_connmark.c +1 -1
index 5019a47b9270..bad1f417c9d9 100644
@@ -68,7 +68,7 @@ static int tcf_connmark(struct sk_buff *skb, const struct tc_action *a,
 	}
 
 	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb),
-			       proto, &tuple))
+			       proto, ca->net, &tuple))
 		goto out;
 
 	zone.id = ca->zone;
Signature · 1 line
2.2.1
#55 lore raw

PATCH 11/14 ipvs: Read hooknum from state rather than ops->hooknum

In reply to Eric W. Biederman

From: Eric W Biederman <[email protected]>

This should be more cache efficient as state is more likely to be in core, and the netfilter core will stop passing in ops soon.

---
 net/netfilter/ipvs/ip_vs_core.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
net/netfilter/ipvs/ip_vs_core.c +8 -8
index 38fbc194b9cb..0be5f12cb12d 100644
@@ -1286,7 +1286,7 @@ static unsigned int
 ip_vs_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET);
+	return ip_vs_out(state->hook, skb, AF_INET);
 }
 
 /*
@@ -1297,7 +1297,7 @@ static unsigned int
 ip_vs_local_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		   const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET);
+	return ip_vs_out(state->hook, skb, AF_INET);
 }
 
 #ifdef CONFIG_IP_VS_IPV6
@@ -1311,7 +1311,7 @@ static unsigned int
 ip_vs_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET6);
+	return ip_vs_out(state->hook, skb, AF_INET6);
 }
 
 /*
@@ -1322,7 +1322,7 @@ static unsigned int
 ip_vs_local_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		   const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET6);
+	return ip_vs_out(state->hook, skb, AF_INET6);
 }
 
 #endif
@@ -1779,7 +1779,7 @@ static unsigned int
 ip_vs_remote_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET);
+	return ip_vs_in(state->hook, skb, AF_INET);
 }
 
 /*
@@ -1790,7 +1790,7 @@ static unsigned int
 ip_vs_local_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET);
+	return ip_vs_in(state->hook, skb, AF_INET);
 }
 
 #ifdef CONFIG_IP_VS_IPV6
@@ -1803,7 +1803,7 @@ static unsigned int
 ip_vs_remote_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET6);
+	return ip_vs_in(state->hook, skb, AF_INET6);
 }
 
 /*
@@ -1814,7 +1814,7 @@ static unsigned int
 ip_vs_local_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET6);
+	return ip_vs_in(state->hook, skb, AF_INET6);
 }
 
 #endif
Signature · 1 line
2.2.1
#56 lore raw

PATCH 12/14 netfilter: Pass priv instead of nf_hook_ops to netfilter hooks

In reply to Eric W. Biederman

From: Eric W Biederman <[email protected]>

Only pass the void *priv parameter out of the nf_hook_ops. That is all any of the functions are interested now, and by limiting what is passed it becomes simpler to change implementation details.

---
 include/linux/netfilter.h                      |  2 +-
 include/net/netfilter/br_netfilter.h           |  2 +-
 include/net/netfilter/nf_nat_l3proto.h         | 32 +++++++++++++-------------
 include/net/netfilter/nf_tables.h              |  3 +--
 net/bridge/br_netfilter_hooks.c                | 14 +++++------
 net/bridge/br_netfilter_ipv6.c                 |  2 +-
 net/bridge/netfilter/ebtable_filter.c          |  4 ++--
 net/bridge/netfilter/ebtable_nat.c             |  4 ++--
 net/bridge/netfilter/nf_tables_bridge.c        |  4 ++--
 net/decnet/netfilter/dn_rtmsg.c                |  2 +-
 net/ipv4/netfilter/arptable_filter.c           |  2 +-
 net/ipv4/netfilter/ipt_CLUSTERIP.c             |  2 +-
 net/ipv4/netfilter/ipt_SYNPROXY.c              |  2 +-
 net/ipv4/netfilter/iptable_filter.c            |  2 +-
 net/ipv4/netfilter/iptable_mangle.c            |  2 +-
 net/ipv4/netfilter/iptable_nat.c               | 18 +++++++--------
 net/ipv4/netfilter/iptable_raw.c               |  2 +-
 net/ipv4/netfilter/iptable_security.c          |  2 +-
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c |  8 +++----
 net/ipv4/netfilter/nf_defrag_ipv4.c            |  2 +-
 net/ipv4/netfilter/nf_nat_l3proto_ipv4.c       | 24 +++++++++----------
 net/ipv4/netfilter/nf_tables_arp.c             |  4 ++--
 net/ipv4/netfilter/nf_tables_ipv4.c            |  8 +++----
 net/ipv4/netfilter/nft_chain_nat_ipv4.c        | 20 ++++++++--------
 net/ipv4/netfilter/nft_chain_route_ipv4.c      |  4 ++--
 net/ipv6/netfilter/ip6t_SYNPROXY.c             |  2 +-
 net/ipv6/netfilter/ip6table_filter.c           |  2 +-
 net/ipv6/netfilter/ip6table_mangle.c           |  2 +-
 net/ipv6/netfilter/ip6table_nat.c              | 18 +++++++--------
 net/ipv6/netfilter/ip6table_raw.c              |  2 +-
 net/ipv6/netfilter/ip6table_security.c         |  2 +-
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c |  8 +++----
 net/ipv6/netfilter/nf_defrag_ipv6_hooks.c      |  2 +-
 net/ipv6/netfilter/nf_nat_l3proto_ipv6.c       | 24 +++++++++----------
 net/ipv6/netfilter/nf_tables_ipv6.c            |  8 +++----
 net/ipv6/netfilter/nft_chain_nat_ipv6.c        | 20 ++++++++--------
 net/ipv6/netfilter/nft_chain_route_ipv6.c      |  4 ++--
 net/netfilter/core.c                           |  2 +-
 net/netfilter/ipvs/ip_vs_core.c                | 24 +++++++++----------
 net/netfilter/nf_tables_core.c                 |  4 ++--
 net/netfilter/nf_tables_netdev.c               |  4 ++--
 security/selinux/hooks.c                       | 10 ++++----
 security/smack/smack_netfilter.c               |  4 ++--
 43 files changed, 156 insertions(+), 157 deletions(-)
include/linux/netfilter.h +1 -1
index 0b4d4560f33d..987c74cd523c 100644
@@ -80,7 +80,7 @@ static inline void nf_hook_state_init(struct nf_hook_state *p,
 	p->okfn = okfn;
 }
 
-typedef unsigned int nf_hookfn(const struct nf_hook_ops *ops,
+typedef unsigned int nf_hookfn(void *priv,
 			       struct sk_buff *skb,
 			       const struct nf_hook_state *state);
 
include/net/netfilter/br_netfilter.h +1 -1
index 8fe266504900..c93c75fa41ad 100644
@@ -46,7 +46,7 @@ void br_netfilter_enable(void);
 
 #if IS_ENABLED(CONFIG_IPV6)
 int br_validate_ipv6(struct sk_buff *skb);
-unsigned int br_nf_pre_routing_ipv6(const struct nf_hook_ops *ops,
+unsigned int br_nf_pre_routing_ipv6(void *priv,
 				    struct sk_buff *skb,
 				    const struct nf_hook_state *state);
 #else
include/net/netfilter/nf_nat_l3proto.h +16 -16
index a3127325f624..aef3e5fc9fd9 100644
@@ -43,31 +43,31 @@ int nf_nat_icmp_reply_translation(struct sk_buff *skb, struct nf_conn *ct,
 				  enum ip_conntrack_info ctinfo,
 				  unsigned int hooknum);
 
-unsigned int nf_nat_ipv4_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
+unsigned int nf_nat_ipv4_in(void *priv, struct sk_buff *skb,
 			    const struct nf_hook_state *state,
-			    unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+			    unsigned int (*do_chain)(void *priv,
 						     struct sk_buff *skb,
 						     const struct nf_hook_state *state,
 						     struct nf_conn *ct));
 
-unsigned int nf_nat_ipv4_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
+unsigned int nf_nat_ipv4_out(void *priv, struct sk_buff *skb,
 			     const struct nf_hook_state *state,
-			     unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+			     unsigned int (*do_chain)(void *priv,
 						      struct sk_buff *skb,
 						      const struct nf_hook_state *state,
 						      struct nf_conn *ct));
 
-unsigned int nf_nat_ipv4_local_fn(const struct nf_hook_ops *ops,
+unsigned int nf_nat_ipv4_local_fn(void *priv,
 				  struct sk_buff *skb,
 				  const struct nf_hook_state *state,
-				  unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+				  unsigned int (*do_chain)(void *priv,
 							   struct sk_buff *skb,
 							   const struct nf_hook_state *state,
 							   struct nf_conn *ct));
 
-unsigned int nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
+unsigned int nf_nat_ipv4_fn(void *priv, struct sk_buff *skb,
 			    const struct nf_hook_state *state,
-			    unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+			    unsigned int (*do_chain)(void *priv,
 						     struct sk_buff *skb,
 						     const struct nf_hook_state *state,
 						     struct nf_conn *ct));
@@ -76,31 +76,31 @@ int nf_nat_icmpv6_reply_translation(struct sk_buff *skb, struct nf_conn *ct,
 				    enum ip_conntrack_info ctinfo,
 				    unsigned int hooknum, unsigned int hdrlen);
 
-unsigned int nf_nat_ipv6_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
+unsigned int nf_nat_ipv6_in(void *priv, struct sk_buff *skb,
 			    const struct nf_hook_state *state,
-			    unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+			    unsigned int (*do_chain)(void *priv,
 						     struct sk_buff *skb,
 						     const struct nf_hook_state *state,
 						     struct nf_conn *ct));
 
-unsigned int nf_nat_ipv6_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
+unsigned int nf_nat_ipv6_out(void *priv, struct sk_buff *skb,
 			     const struct nf_hook_state *state,
-			     unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+			     unsigned int (*do_chain)(void *priv,
 						      struct sk_buff *skb,
 						      const struct nf_hook_state *state,
 						      struct nf_conn *ct));
 
-unsigned int nf_nat_ipv6_local_fn(const struct nf_hook_ops *ops,
+unsigned int nf_nat_ipv6_local_fn(void *priv,
 				  struct sk_buff *skb,
 				  const struct nf_hook_state *state,
-				  unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+				  unsigned int (*do_chain)(void *priv,
 							   struct sk_buff *skb,
 							   const struct nf_hook_state *state,
 							   struct nf_conn *ct));
 
-unsigned int nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
+unsigned int nf_nat_ipv6_fn(void *priv, struct sk_buff *skb,
 			    const struct nf_hook_state *state,
-			    unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+			    unsigned int (*do_chain)(void *priv,
 						     struct sk_buff *skb,
 						     const struct nf_hook_state *state,
 						     struct nf_conn *ct));
include/net/netfilter/nf_tables.h +1 -2
index 42e239e55aa3..c9149cc0a02d 100644
@@ -816,8 +816,7 @@ int nft_register_basechain(struct nft_base_chain *basechain,
 void nft_unregister_basechain(struct nft_base_chain *basechain,
 			      unsigned int hook_nops);
 
-unsigned int nft_do_chain(struct nft_pktinfo *pkt,
-			  const struct nf_hook_ops *ops);
+unsigned int nft_do_chain(struct nft_pktinfo *pkt, void *priv);
 
 /**
  *	struct nft_table - nf_tables table
net/bridge/br_netfilter_hooks.c +7 -7
index e6e76bbdc82f..e21e44c13e07 100644
@@ -464,7 +464,7 @@ struct net_device *setup_pre_routing(struct sk_buff *skb)
  * receiving device) to make netfilter happy, the REDIRECT
  * target in particular.  Save the original destination IP
  * address to be able to detect DNAT afterwards. */
-static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops,
+static unsigned int br_nf_pre_routing(void *priv,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
@@ -486,7 +486,7 @@ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops,
 			return NF_ACCEPT;
 
 		nf_bridge_pull_encap_header_rcsum(skb);
-		return br_nf_pre_routing_ipv6(ops, skb, state);
+		return br_nf_pre_routing_ipv6(priv, skb, state);
 	}
 
 	if (!brnf_call_iptables && !br->nf_call_iptables)
@@ -526,7 +526,7 @@ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops,
  * took place when the packet entered the bridge), but we
  * register an IPv4 PRE_ROUTING 'sabotage' hook that will
  * prevent this from happening. */
-static unsigned int br_nf_local_in(const struct nf_hook_ops *ops,
+static unsigned int br_nf_local_in(void *priv,
 				   struct sk_buff *skb,
 				   const struct nf_hook_state *state)
 {
@@ -570,7 +570,7 @@ static int br_nf_forward_finish(struct net *net, struct sock *sk, struct sk_buff
  * but we are still able to filter on the 'real' indev/outdev
  * because of the physdev module. For ARP, indev and outdev are the
  * bridge ports. */
-static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops,
+static unsigned int br_nf_forward_ip(void *priv,
 				     struct sk_buff *skb,
 				     const struct nf_hook_state *state)
 {
@@ -633,7 +633,7 @@ static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops,
 	return NF_STOLEN;
 }
 
-static unsigned int br_nf_forward_arp(const struct nf_hook_ops *ops,
+static unsigned int br_nf_forward_arp(void *priv,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
@@ -801,7 +801,7 @@ static int br_nf_dev_queue_xmit(struct net *net, struct sock *sk, struct sk_buff
 }
 
 /* PF_BRIDGE/POST_ROUTING ********************************************/
-static unsigned int br_nf_post_routing(const struct nf_hook_ops *ops,
+static unsigned int br_nf_post_routing(void *priv,
 				       struct sk_buff *skb,
 				       const struct nf_hook_state *state)
 {
@@ -850,7 +850,7 @@ static unsigned int br_nf_post_routing(const struct nf_hook_ops *ops,
 /* IP/SABOTAGE *****************************************************/
 /* Don't hand locally destined packets to PF_INET(6)/PRE_ROUTING
  * for the second time. */
-static unsigned int ip_sabotage_in(const struct nf_hook_ops *ops,
+static unsigned int ip_sabotage_in(void *priv,
 				   struct sk_buff *skb,
 				   const struct nf_hook_state *state)
 {
net/bridge/br_netfilter_ipv6.c +1 -1
index e4dbbe44c724..c51cc3fd50d9 100644
@@ -218,7 +218,7 @@ static int br_nf_pre_routing_finish_ipv6(struct net *net, struct sock *sk, struc
 /* Replicate the checks that IPv6 does on packet reception and pass the packet
  * to ip6tables.
  */
-unsigned int br_nf_pre_routing_ipv6(const struct nf_hook_ops *ops,
+unsigned int br_nf_pre_routing_ipv6(void *priv,
 				    struct sk_buff *skb,
 				    const struct nf_hook_state *state)
 {
net/bridge/netfilter/ebtable_filter.c +2 -2
index 118ce40ac181..f9242dffa65e 100644
@@ -57,14 +57,14 @@ static const struct ebt_table frame_filter = {
 };
 
 static unsigned int
-ebt_in_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ebt_in_hook(void *priv, struct sk_buff *skb,
 	    const struct nf_hook_state *state)
 {
 	return ebt_do_table(skb, state, state->net->xt.frame_filter);
 }
 
 static unsigned int
-ebt_out_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ebt_out_hook(void *priv, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
 	return ebt_do_table(skb, state, state->net->xt.frame_filter);
net/bridge/netfilter/ebtable_nat.c +2 -2
index 56c3329d6c37..4bbefe03ab58 100644
@@ -57,14 +57,14 @@ static struct ebt_table frame_nat = {
 };
 
 static unsigned int
-ebt_nat_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ebt_nat_in(void *priv, struct sk_buff *skb,
 	   const struct nf_hook_state *state)
 {
 	return ebt_do_table(skb, state, state->net->xt.frame_nat);
 }
 
 static unsigned int
-ebt_nat_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ebt_nat_out(void *priv, struct sk_buff *skb,
 	    const struct nf_hook_state *state)
 {
 	return ebt_do_table(skb, state, state->net->xt.frame_nat);
net/bridge/netfilter/nf_tables_bridge.c +2 -2
index 318d825e4207..62f6b1b19589 100644
@@ -87,7 +87,7 @@ static inline void nft_bridge_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
 }
 
 static unsigned int
-nft_do_chain_bridge(const struct nf_hook_ops *ops,
+nft_do_chain_bridge(void *priv,
 		    struct sk_buff *skb,
 		    const struct nf_hook_state *state)
 {
@@ -105,7 +105,7 @@ nft_do_chain_bridge(const struct nf_hook_ops *ops,
 		break;
 	}
 
-	return nft_do_chain(&pkt, ops);
+	return nft_do_chain(&pkt, priv);
 }
 
 static struct nft_af_info nft_af_bridge __read_mostly = {
net/decnet/netfilter/dn_rtmsg.c +1 -1
index af34fc9bdf69..85f2fdc360c2 100644
@@ -87,7 +87,7 @@ static void dnrmg_send_peer(struct sk_buff *skb)
 }
 
 
-static unsigned int dnrmg_hook(const struct nf_hook_ops *ops,
+static unsigned int dnrmg_hook(void *priv,
 			struct sk_buff *skb,
 			const struct nf_hook_state *state)
 {
net/ipv4/netfilter/arptable_filter.c +1 -1
index 1352e12d4068..1897ee160920 100644
@@ -27,7 +27,7 @@ static const struct xt_table packet_filter = {
 
 /* The work comes in here from netfilter.c */
 static unsigned int
-arptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
+arptable_filter_hook(void *priv, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
 	return arpt_do_table(skb, state, state->net->ipv4.arptable_filter);
net/ipv4/netfilter/ipt_CLUSTERIP.c +1 -1
index 69157d8eba95..3f32c03e8b2e 100644
@@ -507,7 +507,7 @@ static void arp_print(struct arp_payload *payload)
 #endif
 
 static unsigned int
-arp_mangle(const struct nf_hook_ops *ops,
+arp_mangle(void *priv,
 	   struct sk_buff *skb,
 	   const struct nf_hook_state *state)
 {
net/ipv4/netfilter/ipt_SYNPROXY.c +1 -1
index dfab314981e9..d7021f28c3f0 100644
@@ -299,7 +299,7 @@ synproxy_tg4(struct sk_buff *skb, const struct xt_action_param *par)
 	return XT_CONTINUE;
 }
 
-static unsigned int ipv4_synproxy_hook(const struct nf_hook_ops *ops,
+static unsigned int ipv4_synproxy_hook(void *priv,
 				       struct sk_buff *skb,
 				       const struct nf_hook_state *nhs)
 {
net/ipv4/netfilter/iptable_filter.c +1 -1
index 02d4c5395d6e..397ef2dd133e 100644
@@ -33,7 +33,7 @@ static const struct xt_table packet_filter = {
 };
 
 static unsigned int
-iptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
+iptable_filter_hook(void *priv, struct sk_buff *skb,
 		    const struct nf_hook_state *state)
 {
 	if (state->hook == NF_INET_LOCAL_OUT &&
net/ipv4/netfilter/iptable_mangle.c +1 -1
index dc2ff6884999..2d6fc911866f 100644
@@ -78,7 +78,7 @@ ipt_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state)
 
 /* The work comes in here from netfilter.c. */
 static unsigned int
-iptable_mangle_hook(const struct nf_hook_ops *ops,
+iptable_mangle_hook(void *priv,
 		     struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
net/ipv4/netfilter/iptable_nat.c +9 -9
index 8ff63ac1f0d6..3a2e4d830a0b 100644
@@ -28,7 +28,7 @@ static const struct xt_table nf_nat_ipv4_table = {
 	.af		= NFPROTO_IPV4,
 };
 
-static unsigned int iptable_nat_do_chain(const struct nf_hook_ops *ops,
+static unsigned int iptable_nat_do_chain(void *priv,
 					 struct sk_buff *skb,
 					 const struct nf_hook_state *state,
 					 struct nf_conn *ct)
@@ -36,32 +36,32 @@ static unsigned int iptable_nat_do_chain(const struct nf_hook_ops *ops,
 	return ipt_do_table(skb, state, state->net->ipv4.nat_table);
 }
 
-static unsigned int iptable_nat_ipv4_fn(const struct nf_hook_ops *ops,
+static unsigned int iptable_nat_ipv4_fn(void *priv,
 					struct sk_buff *skb,
 					const struct nf_hook_state *state)
 {
-	return nf_nat_ipv4_fn(ops, skb, state, iptable_nat_do_chain);
+	return nf_nat_ipv4_fn(priv, skb, state, iptable_nat_do_chain);
 }
 
-static unsigned int iptable_nat_ipv4_in(const struct nf_hook_ops *ops,
+static unsigned int iptable_nat_ipv4_in(void *priv,
 					struct sk_buff *skb,
 					const struct nf_hook_state *state)
 {
-	return nf_nat_ipv4_in(ops, skb, state, iptable_nat_do_chain);
+	return nf_nat_ipv4_in(priv, skb, state, iptable_nat_do_chain);
 }
 
-static unsigned int iptable_nat_ipv4_out(const struct nf_hook_ops *ops,
+static unsigned int iptable_nat_ipv4_out(void *priv,
 					 struct sk_buff *skb,
 					 const struct nf_hook_state *state)
 {
-	return nf_nat_ipv4_out(ops, skb, state, iptable_nat_do_chain);
+	return nf_nat_ipv4_out(priv, skb, state, iptable_nat_do_chain);
 }
 
-static unsigned int iptable_nat_ipv4_local_fn(const struct nf_hook_ops *ops,
+static unsigned int iptable_nat_ipv4_local_fn(void *priv,
 					      struct sk_buff *skb,
 					      const struct nf_hook_state *state)
 {
-	return nf_nat_ipv4_local_fn(ops, skb, state, iptable_nat_do_chain);
+	return nf_nat_ipv4_local_fn(priv, skb, state, iptable_nat_do_chain);
 }
 
 static struct nf_hook_ops nf_nat_ipv4_ops[] __read_mostly = {
net/ipv4/netfilter/iptable_raw.c +1 -1
index bbb0523d87de..1ba02811acb0 100644
@@ -20,7 +20,7 @@ static const struct xt_table packet_raw = {
 
 /* The work comes in here from netfilter.c. */
 static unsigned int
-iptable_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
+iptable_raw_hook(void *priv, struct sk_buff *skb,
 		 const struct nf_hook_state *state)
 {
 	if (state->hook == NF_INET_LOCAL_OUT &&
net/ipv4/netfilter/iptable_security.c +1 -1
index b92417038705..f534e2f05bad 100644
@@ -37,7 +37,7 @@ static const struct xt_table security_table = {
 };
 
 static unsigned int
-iptable_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
+iptable_security_hook(void *priv, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
 	if (state->hook == NF_INET_LOCAL_OUT &&
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +4 -4
index 15749cc5cf2b..752fb40adcf8 100644
@@ -92,7 +92,7 @@ static int ipv4_get_l4proto(const struct sk_buff *skb, unsigned int nhoff,
 	return NF_ACCEPT;
 }
 
-static unsigned int ipv4_helper(const struct nf_hook_ops *ops,
+static unsigned int ipv4_helper(void *priv,
 				struct sk_buff *skb,
 				const struct nf_hook_state *state)
 {
@@ -119,7 +119,7 @@ static unsigned int ipv4_helper(const struct nf_hook_ops *ops,
 			    ct, ctinfo);
 }
 
-static unsigned int ipv4_confirm(const struct nf_hook_ops *ops,
+static unsigned int ipv4_confirm(void *priv,
 				 struct sk_buff *skb,
 				 const struct nf_hook_state *state)
 {
@@ -143,14 +143,14 @@ out:
 	return nf_conntrack_confirm(skb);
 }
 
-static unsigned int ipv4_conntrack_in(const struct nf_hook_ops *ops,
+static unsigned int ipv4_conntrack_in(void *priv,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
 	return nf_conntrack_in(state->net, PF_INET, state->hook, skb);
 }
 
-static unsigned int ipv4_conntrack_local(const struct nf_hook_ops *ops,
+static unsigned int ipv4_conntrack_local(void *priv,
 					 struct sk_buff *skb,
 					 const struct nf_hook_state *state)
 {
net/ipv4/netfilter/nf_defrag_ipv4.c +1 -1
index 8aea536d2e83..b246346ee849 100644
@@ -61,7 +61,7 @@ static enum ip_defrag_users nf_ct_defrag_user(unsigned int hooknum,
 		return IP_DEFRAG_CONNTRACK_OUT + zone_id;
 }
 
-static unsigned int ipv4_conntrack_defrag(const struct nf_hook_ops *ops,
+static unsigned int ipv4_conntrack_defrag(void *priv,
 					  struct sk_buff *skb,
 					  const struct nf_hook_state *state)
 {
net/ipv4/netfilter/nf_nat_l3proto_ipv4.c +12 -12
index 16da45a76dac..8593a9d88619 100644
@@ -255,9 +255,9 @@ int nf_nat_icmp_reply_translation(struct sk_buff *skb,
 EXPORT_SYMBOL_GPL(nf_nat_icmp_reply_translation);
 
 unsigned int
-nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
+nf_nat_ipv4_fn(void *priv, struct sk_buff *skb,
 	       const struct nf_hook_state *state,
-	       unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+	       unsigned int (*do_chain)(void *priv,
 					struct sk_buff *skb,
 					const struct nf_hook_state *state,
 					struct nf_conn *ct))
@@ -308,7 +308,7 @@ nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		if (!nf_nat_initialized(ct, maniptype)) {
 			unsigned int ret;
 
-			ret = do_chain(ops, skb, state, ct);
+			ret = do_chain(priv, skb, state, ct);
 			if (ret != NF_ACCEPT)
 				return ret;
 
@@ -345,9 +345,9 @@ oif_changed:
 EXPORT_SYMBOL_GPL(nf_nat_ipv4_fn);
 
 unsigned int
-nf_nat_ipv4_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
+nf_nat_ipv4_in(void *priv, struct sk_buff *skb,
 	       const struct nf_hook_state *state,
-	       unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+	       unsigned int (*do_chain)(void *priv,
 					 struct sk_buff *skb,
 					 const struct nf_hook_state *state,
 					 struct nf_conn *ct))
@@ -355,7 +355,7 @@ nf_nat_ipv4_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	unsigned int ret;
 	__be32 daddr = ip_hdr(skb)->daddr;
 
-	ret = nf_nat_ipv4_fn(ops, skb, state, do_chain);
+	ret = nf_nat_ipv4_fn(priv, skb, state, do_chain);
 	if (ret != NF_DROP && ret != NF_STOLEN &&
 	    daddr != ip_hdr(skb)->daddr)
 		skb_dst_drop(skb);
@@ -365,9 +365,9 @@ nf_nat_ipv4_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
 EXPORT_SYMBOL_GPL(nf_nat_ipv4_in);
 
 unsigned int
-nf_nat_ipv4_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
+nf_nat_ipv4_out(void *priv, struct sk_buff *skb,
 		const struct nf_hook_state *state,
-		unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+		unsigned int (*do_chain)(void *priv,
 					  struct sk_buff *skb,
 					  const struct nf_hook_state *state,
 					  struct nf_conn *ct))
@@ -384,7 +384,7 @@ nf_nat_ipv4_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	    ip_hdrlen(skb) < sizeof(struct iphdr))
 		return NF_ACCEPT;
 
-	ret = nf_nat_ipv4_fn(ops, skb, state, do_chain);
+	ret = nf_nat_ipv4_fn(priv, skb, state, do_chain);
 #ifdef CONFIG_XFRM
 	if (ret != NF_DROP && ret != NF_STOLEN &&
 	    !(IPCB(skb)->flags & IPSKB_XFRM_TRANSFORMED) &&
@@ -407,9 +407,9 @@ nf_nat_ipv4_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
 EXPORT_SYMBOL_GPL(nf_nat_ipv4_out);
 
 unsigned int
-nf_nat_ipv4_local_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
+nf_nat_ipv4_local_fn(void *priv, struct sk_buff *skb,
 		     const struct nf_hook_state *state,
-		     unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+		     unsigned int (*do_chain)(void *priv,
 					       struct sk_buff *skb,
 					       const struct nf_hook_state *state,
 					       struct nf_conn *ct))
@@ -424,7 +424,7 @@ nf_nat_ipv4_local_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	    ip_hdrlen(skb) < sizeof(struct iphdr))
 		return NF_ACCEPT;
 
-	ret = nf_nat_ipv4_fn(ops, skb, state, do_chain);
+	ret = nf_nat_ipv4_fn(priv, skb, state, do_chain);
 	if (ret != NF_DROP && ret != NF_STOLEN &&
 	    (ct = nf_ct_get(skb, &ctinfo)) != NULL) {
 		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
net/ipv4/netfilter/nf_tables_arp.c +2 -2
index 883bbf83fe09..9d09d4f59545 100644
@@ -15,7 +15,7 @@
 #include <net/netfilter/nf_tables.h>
 
 static unsigned int
-nft_do_chain_arp(const struct nf_hook_ops *ops,
+nft_do_chain_arp(void *priv,
 		  struct sk_buff *skb,
 		  const struct nf_hook_state *state)
 {
@@ -23,7 +23,7 @@ nft_do_chain_arp(const struct nf_hook_ops *ops,
 
 	nft_set_pktinfo(&pkt, skb, state);
 
-	return nft_do_chain(&pkt, ops);
+	return nft_do_chain(&pkt, priv);
 }
 
 static struct nft_af_info nft_af_arp __read_mostly = {
net/ipv4/netfilter/nf_tables_ipv4.c +4 -4
index 805be5c9fcc3..ca9dc3c46c4f 100644
@@ -18,7 +18,7 @@
 #include <net/ip.h>
 #include <net/netfilter/nf_tables_ipv4.h>
 
-static unsigned int nft_do_chain_ipv4(const struct nf_hook_ops *ops,
+static unsigned int nft_do_chain_ipv4(void *priv,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
@@ -26,10 +26,10 @@ static unsigned int nft_do_chain_ipv4(const struct nf_hook_ops *ops,
 
 	nft_set_pktinfo_ipv4(&pkt, skb, state);
 
-	return nft_do_chain(&pkt, ops);
+	return nft_do_chain(&pkt, priv);
 }
 
-static unsigned int nft_ipv4_output(const struct nf_hook_ops *ops,
+static unsigned int nft_ipv4_output(void *priv,
 				    struct sk_buff *skb,
 				    const struct nf_hook_state *state)
 {
@@ -41,7 +41,7 @@ static unsigned int nft_ipv4_output(const struct nf_hook_ops *ops,
 		return NF_ACCEPT;
 	}
 
-	return nft_do_chain_ipv4(ops, skb, state);
+	return nft_do_chain_ipv4(priv, skb, state);
 }
 
 struct nft_af_info nft_af_ipv4 __read_mostly = {
net/ipv4/netfilter/nft_chain_nat_ipv4.c +10 -10
index c3ffecf28d38..f5c66a7a4bf2 100644
@@ -26,7 +26,7 @@
 #include <net/netfilter/nf_nat_l3proto.h>
 #include <net/ip.h>
 
-static unsigned int nft_nat_do_chain(const struct nf_hook_ops *ops,
+static unsigned int nft_nat_do_chain(void *priv,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state,
 				      struct nf_conn *ct)
@@ -35,35 +35,35 @@ static unsigned int nft_nat_do_chain(const struct nf_hook_ops *ops,
 
 	nft_set_pktinfo_ipv4(&pkt, skb, state);
 
-	return nft_do_chain(&pkt, ops);
+	return nft_do_chain(&pkt, priv);
 }
 
-static unsigned int nft_nat_ipv4_fn(const struct nf_hook_ops *ops,
+static unsigned int nft_nat_ipv4_fn(void *priv,
 				    struct sk_buff *skb,
 				    const struct nf_hook_state *state)
 {
-	return nf_nat_ipv4_fn(ops, skb, state, nft_nat_do_chain);
+	return nf_nat_ipv4_fn(priv, skb, state, nft_nat_do_chain);
 }
 
-static unsigned int nft_nat_ipv4_in(const struct nf_hook_ops *ops,
+static unsigned int nft_nat_ipv4_in(void *priv,
 				    struct sk_buff *skb,
 				    const struct nf_hook_state *state)
 {
-	return nf_nat_ipv4_in(ops, skb, state, nft_nat_do_chain);
+	return nf_nat_ipv4_in(priv, skb, state, nft_nat_do_chain);
 }
 
-static unsigned int nft_nat_ipv4_out(const struct nf_hook_ops *ops,
+static unsigned int nft_nat_ipv4_out(void *priv,
 				     struct sk_buff *skb,
 				     const struct nf_hook_state *state)
 {
-	return nf_nat_ipv4_out(ops, skb, state, nft_nat_do_chain);
+	return nf_nat_ipv4_out(priv, skb, state, nft_nat_do_chain);
 }
 
-static unsigned int nft_nat_ipv4_local_fn(const struct nf_hook_ops *ops,
+static unsigned int nft_nat_ipv4_local_fn(void *priv,
 					  struct sk_buff *skb,
 					  const struct nf_hook_state *state)
 {
-	return nf_nat_ipv4_local_fn(ops, skb, state, nft_nat_do_chain);
+	return nf_nat_ipv4_local_fn(priv, skb, state, nft_nat_do_chain);
 }
 
 static const struct nf_chain_type nft_chain_nat_ipv4 = {
net/ipv4/netfilter/nft_chain_route_ipv4.c +2 -2
index 2a1e3d8a3e43..9f486b302108 100644
@@ -21,7 +21,7 @@
 #include <net/route.h>
 #include <net/ip.h>
 
-static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops,
+static unsigned int nf_route_table_hook(void *priv,
 					struct sk_buff *skb,
 					const struct nf_hook_state *state)
 {
@@ -45,7 +45,7 @@ static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops,
 	daddr = iph->daddr;
 	tos = iph->tos;
 
-	ret = nft_do_chain(&pkt, ops);
+	ret = nft_do_chain(&pkt, priv);
 	if (ret != NF_DROP && ret != NF_QUEUE) {
 		iph = ip_hdr(skb);
 
net/ipv6/netfilter/ip6t_SYNPROXY.c +1 -1
index 41451809b37c..c2356602158a 100644
@@ -316,7 +316,7 @@ synproxy_tg6(struct sk_buff *skb, const struct xt_action_param *par)
 	return XT_CONTINUE;
 }
 
-static unsigned int ipv6_synproxy_hook(const struct nf_hook_ops *ops,
+static unsigned int ipv6_synproxy_hook(void *priv,
 				       struct sk_buff *skb,
 				       const struct nf_hook_state *nhs)
 {
net/ipv6/netfilter/ip6table_filter.c +1 -1
index a7327f61b90c..8b277b983ca5 100644
@@ -32,7 +32,7 @@ static const struct xt_table packet_filter = {
 
 /* The work comes in here from netfilter.c. */
 static unsigned int
-ip6table_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip6table_filter_hook(void *priv, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
 	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_filter);
net/ipv6/netfilter/ip6table_mangle.c +1 -1
index c2e061dcedf3..8745b592b2f6 100644
@@ -75,7 +75,7 @@ ip6t_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state)
 
 /* The work comes in here from netfilter.c. */
 static unsigned int
-ip6table_mangle_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip6table_mangle_hook(void *priv, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
 	if (state->hook == NF_INET_LOCAL_OUT)
net/ipv6/netfilter/ip6table_nat.c +9 -9
index efa6754c4d06..abea175d5853 100644
@@ -30,7 +30,7 @@ static const struct xt_table nf_nat_ipv6_table = {
 	.af		= NFPROTO_IPV6,
 };
 
-static unsigned int ip6table_nat_do_chain(const struct nf_hook_ops *ops,
+static unsigned int ip6table_nat_do_chain(void *priv,
 					  struct sk_buff *skb,
 					  const struct nf_hook_state *state,
 					  struct nf_conn *ct)
@@ -38,32 +38,32 @@ static unsigned int ip6table_nat_do_chain(const struct nf_hook_ops *ops,
 	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_nat);
 }
 
-static unsigned int ip6table_nat_fn(const struct nf_hook_ops *ops,
+static unsigned int ip6table_nat_fn(void *priv,
 				    struct sk_buff *skb,
 				    const struct nf_hook_state *state)
 {
-	return nf_nat_ipv6_fn(ops, skb, state, ip6table_nat_do_chain);
+	return nf_nat_ipv6_fn(priv, skb, state, ip6table_nat_do_chain);
 }
 
-static unsigned int ip6table_nat_in(const struct nf_hook_ops *ops,
+static unsigned int ip6table_nat_in(void *priv,
 				    struct sk_buff *skb,
 				    const struct nf_hook_state *state)
 {
-	return nf_nat_ipv6_in(ops, skb, state, ip6table_nat_do_chain);
+	return nf_nat_ipv6_in(priv, skb, state, ip6table_nat_do_chain);
 }
 
-static unsigned int ip6table_nat_out(const struct nf_hook_ops *ops,
+static unsigned int ip6table_nat_out(void *priv,
 				     struct sk_buff *skb,
 				     const struct nf_hook_state *state)
 {
-	return nf_nat_ipv6_out(ops, skb, state, ip6table_nat_do_chain);
+	return nf_nat_ipv6_out(priv, skb, state, ip6table_nat_do_chain);
 }
 
-static unsigned int ip6table_nat_local_fn(const struct nf_hook_ops *ops,
+static unsigned int ip6table_nat_local_fn(void *priv,
 					  struct sk_buff *skb,
 					  const struct nf_hook_state *state)
 {
-	return nf_nat_ipv6_local_fn(ops, skb, state, ip6table_nat_do_chain);
+	return nf_nat_ipv6_local_fn(priv, skb, state, ip6table_nat_do_chain);
 }
 
 static struct nf_hook_ops nf_nat_ipv6_ops[] __read_mostly = {
net/ipv6/netfilter/ip6table_raw.c +1 -1
index fac6ad7c0a7c..9021963565c3 100644
@@ -19,7 +19,7 @@ static const struct xt_table packet_raw = {
 
 /* The work comes in here from netfilter.c. */
 static unsigned int
-ip6table_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip6table_raw_hook(void *priv, struct sk_buff *skb,
 		  const struct nf_hook_state *state)
 {
 	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_raw);
net/ipv6/netfilter/ip6table_security.c +1 -1
index 96c94fc240c8..0d856fedfeb0 100644
@@ -36,7 +36,7 @@ static const struct xt_table security_table = {
 };
 
 static unsigned int
-ip6table_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip6table_security_hook(void *priv, struct sk_buff *skb,
 		       const struct nf_hook_state *state)
 {
 	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_security);
net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +4 -4
index 339be1d59afc..dd83ad42f8f6 100644
@@ -95,7 +95,7 @@ static int ipv6_get_l4proto(const struct sk_buff *skb, unsigned int nhoff,
 	return NF_ACCEPT;
 }
 
-static unsigned int ipv6_helper(const struct nf_hook_ops *ops,
+static unsigned int ipv6_helper(void *priv,
 				struct sk_buff *skb,
 				const struct nf_hook_state *state)
 {
@@ -131,7 +131,7 @@ static unsigned int ipv6_helper(const struct nf_hook_ops *ops,
 	return helper->help(skb, protoff, ct, ctinfo);
 }
 
-static unsigned int ipv6_confirm(const struct nf_hook_ops *ops,
+static unsigned int ipv6_confirm(void *priv,
 				 struct sk_buff *skb,
 				 const struct nf_hook_state *state)
 {
@@ -165,14 +165,14 @@ out:
 	return nf_conntrack_confirm(skb);
 }
 
-static unsigned int ipv6_conntrack_in(const struct nf_hook_ops *ops,
+static unsigned int ipv6_conntrack_in(void *priv,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
 	return nf_conntrack_in(state->net, PF_INET6, state->hook, skb);
 }
 
-static unsigned int ipv6_conntrack_local(const struct nf_hook_ops *ops,
+static unsigned int ipv6_conntrack_local(void *priv,
 					 struct sk_buff *skb,
 					 const struct nf_hook_state *state)
 {
net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +1 -1
index a9c08520596b..a99baf63eccf 100644
@@ -51,7 +51,7 @@ static enum ip6_defrag_users nf_ct6_defrag_user(unsigned int hooknum,
 		return IP6_DEFRAG_CONNTRACK_OUT + zone_id;
 }
 
-static unsigned int ipv6_defrag(const struct nf_hook_ops *ops,
+static unsigned int ipv6_defrag(void *priv,
 				struct sk_buff *skb,
 				const struct nf_hook_state *state)
 {
net/ipv6/netfilter/nf_nat_l3proto_ipv6.c +12 -12
index 8bc94907dbd9..357f57ba47e4 100644
@@ -262,9 +262,9 @@ int nf_nat_icmpv6_reply_translation(struct sk_buff *skb,
 EXPORT_SYMBOL_GPL(nf_nat_icmpv6_reply_translation);
 
 unsigned int
-nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
+nf_nat_ipv6_fn(void *priv, struct sk_buff *skb,
 	       const struct nf_hook_state *state,
-	       unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+	       unsigned int (*do_chain)(void *priv,
 					struct sk_buff *skb,
 					const struct nf_hook_state *state,
 					struct nf_conn *ct))
@@ -317,7 +317,7 @@ nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		if (!nf_nat_initialized(ct, maniptype)) {
 			unsigned int ret;
 
-			ret = do_chain(ops, skb, state, ct);
+			ret = do_chain(priv, skb, state, ct);
 			if (ret != NF_ACCEPT)
 				return ret;
 
@@ -353,9 +353,9 @@ oif_changed:
 EXPORT_SYMBOL_GPL(nf_nat_ipv6_fn);
 
 unsigned int
-nf_nat_ipv6_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
+nf_nat_ipv6_in(void *priv, struct sk_buff *skb,
 	       const struct nf_hook_state *state,
-	       unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+	       unsigned int (*do_chain)(void *priv,
 					struct sk_buff *skb,
 					const struct nf_hook_state *state,
 					struct nf_conn *ct))
@@ -363,7 +363,7 @@ nf_nat_ipv6_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	unsigned int ret;
 	struct in6_addr daddr = ipv6_hdr(skb)->daddr;
 
-	ret = nf_nat_ipv6_fn(ops, skb, state, do_chain);
+	ret = nf_nat_ipv6_fn(priv, skb, state, do_chain);
 	if (ret != NF_DROP && ret != NF_STOLEN &&
 	    ipv6_addr_cmp(&daddr, &ipv6_hdr(skb)->daddr))
 		skb_dst_drop(skb);
@@ -373,9 +373,9 @@ nf_nat_ipv6_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
 EXPORT_SYMBOL_GPL(nf_nat_ipv6_in);
 
 unsigned int
-nf_nat_ipv6_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
+nf_nat_ipv6_out(void *priv, struct sk_buff *skb,
 		const struct nf_hook_state *state,
-		unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+		unsigned int (*do_chain)(void *priv,
 					 struct sk_buff *skb,
 					 const struct nf_hook_state *state,
 					 struct nf_conn *ct))
@@ -391,7 +391,7 @@ nf_nat_ipv6_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	if (skb->len < sizeof(struct ipv6hdr))
 		return NF_ACCEPT;
 
-	ret = nf_nat_ipv6_fn(ops, skb, state, do_chain);
+	ret = nf_nat_ipv6_fn(priv, skb, state, do_chain);
 #ifdef CONFIG_XFRM
 	if (ret != NF_DROP && ret != NF_STOLEN &&
 	    !(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) &&
@@ -414,9 +414,9 @@ nf_nat_ipv6_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
 EXPORT_SYMBOL_GPL(nf_nat_ipv6_out);
 
 unsigned int
-nf_nat_ipv6_local_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
+nf_nat_ipv6_local_fn(void *priv, struct sk_buff *skb,
 		     const struct nf_hook_state *state,
-		     unsigned int (*do_chain)(const struct nf_hook_ops *ops,
+		     unsigned int (*do_chain)(void *priv,
 					      struct sk_buff *skb,
 					      const struct nf_hook_state *state,
 					      struct nf_conn *ct))
@@ -430,7 +430,7 @@ nf_nat_ipv6_local_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	if (skb->len < sizeof(struct ipv6hdr))
 		return NF_ACCEPT;
 
-	ret = nf_nat_ipv6_fn(ops, skb, state, do_chain);
+	ret = nf_nat_ipv6_fn(priv, skb, state, do_chain);
 	if (ret != NF_DROP && ret != NF_STOLEN &&
 	    (ct = nf_ct_get(skb, &ctinfo)) != NULL) {
 		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
net/ipv6/netfilter/nf_tables_ipv6.c +4 -4
index 41340b794f9b..120ea9131be0 100644
@@ -16,7 +16,7 @@
 #include <net/netfilter/nf_tables.h>
 #include <net/netfilter/nf_tables_ipv6.h>
 
-static unsigned int nft_do_chain_ipv6(const struct nf_hook_ops *ops,
+static unsigned int nft_do_chain_ipv6(void *priv,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
@@ -26,10 +26,10 @@ static unsigned int nft_do_chain_ipv6(const struct nf_hook_ops *ops,
 	if (nft_set_pktinfo_ipv6(&pkt, skb, state) < 0)
 		return NF_DROP;
 
-	return nft_do_chain(&pkt, ops);
+	return nft_do_chain(&pkt, priv);
 }
 
-static unsigned int nft_ipv6_output(const struct nf_hook_ops *ops,
+static unsigned int nft_ipv6_output(void *priv,
 				    struct sk_buff *skb,
 				    const struct nf_hook_state *state)
 {
@@ -40,7 +40,7 @@ static unsigned int nft_ipv6_output(const struct nf_hook_ops *ops,
 		return NF_ACCEPT;
 	}
 
-	return nft_do_chain_ipv6(ops, skb, state);
+	return nft_do_chain_ipv6(priv, skb, state);
 }
 
 struct nft_af_info nft_af_ipv6 __read_mostly = {
net/ipv6/netfilter/nft_chain_nat_ipv6.c +10 -10
index e96feaefeb14..443cd306c0b0 100644
@@ -24,7 +24,7 @@
 #include <net/netfilter/nf_nat_l3proto.h>
 #include <net/ipv6.h>
 
-static unsigned int nft_nat_do_chain(const struct nf_hook_ops *ops,
+static unsigned int nft_nat_do_chain(void *priv,
 				     struct sk_buff *skb,
 				     const struct nf_hook_state *state,
 				     struct nf_conn *ct)
@@ -33,35 +33,35 @@ static unsigned int nft_nat_do_chain(const struct nf_hook_ops *ops,
 
 	nft_set_pktinfo_ipv6(&pkt, skb, state);
 
-	return nft_do_chain(&pkt, ops);
+	return nft_do_chain(&pkt, priv);
 }
 
-static unsigned int nft_nat_ipv6_fn(const struct nf_hook_ops *ops,
+static unsigned int nft_nat_ipv6_fn(void *priv,
 				    struct sk_buff *skb,
 				    const struct nf_hook_state *state)
 {
-	return nf_nat_ipv6_fn(ops, skb, state, nft_nat_do_chain);
+	return nf_nat_ipv6_fn(priv, skb, state, nft_nat_do_chain);
 }
 
-static unsigned int nft_nat_ipv6_in(const struct nf_hook_ops *ops,
+static unsigned int nft_nat_ipv6_in(void *priv,
 				    struct sk_buff *skb,
 				    const struct nf_hook_state *state)
 {
-	return nf_nat_ipv6_in(ops, skb, state, nft_nat_do_chain);
+	return nf_nat_ipv6_in(priv, skb, state, nft_nat_do_chain);
 }
 
-static unsigned int nft_nat_ipv6_out(const struct nf_hook_ops *ops,
+static unsigned int nft_nat_ipv6_out(void *priv,
 				     struct sk_buff *skb,
 				     const struct nf_hook_state *state)
 {
-	return nf_nat_ipv6_out(ops, skb, state, nft_nat_do_chain);
+	return nf_nat_ipv6_out(priv, skb, state, nft_nat_do_chain);
 }
 
-static unsigned int nft_nat_ipv6_local_fn(const struct nf_hook_ops *ops,
+static unsigned int nft_nat_ipv6_local_fn(void *priv,
 					  struct sk_buff *skb,
 					  const struct nf_hook_state *state)
 {
-	return nf_nat_ipv6_local_fn(ops, skb, state, nft_nat_do_chain);
+	return nf_nat_ipv6_local_fn(priv, skb, state, nft_nat_do_chain);
 }
 
 static const struct nf_chain_type nft_chain_nat_ipv6 = {
net/ipv6/netfilter/nft_chain_route_ipv6.c +2 -2
index d1bcd2ed7bcc..d42bbc1d7555 100644
@@ -22,7 +22,7 @@
 #include <net/netfilter/nf_tables_ipv6.h>
 #include <net/route.h>
 
-static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops,
+static unsigned int nf_route_table_hook(void *priv,
 					struct sk_buff *skb,
 					const struct nf_hook_state *state)
 {
@@ -45,7 +45,7 @@ static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops,
 	/* flowlabel and prio (includes version, which shouldn't change either */
 	flowlabel = *((u32 *)ipv6_hdr(skb));
 
-	ret = nft_do_chain(&pkt, ops);
+	ret = nft_do_chain(&pkt, priv);
 	if (ret != NF_DROP && ret != NF_QUEUE &&
 	    (memcmp(&ipv6_hdr(skb)->saddr, &saddr, sizeof(saddr)) ||
 	     memcmp(&ipv6_hdr(skb)->daddr, &daddr, sizeof(daddr)) ||
net/netfilter/core.c +1 -1
index 8e47f8113495..2e907335ee81 100644
@@ -269,7 +269,7 @@ unsigned int nf_iterate(struct list_head *head,
 		/* Optimization: we don't need to hold module
 		   reference here, since function can't sleep. --RR */
 repeat:
-		verdict = (*elemp)->hook(*elemp, skb, state);
+		verdict = (*elemp)->hook((*elemp)->priv, skb, state);
 		if (verdict != NF_ACCEPT) {
 #ifdef CONFIG_NETFILTER_DEBUG
 			if (unlikely((verdict & NF_VERDICT_MASK)
net/netfilter/ipvs/ip_vs_core.c +12 -12
index 0be5f12cb12d..59b8a7a807ec 100644
@@ -1283,7 +1283,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af)
  *	Check if packet is reply for established ip_vs_conn.
  */
 static unsigned int
-ip_vs_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_reply4(void *priv, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
 	return ip_vs_out(state->hook, skb, AF_INET);
@@ -1294,7 +1294,7 @@ ip_vs_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Check if packet is reply for established ip_vs_conn.
  */
 static unsigned int
-ip_vs_local_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_local_reply4(void *priv, struct sk_buff *skb,
 		   const struct nf_hook_state *state)
 {
 	return ip_vs_out(state->hook, skb, AF_INET);
@@ -1308,7 +1308,7 @@ ip_vs_local_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Check if packet is reply for established ip_vs_conn.
  */
 static unsigned int
-ip_vs_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_reply6(void *priv, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
 	return ip_vs_out(state->hook, skb, AF_INET6);
@@ -1319,7 +1319,7 @@ ip_vs_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Check if packet is reply for established ip_vs_conn.
  */
 static unsigned int
-ip_vs_local_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_local_reply6(void *priv, struct sk_buff *skb,
 		   const struct nf_hook_state *state)
 {
 	return ip_vs_out(state->hook, skb, AF_INET6);
@@ -1776,7 +1776,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af)
  *	Schedule and forward packets from remote clients
  */
 static unsigned int
-ip_vs_remote_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_remote_request4(void *priv, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
 	return ip_vs_in(state->hook, skb, AF_INET);
@@ -1787,7 +1787,7 @@ ip_vs_remote_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Schedule and forward packets from local clients
  */
 static unsigned int
-ip_vs_local_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_local_request4(void *priv, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
 	return ip_vs_in(state->hook, skb, AF_INET);
@@ -1800,7 +1800,7 @@ ip_vs_local_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Schedule and forward packets from remote clients
  */
 static unsigned int
-ip_vs_remote_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_remote_request6(void *priv, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
 	return ip_vs_in(state->hook, skb, AF_INET6);
@@ -1811,7 +1811,7 @@ ip_vs_remote_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Schedule and forward packets from local clients
  */
 static unsigned int
-ip_vs_local_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_local_request6(void *priv, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
 	return ip_vs_in(state->hook, skb, AF_INET6);
@@ -1830,7 +1830,7 @@ ip_vs_local_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *      and send them to ip_vs_in_icmp.
  */
 static unsigned int
-ip_vs_forward_icmp(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_forward_icmp(void *priv, struct sk_buff *skb,
 		   const struct nf_hook_state *state)
 {
 	int r;
@@ -1846,12 +1846,12 @@ ip_vs_forward_icmp(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	if (unlikely(sysctl_backup_only(ipvs) || !ipvs->enable))
 		return NF_ACCEPT;
 
-	return ip_vs_in_icmp(skb, &r, ops->hooknum);
+	return ip_vs_in_icmp(skb, &r, state->hook);
 }
 
 #ifdef CONFIG_IP_VS_IPV6
 static unsigned int
-ip_vs_forward_icmp_v6(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_forward_icmp_v6(void *priv, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
 	int r;
@@ -1869,7 +1869,7 @@ ip_vs_forward_icmp_v6(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	if (unlikely(sysctl_backup_only(ipvs) || !ipvs->enable))
 		return NF_ACCEPT;
 
-	return ip_vs_in_icmp_v6(skb, &r, ops->hooknum, &iphdr);
+	return ip_vs_in_icmp_v6(skb, &r, state->hook, &iphdr);
 }
 #endif
 
net/netfilter/nf_tables_core.c +2 -2
index e5c1f332e45e..f3695a497408 100644
@@ -109,9 +109,9 @@ struct nft_jumpstack {
 };
 
 unsigned int
-nft_do_chain(struct nft_pktinfo *pkt, const struct nf_hook_ops *ops)
+nft_do_chain(struct nft_pktinfo *pkt, void *priv)
 {
-	const struct nft_chain *chain = ops->priv, *basechain = chain;
+	const struct nft_chain *chain = priv, *basechain = chain;
 	const struct net *net = pkt->net;
 	const struct nft_rule *rule;
 	const struct nft_expr *expr, *last;
net/netfilter/nf_tables_netdev.c +2 -2
index db416a3396e9..7b9c053ba750 100644
@@ -89,7 +89,7 @@ static inline void nft_netdev_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
 }
 
 static unsigned int
-nft_do_chain_netdev(const struct nf_hook_ops *ops, struct sk_buff *skb,
+nft_do_chain_netdev(void *priv, struct sk_buff *skb,
 		    const struct nf_hook_state *state)
 {
 	struct nft_pktinfo pkt;
@@ -106,7 +106,7 @@ nft_do_chain_netdev(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		break;
 	}
 
-	return nft_do_chain(&pkt, ops);
+	return nft_do_chain(&pkt, priv);
 }
 
 static struct nft_af_info nft_af_netdev __read_mostly = {
security/selinux/hooks.c +5 -5
index e4369d86e588..64340160f4ac 100644
@@ -4866,7 +4866,7 @@ static unsigned int selinux_ip_forward(struct sk_buff *skb,
 	return NF_ACCEPT;
 }
 
-static unsigned int selinux_ipv4_forward(const struct nf_hook_ops *ops,
+static unsigned int selinux_ipv4_forward(void *priv,
 					 struct sk_buff *skb,
 					 const struct nf_hook_state *state)
 {
@@ -4874,7 +4874,7 @@ static unsigned int selinux_ipv4_forward(const struct nf_hook_ops *ops,
 }
 
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
-static unsigned int selinux_ipv6_forward(const struct nf_hook_ops *ops,
+static unsigned int selinux_ipv6_forward(void *priv,
 					 struct sk_buff *skb,
 					 const struct nf_hook_state *state)
 {
@@ -4924,7 +4924,7 @@ static unsigned int selinux_ip_output(struct sk_buff *skb,
 	return NF_ACCEPT;
 }
 
-static unsigned int selinux_ipv4_output(const struct nf_hook_ops *ops,
+static unsigned int selinux_ipv4_output(void *priv,
 					struct sk_buff *skb,
 					const struct nf_hook_state *state)
 {
@@ -5099,7 +5099,7 @@ static unsigned int selinux_ip_postroute(struct sk_buff *skb,
 	return NF_ACCEPT;
 }
 
-static unsigned int selinux_ipv4_postroute(const struct nf_hook_ops *ops,
+static unsigned int selinux_ipv4_postroute(void *priv,
 					   struct sk_buff *skb,
 					   const struct nf_hook_state *state)
 {
@@ -5107,7 +5107,7 @@ static unsigned int selinux_ipv4_postroute(const struct nf_hook_ops *ops,
 }
 
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
-static unsigned int selinux_ipv6_postroute(const struct nf_hook_ops *ops,
+static unsigned int selinux_ipv6_postroute(void *priv,
 					   struct sk_buff *skb,
 					   const struct nf_hook_state *state)
 {
security/smack/smack_netfilter.c +2 -2
index a455cfc9ec1f..a9e41da05d28 100644
@@ -21,7 +21,7 @@
 
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
 
-static unsigned int smack_ipv6_output(const struct nf_hook_ops *ops,
+static unsigned int smack_ipv6_output(void *priv,
 					struct sk_buff *skb,
 					const struct nf_hook_state *state)
 {
@@ -38,7 +38,7 @@ static unsigned int smack_ipv6_output(const struct nf_hook_ops *ops,
 }
 #endif	/* IPV6 */
 
-static unsigned int smack_ipv4_output(const struct nf_hook_ops *ops,
+static unsigned int smack_ipv4_output(void *priv,
 					struct sk_buff *skb,
 					const struct nf_hook_state *state)
 {
Signature · 1 line
2.2.1
#57 lore raw

PATCH 13/14 netfilter: Pass net into nf_xfrm_me_harder

In reply to Eric W. Biederman

Instead of calling dev_net on a likley looking network device pass state->net into nf_xfrm_me_harder.

---
 include/net/netfilter/nf_nat_core.h      | 2 +-
 net/ipv4/netfilter/nf_nat_l3proto_ipv4.c | 4 ++--
 net/ipv6/netfilter/nf_nat_l3proto_ipv6.c | 4 ++--
 net/netfilter/nf_nat_core.c              | 4 ++--
 4 files changed, 7 insertions(+), 7 deletions(-)
include/net/netfilter/nf_nat_core.h +1 -1
index fbfd1ba4254e..186c54138f35 100644
@@ -10,7 +10,7 @@
 unsigned int nf_nat_packet(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
 			   unsigned int hooknum, struct sk_buff *skb);
 
-int nf_xfrm_me_harder(struct sk_buff *skb, unsigned int family);
+int nf_xfrm_me_harder(struct net *net, struct sk_buff *skb, unsigned int family);
 
 static inline int nf_nat_initialized(struct nf_conn *ct,
 				     enum nf_nat_manip_type manip)
net/ipv4/netfilter/nf_nat_l3proto_ipv4.c +2 -2
index 8593a9d88619..bc3b9dcbf080 100644
@@ -396,7 +396,7 @@ nf_nat_ipv4_out(void *priv, struct sk_buff *skb,
 		    (ct->tuplehash[dir].tuple.dst.protonum != IPPROTO_ICMP &&
 		     ct->tuplehash[dir].tuple.src.u.all !=
 		     ct->tuplehash[!dir].tuple.dst.u.all)) {
-			err = nf_xfrm_me_harder(skb, AF_INET);
+			err = nf_xfrm_me_harder(state->net, skb, AF_INET);
 			if (err < 0)
 				ret = NF_DROP_ERR(err);
 		}
@@ -440,7 +440,7 @@ nf_nat_ipv4_local_fn(void *priv, struct sk_buff *skb,
 			 ct->tuplehash[dir].tuple.dst.protonum != IPPROTO_ICMP &&
 			 ct->tuplehash[dir].tuple.dst.u.all !=
 			 ct->tuplehash[!dir].tuple.src.u.all) {
-			err = nf_xfrm_me_harder(skb, AF_INET);
+			err = nf_xfrm_me_harder(state->net, skb, AF_INET);
 			if (err < 0)
 				ret = NF_DROP_ERR(err);
 		}
net/ipv6/netfilter/nf_nat_l3proto_ipv6.c +2 -2
index 357f57ba47e4..18e835ffbef3 100644
@@ -403,7 +403,7 @@ nf_nat_ipv6_out(void *priv, struct sk_buff *skb,
 		    (ct->tuplehash[dir].tuple.dst.protonum != IPPROTO_ICMPV6 &&
 		     ct->tuplehash[dir].tuple.src.u.all !=
 		     ct->tuplehash[!dir].tuple.dst.u.all)) {
-			err = nf_xfrm_me_harder(skb, AF_INET6);
+			err = nf_xfrm_me_harder(state->net, skb, AF_INET6);
 			if (err < 0)
 				ret = NF_DROP_ERR(err);
 		}
@@ -446,7 +446,7 @@ nf_nat_ipv6_local_fn(void *priv, struct sk_buff *skb,
 			 ct->tuplehash[dir].tuple.dst.protonum != IPPROTO_ICMPV6 &&
 			 ct->tuplehash[dir].tuple.dst.u.all !=
 			 ct->tuplehash[!dir].tuple.src.u.all) {
-			err = nf_xfrm_me_harder(skb, AF_INET6);
+			err = nf_xfrm_me_harder(state->net, skb, AF_INET6);
 			if (err < 0)
 				ret = NF_DROP_ERR(err);
 		}
net/netfilter/nf_nat_core.c +2 -2
index 5113dfd39df9..06a9f45771ab 100644
@@ -83,7 +83,7 @@ out:
 	rcu_read_unlock();
 }
 
-int nf_xfrm_me_harder(struct sk_buff *skb, unsigned int family)
+int nf_xfrm_me_harder(struct net *net, struct sk_buff *skb, unsigned int family)
 {
 	struct flowi fl;
 	unsigned int hh_len;
@@ -99,7 +99,7 @@ int nf_xfrm_me_harder(struct sk_buff *skb, unsigned int family)
 		dst = ((struct xfrm_dst *)dst)->route;
 	dst_hold(dst);
 
-	dst = xfrm_lookup(dev_net(dst->dev), dst, &fl, skb->sk, 0);
+	dst = xfrm_lookup(net, dst, &fl, skb->sk, 0);
 	if (IS_ERR(dst))
 		return PTR_ERR(dst);
 
Signature · 1 line
2.2.1
#58 lore raw

PATCH 14/14 netfilter: Use nf_ct_net instead of dev_net(out) in nf_nat_masquerade_ipv6

In reply to Eric W. Biederman

Use nf_ct_net(ct) instead of guessing that the netdevice out can reliably report the network namespace the conntrack operation is happening in.

---
 net/ipv6/netfilter/nf_nat_masquerade_ipv6.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/netfilter/nf_nat_masquerade_ipv6.c +1 -1
index 7745609665cd..31ba7ca19757 100644
@@ -34,7 +34,7 @@ nf_nat_masquerade_ipv6(struct sk_buff *skb, const struct nf_nat_range *range,
 	NF_CT_ASSERT(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED ||
 			    ctinfo == IP_CT_RELATED_REPLY));
 
-	if (ipv6_dev_get_saddr(dev_net(out), out,
+	if (ipv6_dev_get_saddr(nf_ct_net(ct), out,
 			       &ipv6_hdr(skb)->daddr, 0, &src) < 0)
 		return NF_DROP;
 
Signature · 1 line
2.2.1
Acked-by on the v1 series
#59 lore raw

Re: [PATCH next 0/14] netfilter: Stop guessing net

In reply to Eric W. Biederman
Quoted 10 lines · Eric W. Biederman

This is the next installment of my work to pass struct net through the output path so the code does not need to guess how to figure out which network namespace it is in, and ultimately routes can have output devices in another network namespace.

This round the changes focus on using nf_hook_state->net in the netfilter functions, and not passing nf_hook_ops to the functions on the netfilter chains.

Quoted 8 lines

Given the way per network namespace operations are implemented callers can not rely on addresses or games with container_of. None of the netfilter hooks do but we should remove the temptation. Plus removing the extra parameter and the extra jump to get at values should reduce the number of cache line misses.

At a practical matter this is the bulk of the work that other pieces of code in the netfilter stack will depend on.

Great!

LGTM.

--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#60 lore raw

Re: [PATCH next 0/14] netfilter: Stop guessing net

Quoted 17 lines · Nicolas Dichtel
Quoted 16 lines · Eric W. Biederman

This is the next installment of my work to pass struct net through the output path so the code does not need to guess how to figure out which network namespace it is in, and ultimately routes can have output devices in another network namespace.

This round the changes focus on using nf_hook_state->net in the netfilter functions, and not passing nf_hook_ops to the functions on the netfilter chains.

Given the way per network namespace operations are implemented callers can not rely on addresses or games with container_of. None of the netfilter hooks do but we should remove the temptation. Plus removing the extra parameter and the extra jump to get at values should reduce the number of cache line misses.

Quoted 6 lines
Quoted 2 lines

At a practical matter this is the bulk of the work that other pieces of code in the netfilter stack will depend on.

Great!

LGTM.

Grr. A build bot failed to build this patchset and let me know. Apparently a gremlin got in and disabled CONFIG_NET_ACT_CONNMARK in my test config.

I will respin in a minute.

Eric

--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#61 lore raw

cover netfilter: Stop guessing net (take 2)

In reply to Eric W. Biederman

This is the next installment of my work to pass struct net through the output path so the code does not need to guess how to figure out which network namespace it is in, and ultimately routes can have output devices in another network namespace.

This round the changes focus on using nf_hook_state->net in the netfilter functions, and not passing nf_hook_ops to the functions on the netfilter chains.

Given the way per network namespace operations are implemented callers can not rely on addresses or games with container_of. None of the netfilter hooks do but we should remove the temptation. Plus removing the extra parameter and the extra jump to get at values should reduce the number of cache line misses.

At a practical matter this is the bulk of the work that other pieces of code in the netfilter stack will depend on.

Since the last round I have fixed my test config to build act_connmark and added the patch that modifies act_connmark to track which network namespace it was it.

The changes are also available against net-next at:
git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/net-next.git master

My entire pending set of changes for those who want to look ahead is at:
git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/net-next.git for-testing

Eric

Eric W. Biederman (15):
ebtables: Simplify the arguments to ebt_do_table
inet netfilter: Remove hook from ip6t_do_table, arp_do_table, ipt_do_table inet netfilter: Prefer state->hook to ops->hooknum
nftables: kill nft_pktinfo.ops
x_tables: Pass struct net in xt_action_param
x_tables: Use par->net instead of computing from the passed net devices
nftables: Pass struct net in nft_pktinfo
nftables: Use pkt->net instead of computing net from the passed net_devices
net: Pass net to nf_dup_ipv4 and nf_dup_ipv6
act_connmark: Remember the struct net instead of guessing it.
nf_conntrack: Add a struct net parameter to l4_pkt_to_tuple
ipvs: Read hooknum from state rather than ops->hooknum
netfilter: Pass priv instead of nf_hook_ops to netfilter hooks
netfilter: Pass net into nf_xfrm_me_harder
netfilter: Use nf_ct_net instead of dev_net(out) in nf_nat_masquerade_ipv6

 include/linux/netfilter.h                      |  2 +-
 include/linux/netfilter/x_tables.h             |  3 +-
 include/linux/netfilter_arp/arp_tables.h       |  1 -
 include/linux/netfilter_bridge/ebtables.h      |  6 ++--
 include/linux/netfilter_ipv4/ip_tables.h       |  1 -
 include/linux/netfilter_ipv6/ip6_tables.h      |  1 -
 include/net/netfilter/br_netfilter.h           |  2 +-
 include/net/netfilter/ipv4/nf_dup_ipv4.h       |  2 +-
 include/net/netfilter/ipv6/nf_dup_ipv6.h       |  2 +-
 include/net/netfilter/nf_conntrack.h           |  3 +-
 include/net/netfilter/nf_conntrack_core.h      |  1 +
 include/net/netfilter/nf_conntrack_l4proto.h   |  2 +-
 include/net/netfilter/nf_nat_core.h            |  2 +-
 include/net/netfilter/nf_nat_l3proto.h         | 32 ++++++++++----------
 include/net/netfilter/nf_tables.h              | 14 ++++-----
 include/net/netfilter/nf_tables_ipv4.h         |  3 +-
 include/net/netfilter/nf_tables_ipv6.h         |  3 +-
 include/net/tc_act/tc_connmark.h               |  1 +
 net/bridge/br_netfilter_hooks.c                | 14 ++++-----
 net/bridge/br_netfilter_ipv6.c                 |  2 +-
 net/bridge/netfilter/ebt_log.c                 |  2 +-
 net/bridge/netfilter/ebt_nflog.c               |  2 +-
 net/bridge/netfilter/ebtable_broute.c          |  8 +++--
 net/bridge/netfilter/ebtable_filter.c          | 10 +++---
 net/bridge/netfilter/ebtable_nat.c             | 10 +++---
 net/bridge/netfilter/ebtables.c                | 14 +++++----
 net/bridge/netfilter/nf_tables_bridge.c        | 20 ++++++------
 net/bridge/netfilter/nft_reject_bridge.c       | 19 ++++++------
 net/decnet/netfilter/dn_rtmsg.c                |  2 +-
 net/ipv4/netfilter/arp_tables.c                |  3 +-
 net/ipv4/netfilter/arptable_filter.c           |  5 ++-
 net/ipv4/netfilter/ip_tables.c                 |  3 +-
 net/ipv4/netfilter/ipt_CLUSTERIP.c             |  2 +-
 net/ipv4/netfilter/ipt_SYNPROXY.c              |  4 +--
 net/ipv4/netfilter/ipt_rpfilter.c              |  5 ++-
 net/ipv4/netfilter/iptable_filter.c            |  7 ++---
 net/ipv4/netfilter/iptable_mangle.c            | 14 ++++-----
 net/ipv4/netfilter/iptable_nat.c               | 21 ++++++-------
 net/ipv4/netfilter/iptable_raw.c               |  7 ++---
 net/ipv4/netfilter/iptable_security.c          |  7 ++---
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 12 ++++----
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   |  4 +--
 net/ipv4/netfilter/nf_defrag_ipv4.c            |  4 +--
 net/ipv4/netfilter/nf_dup_ipv4.c               | 23 +++-----------
 net/ipv4/netfilter/nf_nat_l3proto_ipv4.c       | 42 +++++++++++++-------------
 net/ipv4/netfilter/nf_tables_arp.c             |  6 ++--
 net/ipv4/netfilter/nf_tables_ipv4.c            | 10 +++---
 net/ipv4/netfilter/nft_chain_nat_ipv4.c        | 22 +++++++-------
 net/ipv4/netfilter/nft_chain_route_ipv4.c      |  6 ++--
 net/ipv4/netfilter/nft_dup_ipv4.c              |  2 +-
 net/ipv4/netfilter/nft_masq_ipv4.c             |  2 +-
 net/ipv4/netfilter/nft_redir_ipv4.c            |  2 +-
 net/ipv4/netfilter/nft_reject_ipv4.c           |  5 ++-
 net/ipv6/netfilter/ip6_tables.c                |  3 +-
 net/ipv6/netfilter/ip6t_REJECT.c               |  2 +-
 net/ipv6/netfilter/ip6t_SYNPROXY.c             |  4 +--
 net/ipv6/netfilter/ip6t_rpfilter.c             |  6 ++--
 net/ipv6/netfilter/ip6table_filter.c           |  5 ++-
 net/ipv6/netfilter/ip6table_mangle.c           | 14 ++++-----
 net/ipv6/netfilter/ip6table_nat.c              | 21 ++++++-------
 net/ipv6/netfilter/ip6table_raw.c              |  5 ++-
 net/ipv6/netfilter/ip6table_security.c         |  5 ++-
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 12 ++++----
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c |  3 +-
 net/ipv6/netfilter/nf_defrag_ipv6_hooks.c      |  6 ++--
 net/ipv6/netfilter/nf_dup_ipv6.c               | 23 +++-----------
 net/ipv6/netfilter/nf_nat_l3proto_ipv6.c       | 42 +++++++++++++-------------
 net/ipv6/netfilter/nf_nat_masquerade_ipv6.c    |  2 +-
 net/ipv6/netfilter/nf_tables_ipv6.c            | 10 +++---
 net/ipv6/netfilter/nft_chain_nat_ipv6.c        | 22 +++++++-------
 net/ipv6/netfilter/nft_chain_route_ipv6.c      |  6 ++--
 net/ipv6/netfilter/nft_dup_ipv6.c              |  2 +-
 net/ipv6/netfilter/nft_redir_ipv6.c            |  3 +-
 net/ipv6/netfilter/nft_reject_ipv6.c           |  7 ++---
 net/netfilter/core.c                           |  2 +-
 net/netfilter/ipset/ip_set_core.c              |  9 ++----
 net/netfilter/ipvs/ip_vs_core.c                | 40 ++++++++++++------------
 net/netfilter/nf_conntrack_core.c              | 10 +++---
 net/netfilter/nf_conntrack_proto_dccp.c        |  2 +-
 net/netfilter/nf_conntrack_proto_generic.c     |  2 +-
 net/netfilter/nf_conntrack_proto_gre.c         |  3 +-
 net/netfilter/nf_conntrack_proto_sctp.c        |  2 +-
 net/netfilter/nf_conntrack_proto_tcp.c         |  2 +-
 net/netfilter/nf_conntrack_proto_udp.c         |  1 +
 net/netfilter/nf_conntrack_proto_udplite.c     |  1 +
 net/netfilter/nf_nat_core.c                    |  4 +--
 net/netfilter/nf_tables_core.c                 | 10 +++---
 net/netfilter/nf_tables_netdev.c               | 20 ++++++------
 net/netfilter/nft_log.c                        |  3 +-
 net/netfilter/nft_meta.c                       |  4 +--
 net/netfilter/nft_queue.c                      |  2 +-
 net/netfilter/nft_reject_inet.c                | 19 ++++++------
 net/netfilter/xt_LOG.c                         |  2 +-
 net/netfilter/xt_NFLOG.c                       |  2 +-
 net/netfilter/xt_TCPMSS.c                      |  2 +-
 net/netfilter/xt_TEE.c                         |  4 +--
 net/netfilter/xt_TPROXY.c                      | 24 +++++++--------
 net/netfilter/xt_addrtype.c                    |  4 +--
 net/netfilter/xt_connlimit.c                   |  4 +--
 net/netfilter/xt_osf.c                         |  2 +-
 net/netfilter/xt_recent.c                      |  2 +-
 net/netfilter/xt_socket.c                      | 14 +++++----
 net/openvswitch/conntrack.c                    |  2 +-
 net/sched/act_connmark.c                       |  5 +--
 net/sched/act_ipt.c                            |  1 +
 net/sched/em_ipset.c                           |  1 +
 security/selinux/hooks.c                       | 10 +++---
 security/smack/smack_netfilter.c               |  4 +--
 108 files changed, 389 insertions(+), 435 deletions(-)
#62 lore raw

PATCH 1/15 ebtables: Simplify the arguments to ebt_do_table

Nearly everything thing of interest to ebt_do_table is already present in nf_hook_state. Simplify ebt_do_table by just passing in the skb, nf_hook_state, and the table. This make the code easier to read and maintenance easier.

To support this create an nf_hook_state on the stack in ebt_broute (the only caller without a nf_hook_state already available). This new nf_hook_state adds no new computations to ebt_broute, but does use a few more bytes of stack.

---
 include/linux/netfilter_bridge/ebtables.h |  6 +++---
 net/bridge/netfilter/ebtable_broute.c     |  8 ++++++--
 net/bridge/netfilter/ebtable_filter.c     |  6 ++----
 net/bridge/netfilter/ebtable_nat.c        |  6 ++----
 net/bridge/netfilter/ebtables.c           | 13 +++++++------
 5 files changed, 20 insertions(+), 19 deletions(-)
include/linux/netfilter_bridge/ebtables.h +3 -3
index 8ca6d6464ea3..2ea517c7c6b9 100644
@@ -111,9 +111,9 @@ struct ebt_table {
 extern struct ebt_table *ebt_register_table(struct net *net,
 					    const struct ebt_table *table);
 extern void ebt_unregister_table(struct net *net, struct ebt_table *table);
-extern unsigned int ebt_do_table(unsigned int hook, struct sk_buff *skb,
-   const struct net_device *in, const struct net_device *out,
-   struct ebt_table *table);
+extern unsigned int ebt_do_table(struct sk_buff *skb,
+				 const struct nf_hook_state *state,
+				 struct ebt_table *table);
 
 /* Used in the kernel match() functions */
 #define FWINV(bool,invflg) ((bool) ^ !!(info->invflags & invflg))
net/bridge/netfilter/ebtable_broute.c +6 -2
index d2cdf5d6e98c..ec94c6f1ae88 100644
@@ -50,10 +50,14 @@ static const struct ebt_table broute_table = {
 
 static int ebt_broute(struct sk_buff *skb)
 {
+	struct nf_hook_state state;
 	int ret;
 
-	ret = ebt_do_table(NF_BR_BROUTING, skb, skb->dev, NULL,
-			   dev_net(skb->dev)->xt.broute_table);
+	nf_hook_state_init(&state, NULL, NF_BR_BROUTING, INT_MIN,
+			   NFPROTO_BRIDGE, skb->dev, NULL, NULL,
+			   dev_net(skb->dev), NULL);
+
+	ret = ebt_do_table(skb, &state, state.net->xt.broute_table);
 	if (ret == NF_DROP)
 		return 1; /* route it */
 	return 0; /* bridge it */
net/bridge/netfilter/ebtable_filter.c +2 -4
index ab20d6ed6e2f..118ce40ac181 100644
@@ -60,16 +60,14 @@ static unsigned int
 ebt_in_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	    const struct nf_hook_state *state)
 {
-	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    state->net->xt.frame_filter);
+	return ebt_do_table(skb, state, state->net->xt.frame_filter);
 }
 
 static unsigned int
 ebt_out_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
-	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    state->net->xt.frame_filter);
+	return ebt_do_table(skb, state, state->net->xt.frame_filter);
 }
 
 static struct nf_hook_ops ebt_ops_filter[] __read_mostly = {
net/bridge/netfilter/ebtable_nat.c +2 -4
index ad81a5a65644..56c3329d6c37 100644
@@ -60,16 +60,14 @@ static unsigned int
 ebt_nat_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	   const struct nf_hook_state *state)
 {
-	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    state->net->xt.frame_nat);
+	return ebt_do_table(skb, state, state->net->xt.frame_nat);
 }
 
 static unsigned int
 ebt_nat_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	    const struct nf_hook_state *state)
 {
-	return ebt_do_table(ops->hooknum, skb, state->in, state->out,
-			    state->net->xt.frame_nat);
+	return ebt_do_table(skb, state, state->net->xt.frame_nat);
 }
 
 static struct nf_hook_ops ebt_ops_nat[] __read_mostly = {
net/bridge/netfilter/ebtables.c +7 -6
index 48b6b01295de..8d5a3975b963 100644
@@ -183,10 +183,11 @@ struct ebt_entry *ebt_next_entry(const struct ebt_entry *entry)
 }
 
 /* Do some firewalling */
-unsigned int ebt_do_table (unsigned int hook, struct sk_buff *skb,
-   const struct net_device *in, const struct net_device *out,
-   struct ebt_table *table)
+unsigned int ebt_do_table(struct sk_buff *skb,
+			  const struct nf_hook_state *state,
+			  struct ebt_table *table)
 {
+	unsigned int hook = state->hook;
 	int i, nentries;
 	struct ebt_entry *point;
 	struct ebt_counter *counter_base, *cb_base;
@@ -199,8 +200,8 @@ unsigned int ebt_do_table (unsigned int hook, struct sk_buff *skb,
 	struct xt_action_param acpar;
 
 	acpar.family  = NFPROTO_BRIDGE;
-	acpar.in      = in;
-	acpar.out     = out;
+	acpar.in      = state->in;
+	acpar.out     = state->out;
 	acpar.hotdrop = false;
 	acpar.hooknum = hook;
 
@@ -220,7 +221,7 @@ unsigned int ebt_do_table (unsigned int hook, struct sk_buff *skb,
 	base = private->entries;
 	i = 0;
 	while (i < nentries) {
-		if (ebt_basic_match(point, skb, in, out))
+		if (ebt_basic_match(point, skb, state->in, state->out))
 			goto letscontinue;
 
 		if (EBT_MATCH_ITERATE(point, ebt_do_match, skb, &acpar) != 0)
Signature · 1 line
2.2.1
#63 lore raw

PATCH 2/15 inet netfilter: Remove hook from ip6t_do_table, arp_do_table, ipt_do_table

In reply to Eric W. Biederman

The values of ops->hooknum and state->hook are guaraneted to be equal making the hook argument to ip6t_do_table, arp_do_table, and ipt_do_table is unnecessary. Remove the unnecessary hook argument.

In the callers use state->hook instead of ops->hooknum for clarity and to reduce the number of cachelines the callers touch.

---
 include/linux/netfilter_arp/arp_tables.h  |  1 -
 include/linux/netfilter_ipv4/ip_tables.h  |  1 -
 include/linux/netfilter_ipv6/ip6_tables.h |  1 -
 net/ipv4/netfilter/arp_tables.c           |  2 +-
 net/ipv4/netfilter/arptable_filter.c      |  3 +--
 net/ipv4/netfilter/ip_tables.c            |  2 +-
 net/ipv4/netfilter/iptable_filter.c       |  5 ++---
 net/ipv4/netfilter/iptable_mangle.c       | 12 +++++-------
 net/ipv4/netfilter/iptable_nat.c          |  3 +--
 net/ipv4/netfilter/iptable_raw.c          |  5 ++---
 net/ipv4/netfilter/iptable_security.c     |  5 ++---
 net/ipv6/netfilter/ip6_tables.c           |  2 +-
 net/ipv6/netfilter/ip6table_filter.c      |  3 +--
 net/ipv6/netfilter/ip6table_mangle.c      | 12 +++++-------
 net/ipv6/netfilter/ip6table_nat.c         |  3 +--
 net/ipv6/netfilter/ip6table_raw.c         |  3 +--
 net/ipv6/netfilter/ip6table_security.c    |  3 +--
 17 files changed, 25 insertions(+), 41 deletions(-)
include/linux/netfilter_arp/arp_tables.h +0 -1
index c22a7fb8d0df..6f074db2f23d 100644
@@ -53,7 +53,6 @@ extern struct xt_table *arpt_register_table(struct net *net,
 					    const struct arpt_replace *repl);
 extern void arpt_unregister_table(struct xt_table *table);
 extern unsigned int arpt_do_table(struct sk_buff *skb,
-				  unsigned int hook,
 				  const struct nf_hook_state *state,
 				  struct xt_table *table);
 
include/linux/netfilter_ipv4/ip_tables.h +0 -1
index 4073510da485..aa598f942c01 100644
@@ -64,7 +64,6 @@ struct ipt_error {
 
 extern void *ipt_alloc_initial_table(const struct xt_table *);
 extern unsigned int ipt_do_table(struct sk_buff *skb,
-				 unsigned int hook,
 				 const struct nf_hook_state *state,
 				 struct xt_table *table);
 
include/linux/netfilter_ipv6/ip6_tables.h +0 -1
index b40d2b635778..0f76e5c674f9 100644
@@ -30,7 +30,6 @@ extern struct xt_table *ip6t_register_table(struct net *net,
 					    const struct ip6t_replace *repl);
 extern void ip6t_unregister_table(struct net *net, struct xt_table *table);
 extern unsigned int ip6t_do_table(struct sk_buff *skb,
-				  unsigned int hook,
 				  const struct nf_hook_state *state,
 				  struct xt_table *table);
 
net/ipv4/netfilter/arp_tables.c +1 -1
index 8f87fc38ccde..10eb2b297450 100644
@@ -247,10 +247,10 @@ struct arpt_entry *arpt_next_entry(const struct arpt_entry *entry)
 }
 
 unsigned int arpt_do_table(struct sk_buff *skb,
-			   unsigned int hook,
 			   const struct nf_hook_state *state,
 			   struct xt_table *table)
 {
+	unsigned int hook = state->hook;
 	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
 	unsigned int verdict = NF_DROP;
 	const struct arphdr *arp;
net/ipv4/netfilter/arptable_filter.c +1 -2
index d217e4c19645..1352e12d4068 100644
@@ -30,8 +30,7 @@ static unsigned int
 arptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	return arpt_do_table(skb, ops->hooknum, state,
-			     state->net->ipv4.arptable_filter);
+	return arpt_do_table(skb, state, state->net->ipv4.arptable_filter);
 }
 
 static struct nf_hook_ops *arpfilter_ops __read_mostly;
net/ipv4/netfilter/ip_tables.c +1 -1
index 5d514eac4c31..2b049e135de8 100644
@@ -285,10 +285,10 @@ struct ipt_entry *ipt_next_entry(const struct ipt_entry *entry)
 /* Returns one of the generic firewall policies, like NF_ACCEPT. */
 unsigned int
 ipt_do_table(struct sk_buff *skb,
-	     unsigned int hook,
 	     const struct nf_hook_state *state,
 	     struct xt_table *table)
 {
+	unsigned int hook = state->hook;
 	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
 	const struct iphdr *ip;
 	/* Initializing verdict to NF_DROP keeps gcc happy. */
net/ipv4/netfilter/iptable_filter.c +2 -3
index 32feff32b116..02d4c5395d6e 100644
@@ -36,14 +36,13 @@ static unsigned int
 iptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		    const struct nf_hook_state *state)
 {
-	if (ops->hooknum == NF_INET_LOCAL_OUT &&
+	if (state->hook == NF_INET_LOCAL_OUT &&
 	    (skb->len < sizeof(struct iphdr) ||
 	     ip_hdrlen(skb) < sizeof(struct iphdr)))
 		/* root is playing with raw sockets. */
 		return NF_ACCEPT;
 
-	return ipt_do_table(skb, ops->hooknum, state,
-			    state->net->ipv4.iptable_filter);
+	return ipt_do_table(skb, state, state->net->ipv4.iptable_filter);
 }
 
 static struct nf_hook_ops *filter_ops __read_mostly;
net/ipv4/netfilter/iptable_mangle.c +5 -7
index 4a5150fc9510..dc2ff6884999 100644
@@ -58,8 +58,7 @@ ipt_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state)
 	daddr = iph->daddr;
 	tos = iph->tos;
 
-	ret = ipt_do_table(skb, NF_INET_LOCAL_OUT, state,
-			   state->net->ipv4.iptable_mangle);
+	ret = ipt_do_table(skb, state, state->net->ipv4.iptable_mangle);
 	/* Reroute for ANY change. */
 	if (ret != NF_DROP && ret != NF_STOLEN) {
 		iph = ip_hdr(skb);
@@ -83,14 +82,13 @@ iptable_mangle_hook(const struct nf_hook_ops *ops,
 		     struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	if (ops->hooknum == NF_INET_LOCAL_OUT)
+	if (state->hook == NF_INET_LOCAL_OUT)
 		return ipt_mangle_out(skb, state);
-	if (ops->hooknum == NF_INET_POST_ROUTING)
-		return ipt_do_table(skb, ops->hooknum, state,
+	if (state->hook == NF_INET_POST_ROUTING)
+		return ipt_do_table(skb, state,
 				    state->net->ipv4.iptable_mangle);
 	/* PREROUTING/INPUT/FORWARD: */
-	return ipt_do_table(skb, ops->hooknum, state,
-			    state->net->ipv4.iptable_mangle);
+	return ipt_do_table(skb, state, state->net->ipv4.iptable_mangle);
 }
 
 static struct nf_hook_ops *mangle_ops __read_mostly;
net/ipv4/netfilter/iptable_nat.c +1 -2
index 4f4c64f81169..8ff63ac1f0d6 100644
@@ -33,8 +33,7 @@ static unsigned int iptable_nat_do_chain(const struct nf_hook_ops *ops,
 					 const struct nf_hook_state *state,
 					 struct nf_conn *ct)
 {
-	return ipt_do_table(skb, ops->hooknum, state,
-			    state->net->ipv4.nat_table);
+	return ipt_do_table(skb, state, state->net->ipv4.nat_table);
 }
 
 static unsigned int iptable_nat_ipv4_fn(const struct nf_hook_ops *ops,
net/ipv4/netfilter/iptable_raw.c +2 -3
index 20126e469ffb..bbb0523d87de 100644
@@ -23,14 +23,13 @@ static unsigned int
 iptable_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		 const struct nf_hook_state *state)
 {
-	if (ops->hooknum == NF_INET_LOCAL_OUT &&
+	if (state->hook == NF_INET_LOCAL_OUT &&
 	    (skb->len < sizeof(struct iphdr) ||
 	     ip_hdrlen(skb) < sizeof(struct iphdr)))
 		/* root is playing with raw sockets. */
 		return NF_ACCEPT;
 
-	return ipt_do_table(skb, ops->hooknum, state,
-			    state->net->ipv4.iptable_raw);
+	return ipt_do_table(skb, state, state->net->ipv4.iptable_raw);
 }
 
 static struct nf_hook_ops *rawtable_ops __read_mostly;
net/ipv4/netfilter/iptable_security.c +2 -3
index 82fefd609b85..b92417038705 100644
@@ -40,14 +40,13 @@ static unsigned int
 iptable_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
-	if (ops->hooknum == NF_INET_LOCAL_OUT &&
+	if (state->hook == NF_INET_LOCAL_OUT &&
 	    (skb->len < sizeof(struct iphdr) ||
 	     ip_hdrlen(skb) < sizeof(struct iphdr)))
 		/* Somebody is playing with raw sockets. */
 		return NF_ACCEPT;
 
-	return ipt_do_table(skb, ops->hooknum, state,
-			    state->net->ipv4.iptable_security);
+	return ipt_do_table(skb, state, state->net->ipv4.iptable_security);
 }
 
 static struct nf_hook_ops *sectbl_ops __read_mostly;
net/ipv6/netfilter/ip6_tables.c +1 -1
index cd9b401231d3..da6446b6e3f9 100644
@@ -314,10 +314,10 @@ ip6t_next_entry(const struct ip6t_entry *entry)
 /* Returns one of the generic firewall policies, like NF_ACCEPT. */
 unsigned int
 ip6t_do_table(struct sk_buff *skb,
-	      unsigned int hook,
 	      const struct nf_hook_state *state,
 	      struct xt_table *table)
 {
+	unsigned int hook = state->hook;
 	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
 	/* Initializing verdict to NF_DROP keeps gcc happy. */
 	unsigned int verdict = NF_DROP;
net/ipv6/netfilter/ip6table_filter.c +1 -2
index 2449005fb5dc..a7327f61b90c 100644
@@ -35,8 +35,7 @@ static unsigned int
 ip6table_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	return ip6t_do_table(skb, ops->hooknum, state,
-			     state->net->ipv6.ip6table_filter);
+	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_filter);
 }
 
 static struct nf_hook_ops *filter_ops __read_mostly;
net/ipv6/netfilter/ip6table_mangle.c +5 -7
index a46dbf097d29..c2e061dcedf3 100644
@@ -57,8 +57,7 @@ ip6t_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state)
 	/* flowlabel and prio (includes version, which shouldn't change either */
 	flowlabel = *((u_int32_t *)ipv6_hdr(skb));
 
-	ret = ip6t_do_table(skb, NF_INET_LOCAL_OUT, state,
-			    state->net->ipv6.ip6table_mangle);
+	ret = ip6t_do_table(skb, state, state->net->ipv6.ip6table_mangle);
 
 	if (ret != NF_DROP && ret != NF_STOLEN &&
 	    (!ipv6_addr_equal(&ipv6_hdr(skb)->saddr, &saddr) ||
@@ -79,14 +78,13 @@ static unsigned int
 ip6table_mangle_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	if (ops->hooknum == NF_INET_LOCAL_OUT)
+	if (state->hook == NF_INET_LOCAL_OUT)
 		return ip6t_mangle_out(skb, state);
-	if (ops->hooknum == NF_INET_POST_ROUTING)
-		return ip6t_do_table(skb, ops->hooknum, state,
+	if (state->hook == NF_INET_POST_ROUTING)
+		return ip6t_do_table(skb, state,
 				     state->net->ipv6.ip6table_mangle);
 	/* INPUT/FORWARD */
-	return ip6t_do_table(skb, ops->hooknum, state,
-			     state->net->ipv6.ip6table_mangle);
+	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_mangle);
 }
 
 static struct nf_hook_ops *mangle_ops __read_mostly;
net/ipv6/netfilter/ip6table_nat.c +1 -2
index a56451de127f..efa6754c4d06 100644
@@ -35,8 +35,7 @@ static unsigned int ip6table_nat_do_chain(const struct nf_hook_ops *ops,
 					  const struct nf_hook_state *state,
 					  struct nf_conn *ct)
 {
-	return ip6t_do_table(skb, ops->hooknum, state,
-			     state->net->ipv6.ip6table_nat);
+	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_nat);
 }
 
 static unsigned int ip6table_nat_fn(const struct nf_hook_ops *ops,
net/ipv6/netfilter/ip6table_raw.c +1 -2
index 18e831e35782..fac6ad7c0a7c 100644
@@ -22,8 +22,7 @@ static unsigned int
 ip6table_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		  const struct nf_hook_state *state)
 {
-	return ip6t_do_table(skb, ops->hooknum, state,
-			     state->net->ipv6.ip6table_raw);
+	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_raw);
 }
 
 static struct nf_hook_ops *rawtable_ops __read_mostly;
net/ipv6/netfilter/ip6table_security.c +1 -2
index 83bc96ae5d73..96c94fc240c8 100644
@@ -39,8 +39,7 @@ static unsigned int
 ip6table_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		       const struct nf_hook_state *state)
 {
-	return ip6t_do_table(skb, ops->hooknum, state,
-			     state->net->ipv6.ip6table_security);
+	return ip6t_do_table(skb, state, state->net->ipv6.ip6table_security);
 }
 
 static struct nf_hook_ops *sectbl_ops __read_mostly;
Signature · 1 line
2.2.1
#64 lore raw

PATCH 3/15 inet netfilter: Prefer state->hook to ops->hooknum

In reply to Eric W. Biederman

The values of nf_hook_state.hook and nf_hook_ops.hooknum must be the same by definition.

We are more likely to access the fields in nf_hook_state over the fields in nf_hook_ops so with a little luck this results in fewer cache line misses, and slightly more consistent code.

---
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c |  4 ++--
 net/ipv4/netfilter/nf_defrag_ipv4.c            |  2 +-
 net/ipv4/netfilter/nf_nat_l3proto_ipv4.c       | 14 +++++++-------
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c |  4 ++--
 net/ipv6/netfilter/nf_defrag_ipv6_hooks.c      |  4 ++--
 net/ipv6/netfilter/nf_nat_l3proto_ipv6.c       | 14 +++++++-------
 6 files changed, 21 insertions(+), 21 deletions(-)
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +2 -2
index 9564684876c9..15749cc5cf2b 100644
@@ -147,7 +147,7 @@ static unsigned int ipv4_conntrack_in(const struct nf_hook_ops *ops,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
-	return nf_conntrack_in(state->net, PF_INET, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET, state->hook, skb);
 }
 
 static unsigned int ipv4_conntrack_local(const struct nf_hook_ops *ops,
@@ -158,7 +158,7 @@ static unsigned int ipv4_conntrack_local(const struct nf_hook_ops *ops,
 	if (skb->len < sizeof(struct iphdr) ||
 	    ip_hdrlen(skb) < sizeof(struct iphdr))
 		return NF_ACCEPT;
-	return nf_conntrack_in(state->net, PF_INET, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET, state->hook, skb);
 }
 
 /* Connection tracking may drop packets, but never alters them, so
net/ipv4/netfilter/nf_defrag_ipv4.c +1 -1
index 9306ec4fab41..8aea536d2e83 100644
@@ -83,7 +83,7 @@ static unsigned int ipv4_conntrack_defrag(const struct nf_hook_ops *ops,
 	/* Gather fragments. */
 	if (ip_is_fragment(ip_hdr(skb))) {
 		enum ip_defrag_users user =
-			nf_ct_defrag_user(ops->hooknum, skb);
+			nf_ct_defrag_user(state->hook, skb);
 
 		if (nf_ct_ipv4_gather_frags(skb, user))
 			return NF_STOLEN;
net/ipv4/netfilter/nf_nat_l3proto_ipv4.c +7 -7
index 22f4579b0c2a..16da45a76dac 100644
@@ -266,7 +266,7 @@ nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn_nat *nat;
 	/* maniptype == SRC for postrouting. */
-	enum nf_nat_manip_type maniptype = HOOK2MANIP(ops->hooknum);
+	enum nf_nat_manip_type maniptype = HOOK2MANIP(state->hook);
 
 	/* We never see fragments: conntrack defrags on pre-routing
 	 * and local-out, and nf_nat_out protects post-routing.
@@ -295,7 +295,7 @@ nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	case IP_CT_RELATED_REPLY:
 		if (ip_hdr(skb)->protocol == IPPROTO_ICMP) {
 			if (!nf_nat_icmp_reply_translation(skb, ct, ctinfo,
-							   ops->hooknum))
+							   state->hook))
 				return NF_DROP;
 			else
 				return NF_ACCEPT;
@@ -312,17 +312,17 @@ nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 			if (ret != NF_ACCEPT)
 				return ret;
 
-			if (nf_nat_initialized(ct, HOOK2MANIP(ops->hooknum)))
+			if (nf_nat_initialized(ct, HOOK2MANIP(state->hook)))
 				break;
 
-			ret = nf_nat_alloc_null_binding(ct, ops->hooknum);
+			ret = nf_nat_alloc_null_binding(ct, state->hook);
 			if (ret != NF_ACCEPT)
 				return ret;
 		} else {
 			pr_debug("Already setup manip %s for ct %p\n",
 				 maniptype == NF_NAT_MANIP_SRC ? "SRC" : "DST",
 				 ct);
-			if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat,
+			if (nf_nat_oif_changed(state->hook, ctinfo, nat,
 					       state->out))
 				goto oif_changed;
 		}
@@ -332,11 +332,11 @@ nf_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		/* ESTABLISHED */
 		NF_CT_ASSERT(ctinfo == IP_CT_ESTABLISHED ||
 			     ctinfo == IP_CT_ESTABLISHED_REPLY);
-		if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat, state->out))
+		if (nf_nat_oif_changed(state->hook, ctinfo, nat, state->out))
 			goto oif_changed;
 	}
 
-	return nf_nat_packet(ct, ctinfo, ops->hooknum, skb);
+	return nf_nat_packet(ct, ctinfo, state->hook, skb);
 
 oif_changed:
 	nf_ct_kill_acct(ct, ctinfo, skb);
net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +2 -2
index 1ef1b79def56..339be1d59afc 100644
@@ -169,7 +169,7 @@ static unsigned int ipv6_conntrack_in(const struct nf_hook_ops *ops,
 				      struct sk_buff *skb,
 				      const struct nf_hook_state *state)
 {
-	return nf_conntrack_in(state->net, PF_INET6, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET6, state->hook, skb);
 }
 
 static unsigned int ipv6_conntrack_local(const struct nf_hook_ops *ops,
@@ -181,7 +181,7 @@ static unsigned int ipv6_conntrack_local(const struct nf_hook_ops *ops,
 		net_notice_ratelimited("ipv6_conntrack_local: packet too short\n");
 		return NF_ACCEPT;
 	}
-	return nf_conntrack_in(state->net, PF_INET6, ops->hooknum, skb);
+	return nf_conntrack_in(state->net, PF_INET6, state->hook, skb);
 }
 
 static struct nf_hook_ops ipv6_conntrack_ops[] __read_mostly = {
net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +2 -2
index 6b576be3c83e..a9c08520596b 100644
@@ -63,7 +63,7 @@ static unsigned int ipv6_defrag(const struct nf_hook_ops *ops,
 		return NF_ACCEPT;
 #endif
 
-	reasm = nf_ct_frag6_gather(skb, nf_ct6_defrag_user(ops->hooknum, skb));
+	reasm = nf_ct_frag6_gather(skb, nf_ct6_defrag_user(state->hook, skb));
 	/* queued */
 	if (reasm == NULL)
 		return NF_STOLEN;
@@ -74,7 +74,7 @@ static unsigned int ipv6_defrag(const struct nf_hook_ops *ops,
 
 	nf_ct_frag6_consume_orig(reasm);
 
-	NF_HOOK_THRESH(NFPROTO_IPV6, ops->hooknum, state->net, state->sk, reasm,
+	NF_HOOK_THRESH(NFPROTO_IPV6, state->hook, state->net, state->sk, reasm,
 		       state->in, state->out,
 		       state->okfn, NF_IP6_PRI_CONNTRACK_DEFRAG + 1);
 
net/ipv6/netfilter/nf_nat_l3proto_ipv6.c +7 -7
index 70fbaed49edb..8bc94907dbd9 100644
@@ -272,7 +272,7 @@ nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	struct nf_conn *ct;
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn_nat *nat;
-	enum nf_nat_manip_type maniptype = HOOK2MANIP(ops->hooknum);
+	enum nf_nat_manip_type maniptype = HOOK2MANIP(state->hook);
 	__be16 frag_off;
 	int hdrlen;
 	u8 nexthdr;
@@ -303,7 +303,7 @@ nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 
 		if (hdrlen >= 0 && nexthdr == IPPROTO_ICMPV6) {
 			if (!nf_nat_icmpv6_reply_translation(skb, ct, ctinfo,
-							     ops->hooknum,
+							     state->hook,
 							     hdrlen))
 				return NF_DROP;
 			else
@@ -321,17 +321,17 @@ nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 			if (ret != NF_ACCEPT)
 				return ret;
 
-			if (nf_nat_initialized(ct, HOOK2MANIP(ops->hooknum)))
+			if (nf_nat_initialized(ct, HOOK2MANIP(state->hook)))
 				break;
 
-			ret = nf_nat_alloc_null_binding(ct, ops->hooknum);
+			ret = nf_nat_alloc_null_binding(ct, state->hook);
 			if (ret != NF_ACCEPT)
 				return ret;
 		} else {
 			pr_debug("Already setup manip %s for ct %p\n",
 				 maniptype == NF_NAT_MANIP_SRC ? "SRC" : "DST",
 				 ct);
-			if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat, state->out))
+			if (nf_nat_oif_changed(state->hook, ctinfo, nat, state->out))
 				goto oif_changed;
 		}
 		break;
@@ -340,11 +340,11 @@ nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		/* ESTABLISHED */
 		NF_CT_ASSERT(ctinfo == IP_CT_ESTABLISHED ||
 			     ctinfo == IP_CT_ESTABLISHED_REPLY);
-		if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat, state->out))
+		if (nf_nat_oif_changed(state->hook, ctinfo, nat, state->out))
 			goto oif_changed;
 	}
 
-	return nf_nat_packet(ct, ctinfo, ops->hooknum, skb);
+	return nf_nat_packet(ct, ctinfo, state->hook, skb);
 
 oif_changed:
 	nf_ct_kill_acct(ct, ctinfo, skb);
Signature · 1 line
2.2.1
#65 lore raw

PATCH 4/15 nftables: kill nft_pktinfo.ops

In reply to Eric W. Biederman

- Add nft_pktinfo.pf to replace ops->pf
- Add nft_pktinfo.hook to replace ops->hooknum

This simplifies the code, makes it more readable, and likely reduces cache line misses. Maintainability is enhanced as the details of nft_hook_ops are of no concern to the recpients of nft_pktinfo.

---
 include/net/netfilter/nf_tables.h         |  9 ++++-----
 include/net/netfilter/nf_tables_ipv4.h    |  3 +--
 include/net/netfilter/nf_tables_ipv6.h    |  3 +--
 net/bridge/netfilter/nf_tables_bridge.c   | 16 +++++++---------
 net/bridge/netfilter/nft_reject_bridge.c  | 12 ++++++------
 net/ipv4/netfilter/nf_tables_arp.c        |  2 +-
 net/ipv4/netfilter/nf_tables_ipv4.c       |  2 +-
 net/ipv4/netfilter/nft_chain_nat_ipv4.c   |  2 +-
 net/ipv4/netfilter/nft_chain_route_ipv4.c |  2 +-
 net/ipv4/netfilter/nft_dup_ipv4.c         |  2 +-
 net/ipv4/netfilter/nft_masq_ipv4.c        |  2 +-
 net/ipv4/netfilter/nft_redir_ipv4.c       |  2 +-
 net/ipv4/netfilter/nft_reject_ipv4.c      |  5 ++---
 net/ipv6/netfilter/nf_tables_ipv6.c       |  2 +-
 net/ipv6/netfilter/nft_chain_nat_ipv6.c   |  2 +-
 net/ipv6/netfilter/nft_chain_route_ipv6.c |  2 +-
 net/ipv6/netfilter/nft_dup_ipv6.c         |  2 +-
 net/ipv6/netfilter/nft_redir_ipv6.c       |  3 +--
 net/ipv6/netfilter/nft_reject_ipv6.c      |  5 ++---
 net/netfilter/nf_tables_core.c            |  2 +-
 net/netfilter/nf_tables_netdev.c          | 16 +++++++---------
 net/netfilter/nft_log.c                   |  2 +-
 net/netfilter/nft_meta.c                  |  4 ++--
 net/netfilter/nft_queue.c                 |  2 +-
 net/netfilter/nft_reject_inet.c           | 14 +++++++-------
 25 files changed, 54 insertions(+), 64 deletions(-)
include/net/netfilter/nf_tables.h +4 -5
index aa8bee72c9d3..c0899f97ff8d 100644
@@ -16,7 +16,8 @@ struct nft_pktinfo {
 	struct sk_buff			*skb;
 	const struct net_device		*in;
 	const struct net_device		*out;
-	const struct nf_hook_ops	*ops;
+	u8				pf;
+	u8				hook;
 	u8				nhoff;
 	u8				thoff;
 	u8				tprot;
@@ -25,16 +26,14 @@ struct nft_pktinfo {
 };
 
 static inline void nft_set_pktinfo(struct nft_pktinfo *pkt,
-				   const struct nf_hook_ops *ops,
 				   struct sk_buff *skb,
 				   const struct nf_hook_state *state)
 {
 	pkt->skb = skb;
 	pkt->in = pkt->xt.in = state->in;
 	pkt->out = pkt->xt.out = state->out;
-	pkt->ops = ops;
-	pkt->xt.hooknum = ops->hooknum;
-	pkt->xt.family = ops->pf;
+	pkt->hook = pkt->xt.hooknum = state->hook;
+	pkt->pf = pkt->xt.family = state->pf;
 }
 
 /**
include/net/netfilter/nf_tables_ipv4.h +1 -2
index 2df7f96902ee..ca6ef6bf775e 100644
@@ -6,13 +6,12 @@
 
 static inline void
 nft_set_pktinfo_ipv4(struct nft_pktinfo *pkt,
-		     const struct nf_hook_ops *ops,
 		     struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
 	struct iphdr *ip;
 
-	nft_set_pktinfo(pkt, ops, skb, state);
+	nft_set_pktinfo(pkt, skb, state);
 
 	ip = ip_hdr(pkt->skb);
 	pkt->tprot = ip->protocol;
include/net/netfilter/nf_tables_ipv6.h +1 -2
index 97db2e3a5e65..8ad39a6a5fe1 100644
@@ -6,14 +6,13 @@
 
 static inline int
 nft_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
-		     const struct nf_hook_ops *ops,
 		     struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
 	int protohdr, thoff = 0;
 	unsigned short frag_off;
 
-	nft_set_pktinfo(pkt, ops, skb, state);
+	nft_set_pktinfo(pkt, skb, state);
 
 	protohdr = ipv6_find_hdr(pkt->skb, &thoff, -1, &frag_off, NULL);
 	/* If malformed, drop it */
net/bridge/netfilter/nf_tables_bridge.c +7 -9
index a343e62442b1..318d825e4207 100644
@@ -65,27 +65,25 @@ int nft_bridge_ip6hdr_validate(struct sk_buff *skb)
 EXPORT_SYMBOL_GPL(nft_bridge_ip6hdr_validate);
 
 static inline void nft_bridge_set_pktinfo_ipv4(struct nft_pktinfo *pkt,
-					       const struct nf_hook_ops *ops,
 					       struct sk_buff *skb,
 					       const struct nf_hook_state *state)
 {
 	if (nft_bridge_iphdr_validate(skb))
-		nft_set_pktinfo_ipv4(pkt, ops, skb, state);
+		nft_set_pktinfo_ipv4(pkt, skb, state);
 	else
-		nft_set_pktinfo(pkt, ops, skb, state);
+		nft_set_pktinfo(pkt, skb, state);
 }
 
 static inline void nft_bridge_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
-					       const struct nf_hook_ops *ops,
 					       struct sk_buff *skb,
 					       const struct nf_hook_state *state)
 {
 #if IS_ENABLED(CONFIG_IPV6)
 	if (nft_bridge_ip6hdr_validate(skb) &&
-	    nft_set_pktinfo_ipv6(pkt, ops, skb, state) == 0)
+	    nft_set_pktinfo_ipv6(pkt, skb, state) == 0)
 		return;
 #endif
-	nft_set_pktinfo(pkt, ops, skb, state);
+	nft_set_pktinfo(pkt, skb, state);
 }
 
 static unsigned int
@@ -97,13 +95,13 @@ nft_do_chain_bridge(const struct nf_hook_ops *ops,
 
 	switch (eth_hdr(skb)->h_proto) {
 	case htons(ETH_P_IP):
-		nft_bridge_set_pktinfo_ipv4(&pkt, ops, skb, state);
+		nft_bridge_set_pktinfo_ipv4(&pkt, skb, state);
 		break;
 	case htons(ETH_P_IPV6):
-		nft_bridge_set_pktinfo_ipv6(&pkt, ops, skb, state);
+		nft_bridge_set_pktinfo_ipv6(&pkt, skb, state);
 		break;
 	default:
-		nft_set_pktinfo(&pkt, ops, skb, state);
+		nft_set_pktinfo(&pkt, skb, state);
 		break;
 	}
 
net/bridge/netfilter/nft_reject_bridge.c +6 -6
index 858d848564ee..cee92612b2cc 100644
@@ -273,16 +273,16 @@ static void nft_reject_bridge_eval(const struct nft_expr *expr,
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
 			nft_reject_br_send_v4_unreach(pkt->skb, pkt->in,
-						      pkt->ops->hooknum,
+						      pkt->hook,
 						      priv->icmp_code);
 			break;
 		case NFT_REJECT_TCP_RST:
 			nft_reject_br_send_v4_tcp_reset(pkt->skb, pkt->in,
-							pkt->ops->hooknum);
+							pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
 			nft_reject_br_send_v4_unreach(pkt->skb, pkt->in,
-						      pkt->ops->hooknum,
+						      pkt->hook,
 						      nft_reject_icmp_code(priv->icmp_code));
 			break;
 		}
@@ -291,16 +291,16 @@ static void nft_reject_bridge_eval(const struct nft_expr *expr,
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
 			nft_reject_br_send_v6_unreach(net, pkt->skb, pkt->in,
-						      pkt->ops->hooknum,
+						      pkt->hook,
 						      priv->icmp_code);
 			break;
 		case NFT_REJECT_TCP_RST:
 			nft_reject_br_send_v6_tcp_reset(net, pkt->skb, pkt->in,
-							pkt->ops->hooknum);
+							pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
 			nft_reject_br_send_v6_unreach(net, pkt->skb, pkt->in,
-						      pkt->ops->hooknum,
+						      pkt->hook,
 						      nft_reject_icmpv6_code(priv->icmp_code));
 			break;
 		}
net/ipv4/netfilter/nf_tables_arp.c +1 -1
index 8412268bbad1..883bbf83fe09 100644
@@ -21,7 +21,7 @@ nft_do_chain_arp(const struct nf_hook_ops *ops,
 {
 	struct nft_pktinfo pkt;
 
-	nft_set_pktinfo(&pkt, ops, skb, state);
+	nft_set_pktinfo(&pkt, skb, state);
 
 	return nft_do_chain(&pkt, ops);
 }
net/ipv4/netfilter/nf_tables_ipv4.c +1 -1
index aa180d3a69a5..805be5c9fcc3 100644
@@ -24,7 +24,7 @@ static unsigned int nft_do_chain_ipv4(const struct nf_hook_ops *ops,
 {
 	struct nft_pktinfo pkt;
 
-	nft_set_pktinfo_ipv4(&pkt, ops, skb, state);
+	nft_set_pktinfo_ipv4(&pkt, skb, state);
 
 	return nft_do_chain(&pkt, ops);
 }
net/ipv4/netfilter/nft_chain_nat_ipv4.c +1 -1
index bf5c30ae14e4..c3ffecf28d38 100644
@@ -33,7 +33,7 @@ static unsigned int nft_nat_do_chain(const struct nf_hook_ops *ops,
 {
 	struct nft_pktinfo pkt;
 
-	nft_set_pktinfo_ipv4(&pkt, ops, skb, state);
+	nft_set_pktinfo_ipv4(&pkt, skb, state);
 
 	return nft_do_chain(&pkt, ops);
 }
net/ipv4/netfilter/nft_chain_route_ipv4.c +1 -1
index e335b0afdaf3..2a1e3d8a3e43 100644
@@ -37,7 +37,7 @@ static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops,
 	    ip_hdrlen(skb) < sizeof(struct iphdr))
 		return NF_ACCEPT;
 
-	nft_set_pktinfo_ipv4(&pkt, ops, skb, state);
+	nft_set_pktinfo_ipv4(&pkt, skb, state);
 
 	mark = skb->mark;
 	iph = ip_hdr(skb);
net/ipv4/netfilter/nft_dup_ipv4.c +1 -1
index b45932d43b69..30bcf820e8bd 100644
@@ -30,7 +30,7 @@ static void nft_dup_ipv4_eval(const struct nft_expr *expr,
 	};
 	int oif = regs->data[priv->sreg_dev];
 
-	nf_dup_ipv4(pkt->skb, pkt->ops->hooknum, &gw, oif);
+	nf_dup_ipv4(pkt->skb, pkt->hook, &gw, oif);
 }
 
 static int nft_dup_ipv4_init(const struct nft_ctx *ctx,
net/ipv4/netfilter/nft_masq_ipv4.c +1 -1
index 40e414c4ca56..b72ffc58e255 100644
@@ -26,7 +26,7 @@ static void nft_masq_ipv4_eval(const struct nft_expr *expr,
 	memset(&range, 0, sizeof(range));
 	range.flags = priv->flags;
 
-	regs->verdict.code = nf_nat_masquerade_ipv4(pkt->skb, pkt->ops->hooknum,
+	regs->verdict.code = nf_nat_masquerade_ipv4(pkt->skb, pkt->hook,
 						    &range, pkt->out);
 }
 
net/ipv4/netfilter/nft_redir_ipv4.c +1 -1
index d8d795df9c13..c09d4381427e 100644
@@ -36,7 +36,7 @@ static void nft_redir_ipv4_eval(const struct nft_expr *expr,
 	mr.range[0].flags |= priv->flags;
 
 	regs->verdict.code = nf_nat_redirect_ipv4(pkt->skb, &mr,
-						  pkt->ops->hooknum);
+						  pkt->hook);
 }
 
 static struct nft_expr_type nft_redir_ipv4_type;
net/ipv4/netfilter/nft_reject_ipv4.c +2 -3
index b07e58b51158..c1582e03b628 100644
@@ -27,11 +27,10 @@ static void nft_reject_ipv4_eval(const struct nft_expr *expr,
 
 	switch (priv->type) {
 	case NFT_REJECT_ICMP_UNREACH:
-		nf_send_unreach(pkt->skb, priv->icmp_code,
-				pkt->ops->hooknum);
+		nf_send_unreach(pkt->skb, priv->icmp_code, pkt->hook);
 		break;
 	case NFT_REJECT_TCP_RST:
-		nf_send_reset(pkt->skb, pkt->ops->hooknum);
+		nf_send_reset(pkt->skb, pkt->hook);
 		break;
 	default:
 		break;
net/ipv6/netfilter/nf_tables_ipv6.c +1 -1
index c8148ba76d1a..41340b794f9b 100644
@@ -23,7 +23,7 @@ static unsigned int nft_do_chain_ipv6(const struct nf_hook_ops *ops,
 	struct nft_pktinfo pkt;
 
 	/* malformed packet, drop it */
-	if (nft_set_pktinfo_ipv6(&pkt, ops, skb, state) < 0)
+	if (nft_set_pktinfo_ipv6(&pkt, skb, state) < 0)
 		return NF_DROP;
 
 	return nft_do_chain(&pkt, ops);
net/ipv6/netfilter/nft_chain_nat_ipv6.c +1 -1
index 951bb458b7bd..e96feaefeb14 100644
@@ -31,7 +31,7 @@ static unsigned int nft_nat_do_chain(const struct nf_hook_ops *ops,
 {
 	struct nft_pktinfo pkt;
 
-	nft_set_pktinfo_ipv6(&pkt, ops, skb, state);
+	nft_set_pktinfo_ipv6(&pkt, skb, state);
 
 	return nft_do_chain(&pkt, ops);
 }
net/ipv6/netfilter/nft_chain_route_ipv6.c +1 -1
index 0dafdaac5e17..d1bcd2ed7bcc 100644
@@ -33,7 +33,7 @@ static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops,
 	u32 mark, flowlabel;
 
 	/* malformed packet, drop it */
-	if (nft_set_pktinfo_ipv6(&pkt, ops, skb, state) < 0)
+	if (nft_set_pktinfo_ipv6(&pkt, skb, state) < 0)
 		return NF_DROP;
 
 	/* save source/dest address, mark, hoplimit, flowlabel, priority */
net/ipv6/netfilter/nft_dup_ipv6.c +1 -1
index 0eaa4f65fdea..c81204faf15d 100644
@@ -28,7 +28,7 @@ static void nft_dup_ipv6_eval(const struct nft_expr *expr,
 	struct in6_addr *gw = (struct in6_addr *)&regs->data[priv->sreg_addr];
 	int oif = regs->data[priv->sreg_dev];
 
-	nf_dup_ipv6(pkt->skb, pkt->ops->hooknum, gw, oif);
+	nf_dup_ipv6(pkt->skb, pkt->hook, gw, oif);
 }
 
 static int nft_dup_ipv6_init(const struct nft_ctx *ctx,
net/ipv6/netfilter/nft_redir_ipv6.c +1 -2
index effd393bd517..aca44e89a881 100644
@@ -35,8 +35,7 @@ static void nft_redir_ipv6_eval(const struct nft_expr *expr,
 
 	range.flags |= priv->flags;
 
-	regs->verdict.code = nf_nat_redirect_ipv6(pkt->skb, &range,
-						  pkt->ops->hooknum);
+	regs->verdict.code = nf_nat_redirect_ipv6(pkt->skb, &range, pkt->hook);
 }
 
 static struct nft_expr_type nft_redir_ipv6_type;
net/ipv6/netfilter/nft_reject_ipv6.c +2 -3
index d0d1540ecf87..ffcac7d5da43 100644
@@ -28,11 +28,10 @@ static void nft_reject_ipv6_eval(const struct nft_expr *expr,
 
 	switch (priv->type) {
 	case NFT_REJECT_ICMP_UNREACH:
-		nf_send_unreach6(net, pkt->skb, priv->icmp_code,
-				 pkt->ops->hooknum);
+		nf_send_unreach6(net, pkt->skb, priv->icmp_code, pkt->hook);
 		break;
 	case NFT_REJECT_TCP_RST:
-		nf_send_reset6(net, pkt->skb, pkt->ops->hooknum);
+		nf_send_reset6(net, pkt->skb, pkt->hook);
 		break;
 	default:
 		break;
net/netfilter/nf_tables_core.c +1 -1
index 05d0b03530f6..539083099c0d 100644
@@ -50,7 +50,7 @@ static void __nft_trace_packet(const struct nft_pktinfo *pkt,
 {
 	struct net *net = dev_net(pkt->in ? pkt->in : pkt->out);
 
-	nf_log_trace(net, pkt->xt.family, pkt->ops->hooknum, pkt->skb, pkt->in,
+	nf_log_trace(net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
 		     pkt->out, &trace_loginfo, "TRACE: %s:%s:%s:%u ",
 		     chain->table->name, chain->name, comments[type],
 		     rulenum);
net/netfilter/nf_tables_netdev.c +7 -9
index 2cae4d4a03b7..db416a3396e9 100644
@@ -17,13 +17,13 @@
 
 static inline void
 nft_netdev_set_pktinfo_ipv4(struct nft_pktinfo *pkt,
-			    const struct nf_hook_ops *ops, struct sk_buff *skb,
+			    struct sk_buff *skb,
 			    const struct nf_hook_state *state)
 {
 	struct iphdr *iph, _iph;
 	u32 len, thoff;
 
-	nft_set_pktinfo(pkt, ops, skb, state);
+	nft_set_pktinfo(pkt, skb, state);
 
 	iph = skb_header_pointer(skb, skb_network_offset(skb), sizeof(*iph),
 				 &_iph);
@@ -48,7 +48,6 @@ nft_netdev_set_pktinfo_ipv4(struct nft_pktinfo *pkt,
 
 static inline void
 __nft_netdev_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
-			      const struct nf_hook_ops *ops,
 			      struct sk_buff *skb,
 			      const struct nf_hook_state *state)
 {
@@ -82,12 +81,11 @@ __nft_netdev_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
 }
 
 static inline void nft_netdev_set_pktinfo_ipv6(struct nft_pktinfo *pkt,
-					       const struct nf_hook_ops *ops,
 					       struct sk_buff *skb,
 					       const struct nf_hook_state *state)
 {
-	nft_set_pktinfo(pkt, ops, skb, state);
-	__nft_netdev_set_pktinfo_ipv6(pkt, ops, skb, state);
+	nft_set_pktinfo(pkt, skb, state);
+	__nft_netdev_set_pktinfo_ipv6(pkt, skb, state);
 }
 
 static unsigned int
@@ -98,13 +96,13 @@ nft_do_chain_netdev(const struct nf_hook_ops *ops, struct sk_buff *skb,
 
 	switch (eth_hdr(skb)->h_proto) {
 	case htons(ETH_P_IP):
-		nft_netdev_set_pktinfo_ipv4(&pkt, ops, skb, state);
+		nft_netdev_set_pktinfo_ipv4(&pkt, skb, state);
 		break;
 	case htons(ETH_P_IPV6):
-		nft_netdev_set_pktinfo_ipv6(&pkt, ops, skb, state);
+		nft_netdev_set_pktinfo_ipv6(&pkt, skb, state);
 		break;
 	default:
-		nft_set_pktinfo(&pkt, ops, skb, state);
+		nft_set_pktinfo(&pkt, skb, state);
 		break;
 	}
 
net/netfilter/nft_log.c +1 -1
index a13d6a386d63..c7c7df85f0b7 100644
@@ -33,7 +33,7 @@ static void nft_log_eval(const struct nft_expr *expr,
 	const struct nft_log *priv = nft_expr_priv(expr);
 	struct net *net = dev_net(pkt->in ? pkt->in : pkt->out);
 
-	nf_log_packet(net, pkt->ops->pf, pkt->ops->hooknum, pkt->skb, pkt->in,
+	nf_log_packet(net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
 		      pkt->out, &priv->loginfo, "%s", priv->prefix);
 }
 
net/netfilter/nft_meta.c +2 -2
index cb2f13ebb5a6..e4ad2c24bc41 100644
@@ -42,7 +42,7 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 		*(__be16 *)dest = skb->protocol;
 		break;
 	case NFT_META_NFPROTO:
-		*dest = pkt->ops->pf;
+		*dest = pkt->pf;
 		break;
 	case NFT_META_L4PROTO:
 		*dest = pkt->tprot;
@@ -135,7 +135,7 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 			break;
 		}
 
-		switch (pkt->ops->pf) {
+		switch (pkt->pf) {
 		case NFPROTO_IPV4:
 			if (ipv4_is_multicast(ip_hdr(skb)->daddr))
 				*dest = PACKET_MULTICAST;
net/netfilter/nft_queue.c +1 -1
index 96805d21d618..61d216eb7917 100644
@@ -42,7 +42,7 @@ static void nft_queue_eval(const struct nft_expr *expr,
 			queue = priv->queuenum + cpu % priv->queues_total;
 		} else {
 			queue = nfqueue_hash(pkt->skb, queue,
-					     priv->queues_total, pkt->ops->pf,
+					     priv->queues_total, pkt->pf,
 					     jhash_initval);
 		}
 	}
net/netfilter/nft_reject_inet.c +7 -7
index 635dbba93d01..dea6750af6ff 100644
@@ -24,20 +24,20 @@ static void nft_reject_inet_eval(const struct nft_expr *expr,
 	struct nft_reject *priv = nft_expr_priv(expr);
 	struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out);
 
-	switch (pkt->ops->pf) {
+	switch (pkt->pf) {
 	case NFPROTO_IPV4:
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
 			nf_send_unreach(pkt->skb, priv->icmp_code,
-					pkt->ops->hooknum);
+					pkt->hook);
 			break;
 		case NFT_REJECT_TCP_RST:
-			nf_send_reset(pkt->skb, pkt->ops->hooknum);
+			nf_send_reset(pkt->skb, pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
 			nf_send_unreach(pkt->skb,
 					nft_reject_icmp_code(priv->icmp_code),
-					pkt->ops->hooknum);
+					pkt->hook);
 			break;
 		}
 		break;
@@ -45,15 +45,15 @@ static void nft_reject_inet_eval(const struct nft_expr *expr,
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
 			nf_send_unreach6(net, pkt->skb, priv->icmp_code,
-					 pkt->ops->hooknum);
+					 pkt->hook);
 			break;
 		case NFT_REJECT_TCP_RST:
-			nf_send_reset6(net, pkt->skb, pkt->ops->hooknum);
+			nf_send_reset6(net, pkt->skb, pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
 			nf_send_unreach6(net, pkt->skb,
 					 nft_reject_icmpv6_code(priv->icmp_code),
-					 pkt->ops->hooknum);
+					 pkt->hook);
 			break;
 		}
 		break;
Signature · 1 line
2.2.1
#66 lore raw

PATCH 5/15 x_tables: Pass struct net in xt_action_param

In reply to Eric W. Biederman

As xt_action_param lives on the stack this does not bloat any persistent data structures.

This is a first step in making netfilter code that needs to know which network namespace it is executing in simpler.

---
 include/linux/netfilter/x_tables.h | 3 ++-
 include/net/netfilter/nf_tables.h  | 1 +
 net/bridge/netfilter/ebtables.c    | 1 +
 net/ipv4/netfilter/arp_tables.c    | 1 +
 net/ipv4/netfilter/ip_tables.c     | 1 +
 net/ipv6/netfilter/ip6_tables.c    | 1 +
 net/sched/act_ipt.c                | 1 +
 net/sched/em_ipset.c               | 1 +
 8 files changed, 9 insertions(+), 1 deletion(-)
include/linux/netfilter/x_tables.h +2 -1
index b006b719183f..c5577410c25d 100644
@@ -13,6 +13,7 @@
  * @target:	the target extension
  * @matchinfo:	per-match data
  * @targetinfo:	per-target data
+ * @net		network namespace through which the action was invoked
  * @in:		input netdevice
  * @out:	output netdevice
  * @fragoff:	packet is a fragment, this is the data offset
@@ -24,7 +25,6 @@
  * Fields written to by extensions:
  *
  * @hotdrop:	drop packet if we had inspection problems
- * Network namespace obtainable using dev_net(in/out)
  */
 struct xt_action_param {
 	union {
@@ -34,6 +34,7 @@ struct xt_action_param {
 	union {
 		const void *matchinfo, *targinfo;
 	};
+	struct net *net;
 	const struct net_device *in, *out;
 	int fragoff;
 	unsigned int thoff;
include/net/netfilter/nf_tables.h +1 -0
index c0899f97ff8d..c0516529e8a0 100644
@@ -30,6 +30,7 @@ static inline void nft_set_pktinfo(struct nft_pktinfo *pkt,
 				   const struct nf_hook_state *state)
 {
 	pkt->skb = skb;
+	pkt->xt.net = state->net;
 	pkt->in = pkt->xt.in = state->in;
 	pkt->out = pkt->xt.out = state->out;
 	pkt->hook = pkt->xt.hooknum = state->hook;
net/bridge/netfilter/ebtables.c +1 -0
index 8d5a3975b963..f46ca417bf2d 100644
@@ -200,6 +200,7 @@ unsigned int ebt_do_table(struct sk_buff *skb,
 	struct xt_action_param acpar;
 
 	acpar.family  = NFPROTO_BRIDGE;
+	acpar.net     = state->net;
 	acpar.in      = state->in;
 	acpar.out     = state->out;
 	acpar.hotdrop = false;
net/ipv4/netfilter/arp_tables.c +1 -0
index 10eb2b297450..2dad3e1c5f11 100644
@@ -285,6 +285,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 	 */
 	e = get_entry(table_base, private->hook_entry[hook]);
 
+	acpar.net     = state->net;
 	acpar.in      = state->in;
 	acpar.out     = state->out;
 	acpar.hooknum = hook;
net/ipv4/netfilter/ip_tables.c +1 -0
index 2b049e135de8..42d0946956db 100644
@@ -315,6 +315,7 @@ ipt_do_table(struct sk_buff *skb,
 	acpar.fragoff = ntohs(ip->frag_off) & IP_OFFSET;
 	acpar.thoff   = ip_hdrlen(skb);
 	acpar.hotdrop = false;
+	acpar.net     = state->net;
 	acpar.in      = state->in;
 	acpar.out     = state->out;
 	acpar.family  = NFPROTO_IPV4;
net/ipv6/netfilter/ip6_tables.c +1 -0
index da6446b6e3f9..80e3bd72b715 100644
@@ -340,6 +340,7 @@ ip6t_do_table(struct sk_buff *skb,
 	 * rule is also a fragment-specific rule, non-fragments won't
 	 * match it. */
 	acpar.hotdrop = false;
+	acpar.net     = state->net;
 	acpar.in      = state->in;
 	acpar.out     = state->out;
 	acpar.family  = NFPROTO_IPV6;
net/sched/act_ipt.c +1 -0
index 99c9cc1c7af9..d05869646515 100644
@@ -189,6 +189,7 @@ static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a,
 	 * worry later - danger - this API seems to have changed
 	 * from earlier kernels
 	 */
+	par.net	     = dev_net(skb->dev);
 	par.in       = skb->dev;
 	par.out      = NULL;
 	par.hooknum  = ipt->tcfi_hook;
net/sched/em_ipset.c +1 -0
index df0328ba6a48..c66ca9400ab4 100644
@@ -95,6 +95,7 @@ static int em_ipset_match(struct sk_buff *skb, struct tcf_ematch *em,
 	if (skb->skb_iif)
 		indev = dev_get_by_index_rcu(em->net, skb->skb_iif);
 
+	acpar.net     = em->net;
 	acpar.in      = indev ? indev : dev;
 	acpar.out     = dev;
 
Signature · 1 line
2.2.1
#67 lore raw

PATCH 6/15 x_tables: Use par->net instead of computing from the passed net devices

In reply to Eric W. Biederman
---
 net/bridge/netfilter/ebt_log.c     |  2 +-
 net/bridge/netfilter/ebt_nflog.c   |  2 +-
 net/ipv4/netfilter/ipt_SYNPROXY.c  |  2 +-
 net/ipv4/netfilter/ipt_rpfilter.c  |  5 ++---
 net/ipv6/netfilter/ip6t_REJECT.c   |  2 +-
 net/ipv6/netfilter/ip6t_SYNPROXY.c |  2 +-
 net/ipv6/netfilter/ip6t_rpfilter.c |  6 +++---
 net/netfilter/ipset/ip_set_core.c  |  9 +++------
 net/netfilter/xt_LOG.c             |  2 +-
 net/netfilter/xt_NFLOG.c           |  2 +-
 net/netfilter/xt_TCPMSS.c          |  2 +-
 net/netfilter/xt_TPROXY.c          | 24 ++++++++++++------------
 net/netfilter/xt_addrtype.c        |  4 ++--
 net/netfilter/xt_connlimit.c       |  2 +-
 net/netfilter/xt_osf.c             |  2 +-
 net/netfilter/xt_recent.c          |  2 +-
 net/netfilter/xt_socket.c          | 14 ++++++++------
 17 files changed, 41 insertions(+), 43 deletions(-)
net/bridge/netfilter/ebt_log.c +1 -1
index 17f2e4bc2a29..0ad639a96142 100644
@@ -180,7 +180,7 @@ ebt_log_tg(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct ebt_log_info *info = par->targinfo;
 	struct nf_loginfo li;
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 
 	li.type = NF_LOG_TYPE_LOG;
 	li.u.log.level = info->loglevel;
net/bridge/netfilter/ebt_nflog.c +1 -1
index 59ac7952010d..54816150608e 100644
@@ -24,7 +24,7 @@ ebt_nflog_tg(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct ebt_nflog_info *info = par->targinfo;
 	struct nf_loginfo li;
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 
 	li.type = NF_LOG_TYPE_ULOG;
 	li.u.ulog.copy_len = info->len;
net/ipv4/netfilter/ipt_SYNPROXY.c +1 -1
index f471a0628c75..dfab314981e9 100644
@@ -258,7 +258,7 @@ static unsigned int
 synproxy_tg4(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_synproxy_info *info = par->targinfo;
-	struct synproxy_net *snet = synproxy_pernet(dev_net(par->in));
+	struct synproxy_net *snet = synproxy_pernet(par->net);
 	struct synproxy_options opts = {};
 	struct tcphdr *th, _th;
 
net/ipv4/netfilter/ipt_rpfilter.c +2 -3
index 8618fd150c96..74dd6671b66d 100644
@@ -32,12 +32,11 @@ static __be32 rpfilter_get_saddr(__be32 addr)
 	return addr;
 }
 
-static bool rpfilter_lookup_reverse(struct flowi4 *fl4,
+static bool rpfilter_lookup_reverse(struct net *net, struct flowi4 *fl4,
 				const struct net_device *dev, u8 flags)
 {
 	struct fib_result res;
 	bool dev_match;
-	struct net *net = dev_net(dev);
 	int ret __maybe_unused;
 
 	if (fib_lookup(net, fl4, &res, FIB_LOOKUP_IGNORE_LINKSTATE))
@@ -98,7 +97,7 @@ static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par)
 	flow.flowi4_tos = RT_TOS(iph->tos);
 	flow.flowi4_scope = RT_SCOPE_UNIVERSE;
 
-	return rpfilter_lookup_reverse(&flow, par->in, info->flags) ^ invert;
+	return rpfilter_lookup_reverse(par->net, &flow, par->in, info->flags) ^ invert;
 }
 
 static int rpfilter_check(const struct xt_mtchk_param *par)
net/ipv6/netfilter/ip6t_REJECT.c +1 -1
index 0ed841a3fa33..db29bbf41b59 100644
@@ -39,7 +39,7 @@ static unsigned int
 reject_tg6(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct ip6t_reject_info *reject = par->targinfo;
-	struct net *net = dev_net((par->in != NULL) ? par->in : par->out);
+	struct net *net = par->net;
 
 	switch (reject->with) {
 	case IP6T_ICMP6_NO_ROUTE:
net/ipv6/netfilter/ip6t_SYNPROXY.c +1 -1
index 4c9f3e79d75f..41451809b37c 100644
@@ -275,7 +275,7 @@ static unsigned int
 synproxy_tg6(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_synproxy_info *info = par->targinfo;
-	struct synproxy_net *snet = synproxy_pernet(dev_net(par->in));
+	struct synproxy_net *snet = synproxy_pernet(par->net);
 	struct synproxy_options opts = {};
 	struct tcphdr *th, _th;
 
net/ipv6/netfilter/ip6t_rpfilter.c +3 -3
index 790e0c6b19e1..1ee1b25df096 100644
@@ -26,7 +26,7 @@ static bool rpfilter_addr_unicast(const struct in6_addr *addr)
 	return addr_type & IPV6_ADDR_UNICAST;
 }
 
-static bool rpfilter_lookup_reverse6(const struct sk_buff *skb,
+static bool rpfilter_lookup_reverse6(struct net *net, const struct sk_buff *skb,
 				     const struct net_device *dev, u8 flags)
 {
 	struct rt6_info *rt;
@@ -53,7 +53,7 @@ static bool rpfilter_lookup_reverse6(const struct sk_buff *skb,
 		lookup_flags |= RT6_LOOKUP_F_IFACE;
 	}
 
-	rt = (void *) ip6_route_lookup(dev_net(dev), &fl6, lookup_flags);
+	rt = (void *) ip6_route_lookup(net, &fl6, lookup_flags);
 	if (rt->dst.error)
 		goto out;
 
@@ -93,7 +93,7 @@ static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par)
 	if (unlikely(saddrtype == IPV6_ADDR_ANY))
 		return true ^ invert; /* not routable: forward path will drop it */
 
-	return rpfilter_lookup_reverse6(skb, par->in, info->flags) ^ invert;
+	return rpfilter_lookup_reverse6(par->net, skb, par->in, info->flags) ^ invert;
 }
 
 static int rpfilter_check(const struct xt_mtchk_param *par)
net/netfilter/ipset/ip_set_core.c +3 -6
index 338b4047776f..69ab9c2634e1 100644
@@ -519,8 +519,7 @@ int
 ip_set_test(ip_set_id_t index, const struct sk_buff *skb,
 	    const struct xt_action_param *par, struct ip_set_adt_opt *opt)
 {
-	struct ip_set *set = ip_set_rcu_get(
-			dev_net(par->in ? par->in : par->out), index);
+	struct ip_set *set = ip_set_rcu_get(par->net, index);
 	int ret = 0;
 
 	BUG_ON(!set);
@@ -558,8 +557,7 @@ int
 ip_set_add(ip_set_id_t index, const struct sk_buff *skb,
 	   const struct xt_action_param *par, struct ip_set_adt_opt *opt)
 {
-	struct ip_set *set = ip_set_rcu_get(
-			dev_net(par->in ? par->in : par->out), index);
+	struct ip_set *set = ip_set_rcu_get(par->net, index);
 	int ret;
 
 	BUG_ON(!set);
@@ -581,8 +579,7 @@ int
 ip_set_del(ip_set_id_t index, const struct sk_buff *skb,
 	   const struct xt_action_param *par, struct ip_set_adt_opt *opt)
 {
-	struct ip_set *set = ip_set_rcu_get(
-			dev_net(par->in ? par->in : par->out), index);
+	struct ip_set *set = ip_set_rcu_get(par->net, index);
 	int ret = 0;
 
 	BUG_ON(!set);
net/netfilter/xt_LOG.c +1 -1
index c13b79440ede..1763ab82bcd7 100644
@@ -33,7 +33,7 @@ log_tg(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_log_info *loginfo = par->targinfo;
 	struct nf_loginfo li;
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 
 	li.type = NF_LOG_TYPE_LOG;
 	li.u.log.level = loginfo->level;
net/netfilter/xt_NFLOG.c +1 -1
index fb7497c928a0..a1fa2c800cb9 100644
@@ -26,7 +26,7 @@ nflog_tg(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_nflog_info *info = par->targinfo;
 	struct nf_loginfo li;
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 
 	li.type		     = NF_LOG_TYPE_ULOG;
 	li.u.ulog.copy_len   = info->len;
net/netfilter/xt_TCPMSS.c +1 -1
index 8c02501a530f..b7c43def0dc6 100644
@@ -108,7 +108,7 @@ tcpmss_mangle_packet(struct sk_buff *skb,
 		return -1;
 
 	if (info->mss == XT_TCPMSS_CLAMP_PMTU) {
-		struct net *net = dev_net(par->in ? par->in : par->out);
+		struct net *net = par->net;
 		unsigned int in_mtu = tcpmss_reverse_mtu(net, skb, family);
 
 		if (dst_mtu(skb_dst(skb)) <= minlen) {
net/netfilter/xt_TPROXY.c +12 -12
index d0c96c5ae29a..3ab591e73ec0 100644
@@ -250,8 +250,8 @@ nf_tproxy_get_sock_v6(struct net *net, const u8 protocol,
  * no such listener is found, or NULL if the TCP header is incomplete.
  */
 static struct sock *
-tproxy_handle_time_wait4(struct sk_buff *skb, __be32 laddr, __be16 lport,
-			struct sock *sk)
+tproxy_handle_time_wait4(struct net *net, struct sk_buff *skb,
+			 __be32 laddr, __be16 lport, struct sock *sk)
 {
 	const struct iphdr *iph = ip_hdr(skb);
 	struct tcphdr _hdr, *hp;
@@ -267,7 +267,7 @@ tproxy_handle_time_wait4(struct sk_buff *skb, __be32 laddr, __be16 lport,
 		 * to a listener socket if there's one */
 		struct sock *sk2;
 
-		sk2 = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol,
+		sk2 = nf_tproxy_get_sock_v4(net, iph->protocol,
 					    iph->saddr, laddr ? laddr : iph->daddr,
 					    hp->source, lport ? lport : hp->dest,
 					    skb->dev, NFT_LOOKUP_LISTENER);
@@ -290,7 +290,7 @@ nf_tproxy_assign_sock(struct sk_buff *skb, struct sock *sk)
 }
 
 static unsigned int
-tproxy_tg4(struct sk_buff *skb, __be32 laddr, __be16 lport,
+tproxy_tg4(struct net *net, struct sk_buff *skb, __be32 laddr, __be16 lport,
 	   u_int32_t mark_mask, u_int32_t mark_value)
 {
 	const struct iphdr *iph = ip_hdr(skb);
@@ -305,7 +305,7 @@ tproxy_tg4(struct sk_buff *skb, __be32 laddr, __be16 lport,
 	 * addresses, this happens if the redirect already happened
 	 * and the current packet belongs to an already established
 	 * connection */
-	sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol,
+	sk = nf_tproxy_get_sock_v4(net, iph->protocol,
 				   iph->saddr, iph->daddr,
 				   hp->source, hp->dest,
 				   skb->dev, NFT_LOOKUP_ESTABLISHED);
@@ -317,11 +317,11 @@ tproxy_tg4(struct sk_buff *skb, __be32 laddr, __be16 lport,
 	/* UDP has no TCP_TIME_WAIT state, so we never enter here */
 	if (sk && sk->sk_state == TCP_TIME_WAIT)
 		/* reopening a TIME_WAIT connection needs special handling */
-		sk = tproxy_handle_time_wait4(skb, laddr, lport, sk);
+		sk = tproxy_handle_time_wait4(net, skb, laddr, lport, sk);
 	else if (!sk)
 		/* no, there's no established connection, check if
 		 * there's a listener on the redirected addr/port */
-		sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol,
+		sk = nf_tproxy_get_sock_v4(net, iph->protocol,
 					   iph->saddr, laddr,
 					   hp->source, lport,
 					   skb->dev, NFT_LOOKUP_LISTENER);
@@ -351,7 +351,7 @@ tproxy_tg4_v0(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_tproxy_target_info *tgi = par->targinfo;
 
-	return tproxy_tg4(skb, tgi->laddr, tgi->lport, tgi->mark_mask, tgi->mark_value);
+	return tproxy_tg4(par->net, skb, tgi->laddr, tgi->lport, tgi->mark_mask, tgi->mark_value);
 }
 
 static unsigned int
@@ -359,7 +359,7 @@ tproxy_tg4_v1(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_tproxy_target_info_v1 *tgi = par->targinfo;
 
-	return tproxy_tg4(skb, tgi->laddr.ip, tgi->lport, tgi->mark_mask, tgi->mark_value);
+	return tproxy_tg4(par->net, skb, tgi->laddr.ip, tgi->lport, tgi->mark_mask, tgi->mark_value);
 }
 
 #ifdef XT_TPROXY_HAVE_IPV6
@@ -429,7 +429,7 @@ tproxy_handle_time_wait6(struct sk_buff *skb, int tproto, int thoff,
 		 * to a listener socket if there's one */
 		struct sock *sk2;
 
-		sk2 = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto,
+		sk2 = nf_tproxy_get_sock_v6(par->net, tproto,
 					    &iph->saddr,
 					    tproxy_laddr6(skb, &tgi->laddr.in6, &iph->daddr),
 					    hp->source,
@@ -472,7 +472,7 @@ tproxy_tg6_v1(struct sk_buff *skb, const struct xt_action_param *par)
 	 * addresses, this happens if the redirect already happened
 	 * and the current packet belongs to an already established
 	 * connection */
-	sk = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto,
+	sk = nf_tproxy_get_sock_v6(par->net, tproto,
 				   &iph->saddr, &iph->daddr,
 				   hp->source, hp->dest,
 				   par->in, NFT_LOOKUP_ESTABLISHED);
@@ -487,7 +487,7 @@ tproxy_tg6_v1(struct sk_buff *skb, const struct xt_action_param *par)
 	else if (!sk)
 		/* no there's no established connection, check if
 		 * there's a listener on the redirected addr/port */
-		sk = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto,
+		sk = nf_tproxy_get_sock_v6(par->net, tproto,
 					   &iph->saddr, laddr,
 					   hp->source, lport,
 					   par->in, NFT_LOOKUP_LISTENER);
net/netfilter/xt_addrtype.c +2 -2
index 5b4743cc0436..11d6091991a4 100644
@@ -125,7 +125,7 @@ static inline bool match_type(struct net *net, const struct net_device *dev,
 static bool
 addrtype_mt_v0(const struct sk_buff *skb, struct xt_action_param *par)
 {
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 	const struct xt_addrtype_info *info = par->matchinfo;
 	const struct iphdr *iph = ip_hdr(skb);
 	bool ret = true;
@@ -143,7 +143,7 @@ addrtype_mt_v0(const struct sk_buff *skb, struct xt_action_param *par)
 static bool
 addrtype_mt_v1(const struct sk_buff *skb, struct xt_action_param *par)
 {
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 	const struct xt_addrtype_info_v1 *info = par->matchinfo;
 	const struct iphdr *iph;
 	const struct net_device *dev = NULL;
net/netfilter/xt_connlimit.c +1 -1
index 075d89d94d28..213db252e5be 100644
@@ -317,7 +317,7 @@ static int count_them(struct net *net,
 static bool
 connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par)
 {
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 	const struct xt_connlimit_info *info = par->matchinfo;
 	union nf_inet_addr addr;
 	struct nf_conntrack_tuple tuple;
net/netfilter/xt_osf.c +1 -1
index 0778855ea5e7..df8801e02a32 100644
@@ -200,7 +200,7 @@ xt_osf_match_packet(const struct sk_buff *skb, struct xt_action_param *p)
 	unsigned char opts[MAX_IPOPTLEN];
 	const struct xt_osf_finger *kf;
 	const struct xt_osf_user_finger *f;
-	struct net *net = dev_net(p->in ? p->in : p->out);
+	struct net *net = p->net;
 
 	if (!info)
 		return false;
net/netfilter/xt_recent.c +1 -1
index 45e1b30e4fb2..d725a27743a1 100644
@@ -237,7 +237,7 @@ static void recent_table_flush(struct recent_table *t)
 static bool
 recent_mt(const struct sk_buff *skb, struct xt_action_param *par)
 {
-	struct net *net = dev_net(par->in ? par->in : par->out);
+	struct net *net = par->net;
 	struct recent_net *recent_net = recent_pernet(net);
 	const struct xt_recent_mtinfo_v1 *info = par->matchinfo;
 	struct recent_table *t;
net/netfilter/xt_socket.c +8 -6
index 43e26c881100..2ec08f04b816 100644
@@ -143,7 +143,8 @@ static bool xt_socket_sk_is_transparent(struct sock *sk)
 	}
 }
 
-static struct sock *xt_socket_lookup_slow_v4(const struct sk_buff *skb,
+static struct sock *xt_socket_lookup_slow_v4(struct net *net,
+					     const struct sk_buff *skb,
 					     const struct net_device *indev)
 {
 	const struct iphdr *iph = ip_hdr(skb);
@@ -197,7 +198,7 @@ static struct sock *xt_socket_lookup_slow_v4(const struct sk_buff *skb,
 	}
 #endif
 
-	return xt_socket_get_sock_v4(dev_net(skb->dev), protocol, saddr, daddr,
+	return xt_socket_get_sock_v4(net, protocol, saddr, daddr,
 				     sport, dport, indev);
 }
 
@@ -209,7 +210,7 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par,
 	struct sock *sk = skb->sk;
 
 	if (!sk)
-		sk = xt_socket_lookup_slow_v4(skb, par->in);
+		sk = xt_socket_lookup_slow_v4(par->net, skb, par->in);
 	if (sk) {
 		bool wildcard;
 		bool transparent = true;
@@ -335,7 +336,8 @@ xt_socket_get_sock_v6(struct net *net, const u8 protocol,
 	return NULL;
 }
 
-static struct sock *xt_socket_lookup_slow_v6(const struct sk_buff *skb,
+static struct sock *xt_socket_lookup_slow_v6(struct net *net,
+					     const struct sk_buff *skb,
 					     const struct net_device *indev)
 {
 	__be16 uninitialized_var(dport), uninitialized_var(sport);
@@ -371,7 +373,7 @@ static struct sock *xt_socket_lookup_slow_v6(const struct sk_buff *skb,
 		return NULL;
 	}
 
-	return xt_socket_get_sock_v6(dev_net(skb->dev), tproto, saddr, daddr,
+	return xt_socket_get_sock_v6(net, tproto, saddr, daddr,
 				     sport, dport, indev);
 }
 
@@ -383,7 +385,7 @@ socket_mt6_v1_v2_v3(const struct sk_buff *skb, struct xt_action_param *par)
 	struct sock *sk = skb->sk;
 
 	if (!sk)
-		sk = xt_socket_lookup_slow_v6(skb, par->in);
+		sk = xt_socket_lookup_slow_v6(par->net, skb, par->in);
 	if (sk) {
 		bool wildcard;
 		bool transparent = true;
Signature · 1 line
2.2.1
#68 lore raw

PATCH 7/15 nftables: Pass struct net in nft_pktinfo

In reply to Eric W. Biederman

nft_pktinfo is passed on the stack so this does not bloat any in core data structures.

By centrally computing this information this makes maintence of the code simpler, and understading of the code easier.

---
 include/net/netfilter/nf_tables.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
include/net/netfilter/nf_tables.h +2 -1
index c0516529e8a0..42e239e55aa3 100644
@@ -14,6 +14,7 @@
 
 struct nft_pktinfo {
 	struct sk_buff			*skb;
+	struct net			*net;
 	const struct net_device		*in;
 	const struct net_device		*out;
 	u8				pf;
@@ -30,7 +31,7 @@ static inline void nft_set_pktinfo(struct nft_pktinfo *pkt,
 				   const struct nf_hook_state *state)
 {
 	pkt->skb = skb;
-	pkt->xt.net = state->net;
+	pkt->net = pkt->xt.net = state->net;
 	pkt->in = pkt->xt.in = state->in;
 	pkt->out = pkt->xt.out = state->out;
 	pkt->hook = pkt->xt.hooknum = state->hook;
Signature · 1 line
2.2.1
#69 lore raw

PATCH 8/15 nftables: Use pkt->net instead of computing net from the passed net_devices

In reply to Eric W. Biederman
---
 net/bridge/netfilter/nft_reject_bridge.c | 13 ++++++-------
 net/ipv6/netfilter/nft_reject_ipv6.c     |  6 +++---
 net/netfilter/nf_tables_core.c           |  6 ++----
 net/netfilter/nft_log.c                  |  3 +--
 net/netfilter/nft_reject_inet.c          |  7 +++----
 5 files changed, 15 insertions(+), 20 deletions(-)
net/bridge/netfilter/nft_reject_bridge.c +6 -7
index cee92612b2cc..fdba3d9fbff3 100644
@@ -261,7 +261,6 @@ static void nft_reject_bridge_eval(const struct nft_expr *expr,
 				   const struct nft_pktinfo *pkt)
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
-	struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out);
 	const unsigned char *dest = eth_hdr(pkt->skb)->h_dest;
 
 	if (is_broadcast_ether_addr(dest) ||
@@ -290,17 +289,17 @@ static void nft_reject_bridge_eval(const struct nft_expr *expr,
 	case htons(ETH_P_IPV6):
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
-			nft_reject_br_send_v6_unreach(net, pkt->skb, pkt->in,
-						      pkt->hook,
+			nft_reject_br_send_v6_unreach(pkt->net, pkt->skb,
+						      pkt->in, pkt->hook,
 						      priv->icmp_code);
 			break;
 		case NFT_REJECT_TCP_RST:
-			nft_reject_br_send_v6_tcp_reset(net, pkt->skb, pkt->in,
-							pkt->hook);
+			nft_reject_br_send_v6_tcp_reset(pkt->net, pkt->skb,
+							pkt->in, pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
-			nft_reject_br_send_v6_unreach(net, pkt->skb, pkt->in,
-						      pkt->hook,
+			nft_reject_br_send_v6_unreach(pkt->net, pkt->skb,
+						      pkt->in, pkt->hook,
 						      nft_reject_icmpv6_code(priv->icmp_code));
 			break;
 		}
net/ipv6/netfilter/nft_reject_ipv6.c +3 -3
index ffcac7d5da43..533cd5719c59 100644
@@ -24,14 +24,14 @@ static void nft_reject_ipv6_eval(const struct nft_expr *expr,
 				 const struct nft_pktinfo *pkt)
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
-	struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out);
 
 	switch (priv->type) {
 	case NFT_REJECT_ICMP_UNREACH:
-		nf_send_unreach6(net, pkt->skb, priv->icmp_code, pkt->hook);
+		nf_send_unreach6(pkt->net, pkt->skb, priv->icmp_code,
+				 pkt->hook);
 		break;
 	case NFT_REJECT_TCP_RST:
-		nf_send_reset6(net, pkt->skb, pkt->hook);
+		nf_send_reset6(pkt->net, pkt->skb, pkt->hook);
 		break;
 	default:
 		break;
net/netfilter/nf_tables_core.c +2 -4
index 539083099c0d..e5c1f332e45e 100644
@@ -48,9 +48,7 @@ static void __nft_trace_packet(const struct nft_pktinfo *pkt,
 			       const struct nft_chain *chain,
 			       int rulenum, enum nft_trace type)
 {
-	struct net *net = dev_net(pkt->in ? pkt->in : pkt->out);
-
-	nf_log_trace(net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
+	nf_log_trace(pkt->net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
 		     pkt->out, &trace_loginfo, "TRACE: %s:%s:%s:%u ",
 		     chain->table->name, chain->name, comments[type],
 		     rulenum);
@@ -114,7 +112,7 @@ unsigned int
 nft_do_chain(struct nft_pktinfo *pkt, const struct nf_hook_ops *ops)
 {
 	const struct nft_chain *chain = ops->priv, *basechain = chain;
-	const struct net *net = dev_net(pkt->in ? pkt->in : pkt->out);
+	const struct net *net = pkt->net;
 	const struct nft_rule *rule;
 	const struct nft_expr *expr, *last;
 	struct nft_regs regs;
net/netfilter/nft_log.c +1 -2
index c7c7df85f0b7..319c22b4bca2 100644
@@ -31,9 +31,8 @@ static void nft_log_eval(const struct nft_expr *expr,
 			 const struct nft_pktinfo *pkt)
 {
 	const struct nft_log *priv = nft_expr_priv(expr);
-	struct net *net = dev_net(pkt->in ? pkt->in : pkt->out);
 
-	nf_log_packet(net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
+	nf_log_packet(pkt->net, pkt->pf, pkt->hook, pkt->skb, pkt->in,
 		      pkt->out, &priv->loginfo, "%s", priv->prefix);
 }
 
net/netfilter/nft_reject_inet.c +3 -4
index dea6750af6ff..0bc19f97e238 100644
@@ -22,7 +22,6 @@ static void nft_reject_inet_eval(const struct nft_expr *expr,
 				 const struct nft_pktinfo *pkt)
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
-	struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out);
 
 	switch (pkt->pf) {
 	case NFPROTO_IPV4:
@@ -44,14 +43,14 @@ static void nft_reject_inet_eval(const struct nft_expr *expr,
 	case NFPROTO_IPV6:
 		switch (priv->type) {
 		case NFT_REJECT_ICMP_UNREACH:
-			nf_send_unreach6(net, pkt->skb, priv->icmp_code,
+			nf_send_unreach6(pkt->net, pkt->skb, priv->icmp_code,
 					 pkt->hook);
 			break;
 		case NFT_REJECT_TCP_RST:
-			nf_send_reset6(net, pkt->skb, pkt->hook);
+			nf_send_reset6(pkt->net, pkt->skb, pkt->hook);
 			break;
 		case NFT_REJECT_ICMPX_UNREACH:
-			nf_send_unreach6(net, pkt->skb,
+			nf_send_unreach6(pkt->net, pkt->skb,
 					 nft_reject_icmpv6_code(priv->icmp_code),
 					 pkt->hook);
 			break;
Signature · 1 line
2.2.1
#70 lore raw

PATCH 9/15 net: Pass net to nf_dup_ipv4 and nf_dup_ipv6

In reply to Eric W. Biederman

This allows them to stop guessing the network namespace with pick_net.

---
 include/net/netfilter/ipv4/nf_dup_ipv4.h |  2 +-
 include/net/netfilter/ipv6/nf_dup_ipv6.h |  2 +-
 net/ipv4/netfilter/nf_dup_ipv4.c         | 23 ++++-------------------
 net/ipv4/netfilter/nft_dup_ipv4.c        |  2 +-
 net/ipv6/netfilter/nf_dup_ipv6.c         | 23 ++++-------------------
 net/ipv6/netfilter/nft_dup_ipv6.c        |  2 +-
 net/netfilter/xt_TEE.c                   |  4 ++--
 7 files changed, 14 insertions(+), 44 deletions(-)
include/net/netfilter/ipv4/nf_dup_ipv4.h +1 -1
index 42008f10dfc4..0a14733e8b82 100644
@@ -1,7 +1,7 @@
 #ifndef _NF_DUP_IPV4_H_
 #define _NF_DUP_IPV4_H_
 
-void nf_dup_ipv4(struct sk_buff *skb, unsigned int hooknum,
+void nf_dup_ipv4(struct net *net, struct sk_buff *skb, unsigned int hooknum,
 		 const struct in_addr *gw, int oif);
 
 #endif /* _NF_DUP_IPV4_H_ */
include/net/netfilter/ipv6/nf_dup_ipv6.h +1 -1
index ed6bd66fa5a0..fa6237b382a3 100644
@@ -1,7 +1,7 @@
 #ifndef _NF_DUP_IPV6_H_
 #define _NF_DUP_IPV6_H_
 
-void nf_dup_ipv6(struct sk_buff *skb, unsigned int hooknum,
+void nf_dup_ipv6(struct net *net, struct sk_buff *skb, unsigned int hooknum,
 		 const struct in6_addr *gw, int oif);
 
 #endif /* _NF_DUP_IPV6_H_ */
net/ipv4/netfilter/nf_dup_ipv4.c +4 -19
index 2d79e6e8d934..ce2a59e5c665 100644
@@ -23,25 +23,10 @@
 #include <net/netfilter/nf_conntrack.h>
 #endif
 
-static struct net *pick_net(struct sk_buff *skb)
-{
-#ifdef CONFIG_NET_NS
-	const struct dst_entry *dst;
-
-	if (skb->dev != NULL)
-		return dev_net(skb->dev);
-	dst = skb_dst(skb);
-	if (dst != NULL && dst->dev != NULL)
-		return dev_net(dst->dev);
-#endif
-	return &init_net;
-}
-
-static bool nf_dup_ipv4_route(struct sk_buff *skb, const struct in_addr *gw,
-			      int oif)
+static bool nf_dup_ipv4_route(struct net *net, struct sk_buff *skb,
+			      const struct in_addr *gw, int oif)
 {
 	const struct iphdr *iph = ip_hdr(skb);
-	struct net *net = pick_net(skb);
 	struct rtable *rt;
 	struct flowi4 fl4;
 
@@ -65,7 +50,7 @@ static bool nf_dup_ipv4_route(struct sk_buff *skb, const struct in_addr *gw,
 	return true;
 }
 
-void nf_dup_ipv4(struct sk_buff *skb, unsigned int hooknum,
+void nf_dup_ipv4(struct net *net, struct sk_buff *skb, unsigned int hooknum,
 		 const struct in_addr *gw, int oif)
 {
 	struct iphdr *iph;
@@ -105,7 +90,7 @@ void nf_dup_ipv4(struct sk_buff *skb, unsigned int hooknum,
 		--iph->ttl;
 	ip_send_check(iph);
 
-	if (nf_dup_ipv4_route(skb, gw, oif)) {
+	if (nf_dup_ipv4_route(net, skb, gw, oif)) {
 		__this_cpu_write(nf_skb_duplicated, true);
 		ip_local_out(skb);
 		__this_cpu_write(nf_skb_duplicated, false);
net/ipv4/netfilter/nft_dup_ipv4.c +1 -1
index 30bcf820e8bd..bf855e64fc45 100644
@@ -30,7 +30,7 @@ static void nft_dup_ipv4_eval(const struct nft_expr *expr,
 	};
 	int oif = regs->data[priv->sreg_dev];
 
-	nf_dup_ipv4(pkt->skb, pkt->hook, &gw, oif);
+	nf_dup_ipv4(pkt->net, pkt->skb, pkt->hook, &gw, oif);
 }
 
 static int nft_dup_ipv4_init(const struct nft_ctx *ctx,
net/ipv6/netfilter/nf_dup_ipv6.c +4 -19
index c8ab626556a0..ee0d9a5b16c3 100644
@@ -19,25 +19,10 @@
 #include <net/netfilter/nf_conntrack.h>
 #endif
 
-static struct net *pick_net(struct sk_buff *skb)
-{
-#ifdef CONFIG_NET_NS
-	const struct dst_entry *dst;
-
-	if (skb->dev != NULL)
-		return dev_net(skb->dev);
-	dst = skb_dst(skb);
-	if (dst != NULL && dst->dev != NULL)
-		return dev_net(dst->dev);
-#endif
-	return &init_net;
-}
-
-static bool nf_dup_ipv6_route(struct sk_buff *skb, const struct in6_addr *gw,
-			      int oif)
+static bool nf_dup_ipv6_route(struct net *net, struct sk_buff *skb,
+			      const struct in6_addr *gw, int oif)
 {
 	const struct ipv6hdr *iph = ipv6_hdr(skb);
-	struct net *net = pick_net(skb);
 	struct dst_entry *dst;
 	struct flowi6 fl6;
 
@@ -61,7 +46,7 @@ static bool nf_dup_ipv6_route(struct sk_buff *skb, const struct in6_addr *gw,
 	return true;
 }
 
-void nf_dup_ipv6(struct sk_buff *skb, unsigned int hooknum,
+void nf_dup_ipv6(struct net *net, struct sk_buff *skb, unsigned int hooknum,
 		 const struct in6_addr *gw, int oif)
 {
 	if (this_cpu_read(nf_skb_duplicated))
@@ -81,7 +66,7 @@ void nf_dup_ipv6(struct sk_buff *skb, unsigned int hooknum,
 		struct ipv6hdr *iph = ipv6_hdr(skb);
 		--iph->hop_limit;
 	}
-	if (nf_dup_ipv6_route(skb, gw, oif)) {
+	if (nf_dup_ipv6_route(net, skb, gw, oif)) {
 		__this_cpu_write(nf_skb_duplicated, true);
 		ip6_local_out(skb);
 		__this_cpu_write(nf_skb_duplicated, false);
net/ipv6/netfilter/nft_dup_ipv6.c +1 -1
index c81204faf15d..8bfd470cbe72 100644
@@ -28,7 +28,7 @@ static void nft_dup_ipv6_eval(const struct nft_expr *expr,
 	struct in6_addr *gw = (struct in6_addr *)&regs->data[priv->sreg_addr];
 	int oif = regs->data[priv->sreg_dev];
 
-	nf_dup_ipv6(pkt->skb, pkt->hook, gw, oif);
+	nf_dup_ipv6(pkt->net, pkt->skb, pkt->hook, gw, oif);
 }
 
 static int nft_dup_ipv6_init(const struct nft_ctx *ctx,
net/netfilter/xt_TEE.c +2 -2
index fd980aa7715d..899b06115fc5 100644
@@ -32,7 +32,7 @@ tee_tg4(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_tee_tginfo *info = par->targinfo;
 
-	nf_dup_ipv4(skb, par->hooknum, &info->gw.in, info->priv->oif);
+	nf_dup_ipv4(par->net, skb, par->hooknum, &info->gw.in, info->priv->oif);
 
 	return XT_CONTINUE;
 }
@@ -43,7 +43,7 @@ tee_tg6(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	const struct xt_tee_tginfo *info = par->targinfo;
 
-	nf_dup_ipv6(skb, par->hooknum, &info->gw.in6, info->priv->oif);
+	nf_dup_ipv6(par->net, skb, par->hooknum, &info->gw.in6, info->priv->oif);
 
 	return XT_CONTINUE;
 }
Signature · 1 line
2.2.1
#71 lore raw

PATCH 10/15 act_connmark: Remember the struct net instead of guessing it.

In reply to Eric W. Biederman

Stop guessing the struct net instead of remember it. Guessing is just silly and will be problematic in the future when I implement routes between network namespaces.

---
 include/net/tc_act/tc_connmark.h | 1 +
 net/sched/act_connmark.c         | 3 ++-
 2 files changed, 3 insertions(+), 1 deletion(-)
include/net/tc_act/tc_connmark.h +1 -0
index 5c1104c2e24f..02caa406611b 100644
@@ -5,6 +5,7 @@
 
 struct tcf_connmark_info {
 	struct tcf_common common;
+	struct net *net;
 	u16 zone;
 };
 
net/sched/act_connmark.c +2 -1
index 5019a47b9270..413ac39147d8 100644
@@ -74,7 +74,7 @@ static int tcf_connmark(struct sk_buff *skb, const struct tc_action *a,
 	zone.id = ca->zone;
 	zone.dir = NF_CT_DEFAULT_ZONE_DIR;
 
-	thash = nf_conntrack_find_get(dev_net(skb->dev), &zone, &tuple);
+	thash = nf_conntrack_find_get(ca->net, &zone, &tuple);
 	if (!thash)
 		goto out;
 
@@ -119,6 +119,7 @@ static int tcf_connmark_init(struct net *net, struct nlattr *nla,
 
 		ci = to_connmark(a);
 		ci->tcf_action = parm->action;
+		ci->net = net;
 		ci->zone = parm->zone;
 
 		tcf_hash_insert(a);
Signature · 1 line
2.2.1
#72 lore raw

PATCH 11/15 nf_conntrack: Add a struct net parameter to l4_pkt_to_tuple

In reply to Eric W. Biederman

As gre does not have the srckey in the packet gre_pkt_to_tuple needs to perform a lookup in it's per network namespace tables.

Pass in the proper network namespace to all pkt_to_tuple implementations to ensure gre (and any similar protocols) can get this right.

---
 include/net/netfilter/nf_conntrack.h           |  3 ++-
 include/net/netfilter/nf_conntrack_core.h      |  1 +
 include/net/netfilter/nf_conntrack_l4proto.h   |  2 +-
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   |  4 ++--
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c |  3 ++-
 net/netfilter/nf_conntrack_core.c              | 10 ++++++----
 net/netfilter/nf_conntrack_proto_dccp.c        |  2 +-
 net/netfilter/nf_conntrack_proto_generic.c     |  2 +-
 net/netfilter/nf_conntrack_proto_gre.c         |  3 +--
 net/netfilter/nf_conntrack_proto_sctp.c        |  2 +-
 net/netfilter/nf_conntrack_proto_tcp.c         |  2 +-
 net/netfilter/nf_conntrack_proto_udp.c         |  1 +
 net/netfilter/nf_conntrack_proto_udplite.c     |  1 +
 net/netfilter/xt_connlimit.c                   |  2 +-
 net/openvswitch/conntrack.c                    |  2 +-
 net/sched/act_connmark.c                       |  2 +-
 16 files changed, 24 insertions(+), 18 deletions(-)
include/net/netfilter/nf_conntrack.h +2 -1
index e8ad46834df8..d642f68a7c73 100644
@@ -191,7 +191,8 @@ int nf_conntrack_hash_check_insert(struct nf_conn *ct);
 bool nf_ct_delete(struct nf_conn *ct, u32 pid, int report);
 
 bool nf_ct_get_tuplepr(const struct sk_buff *skb, unsigned int nhoff,
-		       u_int16_t l3num, struct nf_conntrack_tuple *tuple);
+		       u_int16_t l3num, struct net *net,
+		       struct nf_conntrack_tuple *tuple);
 bool nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse,
 			  const struct nf_conntrack_tuple *orig);
 
include/net/netfilter/nf_conntrack_core.h +1 -0
index c03f9c42b3cd..788ef58a66b9 100644
@@ -41,6 +41,7 @@ void nf_conntrack_cleanup_end(void);
 
 bool nf_ct_get_tuple(const struct sk_buff *skb, unsigned int nhoff,
 		     unsigned int dataoff, u_int16_t l3num, u_int8_t protonum,
+		     struct net *net,
 		     struct nf_conntrack_tuple *tuple,
 		     const struct nf_conntrack_l3proto *l3proto,
 		     const struct nf_conntrack_l4proto *l4proto);
include/net/netfilter/nf_conntrack_l4proto.h +1 -1
index 1f7061313d54..956d8a6ac069 100644
@@ -26,7 +26,7 @@ struct nf_conntrack_l4proto {
 	/* Try to fill in the third arg: dataoff is offset past network protocol
            hdr.  Return true if possible. */
 	bool (*pkt_to_tuple)(const struct sk_buff *skb, unsigned int dataoff,
-			     struct nf_conntrack_tuple *tuple);
+			     struct net *net, struct nf_conntrack_tuple *tuple);
 
 	/* Invert the per-proto part of the tuple: ie. turn xmit into reply.
 	 * Some packets can't be inverted: return 0 in that case.
net/ipv4/netfilter/nf_conntrack_proto_icmp.c +2 -2
index cdde3ec496e9..c567e1b5d799 100644
@@ -30,7 +30,7 @@ static inline struct nf_icmp_net *icmp_pernet(struct net *net)
 }
 
 static bool icmp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			      struct nf_conntrack_tuple *tuple)
+			      struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	const struct icmphdr *hp;
 	struct icmphdr _hdr;
@@ -144,7 +144,7 @@ icmp_error_message(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb,
 	if (!nf_ct_get_tuplepr(skb,
 			       skb_network_offset(skb) + ip_hdrlen(skb)
 						       + sizeof(struct icmphdr),
-			       PF_INET, &origtuple)) {
+			       PF_INET, net, &origtuple)) {
 		pr_debug("icmp_error_message: failed to get tuple\n");
 		return -NF_ACCEPT;
 	}
net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +2 -1
index 0e6fae103d33..d3b797446cea 100644
@@ -36,6 +36,7 @@ static inline struct nf_icmp_net *icmpv6_pernet(struct net *net)
 
 static bool icmpv6_pkt_to_tuple(const struct sk_buff *skb,
 				unsigned int dataoff,
+				struct net *net,
 				struct nf_conntrack_tuple *tuple)
 {
 	const struct icmp6hdr *hp;
@@ -159,7 +160,7 @@ icmpv6_error_message(struct net *net, struct nf_conn *tmpl,
 			       skb_network_offset(skb)
 				+ sizeof(struct ipv6hdr)
 				+ sizeof(struct icmp6hdr),
-			       PF_INET6, &origtuple)) {
+			       PF_INET6, net, &origtuple)) {
 		pr_debug("icmpv6_error: Can't get tuple\n");
 		return -NF_ACCEPT;
 	}
net/netfilter/nf_conntrack_core.c +6 -4
index c09d6c7198f6..09d1d19b2ab9 100644
@@ -168,6 +168,7 @@ nf_ct_get_tuple(const struct sk_buff *skb,
 		unsigned int dataoff,
 		u_int16_t l3num,
 		u_int8_t protonum,
+		struct net *net,
 		struct nf_conntrack_tuple *tuple,
 		const struct nf_conntrack_l3proto *l3proto,
 		const struct nf_conntrack_l4proto *l4proto)
@@ -181,12 +182,13 @@ nf_ct_get_tuple(const struct sk_buff *skb,
 	tuple->dst.protonum = protonum;
 	tuple->dst.dir = IP_CT_DIR_ORIGINAL;
 
-	return l4proto->pkt_to_tuple(skb, dataoff, tuple);
+	return l4proto->pkt_to_tuple(skb, dataoff, net, tuple);
 }
 EXPORT_SYMBOL_GPL(nf_ct_get_tuple);
 
 bool nf_ct_get_tuplepr(const struct sk_buff *skb, unsigned int nhoff,
-		       u_int16_t l3num, struct nf_conntrack_tuple *tuple)
+		       u_int16_t l3num,
+		       struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	struct nf_conntrack_l3proto *l3proto;
 	struct nf_conntrack_l4proto *l4proto;
@@ -205,7 +207,7 @@ bool nf_ct_get_tuplepr(const struct sk_buff *skb, unsigned int nhoff,
 
 	l4proto = __nf_ct_l4proto_find(l3num, protonum);
 
-	ret = nf_ct_get_tuple(skb, nhoff, protoff, l3num, protonum, tuple,
+	ret = nf_ct_get_tuple(skb, nhoff, protoff, l3num, protonum, net, tuple,
 			      l3proto, l4proto);
 
 	rcu_read_unlock();
@@ -1029,7 +1031,7 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl,
 	u32 hash;
 
 	if (!nf_ct_get_tuple(skb, skb_network_offset(skb),
-			     dataoff, l3num, protonum, &tuple, l3proto,
+			     dataoff, l3num, protonum, net, &tuple, l3proto,
 			     l4proto)) {
 		pr_debug("resolve_normal_ct: Can't get tuple\n");
 		return NULL;
net/netfilter/nf_conntrack_proto_dccp.c +1 -1
index 6dd995c7c72b..fce1b1cca32d 100644
@@ -398,7 +398,7 @@ static inline struct dccp_net *dccp_pernet(struct net *net)
 }
 
 static bool dccp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			      struct nf_conntrack_tuple *tuple)
+			      struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	struct dccp_hdr _hdr, *dh;
 
net/netfilter/nf_conntrack_proto_generic.c +1 -1
index 2281be419a74..86dc752e5349 100644
@@ -45,7 +45,7 @@ static inline struct nf_generic_net *generic_pernet(struct net *net)
 
 static bool generic_pkt_to_tuple(const struct sk_buff *skb,
 				 unsigned int dataoff,
-				 struct nf_conntrack_tuple *tuple)
+				 struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	tuple->src.u.all = 0;
 	tuple->dst.u.all = 0;
net/netfilter/nf_conntrack_proto_gre.c +1 -2
index 7648674f29c3..a96451a7af20 100644
@@ -190,9 +190,8 @@ static bool gre_invert_tuple(struct nf_conntrack_tuple *tuple,
 
 /* gre hdr info to tuple */
 static bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			     struct nf_conntrack_tuple *tuple)
+			     struct net *net, struct nf_conntrack_tuple *tuple)
 {
-	struct net *net = dev_net(skb->dev ? skb->dev : skb_dst(skb)->dev);
 	const struct gre_hdr_pptp *pgrehdr;
 	struct gre_hdr_pptp _pgrehdr;
 	__be16 srckey;
net/netfilter/nf_conntrack_proto_sctp.c +1 -1
index 67197731eb68..9578a7c371ef 100644
@@ -156,7 +156,7 @@ static inline struct sctp_net *sctp_pernet(struct net *net)
 }
 
 static bool sctp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			      struct nf_conntrack_tuple *tuple)
+			      struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	const struct sctphdr *hp;
 	struct sctphdr _hdr;
net/netfilter/nf_conntrack_proto_tcp.c +1 -1
index 70383de72054..278f3b9356ef 100644
@@ -277,7 +277,7 @@ static inline struct nf_tcp_net *tcp_pernet(struct net *net)
 }
 
 static bool tcp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			     struct nf_conntrack_tuple *tuple)
+			     struct net *net, struct nf_conntrack_tuple *tuple)
 {
 	const struct tcphdr *hp;
 	struct tcphdr _hdr;
net/netfilter/nf_conntrack_proto_udp.c +1 -0
index 6957281ffee5..478f92f834b6 100644
@@ -38,6 +38,7 @@ static inline struct nf_udp_net *udp_pernet(struct net *net)
 
 static bool udp_pkt_to_tuple(const struct sk_buff *skb,
 			     unsigned int dataoff,
+			     struct net *net,
 			     struct nf_conntrack_tuple *tuple)
 {
 	const struct udphdr *hp;
net/netfilter/nf_conntrack_proto_udplite.c +1 -0
index c5903d1649f9..1ac8ee13a873 100644
@@ -48,6 +48,7 @@ static inline struct udplite_net *udplite_pernet(struct net *net)
 
 static bool udplite_pkt_to_tuple(const struct sk_buff *skb,
 				 unsigned int dataoff,
+				 struct net *net,
 				 struct nf_conntrack_tuple *tuple)
 {
 	const struct udphdr *hp;
net/netfilter/xt_connlimit.c +1 -1
index 213db252e5be..99bbc829868d 100644
@@ -332,7 +332,7 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par)
 		tuple_ptr = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
 		zone = nf_ct_zone(ct);
 	} else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb),
-				    par->family, &tuple)) {
+				      par->family, net, &tuple)) {
 		goto hotdrop;
 	}
 
net/openvswitch/conntrack.c +1 -1
index e8e524ad8a01..aaf5cbd6d9ae 100644
@@ -345,7 +345,7 @@ ovs_ct_expect_find(struct net *net, const struct nf_conntrack_zone *zone,
 {
 	struct nf_conntrack_tuple tuple;
 
-	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), proto, &tuple))
+	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), proto, net, &tuple))
 		return NULL;
 	return __nf_ct_expect_find(net, zone, &tuple);
 }
net/sched/act_connmark.c +1 -1
index 413ac39147d8..bb41699c6c49 100644
@@ -68,7 +68,7 @@ static int tcf_connmark(struct sk_buff *skb, const struct tc_action *a,
 	}
 
 	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb),
-			       proto, &tuple))
+			       proto, ca->net, &tuple))
 		goto out;
 
 	zone.id = ca->zone;
Signature · 1 line
2.2.1
#73 lore raw

PATCH 12/15 ipvs: Read hooknum from state rather than ops->hooknum

In reply to Eric W. Biederman

This should be more cache efficient as state is more likely to be in core, and the netfilter core will stop passing in ops soon.

---
 net/netfilter/ipvs/ip_vs_core.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
net/netfilter/ipvs/ip_vs_core.c +8 -8
index 38fbc194b9cb..0be5f12cb12d 100644
@@ -1286,7 +1286,7 @@ static unsigned int
 ip_vs_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET);
+	return ip_vs_out(state->hook, skb, AF_INET);
 }
 
 /*
@@ -1297,7 +1297,7 @@ static unsigned int
 ip_vs_local_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		   const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET);
+	return ip_vs_out(state->hook, skb, AF_INET);
 }
 
 #ifdef CONFIG_IP_VS_IPV6
@@ -1311,7 +1311,7 @@ static unsigned int
 ip_vs_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET6);
+	return ip_vs_out(state->hook, skb, AF_INET6);
 }
 
 /*
@@ -1322,7 +1322,7 @@ static unsigned int
 ip_vs_local_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		   const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET6);
+	return ip_vs_out(state->hook, skb, AF_INET6);
 }
 
 #endif
@@ -1779,7 +1779,7 @@ static unsigned int
 ip_vs_remote_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET);
+	return ip_vs_in(state->hook, skb, AF_INET);
 }
 
 /*
@@ -1790,7 +1790,7 @@ static unsigned int
 ip_vs_local_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET);
+	return ip_vs_in(state->hook, skb, AF_INET);
 }
 
 #ifdef CONFIG_IP_VS_IPV6
@@ -1803,7 +1803,7 @@ static unsigned int
 ip_vs_remote_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET6);
+	return ip_vs_in(state->hook, skb, AF_INET6);
 }
 
 /*
@@ -1814,7 +1814,7 @@ static unsigned int
 ip_vs_local_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET6);
+	return ip_vs_in(state->hook, skb, AF_INET6);
 }
 
 #endif
Signature · 1 line
2.2.1