[IPSEC]: Merge most of the output path

As part of the work on asynchrnous cryptographic operations, we need
to be able to resume from the spot where they occur.  As such, it
helps if we isolate them to one spot.

This patch moves most of the remaining family-specific processing into
the common output code.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/dst.h

@@ -98,6 +98,7 @@ struct dst_ops
 	struct dst_entry *	(*negative_advice)(struct dst_entry *);
 	void			(*link_failure)(struct sk_buff *);
 	void			(*update_pmtu)(struct dst_entry *dst, u32 mtu);
+	int			(*local_out)(struct sk_buff *skb);
 	int			entry_size;
 
 	atomic_t		entries;

include/net/xfrm.h

@@ -259,6 +259,7 @@ struct xfrm_state_afinfo {
 	unsigned int		family;
 	unsigned int		proto;
 	unsigned int		eth_proto;
+	unsigned int		nf_post_routing;
 	struct module		*owner;
 	struct xfrm_type	*type_map[IPPROTO_MAX];
 	struct xfrm_mode	*mode_map[XFRM_MODE_MAX];

net/ipv4/route.c

@@ -166,6 +166,7 @@ static struct dst_ops ipv4_dst_ops = {
 	.negative_advice =	ipv4_negative_advice,
 	.link_failure =		ipv4_link_failure,
 	.update_pmtu =		ip_rt_update_pmtu,
+	.local_out =		ip_local_out,
 	.entry_size =		sizeof(struct rtable),
 };
 

net/ipv4/xfrm4_output.c

@@ -59,7 +59,7 @@ int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 		return err;
 
 	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
-	IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE;
+	IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED;
 
 	skb->protocol = htons(ETH_P_IP);
 
@@ -67,87 +67,19 @@ int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 }
 EXPORT_SYMBOL(xfrm4_prepare_output);
 
-static inline int xfrm4_output_one(struct sk_buff *skb)
-{
-	int err;
-
-	err = xfrm_output(skb);
-	if (err)
-		goto error_nolock;
-
-	IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED;
-	err = 0;
-
-out_exit:
-	return err;
-error_nolock:
-	kfree_skb(skb);
-	goto out_exit;
-}
-
-static int xfrm4_output_finish2(struct sk_buff *skb)
-{
-	int err;
-
-	while (likely((err = xfrm4_output_one(skb)) == 0)) {
-		nf_reset(skb);
-
-		err = __ip_local_out(skb);
-		if (unlikely(err != 1))
-			break;
-
-		if (!skb->dst->xfrm)
-			return dst_output(skb);
-
-		err = nf_hook(PF_INET, NF_IP_POST_ROUTING, skb, NULL,
-			      skb->dst->dev, xfrm4_output_finish2);
-		if (unlikely(err != 1))
-			break;
-	}
-
-	return err;
-}
-
 static int xfrm4_output_finish(struct sk_buff *skb)
 {
-	struct sk_buff *segs;
-
 #ifdef CONFIG_NETFILTER
 	if (!skb->dst->xfrm) {
 		IPCB(skb)->flags |= IPSKB_REROUTED;
 		return dst_output(skb);
 	}
-#endif
 
-	if (!skb_is_gso(skb))
-		return xfrm4_output_finish2(skb);
+	IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED;
+#endif
 
 	skb->protocol = htons(ETH_P_IP);
-	segs = skb_gso_segment(skb, 0);
-	kfree_skb(skb);
-	if (unlikely(IS_ERR(segs)))
-		return PTR_ERR(segs);
-
-	do {
-		struct sk_buff *nskb = segs->next;
-		int err;
-
-		segs->next = NULL;
-		err = xfrm4_output_finish2(segs);
-
-		if (unlikely(err)) {
-			while ((segs = nskb)) {
-				nskb = segs->next;
-				segs->next = NULL;
-				kfree_skb(segs);
-			}
-			return err;
-		}
-
-		segs = nskb;
-	} while (segs);
-
-	return 0;
+	return xfrm_output(skb);
 }
 
 int xfrm4_output(struct sk_buff *skb)

net/ipv4/xfrm4_policy.c

@@ -237,6 +237,7 @@ static struct dst_ops xfrm4_dst_ops = {
 	.update_pmtu =		xfrm4_update_pmtu,
 	.destroy =		xfrm4_dst_destroy,
 	.ifdown =		xfrm4_dst_ifdown,
+	.local_out =		__ip_local_out,
 	.gc_thresh =		1024,
 	.entry_size =		sizeof(struct xfrm_dst),
 };

net/ipv4/xfrm4_state.c

@@ -11,6 +11,7 @@
 #include <net/xfrm.h>
 #include <linux/pfkeyv2.h>
 #include <linux/ipsec.h>
+#include <linux/netfilter_ipv4.h>
 
 static struct xfrm_state_afinfo xfrm4_state_afinfo;
 
@@ -66,6 +67,7 @@ static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.family			= AF_INET,
 	.proto			= IPPROTO_IPIP,
 	.eth_proto		= htons(ETH_P_IP),
+	.nf_post_routing	= NF_IP_POST_ROUTING,
 	.owner			= THIS_MODULE,
 	.init_flags		= xfrm4_init_flags,
 	.init_tempsel		= __xfrm4_init_tempsel,

net/ipv6/route.c

@@ -113,6 +113,7 @@ static struct dst_ops ip6_dst_ops = {
 	.negative_advice	=	ip6_negative_advice,
 	.link_failure		=	ip6_link_failure,
 	.update_pmtu		=	ip6_rt_update_pmtu,
+	.local_out		=	ip6_local_out,
 	.entry_size		=	sizeof(struct rt6_info),
 };
 

net/ipv6/xfrm6_output.c

@@ -66,6 +66,9 @@ int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 		return err;
 
 	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
+#ifdef CONFIG_NETFILTER
+	IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED;
+#endif
 
 	skb->protocol = htons(ETH_P_IPV6);
 
@@ -73,80 +76,14 @@ int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 }
 EXPORT_SYMBOL(xfrm6_prepare_output);
 
-static inline int xfrm6_output_one(struct sk_buff *skb)
-{
-	int err;
-
-	err = xfrm_output(skb);
-	if (err)
-		goto error_nolock;
-
-	IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED;
-	err = 0;
-
-out_exit:
-	return err;
-error_nolock:
-	kfree_skb(skb);
-	goto out_exit;
-}
-
-static int xfrm6_output_finish2(struct sk_buff *skb)
-{
-	int err;
-
-	while (likely((err = xfrm6_output_one(skb)) == 0)) {
-		nf_reset(skb);
-
-		err = __ip6_local_out(skb);
-		if (unlikely(err != 1))
-			break;
-
-		if (!skb->dst->xfrm)
-			return dst_output(skb);
-
-		err = nf_hook(PF_INET6, NF_IP6_POST_ROUTING, skb, NULL,
-			      skb->dst->dev, xfrm6_output_finish2);
-		if (unlikely(err != 1))
-			break;
-	}
-
-	return err;
-}
-
 static int xfrm6_output_finish(struct sk_buff *skb)
 {
-	struct sk_buff *segs;
-
-	if (!skb_is_gso(skb))
-		return xfrm6_output_finish2(skb);
+#ifdef CONFIG_NETFILTER
+	IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED;
+#endif
 
 	skb->protocol = htons(ETH_P_IPV6);
-	segs = skb_gso_segment(skb, 0);
-	kfree_skb(skb);
-	if (unlikely(IS_ERR(segs)))
-		return PTR_ERR(segs);
-
-	do {
-		struct sk_buff *nskb = segs->next;
-		int err;
-
-		segs->next = NULL;
-		err = xfrm6_output_finish2(segs);
-
-		if (unlikely(err)) {
-			while ((segs = nskb)) {
-				nskb = segs->next;
-				segs->next = NULL;
-				kfree_skb(segs);
-			}
-			return err;
-		}
-
-		segs = nskb;
-	} while (segs);
-
-	return 0;
+	return xfrm_output(skb);
 }
 
 int xfrm6_output(struct sk_buff *skb)

net/ipv6/xfrm6_policy.c

@@ -252,6 +252,7 @@ static struct dst_ops xfrm6_dst_ops = {
 	.update_pmtu =		xfrm6_update_pmtu,
 	.destroy =		xfrm6_dst_destroy,
 	.ifdown =		xfrm6_dst_ifdown,
+	.local_out =		__ip6_local_out,
 	.gc_thresh =		1024,
 	.entry_size =		sizeof(struct xfrm_dst),
 };

net/ipv6/xfrm6_state.c

@@ -14,6 +14,7 @@
 #include <net/xfrm.h>
 #include <linux/pfkeyv2.h>
 #include <linux/ipsec.h>
+#include <linux/netfilter_ipv6.h>
 #include <net/dsfield.h>
 #include <net/ipv6.h>
 #include <net/addrconf.h>
@@ -189,6 +190,7 @@ static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.family			= AF_INET6,
 	.proto			= IPPROTO_IPV6,
 	.eth_proto		= htons(ETH_P_IPV6),
+	.nf_post_routing	= NF_IP6_POST_ROUTING,
 	.owner			= THIS_MODULE,
 	.init_tempsel		= __xfrm6_init_tempsel,
 	.tmpl_sort		= __xfrm6_tmpl_sort,

net/xfrm/xfrm_output.c

@@ -12,6 +12,7 @@
 #include <linux/errno.h>
 #include <linux/module.h>
 #include <linux/netdevice.h>
+#include <linux/netfilter.h>
 #include <linux/skbuff.h>
 #include <linux/spinlock.h>
 #include <net/dst.h>
@@ -40,7 +41,7 @@ static int xfrm_state_check(struct xfrm_state *x, struct sk_buff *skb)
 	return err;
 }
 
-int xfrm_output(struct sk_buff *skb)
+static int xfrm_output_one(struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb->dst;
 	struct xfrm_state *x = dst->xfrm;
@@ -87,10 +88,73 @@ int xfrm_output(struct sk_buff *skb)
 
 	err = 0;
 
-error_nolock:
+out_exit:
 	return err;
 error:
 	spin_unlock_bh(&x->lock);
-	goto error_nolock;
+error_nolock:
+	kfree_skb(skb);
+	goto out_exit;
+}
+
+static int xfrm_output2(struct sk_buff *skb)
+{
+	int err;
+
+	while (likely((err = xfrm_output_one(skb)) == 0)) {
+		struct xfrm_state *x;
+
+		nf_reset(skb);
+
+		err = skb->dst->ops->local_out(skb);
+		if (unlikely(err != 1))
+			break;
+
+		x = skb->dst->xfrm;
+		if (!x)
+			return dst_output(skb);
+
+		err = nf_hook(x->inner_mode->afinfo->family,
+			      x->inner_mode->afinfo->nf_post_routing, skb,
+			      NULL, skb->dst->dev, xfrm_output2);
+		if (unlikely(err != 1))
+			break;
+	}
+
+	return err;
+}
+
+int xfrm_output(struct sk_buff *skb)
+{
+	struct sk_buff *segs;
+
+	if (!skb_is_gso(skb))
+		return xfrm_output2(skb);
+
+	segs = skb_gso_segment(skb, 0);
+	kfree_skb(skb);
+	if (unlikely(IS_ERR(segs)))
+		return PTR_ERR(segs);
+
+	do {
+		struct sk_buff *nskb = segs->next;
+		int err;
+
+		segs->next = NULL;
+		err = xfrm_output2(segs);
+
+		if (unlikely(err)) {
+			while ((segs = nskb)) {
+				nskb = segs->next;
+				segs->next = NULL;
+				kfree_skb(segs);
+			}
+			return err;
+		}
+
+		segs = nskb;
+	} while (segs);
+
+	return 0;
 }
 EXPORT_SYMBOL_GPL(xfrm_output);