beet-interfamily-sleep-loopback-2.6.21.3.patch

来自「HIP：Host Identity Protocol」· PATCH 代码 · 共 840 行 · 第 1/2 页

 static inline void
 __xfrm6_bundle_len_inc(int *len, int *nflen, struct xfrm_state *x)
 {
-	if (x->type->flags & XFRM_TYPE_NON_FRAGMENT)
+	if (x->type && x->type->flags & XFRM_TYPE_NON_FRAGMENT)
 		*nflen += x->props.header_len;
 	else
 		*len += x->props.header_len;
@@ -114,7 +115,7 @@
 static inline void
 __xfrm6_bundle_len_dec(int *len, int *nflen, struct xfrm_state *x)
 {
-	if (x->type->flags & XFRM_TYPE_NON_FRAGMENT)
+	if (x->type && x->type->flags & XFRM_TYPE_NON_FRAGMENT)
 		*nflen -= x->props.header_len;
 	else
 		*len -= x->props.header_len;
@@ -131,6 +132,7 @@
 	struct dst_entry *dst, *dst_prev;
 	struct rt6_info *rt0 = (struct rt6_info*)(*dst_p);
 	struct rt6_info *rt  = rt0;
+        unsigned short encap_family = 0, beet = 0;
 	struct flowi fl_tunnel = {
 		.nl_u = {
 			.ip6_u = {
@@ -139,6 +141,10 @@
 			}
 		}
 	};
+	union {
+		struct in6_addr *in6;
+		struct in_addr *in;
+	} remote, local;
 	int i;
 	int err = 0;
 	int header_len = 0;
@@ -175,20 +181,35 @@
 		dst1->next = dst_prev;
 		dst_prev = dst1;
 
+		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
+			encap_family = xfrm[i]->props.family;
+			beet = (xfrm[i]->props.mode == XFRM_MODE_BEET);
+			if (encap_family == AF_INET6) {
+				remote.in6 = __xfrm6_bundle_addr_remote(xfrm[i], remote.in6);
+				local.in6  = __xfrm6_bundle_addr_local(xfrm[i], local.in6);
+			} else if (encap_family == AF_INET) {
+				remote.in = (struct in_addr*)&xfrm[i]->id.daddr.a4;
+				local.in = (struct in_addr*)&xfrm[i]->props.saddr.a4;
+			} else {
+				BUG_ON(1);
+			}
+		}
+
 		__xfrm6_bundle_len_inc(&header_len, &nfheader_len, xfrm[i]);
 		trailer_len += xfrm[i]->props.trailer_len;
 
 		if (xfrm[i]->props.mode == XFRM_MODE_TUNNEL ||
+                    xfrm[i]->props.mode == XFRM_MODE_BEET ||
 		    xfrm[i]->props.mode == XFRM_MODE_ROUTEOPTIMIZATION) {
-			unsigned short encap_family = xfrm[i]->props.family;
 			switch(encap_family) {
 			case AF_INET:
 				fl_tunnel.fl4_dst = xfrm[i]->id.daddr.a4;
 				fl_tunnel.fl4_src = xfrm[i]->props.saddr.a4;
+                                fl_tunnel.fl4_tos = 0;
+                                fl_tunnel.fl4_scope = 0;
 				break;
 			case AF_INET6:
 				ipv6_addr_copy(&fl_tunnel.fl6_dst, __xfrm6_bundle_addr_remote(xfrm[i], &fl->fl6_dst));
-
 				ipv6_addr_copy(&fl_tunnel.fl6_src, __xfrm6_bundle_addr_local(xfrm[i], &fl->fl6_src));
 				break;
 			default:
@@ -259,6 +280,13 @@
 	}
 
 	xfrm_init_pmtu(dst);
+
+	if (beet && encap_family == AF_INET) {
+		int delta = sizeof(struct ipv6hdr) - sizeof(struct iphdr);
+		u32 mtu = dst_mtu(dst);
+		xfrm6_update_pmtu(dst, mtu + delta);
+	}
+
 	return 0;
 
 error:
diff -urN  a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c
--- a/net/ipv6/xfrm6_tunnel.c 2007-05-25 16:36:44.000000000 +0300
+++ b/net/ipv6/xfrm6_tunnel.c 2007-05-25 16:36:44.000000000 +0300
@@ -307,7 +307,7 @@
 
 static int xfrm6_tunnel_init_state(struct xfrm_state *x)
 {
-	if (x->props.mode != XFRM_MODE_TUNNEL)
+	if (x->props.mode != XFRM_MODE_TUNNEL || x->props.mode != XFRM_MODE_BEET)
 		return -EINVAL;
 
 	if (x->encap)
diff -urN  a/net/ipv6/xfrm6_mode_tunnel.c b/net/ipv6/xfrm6_mode_tunnel.c
--- a/net/ipv6/xfrm6_mode_tunnel.c 2007-05-25 16:36:44.000000000 +0300
+++ b/net/ipv6/xfrm6_mode_tunnel.c 2007-05-25 16:36:44.000000000 +0300
@@ -80,6 +80,7 @@
 	top_iph->hop_limit = dst_metric(dst->child, RTAX_HOPLIMIT);
 	ipv6_addr_copy(&top_iph->saddr, (struct in6_addr *)&x->props.saddr);
 	ipv6_addr_copy(&top_iph->daddr, (struct in6_addr *)&x->id.daddr);
+        skb->protocol = htons(ETH_P_IPV6);
 	return 0;
 }
 
diff -urN  a/net/ipv6/xfrm6_mode_beet.c b/net/ipv6/xfrm6_mode_beet.c
--- a/net/ipv6/xfrm6_mode_beet.c 2007-05-25 16:36:44.000000000 +0300
+++ b/net/ipv6/xfrm6_mode_beet.c 2007-05-25 16:36:44.000000000 +0300
@@ -6,6 +6,7 @@
  *                    Herbert Xu     <[e-mail]>
  *                    Abhinav Pathak <[e-mail]>
  *                    Jeff Ahrenholz <[e-mail]>
+ *                    Joakim Koskela <[e-mail]>
  */
 
 #include <linux/init.h>
@@ -33,37 +34,154 @@
  */
 static int xfrm6_beet_output(struct xfrm_state *x, struct sk_buff *skb)
 {
-	struct ipv6hdr *iph, *top_iph;
-	u8 *prevhdr;
-	int hdr_len;
-
-	skb_push(skb, x->props.header_len);
-	iph = skb->nh.ipv6h;
-
-	hdr_len = ip6_find_1stfragopt(skb, &prevhdr);
-	skb->nh.raw = prevhdr - x->props.header_len;
-	skb->h.raw = skb->data + hdr_len;
-	memmove(skb->data, iph, hdr_len);
 
-	skb->nh.raw = skb->data;
-	top_iph = skb->nh.ipv6h;
-	skb->nh.raw = &top_iph->nexthdr;
-	skb->h.ipv6h = top_iph + 1;
-
-	ipv6_addr_copy(&top_iph->saddr, (struct in6_addr *)&x->props.saddr);
-	ipv6_addr_copy(&top_iph->daddr, (struct in6_addr *)&x->id.daddr);
+        struct dst_entry *dst = skb->dst;
+        int hdrlen;
+        struct iphdr *iphv4, *top_iphv4;
+        struct ipv6hdr *iphv6, *top_iphv6;
+	
+	if (skb->nh.iph->version == 6) {
+                
+		u8 *prevhdr;
+		int hdr_len;
+
+                /* 6-6 */
+
+                hdrlen = x->props.header_len - IPV4_BEET_PHMAXLEN;
+                skb_push(skb, hdrlen);
+                iphv6 = skb->nh.ipv6h;
+
+		hdr_len = ip6_find_1stfragopt(skb, &prevhdr);
+		skb->nh.raw = prevhdr - hdrlen;
+		skb->h.raw = skb->data + hdr_len;
+		memmove(skb->data, iphv6, hdr_len);
+                
+                skb->nh.raw = skb->data;
+                top_iphv6 = skb->nh.ipv6h;
+		skb->nh.raw = &top_iphv6->nexthdr;
+		skb->h.ipv6h = top_iphv6 + 1;
+		ipv6_addr_copy(&top_iphv6->saddr, (struct in6_addr *)&x->props.saddr);
+		ipv6_addr_copy(&top_iphv6->daddr, (struct in6_addr *)&x->id.daddr);
+
+                skb->protocol = htons(ETH_P_IPV6);
+                
+	} else if (skb->nh.iph->version == 4) {
+
+                int flags;
+                int optlen;
+                int dsfield;
+                u8 protocol;
+                int delta = sizeof(struct ipv6hdr) - sizeof(struct iphdr);
+
+                /* Inner = 4, Outer = 6*/
+
+                iphv4 = skb->nh.iph;
+                skb->h.ipiph = iphv4;
+
+                hdrlen = x->props.header_len;
+
+                optlen = iphv4->ihl * 4 - sizeof(*iphv4);
+                
+                if (!optlen) {
+                        hdrlen -= IPV4_BEET_PHMAXLEN;
+                } else {
+                        skb->h.raw -= (IPV4_BEET_PHMAXLEN - (optlen & 4));
+                        hdrlen -= optlen & 4;
+                }
+
+                skb->nh.raw = skb_push(skb, hdrlen);
+                
+                top_iphv4 = skb->nh.iph;
+                hdrlen = iphv4->ihl * 4 - optlen;
+                skb->h.raw += hdrlen;                   
+                if (unlikely(optlen)) {
+                        struct ip_beet_phdr *ph;
+                
+                        BUG_ON(optlen < 0);
+                
+                        ph = (struct ip_beet_phdr *)skb->h.raw;
+                        ph->padlen = 4 - (optlen & 4);
+                        ph->hdrlen = (optlen + ph->padlen + sizeof(*ph)) / 8;
+                        ph->nexthdr = iphv4->protocol;
+                        top_iphv4->protocol = IPPROTO_BEETPH;
+                        top_iphv4->ihl = sizeof(struct iphdr) / 4;
+                }
+                
+                if (unlikely(optlen))
+                        protocol = top_iphv4->protocol;
+                else
+                        protocol = iphv4->protocol;
+                
+                if (skb_headroom(skb) <=  2*delta){
+                        if (pskb_expand_head(skb, delta,0, GFP_ATOMIC))
+                                return -ENOMEM;
+                }
+                
+                skb->nh.raw = skb_push(skb, delta);
+
+                top_iphv6 = skb->nh.ipv6h;
+                skb->h.ipv6h = top_iphv6 + 1;
+                /* DS disclosed */
+                top_iphv6->version = 6;
+                top_iphv6->priority = 0;
+                top_iphv6->flow_lbl[0] = 0;
+                top_iphv6->flow_lbl[1] = 0;
+                top_iphv6->flow_lbl[2] = 0;
+                dsfield = ipv6_get_dsfield(top_iphv6);
+                dsfield = INET_ECN_encapsulate(dsfield, dsfield);
+                flags = x->props.flags;
+                if (flags & XFRM_STATE_NOECN)
+                        dsfield &= ~INET_ECN_MASK;
+                ipv6_change_dsfield(top_iphv6, 0, dsfield);
+
+		top_iphv6->nexthdr = protocol;
+                top_iphv6->hop_limit = dst_metric(dst->child, RTAX_HOPLIMIT);
+                top_iphv6->payload_len = htons(skb->len - sizeof(struct ipv6hdr));
+                ipv6_addr_copy(&top_iphv6->saddr,(struct in6_addr *)&x->props.saddr);
+                ipv6_addr_copy(&top_iphv6->daddr, (struct in6_addr *)&x->id.daddr);
+                skb->nh.raw = &top_iphv6->nexthdr;
+
+                skb->protocol = htons(ETH_P_IPV6);
+	} else
+		BUG_ON(1);
 
 	return 0;
 }
 
 static int xfrm6_beet_input(struct xfrm_state *x, struct sk_buff *skb)
 {
-	struct ipv6hdr *ip6h;
-	int size = sizeof(struct ipv6hdr);
+	struct ip_beet_phdr *ph = (struct ip_beet_phdr*)(skb->h.raw);
+	int size = (x->sel.family == AF_INET) ? sizeof(struct iphdr) : sizeof(struct ipv6hdr);
+	int delta = sizeof(struct ipv6hdr) - sizeof(struct iphdr);
+	__u8 proto = skb->nh.ipv6h->nexthdr, hops = skb->nh.ipv6h->hop_limit;
+	__u8 ph_nexthdr = 0;
+	int phlen = 0;
+	int optlen = 0;
+
 	int err = -EINVAL;
 
-	if (!pskb_may_pull(skb, sizeof(struct ipv6hdr)))
-		goto out;
+	if (x->sel.family == AF_INET) {
+		/* Inner = IPv4, therefore the IPhdr must be shrunk */
+		/* Inner = 4, Outer = 6 */
+		if (unlikely(proto == IPPROTO_BEETPH)) {
+			if (!pskb_may_pull(skb, sizeof(*ph)))
+				goto out;
+			phlen = ph->hdrlen * 8;
+			optlen = phlen - ph->padlen - sizeof(*ph);
+
+			if (optlen < 0 || optlen & 3 || optlen > 250)
+				goto out;
+			if (!pskb_may_pull(skb, phlen))
+				goto out;
+
+			proto = ph_nexthdr = ph->nexthdr;
+		}
+		skb->nh.raw += delta;
+	}
+        
+        if (skb_cloned(skb) &&
+            pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
+                goto out;
 
 	skb_push(skb, size);
 	memmove(skb->data, skb->nh.raw, size);
@@ -71,11 +189,36 @@
 
 	skb->mac.raw = memmove(skb->data - skb->mac_len,
 			       skb->mac.raw, skb->mac_len);
-
-	ip6h = skb->nh.ipv6h;
-	ip6h->payload_len = htons(skb->len - size);
-	ipv6_addr_copy(&ip6h->daddr, (struct in6_addr *) &x->sel.daddr.a6);
-	ipv6_addr_copy(&ip6h->saddr, (struct in6_addr *) &x->sel.saddr.a6);
+	if (unlikely(phlen)) {
+		skb_pull(skb, phlen - optlen);
+		skb->nh.raw = skb->data;
+	}
+	if (x->sel.family == AF_INET6) {
+		struct ipv6hdr *ip6h = skb->nh.ipv6h;
+		ip6h->payload_len = htons(skb->len - size);
+		ipv6_addr_copy(&ip6h->daddr, (struct in6_addr *) &x->sel.daddr.a6);
+		ipv6_addr_copy(&ip6h->saddr, (struct in6_addr *) &x->sel.saddr.a6);
+	} else if (x->sel.family == AF_INET) {
+		struct iphdr *iph = skb->nh.iph;
+		iph->ihl = (sizeof(*iph) + optlen) / 4;
+		iph->version = 4;
+		iph->tos = 0;
+		iph->id = 0;
+		iph->frag_off = 0;
+		iph->ttl = hops;
+		iph->protocol = proto;
+		iph->daddr = x->sel.daddr.a4;
+		iph->saddr = x->sel.saddr.a4;
+		iph->tot_len = htons(skb->len);
+		ip_send_check(iph);
+		skb->protocol = htons(ETH_P_IP);
+		if (unlikely(!optlen)) {
+			skb->h.raw = skb->nh.raw;
+		}
+		dst_release(skb->dst);
+		skb->dst = NULL;
+	} else
+		BUG_ON(1);
 	err = 0;
 out:
 	return err;
diff -urN  a/net/ipv6/esp6.c b/net/ipv6/esp6.c
--- a/net/ipv6/esp6.c 2007-05-25 16:36:44.000000000 +0300
+++ b/net/ipv6/esp6.c 2007-05-25 16:36:44.000000000 +0300
@@ -242,7 +242,7 @@
 	struct esp_data *esp = x->data;
 	u32 blksize = ALIGN(crypto_blkcipher_blocksize(esp->conf.tfm), 4);
 
-	if (x->props.mode == XFRM_MODE_TUNNEL) {
+	if (x->props.mode == XFRM_MODE_TUNNEL || x->props.mode == XFRM_MODE_BEET) {
 		mtu = ALIGN(mtu + 2, blksize);
 	} else {
 		/* The worst case. */
@@ -365,6 +365,8 @@
 	x->props.header_len = sizeof(struct ipv6_esp_hdr) + esp->conf.ivlen;
 	if (x->props.mode == XFRM_MODE_TUNNEL)
 		x->props.header_len += sizeof(struct ipv6hdr);
+	else if (x->props.mode == XFRM_MODE_BEET)
+		x->props.header_len += IPV4_BEET_PHMAXLEN;
 	x->data = esp;
 	return 0;
 
diff -urN  a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
--- a/net/ipv6/xfrm6_input.c 2007-05-25 16:36:44.000000000 +0300
+++ b/net/ipv6/xfrm6_input.c 2007-05-25 16:36:44.000000000 +0300
@@ -73,7 +73,7 @@
 		if (x->mode->input(x, skb))
 			goto drop;
 
-		if (x->props.mode == XFRM_MODE_TUNNEL) { /* XXX */
+		if (x->props.mode == XFRM_MODE_TUNNEL || x->props.mode == XFRM_MODE_BEET) { /* XXX */
 			decaps = 1;
 			break;
 		}
diff -urN  a/net/dccp/ipv6.c b/net/dccp/ipv6.c
--- a/net/dccp/ipv6.c 2007-05-25 16:36:44.000000000 +0300
+++ b/net/dccp/ipv6.c 2007-05-25 16:36:44.000000000 +0300
@@ -156,7 +156,7 @@
 				goto out;
 			}
 
-			err = xfrm_lookup(&dst, &fl, sk, 0);
+			err = xfrm_lookup(&dst, &fl, sk, 1);
 			if (err < 0) {
 				sk->sk_err_soft = -err;
 				goto out;
diff -urN  a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
--- a/net/xfrm/xfrm_policy.c 2007-05-25 16:36:44.000000000 +0300
+++ b/net/xfrm/xfrm_policy.c 2007-05-25 16:36:44.000000000 +0300
@@ -1224,7 +1224,7 @@
 		xfrm_address_t *local  = saddr;
 		struct xfrm_tmpl *tmpl = &policy->xfrm_vec[i];
 
-		if (tmpl->mode == XFRM_MODE_TUNNEL) {
+		if (tmpl->mode == XFRM_MODE_TUNNEL || tmpl->mode == XFRM_MODE_BEET) {
 			remote = &tmpl->id.daddr;
 			local = &tmpl->saddr;
 			family = tmpl->encap_family;
diff -urN  a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
--- a/net/xfrm/xfrm_state.c 2007-05-25 16:36:45.000000000 +0300
+++ b/net/xfrm/xfrm_state.c 2007-05-25 16:36:45.000000000 +0300
@@ -558,7 +558,7 @@
 			      selector.
 			 */
 			if (x->km.state == XFRM_STATE_VALID) {
-				if (!xfrm_selector_match(&x->sel, fl, family) ||
+				if (!xfrm_selector_match(&x->sel, fl, x->sel.family) ||
 				    !security_xfrm_state_pol_flow_match(x, pol, fl))
 					continue;
 				if (!best ||
@@ -570,7 +570,7 @@
 				acquire_in_progress = 1;
 			} else if (x->km.state == XFRM_STATE_ERROR ||
 				   x->km.state == XFRM_STATE_EXPIRED) {
-				if (xfrm_selector_match(&x->sel, fl, family) &&
+				if (xfrm_selector_match(&x->sel, fl, x->sel.family) &&
 				    security_xfrm_state_pol_flow_match(x, pol, fl))
 					error = -ESRCH;
 			}