X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=net%2Fipv6%2Fip6_output.c;h=fce2a87e9e4a6f70bb89006303fc03db6b769a77;hb=6a77f38946aaee1cd85eeec6cf4229b204c15071;hp=0057672a30ce8b07803c662502c82626f0d6dbe3;hpb=9213980e6a70d8473e0ffd4b39ab5b6caaba9ff5;p=linux-2.6.git

diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 0057672a3..fce2a87e9 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -54,13 +54,14 @@
 #include <net/rawv6.h>
 #include <net/icmp.h>
 #include <net/xfrm.h>
+#include <net/checksum.h>
 
-static int ip6_fragment(struct sk_buff **pskb, int (*output)(struct sk_buff**));
+static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *));
 
 static __inline__ void ipv6_select_ident(struct sk_buff *skb, struct frag_hdr *fhdr)
 {
 	static u32 ipv6_fragmentation_id = 1;
-	static spinlock_t ip6_id_lock = SPIN_LOCK_UNLOCKED;
+	static DEFINE_SPINLOCK(ip6_id_lock);
 
 	spin_lock_bh(&ip6_id_lock);
 	fhdr->identification = htonl(ipv6_fragmentation_id);
@@ -87,7 +88,7 @@ static inline int ip6_output_finish(struct sk_buff *skb)
 	} else if (dst->neighbour)
 		return dst->neighbour->output(skb);
 
-	IP6_INC_STATS_BH(OutNoRoutes);
+	IP6_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
 	kfree_skb(skb);
 	return -EINVAL;
 
@@ -107,9 +108,8 @@ static int ip6_dev_loopback_xmit(struct sk_buff *newskb)
 }
 
 
-static int ip6_output2(struct sk_buff **pskb)
+static int ip6_output2(struct sk_buff *skb)
 {
-	struct sk_buff *skb = *pskb;
 	struct dst_entry *dst = skb->dst;
 	struct net_device *dev = dst->dev;
 
@@ -133,26 +133,24 @@ static int ip6_output2(struct sk_buff **pskb)
 					ip6_dev_loopback_xmit);
 
 			if (skb->nh.ipv6h->hop_limit == 0) {
-				IP6_INC_STATS(OutDiscards);
+				IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
 				kfree_skb(skb);
 				return 0;
 			}
 		}
 
-		IP6_INC_STATS(OutMcastPkts);
+		IP6_INC_STATS(IPSTATS_MIB_OUTMCASTPKTS);
 	}
 
 	return NF_HOOK(PF_INET6, NF_IP6_POST_ROUTING, skb,NULL, skb->dev,ip6_output_finish);
 }
 
-int ip6_output(struct sk_buff **pskb)
+int ip6_output(struct sk_buff *skb)
 {
-	struct sk_buff *skb = *pskb;
-
-	if ((skb->len > dst_pmtu(skb->dst) || skb_shinfo(skb)->frag_list))
-		return ip6_fragment(pskb, ip6_output2);
+	if (skb->len > dst_pmtu(skb->dst))
+		return ip6_fragment(skb, ip6_output2);
 	else
-		return ip6_output2(pskb);
+		return ip6_output2(skb);
 }
 
 #ifdef CONFIG_NETFILTER
@@ -172,7 +170,7 @@ int ip6_route_me_harder(struct sk_buff *skb)
 	dst = ip6_route_output(skb->sk, &fl);
 
 	if (dst->error) {
-		IP6_INC_STATS(OutNoRoutes);
+		IP6_INC_STATS(IPSTATS_MIB_OUTNOROUTES);
 		LIMIT_NETDEBUG(
 			printk(KERN_DEBUG "ip6_route_me_harder: No more route.\n"));
 		dst_release(dst);
@@ -231,7 +229,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl,
 			kfree_skb(skb);
 			skb = skb2;
 			if (skb == NULL) {	
-				IP6_INC_STATS(OutDiscards);
+				IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
 				return -ENOBUFS;
 			}
 			if (sk)
@@ -265,7 +263,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl,
 
 	mtu = dst_pmtu(dst);
 	if ((skb->len <= mtu) || ipfragok) {
-		IP6_INC_STATS(OutRequests);
+		IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS);
 		return NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, dst->dev, ip6_maybe_reroute);
 	}
 
@@ -273,7 +271,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl,
 		printk(KERN_DEBUG "IPv6: sending pkt_too_big to self\n");
 	skb->dev = dst->dev;
 	icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev);
-	IP6_INC_STATS(FragFails);
+	IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS);
 	kfree_skb(skb);
 	return -EMSGSIZE;
 }
@@ -313,7 +311,7 @@ int ip6_nd_hdr(struct sock *sk, struct sk_buff *skb, struct net_device *dev,
 	return 0;
 }
 
-int ip6_call_ra_chain(struct sk_buff *skb, int sel)
+static int ip6_call_ra_chain(struct sk_buff *skb, int sel)
 {
 	struct ip6_ra_chain *ra;
 	struct sock *last = NULL;
@@ -355,7 +353,7 @@ int ip6_forward(struct sk_buff *skb)
 		goto error;
 
 	if (!xfrm6_policy_check(NULL, XFRM_POLICY_FWD, skb)) {
-		IP6_INC_STATS(InDiscards);
+		IP6_INC_STATS(IPSTATS_MIB_INDISCARDS);
 		goto drop;
 	}
 
@@ -394,7 +392,7 @@ int ip6_forward(struct sk_buff *skb)
 	}
 
 	if (!xfrm6_route_forward(skb)) {
-		IP6_INC_STATS(InDiscards);
+		IP6_INC_STATS(IPSTATS_MIB_INDISCARDS);
 		goto drop;
 	}
 
@@ -432,14 +430,14 @@ int ip6_forward(struct sk_buff *skb)
 		/* Again, force OUTPUT device used as source address */
 		skb->dev = dst->dev;
 		icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, dst_pmtu(dst), skb->dev);
-		IP6_INC_STATS_BH(InTooBigErrors);
-		IP6_INC_STATS_BH(FragFails);
+		IP6_INC_STATS_BH(IPSTATS_MIB_INTOOBIGERRORS);
+		IP6_INC_STATS_BH(IPSTATS_MIB_FRAGFAILS);
 		kfree_skb(skb);
 		return -EMSGSIZE;
 	}
 
 	if (skb_cow(skb, dst->dev->hard_header_len)) {
-		IP6_INC_STATS(OutDiscards);
+		IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
 		goto drop;
 	}
 
@@ -449,11 +447,11 @@ int ip6_forward(struct sk_buff *skb)
  
 	hdr->hop_limit--;
 
-	IP6_INC_STATS_BH(OutForwDatagrams);
+	IP6_INC_STATS_BH(IPSTATS_MIB_OUTFORWDATAGRAMS);
 	return NF_HOOK(PF_INET6,NF_IP6_FORWARD, skb, skb->dev, dst->dev, ip6_forward_finish);
 
 error:
-	IP6_INC_STATS_BH(InAddrErrors);
+	IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS);
 drop:
 	kfree_skb(skb);
 	return -EINVAL;
@@ -465,6 +463,7 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from)
 	to->priority = from->priority;
 	to->protocol = from->protocol;
 	to->security = from->security;
+	dst_release(to->dst);
 	to->dst = dst_clone(from->dst);
 	to->dev = from->dev;
 
@@ -476,6 +475,7 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from)
 	/* Connection association is same as pre-frag packet */
 	to->nfct = from->nfct;
 	nf_conntrack_get(to->nfct);
+	to->nfctinfo = from->nfctinfo;
 #ifdef CONFIG_BRIDGE_NETFILTER
 	nf_bridge_put(to->nf_bridge);
 	to->nf_bridge = from->nf_bridge;
@@ -516,10 +516,10 @@ int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr)
 	return offset;
 }
 
-static int ip6_fragment(struct sk_buff **pskb, int (*output)(struct sk_buff**))
+static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 {
 	struct net_device *dev;
-	struct sk_buff *frag, *skb = *pskb;
+	struct sk_buff *frag;
 	struct rt6_info *rt = (struct rt6_info*)skb->dst;
 	struct ipv6hdr *tmp_hdr;
 	struct frag_hdr *fh;
@@ -561,12 +561,12 @@ static int ip6_fragment(struct sk_buff **pskb, int (*output)(struct sk_buff**))
 		err = 0;
 		offset = 0;
 		frag = skb_shinfo(skb)->frag_list;
-		skb_shinfo(skb)->frag_list = 0;
+		skb_shinfo(skb)->frag_list = NULL;
 		/* BUILD HEADER */
 
 		tmp_hdr = kmalloc(hlen, GFP_ATOMIC);
 		if (!tmp_hdr) {
-			IP6_INC_STATS(FragFails);
+			IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS);
 			return -ENOMEM;
 		}
 
@@ -593,6 +593,7 @@ static int ip6_fragment(struct sk_buff **pskb, int (*output)(struct sk_buff**))
 			/* Prepare header of the next frame,
 			 * before previous one went down. */
 			if (frag) {
+				frag->ip_summed = CHECKSUM_NONE;
 				frag->h.raw = frag->data;
 				fh = (struct frag_hdr*)__skb_push(frag, sizeof(struct frag_hdr));
 				frag->nh.raw = __skb_push(frag, hlen);
@@ -607,13 +608,11 @@ static int ip6_fragment(struct sk_buff **pskb, int (*output)(struct sk_buff**))
 				frag->nh.ipv6h->payload_len = htons(frag->len - sizeof(struct ipv6hdr));
 				ip6_copy_metadata(frag, skb);
 			}
-			err = output(pskb);
-			if (err || !frag) {
-				if (unlikely(skb != *pskb))
-					skb = *pskb;
-				break;
-			}
 			
+			err = output(skb);
+			if (err || !frag)
+				break;
+
 			skb = frag;
 			frag = skb->next;
 			skb->next = NULL;
@@ -623,7 +622,7 @@ static int ip6_fragment(struct sk_buff **pskb, int (*output)(struct sk_buff**))
 			kfree(tmp_hdr);
 
 		if (err == 0) {
-			IP6_INC_STATS(FragOKs);
+			IP6_INC_STATS(IPSTATS_MIB_FRAGOKS);
 			return 0;
 		}
 
@@ -633,7 +632,7 @@ static int ip6_fragment(struct sk_buff **pskb, int (*output)(struct sk_buff**))
 			frag = skb;
 		}
 
-		IP6_INC_STATS(FragFails);
+		IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS);
 		return err;
 	}
 
@@ -666,7 +665,7 @@ slow_path:
 
 		if ((frag = alloc_skb(len+hlen+sizeof(struct frag_hdr)+LL_RESERVED_SPACE(rt->u.dst.dev), GFP_ATOMIC)) == NULL) {
 			NETDEBUG(printk(KERN_INFO "IPv6: frag: no memory for new fragment!\n"));
-			IP6_INC_STATS(FragFails);
+			IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS);
 			err = -ENOMEM;
 			goto fail;
 		}
@@ -724,19 +723,19 @@ slow_path:
 		 *	Put this fragment into the sending queue.
 		 */
 
-		IP6_INC_STATS(FragCreates);
+		IP6_INC_STATS(IPSTATS_MIB_FRAGCREATES);
 
-		err = output(&frag);
+		err = output(frag);
 		if (err)
 			goto fail;
 	}
 	kfree_skb(skb);
-	IP6_INC_STATS(FragOKs);
+	IP6_INC_STATS(IPSTATS_MIB_FRAGOKS);
 	return err;
 
 fail:
 	kfree_skb(skb); 
-	IP6_INC_STATS(FragFails);
+	IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS);
 	return err;
 }
 
@@ -748,7 +747,7 @@ int ip6_dst_lookup(struct sock *sk, struct dst_entry **dst, struct flowi *fl)
 	if (sk) {
 		struct ipv6_pinfo *np = inet6_sk(sk);
 	
-		*dst = __sk_dst_check(sk, np->dst_cookie);
+		*dst = sk_dst_check(sk, np->dst_cookie);
 		if (*dst) {
 			struct rt6_info *rt = (struct rt6_info*)*dst;
 	
@@ -771,13 +770,13 @@ int ip6_dst_lookup(struct sock *sk, struct dst_entry **dst, struct flowi *fl)
 				 */
 	
 			if (((rt->rt6i_dst.plen != 128 ||
-			      ipv6_addr_cmp(&fl->fl6_dst, &rt->rt6i_dst.addr))
+			      !ipv6_addr_equal(&fl->fl6_dst, &rt->rt6i_dst.addr))
 			     && (np->daddr_cache == NULL ||
-				 ipv6_addr_cmp(&fl->fl6_dst, np->daddr_cache)))
+				 !ipv6_addr_equal(&fl->fl6_dst, np->daddr_cache)))
 			    || (fl->oif && fl->oif != (*dst)->dev->ifindex)) {
+				dst_release(*dst);
 				*dst = NULL;
-			} else
-				dst_hold(*dst);
+			}
 		}
 	}
 
@@ -798,10 +797,6 @@ int ip6_dst_lookup(struct sock *sk, struct dst_entry **dst, struct flowi *fl)
 			goto out_err_release;
 		}
 	}
-	if ((err = xfrm_lookup(dst, fl, sk, 0)) < 0) {
-		err = -ENETUNREACH;
-		goto out_err_release;
-        }
 
 	return 0;
 
@@ -816,14 +811,14 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to, int offse
 		    int hlimit, struct ipv6_txoptions *opt, struct flowi *fl, struct rt6_info *rt,
 		    unsigned int flags)
 {
-	struct inet_opt *inet = inet_sk(sk);
+	struct inet_sock *inet = inet_sk(sk);
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct sk_buff *skb;
 	unsigned int maxfraglen, fragheaderlen;
 	int exthdrlen;
 	int hh_len;
 	int mtu;
-	int copy = 0;
+	int copy;
 	int err;
 	int offset = 0;
 	int csummode = CHECKSUM_NONE;
@@ -854,8 +849,8 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to, int offse
 		np->cork.hop_limit = hlimit;
 		inet->cork.fragsize = mtu = dst_pmtu(&rt->u.dst);
 		inet->cork.length = 0;
-		inet->sndmsg_page = NULL;
-		inet->sndmsg_off = 0;
+		sk->sk_sndmsg_page = NULL;
+		sk->sk_sndmsg_off = 0;
 		exthdrlen = rt->u.dst.header_len + (opt ? opt->opt_flen : 0);
 		length += exthdrlen;
 		transhdrlen += exthdrlen;
@@ -881,29 +876,79 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to, int offse
 		}
 	}
 
+	/*
+	 * Let's try using as much space as possible.
+	 * Use MTU if total length of the message fits into the MTU.
+	 * Otherwise, we need to reserve fragment header and
+	 * fragment alignment (= 8-15 octects, in total).
+	 *
+	 * Note that we may need to "move" the data from the tail of
+	 * of the buffer to the new fragment when we split 
+	 * the message.
+	 *
+	 * FIXME: It may be fragmented into multiple chunks 
+	 *        at once if non-fragmentable extension headers
+	 *        are too large.
+	 * --yoshfuji 
+	 */
+
 	inet->cork.length += length;
 
 	if ((skb = skb_peek_tail(&sk->sk_write_queue)) == NULL)
 		goto alloc_new_skb;
 
 	while (length > 0) {
-		if ((copy = maxfraglen - skb->len) <= 0) {
+		/* Check if the remaining data fits into current packet. */
+		copy = mtu - skb->len;
+		if (copy < length)
+			copy = maxfraglen - skb->len;
+
+		if (copy <= 0) {
 			char *data;
 			unsigned int datalen;
 			unsigned int fraglen;
+			unsigned int fraggap;
 			unsigned int alloclen;
-			BUG_TRAP(copy == 0);
+			struct sk_buff *skb_prev;
 alloc_new_skb:
-			datalen = maxfraglen - fragheaderlen;
-			if (datalen > length)
-				datalen = length;
+			skb_prev = skb;
+
+			/* There's no room in the current skb */
+			if (skb_prev)
+				fraggap = skb_prev->len - maxfraglen;
+			else
+				fraggap = 0;
+
+			/*
+			 * If remaining data exceeds the mtu,
+			 * we know we need more fragment(s).
+			 */
+			datalen = length + fraggap;
+			if (datalen > mtu - fragheaderlen)
+				datalen = maxfraglen - fragheaderlen;
+
 			fraglen = datalen + fragheaderlen;
 			if ((flags & MSG_MORE) &&
 			    !(rt->u.dst.dev->features&NETIF_F_SG))
-				alloclen = maxfraglen;
+				alloclen = mtu;
 			else
-				alloclen = fraglen;
+				alloclen = datalen + fragheaderlen;
+
+			/*
+			 * The last fragment gets additional space at tail.
+			 * Note: we overallocate on fragments with MSG_MODE
+			 * because we have no idea if we're the last one.
+			 */
+			if (datalen == length + fraggap)
+				alloclen += rt->u.dst.trailer_len;
+
+			/*
+			 * We just reserve space for fragment header.
+			 * Note: this may be overallocation if the message 
+			 * (without MSG_MORE) fits into the MTU.
+			 */
 			alloclen += sizeof(struct frag_hdr);
+
 			if (transhdrlen) {
 				skb = sock_alloc_send_skb(sk,
 						alloclen + hh_len,
@@ -925,7 +970,7 @@ alloc_new_skb:
 			 */
 			skb->ip_summed = csummode;
 			skb->csum = 0;
-			/* reserve 8 byte for fragmentation */
+			/* reserve for fragmentation */
 			skb_reserve(skb, hh_len+sizeof(struct frag_hdr));
 
 			/*
@@ -935,15 +980,29 @@ alloc_new_skb:
 			skb->nh.raw = data + exthdrlen;
 			data += fragheaderlen;
 			skb->h.raw = data + exthdrlen;
-			copy = datalen - transhdrlen;
-			if (copy > 0 && getfrag(from, data + transhdrlen, offset, copy, 0, skb) < 0) {
+
+			if (fraggap) {
+				skb->csum = skb_copy_and_csum_bits(
+					skb_prev, maxfraglen,
+					data + transhdrlen, fraggap, 0);
+				skb_prev->csum = csum_sub(skb_prev->csum,
+							  skb->csum);
+				data += fraggap;
+				skb_trim(skb_prev, maxfraglen);
+			}
+			copy = datalen - transhdrlen - fraggap;
+			if (copy < 0) {
+				err = -EINVAL;
+				kfree_skb(skb);
+				goto error;
+			} else if (copy > 0 && getfrag(from, data + transhdrlen, offset, copy, fraggap, skb) < 0) {
 				err = -EFAULT;
 				kfree_skb(skb);
 				goto error;
 			}
 
 			offset += copy;
-			length -= datalen;
+			length -= datalen - fraggap;
 			transhdrlen = 0;
 			exthdrlen = 0;
 			csummode = CHECKSUM_NONE;
@@ -971,8 +1030,8 @@ alloc_new_skb:
 		} else {
 			int i = skb_shinfo(skb)->nr_frags;
 			skb_frag_t *frag = &skb_shinfo(skb)->frags[i-1];
-			struct page *page = inet->sndmsg_page;
-			int off = inet->sndmsg_off;
+			struct page *page = sk->sk_sndmsg_page;
+			int off = sk->sk_sndmsg_off;
 			unsigned int left;
 
 			if (page && (left = PAGE_SIZE - off) > 0) {
@@ -984,7 +1043,7 @@ alloc_new_skb:
 						goto error;
 					}
 					get_page(page);
-					skb_fill_page_desc(skb, i, page, inet->sndmsg_off, 0);
+					skb_fill_page_desc(skb, i, page, sk->sk_sndmsg_off, 0);
 					frag = &skb_shinfo(skb)->frags[i];
 				}
 			} else if(i < MAX_SKB_FRAGS) {
@@ -995,8 +1054,8 @@ alloc_new_skb:
 					err = -ENOMEM;
 					goto error;
 				}
-				inet->sndmsg_page = page;
-				inet->sndmsg_off = 0;
+				sk->sk_sndmsg_page = page;
+				sk->sk_sndmsg_off = 0;
 
 				skb_fill_page_desc(skb, i, page, 0, 0);
 				frag = &skb_shinfo(skb)->frags[i];
@@ -1010,7 +1069,7 @@ alloc_new_skb:
 				err = -EFAULT;
 				goto error;
 			}
-			inet->sndmsg_off += copy;
+			sk->sk_sndmsg_off += copy;
 			frag->size += copy;
 			skb->len += copy;
 			skb->data_len += copy;
@@ -1021,7 +1080,7 @@ alloc_new_skb:
 	return 0;
 error:
 	inet->cork.length -= length;
-	IP6_INC_STATS(OutDiscards);
+	IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
 	return err;
 }
 
@@ -1030,7 +1089,7 @@ int ip6_push_pending_frames(struct sock *sk)
 	struct sk_buff *skb, *tmp_skb;
 	struct sk_buff **tail_skb;
 	struct in6_addr final_dst_buf, *final_dst = &final_dst_buf;
-	struct inet_opt *inet = inet_sk(sk);
+	struct inet_sock *inet = inet_sk(sk);
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct ipv6hdr *hdr;
 	struct ipv6_txoptions *opt = np->cork.opt;
@@ -1081,7 +1140,7 @@ int ip6_push_pending_frames(struct sock *sk)
 	ipv6_addr_copy(&hdr->daddr, final_dst);
 
 	skb->dst = dst_clone(&rt->u.dst);
-	IP6_INC_STATS(OutRequests);	
+	IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS);	
 	err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dst->dev, dst_output);
 	if (err) {
 		if (err > 0)
@@ -1108,12 +1167,12 @@ error:
 
 void ip6_flush_pending_frames(struct sock *sk)
 {
-	struct inet_opt *inet = inet_sk(sk);
+	struct inet_sock *inet = inet_sk(sk);
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct sk_buff *skb;
 
 	while ((skb = __skb_dequeue_tail(&sk->sk_write_queue)) != NULL) {
-		IP6_INC_STATS(OutDiscards);
+		IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
 		kfree_skb(skb);
 	}