X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=net%2Fipv6%2Fip6_output.c;h=fce2a87e9e4a6f70bb89006303fc03db6b769a77;hb=6a77f38946aaee1cd85eeec6cf4229b204c15071;hp=81eed71ddfe3fd5a82755306b2506b5392615427;hpb=9bf4aaab3e101692164d49b7ca357651eb691cb6;p=linux-2.6.git diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 81eed71dd..fce2a87e9 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -54,13 +54,14 @@ #include #include #include +#include -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); @@ -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; @@ -145,14 +145,12 @@ static int ip6_output2(struct sk_buff **pskb) 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 @@ -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; @@ -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; @@ -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); @@ -608,7 +609,7 @@ static int ip6_fragment(struct sk_buff **pskb, int (*output)(struct sk_buff**)) ip6_copy_metadata(frag, skb); } - err = output(&skb); + err = output(skb); if (err || !frag) break; @@ -724,7 +725,7 @@ slow_path: IP6_INC_STATS(IPSTATS_MIB_FRAGCREATES); - err = output(&frag); + err = output(frag); if (err) goto fail; } @@ -746,7 +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; @@ -769,13 +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); + } } } @@ -796,10 +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; @@ -814,14 +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; @@ -879,29 +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, @@ -923,7 +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)); /* @@ -933,15 +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; @@ -1028,7 +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; @@ -1106,7 +1167,7 @@ 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;