X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=net%2Fipv6%2Fip6_output.c;h=33260cfacd35a1215b7c6f4e7b2b255acdb86f39;hb=c7b5ebbddf7bcd3651947760f423e3783bbe6573;hp=d6f0db4aaa069a781b3497a622388ec725e0dbb5;hpb=5273a3df6485dc2ad6aa7ddd441b9a21970f003b;p=linux-2.6.git diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index d6f0db4aa..33260cfac 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -54,8 +54,9 @@ #include #include #include +#include -static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*)); +static int ip6_fragment(struct sk_buff **pskb, int (*output)(struct sk_buff**)); static __inline__ void ipv6_select_ident(struct sk_buff *skb, struct frag_hdr *fhdr) { @@ -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(Ip6OutNoRoutes); + IP6_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES); kfree_skb(skb); return -EINVAL; @@ -107,8 +108,9 @@ static int ip6_dev_loopback_xmit(struct sk_buff *newskb) } -int ip6_output2(struct sk_buff *skb) +static int ip6_output2(struct sk_buff **pskb) { + struct sk_buff *skb = *pskb; struct dst_entry *dst = skb->dst; struct net_device *dev = dst->dev; @@ -132,24 +134,26 @@ int ip6_output2(struct sk_buff *skb) ip6_dev_loopback_xmit); if (skb->nh.ipv6h->hop_limit == 0) { - IP6_INC_STATS(Ip6OutDiscards); + IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); return 0; } } - IP6_INC_STATS(Ip6OutMcastPkts); + 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 *skb) +int ip6_output(struct sk_buff **pskb) { + struct sk_buff *skb = *pskb; + if ((skb->len > dst_pmtu(skb->dst) || skb_shinfo(skb)->frag_list)) - return ip6_fragment(skb, ip6_output2); + return ip6_fragment(pskb, ip6_output2); else - return ip6_output2(skb); + return ip6_output2(pskb); } #ifdef CONFIG_NETFILTER @@ -169,7 +173,7 @@ int ip6_route_me_harder(struct sk_buff *skb) dst = ip6_route_output(skb->sk, &fl); if (dst->error) { - IP6_INC_STATS(Ip6OutNoRoutes); + IP6_INC_STATS(IPSTATS_MIB_OUTNOROUTES); LIMIT_NETDEBUG( printk(KERN_DEBUG "ip6_route_me_harder: No more route.\n")); dst_release(dst); @@ -228,7 +232,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, kfree_skb(skb); skb = skb2; if (skb == NULL) { - IP6_INC_STATS(Ip6OutDiscards); + IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); return -ENOBUFS; } if (sk) @@ -262,7 +266,7 @@ 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(Ip6OutRequests); + IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); return NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, dst->dev, ip6_maybe_reroute); } @@ -270,7 +274,7 @@ 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(Ip6FragFails); + IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return -EMSGSIZE; } @@ -346,13 +350,13 @@ int ip6_forward(struct sk_buff *skb) { struct dst_entry *dst = skb->dst; struct ipv6hdr *hdr = skb->nh.ipv6h; - struct inet6_skb_parm *opt =(struct inet6_skb_parm*)skb->cb; + struct inet6_skb_parm *opt = IP6CB(skb); if (ipv6_devconf.forwarding == 0) goto error; if (!xfrm6_policy_check(NULL, XFRM_POLICY_FWD, skb)) { - IP6_INC_STATS(Ip6InDiscards); + IP6_INC_STATS(IPSTATS_MIB_INDISCARDS); goto drop; } @@ -391,7 +395,7 @@ int ip6_forward(struct sk_buff *skb) } if (!xfrm6_route_forward(skb)) { - IP6_INC_STATS(Ip6InDiscards); + IP6_INC_STATS(IPSTATS_MIB_INDISCARDS); goto drop; } @@ -429,14 +433,14 @@ 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(Ip6InTooBigErrors); - IP6_INC_STATS_BH(Ip6FragFails); + 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(Ip6OutDiscards); + IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); goto drop; } @@ -446,11 +450,11 @@ int ip6_forward(struct sk_buff *skb) hdr->hop_limit--; - IP6_INC_STATS_BH(Ip6OutForwDatagrams); + 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(Ip6InAddrErrors); + IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS); drop: kfree_skb(skb); return -EINVAL; @@ -473,6 +477,7 @@ 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; @@ -513,11 +518,11 @@ int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr) return offset; } -static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*)) +static int ip6_fragment(struct sk_buff **pskb, int (*output)(struct sk_buff**)) { struct net_device *dev; + struct sk_buff *frag, *skb = *pskb; struct rt6_info *rt = (struct rt6_info*)skb->dst; - struct sk_buff *frag; struct ipv6hdr *tmp_hdr; struct frag_hdr *fh; unsigned int mtu, hlen, left, len; @@ -558,12 +563,12 @@ static int ip6_fragment(struct sk_buff *skb, 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(Ip6FragFails); + IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS); return -ENOMEM; } @@ -604,8 +609,8 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*)) frag->nh.ipv6h->payload_len = htons(frag->len - sizeof(struct ipv6hdr)); ip6_copy_metadata(frag, skb); } - err = output(skb); - + + err = output(&skb); if (err || !frag) break; @@ -618,7 +623,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*)) kfree(tmp_hdr); if (err == 0) { - IP6_INC_STATS(Ip6FragOKs); + IP6_INC_STATS(IPSTATS_MIB_FRAGOKS); return 0; } @@ -628,7 +633,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*)) frag = skb; } - IP6_INC_STATS(Ip6FragFails); + IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS); return err; } @@ -661,7 +666,7 @@ 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(Ip6FragFails); + IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS); err = -ENOMEM; goto fail; } @@ -719,19 +724,19 @@ slow_path: * Put this fragment into the sending queue. */ - IP6_INC_STATS(Ip6FragCreates); + IP6_INC_STATS(IPSTATS_MIB_FRAGCREATES); - err = output(frag); + err = output(&frag); if (err) goto fail; } kfree_skb(skb); - IP6_INC_STATS(Ip6FragOKs); + IP6_INC_STATS(IPSTATS_MIB_FRAGOKS); return err; fail: kfree_skb(skb); - IP6_INC_STATS(Ip6FragFails); + IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS); return err; } @@ -793,10 +798,6 @@ 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; @@ -818,7 +819,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to, int offse int exthdrlen; int hh_len; int mtu; - int copy = 0; + int copy; int err; int offset = 0; int csummode = CHECKSUM_NONE; @@ -849,8 +850,8 @@ 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; @@ -876,29 +877,79 @@ 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, @@ -920,7 +971,7 @@ 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)); /* @@ -930,15 +981,29 @@ 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; @@ -966,8 +1031,8 @@ 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) { @@ -979,7 +1044,7 @@ 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) { @@ -990,8 +1055,8 @@ 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]; @@ -1005,7 +1070,7 @@ 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; @@ -1016,7 +1081,7 @@ alloc_new_skb: return 0; error: inet->cork.length -= length; - IP6_INC_STATS(Ip6OutDiscards); + IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); return err; } @@ -1076,7 +1141,7 @@ 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(Ip6OutRequests); + 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) @@ -1108,7 +1173,7 @@ void ip6_flush_pending_frames(struct sock *sk) struct sk_buff *skb; while ((skb = __skb_dequeue_tail(&sk->sk_write_queue)) != NULL) { - IP6_INC_STATS(Ip6OutDiscards); + IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); }