From 56d417b12e57dfe11c9b7ba4bea3882c62a55815 Mon Sep 17 00:00:00 2001 From: Brian Haley Date: Mon, 1 Jun 2009 03:07:33 -0700 Subject: IPv6: Add 'autoconf' and 'disable_ipv6' module parameters Add 'autoconf' and 'disable_ipv6' parameters to the IPv6 module. The first controls if IPv6 addresses are autoconfigured from prefixes received in Router Advertisements. The IPv6 loopback (::1) and link-local addresses are still configured. The second controls if IPv6 addresses are desired at all. No IPv6 addresses will be added to any interfaces. Signed-off-by: Brian Haley Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 83 +++++++++++++++++++++++++++++++++++++++++++++++++---- net/ipv6/af_inet6.c | 22 ++++++++++---- 2 files changed, 95 insertions(+), 10 deletions(-) (limited to 'net/ipv6') diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 31938e5fb220..c3488372f12d 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -591,7 +591,6 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, int pfxlen, { struct inet6_ifaddr *ifa = NULL; struct rt6_info *rt; - struct net *net = dev_net(idev->dev); int hash; int err = 0; int addr_type = ipv6_addr_type(addr); @@ -608,7 +607,7 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, int pfxlen, goto out2; } - if (idev->cnf.disable_ipv6 || net->ipv6.devconf_all->disable_ipv6) { + if (idev->cnf.disable_ipv6) { err = -EACCES; goto out2; } @@ -1752,6 +1751,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len) __u32 prefered_lft; int addr_type; struct inet6_dev *in6_dev; + struct net *net = dev_net(dev); pinfo = (struct prefix_info *) opt; @@ -1809,7 +1809,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len) if (addrconf_finite_timeout(rt_expires)) rt_expires *= HZ; - rt = rt6_lookup(dev_net(dev), &pinfo->prefix, NULL, + rt = rt6_lookup(net, &pinfo->prefix, NULL, dev->ifindex, 1); if (rt && addrconf_is_prefix_route(rt)) { @@ -1846,7 +1846,6 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len) struct inet6_ifaddr * ifp; struct in6_addr addr; int create = 0, update_lft = 0; - struct net *net = dev_net(dev); if (pinfo->prefix_len == 64) { memcpy(&addr, &pinfo->prefix, 8); @@ -3988,6 +3987,75 @@ static int addrconf_sysctl_forward_strategy(ctl_table *table, return addrconf_fixup_forwarding(table, valp, val); } +static void dev_disable_change(struct inet6_dev *idev) +{ + if (!idev || !idev->dev) + return; + + if (idev->cnf.disable_ipv6) + addrconf_notify(NULL, NETDEV_DOWN, idev->dev); + else + addrconf_notify(NULL, NETDEV_UP, idev->dev); +} + +static void addrconf_disable_change(struct net *net, __s32 newf) +{ + struct net_device *dev; + struct inet6_dev *idev; + + read_lock(&dev_base_lock); + for_each_netdev(net, dev) { + rcu_read_lock(); + idev = __in6_dev_get(dev); + if (idev) { + int changed = (!idev->cnf.disable_ipv6) ^ (!newf); + idev->cnf.disable_ipv6 = newf; + if (changed) + dev_disable_change(idev); + } + rcu_read_unlock(); + } + read_unlock(&dev_base_lock); +} + +static int addrconf_disable_ipv6(struct ctl_table *table, int *p, int old) +{ + struct net *net; + + net = (struct net *)table->extra2; + + if (p == &net->ipv6.devconf_dflt->disable_ipv6) + return 0; + + if (!rtnl_trylock()) + return restart_syscall(); + + if (p == &net->ipv6.devconf_all->disable_ipv6) { + __s32 newf = net->ipv6.devconf_all->disable_ipv6; + net->ipv6.devconf_dflt->disable_ipv6 = newf; + addrconf_disable_change(net, newf); + } else if ((!*p) ^ (!old)) + dev_disable_change((struct inet6_dev *)table->extra1); + + rtnl_unlock(); + return 0; +} + +static +int addrconf_sysctl_disable(ctl_table *ctl, int write, struct file * filp, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int *valp = ctl->data; + int val = *valp; + int ret; + + ret = proc_dointvec(ctl, write, filp, buffer, lenp, ppos); + + if (write) + ret = addrconf_disable_ipv6(ctl, valp, val); + return ret; +} + static struct addrconf_sysctl_table { struct ctl_table_header *sysctl_header; @@ -4225,7 +4293,8 @@ static struct addrconf_sysctl_table .data = &ipv6_devconf.disable_ipv6, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = addrconf_sysctl_disable, + .strategy = sysctl_intvec, }, { .ctl_name = CTL_UNNUMBERED, @@ -4346,6 +4415,10 @@ static int addrconf_init_net(struct net *net) dflt = kmemdup(dflt, sizeof(ipv6_devconf_dflt), GFP_KERNEL); if (dflt == NULL) goto err_alloc_dflt; + } else { + /* these will be inherited by all namespaces */ + dflt->autoconf = ipv6_defaults.autoconf; + dflt->disable_ipv6 = ipv6_defaults.disable_ipv6; } net->ipv6.devconf_all = all; diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index b6215be0963f..85b3d0036afd 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -72,9 +72,21 @@ MODULE_LICENSE("GPL"); static struct list_head inetsw6[SOCK_MAX]; static DEFINE_SPINLOCK(inetsw6_lock); -static int disable_ipv6 = 0; -module_param_named(disable, disable_ipv6, int, 0); -MODULE_PARM_DESC(disable, "Disable IPv6 such that it is non-functional"); +struct ipv6_params ipv6_defaults = { + .disable_ipv6 = 0, + .autoconf = 1, +}; + +static int disable_ipv6_mod = 0; + +module_param_named(disable, disable_ipv6_mod, int, 0444); +MODULE_PARM_DESC(disable, "Disable IPv6 module such that it is non-functional"); + +module_param_named(disable_ipv6, ipv6_defaults.disable_ipv6, int, 0444); +MODULE_PARM_DESC(disable_ipv6, "Disable IPv6 on all interfaces"); + +module_param_named(autoconf, ipv6_defaults.autoconf, int, 0444); +MODULE_PARM_DESC(autoconf, "Enable IPv6 address autoconfiguration on all interfaces"); static __inline__ struct ipv6_pinfo *inet6_sk_generic(struct sock *sk) { @@ -1038,7 +1050,7 @@ static int __init inet6_init(void) for(r = &inetsw6[0]; r < &inetsw6[SOCK_MAX]; ++r) INIT_LIST_HEAD(r); - if (disable_ipv6) { + if (disable_ipv6_mod) { printk(KERN_INFO "IPv6: Loaded, but administratively disabled, " "reboot required to enable\n"); @@ -1227,7 +1239,7 @@ module_init(inet6_init); static void __exit inet6_exit(void) { - if (disable_ipv6) + if (disable_ipv6_mod) return; /* First of all disallow new sockets creation. */ -- cgit v1.2.3 From dae9de8e13a8a5154688e4c788e65399b4718707 Mon Sep 17 00:00:00 2001 From: Brian Haley Date: Tue, 2 Jun 2009 00:20:26 -0700 Subject: IPv6: Print error value when skb allocation fails Print-out the error value when sock_alloc_send_skb() fails in the IPv6 neighbor discovery code - can be useful for debugging. Signed-off-by: Brian Haley Signed-off-by: David S. Miller --- net/ipv6/ndisc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net/ipv6') diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index e09f12ee57cf..1d13d9964985 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -465,8 +465,8 @@ struct sk_buff *ndisc_build_skb(struct net_device *dev, 1, &err); if (!skb) { ND_PRINTK0(KERN_ERR - "ICMPv6 ND: %s() failed to allocate an skb.\n", - __func__); + "ICMPv6 ND: %s() failed to allocate an skb, err=%d.\n", + __func__, err); return NULL; } @@ -1562,8 +1562,8 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, 1, &err); if (buff == NULL) { ND_PRINTK0(KERN_ERR - "ICMPv6 Redirect: %s() failed to allocate an skb.\n", - __func__); + "ICMPv6 Redirect: %s() failed to allocate an skb, err=%d.\n", + __func__, err); goto release; } -- cgit v1.2.3 From adf30907d63893e4208dfe3f5c88ae12bc2f25d5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 2 Jun 2009 05:19:30 +0000 Subject: net: skb->dst accessors Define three accessors to get/set dst attached to a skb struct dst_entry *skb_dst(const struct sk_buff *skb) void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst) void skb_dst_drop(struct sk_buff *skb) This one should replace occurrences of : dst_release(skb->dst) skb->dst = NULL; Delete skb->dst field Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/exthdrs.c | 40 +++++++++++++-------------- net/ipv6/inet6_connection_sock.c | 2 +- net/ipv6/ip6_input.c | 12 ++++---- net/ipv6/ip6_output.c | 60 ++++++++++++++++++++-------------------- net/ipv6/ip6_tunnel.c | 26 ++++++++--------- net/ipv6/ip6mr.c | 13 ++++----- net/ipv6/mcast.c | 17 +++++++----- net/ipv6/ndisc.c | 4 +-- net/ipv6/netfilter.c | 16 +++++++---- net/ipv6/netfilter/ip6t_REJECT.c | 2 +- net/ipv6/raw.c | 2 +- net/ipv6/reassembly.c | 22 +++++++-------- net/ipv6/route.c | 12 ++++---- net/ipv6/sit.c | 21 +++++++------- net/ipv6/tcp_ipv6.c | 8 ++++-- net/ipv6/udp.c | 7 ++--- net/ipv6/xfrm6_mode_tunnel.c | 4 +-- net/ipv6/xfrm6_output.c | 4 +-- 18 files changed, 139 insertions(+), 133 deletions(-) (limited to 'net/ipv6') diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 1c7f400a3cfe..4aae658e5501 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -277,7 +277,7 @@ static int ipv6_destopt_rcv(struct sk_buff *skb) if (!pskb_may_pull(skb, skb_transport_offset(skb) + 8) || !pskb_may_pull(skb, (skb_transport_offset(skb) + ((skb_transport_header(skb)[1] + 1) << 3)))) { - IP6_INC_STATS_BH(dev_net(skb->dst->dev), ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); kfree_skb(skb); return -1; @@ -288,7 +288,7 @@ static int ipv6_destopt_rcv(struct sk_buff *skb) dstbuf = opt->dst1; #endif - dst = dst_clone(skb->dst); + dst = dst_clone(skb_dst(skb)); if (ip6_parse_tlv(tlvprocdestopt_lst, skb)) { dst_release(dst); skb->transport_header += (skb_transport_header(skb)[1] + 1) << 3; @@ -333,7 +333,7 @@ static int ipv6_rthdr_rcv(struct sk_buff *skb) if (!pskb_may_pull(skb, skb_transport_offset(skb) + 8) || !pskb_may_pull(skb, (skb_transport_offset(skb) + ((skb_transport_header(skb)[1] + 1) << 3)))) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); kfree_skb(skb); return -1; @@ -343,7 +343,7 @@ static int ipv6_rthdr_rcv(struct sk_buff *skb) if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr) || skb->pkt_type != PACKET_HOST) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; @@ -358,7 +358,7 @@ looped_back: * processed by own */ if (!addr) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; @@ -384,7 +384,7 @@ looped_back: goto unknown_rh; /* Silently discard invalid RTH type 2 */ if (hdr->hdrlen != 2 || hdr->segments_left != 1) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); kfree_skb(skb); return -1; @@ -403,7 +403,7 @@ looped_back: n = hdr->hdrlen >> 1; if (hdr->segments_left > n) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, ((&hdr->segments_left) - @@ -417,7 +417,7 @@ looped_back: if (skb_cloned(skb)) { /* the copy is a forwarded packet */ if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); return -1; @@ -440,13 +440,13 @@ looped_back: if (xfrm6_input_addr(skb, (xfrm_address_t *)addr, (xfrm_address_t *)&ipv6_hdr(skb)->saddr, IPPROTO_ROUTING) < 0) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; } - if (!ipv6_chk_home_addr(dev_net(skb->dst->dev), addr)) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + if (!ipv6_chk_home_addr(dev_net(skb_dst(skb)->dev), addr)) { + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; @@ -458,7 +458,7 @@ looped_back: } if (ipv6_addr_is_multicast(addr)) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; @@ -468,17 +468,17 @@ looped_back: ipv6_addr_copy(addr, &ipv6_hdr(skb)->daddr); ipv6_addr_copy(&ipv6_hdr(skb)->daddr, &daddr); - dst_release(xchg(&skb->dst, NULL)); + skb_dst_drop(skb); ip6_route_input(skb); - if (skb->dst->error) { + if (skb_dst(skb)->error) { skb_push(skb, skb->data - skb_network_header(skb)); dst_input(skb); return -1; } - if (skb->dst->dev->flags&IFF_LOOPBACK) { + if (skb_dst(skb)->dev->flags&IFF_LOOPBACK) { if (ipv6_hdr(skb)->hop_limit <= 1) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, 0, skb->dev); @@ -494,7 +494,7 @@ looped_back: return -1; unknown_rh: - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->type) - skb_network_header(skb)); return -1; @@ -552,11 +552,11 @@ void ipv6_exthdrs_exit(void) **********************************/ /* - * Note: we cannot rely on skb->dst before we assign it in ip6_route_input(). + * Note: we cannot rely on skb_dst(skb) before we assign it in ip6_route_input(). */ static inline struct inet6_dev *ipv6_skb_idev(struct sk_buff *skb) { - return skb->dst ? ip6_dst_idev(skb->dst) : __in6_dev_get(skb->dev); + return skb_dst(skb) ? ip6_dst_idev(skb_dst(skb)) : __in6_dev_get(skb->dev); } /* Router Alert as of RFC 2711 */ @@ -581,7 +581,7 @@ static int ipv6_hop_jumbo(struct sk_buff *skb, int optoff) { const unsigned char *nh = skb_network_header(skb); u32 pkt_len; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); if (nh[optoff + 1] != 4 || (optoff & 3) != 2) { LIMIT_NETDEBUG(KERN_DEBUG "ipv6_hop_jumbo: wrong jumbo opt length/alignment %d\n", diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index 3c3732d50c1a..cc4797dd8325 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -228,7 +228,7 @@ int inet6_csk_xmit(struct sk_buff *skb, int ipfragok) __inet6_csk_dst_store(sk, dst, NULL, NULL); } - skb->dst = dst_clone(dst); + skb_dst_set(skb, dst_clone(dst)); /* Restore final destination back after routing done */ ipv6_addr_copy(&fl.fl6_dst, &np->daddr); diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index bc1a920c34a1..c3a07d75b5f5 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -48,7 +48,7 @@ inline int ip6_rcv_finish( struct sk_buff *skb) { - if (skb->dst == NULL) + if (skb_dst(skb) == NULL) ip6_route_input(skb); return dst_input(skb); @@ -91,7 +91,7 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt * arrived via the sending interface (ethX), because of the * nature of scoping architecture. --yoshfuji */ - IP6CB(skb)->iif = skb->dst ? ip6_dst_idev(skb->dst)->dev->ifindex : dev->ifindex; + IP6CB(skb)->iif = skb_dst(skb) ? ip6_dst_idev(skb_dst(skb))->dev->ifindex : dev->ifindex; if (unlikely(!pskb_may_pull(skb, sizeof(*hdr)))) goto err; @@ -161,7 +161,7 @@ static int ip6_input_finish(struct sk_buff *skb) int nexthdr, raw; u8 hash; struct inet6_dev *idev; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); /* * Parse extension headers @@ -169,7 +169,7 @@ static int ip6_input_finish(struct sk_buff *skb) rcu_read_lock(); resubmit: - idev = ip6_dst_idev(skb->dst); + idev = ip6_dst_idev(skb_dst(skb)); if (!pskb_pull(skb, skb_transport_offset(skb))) goto discard; nhoff = IP6CB(skb)->nhoff; @@ -242,8 +242,8 @@ int ip6_mc_input(struct sk_buff *skb) struct ipv6hdr *hdr; int deliver; - IP6_UPD_PO_STATS_BH(dev_net(skb->dst->dev), - ip6_dst_idev(skb->dst), IPSTATS_MIB_INMCAST, + IP6_UPD_PO_STATS_BH(dev_net(skb_dst(skb)->dev), + ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INMCAST, skb->len); hdr = ipv6_hdr(skb); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 735a2bf4b5f1..c8dc8e5a822f 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -78,7 +78,7 @@ int __ip6_local_out(struct sk_buff *skb) len = 0; ipv6_hdr(skb)->payload_len = htons(len); - return nf_hook(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, skb->dst->dev, + return nf_hook(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, skb_dst(skb)->dev, dst_output); } @@ -96,7 +96,7 @@ EXPORT_SYMBOL_GPL(ip6_local_out); static int ip6_output_finish(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); if (dst->hh) return neigh_hh_output(dst->hh, skb); @@ -117,7 +117,7 @@ static int ip6_dev_loopback_xmit(struct sk_buff *newskb) __skb_pull(newskb, skb_network_offset(newskb)); newskb->pkt_type = PACKET_LOOPBACK; newskb->ip_summed = CHECKSUM_UNNECESSARY; - WARN_ON(!newskb->dst); + WARN_ON(!skb_dst(newskb)); netif_rx(newskb); return 0; @@ -126,7 +126,7 @@ static int ip6_dev_loopback_xmit(struct sk_buff *newskb) static int ip6_output2(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct net_device *dev = dst->dev; skb->protocol = htons(ETH_P_IPV6); @@ -134,7 +134,7 @@ static int ip6_output2(struct sk_buff *skb) if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr)) { struct ipv6_pinfo* np = skb->sk ? inet6_sk(skb->sk) : NULL; - struct inet6_dev *idev = ip6_dst_idev(skb->dst); + struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb)); if (!(dev->flags & IFF_LOOPBACK) && (!np || np->mc_loop) && ((mroute6_socket(dev_net(dev)) && @@ -172,21 +172,21 @@ static inline int ip6_skb_dst_mtu(struct sk_buff *skb) struct ipv6_pinfo *np = skb->sk ? inet6_sk(skb->sk) : NULL; return (np && np->pmtudisc == IPV6_PMTUDISC_PROBE) ? - skb->dst->dev->mtu : dst_mtu(skb->dst); + skb_dst(skb)->dev->mtu : dst_mtu(skb_dst(skb)); } int ip6_output(struct sk_buff *skb) { - struct inet6_dev *idev = ip6_dst_idev(skb->dst); + struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb)); if (unlikely(idev->cnf.disable_ipv6)) { - IP6_INC_STATS(dev_net(skb->dst->dev), idev, + IP6_INC_STATS(dev_net(skb_dst(skb)->dev), idev, IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); return 0; } if ((skb->len > ip6_skb_dst_mtu(skb) && !skb_is_gso(skb)) || - dst_allfrag(skb->dst)) + dst_allfrag(skb_dst(skb))) return ip6_fragment(skb, ip6_output2); else return ip6_output2(skb); @@ -202,7 +202,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, struct net *net = sock_net(sk); struct ipv6_pinfo *np = inet6_sk(sk); struct in6_addr *first_hop = &fl->fl6_dst; - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct ipv6hdr *hdr; u8 proto = fl->proto; int seg_len = skb->len; @@ -222,7 +222,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, if (skb_headroom(skb) < head_room) { struct sk_buff *skb2 = skb_realloc_headroom(skb, head_room); if (skb2 == NULL) { - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); return -ENOBUFS; @@ -276,7 +276,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, mtu = dst_mtu(dst); if ((skb->len <= mtu) || skb->local_df || skb_is_gso(skb)) { - IP6_UPD_PO_STATS(net, ip6_dst_idev(skb->dst), + IP6_UPD_PO_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUT, skb->len); return NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, dst->dev, dst_output); @@ -286,7 +286,7 @@ 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(net, ip6_dst_idev(skb->dst), IPSTATS_MIB_FRAGFAILS); + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return -EMSGSIZE; } @@ -416,7 +416,7 @@ static inline int ip6_forward_finish(struct sk_buff *skb) int ip6_forward(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct ipv6hdr *hdr = ipv6_hdr(skb); struct inet6_skb_parm *opt = IP6CB(skb); struct net *net = dev_net(dst->dev); @@ -485,7 +485,7 @@ int ip6_forward(struct sk_buff *skb) IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_INDISCARDS); goto drop; } - dst = skb->dst; + dst = skb_dst(skb); /* IPv6 specs say nothing about it, but it is clear that we cannot send redirects to source routed frames. @@ -566,8 +566,8 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) to->pkt_type = from->pkt_type; to->priority = from->priority; to->protocol = from->protocol; - dst_release(to->dst); - to->dst = dst_clone(from->dst); + skb_dst_drop(to); + skb_dst_set(to, dst_clone(skb_dst(from))); to->dev = from->dev; to->mark = from->mark; @@ -624,7 +624,7 @@ int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr) static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) { struct sk_buff *frag; - struct rt6_info *rt = (struct rt6_info*)skb->dst; + struct rt6_info *rt = (struct rt6_info*)skb_dst(skb); struct ipv6_pinfo *np = skb->sk ? inet6_sk(skb->sk) : NULL; struct ipv6hdr *tmp_hdr; struct frag_hdr *fh; @@ -632,7 +632,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) __be32 frag_id = 0; int ptr, offset = 0, err=0; u8 *prevhdr, nexthdr = 0; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); hlen = ip6_find_1stfragopt(skb, &prevhdr); nexthdr = *prevhdr; @@ -644,9 +644,9 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) * check should be redundant, but it's free.) */ if (!skb->local_df) { - skb->dev = skb->dst->dev; + skb->dev = skb_dst(skb)->dev; icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return -EMSGSIZE; @@ -696,7 +696,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) *prevhdr = NEXTHDR_FRAGMENT; tmp_hdr = kmemdup(skb_network_header(skb), hlen, GFP_ATOMIC); if (!tmp_hdr) { - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); return -ENOMEM; } @@ -809,7 +809,7 @@ slow_path: if ((frag = alloc_skb(len+hlen+sizeof(struct frag_hdr)+LL_ALLOCATED_SPACE(rt->u.dst.dev), GFP_ATOMIC)) == NULL) { NETDEBUG(KERN_INFO "IPv6: frag: no memory for new fragment!\n"); - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); err = -ENOMEM; goto fail; @@ -873,16 +873,16 @@ slow_path: if (err) goto fail; - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGCREATES); } - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGOKS); kfree_skb(skb); return err; fail: - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return err; @@ -1516,10 +1516,10 @@ int ip6_push_pending_frames(struct sock *sk) skb->priority = sk->sk_priority; skb->mark = sk->sk_mark; - skb->dst = dst_clone(&rt->u.dst); + skb_dst_set(skb, dst_clone(&rt->u.dst)); IP6_UPD_PO_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len); if (proto == IPPROTO_ICMPV6) { - struct inet6_dev *idev = ip6_dst_idev(skb->dst); + struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb)); ICMP6MSGOUT_INC_STATS_BH(net, idev, icmp6_hdr(skb)->icmp6_type); ICMP6_INC_STATS_BH(net, idev, ICMP6_MIB_OUTMSGS); @@ -1545,8 +1545,8 @@ void ip6_flush_pending_frames(struct sock *sk) struct sk_buff *skb; while ((skb = __skb_dequeue_tail(&sk->sk_write_queue)) != NULL) { - if (skb->dst) - IP6_INC_STATS(sock_net(sk), ip6_dst_idev(skb->dst), + if (skb_dst(skb)) + IP6_INC_STATS(sock_net(sk), ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); } diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index af256d47fd35..404d16a97d5c 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -532,8 +532,8 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (!skb2) return 0; - dst_release(skb2->dst); - skb2->dst = NULL; + skb_dst_drop(skb2); + skb_pull(skb2, offset); skb_reset_network_header(skb2); eiph = ip_hdr(skb2); @@ -560,21 +560,21 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, ip_rt_put(rt); goto out; } - skb2->dst = (struct dst_entry *)rt; + skb_dst_set(skb2, (struct dst_entry *)rt); } else { ip_rt_put(rt); if (ip_route_input(skb2, eiph->daddr, eiph->saddr, eiph->tos, skb2->dev) || - skb2->dst->dev->type != ARPHRD_TUNNEL) + skb_dst(skb2)->dev->type != ARPHRD_TUNNEL) goto out; } /* change mtu on this route */ if (rel_type == ICMP_DEST_UNREACH && rel_code == ICMP_FRAG_NEEDED) { - if (rel_info > dst_mtu(skb2->dst)) + if (rel_info > dst_mtu(skb_dst(skb2))) goto out; - skb2->dst->ops->update_pmtu(skb2->dst, rel_info); + skb_dst(skb2)->ops->update_pmtu(skb_dst(skb2), rel_info); } icmp_send(skb2, rel_type, rel_code, htonl(rel_info)); @@ -606,8 +606,7 @@ ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (!skb2) return 0; - dst_release(skb2->dst); - skb2->dst = NULL; + skb_dst_drop(skb2); skb_pull(skb2, offset); skb_reset_network_header(skb2); @@ -720,8 +719,7 @@ static int ip6_tnl_rcv(struct sk_buff *skb, __u16 protocol, skb->pkt_type = PACKET_HOST; memset(skb->cb, 0, sizeof(struct inet6_skb_parm)); skb->dev = t->dev; - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); nf_reset(skb); dscp_ecn_decapsulate(t, ipv6h, skb); @@ -885,8 +883,8 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, } if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; - if (skb->dst) - skb->dst->ops->update_pmtu(skb->dst, mtu); + if (skb_dst(skb)) + skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); if (skb->len > mtu) { *pmtu = mtu; err = -EMSGSIZE; @@ -910,8 +908,8 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, kfree_skb(skb); skb = new_skb; } - dst_release(skb->dst); - skb->dst = dst_clone(dst); + skb_dst_drop(skb); + skb_dst_set(skb, dst_clone(dst)); skb->transport_header = skb->network_header; diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 228be551e9c1..a35d8fc55b04 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -398,10 +398,9 @@ static int pim6_rcv(struct sk_buff *skb) skb->protocol = htons(ETH_P_IPV6); skb->ip_summed = 0; skb->pkt_type = PACKET_HOST; - dst_release(skb->dst); + skb_dst_drop(skb); reg_dev->stats.rx_bytes += skb->len; reg_dev->stats.rx_packets++; - skb->dst = NULL; nf_reset(skb); netif_rx(skb); dev_put(reg_dev); @@ -849,7 +848,7 @@ static int ip6mr_cache_report(struct net *net, struct sk_buff *pkt, mifi_t mifi, ipv6_addr_copy(&msg->im6_src, &ipv6_hdr(pkt)->saddr); ipv6_addr_copy(&msg->im6_dst, &ipv6_hdr(pkt)->daddr); - skb->dst = dst_clone(pkt->dst); + skb_dst_set(skb, dst_clone(skb_dst(pkt))); skb->ip_summed = CHECKSUM_UNNECESSARY; } @@ -1487,7 +1486,7 @@ int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg) static inline int ip6mr_forward2_finish(struct sk_buff *skb) { - IP6_INC_STATS_BH(dev_net(skb->dst->dev), ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTFORWDATAGRAMS); return dst_output(skb); } @@ -1532,8 +1531,8 @@ static int ip6mr_forward2(struct sk_buff *skb, struct mfc6_cache *c, int vifi) if (!dst) goto out_free; - dst_release(skb->dst); - skb->dst = dst; + skb_dst_drop(skb); + skb_dst_set(skb, dst); /* * RFC1584 teaches, that DVMRP/PIM router must deliver packets locally @@ -1722,7 +1721,7 @@ int ip6mr_get_route(struct net *net, { int err; struct mfc6_cache *cache; - struct rt6_info *rt = (struct rt6_info *)skb->dst; + struct rt6_info *rt = (struct rt6_info *)skb_dst(skb); read_lock(&mrt_lock); cache = ip6mr_cache_find(net, &rt->rt6i_src.addr, &rt->rt6i_dst.addr); diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 4b48819a5b8d..4b264ed40a8c 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1448,6 +1448,7 @@ static void mld_sendpack(struct sk_buff *skb) struct net *net = dev_net(skb->dev); int err; struct flowi fl; + struct dst_entry *dst; IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len); @@ -1459,9 +1460,9 @@ static void mld_sendpack(struct sk_buff *skb) IPPROTO_ICMPV6, csum_partial(skb_transport_header(skb), mldlen, 0)); - skb->dst = icmp6_dst_alloc(skb->dev, NULL, &ipv6_hdr(skb)->daddr); + dst = icmp6_dst_alloc(skb->dev, NULL, &ipv6_hdr(skb)->daddr); - if (!skb->dst) { + if (!dst) { err = -ENOMEM; goto err_out; } @@ -1470,7 +1471,8 @@ static void mld_sendpack(struct sk_buff *skb) &ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->dev->ifindex); - err = xfrm_lookup(net, &skb->dst, &fl, NULL, 0); + err = xfrm_lookup(net, &dst, &fl, NULL, 0); + skb_dst_set(skb, dst); if (err) goto err_out; @@ -1775,6 +1777,7 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) IPV6_TLV_ROUTERALERT, 2, 0, 0, IPV6_TLV_PADN, 0 }; struct flowi fl; + struct dst_entry *dst; if (type == ICMPV6_MGM_REDUCTION) snd_addr = &in6addr_linklocal_allrouters; @@ -1828,8 +1831,8 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) idev = in6_dev_get(skb->dev); - skb->dst = icmp6_dst_alloc(skb->dev, NULL, &ipv6_hdr(skb)->daddr); - if (!skb->dst) { + dst = icmp6_dst_alloc(skb->dev, NULL, &ipv6_hdr(skb)->daddr); + if (!dst) { err = -ENOMEM; goto err_out; } @@ -1838,11 +1841,11 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) &ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->dev->ifindex); - err = xfrm_lookup(net, &skb->dst, &fl, NULL, 0); + err = xfrm_lookup(net, &dst, &fl, NULL, 0); if (err) goto err_out; - + skb_dst_set(skb, dst); err = NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, skb->dev, dst_output); out: diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 1d13d9964985..9eb68e92cc18 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -530,7 +530,7 @@ void ndisc_send_skb(struct sk_buff *skb, return; } - skb->dst = dst; + skb_dst_set(skb, dst); idev = in6_dev_get(dst->dev); IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len); @@ -1612,7 +1612,7 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, len, IPPROTO_ICMPV6, csum_partial(icmph, len, 0)); - buff->dst = dst; + skb_dst_set(buff, dst); idev = in6_dev_get(dst->dev); IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len); err = NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, buff, NULL, dst->dev, diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 834cea69fb53..d5ed92b14346 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -12,7 +12,7 @@ int ip6_route_me_harder(struct sk_buff *skb) { - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); struct ipv6hdr *iph = ipv6_hdr(skb); struct dst_entry *dst; struct flowi fl = { @@ -28,9 +28,15 @@ int ip6_route_me_harder(struct sk_buff *skb) #ifdef CONFIG_XFRM if (!(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) && - xfrm_decode_session(skb, &fl, AF_INET6) == 0) - if (xfrm_lookup(net, &skb->dst, &fl, skb->sk, 0)) + xfrm_decode_session(skb, &fl, AF_INET6) == 0) { + struct dst_entry *dst2 = skb_dst(skb); + + if (xfrm_lookup(net, &dst2, &fl, skb->sk, 0)) { + skb_dst_set(skb, NULL); return -1; + } + skb_dst_set(skb, dst2); + } #endif if (dst->error) { @@ -41,9 +47,9 @@ int ip6_route_me_harder(struct sk_buff *skb) } /* Drop old route. */ - dst_release(skb->dst); + skb_dst_drop(skb); - skb->dst = dst; + skb_dst_set(skb, dst); return 0; } EXPORT_SYMBOL(ip6_route_me_harder); diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c index 5a2d0a41694a..5a7f00cd15ce 100644 --- a/net/ipv6/netfilter/ip6t_REJECT.c +++ b/net/ipv6/netfilter/ip6t_REJECT.c @@ -112,7 +112,7 @@ static void send_reset(struct net *net, struct sk_buff *oldskb) return; } - nskb->dst = dst; + skb_dst_set(nskb, dst); skb_reserve(nskb, hh_len + dst->header_len); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index e99307fba0b1..36a090d87a3d 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -625,7 +625,7 @@ static int rawv6_send_hdrinc(struct sock *sk, void *from, int length, skb->priority = sk->sk_priority; skb->mark = sk->sk_mark; - skb->dst = dst_clone(&rt->u.dst); + skb_dst_set(skb, dst_clone(&rt->u.dst)); skb_put(skb, length); skb_reset_network_header(skb); diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index e9ac7a12f595..54a387d31e1a 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -267,7 +267,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, struct sk_buff *prev, *next; struct net_device *dev; int offset, end; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); if (fq->q.last_in & INET_FRAG_COMPLETE) goto err; @@ -277,7 +277,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, ((u8 *)(fhdr + 1) - (u8 *)(ipv6_hdr(skb) + 1))); if ((unsigned int)end > IPV6_MAXPLEN) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, ((u8 *)&fhdr->frag_off - @@ -310,7 +310,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, /* RFC2460 says always send parameter problem in * this case. -DaveM */ - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, offsetof(struct ipv6hdr, payload_len)); @@ -434,7 +434,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, return -1; err: - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_REASMFAILS); kfree_skb(skb); return -1; @@ -576,9 +576,9 @@ static int ipv6_frag_rcv(struct sk_buff *skb) struct frag_hdr *fhdr; struct frag_queue *fq; struct ipv6hdr *hdr = ipv6_hdr(skb); - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMREQDS); + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_REASMREQDS); /* Jumbo payload inhibits frag. header */ if (hdr->payload_len==0) @@ -595,17 +595,17 @@ static int ipv6_frag_rcv(struct sk_buff *skb) /* It is not a fragmented frame */ skb->transport_header += sizeof(struct frag_hdr); IP6_INC_STATS_BH(net, - ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMOKS); + ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_REASMOKS); IP6CB(skb)->nhoff = (u8 *)fhdr - skb_network_header(skb); return 1; } if (atomic_read(&net->ipv6.frags.mem) > net->ipv6.frags.high_thresh) - ip6_evictor(net, ip6_dst_idev(skb->dst)); + ip6_evictor(net, ip6_dst_idev(skb_dst(skb))); if ((fq = fq_find(net, fhdr->identification, &hdr->saddr, &hdr->daddr, - ip6_dst_idev(skb->dst))) != NULL) { + ip6_dst_idev(skb_dst(skb)))) != NULL) { int ret; spin_lock(&fq->q.lock); @@ -617,12 +617,12 @@ static int ipv6_frag_rcv(struct sk_buff *skb) return ret; } - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMFAILS); + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_REASMFAILS); kfree_skb(skb); return -1; fail_hdr: - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, skb_network_header_len(skb)); return -1; } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 032a5ec391c5..658293ea05ba 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -800,7 +800,7 @@ void ip6_route_input(struct sk_buff *skb) if (rt6_need_strict(&iph->daddr) && skb->dev->type != ARPHRD_PIMREG) flags |= RT6_LOOKUP_F_IFACE; - skb->dst = fib6_rule_lookup(net, &fl, flags, ip6_pol_route_input); + skb_dst_set(skb, fib6_rule_lookup(net, &fl, flags, ip6_pol_route_input)); } static struct rt6_info *ip6_pol_route_output(struct net *net, struct fib6_table *table, @@ -911,7 +911,7 @@ static void ip6_link_failure(struct sk_buff *skb) icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0, skb->dev); - rt = (struct rt6_info *) skb->dst; + rt = (struct rt6_info *) skb_dst(skb); if (rt) { if (rt->rt6i_flags&RTF_CACHE) { dst_set_expires(&rt->u.dst, 0); @@ -1868,7 +1868,7 @@ int ipv6_route_ioctl(struct net *net, unsigned int cmd, void __user *arg) static int ip6_pkt_drop(struct sk_buff *skb, int code, int ipstats_mib_noroutes) { int type; - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); switch (ipstats_mib_noroutes) { case IPSTATS_MIB_INNOROUTES: type = ipv6_addr_type(&ipv6_hdr(skb)->daddr); @@ -1895,7 +1895,7 @@ static int ip6_pkt_discard(struct sk_buff *skb) static int ip6_pkt_discard_out(struct sk_buff *skb) { - skb->dev = skb->dst->dev; + skb->dev = skb_dst(skb)->dev; return ip6_pkt_drop(skb, ICMPV6_NOROUTE, IPSTATS_MIB_OUTNOROUTES); } @@ -1908,7 +1908,7 @@ static int ip6_pkt_prohibit(struct sk_buff *skb) static int ip6_pkt_prohibit_out(struct sk_buff *skb) { - skb->dev = skb->dst->dev; + skb->dev = skb_dst(skb)->dev; return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED, IPSTATS_MIB_OUTNOROUTES); } @@ -2366,7 +2366,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr)); rt = (struct rt6_info*) ip6_route_output(net, NULL, &fl); - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); err = rt6_fill_node(net, skb, rt, &fl.fl6_dst, &fl.fl6_src, iif, RTM_NEWROUTE, NETLINK_CB(in_skb).pid, diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index b3a59bd40f01..68e52308e552 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -575,8 +575,7 @@ static int ipip6_rcv(struct sk_buff *skb) tunnel->dev->stats.rx_packets++; tunnel->dev->stats.rx_bytes += skb->len; skb->dev = tunnel->dev; - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); nf_reset(skb); ipip6_ecn_decapsulate(iph, skb); netif_rx(skb); @@ -638,8 +637,8 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (dev->priv_flags & IFF_ISATAP) { struct neighbour *neigh = NULL; - if (skb->dst) - neigh = skb->dst->neighbour; + if (skb_dst(skb)) + neigh = skb_dst(skb)->neighbour; if (neigh == NULL) { if (net_ratelimit()) @@ -663,8 +662,8 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (!dst) { struct neighbour *neigh = NULL; - if (skb->dst) - neigh = skb->dst->neighbour; + if (skb_dst(skb)) + neigh = skb_dst(skb)->neighbour; if (neigh == NULL) { if (net_ratelimit()) @@ -714,7 +713,7 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (tiph->frag_off) mtu = dst_mtu(&rt->u.dst) - sizeof(struct iphdr); else - mtu = skb->dst ? dst_mtu(skb->dst) : dev->mtu; + mtu = skb_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu; if (mtu < 68) { stats->collisions++; @@ -723,8 +722,8 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) } if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; - if (tunnel->parms.iph.daddr && skb->dst) - skb->dst->ops->update_pmtu(skb->dst, mtu); + if (tunnel->parms.iph.daddr && skb_dst(skb)) + skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); if (skb->len > mtu) { icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, dev); @@ -768,8 +767,8 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) skb_reset_network_header(skb); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); IPCB(skb)->flags = 0; - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* * Push down and install the IPIP header. diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index ea37741062a9..53b6a4192b16 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -981,9 +981,10 @@ static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win, struct tcphdr *th = tcp_hdr(skb), *t1; struct sk_buff *buff; struct flowi fl; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); struct sock *ctl_sk = net->ipv6.tcp_sk; unsigned int tot_len = sizeof(struct tcphdr); + struct dst_entry *dst; __be32 *topt; if (ts) @@ -1052,8 +1053,9 @@ static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win, * Underlying function will use this to retrieve the network * namespace */ - if (!ip6_dst_lookup(ctl_sk, &buff->dst, &fl)) { - if (xfrm_lookup(net, &buff->dst, &fl, NULL, 0) >= 0) { + if (!ip6_dst_lookup(ctl_sk, &dst, &fl)) { + if (xfrm_lookup(net, &dst, &fl, NULL, 0) >= 0) { + skb_dst_set(buff, dst); ip6_xmit(ctl_sk, buff, &fl, NULL, 0); TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS); if (rst) diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 8905712cfbb8..fc333d854728 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -177,10 +177,9 @@ static struct sock *__udp6_lib_lookup_skb(struct sk_buff *skb, if (unlikely(sk = skb_steal_sock(skb))) return sk; - else - return __udp6_lib_lookup(dev_net(skb->dst->dev), &iph->saddr, sport, - &iph->daddr, dport, inet6_iif(skb), - udptable); + return __udp6_lib_lookup(dev_net(skb_dst(skb)->dev), &iph->saddr, sport, + &iph->daddr, dport, inet6_iif(skb), + udptable); } /* diff --git a/net/ipv6/xfrm6_mode_tunnel.c b/net/ipv6/xfrm6_mode_tunnel.c index e20529b4c825..3927832227b9 100644 --- a/net/ipv6/xfrm6_mode_tunnel.c +++ b/net/ipv6/xfrm6_mode_tunnel.c @@ -31,7 +31,7 @@ static inline void ipip6_ecn_decapsulate(struct sk_buff *skb) */ static int xfrm6_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct ipv6hdr *top_iph; int dsfield; @@ -45,7 +45,7 @@ static int xfrm6_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) memcpy(top_iph->flow_lbl, XFRM_MODE_SKB_CB(skb)->flow_lbl, sizeof(top_iph->flow_lbl)); - top_iph->nexthdr = xfrm_af2proto(skb->dst->ops->family); + top_iph->nexthdr = xfrm_af2proto(skb_dst(skb)->ops->family); dsfield = XFRM_MODE_SKB_CB(skb)->tos; dsfield = INET_ECN_encapsulate(dsfield, dsfield); diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c index 5ee5a031bc93..c4f4eef032a3 100644 --- a/net/ipv6/xfrm6_output.c +++ b/net/ipv6/xfrm6_output.c @@ -30,7 +30,7 @@ EXPORT_SYMBOL(xfrm6_find_1stfragopt); static int xfrm6_tunnel_check_size(struct sk_buff *skb) { int mtu, ret = 0; - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); mtu = dst_mtu(dst); if (mtu < IPV6_MIN_MTU) @@ -90,6 +90,6 @@ static int xfrm6_output_finish(struct sk_buff *skb) int xfrm6_output(struct sk_buff *skb) { - return NF_HOOK(PF_INET6, NF_INET_POST_ROUTING, skb, NULL, skb->dst->dev, + return NF_HOOK(PF_INET6, NF_INET_POST_ROUTING, skb, NULL, skb_dst(skb)->dev, xfrm6_output_finish); } -- cgit v1.2.3 From 4d9092bb41c8fdca45513461587d8b4ad3918f74 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 9 Jun 2009 00:20:05 -0700 Subject: ipv6: Use frag list abstraction interfaces. Signed-off-by: David S. Miller --- net/ipv6/ip6_output.c | 6 +++--- net/ipv6/reassembly.c | 4 ++-- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'net/ipv6') diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index c8dc8e5a822f..db6c7224a862 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -658,7 +658,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) } mtu -= hlen + sizeof(struct frag_hdr); - if (skb_shinfo(skb)->frag_list) { + if (skb_has_frags(skb)) { int first_len = skb_pagelen(skb); int truesizes = 0; @@ -667,7 +667,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) skb_cloned(skb)) goto slow_path; - for (frag = skb_shinfo(skb)->frag_list; frag; frag = frag->next) { + skb_walk_frags(skb, frag) { /* Correct geometry. */ if (frag->len > mtu || ((frag->len & 7) && frag->next) || @@ -690,7 +690,7 @@ 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 = NULL; + skb_frag_list_init(skb); /* BUILD HEADER */ *prevhdr = NEXTHDR_FRAGMENT; diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 54a387d31e1a..2642a41a8535 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -494,7 +494,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev, /* If the first fragment is fragmented itself, we split * it to two chunks: the first with data and paged part * and the second, holding only fragments. */ - if (skb_shinfo(head)->frag_list) { + if (skb_has_frags(head)) { struct sk_buff *clone; int i, plen = 0; @@ -503,7 +503,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev, clone->next = head->next; head->next = clone; skb_shinfo(clone)->frag_list = skb_shinfo(head)->frag_list; - skb_shinfo(head)->frag_list = NULL; + skb_frag_list_init(head); for (i=0; inr_frags; i++) plen += skb_shinfo(head)->frags[i].size; clone->len = clone->data_len = head->data_len - plen; -- cgit v1.2.3 From 343a99724e4431d8618bea2eb7552e12c965425a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 9 Jun 2009 00:23:58 -0700 Subject: netfilter: Use frag list abstraction interfaces. Signed-off-by: David S. Miller --- net/ipv6/netfilter/nf_conntrack_reasm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/ipv6') diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 058a5e4a60c3..f3aba255ad9f 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -409,7 +409,7 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev) /* If the first fragment is fragmented itself, we split * it to two chunks: the first with data and paged part * and the second, holding only fragments. */ - if (skb_shinfo(head)->frag_list) { + if (skb_has_frags(head)) { struct sk_buff *clone; int i, plen = 0; @@ -420,7 +420,7 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev) clone->next = head->next; head->next = clone; skb_shinfo(clone)->frag_list = skb_shinfo(head)->frag_list; - skb_shinfo(head)->frag_list = NULL; + skb_frag_list_init(head); for (i=0; inr_frags; i++) plen += skb_shinfo(head)->frags[i].size; clone->len = clone->data_len = head->data_len - plen; -- cgit v1.2.3 From 2b85a34e911bf483c27cfdd124aeb1605145dc80 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 11 Jun 2009 02:55:43 -0700 Subject: net: No more expensive sock_hold()/sock_put() on each tx One of the problem with sock memory accounting is it uses a pair of sock_hold()/sock_put() for each transmitted packet. This slows down bidirectional flows because the receive path also needs to take a refcount on socket and might use a different cpu than transmit path or transmit completion path. So these two atomic operations also trigger cache line bounces. We can see this in tx or tx/rx workloads (media gateways for example), where sock_wfree() can be in top five functions in profiles. We use this sock_hold()/sock_put() so that sock freeing is delayed until all tx packets are completed. As we also update sk_wmem_alloc, we could offset sk_wmem_alloc by one unit at init time, until sk_free() is called. Once sk_free() is called, we atomic_dec_and_test(sk_wmem_alloc) to decrement initial offset and atomicaly check if any packets are in flight. skb_set_owner_w() doesnt call sock_hold() anymore sock_wfree() doesnt call sock_put() anymore, but check if sk_wmem_alloc reached 0 to perform the final freeing. Drawback is that a skb->truesize error could lead to unfreeable sockets, or even worse, prematurely calling __sk_free() on a live socket. Nice speedups on SMP. tbench for example, going from 2691 MB/s to 2711 MB/s on my 8 cpu dev machine, even if tbench was not really hitting sk_refcnt contention point. 5 % speedup on a UDP transmit workload (depends on number of flows), lowering TX completion cpu usage. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/ip6_output.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net/ipv6') diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index db6c7224a862..7c76e3d18215 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -680,7 +680,6 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) BUG_ON(frag->sk); if (skb->sk) { - sock_hold(skb->sk); frag->sk = skb->sk; frag->destructor = sock_wfree; truesizes += frag->truesize; -- cgit v1.2.3