From 93f154b594fe47e4a7e5358b309add449a046cd3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 18 May 2009 22:19:19 -0700 Subject: net: release dst entry in dev_hard_start_xmit() One point of contention in high network loads is the dst_release() performed when a transmited skb is freed. This is because NIC tx completion calls dev_kree_skb() long after original call to dev_queue_xmit(skb). CPU cache is cold and the atomic op in dst_release() stalls. On SMP, this is quite visible if one CPU is 100% handling softirqs for a network device, since dst_clone() is done by other cpus, involving cache line ping pongs. It seems right place to release dst is in dev_hard_start_xmit(), for most devices but ones that are virtual, and some exceptions. David Miller suggested to define a new device flag, set in alloc_netdev_mq() (so that most devices set it at init time), and carefuly unset in devices which dont want a NULL skb->dst in their ndo_start_xmit(). List of devices that must clear this flag is : - loopback device, because it calls netif_rx() and quoting Patrick : "ip_route_input() doesn't accept loopback addresses, so loopback packets already need to have a dst_entry attached." - appletalk/ipddp.c : needs skb->dst in its xmit function - And all devices that call again dev_queue_xmit() from their xmit function (as some classifiers need skb->dst) : bonding, vlan, macvlan, eql, ifb, hdlc_fr Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/appletalk/ipddp.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers/net/appletalk/ipddp.c') diff --git a/drivers/net/appletalk/ipddp.c b/drivers/net/appletalk/ipddp.c index da64ba88d7f..f939e92fcf8 100644 --- a/drivers/net/appletalk/ipddp.c +++ b/drivers/net/appletalk/ipddp.c @@ -71,6 +71,7 @@ static struct net_device * __init ipddp_init(void) if (!dev) return ERR_PTR(-ENOMEM); + dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; strcpy(dev->name, "ipddp%d"); if (version_printed++ == 0) -- cgit v1.2.3-70-g09d2 From 5615968a70845157adaffc11062c997d045339ee Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 27 May 2009 16:56:47 -0700 Subject: appletalk: Add proper locking around IPDDP routing table. Signed-off-by: David S. Miller --- drivers/net/appletalk/ipddp.c | 40 ++++++++++++++++++++++++++++++++-------- 1 file changed, 32 insertions(+), 8 deletions(-) (limited to 'drivers/net/appletalk/ipddp.c') diff --git a/drivers/net/appletalk/ipddp.c b/drivers/net/appletalk/ipddp.c index f939e92fcf8..9832b757f10 100644 --- a/drivers/net/appletalk/ipddp.c +++ b/drivers/net/appletalk/ipddp.c @@ -39,6 +39,7 @@ static const char version[] = KERN_INFO "ipddp.c:v0.01 8/28/97 Bradford W. Johnson \n"; static struct ipddp_route *ipddp_route_list; +static DEFINE_SPINLOCK(ipddp_route_lock); #ifdef CONFIG_IPDDP_ENCAP static int ipddp_mode = IPDDP_ENCAP; @@ -50,7 +51,7 @@ static int ipddp_mode = IPDDP_DECAP; static int ipddp_xmit(struct sk_buff *skb, struct net_device *dev); static int ipddp_create(struct ipddp_route *new_rt); static int ipddp_delete(struct ipddp_route *rt); -static struct ipddp_route* ipddp_find_route(struct ipddp_route *rt); +static struct ipddp_route* __ipddp_find_route(struct ipddp_route *rt); static int ipddp_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd); static const struct net_device_ops ipddp_netdev_ops = { @@ -119,6 +120,8 @@ static int ipddp_xmit(struct sk_buff *skb, struct net_device *dev) struct ipddp_route *rt; struct atalk_addr *our_addr; + spin_lock(&ipddp_route_lock); + /* * Find appropriate route to use, based only on IP number. */ @@ -127,8 +130,10 @@ static int ipddp_xmit(struct sk_buff *skb, struct net_device *dev) if(rt->ip == paddr) break; } - if(rt == NULL) + if(rt == NULL) { + spin_unlock(&ipddp_route_lock); return 0; + } our_addr = atalk_find_dev_addr(rt->dev); @@ -174,6 +179,8 @@ static int ipddp_xmit(struct sk_buff *skb, struct net_device *dev) if(aarp_send_ddp(rt->dev, skb, &rt->at, NULL) < 0) dev_kfree_skb(skb); + spin_unlock(&ipddp_route_lock); + return 0; } @@ -196,7 +203,9 @@ static int ipddp_create(struct ipddp_route *new_rt) return -ENETUNREACH; } - if (ipddp_find_route(rt)) { + spin_lock_bh(&ipddp_route_lock); + if (__ipddp_find_route(rt)) { + spin_unlock_bh(&ipddp_route_lock); kfree(rt); return -EEXIST; } @@ -204,6 +213,8 @@ static int ipddp_create(struct ipddp_route *new_rt) rt->next = ipddp_route_list; ipddp_route_list = rt; + spin_unlock_bh(&ipddp_route_lock); + return 0; } @@ -216,6 +227,7 @@ static int ipddp_delete(struct ipddp_route *rt) struct ipddp_route **r = &ipddp_route_list; struct ipddp_route *tmp; + spin_lock_bh(&ipddp_route_lock); while((tmp = *r) != NULL) { if(tmp->ip == rt->ip @@ -223,19 +235,21 @@ static int ipddp_delete(struct ipddp_route *rt) && tmp->at.s_node == rt->at.s_node) { *r = tmp->next; + spin_unlock_bh(&ipddp_route_lock); kfree(tmp); return 0; } r = &tmp->next; } + spin_unlock_bh(&ipddp_route_lock); return (-ENOENT); } /* * Find a routing entry, we only return a FULL match */ -static struct ipddp_route* ipddp_find_route(struct ipddp_route *rt) +static struct ipddp_route* __ipddp_find_route(struct ipddp_route *rt) { struct ipddp_route *f; @@ -253,7 +267,7 @@ static struct ipddp_route* ipddp_find_route(struct ipddp_route *rt) static int ipddp_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { struct ipddp_route __user *rt = ifr->ifr_data; - struct ipddp_route rcp; + struct ipddp_route rcp, rcp2, *rp; if(!capable(CAP_NET_ADMIN)) return -EPERM; @@ -267,9 +281,19 @@ static int ipddp_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) return (ipddp_create(&rcp)); case SIOCFINDIPDDPRT: - if(copy_to_user(rt, ipddp_find_route(&rcp), sizeof(struct ipddp_route))) - return -EFAULT; - return 0; + spin_lock_bh(&ipddp_route_lock); + rp = __ipddp_find_route(&rcp); + if (rp) + memcpy(&rcp2, rp, sizeof(rcp2)); + spin_unlock_bh(&ipddp_route_lock); + + if (rp) { + if (copy_to_user(rt, &rcp2, + sizeof(struct ipddp_route))) + return -EFAULT; + return 0; + } else + return -ENOENT; case SIOCDELIPDDPRT: return (ipddp_delete(&rcp)); -- cgit v1.2.3-70-g09d2 From 511c3f92ad5b6d9f8f6464be1b4f85f0422be91a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 2 Jun 2009 05:14:27 +0000 Subject: net: skb->rtable accessor Define skb_rtable(const struct sk_buff *skb) accessor to get rtable from skb Delete skb->rtable field Setting rtable is not allowed, just set dst instead as rtable is an alias. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/appletalk/ipddp.c | 2 +- include/linux/skbuff.h | 6 +++++- include/net/route.h | 2 +- net/bridge/br_netfilter.c | 25 +++++++++++++--------- net/dccp/ipv4.c | 6 +++--- net/ipv4/arp.c | 4 ++-- net/ipv4/icmp.c | 10 ++++----- net/ipv4/igmp.c | 2 +- net/ipv4/ip_forward.c | 2 +- net/ipv4/ip_gre.c | 4 ++-- net/ipv4/ip_input.c | 2 +- net/ipv4/ip_options.c | 16 +++++++------- net/ipv4/ip_output.c | 10 ++++----- net/ipv4/ip_sockglue.c | 2 +- net/ipv4/ipip.c | 2 +- net/ipv4/ipmr.c | 6 +++--- net/ipv4/netfilter/ipt_MASQUERADE.c | 2 +- net/ipv4/netfilter/nf_nat_helper.c | 4 ++-- net/ipv4/route.c | 37 ++++++++++++++++++++------------- net/ipv4/tcp_ipv4.c | 4 ++-- net/netfilter/nf_conntrack_netbios_ns.c | 2 +- net/sched/em_meta.c | 4 ++-- net/sctp/protocol.c | 8 +++---- 23 files changed, 89 insertions(+), 73 deletions(-) (limited to 'drivers/net/appletalk/ipddp.c') diff --git a/drivers/net/appletalk/ipddp.c b/drivers/net/appletalk/ipddp.c index 9832b757f10..78cea5e80b1 100644 --- a/drivers/net/appletalk/ipddp.c +++ b/drivers/net/appletalk/ipddp.c @@ -115,7 +115,7 @@ static struct net_device * __init ipddp_init(void) */ static int ipddp_xmit(struct sk_buff *skb, struct net_device *dev) { - __be32 paddr = ((struct rtable*)skb->dst)->rt_gateway; + __be32 paddr = skb_rtable(skb)->rt_gateway; struct ddpehdr *ddp; struct ipddp_route *rt; struct atalk_addr *our_addr; diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index d4d7c666ca6..a3ae3c52583 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -324,7 +324,6 @@ struct sk_buff { union { struct dst_entry *dst; - struct rtable *rtable; unsigned long _skb_dst; }; #ifdef CONFIG_XFRM @@ -427,6 +426,11 @@ extern void skb_dma_unmap(struct device *dev, struct sk_buff *skb, enum dma_data_direction dir); #endif +static inline struct rtable *skb_rtable(const struct sk_buff *skb) +{ + return (struct rtable *)skb->_skb_dst; +} + extern void kfree_skb(struct sk_buff *skb); extern void consume_skb(struct sk_buff *skb); extern void __kfree_skb(struct sk_buff *skb); diff --git a/include/net/route.h b/include/net/route.h index 4e8cae0e584..40f6346ef49 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -210,7 +210,7 @@ static inline struct inet_peer *rt_get_peer(struct rtable *rt) static inline int inet_iif(const struct sk_buff *skb) { - return skb->rtable->rt_iif; + return skb_rtable(skb)->rt_iif; } #endif /* _ROUTE_H */ diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index e4a418fcb35..e0ceb66a9ec 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -228,6 +228,7 @@ int nf_bridge_copy_header(struct sk_buff *skb) static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb) { struct nf_bridge_info *nf_bridge = skb->nf_bridge; + struct rtable *rt; if (nf_bridge->mask & BRNF_PKT_TYPE) { skb->pkt_type = PACKET_OTHERHOST; @@ -235,12 +236,13 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb) } nf_bridge->mask ^= BRNF_NF_BRIDGE_PREROUTING; - skb->rtable = bridge_parent_rtable(nf_bridge->physindev); - if (!skb->rtable) { + rt = bridge_parent_rtable(nf_bridge->physindev); + if (!rt) { kfree_skb(skb); return 0; } - dst_hold(&skb->rtable->u.dst); + dst_hold(&rt->u.dst); + skb->dst = &rt->u.dst; skb->dev = nf_bridge->physindev; nf_bridge_push_encap_header(skb); @@ -338,6 +340,7 @@ static int br_nf_pre_routing_finish(struct sk_buff *skb) struct net_device *dev = skb->dev; struct iphdr *iph = ip_hdr(skb); struct nf_bridge_info *nf_bridge = skb->nf_bridge; + struct rtable *rt; int err; if (nf_bridge->mask & BRNF_PKT_TYPE) { @@ -347,7 +350,6 @@ static int br_nf_pre_routing_finish(struct sk_buff *skb) nf_bridge->mask ^= BRNF_NF_BRIDGE_PREROUTING; if (dnat_took_place(skb)) { if ((err = ip_route_input(skb, iph->daddr, iph->saddr, iph->tos, dev))) { - struct rtable *rt; struct flowi fl = { .nl_u = { .ip4_u = { @@ -404,12 +406,13 @@ bridged_dnat: skb->pkt_type = PACKET_HOST; } } else { - skb->rtable = bridge_parent_rtable(nf_bridge->physindev); - if (!skb->rtable) { + rt = bridge_parent_rtable(nf_bridge->physindev); + if (!rt) { kfree_skb(skb); return 0; } - dst_hold(&skb->rtable->u.dst); + dst_hold(&rt->u.dst); + skb->dst = &rt->u.dst; } skb->dev = nf_bridge->physindev; @@ -628,9 +631,11 @@ static unsigned int br_nf_local_in(unsigned int hook, struct sk_buff *skb, const struct net_device *out, int (*okfn)(struct sk_buff *)) { - if (skb->rtable && skb->rtable == bridge_parent_rtable(in)) { - dst_release(&skb->rtable->u.dst); - skb->rtable = NULL; + struct rtable *rt = skb_rtable(skb); + + if (rt && rt == bridge_parent_rtable(in)) { + dst_release(&rt->u.dst); + skb->dst = NULL; } return NF_ACCEPT; diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index d1dd95289b8..2cf48ba0dbb 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -452,7 +452,7 @@ static struct dst_entry* dccp_v4_route_skb(struct net *net, struct sock *sk, struct sk_buff *skb) { struct rtable *rt; - struct flowi fl = { .oif = skb->rtable->rt_iif, + struct flowi fl = { .oif = skb_rtable(skb)->rt_iif, .nl_u = { .ip4_u = { .daddr = ip_hdr(skb)->saddr, .saddr = ip_hdr(skb)->daddr, @@ -514,7 +514,7 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET) return; - if (rxskb->rtable->rt_type != RTN_LOCAL) + if (skb_rtable(rxskb)->rt_type != RTN_LOCAL) return; dst = dccp_v4_route_skb(net, ctl_sk, rxskb); @@ -567,7 +567,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); /* Never answer to DCCP_PKT_REQUESTs send to broadcast or multicast */ - if (skb->rtable->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) + if (skb_rtable(skb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) return 0; /* discard, don't send a reset here */ if (dccp_bad_service_code(sk, service)) { diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index f11931c1838..816494f271a 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -474,7 +474,7 @@ int arp_find(unsigned char *haddr, struct sk_buff *skb) return 1; } - paddr = skb->rtable->rt_gateway; + paddr = skb_rtable(skb)->rt_gateway; if (arp_set_predefined(inet_addr_type(dev_net(dev), paddr), haddr, paddr, dev)) return 0; @@ -817,7 +817,7 @@ static int arp_process(struct sk_buff *skb) if (arp->ar_op == htons(ARPOP_REQUEST) && ip_route_input(skb, tip, sip, 0, dev) == 0) { - rt = skb->rtable; + rt = skb_rtable(skb); addr_type = rt->rt_type; if (addr_type == RTN_LOCAL) { diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 3f50807237e..94f75efae93 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -356,7 +356,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param, static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) { struct ipcm_cookie ipc; - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct net *net = dev_net(rt->u.dst.dev); struct sock *sk; struct inet_sock *inet; @@ -416,7 +416,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) struct iphdr *iph; int room; struct icmp_bxm icmp_param; - struct rtable *rt = skb_in->rtable; + struct rtable *rt = skb_rtable(skb_in); struct ipcm_cookie ipc; __be32 saddr; u8 tos; @@ -596,7 +596,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) RT_TOS(tos), rt2->u.dst.dev); dst_release(&rt2->u.dst); - rt2 = skb_in->rtable; + rt2 = skb_rtable(skb_in); skb_in->dst = odst; } @@ -926,7 +926,7 @@ static void icmp_address(struct sk_buff *skb) static void icmp_address_reply(struct sk_buff *skb) { - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct net_device *dev = skb->dev; struct in_device *in_dev; struct in_ifaddr *ifa; @@ -970,7 +970,7 @@ static void icmp_discard(struct sk_buff *skb) int icmp_rcv(struct sk_buff *skb) { struct icmphdr *icmph; - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct net *net = dev_net(rt->u.dst.dev); if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) { diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index e6058a50379..afabd2758b6 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -948,7 +948,7 @@ int igmp_rcv(struct sk_buff *skb) case IGMPV2_HOST_MEMBERSHIP_REPORT: case IGMPV3_HOST_MEMBERSHIP_REPORT: /* Is it our report looped back? */ - if (skb->rtable->fl.iif == 0) + if (skb_rtable(skb)->fl.iif == 0) break; /* don't rely on MC router hearing unicast reports */ if (skb->pkt_type == PACKET_MULTICAST || diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index df3fe50bbf0..0761cd9bbd1 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -81,7 +81,7 @@ int ip_forward(struct sk_buff *skb) if (!xfrm4_route_forward(skb)) goto drop; - rt = skb->rtable; + rt = skb_rtable(skb); if (opt->is_strictroute && rt->rt_dst != rt->rt_gateway) goto sr_failed; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 77436e2732e..85ddad45a91 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -602,7 +602,7 @@ static int ipgre_rcv(struct sk_buff *skb) #ifdef CONFIG_NET_IPGRE_BROADCAST if (ipv4_is_multicast(iph->daddr)) { /* Looped back packet, drop it! */ - if (skb->rtable->fl.iif == 0) + if (skb_rtable(skb)->fl.iif == 0) goto drop; stats->multicast++; skb->pkt_type = PACKET_BROADCAST; @@ -704,7 +704,7 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) } if (skb->protocol == htons(ETH_P_IP)) { - rt = skb->rtable; + rt = skb_rtable(skb); if ((dst = rt->rt_gateway) == 0) goto tx_error_icmp; } diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 40f6206b2aa..cea784b0aa4 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -357,7 +357,7 @@ static int ip_rcv_finish(struct sk_buff *skb) if (iph->ihl > 5 && ip_rcv_options(skb)) goto drop; - rt = skb->rtable; + rt = skb_rtable(skb); if (rt->rt_type == RTN_MULTICAST) { IP_UPD_PO_STATS_BH(dev_net(rt->u.dst.dev), IPSTATS_MIB_INMCAST, skb->len); diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index 2c88da6e786..7e1074ffdbd 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -102,7 +102,7 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb) sptr = skb_network_header(skb); dptr = dopt->__data; - daddr = skb->rtable->rt_spec_dst; + daddr = skb_rtable(skb)->rt_spec_dst; if (sopt->rr) { optlen = sptr[sopt->rr+1]; @@ -257,7 +257,7 @@ int ip_options_compile(struct net *net, struct rtable *rt = NULL; if (skb != NULL) { - rt = skb->rtable; + rt = skb_rtable(skb); optptr = (unsigned char *)&(ip_hdr(skb)[1]); } else optptr = opt->__data; @@ -550,7 +550,7 @@ void ip_forward_options(struct sk_buff *skb) { struct ip_options * opt = &(IPCB(skb)->opt); unsigned char * optptr; - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); unsigned char *raw = skb_network_header(skb); if (opt->rr_needaddr) { @@ -598,7 +598,7 @@ int ip_options_rcv_srr(struct sk_buff *skb) __be32 nexthop; struct iphdr *iph = ip_hdr(skb); unsigned char *optptr = skb_network_header(skb) + opt->srr; - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct rtable *rt2; int err; @@ -623,13 +623,13 @@ int ip_options_rcv_srr(struct sk_buff *skb) } memcpy(&nexthop, &optptr[srrptr-1], 4); - rt = skb->rtable; - skb->rtable = NULL; + rt = skb_rtable(skb); + skb->dst = NULL; err = ip_route_input(skb, nexthop, iph->saddr, iph->tos, skb->dev); - rt2 = skb->rtable; + rt2 = skb_rtable(skb); if (err || (rt2->rt_type != RTN_UNICAST && rt2->rt_type != RTN_LOCAL)) { ip_rt_put(rt2); - skb->rtable = rt; + skb->dst = &rt->u.dst; return -EINVAL; } ip_rt_put(rt); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index ea19c37ccc0..8d845ebfcca 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -140,7 +140,7 @@ int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk, __be32 saddr, __be32 daddr, struct ip_options *opt) { struct inet_sock *inet = inet_sk(sk); - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct iphdr *iph; /* Build the IP header. */ @@ -238,7 +238,7 @@ static int ip_finish_output(struct sk_buff *skb) int ip_mc_output(struct sk_buff *skb) { struct sock *sk = skb->sk; - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct net_device *dev = rt->u.dst.dev; /* @@ -319,7 +319,7 @@ int ip_queue_xmit(struct sk_buff *skb, int ipfragok) /* Skip all of this if the packet is already routed, * f.e. by something like SCTP. */ - rt = skb->rtable; + rt = skb_rtable(skb); if (rt != NULL) goto packet_routed; @@ -440,7 +440,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) unsigned int mtu, hlen, left, len, ll_rs, pad; int offset; __be16 not_last_frag; - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); int err = 0; dev = rt->u.dst.dev; @@ -1362,7 +1362,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar } replyopts; struct ipcm_cookie ipc; __be32 daddr; - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); if (ip_options_echo(&replyopts.opt, skb)) return; diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index cb49936856e..fc7993e9061 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -57,7 +57,7 @@ static void ip_cmsg_recv_pktinfo(struct msghdr *msg, struct sk_buff *skb) { struct in_pktinfo info; - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); info.ipi_addr.s_addr = ip_hdr(skb)->daddr; if (rt) { diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index bb2f1b17fbf..0c6e7bf18a4 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -416,7 +416,7 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (!dst) { /* NBMA tunnel */ - if ((rt = skb->rtable) == NULL) { + if ((rt = skb_rtable(skb)) == NULL) { stats->tx_fifo_errors++; goto tx_error; } diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 13e9dd3012b..69dd058283e 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1354,7 +1354,7 @@ static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local if (net->ipv4.vif_table[vif].dev != skb->dev) { int true_vifi; - if (skb->rtable->fl.iif == 0) { + if (skb_rtable(skb)->fl.iif == 0) { /* It is our own packet, looped back. Very complicated situation... @@ -1430,7 +1430,7 @@ int ip_mr_input(struct sk_buff *skb) { struct mfc_cache *cache; struct net *net = dev_net(skb->dev); - int local = skb->rtable->rt_flags&RTCF_LOCAL; + int local = skb_rtable(skb)->rt_flags & RTCF_LOCAL; /* Packet is looped back after forward, it should not be forwarded second time, but still can be delivered locally. @@ -1646,7 +1646,7 @@ int ipmr_get_route(struct net *net, { int err; struct mfc_cache *cache; - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); read_lock(&mrt_lock); cache = ipmr_cache_find(net, rt->rt_src, rt->rt_dst); diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c index f389f60cb10..c0992c75bda 100644 --- a/net/ipv4/netfilter/ipt_MASQUERADE.c +++ b/net/ipv4/netfilter/ipt_MASQUERADE.c @@ -72,7 +72,7 @@ masquerade_tg(struct sk_buff *skb, const struct xt_target_param *par) return NF_ACCEPT; mr = par->targinfo; - rt = skb->rtable; + rt = skb_rtable(skb); newsrc = inet_select_addr(par->out, rt->rt_gateway, RT_SCOPE_UNIVERSE); if (!newsrc) { printk("MASQUERADE: %s ate my IP address\n", par->out->name); diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c index cf7a42bf982..155c008626c 100644 --- a/net/ipv4/netfilter/nf_nat_helper.c +++ b/net/ipv4/netfilter/nf_nat_helper.c @@ -140,7 +140,7 @@ nf_nat_mangle_tcp_packet(struct sk_buff *skb, const char *rep_buffer, unsigned int rep_len) { - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct iphdr *iph; struct tcphdr *tcph; int oldlen, datalen; @@ -218,7 +218,7 @@ nf_nat_mangle_udp_packet(struct sk_buff *skb, const char *rep_buffer, unsigned int rep_len) { - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct iphdr *iph; struct udphdr *udph; int datalen, oldlen; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 28205e5bfa9..f20060ac2f0 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1064,7 +1064,8 @@ work_done: out: return 0; } -static int rt_intern_hash(unsigned hash, struct rtable *rt, struct rtable **rp) +static int rt_intern_hash(unsigned hash, struct rtable *rt, + struct rtable **rp, struct sk_buff *skb) { struct rtable *rth, **rthp; unsigned long now; @@ -1114,7 +1115,10 @@ restart: spin_unlock_bh(rt_hash_lock_addr(hash)); rt_drop(rt); - *rp = rth; + if (rp) + *rp = rth; + else + skb->dst = &rth->u.dst; return 0; } @@ -1210,7 +1214,10 @@ restart: rcu_assign_pointer(rt_hash_table[hash].chain, rt); spin_unlock_bh(rt_hash_lock_addr(hash)); - *rp = rt; + if (rp) + *rp = rt; + else + skb->dst = &rt->u.dst; return 0; } @@ -1407,7 +1414,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw, &netevent); rt_del(hash, rth); - if (!rt_intern_hash(hash, rt, &rt)) + if (!rt_intern_hash(hash, rt, &rt, NULL)) ip_rt_put(rt); goto do_next; } @@ -1473,7 +1480,7 @@ static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst) void ip_rt_send_redirect(struct sk_buff *skb) { - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct in_device *in_dev = in_dev_get(rt->u.dst.dev); if (!in_dev) @@ -1521,7 +1528,7 @@ out: static int ip_error(struct sk_buff *skb) { - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); unsigned long now; int code; @@ -1698,7 +1705,7 @@ static void ipv4_link_failure(struct sk_buff *skb) icmp_send(skb, ICMP_DEST_UNREACH, ICMP_HOST_UNREACH, 0); - rt = skb->rtable; + rt = skb_rtable(skb); if (rt) dst_set_expires(&rt->u.dst, 0); } @@ -1858,7 +1865,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr, in_dev_put(in_dev); hash = rt_hash(daddr, saddr, dev->ifindex, rt_genid(dev_net(dev))); - return rt_intern_hash(hash, rth, &skb->rtable); + return rt_intern_hash(hash, rth, NULL, skb); e_nobufs: in_dev_put(in_dev); @@ -2019,7 +2026,7 @@ static int ip_mkroute_input(struct sk_buff *skb, /* put it into the cache */ hash = rt_hash(daddr, saddr, fl->iif, rt_genid(dev_net(rth->u.dst.dev))); - return rt_intern_hash(hash, rth, &skb->rtable); + return rt_intern_hash(hash, rth, NULL, skb); } /* @@ -2175,7 +2182,7 @@ local_input: } rth->rt_type = res.type; hash = rt_hash(daddr, saddr, fl.iif, rt_genid(net)); - err = rt_intern_hash(hash, rth, &skb->rtable); + err = rt_intern_hash(hash, rth, NULL, skb); goto done; no_route: @@ -2244,7 +2251,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr, dst_use(&rth->u.dst, jiffies); RT_CACHE_STAT_INC(in_hit); rcu_read_unlock(); - skb->rtable = rth; + skb->dst = &rth->u.dst; return 0; } RT_CACHE_STAT_INC(in_hlist_search); @@ -2420,7 +2427,7 @@ static int ip_mkroute_output(struct rtable **rp, if (err == 0) { hash = rt_hash(oldflp->fl4_dst, oldflp->fl4_src, oldflp->oif, rt_genid(dev_net(dev_out))); - err = rt_intern_hash(hash, rth, rp); + err = rt_intern_hash(hash, rth, rp, NULL); } return err; @@ -2763,7 +2770,7 @@ static int rt_fill_info(struct net *net, struct sk_buff *skb, u32 pid, u32 seq, int event, int nowait, unsigned int flags) { - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct rtmsg *r; struct nlmsghdr *nlh; long expires; @@ -2907,7 +2914,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void err = ip_route_input(skb, dst, src, rtm->rtm_tos, dev); local_bh_enable(); - rt = skb->rtable; + rt = skb_rtable(skb); if (err == 0 && rt->u.dst.error) err = -rt->u.dst.error; } else { @@ -2927,7 +2934,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void if (err) goto errout_free; - skb->rtable = rt; + skb->dst = &rt->u.dst; if (rtm->rtm_flags & RTM_F_NOTIFY) rt->rt_flags |= RTCF_NOTIFY; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index fc79e341628..319c8852644 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -546,7 +546,7 @@ static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb) if (th->rst) return; - if (skb->rtable->rt_type != RTN_LOCAL) + if (skb_rtable(skb)->rt_type != RTN_LOCAL) return; /* Swap the send and the receive. */ @@ -1185,7 +1185,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) #endif /* Never answer to SYNs send to broadcast or multicast */ - if (skb->rtable->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) + if (skb_rtable(skb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) goto drop; /* TW buckets are converted to open requests without diff --git a/net/netfilter/nf_conntrack_netbios_ns.c b/net/netfilter/nf_conntrack_netbios_ns.c index 8a3875e36ec..497b2224536 100644 --- a/net/netfilter/nf_conntrack_netbios_ns.c +++ b/net/netfilter/nf_conntrack_netbios_ns.c @@ -48,7 +48,7 @@ static int help(struct sk_buff *skb, unsigned int protoff, { struct nf_conntrack_expect *exp; struct iphdr *iph = ip_hdr(skb); - struct rtable *rt = skb->rtable; + struct rtable *rt = skb_rtable(skb); struct in_device *in_dev; __be32 mask = 0; diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index fad596bf32d..b6b588bed4e 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -258,10 +258,10 @@ META_COLLECTOR(int_rtclassid) META_COLLECTOR(int_rtiif) { - if (unlikely(skb->rtable == NULL)) + if (unlikely(skb_rtable(skb) == NULL)) *err = -1; else - dst->value = skb->rtable->fl.iif; + dst->value = skb_rtable(skb)->fl.iif; } /************************************************************************** diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 8eb3e61cb70..cb2c50dbd42 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -393,7 +393,7 @@ static int sctp_v4_addr_valid(union sctp_addr *addr, return 0; /* Is this a broadcast address? */ - if (skb && skb->rtable->rt_flags & RTCF_BROADCAST) + if (skb && skb_rtable(skb)->rt_flags & RTCF_BROADCAST) return 0; return 1; @@ -572,7 +572,7 @@ static void sctp_v4_get_saddr(struct sctp_sock *sk, /* What interface did this skb arrive on? */ static int sctp_v4_skb_iif(const struct sk_buff *skb) { - return skb->rtable->rt_iif; + return skb_rtable(skb)->rt_iif; } /* Was this packet marked by Explicit Congestion Notification? */ @@ -848,8 +848,8 @@ static inline int sctp_v4_xmit(struct sk_buff *skb, SCTP_DEBUG_PRINTK("%s: skb:%p, len:%d, src:%pI4, dst:%pI4\n", __func__, skb, skb->len, - &skb->rtable->rt_src, - &skb->rtable->rt_dst); + &skb_rtable(skb)->rt_src, + &skb_rtable(skb)->rt_dst); inet->pmtudisc = transport->param_flags & SPP_PMTUD_ENABLE ? IP_PMTUDISC_DO : IP_PMTUDISC_DONT; -- cgit v1.2.3-70-g09d2