aboutsummaryrefslogtreecommitdiffstats
path: root/net
diff options
context:
space:
mode:
Diffstat (limited to 'net')
-rw-r--r--net/appletalk/aarp.c2
-rw-r--r--net/appletalk/ddp.c2
-rw-r--r--net/bridge/br_forward.c3
-rw-r--r--net/bridge/br_input.c4
-rw-r--r--net/bridge/br_netfilter.c38
-rw-r--r--net/core/netfilter.c138
-rw-r--r--net/core/skbuff.c6
-rw-r--r--net/ipv4/Kconfig26
-rw-r--r--net/ipv4/Makefile4
-rw-r--r--net/ipv4/af_inet.c12
-rw-r--r--net/ipv4/ah4.c2
-rw-r--r--net/ipv4/esp4.c2
-rw-r--r--net/ipv4/fib_frontend.c55
-rw-r--r--net/ipv4/fib_trie.c2454
-rw-r--r--net/ipv4/ip_input.c5
-rw-r--r--net/ipv4/ip_output.c11
-rw-r--r--net/ipv4/ipcomp.c11
-rw-r--r--net/ipv4/ipmr.c1
-rw-r--r--net/ipv4/ipvs/ip_vs_xmit.c1
-rw-r--r--net/ipv4/netfilter/arp_tables.c1
-rw-r--r--net/ipv4/netfilter/ip_conntrack_amanda.c7
-rw-r--r--net/ipv4/netfilter/ip_conntrack_core.c107
-rw-r--r--net/ipv4/netfilter/ip_conntrack_ftp.c7
-rw-r--r--net/ipv4/netfilter/ip_conntrack_irc.c7
-rw-r--r--net/ipv4/netfilter/ip_conntrack_proto_sctp.c23
-rw-r--r--net/ipv4/netfilter/ip_conntrack_proto_tcp.c27
-rw-r--r--net/ipv4/netfilter/ip_conntrack_proto_udp.c1
-rw-r--r--net/ipv4/netfilter/ip_conntrack_standalone.c22
-rw-r--r--net/ipv4/netfilter/ip_nat_core.c32
-rw-r--r--net/ipv4/netfilter/ip_nat_helper.c13
-rw-r--r--net/ipv4/netfilter/ip_nat_rule.c4
-rw-r--r--net/ipv4/netfilter/ip_nat_standalone.c5
-rw-r--r--net/ipv4/netfilter/ip_tables.c1
-rw-r--r--net/ipv4/netfilter/ipt_CLUSTERIP.c49
-rw-r--r--net/ipv4/netfilter/ipt_MASQUERADE.c10
-rw-r--r--net/ipv4/netfilter/ipt_REJECT.c13
-rw-r--r--net/ipv4/netfilter/ipt_ULOG.c15
-rw-r--r--net/ipv4/netfilter/ipt_hashlimit.c17
-rw-r--r--net/ipv4/netfilter/ipt_helper.c4
-rw-r--r--net/ipv4/xfrm4_output.c8
-rw-r--r--net/ipv4/xfrm4_state.c9
-rw-r--r--net/ipv4/xfrm4_tunnel.c2
-rw-r--r--net/ipv6/addrconf.c14
-rw-r--r--net/ipv6/ah6.c2
-rw-r--r--net/ipv6/anycast.c4
-rw-r--r--net/ipv6/esp6.c2
-rw-r--r--net/ipv6/ip6_fib.c19
-rw-r--r--net/ipv6/ip6_output.c3
-rw-r--r--net/ipv6/ipcomp6.c9
-rw-r--r--net/ipv6/ipv6_sockglue.c5
-rw-r--r--net/ipv6/mcast.c68
-rw-r--r--net/ipv6/ndisc.c4
-rw-r--r--net/ipv6/netfilter/ip6_tables.c1
-rw-r--r--net/ipv6/netfilter/ip6t_LOG.c54
-rw-r--r--net/ipv6/netfilter/ip6table_raw.c6
-rw-r--r--net/ipv6/route.c78
-rw-r--r--net/ipv6/xfrm6_tunnel.c2
-rw-r--r--net/key/af_key.c16
-rw-r--r--net/sctp/associola.c151
-rw-r--r--net/sctp/endpointola.c1
-rw-r--r--net/sctp/input.c2
-rw-r--r--net/sctp/outqueue.c11
-rw-r--r--net/sctp/sm_make_chunk.c20
-rw-r--r--net/sctp/sm_sideeffect.c105
-rw-r--r--net/sctp/sm_statefuns.c148
-rw-r--r--net/sctp/sm_statetable.c6
-rw-r--r--net/sctp/socket.c405
-rw-r--r--net/sctp/transport.c4
-rw-r--r--net/sunrpc/auth.c6
-rw-r--r--net/sunrpc/auth_gss/auth_gss.c18
-rw-r--r--net/sunrpc/clnt.c205
-rw-r--r--net/sunrpc/pmap_clnt.c9
-rw-r--r--net/sunrpc/sched.c84
-rw-r--r--net/sunrpc/sunrpc_syms.c6
-rw-r--r--net/sunrpc/svc.c36
-rw-r--r--net/sunrpc/xdr.c298
-rw-r--r--net/sunrpc/xprt.c71
-rw-r--r--net/xfrm/xfrm_policy.c1
-rw-r--r--net/xfrm/xfrm_state.c37
-rw-r--r--net/xfrm/xfrm_user.c9
80 files changed, 4144 insertions, 937 deletions
diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c
index 54640c0..10d0404 100644
--- a/net/appletalk/aarp.c
+++ b/net/appletalk/aarp.c
@@ -565,7 +565,7 @@ int aarp_send_ddp(struct net_device *dev, struct sk_buff *skb,
* numbers we just happen to need. Now put the
* length in the lower two.
*/
- *((__u16 *)skb->data) = htons(skb->len);
+ *((__be16 *)skb->data) = htons(skb->len);
ft = 1;
}
/*
diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c
index 876dbac..192b529 100644
--- a/net/appletalk/ddp.c
+++ b/net/appletalk/ddp.c
@@ -401,7 +401,7 @@ out_err:
}
/* Find a match for a specific network:node pair */
-static struct atalk_iface *atalk_find_interface(int net, int node)
+static struct atalk_iface *atalk_find_interface(__be16 net, int node)
{
struct atalk_iface *iface;
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index ef9f209..069253f 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -57,9 +57,6 @@ int br_forward_finish(struct sk_buff *skb)
static void __br_deliver(const struct net_bridge_port *to, struct sk_buff *skb)
{
skb->dev = to->dev;
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug = 0;
-#endif
NF_HOOK(PF_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev,
br_forward_finish);
}
diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index 8f5f2e7..9a45e62 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -23,11 +23,7 @@ const unsigned char bridge_ula[6] = { 0x01, 0x80, 0xc2, 0x00, 0x00, 0x00 };
static int br_pass_frame_up_finish(struct sk_buff *skb)
{
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug = 0;
-#endif
netif_receive_skb(skb);
-
return 0;
}
diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index be03d3a..03ae4ed 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -102,10 +102,6 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb)
{
struct nf_bridge_info *nf_bridge = skb->nf_bridge;
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug ^= (1 << NF_BR_PRE_ROUTING);
-#endif
-
if (nf_bridge->mask & BRNF_PKT_TYPE) {
skb->pkt_type = PACKET_OTHERHOST;
nf_bridge->mask ^= BRNF_PKT_TYPE;
@@ -182,10 +178,6 @@ static void __br_dnat_complain(void)
* --Bart, 20021007 (updated) */
static int br_nf_pre_routing_finish_bridge(struct sk_buff *skb)
{
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug |= (1 << NF_BR_PRE_ROUTING) | (1 << NF_BR_FORWARD);
-#endif
-
if (skb->pkt_type == PACKET_OTHERHOST) {
skb->pkt_type = PACKET_HOST;
skb->nf_bridge->mask |= BRNF_PKT_TYPE;
@@ -207,10 +199,6 @@ static int br_nf_pre_routing_finish(struct sk_buff *skb)
struct iphdr *iph = skb->nh.iph;
struct nf_bridge_info *nf_bridge = skb->nf_bridge;
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug ^= (1 << NF_BR_PRE_ROUTING);
-#endif
-
if (nf_bridge->mask & BRNF_PKT_TYPE) {
skb->pkt_type = PACKET_OTHERHOST;
nf_bridge->mask ^= BRNF_PKT_TYPE;
@@ -382,9 +370,6 @@ static unsigned int br_nf_pre_routing_ipv6(unsigned int hook,
if (hdr->nexthdr == NEXTHDR_HOP && check_hbh_len(skb))
goto inhdr_error;
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug ^= (1 << NF_IP6_PRE_ROUTING);
-#endif
if ((nf_bridge = nf_bridge_alloc(skb)) == NULL)
return NF_DROP;
setup_pre_routing(skb);
@@ -468,9 +453,6 @@ static unsigned int br_nf_pre_routing(unsigned int hook, struct sk_buff **pskb,
skb->ip_summed = CHECKSUM_NONE;
}
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug ^= (1 << NF_IP_PRE_ROUTING);
-#endif
if ((nf_bridge = nf_bridge_alloc(skb)) == NULL)
return NF_DROP;
setup_pre_routing(skb);
@@ -517,10 +499,6 @@ static int br_nf_forward_finish(struct sk_buff *skb)
struct net_device *in;
struct vlan_ethhdr *hdr = vlan_eth_hdr(skb);
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug ^= (1 << NF_BR_FORWARD);
-#endif
-
if (skb->protocol != __constant_htons(ETH_P_ARP) && !IS_VLAN_ARP) {
in = nf_bridge->physindev;
if (nf_bridge->mask & BRNF_PKT_TYPE) {
@@ -566,9 +544,6 @@ static unsigned int br_nf_forward_ip(unsigned int hook, struct sk_buff **pskb,
(*pskb)->nh.raw += VLAN_HLEN;
}
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug ^= (1 << NF_BR_FORWARD);
-#endif
nf_bridge = skb->nf_bridge;
if (skb->pkt_type == PACKET_OTHERHOST) {
skb->pkt_type = PACKET_HOST;
@@ -605,10 +580,6 @@ static unsigned int br_nf_forward_arp(unsigned int hook, struct sk_buff **pskb,
(*pskb)->nh.raw += VLAN_HLEN;
}
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug ^= (1 << NF_BR_FORWARD);
-#endif
-
if (skb->nh.arph->ar_pln != 4) {
if (IS_VLAN_ARP) {
skb_push(*pskb, VLAN_HLEN);
@@ -627,9 +598,6 @@ static unsigned int br_nf_forward_arp(unsigned int hook, struct sk_buff **pskb,
/* PF_BRIDGE/LOCAL_OUT ***********************************************/
static int br_nf_local_out_finish(struct sk_buff *skb)
{
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug &= ~(1 << NF_BR_LOCAL_OUT);
-#endif
if (skb->protocol == __constant_htons(ETH_P_8021Q)) {
skb_push(skb, VLAN_HLEN);
skb->nh.raw -= VLAN_HLEN;
@@ -731,10 +699,6 @@ static unsigned int br_nf_local_out(unsigned int hook, struct sk_buff **pskb,
realoutdev, br_nf_local_out_finish,
NF_IP_PRI_BRIDGE_SABOTAGE_FORWARD + 1);
} else {
-#ifdef CONFIG_NETFILTER_DEBUG
- skb->nf_debug ^= (1 << NF_IP_LOCAL_OUT);
-#endif
-
NF_HOOK_THRESH(pf, NF_IP_LOCAL_OUT, skb, realindev,
realoutdev, br_nf_local_out_finish,
NF_IP_PRI_BRIDGE_SABOTAGE_LOCAL_OUT + 1);
@@ -779,8 +743,6 @@ static unsigned int br_nf_post_routing(unsigned int hook, struct sk_buff **pskb,
printk(KERN_CRIT "br_netfilter: skb->dst == NULL.");
goto print_error;
}
-
- skb->nf_debug ^= (1 << NF_IP_POST_ROUTING);
#endif
/* We assume any code from br_dev_queue_push_xmit onwards doesn't care
diff --git a/net/core/netfilter.c b/net/core/netfilter.c
index 22a8f12..076c156 100644
--- a/net/core/netfilter.c
+++ b/net/core/netfilter.c
@@ -141,136 +141,6 @@ void nf_unregister_sockopt(struct nf_sockopt_ops *reg)
up(&nf_sockopt_mutex);
}
-#ifdef CONFIG_NETFILTER_DEBUG
-#include <net/ip.h>
-#include <net/tcp.h>
-#include <linux/netfilter_ipv4.h>
-
-static void debug_print_hooks_ip(unsigned int nf_debug)
-{
- if (nf_debug & (1 << NF_IP_PRE_ROUTING)) {
- printk("PRE_ROUTING ");
- nf_debug ^= (1 << NF_IP_PRE_ROUTING);
- }
- if (nf_debug & (1 << NF_IP_LOCAL_IN)) {
- printk("LOCAL_IN ");
- nf_debug ^= (1 << NF_IP_LOCAL_IN);
- }
- if (nf_debug & (1 << NF_IP_FORWARD)) {
- printk("FORWARD ");
- nf_debug ^= (1 << NF_IP_FORWARD);
- }
- if (nf_debug & (1 << NF_IP_LOCAL_OUT)) {
- printk("LOCAL_OUT ");
- nf_debug ^= (1 << NF_IP_LOCAL_OUT);
- }
- if (nf_debug & (1 << NF_IP_POST_ROUTING)) {
- printk("POST_ROUTING ");
- nf_debug ^= (1 << NF_IP_POST_ROUTING);
- }
- if (nf_debug)
- printk("Crap bits: 0x%04X", nf_debug);
- printk("\n");
-}
-
-static void nf_dump_skb(int pf, struct sk_buff *skb)
-{
- printk("skb: pf=%i %s dev=%s len=%u\n",
- pf,
- skb->sk ? "(owned)" : "(unowned)",
- skb->dev ? skb->dev->name : "(no dev)",
- skb->len);
- switch (pf) {
- case PF_INET: {
- const struct iphdr *ip = skb->nh.iph;
- __u32 *opt = (__u32 *) (ip + 1);
- int opti;
- __u16 src_port = 0, dst_port = 0;
-
- if (ip->protocol == IPPROTO_TCP
- || ip->protocol == IPPROTO_UDP) {
- struct tcphdr *tcp=(struct tcphdr *)((__u32 *)ip+ip->ihl);
- src_port = ntohs(tcp->source);
- dst_port = ntohs(tcp->dest);
- }
-
- printk("PROTO=%d %u.%u.%u.%u:%hu %u.%u.%u.%u:%hu"
- " L=%hu S=0x%2.2hX I=%hu F=0x%4.4hX T=%hu",
- ip->protocol, NIPQUAD(ip->saddr),
- src_port, NIPQUAD(ip->daddr),
- dst_port,
- ntohs(ip->tot_len), ip->tos, ntohs(ip->id),
- ntohs(ip->frag_off), ip->ttl);
-
- for (opti = 0; opti < (ip->ihl - sizeof(struct iphdr) / 4); opti++)
- printk(" O=0x%8.8X", *opt++);
- printk("\n");
- }
- }
-}
-
-void nf_debug_ip_local_deliver(struct sk_buff *skb)
-{
- /* If it's a loopback packet, it must have come through
- * NF_IP_LOCAL_OUT, NF_IP_RAW_INPUT, NF_IP_PRE_ROUTING and
- * NF_IP_LOCAL_IN. Otherwise, must have gone through
- * NF_IP_RAW_INPUT and NF_IP_PRE_ROUTING. */
- if (!skb->dev) {
- printk("ip_local_deliver: skb->dev is NULL.\n");
- } else {
- if (skb->nf_debug != ((1<<NF_IP_PRE_ROUTING)
- | (1<<NF_IP_LOCAL_IN))) {
- printk("ip_local_deliver: bad skb: ");
- debug_print_hooks_ip(skb->nf_debug);
- nf_dump_skb(PF_INET, skb);
- }
- }
-}
-
-void nf_debug_ip_loopback_xmit(struct sk_buff *newskb)
-{
- if (newskb->nf_debug != ((1 << NF_IP_LOCAL_OUT)
- | (1 << NF_IP_POST_ROUTING))) {
- printk("ip_dev_loopback_xmit: bad owned skb = %p: ",
- newskb);
- debug_print_hooks_ip(newskb->nf_debug);
- nf_dump_skb(PF_INET, newskb);
- }
-}
-
-void nf_debug_ip_finish_output2(struct sk_buff *skb)
-{
- /* If it's owned, it must have gone through the
- * NF_IP_LOCAL_OUT and NF_IP_POST_ROUTING.
- * Otherwise, must have gone through
- * NF_IP_PRE_ROUTING, NF_IP_FORWARD and NF_IP_POST_ROUTING.
- */
- if (skb->sk) {
- if (skb->nf_debug != ((1 << NF_IP_LOCAL_OUT)
- | (1 << NF_IP_POST_ROUTING))) {
- printk("ip_finish_output: bad owned skb = %p: ", skb);
- debug_print_hooks_ip(skb->nf_debug);
- nf_dump_skb(PF_INET, skb);
- }
- } else {
- if (skb->nf_debug != ((1 << NF_IP_PRE_ROUTING)
- | (1 << NF_IP_FORWARD)
- | (1 << NF_IP_POST_ROUTING))) {
- /* Fragments, entunnelled packets, TCP RSTs
- generated by ipt_REJECT will have no
- owners, but still may be local */
- if (skb->nf_debug != ((1 << NF_IP_LOCAL_OUT)
- | (1 << NF_IP_POST_ROUTING))){
- printk("ip_finish_output:"
- " bad unowned skb = %p: ",skb);
- debug_print_hooks_ip(skb->nf_debug);
- nf_dump_skb(PF_INET, skb);
- }
- }
- }
-}
-#endif /*CONFIG_NETFILTER_DEBUG*/
-
/* Call get/setsockopt() */
static int nf_sockopt(struct sock *sk, int pf, int val,
char __user *opt, int *len, int get)
@@ -488,14 +358,6 @@ int nf_hook_slow(int pf, unsigned int hook, struct sk_buff **pskb,
/* We may already have this, but read-locks nest anyway */
rcu_read_lock();
-#ifdef CONFIG_NETFILTER_DEBUG
- if (unlikely((*pskb)->nf_debug & (1 << hook))) {
- printk("nf_hook: hook %i already set.\n", hook);
- nf_dump_skb(pf, *pskb);
- }
- (*pskb)->nf_debug |= (1 << hook);
-#endif
-
elem = &nf_hooks[pf][hook];
next_hook:
verdict = nf_iterate(&nf_hooks[pf][hook], pskb, hook, indev,
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index f65b3de..6d68c03 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -365,9 +365,6 @@ struct sk_buff *skb_clone(struct sk_buff *skb, int gfp_mask)
C(nfct);
nf_conntrack_get(skb->nfct);
C(nfctinfo);
-#ifdef CONFIG_NETFILTER_DEBUG
- C(nf_debug);
-#endif
#ifdef CONFIG_BRIDGE_NETFILTER
C(nf_bridge);
nf_bridge_get(skb->nf_bridge);
@@ -432,9 +429,6 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
new->nfct = old->nfct;
nf_conntrack_get(old->nfct);
new->nfctinfo = old->nfctinfo;
-#ifdef CONFIG_NETFILTER_DEBUG
- new->nf_debug = old->nf_debug;
-#endif
#ifdef CONFIG_BRIDGE_NETFILTER
new->nf_bridge = old->nf_bridge;
nf_bridge_get(old->nf_bridge);
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 6d3e8b1..567b03b 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -1,6 +1,32 @@
#
# IP configuration
#
+choice
+ prompt "Choose IP: FIB lookup"
+ depends on INET
+ default IP_FIB_HASH
+
+config IP_FIB_HASH
+ bool "FIB_HASH"
+ ---help---
+ Current FIB is very proven and good enough for most users.
+
+config IP_FIB_TRIE
+ bool "FIB_TRIE"
+ ---help---
+ Use new experimental LC-trie as FIB lookup algoritm.
+ This improves lookup performance
+
+ LC-trie is described in:
+
+ IP-address lookup using LC-tries. Stefan Nilsson and Gunnar Karlsson
+ IEEE Journal on Selected Areas in Communications, 17(6):1083-1092, June 1999
+ An experimental study of compression methods for dynamic tries
+ Stefan Nilsson and Matti Tikkanen. Algorithmica, 33(1):19-33, 2002.
+ http://www.nada.kth.se/~snilsson/public/papers/dyntrie2/
+
+endchoice
+
config IP_MULTICAST
bool "IP: multicasting"
depends on INET
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 8b37962..65d57d8 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -7,8 +7,10 @@ obj-y := utils.o route.o inetpeer.o protocol.o \
ip_output.o ip_sockglue.o \
tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o tcp_minisocks.o \
datagram.o raw.o udp.o arp.o icmp.o devinet.o af_inet.o igmp.o \
- sysctl_net_ipv4.o fib_frontend.o fib_semantics.o fib_hash.o
+ sysctl_net_ipv4.o fib_frontend.o fib_semantics.o
+obj-$(CONFIG_IP_FIB_HASH) += fib_hash.o
+obj-$(CONFIG_IP_FIB_TRIE) += fib_trie.o
obj-$(CONFIG_PROC_FS) += proc.o
obj-$(CONFIG_IP_MULTIPLE_TABLES) += fib_rules.o
obj-$(CONFIG_IP_MROUTE) += ipmr.o
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 03942f1..658e797 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1119,6 +1119,10 @@ module_init(inet_init);
#ifdef CONFIG_PROC_FS
extern int fib_proc_init(void);
extern void fib_proc_exit(void);
+#ifdef CONFIG_IP_FIB_TRIE
+extern int fib_stat_proc_init(void);
+extern void fib_stat_proc_exit(void);
+#endif
extern int ip_misc_proc_init(void);
extern int raw_proc_init(void);
extern void raw_proc_exit(void);
@@ -1139,11 +1143,19 @@ static int __init ipv4_proc_init(void)
goto out_udp;
if (fib_proc_init())
goto out_fib;
+#ifdef CONFIG_IP_FIB_TRIE
+ if (fib_stat_proc_init())
+ goto out_fib_stat;
+ #endif
if (ip_misc_proc_init())
goto out_misc;
out:
return rc;
out_misc:
+#ifdef CONFIG_IP_FIB_TRIE
+ fib_stat_proc_exit();
+out_fib_stat:
+#endif
fib_proc_exit();
out_fib:
udp4_proc_exit();
diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c
index 0e98f22..514c85b 100644
--- a/net/ipv4/ah4.c
+++ b/net/ipv4/ah4.c
@@ -200,7 +200,7 @@ static void ah4_err(struct sk_buff *skb, u32 info)
xfrm_state_put(x);
}
-static int ah_init_state(struct xfrm_state *x, void *args)
+static int ah_init_state(struct xfrm_state *x)
{
struct ah_data *ahp = NULL;
struct xfrm_algo_desc *aalg_desc;
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index eae84cc..ba57446 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -362,7 +362,7 @@ static void esp_destroy(struct xfrm_state *x)
kfree(esp);
}
-static int esp_init_state(struct xfrm_state *x, void *args)
+static int esp_init_state(struct xfrm_state *x)
{
struct esp_data *esp = NULL;
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 563e7d6..cd8e45a 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -516,6 +516,60 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa)
#undef BRD1_OK
}
+static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb )
+{
+
+ struct fib_result res;
+ struct flowi fl = { .nl_u = { .ip4_u = { .daddr = frn->fl_addr,
+ .fwmark = frn->fl_fwmark,
+ .tos = frn->fl_tos,
+ .scope = frn->fl_scope } } };
+ if (tb) {
+ local_bh_disable();
+
+ frn->tb_id = tb->tb_id;
+ frn->err = tb->tb_lookup(tb, &fl, &res);
+
+ if (!frn->err) {
+ frn->prefixlen = res.prefixlen;
+ frn->nh_sel = res.nh_sel;
+ frn->type = res.type;
+ frn->scope = res.scope;
+ }
+ local_bh_enable();
+ }
+}
+
+static void nl_fib_input(struct sock *sk, int len)
+{
+ struct sk_buff *skb = NULL;
+ struct nlmsghdr *nlh = NULL;
+ struct fib_result_nl *frn;
+ int err;
+ u32 pid;
+ struct fib_table *tb;
+
+ skb = skb_recv_datagram(sk, 0, 0, &err);
+ nlh = (struct nlmsghdr *)skb->data;
+
+ frn = (struct fib_result_nl *) NLMSG_DATA(nlh);
+ tb = fib_get_table(frn->tb_id_in);
+
+ nl_fib_lookup(frn, tb);
+
+ pid = nlh->nlmsg_pid; /*pid of sending process */
+ NETLINK_CB(skb).groups = 0; /* not in mcast group */
+ NETLINK_CB(skb).pid = 0; /* from kernel */
+ NETLINK_CB(skb).dst_pid = pid;
+ NETLINK_CB(skb).dst_groups = 0; /* unicast */
+ netlink_unicast(sk, skb, pid, MSG_DONTWAIT);
+}
+
+static void nl_fib_lookup_init(void)
+{
+ netlink_kernel_create(NETLINK_FIB_LOOKUP, nl_fib_input);
+}
+
static void fib_disable_ip(struct net_device *dev, int force)
{
if (fib_sync_down(0, dev, force))
@@ -604,6 +658,7 @@ void __init ip_fib_init(void)
register_netdevice_notifier(&fib_netdev_notifier);
register_inetaddr_notifier(&fib_inetaddr_notifier);
+ nl_fib_lookup_init();
}
EXPORT_SYMBOL(inet_addr_type);
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
new file mode 100644
index 0000000..0671569
--- /dev/null
+++ b/net/ipv4/fib_trie.c
@@ -0,0 +1,2454 @@
+/*
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * Robert Olsson <robert.olsson@its.uu.se> Uppsala Universitet
+ * & Swedish University of Agricultural Sciences.
+ *
+ * Jens Laas <jens.laas@data.slu.se> Swedish University of
+ * Agricultural Sciences.
+ *
+ * Hans Liss <hans.liss@its.uu.se> Uppsala Universitet
+ *
+ * This work is based on the LPC-trie which is originally descibed in:
+ *
+ * An experimental study of compression methods for dynamic tries
+ * Stefan Nilsson and Matti Tikkanen. Algorithmica, 33(1):19-33, 2002.
+ * http://www.nada.kth.se/~snilsson/public/papers/dyntrie2/
+ *
+ *
+ * IP-address lookup using LC-tries. Stefan Nilsson and Gunnar Karlsson
+ * IEEE Journal on Selected Areas in Communications, 17(6):1083-1092, June 1999
+ *
+ * Version: $Id: fib_trie.c,v 1.3 2005/06/08 14:20:01 robert Exp $
+ *
+ *
+ * Code from fib_hash has been reused which includes the following header:
+ *
+ *
+ * INET An implementation of the TCP/IP protocol suite for the LINUX
+ * operating system. INET is implemented using the BSD Socket
+ * interface as the means of communication with the user level.
+ *
+ * IPv4 FIB: lookup engine and maintenance routines.
+ *
+ *
+ * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#define VERSION "0.323"
+
+#include <linux/config.h>
+#include <asm/uaccess.h>
+#include <asm/system.h>
+#include <asm/bitops.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/string.h>
+#include <linux/socket.h>
+#include <linux/sockios.h>
+#include <linux/errno.h>
+#include <linux/in.h>
+#include <linux/inet.h>
+#include <linux/netdevice.h>
+#include <linux/if_arp.h>
+#include <linux/proc_fs.h>
+#include <linux/skbuff.h>
+#include <linux/netlink.h>
+#include <linux/init.h>
+#include <linux/list.h>
+#include <net/ip.h>
+#include <net/protocol.h>
+#include <net/route.h>
+#include <net/tcp.h>
+#include <net/sock.h>
+#include <net/ip_fib.h>
+#include "fib_lookup.h"
+
+#undef CONFIG_IP_FIB_TRIE_STATS
+#define MAX_CHILDS 16384
+
+#define EXTRACT(p, n, str) ((str)<<(p)>>(32-(n)))
+#define KEYLENGTH (8*sizeof(t_key))
+#define MASK_PFX(k, l) (((l)==0)?0:(k >> (KEYLENGTH-l)) << (KEYLENGTH-l))
+#define TKEY_GET_MASK(offset, bits) (((bits)==0)?0:((t_key)(-1) << (KEYLENGTH - bits) >> offset))
+
+static DEFINE_RWLOCK(fib_lock);
+
+typedef unsigned int t_key;
+
+#define T_TNODE 0
+#define T_LEAF 1
+#define NODE_TYPE_MASK 0x1UL
+#define NODE_PARENT(_node) \
+((struct tnode *)((_node)->_parent & ~NODE_TYPE_MASK))
+#define NODE_SET_PARENT(_node, _ptr) \
+((_node)->_parent = (((unsigned long)(_ptr)) | \
+ ((_node)->_parent & NODE_TYPE_MASK)))
+#define NODE_INIT_PARENT(_node, _type) \
+((_node)->_parent = (_type))
+#define NODE_TYPE(_node) \
+((_node)->_parent & NODE_TYPE_MASK)
+
+#define IS_TNODE(n) (!(n->_parent & T_LEAF))
+#define IS_LEAF(n) (n->_parent & T_LEAF)
+
+struct node {
+ t_key key;
+ unsigned long _parent;
+};
+
+struct leaf {
+ t_key key;
+ unsigned long _parent;
+ struct hlist_head list;
+};
+
+struct leaf_info {
+ struct hlist_node hlist;
+ int plen;
+ struct list_head falh;
+};
+
+struct tnode {
+ t_key key;
+ unsigned long _parent;
+ unsigned short pos:5; /* 2log(KEYLENGTH) bits needed */
+ unsigned short bits:5; /* 2log(KEYLENGTH) bits needed */
+ unsigned short full_children; /* KEYLENGTH bits needed */
+ unsigned short empty_children; /* KEYLENGTH bits needed */
+ struct node *child[0];
+};
+
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+struct trie_use_stats {
+ unsigned int gets;
+ unsigned int backtrack;
+ unsigned int semantic_match_passed;
+ unsigned int semantic_match_miss;
+ unsigned int null_node_hit;
+};
+#endif
+
+struct trie_stat {
+ unsigned int totdepth;
+ unsigned int maxdepth;
+ unsigned int tnodes;
+ unsigned int leaves;
+ unsigned int nullpointers;
+ unsigned int nodesizes[MAX_CHILDS];
+};
+
+struct trie {
+ struct node *trie;
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+ struct trie_use_stats stats;
+#endif
+ int size;
+ unsigned int revision;
+};
+
+static int trie_debug = 0;
+
+static int tnode_full(struct tnode *tn, struct node *n);
+static void put_child(struct trie *t, struct tnode *tn, int i, struct node *n);
+static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, int wasfull);
+static int tnode_child_length(struct tnode *tn);
+static struct node *resize(struct trie *t, struct tnode *tn);
+static struct tnode *inflate(struct trie *t, struct tnode *tn);
+static struct tnode *halve(struct trie *t, struct tnode *tn);
+static void tnode_free(struct tnode *tn);
+static void trie_dump_seq(struct seq_file *seq, struct trie *t);
+extern struct fib_alias *fib_find_alias(struct list_head *fah, u8 tos, u32 prio);
+extern int fib_detect_death(struct fib_info *fi, int order,
+ struct fib_info **last_resort, int *last_idx, int *dflt);
+
+extern void rtmsg_fib(int event, u32 key, struct fib_alias *fa, int z, int tb_id,
+ struct nlmsghdr *n, struct netlink_skb_parms *req);
+
+static kmem_cache_t *fn_alias_kmem;
+static struct trie *trie_local = NULL, *trie_main = NULL;
+
+static void trie_bug(char *err)
+{
+ printk("Trie Bug: %s\n", err);
+ BUG();
+}
+
+static inline struct node *tnode_get_child(struct tnode *tn, int i)
+{
+ if (i >= 1<<tn->bits)
+ trie_bug("tnode_get_child");
+
+ return tn->child[i];
+}
+
+static inline int tnode_child_length(struct tnode *tn)
+{
+ return 1<<tn->bits;
+}
+
+/*
+ _________________________________________________________________
+ | i | i | i | i | i | i | i | N | N | N | S | S | S | S | S | C |
+ ----------------------------------------------------------------
+ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
+
+ _________________________________________________________________
+ | C | C | C | u | u | u | u | u | u | u | u | u | u | u | u | u |
+ -----------------------------------------------------------------
+ 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
+
+ tp->pos = 7
+ tp->bits = 3
+ n->pos = 15
+ n->bits=4
+ KEYLENGTH=32
+*/
+
+static inline t_key tkey_extract_bits(t_key a, int offset, int bits)
+{
+ if (offset < KEYLENGTH)
+ return ((t_key)(a << offset)) >> (KEYLENGTH - bits);
+ else
+ return 0;
+}
+
+static inline int tkey_equals(t_key a, t_key b)
+{
+ return a == b;
+}
+
+static inline int tkey_sub_equals(t_key a, int offset, int bits, t_key b)
+{
+ if (bits == 0 || offset >= KEYLENGTH)
+ return 1;
+ bits = bits > KEYLENGTH ? KEYLENGTH : bits;
+ return ((a ^ b) << offset) >> (KEYLENGTH - bits) == 0;
+}
+
+static inline int tkey_mismatch(t_key a, int offset, t_key b)
+{
+ t_key diff = a ^ b;
+ int i = offset;
+
+ if(!diff)
+ return 0;
+ while((diff << i) >> (KEYLENGTH-1) == 0)
+ i++;
+ return i;
+}
+
+/* Candiate for fib_semantics */
+
+static void fn_free_alias(struct fib_alias *fa)
+{
+ fib_release_info(fa->fa_info);
+ kmem_cache_free(fn_alias_kmem, fa);
+}
+
+/*
+ To understand this stuff, an understanding of keys and all their bits is
+ necessary. Every node in the trie has a key associated with it, but not
+ all of the bits in that key are significant.
+
+ Consider a node 'n' and its parent 'tp'.
+
+ If n is a leaf, every bit in its key is significant. Its presence is
+ necessitaded by path compression, since during a tree traversal (when
+ searching for a leaf - unless we are doing an insertion) we will completely
+ ignore all skipped bits we encounter. Thus we need to verify, at the end of
+ a potentially successful search, that we have indeed been walking the
+ correct key path.
+
+ Note that we can never "miss" the correct key in the tree if present by
+ following the wrong path. Path compression ensures that segments of the key
+ that are the same for all keys with a given prefix are skipped, but the
+ skipped part *is* identical for each node in the subtrie below the skipped
+ bit! trie_insert() in this implementation takes care of that - note the
+ call to tkey_sub_equals() in trie_insert().
+
+ if n is an internal node - a 'tnode' here, the various parts of its key
+ have many different meanings.
+
+ Example:
+ _________________________________________________________________
+ | i | i | i | i | i | i | i | N | N | N | S | S | S | S | S | C |
+ -----------------------------------------------------------------
+ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
+
+ _________________________________________________________________
+ | C | C | C | u | u | u | u | u | u | u | u | u | u | u | u | u |
+ -----------------------------------------------------------------
+ 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
+
+ tp->pos = 7
+ tp->bits = 3
+ n->pos = 15
+ n->bits=4
+
+ First, let's just ignore the bits that come before the parent tp, that is
+ the bits from 0 to (tp->pos-1). They are *known* but at this point we do
+ not use them for anything.
+
+ The bits from (tp->pos) to (tp->pos + tp->bits - 1) - "N", above - are the
+ index into the parent's child array. That is, they will be used to find
+ 'n' among tp's children.
+
+ The bits from (tp->pos + tp->bits) to (n->pos - 1) - "S" - are skipped bits
+ for the node n.
+
+ All the bits we have seen so far are significant to the node n. The rest
+ of the bits are really not needed or indeed known in n->key.
+
+ The bits from (n->pos) to (n->pos + n->bits - 1) - "C" - are the index into
+ n's child array, and will of course be different for each child.
+
+ The rest of the bits, from (n->pos + n->bits) onward, are completely unknown
+ at this point.
+
+*/
+
+static void check_tnode(struct tnode *tn)
+{
+ if(tn && tn->pos+tn->bits > 32) {
+ printk("TNODE ERROR tn=%p, pos=%d, bits=%d\n", tn, tn->pos, tn->bits);
+ }
+}
+
+static int halve_threshold = 25;
+static int inflate_threshold = 50;
+
+static struct leaf *leaf_new(void)
+{
+ struct leaf *l = kmalloc(sizeof(struct leaf), GFP_KERNEL);
+ if(l) {
+ NODE_INIT_PARENT(l, T_LEAF);
+ INIT_HLIST_HEAD(&l->list);
+ }
+ return l;
+}
+
+static struct leaf_info *leaf_info_new(int plen)
+{
+ struct leaf_info *li = kmalloc(sizeof(struct leaf_info), GFP_KERNEL);
+ li->plen = plen;
+ INIT_LIST_HEAD(&li->falh);
+ return li;
+}
+
+static inline void free_leaf(struct leaf *l)
+{
+ kfree(l);
+}
+
+static inline void free_leaf_info(struct leaf_info *li)
+{
+ kfree(li);
+}
+
+static struct tnode* tnode_new(t_key key, int pos, int bits)
+{
+ int nchildren = 1<<bits;
+ int sz = sizeof(struct tnode) + nchildren * sizeof(struct node *);
+ struct tnode *tn = kmalloc(sz, GFP_KERNEL);
+
+ if(tn) {
+ memset(tn, 0, sz);
+ NODE_INIT_PARENT(tn, T_TNODE);
+ tn->pos = pos;
+ tn->bits = bits;
+ tn->key = key;
+ tn->full_children = 0;
+ tn->empty_children = 1<<bits;
+ }
+ if(trie_debug > 0)
+ printk("AT %p s=%u %u\n", tn, (unsigned int) sizeof(struct tnode),
+ (unsigned int) (sizeof(struct node) * 1<<bits));
+ return tn;
+}
+
+static void tnode_free(struct tnode *tn)
+{
+ if(!tn) {
+ trie_bug("tnode_free\n");
+ }
+ if(IS_LEAF(tn)) {
+ free_leaf((struct leaf *)tn);
+ if(trie_debug > 0 )
+ printk("FL %p \n", tn);
+ }
+ else if(IS_TNODE(tn)) {
+ kfree(tn);
+ if(trie_debug > 0 )
+ printk("FT %p \n", tn);
+ }
+ else {
+ trie_bug("tnode_free\n");
+ }
+}
+
+/*
+ * Check whether a tnode 'n' is "full", i.e. it is an internal node
+ * and no bits are skipped. See discussion in dyntree paper p. 6
+ */
+
+static inline int tnode_full(struct tnode *tn, struct node *n)
+{
+ if(n == NULL || IS_LEAF(n))
+ return 0;
+
+ return ((struct tnode *) n)->pos == tn->pos + tn->bits;
+}
+
+static inline void put_child(struct trie *t, struct tnode *tn, int i, struct node *n)
+{
+ tnode_put_child_reorg(tn, i, n, -1);
+}
+
+ /*
+ * Add a child at position i overwriting the old value.
+ * Update the value of full_children and empty_children.
+ */
+
+static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, int wasfull)
+{
+ struct node *chi;
+ int isfull;
+
+ if(i >= 1<<tn->bits) {
+ printk("bits=%d, i=%d\n", tn->bits, i);
+ trie_bug("tnode_put_child_reorg bits");
+ }
+ write_lock_bh(&fib_lock);
+ chi = tn->child[i];
+
+ /* update emptyChildren */
+ if (n == NULL && chi != NULL)
+ tn->empty_children++;
+ else if (n != NULL && chi == NULL)
+ tn->empty_children--;
+
+ /* update fullChildren */
+ if (wasfull == -1)
+ wasfull = tnode_full(tn, chi);
+
+ isfull = tnode_full(tn, n);
+ if (wasfull && !isfull)
+ tn->full_children--;
+
+ else if (!wasfull && isfull)
+ tn->full_children++;
+ if(n)
+ NODE_SET_PARENT(n, tn);
+
+ tn->child[i] = n;
+ write_unlock_bh(&fib_lock);
+}
+
+static struct node *resize(struct trie *t, struct tnode *tn)
+{
+ int i;
+
+ if (!tn)
+ return NULL;
+
+ if(trie_debug)
+ printk("In tnode_resize %p inflate_threshold=%d threshold=%d\n",
+ tn, inflate_threshold, halve_threshold);
+
+ /* No children */
+ if (tn->empty_children == tnode_child_length(tn)) {
+ tnode_free(tn);
+ return NULL;
+ }
+ /* One child */
+ if (tn->empty_children == tnode_child_length(tn) - 1)
+ for (i = 0; i < tnode_child_length(tn); i++) {
+
+ write_lock_bh(&fib_lock);
+ if (tn->child[i] != NULL) {
+
+ /* compress one level */
+ struct node *n = tn->child[i];
+ if(n)
+ NODE_INIT_PARENT(n, NODE_TYPE(n));
+
+ write_unlock_bh(&fib_lock);
+ tnode_free(tn);
+ return n;
+ }
+ write_unlock_bh(&fib_lock);
+ }
+ /*
+ * Double as long as the resulting node has a number of
+ * nonempty nodes that are above the threshold.
+ */
+
+ /*
+ * From "Implementing a dynamic compressed trie" by Stefan Nilsson of
+ * the Helsinki University of Technology and Matti Tikkanen of Nokia
+ * Telecommunications, page 6:
+ * "A node is doubled if the ratio of non-empty children to all
+ * children in the *doubled* node is at least 'high'."
+ *
+ * 'high' in this instance is the variable 'inflate_threshold'. It
+ * is expressed as a percentage, so we multiply it with
+ * tnode_child_length() and instead of multiplying by 2 (since the
+ * child array will be doubled by inflate()) and multiplying
+ * the left-hand side by 100 (to handle the percentage thing) we
+ * multiply the left-hand side by 50.
+ *
+ * The left-hand side may look a bit weird: tnode_child_length(tn)
+ * - tn->empty_children is of course the number of non-null children
+ * in the current node. tn->full_children is the number of "full"
+ * children, that is non-null tnodes with a skip value of 0.
+ * All of those will be doubled in the resulting inflated tnode, so
+ * we just count them one extra time here.
+ *
+ * A clearer way to write this would be:
+ *
+ * to_be_doubled = tn->full_children;
+ * not_to_be_doubled = tnode_child_length(tn) - tn->empty_children -
+ * tn->full_children;
+ *
+ * new_child_length = tnode_child_length(tn) * 2;
+ *
+ * new_fill_factor = 100 * (not_to_be_doubled + 2*to_be_doubled) /
+ * new_child_length;
+ * if (new_fill_factor >= inflate_threshold)
+ *
+ * ...and so on, tho it would mess up the while() loop.
+ *
+ * anyway,
+ * 100 * (not_to_be_doubled + 2*to_be_doubled) / new_child_length >=
+ * inflate_threshold
+ *
+ * avoid a division:
+ * 100 * (not_to_be_doubled + 2*to_be_doubled) >=
+ * inflate_threshold * new_child_length
+ *
+ * expand not_to_be_doubled and to_be_doubled, and shorten:
+ * 100 * (tnode_child_length(tn) - tn->empty_children +
+ * tn->full_children ) >= inflate_threshold * new_child_length
+ *
+ * expand new_child_length:
+ * 100 * (tnode_child_length(tn) - tn->empty_children +
+ * tn->full_children ) >=
+ * inflate_threshold * tnode_child_length(tn) * 2
+ *
+ * shorten again:
+ * 50 * (tn->full_children + tnode_child_length(tn) -
+ * tn->empty_children ) >= inflate_threshold *
+ * tnode_child_length(tn)
+ *
+ */
+
+ check_tnode(tn);
+
+ while ((tn->full_children > 0 &&
+ 50 * (tn->full_children + tnode_child_length(tn) - tn->empty_children) >=
+ inflate_threshold * tnode_child_length(tn))) {
+
+ tn = inflate(t, tn);
+ }
+
+ check_tnode(tn);
+
+ /*
+ * Halve as long as the number of empty children in this
+ * node is above threshold.
+ */
+ while (tn->bits > 1 &&
+ 100 * (tnode_child_length(tn) - tn->empty_children) <
+ halve_threshold * tnode_child_length(tn))
+
+ tn = halve(t, tn);
+
+ /* Only one child remains */
+
+ if (tn->empty_children == tnode_child_length(tn) - 1)
+ for (i = 0; i < tnode_child_length(tn); i++) {
+
+ write_lock_bh(&fib_lock);
+ if (tn->child[i] != NULL) {
+ /* compress one level */
+ struct node *n = tn->child[i];
+
+ if(n)
+ NODE_INIT_PARENT(n, NODE_TYPE(n));
+
+ write_unlock_bh(&fib_lock);
+ tnode_free(tn);
+ return n;
+ }
+ write_unlock_bh(&fib_lock);
+ }
+
+ return (struct node *) tn;
+}
+
+static struct tnode *inflate(struct trie *t, struct tnode *tn)
+{
+ struct tnode *inode;
+ struct tnode *oldtnode = tn;
+ int olen = tnode_child_length(tn);
+ int i;
+
+ if(trie_debug)
+ printk("In inflate\n");
+
+ tn = tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits + 1);
+
+ if (!tn)
+ trie_bug("tnode_new failed");
+
+ for(i = 0; i < olen; i++) {
+ struct node *node = tnode_get_child(oldtnode, i);
+
+ /* An empty child */
+ if (node == NULL)
+ continue;
+
+ /* A leaf or an internal node with skipped bits */
+
+ if(IS_LEAF(node) || ((struct tnode *) node)->pos >
+ tn->pos + tn->bits - 1) {
+ if(tkey_extract_bits(node->key, tn->pos + tn->bits - 1,
+ 1) == 0)
+ put_child(t, tn, 2*i, node);
+ else
+ put_child(t, tn, 2*i+1, node);
+ continue;
+ }
+
+ /* An internal node with two children */
+ inode = (struct tnode *) node;
+
+ if (inode->bits == 1) {
+ put_child(t, tn, 2*i, inode->child[0]);
+ put_child(t, tn, 2*i+1, inode->child[1]);
+
+ tnode_free(inode);
+ }
+
+ /* An internal node with more than two children */
+ else {
+ struct tnode *left, *right;
+ int size, j;
+
+ /* We will replace this node 'inode' with two new
+ * ones, 'left' and 'right', each with half of the
+ * original children. The two new nodes will have
+ * a position one bit further down the key and this
+ * means that the "significant" part of their keys
+ * (see the discussion near the top of this file)
+ * will differ by one bit, which will be "0" in
+ * left's key and "1" in right's key. Since we are
+ * moving the key position by one step, the bit that
+ * we are moving away from - the bit at position
+ * (inode->pos) - is the one that will differ between
+ * left and right. So... we synthesize that bit in the
+ * two new keys.
+ * The mask 'm' below will be a single "one" bit at
+ * the position (inode->pos)
+ */
+
+ t_key m = TKEY_GET_MASK(inode->pos, 1);
+
+ /* Use the old key, but set the new significant
+ * bit to zero.
+ */
+ left = tnode_new(inode->key&(~m), inode->pos + 1,
+ inode->bits - 1);
+
+ if(!left)
+ trie_bug("tnode_new failed");
+
+
+ /* Use the old key, but set the new significant
+ * bit to one.
+ */
+ right = tnode_new(inode->key|m, inode->pos + 1,
+ inode->bits - 1);
+
+ if(!right)
+ trie_bug("tnode_new failed");
+
+ size = tnode_child_length(left);
+ for(j = 0; j < size; j++) {
+ put_child(t, left, j, inode->child[j]);
+ put_child(t, right, j, inode->child[j + size]);
+ }
+ put_child(t, tn, 2*i, resize(t, left));
+ put_child(t, tn, 2*i+1, resize(t, right));
+
+ tnode_free(inode);
+ }
+ }
+ tnode_free(oldtnode);
+ return tn;
+}
+
+static struct tnode *halve(struct trie *t, struct tnode *tn)
+{
+ struct tnode *oldtnode = tn;
+ struct node *left, *right;
+ int i;
+ int olen = tnode_child_length(tn);
+
+ if(trie_debug) printk("In halve\n");
+
+ tn=tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits - 1);
+
+ if(!tn)
+ trie_bug("tnode_new failed");
+
+ for(i = 0; i < olen; i += 2) {
+ left = tnode_get_child(oldtnode, i);
+ right = tnode_get_child(oldtnode, i+1);
+
+ /* At least one of the children is empty */
+ if (left == NULL) {
+ if (right == NULL) /* Both are empty */
+ continue;
+ put_child(t, tn, i/2, right);
+ } else if (right == NULL)
+ put_child(t, tn, i/2, left);
+
+ /* Two nonempty children */
+ else {
+ struct tnode *newBinNode =
+ tnode_new(left->key, tn->pos + tn->bits, 1);
+
+ if(!newBinNode)
+ trie_bug("tnode_new failed");
+
+ put_child(t, newBinNode, 0, left);
+ put_child(t, newBinNode, 1, right);
+ put_child(t, tn, i/2, resize(t, newBinNode));
+ }
+ }
+ tnode_free(oldtnode);
+ return tn;
+}
+
+static void *trie_init(struct trie *t)
+{
+ if(t) {
+ t->size = 0;
+ t->trie = NULL;
+ t->revision = 0;
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+ memset(&t->stats, 0, sizeof(struct trie_use_stats));
+#endif
+ }
+ return t;
+}
+
+static struct leaf_info *find_leaf_info(struct hlist_head *head, int plen)
+{
+ struct hlist_node *node;
+ struct leaf_info *li;
+
+ hlist_for_each_entry(li, node, head, hlist) {
+
+ if ( li->plen == plen )
+ return li;
+ }
+ return NULL;
+}
+
+static inline struct list_head * get_fa_head(struct leaf *l, int plen)
+{
+ struct list_head *fa_head=NULL;
+ struct leaf_info *li = find_leaf_info(&l->list, plen);
+
+ if(li)
+ fa_head = &li->falh;
+
+ return fa_head;
+}
+
+static void insert_leaf_info(struct hlist_head *head, struct leaf_info *new)
+{
+ struct leaf_info *li=NULL, *last=NULL;
+ struct hlist_node *node, *tmp;
+
+ write_lock_bh(&fib_lock);
+
+ if(hlist_empty(head))
+ hlist_add_head(&new->hlist, head);
+ else {
+ hlist_for_each_entry_safe(li, node, tmp, head, hlist) {
+
+ if (new->plen > li->plen)
+ break;
+
+ last = li;
+ }
+ if(last)
+ hlist_add_after(&last->hlist, &new->hlist);
+ else
+ hlist_add_before(&new->hlist, &li->hlist);
+ }
+ write_unlock_bh(&fib_lock);
+}
+
+static struct leaf *
+fib_find_node(struct trie *t, u32 key)
+{
+ int pos;
+ struct tnode *tn;
+ struct node *n;
+
+ pos = 0;
+ n=t->trie;
+
+ while (n != NULL && NODE_TYPE(n) == T_TNODE) {
+ tn = (struct tnode *) n;
+
+ check_tnode(tn);
+
+ if(tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) {
+ pos=tn->pos + tn->bits;
+ n = tnode_get_child(tn, tkey_extract_bits(key, tn->pos, tn->bits));
+ }
+ else
+ break;
+ }
+ /* Case we have found a leaf. Compare prefixes */
+
+ if (n != NULL && IS_LEAF(n) && tkey_equals(key, n->key)) {
+ struct leaf *l = (struct leaf *) n;
+ return l;
+ }
+ return NULL;
+}
+
+static struct node *trie_rebalance(struct trie *t, struct tnode *tn)
+{
+ int i = 0;
+ int wasfull;
+ t_key cindex, key;
+ struct tnode *tp = NULL;
+
+ if(!tn)
+ BUG();
+
+ key = tn->key;
+ i = 0;
+
+ while (tn != NULL && NODE_PARENT(tn) != NULL) {
+
+ if( i > 10 ) {
+ printk("Rebalance tn=%p \n", tn);
+ if(tn) printk("tn->parent=%p \n", NODE_PARENT(tn));
+
+ printk("Rebalance tp=%p \n", tp);
+ if(tp) printk("tp->parent=%p \n", NODE_PARENT(tp));
+ }
+
+ if( i > 12 ) BUG();
+ i++;
+
+ tp = NODE_PARENT(tn);
+ cindex = tkey_extract_bits(key, tp->pos, tp->bits);
+ wasfull = tnode_full(tp, tnode_get_child(tp, cindex));
+ tn = (struct tnode *) resize (t, (struct tnode *)tn);
+ tnode_put_child_reorg((struct tnode *)tp, cindex,(struct node*)tn, wasfull);
+
+ if(!NODE_PARENT(tn))
+ break;
+
+ tn = NODE_PARENT(tn);
+ }
+ /* Handle last (top) tnode */
+ if (IS_TNODE(tn))
+ tn = (struct tnode*) resize(t, (struct tnode *)tn);
+
+ return (struct node*) tn;
+}
+
+static struct list_head *
+fib_insert_node(struct trie *t, u32 key, int plen)
+{
+ int pos, newpos;
+ struct tnode *tp = NULL, *tn = NULL;
+ struct node *n;
+ struct leaf *l;
+ int missbit;
+ struct list_head *fa_head=NULL;
+ struct leaf_info *li;
+ t_key cindex;
+
+ pos = 0;
+ n=t->trie;
+
+ /* If we point to NULL, stop. Either the tree is empty and we should
+ * just put a new leaf in if, or we have reached an empty child slot,
+ * and we should just put our new leaf in that.
+ * If we point to a T_TNODE, check if it matches our key. Note that
+ * a T_TNODE might be skipping any number of bits - its 'pos' need
+ * not be the parent's 'pos'+'bits'!
+ *
+ * If it does match the current key, get pos/bits from it, extract
+ * the index from our key, push the T_TNODE and walk the tree.
+ *
+ * If it doesn't, we have to replace it with a new T_TNODE.
+ *
+ * If we point to a T_LEAF, it might or might not have the same key
+ * as we do. If it does, just change the value, update the T_LEAF's
+ * value, and return it.
+ * If it doesn't, we need to replace it with a T_TNODE.
+ */
+
+ while (n != NULL && NODE_TYPE(n) == T_TNODE) {
+ tn = (struct tnode *) n;
+
+ check_tnode(tn);
+
+ if(tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) {
+ tp = tn;
+ pos=tn->pos + tn->bits;
+ n = tnode_get_child(tn, tkey_extract_bits(key, tn->pos, tn->bits));
+
+ if(n && NODE_PARENT(n) != tn) {
+ printk("BUG tn=%p, n->parent=%p\n", tn, NODE_PARENT(n));
+ BUG();
+ }
+ }
+ else
+ break;
+ }
+
+ /*
+ * n ----> NULL, LEAF or TNODE
+ *
+ * tp is n's (parent) ----> NULL or TNODE
+ */
+
+ if(tp && IS_LEAF(tp))
+ BUG();
+
+ t->revision++;
+
+ /* Case 1: n is a leaf. Compare prefixes */
+
+ if (n != NULL && IS_LEAF(n) && tkey_equals(key, n->key)) {
+ struct leaf *l = ( struct leaf *) n;
+
+ li = leaf_info_new(plen);
+
+ if(! li)
+ BUG();
+
+ fa_head = &li->falh;
+ insert_leaf_info(&l->list, li);
+ goto done;
+ }
+ t->size++;
+ l = leaf_new();
+
+ if(! l)
+ BUG();
+
+ l->key = key;
+ li = leaf_info_new(plen);
+
+ if(! li)
+ BUG();
+
+ fa_head = &li->falh;
+ insert_leaf_info(&l->list, li);
+
+ /* Case 2: n is NULL, and will just insert a new leaf */
+ if (t->trie && n == NULL) {
+
+ NODE_SET_PARENT(l, tp);
+
+ if (!tp)
+ BUG();
+
+ else {
+ cindex = tkey_extract_bits(key, tp->pos, tp->bits);
+ put_child(t, (struct tnode *)tp, cindex, (struct node *)l);
+ }
+ }
+ /* Case 3: n is a LEAF or a TNODE and the key doesn't match. */
+ else {
+ /*
+ * Add a new tnode here
+ * first tnode need some special handling
+ */
+
+ if (tp)
+ pos=tp->pos+tp->bits;
+ else
+ pos=0;
+ if(n) {
+ newpos = tkey_mismatch(key, pos, n->key);
+ tn = tnode_new(n->key, newpos, 1);
+ }
+ else {
+ newpos = 0;
+ tn = tnode_new(key, newpos, 1); /* First tnode */
+ }
+ if(!tn)
+ trie_bug("tnode_pfx_new failed");
+
+ NODE_SET_PARENT(tn, tp);
+
+ missbit=tkey_extract_bits(key, newpos, 1);
+ put_child(t, tn, missbit, (struct node *)l);
+ put_child(t, tn, 1-missbit, n);
+
+ if(tp) {
+ cindex = tkey_extract_bits(key, tp->pos, tp->bits);
+ put_child(t, (struct tnode *)tp, cindex, (struct node *)tn);
+ }
+ else {
+ t->trie = (struct node*) tn; /* First tnode */
+ tp = tn;
+ }
+ }
+ if(tp && tp->pos+tp->bits > 32) {
+ printk("ERROR tp=%p pos=%d, bits=%d, key=%0x plen=%d\n",
+ tp, tp->pos, tp->bits, key, plen);
+ }
+ /* Rebalance the trie */
+ t->trie = trie_rebalance(t, tp);
+done:;
+ return fa_head;
+}
+
+static int
+fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
+ struct nlmsghdr *nlhdr, struct netlink_skb_parms *req)
+{
+ struct trie *t = (struct trie *) tb->tb_data;
+ struct fib_alias *fa, *new_fa;
+ struct list_head *fa_head=NULL;
+ struct fib_info *fi;
+ int plen = r->rtm_dst_len;
+ int type = r->rtm_type;
+ u8 tos = r->rtm_tos;
+ u32 key, mask;
+ int err;
+ struct leaf *l;
+
+ if (plen > 32)
+ return -EINVAL;
+
+ key = 0;
+ if (rta->rta_dst)
+ memcpy(&key, rta->rta_dst, 4);
+
+ key = ntohl(key);
+
+ if(trie_debug)
+ printk("Insert table=%d %08x/%d\n", tb->tb_id, key, plen);
+
+ mask = ntohl( inet_make_mask(plen) );
+
+ if(key & ~mask)
+ return -EINVAL;
+
+ key = key & mask;
+
+ if ((fi = fib_create_info(r, rta, nlhdr, &err)) == NULL)
+ goto err;
+
+ l = fib_find_node(t, key);
+ fa = NULL;
+
+ if(l) {
+ fa_head = get_fa_head(l, plen);
+ fa = fib_find_alias(fa_head, tos, fi->fib_priority);
+ }
+
+ /* Now fa, if non-NULL, points to the first fib alias
+ * with the same keys [prefix,tos,priority], if such key already
+ * exists or to the node before which we will insert new one.
+ *
+ * If fa is NULL, we will need to allocate a new one and
+ * insert to the head of f.
+ *
+ * If f is NULL, no fib node matched the destination key
+ * and we need to allocate a new one of those as well.
+ */
+
+ if (fa &&
+ fa->fa_info->fib_priority == fi->fib_priority) {
+ struct fib_alias *fa_orig;
+
+ err = -EEXIST;
+ if (nlhdr->nlmsg_flags & NLM_F_EXCL)
+ goto out;
+
+ if (nlhdr->nlmsg_flags & NLM_F_REPLACE) {
+ struct fib_info *fi_drop;
+ u8 state;
+
+ write_lock_bh(&fib_lock);
+
+ fi_drop = fa->fa_info;
+ fa->fa_info = fi;
+ fa->fa_type = type;
+ fa->fa_scope = r->rtm_scope;
+ state = fa->fa_state;
+ fa->fa_state &= ~FA_S_ACCESSED;
+
+ write_unlock_bh(&fib_lock);
+
+ fib_release_info(fi_drop);
+ if (state & FA_S_ACCESSED)
+ rt_cache_flush(-1);
+
+ goto succeeded;
+ }
+ /* Error if we find a perfect match which
+ * uses the same scope, type, and nexthop
+ * information.
+ */
+ fa_orig = fa;
+ list_for_each_entry(fa, fa_orig->fa_list.prev, fa_list) {
+ if (fa->fa_tos != tos)
+ break;
+ if (fa->fa_info->fib_priority != fi->fib_priority)
+ break;
+ if (fa->fa_type == type &&
+ fa->fa_scope == r->rtm_scope &&
+ fa->fa_info == fi) {
+ goto out;
+ }
+ }
+ if (!(nlhdr->nlmsg_flags & NLM_F_APPEND))
+ fa = fa_orig;
+ }
+ err = -ENOENT;
+ if (!(nlhdr->nlmsg_flags&NLM_F_CREATE))
+ goto out;
+
+ err = -ENOBUFS;
+ new_fa = kmem_cache_alloc(fn_alias_kmem, SLAB_KERNEL);
+ if (new_fa == NULL)
+ goto out;
+
+ new_fa->fa_info = fi;
+ new_fa->fa_tos = tos;
+ new_fa->fa_type = type;
+ new_fa->fa_scope = r->rtm_scope;
+ new_fa->fa_state = 0;
+#if 0
+ new_fa->dst = NULL;
+#endif
+ /*
+ * Insert new entry to the list.
+ */
+
+ if(!fa_head)
+ fa_head = fib_insert_node(t, key, plen);
+
+ write_lock_bh(&fib_lock);
+
+ list_add_tail(&new_fa->fa_list,
+ (fa ? &fa->fa_list : fa_head));
+
+ write_unlock_bh(&fib_lock);
+
+ rt_cache_flush(-1);
+ rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, tb->tb_id, nlhdr, req);
+succeeded:
+ return 0;
+out:
+ fib_release_info(fi);
+err:;
+ return err;
+}
+
+static inline int check_leaf(struct trie *t, struct leaf *l, t_key key, int *plen, const struct flowi *flp,
+ struct fib_result *res, int *err)
+{
+ int i;
+ t_key mask;
+ struct leaf_info *li;
+ struct hlist_head *hhead = &l->list;
+ struct hlist_node *node;
+
+ hlist_for_each_entry(li, node, hhead, hlist) {
+
+ i = li->plen;
+ mask = ntohl(inet_make_mask(i));
+ if (l->key != (key & mask))
+ continue;
+
+ if (((*err) = fib_semantic_match(&li->falh, flp, res, l->key, mask, i)) == 0) {
+ *plen = i;
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+ t->stats.semantic_match_passed++;
+#endif
+ return 1;
+ }
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+ t->stats.semantic_match_miss++;
+#endif
+ }
+ return 0;
+}
+
+static int
+fn_trie_lookup(struct fib_table *tb, const struct flowi *flp, struct fib_result *res)
+{
+ struct trie *t = (struct trie *) tb->tb_data;
+ int plen, ret = 0;
+ struct node *n;
+ struct tnode *pn;
+ int pos, bits;
+ t_key key=ntohl(flp->fl4_dst);
+ int chopped_off;
+ t_key cindex = 0;
+ int current_prefix_length = KEYLENGTH;
+ n = t->trie;
+
+ read_lock(&fib_lock);
+ if(!n)
+ goto failed;
+
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+ t->stats.gets++;
+#endif
+
+ /* Just a leaf? */
+ if (IS_LEAF(n)) {
+ if( check_leaf(t, (struct leaf *)n, key, &plen, flp, res, &ret) )
+ goto found;
+ goto failed;
+ }
+ pn = (struct tnode *) n;
+ chopped_off = 0;
+
+ while (pn) {
+
+ pos = pn->pos;
+ bits = pn->bits;
+
+ if(!chopped_off)
+ cindex = tkey_extract_bits(MASK_PFX(key, current_prefix_length), pos, bits);
+
+ n = tnode_get_child(pn, cindex);
+
+ if (n == NULL) {
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+ t->stats.null_node_hit++;
+#endif
+ goto backtrace;
+ }
+
+ if (IS_TNODE(n)) {
+#define HL_OPTIMIZE
+#ifdef HL_OPTIMIZE
+ struct tnode *cn = (struct tnode *)n;
+ t_key node_prefix, key_prefix, pref_mismatch;
+ int mp;
+
+ /*
+ * It's a tnode, and we can do some extra checks here if we
+ * like, to avoid descending into a dead-end branch.
+ * This tnode is in the parent's child array at index
+ * key[p_pos..p_pos+p_bits] but potentially with some bits
+ * chopped off, so in reality the index may be just a
+ * subprefix, padded with zero at the end.
+ * We can also take a look at any skipped bits in this
+ * tnode - everything up to p_pos is supposed to be ok,
+ * and the non-chopped bits of the index (se previous
+ * paragraph) are also guaranteed ok, but the rest is
+ * considered unknown.
+ *
+ * The skipped bits are key[pos+bits..cn->pos].
+ */
+
+ /* If current_prefix_length < pos+bits, we are already doing
+ * actual prefix matching, which means everything from
+ * pos+(bits-chopped_off) onward must be zero along some
+ * branch of this subtree - otherwise there is *no* valid
+ * prefix present. Here we can only check the skipped
+ * bits. Remember, since we have already indexed into the
+ * parent's child array, we know that the bits we chopped of
+ * *are* zero.
+ */
+
+ /* NOTA BENE: CHECKING ONLY SKIPPED BITS FOR THE NEW NODE HERE */
+
+ if (current_prefix_length < pos+bits) {
+ if (tkey_extract_bits(cn->key, current_prefix_length,
+ cn->pos - current_prefix_length) != 0 ||
+ !(cn->child[0]))
+ goto backtrace;
+ }
+
+ /*
+ * If chopped_off=0, the index is fully validated and we
+ * only need to look at the skipped bits for this, the new,
+ * tnode. What we actually want to do is to find out if
+ * these skipped bits match our key perfectly, or if we will
+ * have to count on finding a matching prefix further down,
+ * because if we do, we would like to have some way of
+ * verifying the existence of such a prefix at this point.
+ */
+
+ /* The only thing we can do at this point is to verify that
+ * any such matching prefix can indeed be a prefix to our
+ * key, and if the bits in the node we are inspecting that
+ * do not match our key are not ZERO, this cannot be true.
+ * Thus, find out where there is a mismatch (before cn->pos)
+ * and verify that all the mismatching bits are zero in the
+ * new tnode's key.
+ */
+
+ /* Note: We aren't very concerned about the piece of the key
+ * that precede pn->pos+pn->bits, since these have already been
+ * checked. The bits after cn->pos aren't checked since these are
+ * by definition "unknown" at this point. Thus, what we want to
+ * see is if we are about to enter the "prefix matching" state,
+ * and in that case verify that the skipped bits that will prevail
+ * throughout this subtree are zero, as they have to be if we are
+ * to find a matching prefix.
+ */
+
+ node_prefix = MASK_PFX(cn->key, cn->pos);
+ key_prefix = MASK_PFX(key, cn->pos);
+ pref_mismatch = key_prefix^node_prefix;
+ mp = 0;
+
+ /* In short: If skipped bits in this node do not match the search
+ * key, enter the "prefix matching" state.directly.
+ */
+ if (pref_mismatch) {
+ while (!(pref_mismatch & (1<<(KEYLENGTH-1)))) {
+ mp++;
+ pref_mismatch = pref_mismatch <<1;
+ }
+ key_prefix = tkey_extract_bits(cn->key, mp, cn->pos-mp);
+
+ if (key_prefix != 0)
+ goto backtrace;
+
+ if (current_prefix_length >= cn->pos)
+ current_prefix_length=mp;
+ }
+#endif
+ pn = (struct tnode *)n; /* Descend */
+ chopped_off = 0;
+ continue;
+ }
+ if (IS_LEAF(n)) {
+ if( check_leaf(t, (struct leaf *)n, key, &plen, flp, res, &ret))
+ goto found;
+ }
+backtrace:
+ chopped_off++;
+
+ /* As zero don't change the child key (cindex) */
+ while ((chopped_off <= pn->bits) && !(cindex & (1<<(chopped_off-1)))) {
+ chopped_off++;
+ }
+
+ /* Decrease current_... with bits chopped off */
+ if (current_prefix_length > pn->pos + pn->bits - chopped_off)
+ current_prefix_length = pn->pos + pn->bits - chopped_off;
+
+ /*
+ * Either we do the actual chop off according or if we have
+ * chopped off all bits in this tnode walk up to our parent.
+ */
+
+ if(chopped_off <= pn->bits)
+ cindex &= ~(1 << (chopped_off-1));
+ else {
+ if( NODE_PARENT(pn) == NULL)
+ goto failed;
+
+ /* Get Child's index */
+ cindex = tkey_extract_bits(pn->key, NODE_PARENT(pn)->pos, NODE_PARENT(pn)->bits);
+ pn = NODE_PARENT(pn);
+ chopped_off = 0;
+
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+ t->stats.backtrack++;
+#endif
+ goto backtrace;
+ }
+ }
+failed:
+ ret = 1;
+found:
+ read_unlock(&fib_lock);
+ return ret;
+}
+
+static int trie_leaf_remove(struct trie *t, t_key key)
+{
+ t_key cindex;
+ struct tnode *tp = NULL;
+ struct node *n = t->trie;
+ struct leaf *l;
+
+ if(trie_debug)
+ printk("entering trie_leaf_remove(%p)\n", n);
+
+ /* Note that in the case skipped bits, those bits are *not* checked!
+ * When we finish this, we will have NULL or a T_LEAF, and the
+ * T_LEAF may or may not match our key.
+ */
+
+ while (n != NULL && IS_TNODE(n)) {
+ struct tnode *tn = (struct tnode *) n;
+ check_tnode(tn);
+ n = tnode_get_child(tn ,tkey_extract_bits(key, tn->pos, tn->bits));
+
+ if(n && NODE_PARENT(n) != tn) {
+ printk("BUG tn=%p, n->parent=%p\n", tn, NODE_PARENT(n));
+ BUG();
+ }
+ }
+ l = (struct leaf *) n;
+
+ if(!n || !tkey_equals(l->key, key))
+ return 0;
+
+ /*
+ * Key found.
+ * Remove the leaf and rebalance the tree
+ */
+
+ t->revision++;
+ t->size--;
+
+ tp = NODE_PARENT(n);
+ tnode_free((struct tnode *) n);
+
+ if(tp) {
+ cindex = tkey_extract_bits(key, tp->pos, tp->bits);
+ put_child(t, (struct tnode *)tp, cindex, NULL);
+ t->trie = trie_rebalance(t, tp);
+ }
+ else
+ t->trie = NULL;
+
+ return 1;
+}
+
+static int
+fn_trie_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
+ struct nlmsghdr *nlhdr, struct netlink_skb_parms *req)
+{
+ struct trie *t = (struct trie *) tb->tb_data;
+ u32 key, mask;
+ int plen = r->rtm_dst_len;
+ u8 tos = r->rtm_tos;
+ struct fib_alias *fa, *fa_to_delete;
+ struct list_head *fa_head;
+ struct leaf *l;
+
+ if (plen > 32)
+ return -EINVAL;
+
+ key = 0;
+ if (rta->rta_dst)
+ memcpy(&key, rta->rta_dst, 4);
+
+ key = ntohl(key);
+ mask = ntohl( inet_make_mask(plen) );
+
+ if(key & ~mask)
+ return -EINVAL;
+
+ key = key & mask;
+ l = fib_find_node(t, key);
+
+ if(!l)
+ return -ESRCH;
+
+ fa_head = get_fa_head(l, plen);
+ fa = fib_find_alias(fa_head, tos, 0);
+
+ if (!fa)
+ return -ESRCH;
+
+ if (trie_debug)
+ printk("Deleting %08x/%d tos=%d t=%p\n", key, plen, tos, t);
+
+ fa_to_delete = NULL;
+ fa_head = fa->fa_list.prev;
+ list_for_each_entry(fa, fa_head, fa_list) {
+ struct fib_info *fi = fa->fa_info;
+
+ if (fa->fa_tos != tos)
+ break;
+
+ if ((!r->rtm_type ||
+ fa->fa_type == r->rtm_type) &&
+ (r->rtm_scope == RT_SCOPE_NOWHERE ||
+ fa->fa_scope == r->rtm_scope) &&
+ (!r->rtm_protocol ||
+ fi->fib_protocol == r->rtm_protocol) &&
+ fib_nh_match(r, nlhdr, rta, fi) == 0) {
+ fa_to_delete = fa;
+ break;
+ }
+ }
+
+ if (fa_to_delete) {
+ int kill_li = 0;
+ struct leaf_info *li;
+
+ fa = fa_to_delete;
+ rtmsg_fib(RTM_DELROUTE, htonl(key), fa, plen, tb->tb_id, nlhdr, req);
+
+ l = fib_find_node(t, key);
+ li = find_leaf_info(&l->list, plen);
+
+ write_lock_bh(&fib_lock);
+
+ list_del(&fa->fa_list);
+
+ if(list_empty(fa_head)) {
+ hlist_del(&li->hlist);
+ kill_li = 1;
+ }
+ write_unlock_bh(&fib_lock);
+
+ if(kill_li)
+ free_leaf_info(li);
+
+ if(hlist_empty(&l->list))
+ trie_leaf_remove(t, key);
+
+ if (fa->fa_state & FA_S_ACCESSED)
+ rt_cache_flush(-1);
+
+ fn_free_alias(fa);
+ return 0;
+ }
+ return -ESRCH;
+}
+
+static int trie_flush_list(struct trie *t, struct list_head *head)
+{
+ struct fib_alias *fa, *fa_node;
+ int found = 0;
+
+ list_for_each_entry_safe(fa, fa_node, head, fa_list) {
+ struct fib_info *fi = fa->fa_info;
+
+ if (fi && (fi->fib_flags&RTNH_F_DEAD)) {
+
+ write_lock_bh(&fib_lock);
+ list_del(&fa->fa_list);
+ write_unlock_bh(&fib_lock);
+
+ fn_free_alias(fa);
+ found++;
+ }
+ }
+ return found;
+}
+
+static int trie_flush_leaf(struct trie *t, struct leaf *l)
+{
+ int found = 0;
+ struct hlist_head *lih = &l->list;
+ struct hlist_node *node, *tmp;
+ struct leaf_info *li = NULL;
+
+ hlist_for_each_entry_safe(li, node, tmp, lih, hlist) {
+
+ found += trie_flush_list(t, &li->falh);
+
+ if (list_empty(&li->falh)) {
+
+ write_lock_bh(&fib_lock);
+ hlist_del(&li->hlist);
+ write_unlock_bh(&fib_lock);
+
+ free_leaf_info(li);
+ }
+ }
+ return found;
+}
+
+static struct leaf *nextleaf(struct trie *t, struct leaf *thisleaf)
+{
+ struct node *c = (struct node *) thisleaf;
+ struct tnode *p;
+ int idx;
+
+ if(c == NULL) {
+ if(t->trie == NULL)
+ return NULL;
+
+ if (IS_LEAF(t->trie)) /* trie w. just a leaf */
+ return (struct leaf *) t->trie;
+
+ p = (struct tnode*) t->trie; /* Start */
+ }
+ else
+ p = (struct tnode *) NODE_PARENT(c);
+ while (p) {
+ int pos, last;
+
+ /* Find the next child of the parent */
+ if(c)
+ pos = 1 + tkey_extract_bits(c->key, p->pos, p->bits);
+ else
+ pos = 0;
+
+ last = 1 << p->bits;
+ for(idx = pos; idx < last ; idx++) {
+ if( p->child[idx]) {
+
+ /* Decend if tnode */
+
+ while (IS_TNODE(p->child[idx])) {
+ p = (struct tnode*) p->child[idx];
+ idx = 0;
+
+ /* Rightmost non-NULL branch */
+ if( p && IS_TNODE(p) )
+ while ( p->child[idx] == NULL && idx < (1 << p->bits) ) idx++;
+
+ /* Done with this tnode? */
+ if( idx >= (1 << p->bits) || p->child[idx] == NULL )
+ goto up;
+ }
+ return (struct leaf*) p->child[idx];
+ }
+ }
+up:
+ /* No more children go up one step */
+ c = (struct node*) p;
+ p = (struct tnode *) NODE_PARENT(p);
+ }
+ return NULL; /* Ready. Root of trie */
+}
+
+static int fn_trie_flush(struct fib_table *tb)
+{
+ struct trie *t = (struct trie *) tb->tb_data;
+ struct leaf *ll = NULL, *l = NULL;
+ int found = 0, h;
+
+ t->revision++;
+
+ for (h=0; (l = nextleaf(t, l)) != NULL; h++) {
+ found += trie_flush_leaf(t, l);
+
+ if (ll && hlist_empty(&ll->list))
+ trie_leaf_remove(t, ll->key);
+ ll = l;
+ }
+
+ if (ll && hlist_empty(&ll->list))
+ trie_leaf_remove(t, ll->key);
+
+ if(trie_debug)
+ printk("trie_flush found=%d\n", found);
+ return found;
+}
+
+static int trie_last_dflt=-1;
+
+static void
+fn_trie_select_default(struct fib_table *tb, const struct flowi *flp, struct fib_result *res)
+{
+ struct trie *t = (struct trie *) tb->tb_data;
+ int order, last_idx;
+ struct fib_info *fi = NULL;
+ struct fib_info *last_resort;
+ struct fib_alias *fa = NULL;
+ struct list_head *fa_head;
+ struct leaf *l;
+
+ last_idx = -1;
+ last_resort = NULL;
+ order = -1;
+
+ read_lock(&fib_lock);
+
+ l = fib_find_node(t, 0);
+ if(!l)
+ goto out;
+
+ fa_head = get_fa_head(l, 0);
+ if(!fa_head)
+ goto out;
+
+ if (list_empty(fa_head))
+ goto out;
+
+ list_for_each_entry(fa, fa_head, fa_list) {
+ struct fib_info *next_fi = fa->fa_info;
+
+ if (fa->fa_scope != res->scope ||
+ fa->fa_type != RTN_UNICAST)
+ continue;
+
+ if (next_fi->fib_priority > res->fi->fib_priority)
+ break;
+ if (!next_fi->fib_nh[0].nh_gw ||
+ next_fi->fib_nh[0].nh_scope != RT_SCOPE_LINK)
+ continue;
+ fa->fa_state |= FA_S_ACCESSED;
+
+ if (fi == NULL) {
+ if (next_fi != res->fi)
+ break;
+ } else if (!fib_detect_death(fi, order, &last_resort,
+ &last_idx, &trie_last_dflt)) {
+ if (res->fi)
+ fib_info_put(res->fi);
+ res->fi = fi;
+ atomic_inc(&fi->fib_clntref);
+ trie_last_dflt = order;
+ goto out;
+ }
+ fi = next_fi;
+ order++;
+ }
+ if (order <= 0 || fi == NULL) {
+ trie_last_dflt = -1;
+ goto out;
+ }
+
+ if (!fib_detect_death(fi, order, &last_resort, &last_idx, &trie_last_dflt)) {
+ if (res->fi)
+ fib_info_put(res->fi);
+ res->fi = fi;
+ atomic_inc(&fi->fib_clntref);
+ trie_last_dflt = order;
+ goto out;
+ }
+ if (last_idx >= 0) {
+ if (res->fi)
+ fib_info_put(res->fi);
+ res->fi = last_resort;
+ if (last_resort)
+ atomic_inc(&last_resort->fib_clntref);
+ }
+ trie_last_dflt = last_idx;
+ out:;
+ read_unlock(&fib_lock);
+}
+
+static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah, struct fib_table *tb,
+ struct sk_buff *skb, struct netlink_callback *cb)
+{
+ int i, s_i;
+ struct fib_alias *fa;
+
+ u32 xkey=htonl(key);
+
+ s_i=cb->args[3];
+ i = 0;
+
+ list_for_each_entry(fa, fah, fa_list) {
+ if (i < s_i) {
+ i++;
+ continue;
+ }
+ if (fa->fa_info->fib_nh == NULL) {
+ printk("Trie error _fib_nh=NULL in fa[%d] k=%08x plen=%d\n", i, key, plen);
+ i++;
+ continue;
+ }
+ if (fa->fa_info == NULL) {
+ printk("Trie error fa_info=NULL in fa[%d] k=%08x plen=%d\n", i, key, plen);
+ i++;
+ continue;
+ }
+
+ if (fib_dump_info(skb, NETLINK_CB(cb->skb).pid,
+ cb->nlh->nlmsg_seq,
+ RTM_NEWROUTE,
+ tb->tb_id,
+ fa->fa_type,
+ fa->fa_scope,
+ &xkey,
+ plen,
+ fa->fa_tos,
+ fa->fa_info, 0) < 0) {
+ cb->args[3] = i;
+ return -1;
+ }
+ i++;
+ }
+ cb->args[3]=i;
+ return skb->len;
+}
+
+static int fn_trie_dump_plen(struct trie *t, int plen, struct fib_table *tb, struct sk_buff *skb,
+ struct netlink_callback *cb)
+{
+ int h, s_h;
+ struct list_head *fa_head;
+ struct leaf *l = NULL;
+ s_h=cb->args[2];
+
+ for (h=0; (l = nextleaf(t, l)) != NULL; h++) {
+
+ if (h < s_h)
+ continue;
+ if (h > s_h)
+ memset(&cb->args[3], 0,
+ sizeof(cb->args) - 3*sizeof(cb->args[0]));
+
+ fa_head = get_fa_head(l, plen);
+
+ if(!fa_head)
+ continue;
+
+ if(list_empty(fa_head))
+ continue;
+
+ if (fn_trie_dump_fa(l->key, plen, fa_head, tb, skb, cb)<0) {
+ cb->args[2]=h;
+ return -1;
+ }
+ }
+ cb->args[2]=h;
+ return skb->len;
+}
+
+static int fn_trie_dump(struct fib_table *tb, struct sk_buff *skb, struct netlink_callback *cb)
+{
+ int m, s_m;
+ struct trie *t = (struct trie *) tb->tb_data;
+
+ s_m = cb->args[1];
+
+ read_lock(&fib_lock);
+ for (m=0; m<=32; m++) {
+
+ if (m < s_m)
+ continue;
+ if (m > s_m)
+ memset(&cb->args[2], 0,
+ sizeof(cb->args) - 2*sizeof(cb->args[0]));
+
+ if (fn_trie_dump_plen(t, 32-m, tb, skb, cb)<0) {
+ cb->args[1] = m;
+ goto out;
+ }
+ }
+ read_unlock(&fib_lock);
+ cb->args[1] = m;
+ return skb->len;
+ out:
+ read_unlock(&fib_lock);
+ return -1;
+}
+
+/* Fix more generic FIB names for init later */
+
+#ifdef CONFIG_IP_MULTIPLE_TABLES
+struct fib_table * fib_hash_init(int id)
+#else
+struct fib_table * __init fib_hash_init(int id)
+#endif
+{
+ struct fib_table *tb;
+ struct trie *t;
+
+ if (fn_alias_kmem == NULL)
+ fn_alias_kmem = kmem_cache_create("ip_fib_alias",
+ sizeof(struct fib_alias),
+ 0, SLAB_HWCACHE_ALIGN,
+ NULL, NULL);
+
+ tb = kmalloc(sizeof(struct fib_table) + sizeof(struct trie),
+ GFP_KERNEL);
+ if (tb == NULL)
+ return NULL;
+
+ tb->tb_id = id;
+ tb->tb_lookup = fn_trie_lookup;
+ tb->tb_insert = fn_trie_insert;
+ tb->tb_delete = fn_trie_delete;
+ tb->tb_flush = fn_trie_flush;
+ tb->tb_select_default = fn_trie_select_default;
+ tb->tb_dump = fn_trie_dump;
+ memset(tb->tb_data, 0, sizeof(struct trie));
+
+ t = (struct trie *) tb->tb_data;
+
+ trie_init(t);
+
+ if (id == RT_TABLE_LOCAL)
+ trie_local=t;
+ else if (id == RT_TABLE_MAIN)
+ trie_main=t;
+
+ if (id == RT_TABLE_LOCAL)
+ printk("IPv4 FIB: Using LC-trie version %s\n", VERSION);
+
+ return tb;
+}
+
+/* Trie dump functions */
+
+static void putspace_seq(struct seq_file *seq, int n)
+{
+ while (n--) seq_printf(seq, " ");
+}
+
+static void printbin_seq(struct seq_file *seq, unsigned int v, int bits)
+{
+ while (bits--)
+ seq_printf(seq, "%s", (v & (1<<bits))?"1":"0");
+}
+
+static void printnode_seq(struct seq_file *seq, int indent, struct node *n,
+ int pend, int cindex, int bits)
+{
+ putspace_seq(seq, indent);
+ if (IS_LEAF(n))
+ seq_printf(seq, "|");
+ else
+ seq_printf(seq, "+");
+ if (bits) {
+ seq_printf(seq, "%d/", cindex);
+ printbin_seq(seq, cindex, bits);
+ seq_printf(seq, ": ");
+ }
+ else
+ seq_printf(seq, "<root>: ");
+ seq_printf(seq, "%s:%p ", IS_LEAF(n)?"Leaf":"Internal node", n);
+
+ if (IS_LEAF(n))
+ seq_printf(seq, "key=%d.%d.%d.%d\n",
+ n->key >> 24, (n->key >> 16) % 256, (n->key >> 8) % 256, n->key % 256);
+ else {
+ int plen=((struct tnode *)n)->pos;
+ t_key prf=MASK_PFX(n->key, plen);
+ seq_printf(seq, "key=%d.%d.%d.%d/%d\n",
+ prf >> 24, (prf >> 16) % 256, (prf >> 8) % 256, prf % 256, plen);
+ }
+ if (IS_LEAF(n)) {
+ struct leaf *l=(struct leaf *)n;
+ struct fib_alias *fa;
+ int i;
+ for (i=32; i>=0; i--)
+ if(find_leaf_info(&l->list, i)) {
+
+ struct list_head *fa_head = get_fa_head(l, i);
+
+ if(!fa_head)
+ continue;
+
+ if(list_empty(fa_head))
+ continue;
+
+ putspace_seq(seq, indent+2);
+ seq_printf(seq, "{/%d...dumping}\n", i);
+
+
+ list_for_each_entry(fa, fa_head, fa_list) {
+ putspace_seq(seq, indent+2);
+ if (fa->fa_info->fib_nh == NULL) {
+ seq_printf(seq, "Error _fib_nh=NULL\n");
+ continue;
+ }
+ if (fa->fa_info == NULL) {
+ seq_printf(seq, "Error fa_info=NULL\n");
+ continue;
+ }
+
+ seq_printf(seq, "{type=%d scope=%d TOS=%d}\n",
+ fa->fa_type,
+ fa->fa_scope,
+ fa->fa_tos);
+ }
+ }
+ }
+ else if (IS_TNODE(n)) {
+ struct tnode *tn=(struct tnode *)n;
+ putspace_seq(seq, indent); seq_printf(seq, "| ");
+ seq_printf(seq, "{key prefix=%08x/", tn->key&TKEY_GET_MASK(0, tn->pos));
+ printbin_seq(seq, tkey_extract_bits(tn->key, 0, tn->pos), tn->pos);
+ seq_printf(seq, "}\n");
+ putspace_seq(seq, indent); seq_printf(seq, "| ");
+ seq_printf(seq, "{pos=%d", tn->pos);
+ seq_printf(seq, " (skip=%d bits)", tn->pos - pend);
+ seq_printf(seq, " bits=%d (%u children)}\n", tn->bits, (1 << tn->bits));
+ putspace_seq(seq, indent); seq_printf(seq, "| ");
+ seq_printf(seq, "{empty=%d full=%d}\n", tn->empty_children, tn->full_children);
+ }
+}
+
+static void trie_dump_seq(struct seq_file *seq, struct trie *t)
+{
+ struct node *n=t->trie;
+ int cindex=0;
+ int indent=1;
+ int pend=0;
+ int depth = 0;
+
+ read_lock(&fib_lock);
+
+ seq_printf(seq, "------ trie_dump of t=%p ------\n", t);
+ if (n) {
+ printnode_seq(seq, indent, n, pend, cindex, 0);
+ if (IS_TNODE(n)) {
+ struct tnode *tn=(struct tnode *)n;
+ pend = tn->pos+tn->bits;
+ putspace_seq(seq, indent); seq_printf(seq, "\\--\n");
+ indent += 3;
+ depth++;
+
+ while (tn && cindex < (1 << tn->bits)) {
+ if (tn->child[cindex]) {
+
+ /* Got a child */
+
+ printnode_seq(seq, indent, tn->child[cindex], pend, cindex, tn->bits);
+ if (IS_LEAF(tn->child[cindex])) {
+ cindex++;
+
+ }
+ else {
+ /*
+ * New tnode. Decend one level
+ */
+
+ depth++;
+ n=tn->child[cindex];
+ tn=(struct tnode *)n;
+ pend=tn->pos+tn->bits;
+ putspace_seq(seq, indent); seq_printf(seq, "\\--\n");
+ indent+=3;
+ cindex=0;
+ }
+ }
+ else
+ cindex++;
+
+ /*
+ * Test if we are done
+ */
+
+ while (cindex >= (1 << tn->bits)) {
+
+ /*
+ * Move upwards and test for root
+ * pop off all traversed nodes
+ */
+
+ if (NODE_PARENT(tn) == NULL) {
+ tn = NULL;
+ n = NULL;
+ break;
+ }
+ else {
+ cindex = tkey_extract_bits(tn->key, NODE_PARENT(tn)->pos, NODE_PARENT(tn)->bits);
+ tn = NODE_PARENT(tn);
+ cindex++;
+ n=(struct node *)tn;
+ pend=tn->pos+tn->bits;
+ indent-=3;
+ depth--;
+ }
+ }
+ }
+ }
+ else n = NULL;
+ }
+ else seq_printf(seq, "------ trie is empty\n");
+
+ read_unlock(&fib_lock);
+}
+
+static struct trie_stat *trie_stat_new(void)
+{
+ struct trie_stat *s = kmalloc(sizeof(struct trie_stat), GFP_KERNEL);
+ int i;
+
+ if(s) {
+ s->totdepth = 0;
+ s->maxdepth = 0;
+ s->tnodes = 0;
+ s->leaves = 0;
+ s->nullpointers = 0;
+
+ for(i=0; i< MAX_CHILDS; i++)
+ s->nodesizes[i] = 0;
+ }
+ return s;
+}
+
+static struct trie_stat *trie_collect_stats(struct trie *t)
+{
+ struct node *n=t->trie;
+ struct trie_stat *s = trie_stat_new();
+ int cindex = 0;
+ int indent = 1;
+ int pend = 0;
+ int depth = 0;
+
+ read_lock(&fib_lock);
+
+ if (s) {
+ if (n) {
+ if (IS_TNODE(n)) {
+ struct tnode *tn = (struct tnode *)n;
+ pend=tn->pos+tn->bits;
+ indent += 3;
+ s->nodesizes[tn->bits]++;
+ depth++;
+
+ while (tn && cindex < (1 << tn->bits)) {
+ if (tn->child[cindex]) {
+ /* Got a child */
+
+ if (IS_LEAF(tn->child[cindex])) {
+ cindex++;
+
+ /* stats */
+ if (depth > s->maxdepth)
+ s->maxdepth = depth;
+ s->totdepth += depth;
+ s->leaves++;
+ }
+
+ else {
+ /*
+ * New tnode. Decend one level
+ */
+
+ s->tnodes++;
+ s->nodesizes[tn->bits]++;
+ depth++;
+
+ n = tn->child[cindex];
+ tn = (struct tnode *)n;
+ pend = tn->pos+tn->bits;
+
+ indent += 3;
+ cindex = 0;
+ }
+ }
+ else {
+ cindex++;
+ s->nullpointers++;
+ }
+
+ /*
+ * Test if we are done
+ */
+
+ while (cindex >= (1 << tn->bits)) {
+
+ /*
+ * Move upwards and test for root
+ * pop off all traversed nodes
+ */
+
+
+ if (NODE_PARENT(tn) == NULL) {
+ tn = NULL;
+ n = NULL;
+ break;
+ }
+ else {
+ cindex = tkey_extract_bits(tn->key, NODE_PARENT(tn)->pos, NODE_PARENT(tn)->bits);
+ tn = NODE_PARENT(tn);
+ cindex++;
+ n = (struct node *)tn;
+ pend=tn->pos+tn->bits;
+ indent -= 3;
+ depth--;
+ }
+ }
+ }
+ }
+ else n = NULL;
+ }
+ }
+
+ read_unlock(&fib_lock);
+ return s;
+}
+
+#ifdef CONFIG_PROC_FS
+
+static struct fib_alias *fib_triestat_get_first(struct seq_file *seq)
+{
+ return NULL;
+}
+
+static struct fib_alias *fib_triestat_get_next(struct seq_file *seq)
+{
+ return NULL;
+}
+
+static void *fib_triestat_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ void *v = NULL;
+
+ if (ip_fib_main_table)
+ v = *pos ? fib_triestat_get_next(seq) : SEQ_START_TOKEN;
+ return v;
+}
+
+static void *fib_triestat_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ ++*pos;
+ return v == SEQ_START_TOKEN ? fib_triestat_get_first(seq) : fib_triestat_get_next(seq);
+}
+
+static void fib_triestat_seq_stop(struct seq_file *seq, void *v)
+{
+
+}
+
+/*
+ * This outputs /proc/net/fib_triestats
+ *
+ * It always works in backward compatibility mode.
+ * The format of the file is not supposed to be changed.
+ */
+
+static void collect_and_show(struct trie *t, struct seq_file *seq)
+{
+ int bytes = 0; /* How many bytes are used, a ref is 4 bytes */
+ int i, max, pointers;
+ struct trie_stat *stat;
+ int avdepth;
+
+ stat = trie_collect_stats(t);
+
+ bytes=0;
+ seq_printf(seq, "trie=%p\n", t);
+
+ if (stat) {
+ if (stat->leaves)
+ avdepth=stat->totdepth*100 / stat->leaves;
+ else
+ avdepth=0;
+ seq_printf(seq, "Aver depth: %d.%02d\n", avdepth / 100, avdepth % 100 );
+ seq_printf(seq, "Max depth: %4d\n", stat->maxdepth);
+
+ seq_printf(seq, "Leaves: %d\n", stat->leaves);
+ bytes += sizeof(struct leaf) * stat->leaves;
+ seq_printf(seq, "Internal nodes: %d\n", stat->tnodes);
+ bytes += sizeof(struct tnode) * stat->tnodes;
+
+ max = MAX_CHILDS-1;
+
+ while (max >= 0 && stat->nodesizes[max] == 0)
+ max--;
+ pointers = 0;
+
+ for (i = 1; i <= max; i++)
+ if (stat->nodesizes[i] != 0) {
+ seq_printf(seq, " %d: %d", i, stat->nodesizes[i]);
+ pointers += (1<<i) * stat->nodesizes[i];
+ }
+ seq_printf(seq, "\n");
+ seq_printf(seq, "Pointers: %d\n", pointers);
+ bytes += sizeof(struct node *) * pointers;
+ seq_printf(seq, "Null ptrs: %d\n", stat->nullpointers);
+ seq_printf(seq, "Total size: %d kB\n", bytes / 1024);
+
+ kfree(stat);
+ }
+
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+ seq_printf(seq, "Counters:\n---------\n");
+ seq_printf(seq,"gets = %d\n", t->stats.gets);
+ seq_printf(seq,"backtracks = %d\n", t->stats.backtrack);
+ seq_printf(seq,"semantic match passed = %d\n", t->stats.semantic_match_passed);
+ seq_printf(seq,"semantic match miss = %d\n", t->stats.semantic_match_miss);
+ seq_printf(seq,"null node hit= %d\n", t->stats.null_node_hit);
+#ifdef CLEAR_STATS
+ memset(&(t->stats), 0, sizeof(t->stats));
+#endif
+#endif /* CONFIG_IP_FIB_TRIE_STATS */
+}
+
+static int fib_triestat_seq_show(struct seq_file *seq, void *v)
+{
+ char bf[128];
+
+ if (v == SEQ_START_TOKEN) {
+ seq_printf(seq, "Basic info: size of leaf: %Zd bytes, size of tnode: %Zd bytes.\n",
+ sizeof(struct leaf), sizeof(struct tnode));
+ if (trie_local)
+ collect_and_show(trie_local, seq);
+
+ if (trie_main)
+ collect_and_show(trie_main, seq);
+ }
+ else {
+ snprintf(bf, sizeof(bf),
+ "*\t%08X\t%08X", 200, 400);
+
+ seq_printf(seq, "%-127s\n", bf);
+ }
+ return 0;
+}
+
+static struct seq_operations fib_triestat_seq_ops = {
+ .start = fib_triestat_seq_start,
+ .next = fib_triestat_seq_next,
+ .stop = fib_triestat_seq_stop,
+ .show = fib_triestat_seq_show,
+};
+
+static int fib_triestat_seq_open(struct inode *inode, struct file *file)
+{
+ struct seq_file *seq;
+ int rc = -ENOMEM;
+
+ rc = seq_open(file, &fib_triestat_seq_ops);
+ if (rc)
+ goto out_kfree;
+
+ seq = file->private_data;
+out:
+ return rc;
+out_kfree:
+ goto out;
+}
+
+static struct file_operations fib_triestat_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = fib_triestat_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release_private,
+};
+
+int __init fib_stat_proc_init(void)
+{
+ if (!proc_net_fops_create("fib_triestat", S_IRUGO, &fib_triestat_seq_fops))
+ return -ENOMEM;
+ return 0;
+}
+
+void __init fib_stat_proc_exit(void)
+{
+ proc_net_remove("fib_triestat");
+}
+
+static struct fib_alias *fib_trie_get_first(struct seq_file *seq)
+{
+ return NULL;
+}
+
+static struct fib_alias *fib_trie_get_next(struct seq_file *seq)
+{
+ return NULL;
+}
+
+static void *fib_trie_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ void *v = NULL;
+
+ if (ip_fib_main_table)
+ v = *pos ? fib_trie_get_next(seq) : SEQ_START_TOKEN;
+ return v;
+}
+
+static void *fib_trie_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ ++*pos;
+ return v == SEQ_START_TOKEN ? fib_trie_get_first(seq) : fib_trie_get_next(seq);
+}
+
+static void fib_trie_seq_stop(struct seq_file *seq, void *v)
+{
+
+}
+
+/*
+ * This outputs /proc/net/fib_trie.
+ *
+ * It always works in backward compatibility mode.
+ * The format of the file is not supposed to be changed.
+ */
+
+static int fib_trie_seq_show(struct seq_file *seq, void *v)
+{
+ char bf[128];
+
+ if (v == SEQ_START_TOKEN) {
+ if (trie_local)
+ trie_dump_seq(seq, trie_local);
+
+ if (trie_main)
+ trie_dump_seq(seq, trie_main);
+ }
+
+ else {
+ snprintf(bf, sizeof(bf),
+ "*\t%08X\t%08X", 200, 400);
+ seq_printf(seq, "%-127s\n", bf);
+ }
+
+ return 0;
+}
+
+static struct seq_operations fib_trie_seq_ops = {
+ .start = fib_trie_seq_start,
+ .next = fib_trie_seq_next,
+ .stop = fib_trie_seq_stop,
+ .show = fib_trie_seq_show,
+};
+
+static int fib_trie_seq_open(struct inode *inode, struct file *file)
+{
+ struct seq_file *seq;
+ int rc = -ENOMEM;
+
+ rc = seq_open(file, &fib_trie_seq_ops);
+ if (rc)
+ goto out_kfree;
+
+ seq = file->private_data;
+out:
+ return rc;
+out_kfree:
+ goto out;
+}
+
+static struct file_operations fib_trie_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = fib_trie_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release_private,
+};
+
+int __init fib_proc_init(void)
+{
+ if (!proc_net_fops_create("fib_trie", S_IRUGO, &fib_trie_seq_fops))
+ return -ENOMEM;
+ return 0;
+}
+
+void __init fib_proc_exit(void)
+{
+ proc_net_remove("fib_trie");
+}
+
+#endif /* CONFIG_PROC_FS */
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 4e47a26..af2ec88 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -184,6 +184,7 @@ int ip_call_ra_chain(struct sk_buff *skb)
raw_rcv(last, skb2);
}
last = sk;
+ nf_reset(skb);
}
}
@@ -200,10 +201,6 @@ static inline int ip_local_deliver_finish(struct sk_buff *skb)
{
int ihl = skb->nh.iph->ihl*4;
-#ifdef CONFIG_NETFILTER_DEBUG
- nf_debug_ip_local_deliver(skb);
-#endif /*CONFIG_NETFILTER_DEBUG*/
-
__skb_pull(skb, ihl);
/* Free reference early: we don't need it any more, and it may
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 760dc82..ee07aec 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -107,10 +107,6 @@ static int ip_dev_loopback_xmit(struct sk_buff *newskb)
newskb->pkt_type = PACKET_LOOPBACK;
newskb->ip_summed = CHECKSUM_UNNECESSARY;
BUG_TRAP(newskb->dst);
-
-#ifdef CONFIG_NETFILTER_DEBUG
- nf_debug_ip_loopback_xmit(newskb);
-#endif
nf_reset(newskb);
netif_rx(newskb);
return 0;
@@ -192,10 +188,6 @@ static inline int ip_finish_output2(struct sk_buff *skb)
skb = skb2;
}
-#ifdef CONFIG_NETFILTER_DEBUG
- nf_debug_ip_finish_output2(skb);
-#endif /*CONFIG_NETFILTER_DEBUG*/
-
nf_reset(skb);
if (hh) {
@@ -415,9 +407,6 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from)
to->nf_bridge = from->nf_bridge;
nf_bridge_get(to->nf_bridge);
#endif
-#ifdef CONFIG_NETFILTER_DEBUG
- to->nf_debug = from->nf_debug;
-#endif
#endif
}
diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c
index 1a23c52..2065944 100644
--- a/net/ipv4/ipcomp.c
+++ b/net/ipv4/ipcomp.c
@@ -236,15 +236,10 @@ static struct xfrm_state *ipcomp_tunnel_create(struct xfrm_state *x)
t->props.mode = 1;
t->props.saddr.a4 = x->props.saddr.a4;
t->props.flags = x->props.flags;
-
- t->type = xfrm_get_type(IPPROTO_IPIP, t->props.family);
- if (t->type == NULL)
- goto error;
-
- if (t->type->init_state(t, NULL))
+
+ if (xfrm_init_state(t))
goto error;
- t->km.state = XFRM_STATE_VALID;
atomic_set(&t->tunnel_users, 1);
out:
return t;
@@ -422,7 +417,7 @@ static void ipcomp_destroy(struct xfrm_state *x)
kfree(ipcd);
}
-static int ipcomp_init_state(struct xfrm_state *x, void *args)
+static int ipcomp_init_state(struct xfrm_state *x)
{
int err;
struct ipcomp_data *ipcd;
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index e21c049..e4f809a 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1350,6 +1350,7 @@ int ip_mr_input(struct sk_buff *skb)
*/
read_lock(&mrt_lock);
if (mroute_socket) {
+ nf_reset(skb);
raw_rcv(mroute_socket, skb);
read_unlock(&mrt_lock);
return 0;
diff --git a/net/ipv4/ipvs/ip_vs_xmit.c b/net/ipv4/ipvs/ip_vs_xmit.c
index de21da0..a8512a3 100644
--- a/net/ipv4/ipvs/ip_vs_xmit.c
+++ b/net/ipv4/ipvs/ip_vs_xmit.c
@@ -127,7 +127,6 @@ ip_vs_dst_reset(struct ip_vs_dest *dest)
#define IP_VS_XMIT(skb, rt) \
do { \
- nf_reset_debug(skb); \
(skb)->nfcache |= NFC_IPVS_PROPERTY; \
(skb)->ip_summed = CHECKSUM_NONE; \
NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, (skb), NULL, \
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index df79f5e..fa16342 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -60,7 +60,6 @@ static DECLARE_MUTEX(arpt_mutex);
#define ASSERT_READ_LOCK(x) ARP_NF_ASSERT(down_trylock(&arpt_mutex) != 0)
#define ASSERT_WRITE_LOCK(x) ARP_NF_ASSERT(down_trylock(&arpt_mutex) != 0)
-#include <linux/netfilter_ipv4/lockhelp.h>
#include <linux/netfilter_ipv4/listhelp.h>
struct arpt_table_info {
diff --git a/net/ipv4/netfilter/ip_conntrack_amanda.c b/net/ipv4/netfilter/ip_conntrack_amanda.c
index 3dbddd0..a78a320 100644
--- a/net/ipv4/netfilter/ip_conntrack_amanda.c
+++ b/net/ipv4/netfilter/ip_conntrack_amanda.c
@@ -26,7 +26,6 @@
#include <net/checksum.h>
#include <net/udp.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
#include <linux/netfilter_ipv4/ip_conntrack_helper.h>
#include <linux/netfilter_ipv4/ip_conntrack_amanda.h>
@@ -42,7 +41,7 @@ static char *conns[] = { "DATA ", "MESG ", "INDEX " };
/* This is slow, but it's simple. --RR */
static char amanda_buffer[65536];
-static DECLARE_LOCK(amanda_buffer_lock);
+static DEFINE_SPINLOCK(amanda_buffer_lock);
unsigned int (*ip_nat_amanda_hook)(struct sk_buff **pskb,
enum ip_conntrack_info ctinfo,
@@ -76,7 +75,7 @@ static int help(struct sk_buff **pskb,
return NF_ACCEPT;
}
- LOCK_BH(&amanda_buffer_lock);
+ spin_lock_bh(&amanda_buffer_lock);
skb_copy_bits(*pskb, dataoff, amanda_buffer, (*pskb)->len - dataoff);
data = amanda_buffer;
data_limit = amanda_buffer + (*pskb)->len - dataoff;
@@ -134,7 +133,7 @@ static int help(struct sk_buff **pskb,
}
out:
- UNLOCK_BH(&amanda_buffer_lock);
+ spin_unlock_bh(&amanda_buffer_lock);
return ret;
}
diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c
index 09e8246..4b78ebe 100644
--- a/net/ipv4/netfilter/ip_conntrack_core.c
+++ b/net/ipv4/netfilter/ip_conntrack_core.c
@@ -38,10 +38,10 @@
#include <linux/percpu.h>
#include <linux/moduleparam.h>
-/* This rwlock protects the main hash table, protocol/helper/expected
+/* ip_conntrack_lock protects the main hash table, protocol/helper/expected
registrations, conntrack timers*/
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_conntrack_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_conntrack_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
#include <linux/netfilter_ipv4/ip_conntrack.h>
#include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
@@ -57,7 +57,7 @@
#define DEBUGP(format, args...)
#endif
-DECLARE_RWLOCK(ip_conntrack_lock);
+DEFINE_RWLOCK(ip_conntrack_lock);
/* ip_conntrack_standalone needs this */
atomic_t ip_conntrack_count = ATOMIC_INIT(0);
@@ -147,7 +147,7 @@ static void destroy_expect(struct ip_conntrack_expect *exp)
static void unlink_expect(struct ip_conntrack_expect *exp)
{
- MUST_BE_WRITE_LOCKED(&ip_conntrack_lock);
+ ASSERT_WRITE_LOCK(&ip_conntrack_lock);
list_del(&exp->list);
/* Logically in destroy_expect, but we hold the lock here. */
exp->master->expecting--;
@@ -157,9 +157,9 @@ static void expectation_timed_out(unsigned long ul_expect)
{
struct ip_conntrack_expect *exp = (void *)ul_expect;
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
unlink_expect(exp);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
destroy_expect(exp);
}
@@ -209,7 +209,7 @@ clean_from_lists(struct ip_conntrack *ct)
unsigned int ho, hr;
DEBUGP("clean_from_lists(%p)\n", ct);
- MUST_BE_WRITE_LOCKED(&ip_conntrack_lock);
+ ASSERT_WRITE_LOCK(&ip_conntrack_lock);
ho = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
hr = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple);
@@ -240,7 +240,7 @@ destroy_conntrack(struct nf_conntrack *nfct)
if (ip_conntrack_destroyed)
ip_conntrack_destroyed(ct);
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
/* Expectations will have been removed in clean_from_lists,
* except TFTP can create an expectation on the first packet,
* before connection is in the list, so we need to clean here,
@@ -254,7 +254,7 @@ destroy_conntrack(struct nf_conntrack *nfct)
}
CONNTRACK_STAT_INC(delete);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
if (ct->master)
ip_conntrack_put(ct->master);
@@ -268,12 +268,12 @@ static void death_by_timeout(unsigned long ul_conntrack)
{
struct ip_conntrack *ct = (void *)ul_conntrack;
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
/* Inside lock so preempt is disabled on module removal path.
* Otherwise we can get spurious warnings. */
CONNTRACK_STAT_INC(delete_list);
clean_from_lists(ct);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
ip_conntrack_put(ct);
}
@@ -282,7 +282,7 @@ conntrack_tuple_cmp(const struct ip_conntrack_tuple_hash *i,
const struct ip_conntrack_tuple *tuple,
const struct ip_conntrack *ignored_conntrack)
{
- MUST_BE_READ_LOCKED(&ip_conntrack_lock);
+ ASSERT_READ_LOCK(&ip_conntrack_lock);
return tuplehash_to_ctrack(i) != ignored_conntrack
&& ip_ct_tuple_equal(tuple, &i->tuple);
}
@@ -294,7 +294,7 @@ __ip_conntrack_find(const struct ip_conntrack_tuple *tuple,
struct ip_conntrack_tuple_hash *h;
unsigned int hash = hash_conntrack(tuple);
- MUST_BE_READ_LOCKED(&ip_conntrack_lock);
+ ASSERT_READ_LOCK(&ip_conntrack_lock);
list_for_each_entry(h, &ip_conntrack_hash[hash], list) {
if (conntrack_tuple_cmp(h, tuple, ignored_conntrack)) {
CONNTRACK_STAT_INC(found);
@@ -313,11 +313,11 @@ ip_conntrack_find_get(const struct ip_conntrack_tuple *tuple,
{
struct ip_conntrack_tuple_hash *h;
- READ_LOCK(&ip_conntrack_lock);
+ read_lock_bh(&ip_conntrack_lock);
h = __ip_conntrack_find(tuple, ignored_conntrack);
if (h)
atomic_inc(&tuplehash_to_ctrack(h)->ct_general.use);
- READ_UNLOCK(&ip_conntrack_lock);
+ read_unlock_bh(&ip_conntrack_lock);
return h;
}
@@ -352,7 +352,7 @@ __ip_conntrack_confirm(struct sk_buff **pskb)
IP_NF_ASSERT(!is_confirmed(ct));
DEBUGP("Confirming conntrack %p\n", ct);
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
/* See if there's one in the list already, including reverse:
NAT could have grabbed it without realizing, since we're
@@ -380,12 +380,12 @@ __ip_conntrack_confirm(struct sk_buff **pskb)
atomic_inc(&ct->ct_general.use);
set_bit(IPS_CONFIRMED_BIT, &ct->status);
CONNTRACK_STAT_INC(insert);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
return NF_ACCEPT;
}
CONNTRACK_STAT_INC(insert_failed);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
return NF_DROP;
}
@@ -398,9 +398,9 @@ ip_conntrack_tuple_taken(const struct ip_conntrack_tuple *tuple,
{
struct ip_conntrack_tuple_hash *h;
- READ_LOCK(&ip_conntrack_lock);
+ read_lock_bh(&ip_conntrack_lock);
h = __ip_conntrack_find(tuple, ignored_conntrack);
- READ_UNLOCK(&ip_conntrack_lock);
+ read_unlock_bh(&ip_conntrack_lock);
return h != NULL;
}
@@ -419,13 +419,13 @@ static int early_drop(struct list_head *chain)
struct ip_conntrack *ct = NULL;
int dropped = 0;
- READ_LOCK(&ip_conntrack_lock);
+ read_lock_bh(&ip_conntrack_lock);
h = LIST_FIND_B(chain, unreplied, struct ip_conntrack_tuple_hash *);
if (h) {
ct = tuplehash_to_ctrack(h);
atomic_inc(&ct->ct_general.use);
}
- READ_UNLOCK(&ip_conntrack_lock);
+ read_unlock_bh(&ip_conntrack_lock);
if (!ct)
return dropped;
@@ -508,7 +508,7 @@ init_conntrack(const struct ip_conntrack_tuple *tuple,
conntrack->timeout.data = (unsigned long)conntrack;
conntrack->timeout.function = death_by_timeout;
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
exp = find_expectation(tuple);
if (exp) {
@@ -532,7 +532,7 @@ init_conntrack(const struct ip_conntrack_tuple *tuple,
list_add(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL].list, &unconfirmed);
atomic_inc(&ip_conntrack_count);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
if (exp) {
if (exp->expectfn)
@@ -723,17 +723,17 @@ void ip_conntrack_unexpect_related(struct ip_conntrack_expect *exp)
{
struct ip_conntrack_expect *i;
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
/* choose the the oldest expectation to evict */
list_for_each_entry_reverse(i, &ip_conntrack_expect_list, list) {
if (expect_matches(i, exp) && del_timer(&i->timeout)) {
unlink_expect(i);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
destroy_expect(i);
return;
}
}
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
}
struct ip_conntrack_expect *ip_conntrack_expect_alloc(void)
@@ -760,15 +760,11 @@ static void ip_conntrack_expect_insert(struct ip_conntrack_expect *exp)
exp->master->expecting++;
list_add(&exp->list, &ip_conntrack_expect_list);
- if (exp->master->helper->timeout) {
- init_timer(&exp->timeout);
- exp->timeout.data = (unsigned long)exp;
- exp->timeout.function = expectation_timed_out;
- exp->timeout.expires
- = jiffies + exp->master->helper->timeout * HZ;
- add_timer(&exp->timeout);
- } else
- exp->timeout.function = NULL;
+ init_timer(&exp->timeout);
+ exp->timeout.data = (unsigned long)exp;
+ exp->timeout.function = expectation_timed_out;
+ exp->timeout.expires = jiffies + exp->master->helper->timeout * HZ;
+ add_timer(&exp->timeout);
CONNTRACK_STAT_INC(expect_create);
}
@@ -808,7 +804,7 @@ int ip_conntrack_expect_related(struct ip_conntrack_expect *expect)
DEBUGP("tuple: "); DUMP_TUPLE(&expect->tuple);
DEBUGP("mask: "); DUMP_TUPLE(&expect->mask);
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
list_for_each_entry(i, &ip_conntrack_expect_list, list) {
if (expect_matches(i, expect)) {
/* Refresh timer: if it's dying, ignore.. */
@@ -832,7 +828,7 @@ int ip_conntrack_expect_related(struct ip_conntrack_expect *expect)
ip_conntrack_expect_insert(expect);
ret = 0;
out:
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
return ret;
}
@@ -841,7 +837,7 @@ out:
void ip_conntrack_alter_reply(struct ip_conntrack *conntrack,
const struct ip_conntrack_tuple *newreply)
{
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
/* Should be unconfirmed, so not in hash table yet */
IP_NF_ASSERT(!is_confirmed(conntrack));
@@ -851,15 +847,15 @@ void ip_conntrack_alter_reply(struct ip_conntrack *conntrack,
conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply;
if (!conntrack->master && conntrack->expecting == 0)
conntrack->helper = ip_ct_find_helper(newreply);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
}
int ip_conntrack_helper_register(struct ip_conntrack_helper *me)
{
BUG_ON(me->timeout == 0);
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
list_prepend(&helpers, me);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
return 0;
}
@@ -878,7 +874,7 @@ void ip_conntrack_helper_unregister(struct ip_conntrack_helper *me)
struct ip_conntrack_expect *exp, *tmp;
/* Need write lock here, to delete helper. */
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
LIST_DELETE(&helpers, me);
/* Get rid of expectations */
@@ -893,7 +889,7 @@ void ip_conntrack_helper_unregister(struct ip_conntrack_helper *me)
for (i = 0; i < ip_conntrack_htable_size; i++)
LIST_FIND_W(&ip_conntrack_hash[i], unhelp,
struct ip_conntrack_tuple_hash *, me);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
/* Someone could be still looking at the helper in a bh. */
synchronize_net();
@@ -925,14 +921,14 @@ void ip_ct_refresh_acct(struct ip_conntrack *ct,
ct->timeout.expires = extra_jiffies;
ct_add_counters(ct, ctinfo, skb);
} else {
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
/* Need del_timer for race avoidance (may already be dying). */
if (del_timer(&ct->timeout)) {
ct->timeout.expires = jiffies + extra_jiffies;
add_timer(&ct->timeout);
}
ct_add_counters(ct, ctinfo, skb);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
}
}
@@ -940,10 +936,6 @@ void ip_ct_refresh_acct(struct ip_conntrack *ct,
struct sk_buff *
ip_ct_gather_frags(struct sk_buff *skb, u_int32_t user)
{
-#ifdef CONFIG_NETFILTER_DEBUG
- unsigned int olddebug = skb->nf_debug;
-#endif
-
skb_orphan(skb);
local_bh_disable();
@@ -953,12 +945,7 @@ ip_ct_gather_frags(struct sk_buff *skb, u_int32_t user)
if (skb) {
ip_send_check(skb->nh.iph);
skb->nfcache |= NFC_ALTERED;
-#ifdef CONFIG_NETFILTER_DEBUG
- /* Packet path as if nothing had happened. */
- skb->nf_debug = olddebug;
-#endif
}
-
return skb;
}
@@ -997,7 +984,7 @@ get_next_corpse(int (*iter)(struct ip_conntrack *i, void *data),
{
struct ip_conntrack_tuple_hash *h = NULL;
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
for (; *bucket < ip_conntrack_htable_size; (*bucket)++) {
h = LIST_FIND_W(&ip_conntrack_hash[*bucket], do_iter,
struct ip_conntrack_tuple_hash *, iter, data);
@@ -1009,7 +996,7 @@ get_next_corpse(int (*iter)(struct ip_conntrack *i, void *data),
struct ip_conntrack_tuple_hash *, iter, data);
if (h)
atomic_inc(&tuplehash_to_ctrack(h)->ct_general.use);
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
return h;
}
@@ -1201,14 +1188,14 @@ int __init ip_conntrack_init(void)
}
/* Don't NEED lock here, but good form anyway. */
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
for (i = 0; i < MAX_IP_CT_PROTO; i++)
ip_ct_protos[i] = &ip_conntrack_generic_protocol;
/* Sew in builtin protocols. */
ip_ct_protos[IPPROTO_TCP] = &ip_conntrack_protocol_tcp;
ip_ct_protos[IPPROTO_UDP] = &ip_conntrack_protocol_udp;
ip_ct_protos[IPPROTO_ICMP] = &ip_conntrack_protocol_icmp;
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
for (i = 0; i < ip_conntrack_htable_size; i++)
INIT_LIST_HEAD(&ip_conntrack_hash[i]);
diff --git a/net/ipv4/netfilter/ip_conntrack_ftp.c b/net/ipv4/netfilter/ip_conntrack_ftp.c
index dd86503..fea6dd2 100644
--- a/net/ipv4/netfilter/ip_conntrack_ftp.c
+++ b/net/ipv4/netfilter/ip_conntrack_ftp.c
@@ -16,7 +16,6 @@
#include <net/checksum.h>
#include <net/tcp.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
#include <linux/netfilter_ipv4/ip_conntrack_helper.h>
#include <linux/netfilter_ipv4/ip_conntrack_ftp.h>
#include <linux/moduleparam.h>
@@ -28,7 +27,7 @@ MODULE_DESCRIPTION("ftp connection tracking helper");
/* This is slow, but it's simple. --RR */
static char ftp_buffer[65536];
-static DECLARE_LOCK(ip_ftp_lock);
+static DEFINE_SPINLOCK(ip_ftp_lock);
#define MAX_PORTS 8
static int ports[MAX_PORTS];
@@ -319,7 +318,7 @@ static int help(struct sk_buff **pskb,
}
datalen = (*pskb)->len - dataoff;
- LOCK_BH(&ip_ftp_lock);
+ spin_lock_bh(&ip_ftp_lock);
fb_ptr = skb_header_pointer(*pskb, dataoff,
(*pskb)->len - dataoff, ftp_buffer);
BUG_ON(fb_ptr == NULL);
@@ -442,7 +441,7 @@ out_update_nl:
if (ends_in_nl)
update_nl_seq(seq, ct_ftp_info,dir);
out:
- UNLOCK_BH(&ip_ftp_lock);
+ spin_unlock_bh(&ip_ftp_lock);
return ret;
}
diff --git a/net/ipv4/netfilter/ip_conntrack_irc.c b/net/ipv4/netfilter/ip_conntrack_irc.c
index 33cc734..cd98772 100644
--- a/net/ipv4/netfilter/ip_conntrack_irc.c
+++ b/net/ipv4/netfilter/ip_conntrack_irc.c
@@ -29,7 +29,6 @@
#include <net/checksum.h>
#include <net/tcp.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
#include <linux/netfilter_ipv4/ip_conntrack_helper.h>
#include <linux/netfilter_ipv4/ip_conntrack_irc.h>
#include <linux/moduleparam.h>
@@ -41,7 +40,7 @@ static int max_dcc_channels = 8;
static unsigned int dcc_timeout = 300;
/* This is slow, but it's simple. --RR */
static char irc_buffer[65536];
-static DECLARE_LOCK(irc_buffer_lock);
+static DEFINE_SPINLOCK(irc_buffer_lock);
unsigned int (*ip_nat_irc_hook)(struct sk_buff **pskb,
enum ip_conntrack_info ctinfo,
@@ -141,7 +140,7 @@ static int help(struct sk_buff **pskb,
if (dataoff >= (*pskb)->len)
return NF_ACCEPT;
- LOCK_BH(&irc_buffer_lock);
+ spin_lock_bh(&irc_buffer_lock);
ib_ptr = skb_header_pointer(*pskb, dataoff,
(*pskb)->len - dataoff, irc_buffer);
BUG_ON(ib_ptr == NULL);
@@ -237,7 +236,7 @@ static int help(struct sk_buff **pskb,
} /* while data < ... */
out:
- UNLOCK_BH(&irc_buffer_lock);
+ spin_unlock_bh(&irc_buffer_lock);
return ret;
}
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c
index ff8c34a..31d7539 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c
@@ -26,7 +26,6 @@
#include <linux/netfilter_ipv4/ip_conntrack.h>
#include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
#if 0
#define DEBUGP(format, ...) printk(format, ## __VA_ARGS__)
@@ -35,7 +34,7 @@
#endif
/* Protects conntrack->proto.sctp */
-static DECLARE_RWLOCK(sctp_lock);
+static DEFINE_RWLOCK(sctp_lock);
/* FIXME: Examine ipfilter's timeouts and conntrack transitions more
closely. They're more complex. --RR
@@ -199,9 +198,9 @@ static int sctp_print_conntrack(struct seq_file *s,
DEBUGP(__FUNCTION__);
DEBUGP("\n");
- READ_LOCK(&sctp_lock);
+ read_lock_bh(&sctp_lock);
state = conntrack->proto.sctp.state;
- READ_UNLOCK(&sctp_lock);
+ read_unlock_bh(&sctp_lock);
return seq_printf(s, "%s ", sctp_conntrack_names[state]);
}
@@ -343,13 +342,13 @@ static int sctp_packet(struct ip_conntrack *conntrack,
oldsctpstate = newconntrack = SCTP_CONNTRACK_MAX;
for_each_sctp_chunk (skb, sch, _sch, offset, count) {
- WRITE_LOCK(&sctp_lock);
+ write_lock_bh(&sctp_lock);
/* Special cases of Verification tag check (Sec 8.5.1) */
if (sch->type == SCTP_CID_INIT) {
/* Sec 8.5.1 (A) */
if (sh->vtag != 0) {
- WRITE_UNLOCK(&sctp_lock);
+ write_unlock_bh(&sctp_lock);
return -1;
}
} else if (sch->type == SCTP_CID_ABORT) {
@@ -357,7 +356,7 @@ static int sctp_packet(struct ip_conntrack *conntrack,
if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)])
&& !(sh->vtag == conntrack->proto.sctp.vtag
[1 - CTINFO2DIR(ctinfo)])) {
- WRITE_UNLOCK(&sctp_lock);
+ write_unlock_bh(&sctp_lock);
return -1;
}
} else if (sch->type == SCTP_CID_SHUTDOWN_COMPLETE) {
@@ -366,13 +365,13 @@ static int sctp_packet(struct ip_conntrack *conntrack,
&& !(sh->vtag == conntrack->proto.sctp.vtag
[1 - CTINFO2DIR(ctinfo)]
&& (sch->flags & 1))) {
- WRITE_UNLOCK(&sctp_lock);
+ write_unlock_bh(&sctp_lock);
return -1;
}
} else if (sch->type == SCTP_CID_COOKIE_ECHO) {
/* Sec 8.5.1 (D) */
if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)])) {
- WRITE_UNLOCK(&sctp_lock);
+ write_unlock_bh(&sctp_lock);
return -1;
}
}
@@ -384,7 +383,7 @@ static int sctp_packet(struct ip_conntrack *conntrack,
if (newconntrack == SCTP_CONNTRACK_MAX) {
DEBUGP("ip_conntrack_sctp: Invalid dir=%i ctype=%u conntrack=%u\n",
CTINFO2DIR(ctinfo), sch->type, oldsctpstate);
- WRITE_UNLOCK(&sctp_lock);
+ write_unlock_bh(&sctp_lock);
return -1;
}
@@ -396,7 +395,7 @@ static int sctp_packet(struct ip_conntrack *conntrack,
ih = skb_header_pointer(skb, offset + sizeof(sctp_chunkhdr_t),
sizeof(_inithdr), &_inithdr);
if (ih == NULL) {
- WRITE_UNLOCK(&sctp_lock);
+ write_unlock_bh(&sctp_lock);
return -1;
}
DEBUGP("Setting vtag %x for dir %d\n",
@@ -405,7 +404,7 @@ static int sctp_packet(struct ip_conntrack *conntrack,
}
conntrack->proto.sctp.state = newconntrack;
- WRITE_UNLOCK(&sctp_lock);
+ write_unlock_bh(&sctp_lock);
}
ip_ct_refresh_acct(conntrack, ctinfo, skb, *sctp_timeouts[newconntrack]);
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
index 721ddbf..809dfed 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
@@ -36,7 +36,6 @@
#include <linux/netfilter_ipv4.h>
#include <linux/netfilter_ipv4/ip_conntrack.h>
#include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
#if 0
#define DEBUGP printk
@@ -46,7 +45,7 @@
#endif
/* Protects conntrack->proto.tcp */
-static DECLARE_RWLOCK(tcp_lock);
+static DEFINE_RWLOCK(tcp_lock);
/* "Be conservative in what you do,
be liberal in what you accept from others."
@@ -330,9 +329,9 @@ static int tcp_print_conntrack(struct seq_file *s,
{
enum tcp_conntrack state;
- READ_LOCK(&tcp_lock);
+ read_lock_bh(&tcp_lock);
state = conntrack->proto.tcp.state;
- READ_UNLOCK(&tcp_lock);
+ read_unlock_bh(&tcp_lock);
return seq_printf(s, "%s ", tcp_conntrack_names[state]);
}
@@ -738,14 +737,14 @@ void ip_conntrack_tcp_update(struct sk_buff *skb,
end = segment_seq_plus_len(ntohl(tcph->seq), skb->len, iph, tcph);
- WRITE_LOCK(&tcp_lock);
+ write_lock_bh(&tcp_lock);
/*
* We have to worry for the ack in the reply packet only...
*/
if (after(end, conntrack->proto.tcp.seen[dir].td_end))
conntrack->proto.tcp.seen[dir].td_end = end;
conntrack->proto.tcp.last_end = end;
- WRITE_UNLOCK(&tcp_lock);
+ write_unlock_bh(&tcp_lock);
DEBUGP("tcp_update: sender end=%u maxend=%u maxwin=%u scale=%i "
"receiver end=%u maxend=%u maxwin=%u scale=%i\n",
sender->td_end, sender->td_maxend, sender->td_maxwin,
@@ -857,7 +856,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
sizeof(_tcph), &_tcph);
BUG_ON(th == NULL);
- WRITE_LOCK(&tcp_lock);
+ write_lock_bh(&tcp_lock);
old_state = conntrack->proto.tcp.state;
dir = CTINFO2DIR(ctinfo);
index = get_conntrack_index(th);
@@ -879,7 +878,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
* that the client cannot but retransmit its SYN and
* thus initiate a clean new session.
*/
- WRITE_UNLOCK(&tcp_lock);
+ write_unlock_bh(&tcp_lock);
if (LOG_INVALID(IPPROTO_TCP))
nf_log_packet(PF_INET, 0, skb, NULL, NULL,
"ip_ct_tcp: killing out of sync session ");
@@ -894,7 +893,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
conntrack->proto.tcp.last_end =
segment_seq_plus_len(ntohl(th->seq), skb->len, iph, th);
- WRITE_UNLOCK(&tcp_lock);
+ write_unlock_bh(&tcp_lock);
if (LOG_INVALID(IPPROTO_TCP))
nf_log_packet(PF_INET, 0, skb, NULL, NULL,
"ip_ct_tcp: invalid packet ignored ");
@@ -904,7 +903,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
DEBUGP("ip_ct_tcp: Invalid dir=%i index=%u ostate=%u\n",
dir, get_conntrack_index(th),
old_state);
- WRITE_UNLOCK(&tcp_lock);
+ write_unlock_bh(&tcp_lock);
if (LOG_INVALID(IPPROTO_TCP))
nf_log_packet(PF_INET, 0, skb, NULL, NULL,
"ip_ct_tcp: invalid state ");
@@ -918,13 +917,13 @@ static int tcp_packet(struct ip_conntrack *conntrack,
conntrack->proto.tcp.seen[dir].td_end)) {
/* Attempt to reopen a closed connection.
* Delete this connection and look up again. */
- WRITE_UNLOCK(&tcp_lock);
+ write_unlock_bh(&tcp_lock);
if (del_timer(&conntrack->timeout))
conntrack->timeout.function((unsigned long)
conntrack);
return -NF_REPEAT;
} else {
- WRITE_UNLOCK(&tcp_lock);
+ write_unlock_bh(&tcp_lock);
if (LOG_INVALID(IPPROTO_TCP))
nf_log_packet(PF_INET, 0, skb, NULL, NULL,
"ip_ct_tcp: invalid SYN");
@@ -949,7 +948,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
if (!tcp_in_window(&conntrack->proto.tcp, dir, index,
skb, iph, th)) {
- WRITE_UNLOCK(&tcp_lock);
+ write_unlock_bh(&tcp_lock);
return -NF_ACCEPT;
}
in_window:
@@ -972,7 +971,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
timeout = conntrack->proto.tcp.retrans >= ip_ct_tcp_max_retrans
&& *tcp_timeouts[new_state] > ip_ct_tcp_timeout_max_retrans
? ip_ct_tcp_timeout_max_retrans : *tcp_timeouts[new_state];
- WRITE_UNLOCK(&tcp_lock);
+ write_unlock_bh(&tcp_lock);
if (!test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status)) {
/* If only reply is a RST, we can consider ourselves not to
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_udp.c b/net/ipv4/netfilter/ip_conntrack_proto_udp.c
index 5bc28a2..8c1eaba 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_udp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_udp.c
@@ -120,6 +120,7 @@ static int udp_error(struct sk_buff *skb, enum ip_conntrack_info *ctinfo,
* and moreover root might send raw packets.
* FIXME: Source route IP option packets --RR */
if (hooknum == NF_IP_PRE_ROUTING
+ && skb->ip_summed != CHECKSUM_UNNECESSARY
&& csum_tcpudp_magic(iph->saddr, iph->daddr, udplen, IPPROTO_UDP,
skb->ip_summed == CHECKSUM_HW ? skb->csum
: skb_checksum(skb, iph->ihl*4, udplen, 0))) {
diff --git a/net/ipv4/netfilter/ip_conntrack_standalone.c b/net/ipv4/netfilter/ip_conntrack_standalone.c
index bc59f7b..42dc951 100644
--- a/net/ipv4/netfilter/ip_conntrack_standalone.c
+++ b/net/ipv4/netfilter/ip_conntrack_standalone.c
@@ -28,8 +28,8 @@
#include <net/checksum.h>
#include <net/ip.h>
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_conntrack_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_conntrack_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
#include <linux/netfilter_ipv4/ip_conntrack.h>
#include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
@@ -119,7 +119,7 @@ static struct list_head *ct_get_idx(struct seq_file *seq, loff_t pos)
static void *ct_seq_start(struct seq_file *seq, loff_t *pos)
{
- READ_LOCK(&ip_conntrack_lock);
+ read_lock_bh(&ip_conntrack_lock);
return ct_get_idx(seq, *pos);
}
@@ -131,7 +131,7 @@ static void *ct_seq_next(struct seq_file *s, void *v, loff_t *pos)
static void ct_seq_stop(struct seq_file *s, void *v)
{
- READ_UNLOCK(&ip_conntrack_lock);
+ read_unlock_bh(&ip_conntrack_lock);
}
static int ct_seq_show(struct seq_file *s, void *v)
@@ -140,7 +140,7 @@ static int ct_seq_show(struct seq_file *s, void *v)
const struct ip_conntrack *conntrack = tuplehash_to_ctrack(hash);
struct ip_conntrack_protocol *proto;
- MUST_BE_READ_LOCKED(&ip_conntrack_lock);
+ ASSERT_READ_LOCK(&ip_conntrack_lock);
IP_NF_ASSERT(conntrack);
/* we only want to print DIR_ORIGINAL */
@@ -239,7 +239,7 @@ static void *exp_seq_start(struct seq_file *s, loff_t *pos)
/* strange seq_file api calls stop even if we fail,
* thus we need to grab lock since stop unlocks */
- READ_LOCK(&ip_conntrack_lock);
+ read_lock_bh(&ip_conntrack_lock);
if (list_empty(e))
return NULL;
@@ -267,7 +267,7 @@ static void *exp_seq_next(struct seq_file *s, void *v, loff_t *pos)
static void exp_seq_stop(struct seq_file *s, void *v)
{
- READ_UNLOCK(&ip_conntrack_lock);
+ read_unlock_bh(&ip_conntrack_lock);
}
static int exp_seq_show(struct seq_file *s, void *v)
@@ -921,22 +921,22 @@ int ip_conntrack_protocol_register(struct ip_conntrack_protocol *proto)
{
int ret = 0;
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
if (ip_ct_protos[proto->proto] != &ip_conntrack_generic_protocol) {
ret = -EBUSY;
goto out;
}
ip_ct_protos[proto->proto] = proto;
out:
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
return ret;
}
void ip_conntrack_protocol_unregister(struct ip_conntrack_protocol *proto)
{
- WRITE_LOCK(&ip_conntrack_lock);
+ write_lock_bh(&ip_conntrack_lock);
ip_ct_protos[proto->proto] = &ip_conntrack_generic_protocol;
- WRITE_UNLOCK(&ip_conntrack_lock);
+ write_unlock_bh(&ip_conntrack_lock);
/* Somebody could be still looking at the proto in bh. */
synchronize_net();
diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c
index 9fc6f93..739b6dd 100644
--- a/net/ipv4/netfilter/ip_nat_core.c
+++ b/net/ipv4/netfilter/ip_nat_core.c
@@ -22,8 +22,8 @@
#include <linux/udp.h>
#include <linux/jhash.h>
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
#include <linux/netfilter_ipv4/ip_conntrack.h>
#include <linux/netfilter_ipv4/ip_conntrack_core.h>
@@ -41,7 +41,7 @@
#define DEBUGP(format, args...)
#endif
-DECLARE_RWLOCK(ip_nat_lock);
+DEFINE_RWLOCK(ip_nat_lock);
/* Calculated at init based on memory size */
static unsigned int ip_nat_htable_size;
@@ -65,9 +65,9 @@ static void ip_nat_cleanup_conntrack(struct ip_conntrack *conn)
if (!(conn->status & IPS_NAT_DONE_MASK))
return;
- WRITE_LOCK(&ip_nat_lock);
+ write_lock_bh(&ip_nat_lock);
list_del(&conn->nat.info.bysource);
- WRITE_UNLOCK(&ip_nat_lock);
+ write_unlock_bh(&ip_nat_lock);
}
/* We do checksum mangling, so if they were wrong before they're still
@@ -142,7 +142,7 @@ find_appropriate_src(const struct ip_conntrack_tuple *tuple,
unsigned int h = hash_by_src(tuple);
struct ip_conntrack *ct;
- READ_LOCK(&ip_nat_lock);
+ read_lock_bh(&ip_nat_lock);
list_for_each_entry(ct, &bysource[h], nat.info.bysource) {
if (same_src(ct, tuple)) {
/* Copy source part from reply tuple. */
@@ -151,12 +151,12 @@ find_appropriate_src(const struct ip_conntrack_tuple *tuple,
result->dst = tuple->dst;
if (in_range(result, range)) {
- READ_UNLOCK(&ip_nat_lock);
+ read_unlock_bh(&ip_nat_lock);
return 1;
}
}
}
- READ_UNLOCK(&ip_nat_lock);
+ read_unlock_bh(&ip_nat_lock);
return 0;
}
@@ -297,9 +297,9 @@ ip_nat_setup_info(struct ip_conntrack *conntrack,
unsigned int srchash
= hash_by_src(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL]
.tuple);
- WRITE_LOCK(&ip_nat_lock);
+ write_lock_bh(&ip_nat_lock);
list_add(&info->bysource, &bysource[srchash]);
- WRITE_UNLOCK(&ip_nat_lock);
+ write_unlock_bh(&ip_nat_lock);
}
/* It's done. */
@@ -474,23 +474,23 @@ int ip_nat_protocol_register(struct ip_nat_protocol *proto)
{
int ret = 0;
- WRITE_LOCK(&ip_nat_lock);
+ write_lock_bh(&ip_nat_lock);
if (ip_nat_protos[proto->protonum] != &ip_nat_unknown_protocol) {
ret = -EBUSY;
goto out;
}
ip_nat_protos[proto->protonum] = proto;
out:
- WRITE_UNLOCK(&ip_nat_lock);
+ write_unlock_bh(&ip_nat_lock);
return ret;
}
/* Noone stores the protocol anywhere; simply delete it. */
void ip_nat_protocol_unregister(struct ip_nat_protocol *proto)
{
- WRITE_LOCK(&ip_nat_lock);
+ write_lock_bh(&ip_nat_lock);
ip_nat_protos[proto->protonum] = &ip_nat_unknown_protocol;
- WRITE_UNLOCK(&ip_nat_lock);
+ write_unlock_bh(&ip_nat_lock);
/* Someone could be still looking at the proto in a bh. */
synchronize_net();
@@ -509,13 +509,13 @@ int __init ip_nat_init(void)
return -ENOMEM;
/* Sew in builtin protocols. */
- WRITE_LOCK(&ip_nat_lock);
+ write_lock_bh(&ip_nat_lock);
for (i = 0; i < MAX_IP_NAT_PROTO; i++)
ip_nat_protos[i] = &ip_nat_unknown_protocol;
ip_nat_protos[IPPROTO_TCP] = &ip_nat_protocol_tcp;
ip_nat_protos[IPPROTO_UDP] = &ip_nat_protocol_udp;
ip_nat_protos[IPPROTO_ICMP] = &ip_nat_protocol_icmp;
- WRITE_UNLOCK(&ip_nat_lock);
+ write_unlock_bh(&ip_nat_lock);
for (i = 0; i < ip_nat_htable_size; i++) {
INIT_LIST_HEAD(&bysource[i]);
diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c
index 1637b96..158f34f 100644
--- a/net/ipv4/netfilter/ip_nat_helper.c
+++ b/net/ipv4/netfilter/ip_nat_helper.c
@@ -28,8 +28,8 @@
#include <net/tcp.h>
#include <net/udp.h>
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
#include <linux/netfilter_ipv4/ip_conntrack.h>
#include <linux/netfilter_ipv4/ip_conntrack_helper.h>
@@ -47,7 +47,7 @@
#define DUMP_OFFSET(x)
#endif
-static DECLARE_LOCK(ip_nat_seqofs_lock);
+static DEFINE_SPINLOCK(ip_nat_seqofs_lock);
/* Setup TCP sequence correction given this change at this sequence */
static inline void
@@ -70,7 +70,7 @@ adjust_tcp_sequence(u32 seq,
DEBUGP("ip_nat_resize_packet: Seq_offset before: ");
DUMP_OFFSET(this_way);
- LOCK_BH(&ip_nat_seqofs_lock);
+ spin_lock_bh(&ip_nat_seqofs_lock);
/* SYN adjust. If it's uninitialized, or this is after last
* correction, record it: we don't handle more than one
@@ -82,7 +82,7 @@ adjust_tcp_sequence(u32 seq,
this_way->offset_before = this_way->offset_after;
this_way->offset_after += sizediff;
}
- UNLOCK_BH(&ip_nat_seqofs_lock);
+ spin_unlock_bh(&ip_nat_seqofs_lock);
DEBUGP("ip_nat_resize_packet: Seq_offset after: ");
DUMP_OFFSET(this_way);
@@ -142,9 +142,6 @@ static int enlarge_skb(struct sk_buff **pskb, unsigned int extra)
/* Transfer socket to new skb. */
if ((*pskb)->sk)
skb_set_owner_w(nskb, (*pskb)->sk);
-#ifdef CONFIG_NETFILTER_DEBUG
- nskb->nf_debug = (*pskb)->nf_debug;
-#endif
kfree_skb(*pskb);
*pskb = nskb;
return 1;
diff --git a/net/ipv4/netfilter/ip_nat_rule.c b/net/ipv4/netfilter/ip_nat_rule.c
index 581f097..60d70fa 100644
--- a/net/ipv4/netfilter/ip_nat_rule.c
+++ b/net/ipv4/netfilter/ip_nat_rule.c
@@ -19,8 +19,8 @@
#include <net/route.h>
#include <linux/bitops.h>
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
#include <linux/netfilter_ipv4/ip_tables.h>
#include <linux/netfilter_ipv4/ip_nat.h>
diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c
index 79f56f6..bc59d0d 100644
--- a/net/ipv4/netfilter/ip_nat_standalone.c
+++ b/net/ipv4/netfilter/ip_nat_standalone.c
@@ -31,8 +31,8 @@
#include <net/checksum.h>
#include <linux/spinlock.h>
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
#include <linux/netfilter_ipv4/ip_nat.h>
#include <linux/netfilter_ipv4/ip_nat_rule.h>
@@ -373,7 +373,6 @@ static int init_or_cleanup(int init)
cleanup_rule_init:
ip_nat_rule_cleanup();
cleanup_nothing:
- MUST_BE_READ_WRITE_UNLOCKED(&ip_nat_lock);
return ret;
}
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 8a54f92..c88dfcd 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -67,7 +67,6 @@ static DECLARE_MUTEX(ipt_mutex);
/* Must have mutex */
#define ASSERT_READ_LOCK(x) IP_NF_ASSERT(down_trylock(&ipt_mutex) != 0)
#define ASSERT_WRITE_LOCK(x) IP_NF_ASSERT(down_trylock(&ipt_mutex) != 0)
-#include <linux/netfilter_ipv4/lockhelp.h>
#include <linux/netfilter_ipv4/listhelp.h>
#if 0
diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c
index 0f12e3a..dc4362b 100644
--- a/net/ipv4/netfilter/ipt_CLUSTERIP.c
+++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c
@@ -29,7 +29,6 @@
#include <linux/netfilter_ipv4/ip_tables.h>
#include <linux/netfilter_ipv4/ipt_CLUSTERIP.h>
#include <linux/netfilter_ipv4/ip_conntrack.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
#define CLUSTERIP_VERSION "0.6"
@@ -41,6 +40,8 @@
#define DEBUGP
#endif
+#define ASSERT_READ_LOCK(x)
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>");
MODULE_DESCRIPTION("iptables target for CLUSTERIP");
@@ -67,7 +68,7 @@ static LIST_HEAD(clusterip_configs);
/* clusterip_lock protects the clusterip_configs list _AND_ the configurable
* data within all structurses (num_local_nodes, local_nodes[]) */
-static DECLARE_RWLOCK(clusterip_lock);
+static DEFINE_RWLOCK(clusterip_lock);
#ifdef CONFIG_PROC_FS
static struct file_operations clusterip_proc_fops;
@@ -82,9 +83,9 @@ clusterip_config_get(struct clusterip_config *c) {
static inline void
clusterip_config_put(struct clusterip_config *c) {
if (atomic_dec_and_test(&c->refcount)) {
- WRITE_LOCK(&clusterip_lock);
+ write_lock_bh(&clusterip_lock);
list_del(&c->list);
- WRITE_UNLOCK(&clusterip_lock);
+ write_unlock_bh(&clusterip_lock);
dev_mc_delete(c->dev, c->clustermac, ETH_ALEN, 0);
dev_put(c->dev);
kfree(c);
@@ -97,7 +98,7 @@ __clusterip_config_find(u_int32_t clusterip)
{
struct list_head *pos;
- MUST_BE_READ_LOCKED(&clusterip_lock);
+ ASSERT_READ_LOCK(&clusterip_lock);
list_for_each(pos, &clusterip_configs) {
struct clusterip_config *c = list_entry(pos,
struct clusterip_config, list);
@@ -114,14 +115,14 @@ clusterip_config_find_get(u_int32_t clusterip)
{
struct clusterip_config *c;
- READ_LOCK(&clusterip_lock);
+ read_lock_bh(&clusterip_lock);
c = __clusterip_config_find(clusterip);
if (!c) {
- READ_UNLOCK(&clusterip_lock);
+ read_unlock_bh(&clusterip_lock);
return NULL;
}
atomic_inc(&c->refcount);
- READ_UNLOCK(&clusterip_lock);
+ read_unlock_bh(&clusterip_lock);
return c;
}
@@ -160,9 +161,9 @@ clusterip_config_init(struct ipt_clusterip_tgt_info *i, u_int32_t ip,
c->pde->data = c;
#endif
- WRITE_LOCK(&clusterip_lock);
+ write_lock_bh(&clusterip_lock);
list_add(&c->list, &clusterip_configs);
- WRITE_UNLOCK(&clusterip_lock);
+ write_unlock_bh(&clusterip_lock);
return c;
}
@@ -172,25 +173,25 @@ clusterip_add_node(struct clusterip_config *c, u_int16_t nodenum)
{
int i;
- WRITE_LOCK(&clusterip_lock);
+ write_lock_bh(&clusterip_lock);
if (c->num_local_nodes >= CLUSTERIP_MAX_NODES
|| nodenum > CLUSTERIP_MAX_NODES) {
- WRITE_UNLOCK(&clusterip_lock);
+ write_unlock_bh(&clusterip_lock);
return 1;
}
/* check if we alrady have this number in our array */
for (i = 0; i < c->num_local_nodes; i++) {
if (c->local_nodes[i] == nodenum) {
- WRITE_UNLOCK(&clusterip_lock);
+ write_unlock_bh(&clusterip_lock);
return 1;
}
}
c->local_nodes[c->num_local_nodes++] = nodenum;
- WRITE_UNLOCK(&clusterip_lock);
+ write_unlock_bh(&clusterip_lock);
return 0;
}
@@ -199,10 +200,10 @@ clusterip_del_node(struct clusterip_config *c, u_int16_t nodenum)
{
int i;
- WRITE_LOCK(&clusterip_lock);
+ write_lock_bh(&clusterip_lock);
if (c->num_local_nodes <= 1 || nodenum > CLUSTERIP_MAX_NODES) {
- WRITE_UNLOCK(&clusterip_lock);
+ write_unlock_bh(&clusterip_lock);
return 1;
}
@@ -211,12 +212,12 @@ clusterip_del_node(struct clusterip_config *c, u_int16_t nodenum)
int size = sizeof(u_int16_t)*(c->num_local_nodes-(i+1));
memmove(&c->local_nodes[i], &c->local_nodes[i+1], size);
c->num_local_nodes--;
- WRITE_UNLOCK(&clusterip_lock);
+ write_unlock_bh(&clusterip_lock);
return 0;
}
}
- WRITE_UNLOCK(&clusterip_lock);
+ write_unlock_bh(&clusterip_lock);
return 1;
}
@@ -286,21 +287,21 @@ clusterip_responsible(struct clusterip_config *config, u_int32_t hash)
{
int i;
- READ_LOCK(&clusterip_lock);
+ read_lock_bh(&clusterip_lock);
if (config->num_local_nodes == 0) {
- READ_UNLOCK(&clusterip_lock);
+ read_unlock_bh(&clusterip_lock);
return 0;
}
for (i = 0; i < config->num_local_nodes; i++) {
if (config->local_nodes[i] == hash) {
- READ_UNLOCK(&clusterip_lock);
+ read_unlock_bh(&clusterip_lock);
return 1;
}
}
- READ_UNLOCK(&clusterip_lock);
+ read_unlock_bh(&clusterip_lock);
return 0;
}
@@ -578,7 +579,7 @@ static void *clusterip_seq_start(struct seq_file *s, loff_t *pos)
struct clusterip_config *c = pde->data;
unsigned int *nodeidx;
- READ_LOCK(&clusterip_lock);
+ read_lock_bh(&clusterip_lock);
if (*pos >= c->num_local_nodes)
return NULL;
@@ -608,7 +609,7 @@ static void clusterip_seq_stop(struct seq_file *s, void *v)
{
kfree(v);
- READ_UNLOCK(&clusterip_lock);
+ read_unlock_bh(&clusterip_lock);
}
static int clusterip_seq_show(struct seq_file *s, void *v)
diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c
index 57e9f6c..91e7450 100644
--- a/net/ipv4/netfilter/ipt_MASQUERADE.c
+++ b/net/ipv4/netfilter/ipt_MASQUERADE.c
@@ -33,7 +33,7 @@ MODULE_DESCRIPTION("iptables MASQUERADE target module");
#endif
/* Lock protects masq region inside conntrack */
-static DECLARE_RWLOCK(masq_lock);
+static DEFINE_RWLOCK(masq_lock);
/* FIXME: Multiple targets. --RR */
static int
@@ -103,9 +103,9 @@ masquerade_target(struct sk_buff **pskb,
return NF_DROP;
}
- WRITE_LOCK(&masq_lock);
+ write_lock_bh(&masq_lock);
ct->nat.masq_index = out->ifindex;
- WRITE_UNLOCK(&masq_lock);
+ write_unlock_bh(&masq_lock);
/* Transfer from original range. */
newrange = ((struct ip_nat_range)
@@ -122,9 +122,9 @@ device_cmp(struct ip_conntrack *i, void *ifindex)
{
int ret;
- READ_LOCK(&masq_lock);
+ read_lock_bh(&masq_lock);
ret = (i->nat.masq_index == (int)(long)ifindex);
- READ_UNLOCK(&masq_lock);
+ read_unlock_bh(&masq_lock);
return ret;
}
diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c
index 266d649..9156964 100644
--- a/net/ipv4/netfilter/ipt_REJECT.c
+++ b/net/ipv4/netfilter/ipt_REJECT.c
@@ -104,10 +104,12 @@ static inline struct rtable *route_reverse(struct sk_buff *skb,
static void send_reset(struct sk_buff *oldskb, int hook)
{
struct sk_buff *nskb;
+ struct iphdr *iph = oldskb->nh.iph;
struct tcphdr _otcph, *oth, *tcph;
struct rtable *rt;
u_int16_t tmp_port;
u_int32_t tmp_addr;
+ unsigned int tcplen;
int needs_ack;
int hh_len;
@@ -124,7 +126,16 @@ static void send_reset(struct sk_buff *oldskb, int hook)
if (oth->rst)
return;
- /* FIXME: Check checksum --RR */
+ /* Check checksum */
+ tcplen = oldskb->len - iph->ihl * 4;
+ if (((hook != NF_IP_LOCAL_IN && oldskb->ip_summed != CHECKSUM_HW) ||
+ (hook == NF_IP_LOCAL_IN &&
+ oldskb->ip_summed != CHECKSUM_UNNECESSARY)) &&
+ csum_tcpudp_magic(iph->saddr, iph->daddr, tcplen, IPPROTO_TCP,
+ oldskb->ip_summed == CHECKSUM_HW ? oldskb->csum :
+ skb_checksum(oldskb, iph->ihl * 4, tcplen, 0)))
+ return;
+
if ((rt = route_reverse(oldskb, oth, hook)) == NULL)
return;
diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c
index 6f2cefb..52a0076 100644
--- a/net/ipv4/netfilter/ipt_ULOG.c
+++ b/net/ipv4/netfilter/ipt_ULOG.c
@@ -56,7 +56,6 @@
#include <linux/netfilter.h>
#include <linux/netfilter_ipv4/ip_tables.h>
#include <linux/netfilter_ipv4/ipt_ULOG.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
#include <net/sock.h>
#include <linux/bitops.h>
@@ -99,8 +98,8 @@ typedef struct {
static ulog_buff_t ulog_buffers[ULOG_MAXNLGROUPS]; /* array of buffers */
-static struct sock *nflognl; /* our socket */
-static DECLARE_LOCK(ulog_lock); /* spinlock */
+static struct sock *nflognl; /* our socket */
+static DEFINE_SPINLOCK(ulog_lock); /* spinlock */
/* send one ulog_buff_t to userspace */
static void ulog_send(unsigned int nlgroupnum)
@@ -135,9 +134,9 @@ static void ulog_timer(unsigned long data)
/* lock to protect against somebody modifying our structure
* from ipt_ulog_target at the same time */
- LOCK_BH(&ulog_lock);
+ spin_lock_bh(&ulog_lock);
ulog_send(data);
- UNLOCK_BH(&ulog_lock);
+ spin_unlock_bh(&ulog_lock);
}
static struct sk_buff *ulog_alloc_skb(unsigned int size)
@@ -193,7 +192,7 @@ static void ipt_ulog_packet(unsigned int hooknum,
ub = &ulog_buffers[groupnum];
- LOCK_BH(&ulog_lock);
+ spin_lock_bh(&ulog_lock);
if (!ub->skb) {
if (!(ub->skb = ulog_alloc_skb(size)))
@@ -278,7 +277,7 @@ static void ipt_ulog_packet(unsigned int hooknum,
ulog_send(groupnum);
}
- UNLOCK_BH(&ulog_lock);
+ spin_unlock_bh(&ulog_lock);
return;
@@ -288,7 +287,7 @@ nlmsg_failure:
alloc_failure:
PRINTR("ipt_ULOG: Error building netlink message\n");
- UNLOCK_BH(&ulog_lock);
+ spin_unlock_bh(&ulog_lock);
}
static unsigned int ipt_ulog_target(struct sk_buff **pskb,
diff --git a/net/ipv4/netfilter/ipt_hashlimit.c b/net/ipv4/netfilter/ipt_hashlimit.c
index f1937190..564b49b 100644
--- a/net/ipv4/netfilter/ipt_hashlimit.c
+++ b/net/ipv4/netfilter/ipt_hashlimit.c
@@ -37,7 +37,6 @@
#include <linux/netfilter_ipv4/ip_tables.h>
#include <linux/netfilter_ipv4/ipt_hashlimit.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
/* FIXME: this is just for IP_NF_ASSERRT */
#include <linux/netfilter_ipv4/ip_conntrack.h>
@@ -92,7 +91,7 @@ struct ipt_hashlimit_htable {
struct hlist_head hash[0]; /* hashtable itself */
};
-static DECLARE_LOCK(hashlimit_lock); /* protects htables list */
+static DEFINE_SPINLOCK(hashlimit_lock); /* protects htables list */
static DECLARE_MUTEX(hlimit_mutex); /* additional checkentry protection */
static HLIST_HEAD(hashlimit_htables);
static kmem_cache_t *hashlimit_cachep;
@@ -233,9 +232,9 @@ static int htable_create(struct ipt_hashlimit_info *minfo)
hinfo->timer.function = htable_gc;
add_timer(&hinfo->timer);
- LOCK_BH(&hashlimit_lock);
+ spin_lock_bh(&hashlimit_lock);
hlist_add_head(&hinfo->node, &hashlimit_htables);
- UNLOCK_BH(&hashlimit_lock);
+ spin_unlock_bh(&hashlimit_lock);
return 0;
}
@@ -301,15 +300,15 @@ static struct ipt_hashlimit_htable *htable_find_get(char *name)
struct ipt_hashlimit_htable *hinfo;
struct hlist_node *pos;
- LOCK_BH(&hashlimit_lock);
+ spin_lock_bh(&hashlimit_lock);
hlist_for_each_entry(hinfo, pos, &hashlimit_htables, node) {
if (!strcmp(name, hinfo->pde->name)) {
atomic_inc(&hinfo->use);
- UNLOCK_BH(&hashlimit_lock);
+ spin_unlock_bh(&hashlimit_lock);
return hinfo;
}
}
- UNLOCK_BH(&hashlimit_lock);
+ spin_unlock_bh(&hashlimit_lock);
return NULL;
}
@@ -317,9 +316,9 @@ static struct ipt_hashlimit_htable *htable_find_get(char *name)
static void htable_put(struct ipt_hashlimit_htable *hinfo)
{
if (atomic_dec_and_test(&hinfo->use)) {
- LOCK_BH(&hashlimit_lock);
+ spin_lock_bh(&hashlimit_lock);
hlist_del(&hinfo->node);
- UNLOCK_BH(&hashlimit_lock);
+ spin_unlock_bh(&hashlimit_lock);
htable_destroy(hinfo);
}
}
diff --git a/net/ipv4/netfilter/ipt_helper.c b/net/ipv4/netfilter/ipt_helper.c
index 33fdf36..3e7dd01 100644
--- a/net/ipv4/netfilter/ipt_helper.c
+++ b/net/ipv4/netfilter/ipt_helper.c
@@ -53,7 +53,7 @@ match(const struct sk_buff *skb,
return ret;
}
- READ_LOCK(&ip_conntrack_lock);
+ read_lock_bh(&ip_conntrack_lock);
if (!ct->master->helper) {
DEBUGP("ipt_helper: master ct %p has no helper\n",
exp->expectant);
@@ -69,7 +69,7 @@ match(const struct sk_buff *skb,
ret ^= !strncmp(ct->master->helper->name, info->name,
strlen(ct->master->helper->name));
out_unlock:
- READ_UNLOCK(&ip_conntrack_lock);
+ read_unlock_bh(&ip_conntrack_lock);
return ret;
}
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index af2392a..66620a9 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -33,6 +33,7 @@ static void xfrm4_encap(struct sk_buff *skb)
struct dst_entry *dst = skb->dst;
struct xfrm_state *x = dst->xfrm;
struct iphdr *iph, *top_iph;
+ int flags;
iph = skb->nh.iph;
skb->h.ipiph = iph;
@@ -51,10 +52,13 @@ static void xfrm4_encap(struct sk_buff *skb)
/* DS disclosed */
top_iph->tos = INET_ECN_encapsulate(iph->tos, iph->tos);
- if (x->props.flags & XFRM_STATE_NOECN)
+
+ flags = x->props.flags;
+ if (flags & XFRM_STATE_NOECN)
IP_ECN_clear(top_iph);
- top_iph->frag_off = iph->frag_off & htons(IP_DF);
+ top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ?
+ 0 : (iph->frag_off & htons(IP_DF));
if (!top_iph->frag_off)
__ip_select_ident(top_iph, dst, 0);
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 223a2e8..050611d 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -7,12 +7,20 @@
*
*/
+#include <net/ip.h>
#include <net/xfrm.h>
#include <linux/pfkeyv2.h>
#include <linux/ipsec.h>
static struct xfrm_state_afinfo xfrm4_state_afinfo;
+static int xfrm4_init_flags(struct xfrm_state *x)
+{
+ if (ipv4_config.no_pmtu_disc)
+ x->props.flags |= XFRM_STATE_NOPMTUDISC;
+ return 0;
+}
+
static void
__xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl,
struct xfrm_tmpl *tmpl,
@@ -109,6 +117,7 @@ __xfrm4_find_acq(u8 mode, u32 reqid, u8 proto,
static struct xfrm_state_afinfo xfrm4_state_afinfo = {
.family = AF_INET,
.lock = RW_LOCK_UNLOCKED,
+ .init_flags = xfrm4_init_flags,
.init_tempsel = __xfrm4_init_tempsel,
.state_lookup = __xfrm4_state_lookup,
.find_acq = __xfrm4_find_acq,
diff --git a/net/ipv4/xfrm4_tunnel.c b/net/ipv4/xfrm4_tunnel.c
index 413191f..e1fe360 100644
--- a/net/ipv4/xfrm4_tunnel.c
+++ b/net/ipv4/xfrm4_tunnel.c
@@ -84,7 +84,7 @@ static void ipip_err(struct sk_buff *skb, u32 info)
handler->err_handler(skb, &arg);
}
-static int ipip_init_state(struct xfrm_state *x, void *args)
+static int ipip_init_state(struct xfrm_state *x)
{
if (!x->props.mode)
return -EINVAL;
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 47a30c3..14f5c53 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -695,7 +695,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp)
if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) {
if (onlink == 0) {
- ip6_del_rt(rt, NULL, NULL);
+ ip6_del_rt(rt, NULL, NULL, NULL);
rt = NULL;
} else if (!(rt->rt6i_flags & RTF_EXPIRES)) {
rt->rt6i_expires = expires;
@@ -1340,7 +1340,7 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev,
if (dev->type == ARPHRD_SIT && (dev->flags&IFF_POINTOPOINT))
rtmsg.rtmsg_flags |= RTF_NONEXTHOP;
- ip6_route_add(&rtmsg, NULL, NULL);
+ ip6_route_add(&rtmsg, NULL, NULL, NULL);
}
/* Create "default" multicast route to the interface */
@@ -1357,7 +1357,7 @@ static void addrconf_add_mroute(struct net_device *dev)
rtmsg.rtmsg_ifindex = dev->ifindex;
rtmsg.rtmsg_flags = RTF_UP;
rtmsg.rtmsg_type = RTMSG_NEWROUTE;
- ip6_route_add(&rtmsg, NULL, NULL);
+ ip6_route_add(&rtmsg, NULL, NULL, NULL);
}
static void sit_route_add(struct net_device *dev)
@@ -1374,7 +1374,7 @@ static void sit_route_add(struct net_device *dev)
rtmsg.rtmsg_flags = RTF_UP|RTF_NONEXTHOP;
rtmsg.rtmsg_ifindex = dev->ifindex;
- ip6_route_add(&rtmsg, NULL, NULL);
+ ip6_route_add(&rtmsg, NULL, NULL, NULL);
}
static void addrconf_add_lroute(struct net_device *dev)
@@ -1467,7 +1467,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len)
if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) {
if (rt->rt6i_flags&RTF_EXPIRES) {
if (valid_lft == 0) {
- ip6_del_rt(rt, NULL, NULL);
+ ip6_del_rt(rt, NULL, NULL, NULL);
rt = NULL;
} else {
rt->rt6i_expires = rt_expires;
@@ -3094,7 +3094,7 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
switch (event) {
case RTM_NEWADDR:
dst_hold(&ifp->rt->u.dst);
- if (ip6_ins_rt(ifp->rt, NULL, NULL))
+ if (ip6_ins_rt(ifp->rt, NULL, NULL, NULL))
dst_release(&ifp->rt->u.dst);
if (ifp->idev->cnf.forwarding)
addrconf_join_anycast(ifp);
@@ -3104,7 +3104,7 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
addrconf_leave_anycast(ifp);
addrconf_leave_solict(ifp->idev, &ifp->addr);
dst_hold(&ifp->rt->u.dst);
- if (ip6_del_rt(ifp->rt, NULL, NULL))
+ if (ip6_del_rt(ifp->rt, NULL, NULL, NULL))
dst_free(&ifp->rt->u.dst);
else
dst_release(&ifp->rt->u.dst);
diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c
index e3ecf62..986fdfd 100644
--- a/net/ipv6/ah6.c
+++ b/net/ipv6/ah6.c
@@ -339,7 +339,7 @@ static void ah6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
xfrm_state_put(x);
}
-static int ah6_init_state(struct xfrm_state *x, void *args)
+static int ah6_init_state(struct xfrm_state *x)
{
struct ah_data *ahp = NULL;
struct xfrm_algo_desc *aalg_desc;
diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index 5d22ca3..6b72940 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -337,7 +337,7 @@ int ipv6_dev_ac_inc(struct net_device *dev, struct in6_addr *addr)
write_unlock_bh(&idev->lock);
dst_hold(&rt->u.dst);
- if (ip6_ins_rt(rt, NULL, NULL))
+ if (ip6_ins_rt(rt, NULL, NULL, NULL))
dst_release(&rt->u.dst);
addrconf_join_solict(dev, &aca->aca_addr);
@@ -380,7 +380,7 @@ int __ipv6_dev_ac_dec(struct inet6_dev *idev, struct in6_addr *addr)
addrconf_leave_solict(idev, &aca->aca_addr);
dst_hold(&aca->aca_rt->u.dst);
- if (ip6_del_rt(aca->aca_rt, NULL, NULL))
+ if (ip6_del_rt(aca->aca_rt, NULL, NULL, NULL))
dst_free(&aca->aca_rt->u.dst);
else
dst_release(&aca->aca_rt->u.dst);
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index be7095d..324db62 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -296,7 +296,7 @@ static void esp6_destroy(struct xfrm_state *x)
kfree(esp);
}
-static int esp6_init_state(struct xfrm_state *x, void *args)
+static int esp6_init_state(struct xfrm_state *x)
{
struct esp_data *esp = NULL;
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 405740b..1b354aa 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -394,7 +394,7 @@ insert_above:
*/
static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
- struct nlmsghdr *nlh)
+ struct nlmsghdr *nlh, struct netlink_skb_parms *req)
{
struct rt6_info *iter = NULL;
struct rt6_info **ins;
@@ -449,7 +449,7 @@ out:
*ins = rt;
rt->rt6i_node = fn;
atomic_inc(&rt->rt6i_ref);
- inet6_rt_notify(RTM_NEWROUTE, rt, nlh);
+ inet6_rt_notify(RTM_NEWROUTE, rt, nlh, req);
rt6_stats.fib_rt_entries++;
if ((fn->fn_flags & RTN_RTINFO) == 0) {
@@ -479,7 +479,8 @@ void fib6_force_start_gc(void)
* with source addr info in sub-trees
*/
-int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
+int fib6_add(struct fib6_node *root, struct rt6_info *rt,
+ struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
{
struct fib6_node *fn;
int err = -ENOMEM;
@@ -552,7 +553,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nlmsghdr *nlh,
}
#endif
- err = fib6_add_rt2node(fn, rt, nlh);
+ err = fib6_add_rt2node(fn, rt, nlh, req);
if (err == 0) {
fib6_start_gc(rt);
@@ -859,7 +860,7 @@ static struct fib6_node * fib6_repair_tree(struct fib6_node *fn)
}
static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
- struct nlmsghdr *nlh, void *_rtattr)
+ struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
{
struct fib6_walker_t *w;
struct rt6_info *rt = *rtp;
@@ -915,11 +916,11 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
if (atomic_read(&rt->rt6i_ref) != 1) BUG();
}
- inet6_rt_notify(RTM_DELROUTE, rt, nlh);
+ inet6_rt_notify(RTM_DELROUTE, rt, nlh, req);
rt6_release(rt);
}
-int fib6_del(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
+int fib6_del(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
{
struct fib6_node *fn = rt->rt6i_node;
struct rt6_info **rtp;
@@ -944,7 +945,7 @@ int fib6_del(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
for (rtp = &fn->leaf; *rtp; rtp = &(*rtp)->u.next) {
if (*rtp == rt) {
- fib6_del_route(fn, rtp, nlh, _rtattr);
+ fib6_del_route(fn, rtp, nlh, _rtattr, req);
return 0;
}
}
@@ -1073,7 +1074,7 @@ static int fib6_clean_node(struct fib6_walker_t *w)
res = c->func(rt, c->arg);
if (res < 0) {
w->leaf = rt;
- res = fib6_del(rt, NULL, NULL);
+ res = fib6_del(rt, NULL, NULL, NULL);
if (res) {
#if RT6_DEBUG >= 2
printk(KERN_DEBUG "fib6_clean_node: del failed: rt=%p@%p err=%d\n", rt, rt->rt6i_node, res);
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index b78a535..06e7cda 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -484,9 +484,6 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from)
to->nf_bridge = from->nf_bridge;
nf_bridge_get(to->nf_bridge);
#endif
-#ifdef CONFIG_NETFILTER_DEBUG
- to->nf_debug = from->nf_debug;
-#endif
#endif
}
diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c
index 6cde531..423feb4 100644
--- a/net/ipv6/ipcomp6.c
+++ b/net/ipv6/ipcomp6.c
@@ -234,14 +234,9 @@ static struct xfrm_state *ipcomp6_tunnel_create(struct xfrm_state *x)
t->props.mode = 1;
memcpy(t->props.saddr.a6, x->props.saddr.a6, sizeof(struct in6_addr));
- t->type = xfrm_get_type(IPPROTO_IPV6, t->props.family);
- if (t->type == NULL)
+ if (xfrm_init_state(t))
goto error;
- if (t->type->init_state(t, NULL))
- goto error;
-
- t->km.state = XFRM_STATE_VALID;
atomic_set(&t->tunnel_users, 1);
out:
@@ -420,7 +415,7 @@ static void ipcomp6_destroy(struct xfrm_state *x)
xfrm6_tunnel_free_spi((xfrm_address_t *)&x->props.saddr);
}
-static int ipcomp6_init_state(struct xfrm_state *x, void *args)
+static int ipcomp6_init_state(struct xfrm_state *x)
{
int err;
struct ipcomp_data *ipcd;
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 279ab86..f3ef4c3 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -423,11 +423,12 @@ done:
psin6 = (struct sockaddr_in6 *)&greqs.gsr_group;
retv = ipv6_sock_mc_join(sk, greqs.gsr_interface,
&psin6->sin6_addr);
- if (retv)
+ /* prior join w/ different source is ok */
+ if (retv && retv != -EADDRINUSE)
break;
omode = MCAST_INCLUDE;
add = 1;
- } else /*IP_DROP_SOURCE_MEMBERSHIP */ {
+ } else /* MCAST_LEAVE_SOURCE_GROUP */ {
omode = MCAST_INCLUDE;
add = 0;
}
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 393b6e6..562fcd1 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -188,6 +188,16 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr)
if (!ipv6_addr_is_multicast(addr))
return -EINVAL;
+ read_lock_bh(&ipv6_sk_mc_lock);
+ for (mc_lst=np->ipv6_mc_list; mc_lst; mc_lst=mc_lst->next) {
+ if ((ifindex == 0 || mc_lst->ifindex == ifindex) &&
+ ipv6_addr_equal(&mc_lst->addr, addr)) {
+ read_unlock_bh(&ipv6_sk_mc_lock);
+ return -EADDRINUSE;
+ }
+ }
+ read_unlock_bh(&ipv6_sk_mc_lock);
+
mc_lst = sock_kmalloc(sk, sizeof(struct ipv6_mc_socklist), GFP_KERNEL);
if (mc_lst == NULL)
@@ -349,6 +359,7 @@ int ip6_mc_source(int add, int omode, struct sock *sk,
struct ipv6_pinfo *inet6 = inet6_sk(sk);
struct ip6_sf_socklist *psl;
int i, j, rv;
+ int leavegroup = 0;
int err;
if (pgsr->gsr_group.ss_family != AF_INET6 ||
@@ -368,6 +379,7 @@ int ip6_mc_source(int add, int omode, struct sock *sk,
err = -EADDRNOTAVAIL;
+ read_lock_bh(&ipv6_sk_mc_lock);
for (pmc=inet6->ipv6_mc_list; pmc; pmc=pmc->next) {
if (pgsr->gsr_interface && pmc->ifindex != pgsr->gsr_interface)
continue;
@@ -401,6 +413,12 @@ int ip6_mc_source(int add, int omode, struct sock *sk,
if (rv) /* source not found */
goto done;
+ /* special case - (INCLUDE, empty) == LEAVE_GROUP */
+ if (psl->sl_count == 1 && omode == MCAST_INCLUDE) {
+ leavegroup = 1;
+ goto done;
+ }
+
/* update the interface filter */
ip6_mc_del_src(idev, group, omode, 1, source, 1);
@@ -453,9 +471,12 @@ int ip6_mc_source(int add, int omode, struct sock *sk,
/* update the interface list */
ip6_mc_add_src(idev, group, omode, 1, source, 1);
done:
+ read_unlock_bh(&ipv6_sk_mc_lock);
read_unlock_bh(&idev->lock);
in6_dev_put(idev);
dev_put(dev);
+ if (leavegroup)
+ return ipv6_sock_mc_drop(sk, pgsr->gsr_interface, group);
return err;
}
@@ -1280,15 +1301,6 @@ static struct sk_buff *mld_newpack(struct net_device *dev, int size)
return NULL;
skb_reserve(skb, LL_RESERVED_SPACE(dev));
- if (dev->hard_header) {
- unsigned char ha[MAX_ADDR_LEN];
-
- ndisc_mc_map(&mld2_all_mcr, ha, dev, 1);
- if (dev->hard_header(skb, dev, ETH_P_IPV6,ha,NULL,size) < 0) {
- kfree_skb(skb);
- return NULL;
- }
- }
if (ipv6_get_lladdr(dev, &addr_buf)) {
/* <draft-ietf-magma-mld-source-05.txt>:
@@ -1312,6 +1324,30 @@ static struct sk_buff *mld_newpack(struct net_device *dev, int size)
return skb;
}
+static inline int mld_dev_queue_xmit2(struct sk_buff *skb)
+{
+ struct net_device *dev = skb->dev;
+
+ if (dev->hard_header) {
+ unsigned char ha[MAX_ADDR_LEN];
+ int err;
+
+ ndisc_mc_map(&skb->nh.ipv6h->daddr, ha, dev, 1);
+ err = dev->hard_header(skb, dev, ETH_P_IPV6, ha, NULL, skb->len);
+ if (err < 0) {
+ kfree_skb(skb);
+ return err;
+ }
+ }
+ return dev_queue_xmit(skb);
+}
+
+static inline int mld_dev_queue_xmit(struct sk_buff *skb)
+{
+ return NF_HOOK(PF_INET6, NF_IP6_POST_ROUTING, skb, NULL, skb->dev,
+ mld_dev_queue_xmit2);
+}
+
static void mld_sendpack(struct sk_buff *skb)
{
struct ipv6hdr *pip6 = skb->nh.ipv6h;
@@ -1329,7 +1365,7 @@ static void mld_sendpack(struct sk_buff *skb)
pmr->csum = csum_ipv6_magic(&pip6->saddr, &pip6->daddr, mldlen,
IPPROTO_ICMPV6, csum_partial(skb->h.raw, mldlen, 0));
err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dev,
- dev_queue_xmit);
+ mld_dev_queue_xmit);
if (!err) {
ICMP6_INC_STATS(idev,ICMP6_MIB_OUTMSGS);
IP6_INC_STATS(IPSTATS_MIB_OUTMCASTPKTS);
@@ -1635,12 +1671,6 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
}
skb_reserve(skb, LL_RESERVED_SPACE(dev));
- if (dev->hard_header) {
- unsigned char ha[MAX_ADDR_LEN];
- ndisc_mc_map(snd_addr, ha, dev, 1);
- if (dev->hard_header(skb, dev, ETH_P_IPV6, ha, NULL, full_len) < 0)
- goto out;
- }
if (ipv6_get_lladdr(dev, &addr_buf)) {
/* <draft-ietf-magma-mld-source-05.txt>:
@@ -1668,7 +1698,7 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
idev = in6_dev_get(skb->dev);
err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dev,
- dev_queue_xmit);
+ mld_dev_queue_xmit);
if (!err) {
if (type == ICMPV6_MGM_REDUCTION)
ICMP6_INC_STATS(idev, ICMP6_MIB_OUTGROUPMEMBREDUCTIONS);
@@ -1682,10 +1712,6 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
if (likely(idev != NULL))
in6_dev_put(idev);
return;
-
-out:
- IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
- kfree_skb(skb);
}
static int ip6_mc_del1_src(struct ifmcaddr6 *pmc, int sfmode,
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 7c291f4..7ae72d4 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -955,7 +955,7 @@ static void ndisc_recv_na(struct sk_buff *skb)
struct rt6_info *rt;
rt = rt6_get_dflt_router(saddr, dev);
if (rt)
- ip6_del_rt(rt, NULL, NULL);
+ ip6_del_rt(rt, NULL, NULL, NULL);
}
out:
@@ -1096,7 +1096,7 @@ static void ndisc_router_discovery(struct sk_buff *skb)
if (rt && lifetime == 0) {
neigh_clone(neigh);
- ip6_del_rt(rt, NULL, NULL);
+ ip6_del_rt(rt, NULL, NULL, NULL);
rt = NULL;
}
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index c735276..7303451 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -71,7 +71,6 @@ static DECLARE_MUTEX(ip6t_mutex);
/* Must have mutex */
#define ASSERT_READ_LOCK(x) IP_NF_ASSERT(down_trylock(&ip6t_mutex) != 0)
#define ASSERT_WRITE_LOCK(x) IP_NF_ASSERT(down_trylock(&ip6t_mutex) != 0)
-#include <linux/netfilter_ipv4/lockhelp.h>
#include <linux/netfilter_ipv4/listhelp.h>
#if 0
diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c
index bfc3d01..c44685e 100644
--- a/net/ipv6/netfilter/ip6t_LOG.c
+++ b/net/ipv6/netfilter/ip6t_LOG.c
@@ -366,8 +366,6 @@ ip6t_log_packet(unsigned int hooknum,
const char *level_string,
const char *prefix)
{
- struct ipv6hdr *ipv6h = skb->nh.ipv6h;
-
spin_lock_bh(&log_lock);
printk(level_string);
printk("%sIN=%s OUT=%s ",
@@ -377,39 +375,25 @@ ip6t_log_packet(unsigned int hooknum,
if (in && !out) {
/* MAC logging for input chain only. */
printk("MAC=");
- if (skb->dev && skb->dev->hard_header_len && skb->mac.raw != (void*)ipv6h) {
- if (skb->dev->type != ARPHRD_SIT){
- int i;
- unsigned char *p = skb->mac.raw;
- for (i = 0; i < skb->dev->hard_header_len; i++,p++)
- printk("%02x%c", *p,
- i==skb->dev->hard_header_len - 1
- ? ' ':':');
- } else {
- int i;
- unsigned char *p = skb->mac.raw;
- if ( p - (ETH_ALEN*2+2) > skb->head ){
- p -= (ETH_ALEN+2);
- for (i = 0; i < (ETH_ALEN); i++,p++)
- printk("%02x%s", *p,
- i == ETH_ALEN-1 ? "->" : ":");
- p -= (ETH_ALEN*2);
- for (i = 0; i < (ETH_ALEN); i++,p++)
- printk("%02x%c", *p,
- i == ETH_ALEN-1 ? ' ' : ':');
- }
-
- if ((skb->dev->addr_len == 4) &&
- skb->dev->hard_header_len > 20){
- printk("TUNNEL=");
- p = skb->mac.raw + 12;
- for (i = 0; i < 4; i++,p++)
- printk("%3d%s", *p,
- i == 3 ? "->" : ".");
- for (i = 0; i < 4; i++,p++)
- printk("%3d%c", *p,
- i == 3 ? ' ' : '.');
- }
+ if (skb->dev && skb->dev->hard_header_len &&
+ skb->mac.raw != skb->nh.raw) {
+ unsigned char *p = skb->mac.raw;
+ int i;
+
+ if (skb->dev->type == ARPHRD_SIT &&
+ (p -= ETH_HLEN) < skb->head)
+ p = NULL;
+
+ if (p != NULL)
+ for (i = 0; i < skb->dev->hard_header_len; i++)
+ printk("%02x", p[i]);
+ printk(" ");
+
+ if (skb->dev->type == ARPHRD_SIT) {
+ struct iphdr *iph = (struct iphdr *)skb->mac.raw;
+ printk("TUNNEL=%u.%u.%u.%u->%u.%u.%u.%u ",
+ NIPQUAD(iph->saddr),
+ NIPQUAD(iph->daddr));
}
} else
printk(" ");
diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c
index 71407be..c2982ef 100644
--- a/net/ipv6/netfilter/ip6table_raw.c
+++ b/net/ipv6/netfilter/ip6table_raw.c
@@ -129,13 +129,15 @@ static struct nf_hook_ops ip6t_ops[] = {
.hook = ip6t_hook,
.pf = PF_INET6,
.hooknum = NF_IP6_PRE_ROUTING,
- .priority = NF_IP6_PRI_FIRST
+ .priority = NF_IP6_PRI_FIRST,
+ .owner = THIS_MODULE,
},
{
.hook = ip6t_hook,
.pf = PF_INET6,
.hooknum = NF_IP6_LOCAL_OUT,
- .priority = NF_IP6_PRI_FIRST
+ .priority = NF_IP6_PRI_FIRST,
+ .owner = THIS_MODULE,
},
};
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 1f5b226..878789b 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -384,12 +384,13 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
be destroyed.
*/
-int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
+int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
+ void *_rtattr, struct netlink_skb_parms *req)
{
int err;
write_lock_bh(&rt6_lock);
- err = fib6_add(&ip6_routing_table, rt, nlh, _rtattr);
+ err = fib6_add(&ip6_routing_table, rt, nlh, _rtattr, req);
write_unlock_bh(&rt6_lock);
return err;
@@ -400,7 +401,7 @@ int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
*/
static struct rt6_info *rt6_cow(struct rt6_info *ort, struct in6_addr *daddr,
- struct in6_addr *saddr)
+ struct in6_addr *saddr, struct netlink_skb_parms *req)
{
int err;
struct rt6_info *rt;
@@ -432,7 +433,7 @@ static struct rt6_info *rt6_cow(struct rt6_info *ort, struct in6_addr *daddr,
dst_hold(&rt->u.dst);
- err = ip6_ins_rt(rt, NULL, NULL);
+ err = ip6_ins_rt(rt, NULL, NULL, req);
if (err == 0)
return rt;
@@ -491,7 +492,8 @@ restart:
read_unlock_bh(&rt6_lock);
nrt = rt6_cow(rt, &skb->nh.ipv6h->daddr,
- &skb->nh.ipv6h->saddr);
+ &skb->nh.ipv6h->saddr,
+ &NETLINK_CB(skb));
dst_release(&rt->u.dst);
rt = nrt;
@@ -551,7 +553,7 @@ restart:
dst_hold(&rt->u.dst);
read_unlock_bh(&rt6_lock);
- nrt = rt6_cow(rt, &fl->fl6_dst, &fl->fl6_src);
+ nrt = rt6_cow(rt, &fl->fl6_dst, &fl->fl6_src, NULL);
dst_release(&rt->u.dst);
rt = nrt;
@@ -598,7 +600,7 @@ static struct dst_entry *ip6_negative_advice(struct dst_entry *dst)
if (rt) {
if (rt->rt6i_flags & RTF_CACHE)
- ip6_del_rt(rt, NULL, NULL);
+ ip6_del_rt(rt, NULL, NULL, NULL);
else
dst_release(dst);
}
@@ -787,7 +789,8 @@ int ipv6_get_hoplimit(struct net_device *dev)
*
*/
-int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_rtattr)
+int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
+ void *_rtattr, struct netlink_skb_parms *req)
{
int err;
struct rtmsg *r;
@@ -974,7 +977,7 @@ install_route:
rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst));
rt->u.dst.dev = dev;
rt->rt6i_idev = idev;
- return ip6_ins_rt(rt, nlh, _rtattr);
+ return ip6_ins_rt(rt, nlh, _rtattr, req);
out:
if (dev)
@@ -986,7 +989,7 @@ out:
return err;
}
-int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
+int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
{
int err;
@@ -994,7 +997,7 @@ int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
rt6_reset_dflt_pointer(NULL);
- err = fib6_del(rt, nlh, _rtattr);
+ err = fib6_del(rt, nlh, _rtattr, req);
dst_release(&rt->u.dst);
write_unlock_bh(&rt6_lock);
@@ -1002,7 +1005,7 @@ int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
return err;
}
-static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_rtattr)
+static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
{
struct fib6_node *fn;
struct rt6_info *rt;
@@ -1029,7 +1032,7 @@ static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_r
dst_hold(&rt->u.dst);
read_unlock_bh(&rt6_lock);
- return ip6_del_rt(rt, nlh, _rtattr);
+ return ip6_del_rt(rt, nlh, _rtattr, req);
}
}
read_unlock_bh(&rt6_lock);
@@ -1136,11 +1139,11 @@ source_ok:
nrt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev);
nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&nrt->u.dst));
- if (ip6_ins_rt(nrt, NULL, NULL))
+ if (ip6_ins_rt(nrt, NULL, NULL, NULL))
goto out;
if (rt->rt6i_flags&RTF_CACHE) {
- ip6_del_rt(rt, NULL, NULL);
+ ip6_del_rt(rt, NULL, NULL, NULL);
return;
}
@@ -1204,7 +1207,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
2. It is gatewayed route or NONEXTHOP route. Action: clone it.
*/
if (!rt->rt6i_nexthop && !(rt->rt6i_flags & RTF_NONEXTHOP)) {
- nrt = rt6_cow(rt, daddr, saddr);
+ nrt = rt6_cow(rt, daddr, saddr, NULL);
if (!nrt->u.dst.error) {
nrt->u.dst.metrics[RTAX_MTU-1] = pmtu;
if (allfrag)
@@ -1232,7 +1235,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
nrt->u.dst.metrics[RTAX_MTU-1] = pmtu;
if (allfrag)
nrt->u.dst.metrics[RTAX_FEATURES-1] |= RTAX_FEATURE_ALLFRAG;
- ip6_ins_rt(nrt, NULL, NULL);
+ ip6_ins_rt(nrt, NULL, NULL, NULL);
}
out:
@@ -1305,7 +1308,7 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
rtmsg.rtmsg_ifindex = dev->ifindex;
- ip6_route_add(&rtmsg, NULL, NULL);
+ ip6_route_add(&rtmsg, NULL, NULL, NULL);
return rt6_get_dflt_router(gwaddr, dev);
}
@@ -1323,7 +1326,7 @@ restart:
read_unlock_bh(&rt6_lock);
- ip6_del_rt(rt, NULL, NULL);
+ ip6_del_rt(rt, NULL, NULL, NULL);
goto restart;
}
@@ -1349,10 +1352,10 @@ int ipv6_route_ioctl(unsigned int cmd, void __user *arg)
rtnl_lock();
switch (cmd) {
case SIOCADDRT:
- err = ip6_route_add(&rtmsg, NULL, NULL);
+ err = ip6_route_add(&rtmsg, NULL, NULL, NULL);
break;
case SIOCDELRT:
- err = ip6_route_del(&rtmsg, NULL, NULL);
+ err = ip6_route_del(&rtmsg, NULL, NULL, NULL);
break;
default:
err = -EINVAL;
@@ -1546,7 +1549,7 @@ int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
return -EINVAL;
- return ip6_route_del(&rtmsg, nlh, arg);
+ return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb));
}
int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
@@ -1556,7 +1559,7 @@ int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
return -EINVAL;
- return ip6_route_add(&rtmsg, nlh, arg);
+ return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb));
}
struct rt6_rtnl_dump_arg
@@ -1566,12 +1569,9 @@ struct rt6_rtnl_dump_arg
};
static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
- struct in6_addr *dst,
- struct in6_addr *src,
- int iif,
- int type, u32 pid, u32 seq,
- struct nlmsghdr *in_nlh, int prefix,
- unsigned int flags)
+ struct in6_addr *dst, struct in6_addr *src,
+ int iif, int type, u32 pid, u32 seq,
+ int prefix, unsigned int flags)
{
struct rtmsg *rtm;
struct nlmsghdr *nlh;
@@ -1585,10 +1585,6 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
}
}
- if (!pid && in_nlh) {
- pid = in_nlh->nlmsg_pid;
- }
-
nlh = NLMSG_NEW(skb, pid, seq, type, sizeof(*rtm), flags);
rtm = NLMSG_DATA(nlh);
rtm->rtm_family = AF_INET6;
@@ -1675,7 +1671,7 @@ static int rt6_dump_route(struct rt6_info *rt, void *p_arg)
return rt6_fill_node(arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE,
NETLINK_CB(arg->cb->skb).pid, arg->cb->nlh->nlmsg_seq,
- NULL, prefix, NLM_F_MULTI);
+ prefix, NLM_F_MULTI);
}
static int fib6_dump_node(struct fib6_walker_t *w)
@@ -1823,7 +1819,7 @@ int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
&fl.fl6_dst, &fl.fl6_src,
iif,
RTM_NEWROUTE, NETLINK_CB(in_skb).pid,
- nlh->nlmsg_seq, nlh, 0, 0);
+ nlh->nlmsg_seq, 0, 0);
if (err < 0) {
err = -EMSGSIZE;
goto out_free;
@@ -1839,17 +1835,25 @@ out_free:
goto out;
}
-void inet6_rt_notify(int event, struct rt6_info *rt, struct nlmsghdr *nlh)
+void inet6_rt_notify(int event, struct rt6_info *rt, struct nlmsghdr *nlh,
+ struct netlink_skb_parms *req)
{
struct sk_buff *skb;
int size = NLMSG_SPACE(sizeof(struct rtmsg)+256);
+ u32 pid = current->pid;
+ u32 seq = 0;
+ if (req)
+ pid = req->pid;
+ if (nlh)
+ seq = nlh->nlmsg_seq;
+
skb = alloc_skb(size, gfp_any());
if (!skb) {
netlink_set_err(rtnl, 0, RTMGRP_IPV6_ROUTE, ENOBUFS);
return;
}
- if (rt6_fill_node(skb, rt, NULL, NULL, 0, event, 0, 0, nlh, 0, 0) < 0) {
+ if (rt6_fill_node(skb, rt, NULL, NULL, 0, event, pid, seq, 0, 0) < 0) {
kfree_skb(skb);
netlink_set_err(rtnl, 0, RTMGRP_IPV6_ROUTE, EINVAL);
return;
diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c
index ffcadd6..60c26c8 100644
--- a/net/ipv6/xfrm6_tunnel.c
+++ b/net/ipv6/xfrm6_tunnel.c
@@ -466,7 +466,7 @@ static void xfrm6_tunnel_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
return;
}
-static int xfrm6_tunnel_init_state(struct xfrm_state *x, void *args)
+static int xfrm6_tunnel_init_state(struct xfrm_state *x)
{
if (!x->props.mode)
return -EINVAL;
diff --git a/net/key/af_key.c b/net/key/af_key.c
index 98b72f2..4879743 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -690,6 +690,8 @@ static struct sk_buff * pfkey_xfrm_state2msg(struct xfrm_state *x, int add_keys,
sa->sadb_sa_flags |= SADB_SAFLAGS_NOECN;
if (x->props.flags & XFRM_STATE_DECAP_DSCP)
sa->sadb_sa_flags |= SADB_SAFLAGS_DECAP_DSCP;
+ if (x->props.flags & XFRM_STATE_NOPMTUDISC)
+ sa->sadb_sa_flags |= SADB_SAFLAGS_NOPMTUDISC;
/* hard time */
if (hsc & 2) {
@@ -974,6 +976,8 @@ static struct xfrm_state * pfkey_msg2xfrm_state(struct sadb_msg *hdr,
x->props.flags |= XFRM_STATE_NOECN;
if (sa->sadb_sa_flags & SADB_SAFLAGS_DECAP_DSCP)
x->props.flags |= XFRM_STATE_DECAP_DSCP;
+ if (sa->sadb_sa_flags & SADB_SAFLAGS_NOPMTUDISC)
+ x->props.flags |= XFRM_STATE_NOPMTUDISC;
lifetime = (struct sadb_lifetime*) ext_hdrs[SADB_EXT_LIFETIME_HARD-1];
if (lifetime != NULL) {
@@ -1096,17 +1100,11 @@ static struct xfrm_state * pfkey_msg2xfrm_state(struct sadb_msg *hdr,
}
}
- x->type = xfrm_get_type(proto, x->props.family);
- if (x->type == NULL) {
- err = -ENOPROTOOPT;
- goto out;
- }
- if (x->type->init_state(x, NULL)) {
- err = -EINVAL;
+ err = xfrm_init_state(x);
+ if (err)
goto out;
- }
+
x->km.seq = hdr->sadb_msg_seq;
- x->km.state = XFRM_STATE_VALID;
return x;
out:
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index 663843d..7ae6aa7 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -191,10 +191,6 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a
asoc->last_cwr_tsn = asoc->ctsn_ack_point;
asoc->unack_data = 0;
- SCTP_DEBUG_PRINTK("myctsnap for %s INIT as 0x%x.\n",
- asoc->ep->debug_name,
- asoc->ctsn_ack_point);
-
/* ADDIP Section 4.1 Asconf Chunk Procedures
*
* When an endpoint has an ASCONF signaled change to be sent to the
@@ -211,6 +207,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a
/* Make an empty list of remote transport addresses. */
INIT_LIST_HEAD(&asoc->peer.transport_addr_list);
+ asoc->peer.transport_count = 0;
/* RFC 2960 5.1 Normal Establishment of an Association
*
@@ -288,6 +285,7 @@ struct sctp_association *sctp_association_new(const struct sctp_endpoint *ep,
asoc->base.malloced = 1;
SCTP_DBG_OBJCNT_INC(assoc);
+ SCTP_DEBUG_PRINTK("Created asoc %p\n", asoc);
return asoc;
@@ -356,6 +354,8 @@ void sctp_association_free(struct sctp_association *asoc)
sctp_transport_free(transport);
}
+ asoc->peer.transport_count = 0;
+
/* Free any cached ASCONF_ACK chunk. */
if (asoc->addip_last_asconf_ack)
sctp_chunk_free(asoc->addip_last_asconf_ack);
@@ -400,7 +400,7 @@ void sctp_assoc_set_primary(struct sctp_association *asoc,
/* If the primary path is changing, assume that the
* user wants to use this new path.
*/
- if (transport->active)
+ if (transport->state != SCTP_INACTIVE)
asoc->peer.active_path = transport;
/*
@@ -428,10 +428,58 @@ void sctp_assoc_set_primary(struct sctp_association *asoc,
transport->cacc.next_tsn_at_change = asoc->next_tsn;
}
+/* Remove a transport from an association. */
+void sctp_assoc_rm_peer(struct sctp_association *asoc,
+ struct sctp_transport *peer)
+{
+ struct list_head *pos;
+ struct sctp_transport *transport;
+
+ SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_rm_peer:association %p addr: ",
+ " port: %d\n",
+ asoc,
+ (&peer->ipaddr),
+ peer->ipaddr.v4.sin_port);
+
+ /* If we are to remove the current retran_path, update it
+ * to the next peer before removing this peer from the list.
+ */
+ if (asoc->peer.retran_path == peer)
+ sctp_assoc_update_retran_path(asoc);
+
+ /* Remove this peer from the list. */
+ list_del(&peer->transports);
+
+ /* Get the first transport of asoc. */
+ pos = asoc->peer.transport_addr_list.next;
+ transport = list_entry(pos, struct sctp_transport, transports);
+
+ /* Update any entries that match the peer to be deleted. */
+ if (asoc->peer.primary_path == peer)
+ sctp_assoc_set_primary(asoc, transport);
+ if (asoc->peer.active_path == peer)
+ asoc->peer.active_path = transport;
+ if (asoc->peer.last_data_from == peer)
+ asoc->peer.last_data_from = transport;
+
+ /* If we remove the transport an INIT was last sent to, set it to
+ * NULL. Combined with the update of the retran path above, this
+ * will cause the next INIT to be sent to the next available
+ * transport, maintaining the cycle.
+ */
+ if (asoc->init_last_sent_to == peer)
+ asoc->init_last_sent_to = NULL;
+
+ asoc->peer.transport_count--;
+
+ sctp_transport_free(peer);
+}
+
/* Add a transport address to an association. */
struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
const union sctp_addr *addr,
- int gfp)
+ const int gfp,
+ const int peer_state)
{
struct sctp_transport *peer;
struct sctp_sock *sp;
@@ -442,14 +490,25 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
/* AF_INET and AF_INET6 share common port field. */
port = addr->v4.sin_port;
+ SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_add_peer:association %p addr: ",
+ " port: %d state:%s\n",
+ asoc,
+ addr,
+ addr->v4.sin_port,
+ peer_state == SCTP_UNKNOWN?"UNKNOWN":"ACTIVE");
+
/* Set the port if it has not been set yet. */
if (0 == asoc->peer.port)
asoc->peer.port = port;
/* Check to see if this is a duplicate. */
peer = sctp_assoc_lookup_paddr(asoc, addr);
- if (peer)
+ if (peer) {
+ if (peer_state == SCTP_ACTIVE &&
+ peer->state == SCTP_UNKNOWN)
+ peer->state = SCTP_ACTIVE;
return peer;
+ }
peer = sctp_transport_new(addr, gfp);
if (!peer)
@@ -516,8 +575,12 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
/* Set the transport's RTO.initial value */
peer->rto = asoc->rto_initial;
+ /* Set the peer's active state. */
+ peer->state = peer_state;
+
/* Attach the remote transport to our asoc. */
list_add_tail(&peer->transports, &asoc->peer.transport_addr_list);
+ asoc->peer.transport_count++;
/* If we do not yet have a primary path, set one. */
if (!asoc->peer.primary_path) {
@@ -525,8 +588,9 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
asoc->peer.retran_path = peer;
}
- if (asoc->peer.active_path == asoc->peer.retran_path)
+ if (asoc->peer.active_path == asoc->peer.retran_path) {
asoc->peer.retran_path = peer;
+ }
return peer;
}
@@ -537,37 +601,16 @@ void sctp_assoc_del_peer(struct sctp_association *asoc,
{
struct list_head *pos;
struct list_head *temp;
- struct sctp_transport *peer = NULL;
struct sctp_transport *transport;
list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) {
transport = list_entry(pos, struct sctp_transport, transports);
if (sctp_cmp_addr_exact(addr, &transport->ipaddr)) {
- peer = transport;
- list_del(pos);
+ /* Do book keeping for removing the peer and free it. */
+ sctp_assoc_rm_peer(asoc, transport);
break;
}
}
-
- /* The address we want delete is not in the association. */
- if (!peer)
- return;
-
- /* Get the first transport of asoc. */
- pos = asoc->peer.transport_addr_list.next;
- transport = list_entry(pos, struct sctp_transport, transports);
-
- /* Update any entries that match the peer to be deleted. */
- if (asoc->peer.primary_path == peer)
- sctp_assoc_set_primary(asoc, transport);
- if (asoc->peer.active_path == peer)
- asoc->peer.active_path = transport;
- if (asoc->peer.retran_path == peer)
- asoc->peer.retran_path = transport;
- if (asoc->peer.last_data_from == peer)
- asoc->peer.last_data_from = transport;
-
- sctp_transport_free(peer);
}
/* Lookup a transport by address. */
@@ -608,12 +651,12 @@ void sctp_assoc_control_transport(struct sctp_association *asoc,
/* Record the transition on the transport. */
switch (command) {
case SCTP_TRANSPORT_UP:
- transport->active = SCTP_ACTIVE;
+ transport->state = SCTP_ACTIVE;
spc_state = SCTP_ADDR_AVAILABLE;
break;
case SCTP_TRANSPORT_DOWN:
- transport->active = SCTP_INACTIVE;
+ transport->state = SCTP_INACTIVE;
spc_state = SCTP_ADDR_UNREACHABLE;
break;
@@ -643,7 +686,7 @@ void sctp_assoc_control_transport(struct sctp_association *asoc,
list_for_each(pos, &asoc->peer.transport_addr_list) {
t = list_entry(pos, struct sctp_transport, transports);
- if (!t->active)
+ if (t->state == SCTP_INACTIVE)
continue;
if (!first || t->last_time_heard > first->last_time_heard) {
second = first;
@@ -663,7 +706,7 @@ void sctp_assoc_control_transport(struct sctp_association *asoc,
* [If the primary is active but not most recent, bump the most
* recently used transport.]
*/
- if (asoc->peer.primary_path->active &&
+ if (asoc->peer.primary_path->state != SCTP_INACTIVE &&
first != asoc->peer.primary_path) {
second = first;
first = asoc->peer.primary_path;
@@ -958,7 +1001,7 @@ void sctp_assoc_update(struct sctp_association *asoc,
transports);
if (!sctp_assoc_lookup_paddr(asoc, &trans->ipaddr))
sctp_assoc_add_peer(asoc, &trans->ipaddr,
- GFP_ATOMIC);
+ GFP_ATOMIC, SCTP_ACTIVE);
}
asoc->ctsn_ack_point = asoc->next_tsn - 1;
@@ -998,7 +1041,7 @@ void sctp_assoc_update_retran_path(struct sctp_association *asoc)
/* Try to find an active transport. */
- if (t->active) {
+ if (t->state != SCTP_INACTIVE) {
break;
} else {
/* Keep track of the next transport in case
@@ -1019,6 +1062,40 @@ void sctp_assoc_update_retran_path(struct sctp_association *asoc)
}
asoc->peer.retran_path = t;
+
+ SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_update_retran_path:association"
+ " %p addr: ",
+ " port: %d\n",
+ asoc,
+ (&t->ipaddr),
+ t->ipaddr.v4.sin_port);
+}
+
+/* Choose the transport for sending a INIT packet. */
+struct sctp_transport *sctp_assoc_choose_init_transport(
+ struct sctp_association *asoc)
+{
+ struct sctp_transport *t;
+
+ /* Use the retran path. If the last INIT was sent over the
+ * retran path, update the retran path and use it.
+ */
+ if (!asoc->init_last_sent_to) {
+ t = asoc->peer.active_path;
+ } else {
+ if (asoc->init_last_sent_to == asoc->peer.retran_path)
+ sctp_assoc_update_retran_path(asoc);
+ t = asoc->peer.retran_path;
+ }
+
+ SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_update_retran_path:association"
+ " %p addr: ",
+ " port: %d\n",
+ asoc,
+ (&t->ipaddr),
+ t->ipaddr.v4.sin_port);
+
+ return t;
}
/* Choose the transport for sending a SHUTDOWN packet. */
diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c
index 334f617..2ec0320 100644
--- a/net/sctp/endpointola.c
+++ b/net/sctp/endpointola.c
@@ -134,7 +134,6 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep,
ep->last_key = ep->current_key = 0;
ep->key_changed_at = jiffies;
- ep->debug_name = "unnamedEndpoint";
return ep;
}
diff --git a/net/sctp/input.c b/net/sctp/input.c
index fffc880..339f7ac 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -353,7 +353,7 @@ void sctp_icmp_proto_unreachable(struct sock *sk,
sctp_do_sm(SCTP_EVENT_T_OTHER,
SCTP_ST_OTHER(SCTP_EVENT_ICMP_PROTO_UNREACH),
- asoc->state, asoc->ep, asoc, NULL,
+ asoc->state, asoc->ep, asoc, t,
GFP_ATOMIC);
}
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index 1b2d4ad..4eb81a1 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -682,9 +682,9 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
if (!new_transport) {
new_transport = asoc->peer.active_path;
- } else if (!new_transport->active) {
- /* If the chunk is Heartbeat or Heartbeat Ack,
- * send it to chunk->transport, even if it's
+ } else if (new_transport->state == SCTP_INACTIVE) {
+ /* If the chunk is Heartbeat or Heartbeat Ack,
+ * send it to chunk->transport, even if it's
* inactive.
*
* 3.3.6 Heartbeat Acknowledgement:
@@ -840,7 +840,8 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
* Otherwise, we want to use the active path.
*/
new_transport = chunk->transport;
- if (!new_transport || !new_transport->active)
+ if (!new_transport ||
+ new_transport->state == SCTP_INACTIVE)
new_transport = asoc->peer.active_path;
/* Change packets if necessary. */
@@ -1454,7 +1455,7 @@ static void sctp_check_transmitted(struct sctp_outq *q,
/* Mark the destination transport address as
* active if it is not so marked.
*/
- if (!transport->active) {
+ if (transport->state == SCTP_INACTIVE) {
sctp_assoc_control_transport(
transport->asoc,
transport,
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 33ac8bf..5baed9b 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -1830,7 +1830,7 @@ int sctp_process_init(struct sctp_association *asoc, sctp_cid_t cid,
* be a a better choice than any of the embedded addresses.
*/
if (peer_addr)
- if(!sctp_assoc_add_peer(asoc, peer_addr, gfp))
+ if(!sctp_assoc_add_peer(asoc, peer_addr, gfp, SCTP_ACTIVE))
goto nomem;
/* Process the initialization parameters. */
@@ -1841,6 +1841,14 @@ int sctp_process_init(struct sctp_association *asoc, sctp_cid_t cid,
goto clean_up;
}
+ /* Walk list of transports, removing transports in the UNKNOWN state. */
+ list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) {
+ transport = list_entry(pos, struct sctp_transport, transports);
+ if (transport->state == SCTP_UNKNOWN) {
+ sctp_assoc_rm_peer(asoc, transport);
+ }
+ }
+
/* The fixed INIT headers are always in network byte
* order.
*/
@@ -1906,7 +1914,8 @@ int sctp_process_init(struct sctp_association *asoc, sctp_cid_t cid,
* stream sequence number shall be set to 0.
*/
- /* Allocate storage for the negotiated streams if it is not a temporary * association.
+ /* Allocate storage for the negotiated streams if it is not a temporary
+ * association.
*/
if (!asoc->temp) {
int assoc_id;
@@ -1952,6 +1961,9 @@ clean_up:
list_del_init(pos);
sctp_transport_free(transport);
}
+
+ asoc->peer.transport_count = 0;
+
nomem:
return 0;
}
@@ -1995,7 +2007,7 @@ static int sctp_process_param(struct sctp_association *asoc,
af->from_addr_param(&addr, param.addr, asoc->peer.port, 0);
scope = sctp_scope(peer_addr);
if (sctp_in_scope(&addr, scope))
- if (!sctp_assoc_add_peer(asoc, &addr, gfp))
+ if (!sctp_assoc_add_peer(asoc, &addr, gfp, SCTP_ACTIVE))
return 0;
break;
@@ -2396,7 +2408,7 @@ static __u16 sctp_process_asconf_param(struct sctp_association *asoc,
* Due to Resource Shortage'.
*/
- peer = sctp_assoc_add_peer(asoc, &addr, GFP_ATOMIC);
+ peer = sctp_assoc_add_peer(asoc, &addr, GFP_ATOMIC, SCTP_ACTIVE);
if (!peer)
return SCTP_ERROR_RSRC_LOW;
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index f65fa44..778639d 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -414,11 +414,13 @@ static void sctp_do_8_2_transport_strike(struct sctp_association *asoc,
*/
asoc->overall_error_count++;
- if (transport->active &&
+ if (transport->state != SCTP_INACTIVE &&
(transport->error_count++ >= transport->max_retrans)) {
- SCTP_DEBUG_PRINTK("transport_strike: transport "
- "IP:%d.%d.%d.%d failed.\n",
- NIPQUAD(transport->ipaddr.v4.sin_addr));
+ SCTP_DEBUG_PRINTK_IPADDR("transport_strike:association %p",
+ " transport IP: port:%d failed.\n",
+ asoc,
+ (&transport->ipaddr),
+ transport->ipaddr.v4.sin_port);
sctp_assoc_control_transport(asoc, transport,
SCTP_TRANSPORT_DOWN,
SCTP_FAILED_THRESHOLD);
@@ -593,7 +595,7 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds,
/* Mark the destination transport address as active if it is not so
* marked.
*/
- if (!t->active)
+ if (t->state == SCTP_INACTIVE)
sctp_assoc_control_transport(asoc, t, SCTP_TRANSPORT_UP,
SCTP_HEARTBEAT_SUCCESS);
@@ -665,8 +667,11 @@ static void sctp_cmd_new_state(sctp_cmd_seq_t *cmds,
asoc->state = state;
+ SCTP_DEBUG_PRINTK("sctp_cmd_new_state: asoc %p[%s]\n",
+ asoc, sctp_state_tbl[state]);
+
if (sctp_style(sk, TCP)) {
- /* Change the sk->sk_state of a TCP-style socket that has
+ /* Change the sk->sk_state of a TCP-style socket that has
* sucessfully completed a connect() call.
*/
if (sctp_state(asoc, ESTABLISHED) && sctp_sstate(sk, CLOSED))
@@ -678,6 +683,16 @@ static void sctp_cmd_new_state(sctp_cmd_seq_t *cmds,
sk->sk_shutdown |= RCV_SHUTDOWN;
}
+ if (sctp_state(asoc, COOKIE_WAIT)) {
+ /* Reset init timeouts since they may have been
+ * increased due to timer expirations.
+ */
+ asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] =
+ asoc->ep->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT];
+ asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] =
+ asoc->ep->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE];
+ }
+
if (sctp_state(asoc, ESTABLISHED) ||
sctp_state(asoc, CLOSED) ||
sctp_state(asoc, SHUTDOWN_RECEIVED)) {
@@ -1120,10 +1135,10 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
* to be executed only during failed attempts of
* association establishment.
*/
- if ((asoc->peer.retran_path !=
- asoc->peer.primary_path) &&
- (asoc->counters[SCTP_COUNTER_INIT_ERROR] > 0)) {
- sctp_add_cmd_sf(commands,
+ if ((asoc->peer.retran_path !=
+ asoc->peer.primary_path) &&
+ (asoc->init_err_counter > 0)) {
+ sctp_add_cmd_sf(commands,
SCTP_CMD_FORCE_PRIM_RETRAN,
SCTP_NULL());
}
@@ -1237,18 +1252,67 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
sctp_association_put(asoc);
break;
+ case SCTP_CMD_INIT_CHOOSE_TRANSPORT:
+ chunk = cmd->obj.ptr;
+ t = sctp_assoc_choose_init_transport(asoc);
+ asoc->init_last_sent_to = t;
+ chunk->transport = t;
+ t->init_sent_count++;
+ break;
+
case SCTP_CMD_INIT_RESTART:
/* Do the needed accounting and updates
* associated with restarting an initialization
- * timer.
+ * timer. Only multiply the timeout by two if
+ * all transports have been tried at the current
+ * timeout.
+ */
+ t = asoc->init_last_sent_to;
+ asoc->init_err_counter++;
+
+ if (t->init_sent_count > (asoc->init_cycle + 1)) {
+ asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] *= 2;
+ if (asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] >
+ asoc->max_init_timeo) {
+ asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] =
+ asoc->max_init_timeo;
+ }
+ asoc->init_cycle++;
+ SCTP_DEBUG_PRINTK(
+ "T1 INIT Timeout adjustment"
+ " init_err_counter: %d"
+ " cycle: %d"
+ " timeout: %d\n",
+ asoc->init_err_counter,
+ asoc->init_cycle,
+ asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT]);
+ }
+
+ sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART,
+ SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT));
+ break;
+
+ case SCTP_CMD_COOKIEECHO_RESTART:
+ /* Do the needed accounting and updates
+ * associated with restarting an initialization
+ * timer. Only multiply the timeout by two if
+ * all transports have been tried at the current
+ * timeout.
*/
- asoc->counters[SCTP_COUNTER_INIT_ERROR]++;
- asoc->timeouts[cmd->obj.to] *= 2;
- if (asoc->timeouts[cmd->obj.to] >
+ asoc->init_err_counter++;
+
+ asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] *= 2;
+ if (asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] >
asoc->max_init_timeo) {
- asoc->timeouts[cmd->obj.to] =
+ asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] =
asoc->max_init_timeo;
}
+ SCTP_DEBUG_PRINTK(
+ "T1 COOKIE Timeout adjustment"
+ " init_err_counter: %d"
+ " timeout: %d\n",
+ asoc->init_err_counter,
+ asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE]);
/* If we've sent any data bundled with
* COOKIE-ECHO we need to resend.
@@ -1261,7 +1325,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
sctp_add_cmd_sf(commands,
SCTP_CMD_TIMER_RESTART,
- SCTP_TO(cmd->obj.to));
+ SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE));
break;
case SCTP_CMD_INIT_FAILED:
@@ -1273,12 +1337,13 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
subtype, chunk, cmd->obj.u32);
break;
- case SCTP_CMD_COUNTER_INC:
- asoc->counters[cmd->obj.counter]++;
+ case SCTP_CMD_INIT_COUNTER_INC:
+ asoc->init_err_counter++;
break;
- case SCTP_CMD_COUNTER_RESET:
- asoc->counters[cmd->obj.counter] = 0;
+ case SCTP_CMD_INIT_COUNTER_RESET:
+ asoc->init_err_counter = 0;
+ asoc->init_cycle = 0;
break;
case SCTP_CMD_REPORT_DUP:
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 8e01b8f..0581896 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -533,6 +533,9 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(const struct sctp_endpoint *ep,
sctp_add_cmd_sf(commands, SCTP_CMD_PEER_INIT,
SCTP_PEER_INIT(initchunk));
+ /* Reset init error count upon receipt of INIT-ACK. */
+ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_COUNTER_RESET, SCTP_NULL());
+
/* 5.1 C) "A" shall stop the T1-init timer and leave
* COOKIE-WAIT state. "A" shall then ... start the T1-cookie
* timer, and enter the COOKIE-ECHOED state.
@@ -775,8 +778,7 @@ sctp_disposition_t sctp_sf_do_5_1E_ca(const struct sctp_endpoint *ep,
* from the COOKIE-ECHOED state to the COOKIE-WAIT
* state is performed.
*/
- sctp_add_cmd_sf(commands, SCTP_CMD_COUNTER_RESET,
- SCTP_COUNTER(SCTP_COUNTER_INIT_ERROR));
+ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_COUNTER_RESET, SCTP_NULL());
/* RFC 2960 5.1 Normal Establishment of an Association
*
@@ -1019,10 +1021,22 @@ sctp_disposition_t sctp_sf_backbeat_8_3(const struct sctp_endpoint *ep,
link = sctp_assoc_lookup_paddr(asoc, &from_addr);
/* This should never happen, but lets log it if so. */
- if (!link) {
- printk(KERN_WARNING
- "%s: Could not find address %d.%d.%d.%d\n",
- __FUNCTION__, NIPQUAD(from_addr.v4.sin_addr));
+ if (unlikely(!link)) {
+ if (from_addr.sa.sa_family == AF_INET6) {
+ printk(KERN_WARNING
+ "%s association %p could not find address "
+ "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n",
+ __FUNCTION__,
+ asoc,
+ NIP6(from_addr.v6.sin6_addr));
+ } else {
+ printk(KERN_WARNING
+ "%s association %p could not find address "
+ "%u.%u.%u.%u\n",
+ __FUNCTION__,
+ asoc,
+ NIPQUAD(from_addr.v4.sin_addr.s_addr));
+ }
return SCTP_DISPOSITION_DISCARD;
}
@@ -2095,9 +2109,7 @@ static sctp_disposition_t sctp_sf_do_5_2_6_stale(const struct sctp_endpoint *ep,
sctp_errhdr_t *err;
struct sctp_chunk *reply;
struct sctp_bind_addr *bp;
- int attempts;
-
- attempts = asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1;
+ int attempts = asoc->init_err_counter + 1;
if (attempts >= asoc->max_init_attempts) {
sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED,
@@ -2157,8 +2169,7 @@ static sctp_disposition_t sctp_sf_do_5_2_6_stale(const struct sctp_endpoint *ep,
/* Cast away the const modifier, as we want to just
* rerun it through as a sideffect.
*/
- sctp_add_cmd_sf(commands, SCTP_CMD_COUNTER_INC,
- SCTP_COUNTER(SCTP_COUNTER_INIT_ERROR));
+ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_COUNTER_INC, SCTP_NULL());
sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP,
SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE));
@@ -2281,8 +2292,7 @@ sctp_disposition_t sctp_sf_cookie_wait_abort(const struct sctp_endpoint *ep,
if (len >= sizeof(struct sctp_chunkhdr) + sizeof(struct sctp_errhdr))
error = ((sctp_errhdr_t *)chunk->skb->data)->cause;
- sctp_stop_t1_and_abort(commands, error);
- return SCTP_DISPOSITION_ABORT;
+ return sctp_stop_t1_and_abort(commands, error, asoc, chunk->transport);
}
/*
@@ -2294,8 +2304,8 @@ sctp_disposition_t sctp_sf_cookie_wait_icmp_abort(const struct sctp_endpoint *ep
void *arg,
sctp_cmd_seq_t *commands)
{
- sctp_stop_t1_and_abort(commands, SCTP_ERROR_NO_ERROR);
- return SCTP_DISPOSITION_ABORT;
+ return sctp_stop_t1_and_abort(commands, SCTP_ERROR_NO_ERROR, asoc,
+ (struct sctp_transport *)arg);
}
/*
@@ -2318,8 +2328,12 @@ sctp_disposition_t sctp_sf_cookie_echoed_abort(const struct sctp_endpoint *ep,
*
* This is common code called by several sctp_sf_*_abort() functions above.
*/
-void sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, __u16 error)
+sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands,
+ __u16 error,
+ const struct sctp_association *asoc,
+ struct sctp_transport *transport)
{
+ SCTP_DEBUG_PRINTK("ABORT received (INIT).\n");
sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE,
SCTP_STATE(SCTP_STATE_CLOSED));
SCTP_INC_STATS(SCTP_MIB_ABORTEDS);
@@ -2328,6 +2342,7 @@ void sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, __u16 error)
/* CMD_INIT_FAILED will DELETE_TCB. */
sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED,
SCTP_U32(error));
+ return SCTP_DISPOSITION_ABORT;
}
/*
@@ -3805,6 +3820,10 @@ sctp_disposition_t sctp_sf_do_prm_asoc(const struct sctp_endpoint *ep,
sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC,
SCTP_ASOC((struct sctp_association *) asoc));
+ /* Choose transport for INIT. */
+ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_CHOOSE_TRANSPORT,
+ SCTP_CHUNK(repl));
+
/* After sending the INIT, "A" starts the T1-init timer and
* enters the COOKIE-WAIT state.
*/
@@ -4589,7 +4608,7 @@ sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep,
}
/*
- * sctp_sf_t1_timer_expire
+ * sctp_sf_t1_init_timer_expire
*
* Section: 4 Note: 2
* Verification Tag:
@@ -4603,7 +4622,59 @@ sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep,
* endpoint MUST abort the initialization process and report the
* error to SCTP user.
*
- * 3) If the T1-cookie timer expires, the endpoint MUST retransmit
+ * Outputs
+ * (timers, events)
+ *
+ */
+sctp_disposition_t sctp_sf_t1_init_timer_expire(const struct sctp_endpoint *ep,
+ const struct sctp_association *asoc,
+ const sctp_subtype_t type,
+ void *arg,
+ sctp_cmd_seq_t *commands)
+{
+ struct sctp_chunk *repl = NULL;
+ struct sctp_bind_addr *bp;
+ int attempts = asoc->init_err_counter + 1;
+
+ SCTP_DEBUG_PRINTK("Timer T1 expired (INIT).\n");
+
+ if (attempts < asoc->max_init_attempts) {
+ bp = (struct sctp_bind_addr *) &asoc->base.bind_addr;
+ repl = sctp_make_init(asoc, bp, GFP_ATOMIC, 0);
+ if (!repl)
+ return SCTP_DISPOSITION_NOMEM;
+
+ /* Choose transport for INIT. */
+ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_CHOOSE_TRANSPORT,
+ SCTP_CHUNK(repl));
+
+ /* Issue a sideeffect to do the needed accounting. */
+ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_RESTART,
+ SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT));
+
+ sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl));
+ } else {
+ SCTP_DEBUG_PRINTK("Giving up on INIT, attempts: %d"
+ " max_init_attempts: %d\n",
+ attempts, asoc->max_init_attempts);
+ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED,
+ SCTP_U32(SCTP_ERROR_NO_ERROR));
+ return SCTP_DISPOSITION_DELETE_TCB;
+ }
+
+ return SCTP_DISPOSITION_CONSUME;
+}
+
+/*
+ * sctp_sf_t1_cookie_timer_expire
+ *
+ * Section: 4 Note: 2
+ * Verification Tag:
+ * Inputs
+ * (endpoint, asoc)
+ *
+ * RFC 2960 Section 4 Notes
+ * 3) If the T1-cookie timer expires, the endpoint MUST retransmit
* COOKIE ECHO and re-start the T1-cookie timer without changing
* state. This MUST be repeated up to 'Max.Init.Retransmits' times.
* After that, the endpoint MUST abort the initialization process and
@@ -4613,46 +4684,26 @@ sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep,
* (timers, events)
*
*/
-sctp_disposition_t sctp_sf_t1_timer_expire(const struct sctp_endpoint *ep,
+sctp_disposition_t sctp_sf_t1_cookie_timer_expire(const struct sctp_endpoint *ep,
const struct sctp_association *asoc,
const sctp_subtype_t type,
void *arg,
sctp_cmd_seq_t *commands)
{
- struct sctp_chunk *repl;
- struct sctp_bind_addr *bp;
- sctp_event_timeout_t timer = (sctp_event_timeout_t) arg;
- int timeout;
- int attempts;
-
- timeout = asoc->timeouts[timer];
- attempts = asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1;
- repl = NULL;
+ struct sctp_chunk *repl = NULL;
+ int attempts = asoc->init_err_counter + 1;
- SCTP_DEBUG_PRINTK("Timer T1 expired.\n");
+ SCTP_DEBUG_PRINTK("Timer T1 expired (COOKIE-ECHO).\n");
if (attempts < asoc->max_init_attempts) {
- switch (timer) {
- case SCTP_EVENT_TIMEOUT_T1_INIT:
- bp = (struct sctp_bind_addr *) &asoc->base.bind_addr;
- repl = sctp_make_init(asoc, bp, GFP_ATOMIC, 0);
- break;
-
- case SCTP_EVENT_TIMEOUT_T1_COOKIE:
- repl = sctp_make_cookie_echo(asoc, NULL);
- break;
-
- default:
- BUG();
- break;
- };
-
+ repl = sctp_make_cookie_echo(asoc, NULL);
if (!repl)
- goto nomem;
+ return SCTP_DISPOSITION_NOMEM;
/* Issue a sideeffect to do the needed accounting. */
- sctp_add_cmd_sf(commands, SCTP_CMD_INIT_RESTART,
- SCTP_TO(timer));
+ sctp_add_cmd_sf(commands, SCTP_CMD_COOKIEECHO_RESTART,
+ SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE));
+
sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl));
} else {
sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED,
@@ -4661,9 +4712,6 @@ sctp_disposition_t sctp_sf_t1_timer_expire(const struct sctp_endpoint *ep,
}
return SCTP_DISPOSITION_CONSUME;
-
-nomem:
- return SCTP_DISPOSITION_NOMEM;
}
/* RFC2960 9.2 If the timer expires, the endpoint must re-send the SHUTDOWN
diff --git a/net/sctp/sm_statetable.c b/net/sctp/sm_statetable.c
index 8967846..75ef104 100644
--- a/net/sctp/sm_statetable.c
+++ b/net/sctp/sm_statetable.c
@@ -783,7 +783,8 @@ static const sctp_sm_table_entry_t other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_
/* SCTP_STATE_COOKIE_WAIT */ \
{.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \
/* SCTP_STATE_COOKIE_ECHOED */ \
- {.fn = sctp_sf_t1_timer_expire, .name = "sctp_sf_t1_timer_expire"}, \
+ {.fn = sctp_sf_t1_cookie_timer_expire, \
+ .name = "sctp_sf_t1_cookie_timer_expire"}, \
/* SCTP_STATE_ESTABLISHED */ \
{.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \
/* SCTP_STATE_SHUTDOWN_PENDING */ \
@@ -802,7 +803,8 @@ static const sctp_sm_table_entry_t other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_
/* SCTP_STATE_CLOSED */ \
{.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \
/* SCTP_STATE_COOKIE_WAIT */ \
- {.fn = sctp_sf_t1_timer_expire, .name = "sctp_sf_t1_timer_expire"}, \
+ {.fn = sctp_sf_t1_init_timer_expire, \
+ .name = "sctp_sf_t1_init_timer_expire"}, \
/* SCTP_STATE_COOKIE_ECHOED */ \
{.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \
/* SCTP_STATE_ESTABLISHED */ \
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index e6926cb..aad55dc 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -262,18 +262,18 @@ static struct sctp_transport *sctp_addr_id2transport(struct sock *sk,
* sockaddr_in6 [RFC 2553]),
* addr_len - the size of the address structure.
*/
-SCTP_STATIC int sctp_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
+SCTP_STATIC int sctp_bind(struct sock *sk, struct sockaddr *addr, int addr_len)
{
int retval = 0;
sctp_lock_sock(sk);
- SCTP_DEBUG_PRINTK("sctp_bind(sk: %p, uaddr: %p, addr_len: %d)\n",
- sk, uaddr, addr_len);
+ SCTP_DEBUG_PRINTK("sctp_bind(sk: %p, addr: %p, addr_len: %d)\n",
+ sk, addr, addr_len);
/* Disallow binding twice. */
if (!sctp_sk(sk)->ep->base.bind_addr.port)
- retval = sctp_do_bind(sk, (union sctp_addr *)uaddr,
+ retval = sctp_do_bind(sk, (union sctp_addr *)addr,
addr_len);
else
retval = -EINVAL;
@@ -318,23 +318,27 @@ SCTP_STATIC int sctp_do_bind(struct sock *sk, union sctp_addr *addr, int len)
unsigned short snum;
int ret = 0;
- SCTP_DEBUG_PRINTK("sctp_do_bind(sk: %p, newaddr: %p, len: %d)\n",
- sk, addr, len);
-
/* Common sockaddr verification. */
af = sctp_sockaddr_af(sp, addr, len);
- if (!af)
+ if (!af) {
+ SCTP_DEBUG_PRINTK("sctp_do_bind(sk: %p, newaddr: %p, len: %d) EINVAL\n",
+ sk, addr, len);
return -EINVAL;
+ }
+
+ snum = ntohs(addr->v4.sin_port);
+
+ SCTP_DEBUG_PRINTK_IPADDR("sctp_do_bind(sk: %p, new addr: ",
+ ", port: %d, new port: %d, len: %d)\n",
+ sk,
+ addr,
+ bp->port, snum,
+ len);
/* PF specific bind() address verification. */
if (!sp->pf->bind_verify(sp, addr))
return -EADDRNOTAVAIL;
- snum= ntohs(addr->v4.sin_port);
-
- SCTP_DEBUG_PRINTK("sctp_do_bind: port: %d, new port: %d\n",
- bp->port, snum);
-
/* We must either be unbound, or bind to the same port. */
if (bp->port && (snum != bp->port)) {
SCTP_DEBUG_PRINTK("sctp_do_bind:"
@@ -816,7 +820,8 @@ out:
*
* Basically do nothing but copying the addresses from user to kernel
* land and invoking either sctp_bindx_add() or sctp_bindx_rem() on the sk.
- * This is used for tunneling the sctp_bindx() request through sctp_setsockopt() * from userspace.
+ * This is used for tunneling the sctp_bindx() request through sctp_setsockopt()
+ * from userspace.
*
* We don't use copy_from_user() for optimization: we first do the
* sanity checks (buffer size -fast- and access check-healthy
@@ -913,6 +918,243 @@ out:
return err;
}
+/* __sctp_connect(struct sock* sk, struct sockaddr *kaddrs, int addrs_size)
+ *
+ * Common routine for handling connect() and sctp_connectx().
+ * Connect will come in with just a single address.
+ */
+static int __sctp_connect(struct sock* sk,
+ struct sockaddr *kaddrs,
+ int addrs_size)
+{
+ struct sctp_sock *sp;
+ struct sctp_endpoint *ep;
+ struct sctp_association *asoc = NULL;
+ struct sctp_association *asoc2;
+ struct sctp_transport *transport;
+ union sctp_addr to;
+ struct sctp_af *af;
+ sctp_scope_t scope;
+ long timeo;
+ int err = 0;
+ int addrcnt = 0;
+ int walk_size = 0;
+ struct sockaddr *sa_addr;
+ void *addr_buf;
+
+ sp = sctp_sk(sk);
+ ep = sp->ep;
+
+ /* connect() cannot be done on a socket that is already in ESTABLISHED
+ * state - UDP-style peeled off socket or a TCP-style socket that
+ * is already connected.
+ * It cannot be done even on a TCP-style listening socket.
+ */
+ if (sctp_sstate(sk, ESTABLISHED) ||
+ (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING))) {
+ err = -EISCONN;
+ goto out_free;
+ }
+
+ /* Walk through the addrs buffer and count the number of addresses. */
+ addr_buf = kaddrs;
+ while (walk_size < addrs_size) {
+ sa_addr = (struct sockaddr *)addr_buf;
+ af = sctp_get_af_specific(sa_addr->sa_family);
+
+ /* If the address family is not supported or if this address
+ * causes the address buffer to overflow return EINVAL.
+ */
+ if (!af || (walk_size + af->sockaddr_len) > addrs_size) {
+ err = -EINVAL;
+ goto out_free;
+ }
+
+ err = sctp_verify_addr(sk, (union sctp_addr *)sa_addr,
+ af->sockaddr_len);
+ if (err)
+ goto out_free;
+
+ memcpy(&to, sa_addr, af->sockaddr_len);
+ to.v4.sin_port = ntohs(to.v4.sin_port);
+
+ /* Check if there already is a matching association on the
+ * endpoint (other than the one created here).
+ */
+ asoc2 = sctp_endpoint_lookup_assoc(ep, &to, &transport);
+ if (asoc2 && asoc2 != asoc) {
+ if (asoc2->state >= SCTP_STATE_ESTABLISHED)
+ err = -EISCONN;
+ else
+ err = -EALREADY;
+ goto out_free;
+ }
+
+ /* If we could not find a matching association on the endpoint,
+ * make sure that there is no peeled-off association matching
+ * the peer address even on another socket.
+ */
+ if (sctp_endpoint_is_peeled_off(ep, &to)) {
+ err = -EADDRNOTAVAIL;
+ goto out_free;
+ }
+
+ if (!asoc) {
+ /* If a bind() or sctp_bindx() is not called prior to
+ * an sctp_connectx() call, the system picks an
+ * ephemeral port and will choose an address set
+ * equivalent to binding with a wildcard address.
+ */
+ if (!ep->base.bind_addr.port) {
+ if (sctp_autobind(sk)) {
+ err = -EAGAIN;
+ goto out_free;
+ }
+ }
+
+ scope = sctp_scope(&to);
+ asoc = sctp_association_new(ep, sk, scope, GFP_KERNEL);
+ if (!asoc) {
+ err = -ENOMEM;
+ goto out_free;
+ }
+ }
+
+ /* Prime the peer's transport structures. */
+ transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL,
+ SCTP_UNKNOWN);
+ if (!transport) {
+ err = -ENOMEM;
+ goto out_free;
+ }
+
+ addrcnt++;
+ addr_buf += af->sockaddr_len;
+ walk_size += af->sockaddr_len;
+ }
+
+ err = sctp_assoc_set_bind_addr_from_ep(asoc, GFP_KERNEL);
+ if (err < 0) {
+ goto out_free;
+ }
+
+ err = sctp_primitive_ASSOCIATE(asoc, NULL);
+ if (err < 0) {
+ goto out_free;
+ }
+
+ /* Initialize sk's dport and daddr for getpeername() */
+ inet_sk(sk)->dport = htons(asoc->peer.port);
+ af = sctp_get_af_specific(to.sa.sa_family);
+ af->to_sk_daddr(&to, sk);
+
+ timeo = sock_sndtimeo(sk, sk->sk_socket->file->f_flags & O_NONBLOCK);
+ err = sctp_wait_for_connect(asoc, &timeo);
+
+ /* Don't free association on exit. */
+ asoc = NULL;
+
+out_free:
+
+ SCTP_DEBUG_PRINTK("About to exit __sctp_connect() free asoc: %p"
+ " kaddrs: %p err: %d\n",
+ asoc, kaddrs, err);
+ if (asoc)
+ sctp_association_free(asoc);
+ return err;
+}
+
+/* Helper for tunneling sctp_connectx() requests through sctp_setsockopt()
+ *
+ * API 8.9
+ * int sctp_connectx(int sd, struct sockaddr *addrs, int addrcnt);
+ *
+ * If sd is an IPv4 socket, the addresses passed must be IPv4 addresses.
+ * If the sd is an IPv6 socket, the addresses passed can either be IPv4
+ * or IPv6 addresses.
+ *
+ * A single address may be specified as INADDR_ANY or IN6ADDR_ANY, see
+ * Section 3.1.2 for this usage.
+ *
+ * addrs is a pointer to an array of one or more socket addresses. Each
+ * address is contained in its appropriate structure (i.e. struct
+ * sockaddr_in or struct sockaddr_in6) the family of the address type
+ * must be used to distengish the address length (note that this
+ * representation is termed a "packed array" of addresses). The caller
+ * specifies the number of addresses in the array with addrcnt.
+ *
+ * On success, sctp_connectx() returns 0. On failure, sctp_connectx() returns
+ * -1, and sets errno to the appropriate error code.
+ *
+ * For SCTP, the port given in each socket address must be the same, or
+ * sctp_connectx() will fail, setting errno to EINVAL.
+ *
+ * An application can use sctp_connectx to initiate an association with
+ * an endpoint that is multi-homed. Much like sctp_bindx() this call
+ * allows a caller to specify multiple addresses at which a peer can be
+ * reached. The way the SCTP stack uses the list of addresses to set up
+ * the association is implementation dependant. This function only
+ * specifies that the stack will try to make use of all the addresses in
+ * the list when needed.
+ *
+ * Note that the list of addresses passed in is only used for setting up
+ * the association. It does not necessarily equal the set of addresses
+ * the peer uses for the resulting association. If the caller wants to
+ * find out the set of peer addresses, it must use sctp_getpaddrs() to
+ * retrieve them after the association has been set up.
+ *
+ * Basically do nothing but copying the addresses from user to kernel
+ * land and invoking either sctp_connectx(). This is used for tunneling
+ * the sctp_connectx() request through sctp_setsockopt() from userspace.
+ *
+ * We don't use copy_from_user() for optimization: we first do the
+ * sanity checks (buffer size -fast- and access check-healthy
+ * pointer); if all of those succeed, then we can alloc the memory
+ * (expensive operation) needed to copy the data to kernel. Then we do
+ * the copying without checking the user space area
+ * (__copy_from_user()).
+ *
+ * On exit there is no need to do sockfd_put(), sys_setsockopt() does
+ * it.
+ *
+ * sk The sk of the socket
+ * addrs The pointer to the addresses in user land
+ * addrssize Size of the addrs buffer
+ *
+ * Returns 0 if ok, <0 errno code on error.
+ */
+SCTP_STATIC int sctp_setsockopt_connectx(struct sock* sk,
+ struct sockaddr __user *addrs,
+ int addrs_size)
+{
+ int err = 0;
+ struct sockaddr *kaddrs;
+
+ SCTP_DEBUG_PRINTK("%s - sk %p addrs %p addrs_size %d\n",
+ __FUNCTION__, sk, addrs, addrs_size);
+
+ if (unlikely(addrs_size <= 0))
+ return -EINVAL;
+
+ /* Check the user passed a healthy pointer. */
+ if (unlikely(!access_ok(VERIFY_READ, addrs, addrs_size)))
+ return -EFAULT;
+
+ /* Alloc space for the address array in kernel memory. */
+ kaddrs = (struct sockaddr *)kmalloc(addrs_size, GFP_KERNEL);
+ if (unlikely(!kaddrs))
+ return -ENOMEM;
+
+ if (__copy_from_user(kaddrs, addrs, addrs_size)) {
+ err = -EFAULT;
+ } else {
+ err = __sctp_connect(sk, kaddrs, addrs_size);
+ }
+
+ kfree(kaddrs);
+ return err;
+}
+
/* API 3.1.4 close() - UDP Style Syntax
* Applications use close() to perform graceful shutdown (as described in
* Section 10.1 of [SCTP]) on ALL the associations currently represented
@@ -1095,7 +1337,7 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk,
sp = sctp_sk(sk);
ep = sp->ep;
- SCTP_DEBUG_PRINTK("Using endpoint: %s.\n", ep->debug_name);
+ SCTP_DEBUG_PRINTK("Using endpoint: %p.\n", ep);
/* We cannot send a message over a TCP-style listening socket. */
if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING)) {
@@ -1306,7 +1548,7 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk,
}
/* Prime the peer's transport structures. */
- transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL);
+ transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL, SCTP_UNKNOWN);
if (!transport) {
err = -ENOMEM;
goto out_free;
@@ -2208,6 +2450,12 @@ SCTP_STATIC int sctp_setsockopt(struct sock *sk, int level, int optname,
optlen, SCTP_BINDX_REM_ADDR);
break;
+ case SCTP_SOCKOPT_CONNECTX:
+ /* 'optlen' is the size of the addresses buffer. */
+ retval = sctp_setsockopt_connectx(sk, (struct sockaddr __user *)optval,
+ optlen);
+ break;
+
case SCTP_DISABLE_FRAGMENTS:
retval = sctp_setsockopt_disable_fragments(sk, optval, optlen);
break;
@@ -2283,112 +2531,29 @@ out_nounlock:
*
* len: the size of the address.
*/
-SCTP_STATIC int sctp_connect(struct sock *sk, struct sockaddr *uaddr,
+SCTP_STATIC int sctp_connect(struct sock *sk, struct sockaddr *addr,
int addr_len)
{
- struct sctp_sock *sp;
- struct sctp_endpoint *ep;
- struct sctp_association *asoc;
- struct sctp_transport *transport;
- union sctp_addr to;
- struct sctp_af *af;
- sctp_scope_t scope;
- long timeo;
int err = 0;
+ struct sctp_af *af;
sctp_lock_sock(sk);
- SCTP_DEBUG_PRINTK("%s - sk: %p, sockaddr: %p, addr_len: %d)\n",
- __FUNCTION__, sk, uaddr, addr_len);
-
- sp = sctp_sk(sk);
- ep = sp->ep;
-
- /* connect() cannot be done on a socket that is already in ESTABLISHED
- * state - UDP-style peeled off socket or a TCP-style socket that
- * is already connected.
- * It cannot be done even on a TCP-style listening socket.
- */
- if (sctp_sstate(sk, ESTABLISHED) ||
- (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING))) {
- err = -EISCONN;
- goto out_unlock;
- }
-
- err = sctp_verify_addr(sk, (union sctp_addr *)uaddr, addr_len);
- if (err)
- goto out_unlock;
-
- if (addr_len > sizeof(to))
- addr_len = sizeof(to);
- memcpy(&to, uaddr, addr_len);
- to.v4.sin_port = ntohs(to.v4.sin_port);
-
- asoc = sctp_endpoint_lookup_assoc(ep, &to, &transport);
- if (asoc) {
- if (asoc->state >= SCTP_STATE_ESTABLISHED)
- err = -EISCONN;
- else
- err = -EALREADY;
- goto out_unlock;
- }
-
- /* If we could not find a matching association on the endpoint,
- * make sure that there is no peeled-off association matching the
- * peer address even on another socket.
- */
- if (sctp_endpoint_is_peeled_off(ep, &to)) {
- err = -EADDRNOTAVAIL;
- goto out_unlock;
- }
-
- /* If a bind() or sctp_bindx() is not called prior to a connect()
- * call, the system picks an ephemeral port and will choose an address
- * set equivalent to binding with a wildcard address.
- */
- if (!ep->base.bind_addr.port) {
- if (sctp_autobind(sk)) {
- err = -EAGAIN;
- goto out_unlock;
- }
- }
-
- scope = sctp_scope(&to);
- asoc = sctp_association_new(ep, sk, scope, GFP_KERNEL);
- if (!asoc) {
- err = -ENOMEM;
- goto out_unlock;
- }
+ SCTP_DEBUG_PRINTK("%s - sk: %p, sockaddr: %p, addr_len: %d\n",
+ __FUNCTION__, sk, addr, addr_len);
- /* Prime the peer's transport structures. */
- transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL);
- if (!transport) {
- sctp_association_free(asoc);
- goto out_unlock;
- }
- err = sctp_assoc_set_bind_addr_from_ep(asoc, GFP_KERNEL);
- if (err < 0) {
- sctp_association_free(asoc);
- goto out_unlock;
- }
-
- err = sctp_primitive_ASSOCIATE(asoc, NULL);
- if (err < 0) {
- sctp_association_free(asoc);
- goto out_unlock;
+ /* Validate addr_len before calling common connect/connectx routine. */
+ af = sctp_get_af_specific(addr->sa_family);
+ if (!af || addr_len < af->sockaddr_len) {
+ err = -EINVAL;
+ } else {
+ /* Pass correct addr len to common routine (so it knows there
+ * is only one address being passed.
+ */
+ err = __sctp_connect(sk, addr, af->sockaddr_len);
}
- /* Initialize sk's dport and daddr for getpeername() */
- inet_sk(sk)->dport = htons(asoc->peer.port);
- af = sctp_get_af_specific(to.sa.sa_family);
- af->to_sk_daddr(&to, sk);
-
- timeo = sock_sndtimeo(sk, sk->sk_socket->file->f_flags & O_NONBLOCK);
- err = sctp_wait_for_connect(asoc, &timeo);
-
-out_unlock:
sctp_release_sock(sk);
-
return err;
}
@@ -2677,12 +2842,15 @@ static int sctp_getsockopt_sctp_status(struct sock *sk, int len,
/* Map ipv4 address into v4-mapped-on-v6 address. */
sctp_get_pf_specific(sk->sk_family)->addr_v4map(sctp_sk(sk),
(union sctp_addr *)&status.sstat_primary.spinfo_address);
- status.sstat_primary.spinfo_state = transport->active;
+ status.sstat_primary.spinfo_state = transport->state;
status.sstat_primary.spinfo_cwnd = transport->cwnd;
status.sstat_primary.spinfo_srtt = transport->srtt;
status.sstat_primary.spinfo_rto = jiffies_to_msecs(transport->rto);
status.sstat_primary.spinfo_mtu = transport->pmtu;
+ if (status.sstat_primary.spinfo_state == SCTP_UNKNOWN)
+ status.sstat_primary.spinfo_state = SCTP_ACTIVE;
+
if (put_user(len, optlen)) {
retval = -EFAULT;
goto out;
@@ -2733,12 +2901,15 @@ static int sctp_getsockopt_peer_addr_info(struct sock *sk, int len,
return -EINVAL;
pinfo.spinfo_assoc_id = sctp_assoc2id(transport->asoc);
- pinfo.spinfo_state = transport->active;
+ pinfo.spinfo_state = transport->state;
pinfo.spinfo_cwnd = transport->cwnd;
pinfo.spinfo_srtt = transport->srtt;
pinfo.spinfo_rto = jiffies_to_msecs(transport->rto);
pinfo.spinfo_mtu = transport->pmtu;
+ if (pinfo.spinfo_state == SCTP_UNKNOWN)
+ pinfo.spinfo_state = SCTP_ACTIVE;
+
if (put_user(len, optlen)) {
retval = -EFAULT;
goto out;
@@ -3591,7 +3762,8 @@ SCTP_STATIC int sctp_getsockopt(struct sock *sk, int level, int optname,
int retval = 0;
int len;
- SCTP_DEBUG_PRINTK("sctp_getsockopt(sk: %p, ...)\n", sk);
+ SCTP_DEBUG_PRINTK("sctp_getsockopt(sk: %p... optname: %d)\n",
+ sk, optname);
/* I can hardly begin to describe how wrong this is. This is
* so broken as to be worse than useless. The API draft
@@ -4596,8 +4768,7 @@ out:
return err;
do_error:
- if (asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1 >=
- asoc->max_init_attempts)
+ if (asoc->init_err_counter + 1 >= asoc->max_init_attempts)
err = -ETIMEDOUT;
else
err = -ECONNREFUSED;
diff --git a/net/sctp/transport.c b/net/sctp/transport.c
index f30882e..0ec0fde 100644
--- a/net/sctp/transport.c
+++ b/net/sctp/transport.c
@@ -83,7 +83,9 @@ static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer,
peer->last_time_used = jiffies;
peer->last_time_ecne_reduced = jiffies;
- peer->active = SCTP_ACTIVE;
+ peer->init_sent_count = 0;
+
+ peer->state = SCTP_ACTIVE;
peer->hb_allowed = 0;
/* Initialize the default path max_retrans. */
diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c
index 9bcec9b..505e2d4 100644
--- a/net/sunrpc/auth.c
+++ b/net/sunrpc/auth.c
@@ -66,10 +66,10 @@ rpcauth_create(rpc_authflavor_t pseudoflavor, struct rpc_clnt *clnt)
u32 flavor = pseudoflavor_to_flavor(pseudoflavor);
if (flavor >= RPC_AUTH_MAXFLAVOR || !(ops = auth_flavors[flavor]))
- return NULL;
+ return ERR_PTR(-EINVAL);
auth = ops->create(clnt, pseudoflavor);
- if (!auth)
- return NULL;
+ if (IS_ERR(auth))
+ return auth;
if (clnt->cl_auth)
rpcauth_destroy(clnt->cl_auth);
clnt->cl_auth = auth;
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index a33b627..2f7b867 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -660,14 +660,16 @@ gss_create(struct rpc_clnt *clnt, rpc_authflavor_t flavor)
{
struct gss_auth *gss_auth;
struct rpc_auth * auth;
+ int err = -ENOMEM; /* XXX? */
dprintk("RPC: creating GSS authenticator for client %p\n",clnt);
if (!try_module_get(THIS_MODULE))
- return NULL;
+ return ERR_PTR(err);
if (!(gss_auth = kmalloc(sizeof(*gss_auth), GFP_KERNEL)))
goto out_dec;
gss_auth->client = clnt;
+ err = -EINVAL;
gss_auth->mech = gss_mech_get_by_pseudoflavor(flavor);
if (!gss_auth->mech) {
printk(KERN_WARNING "%s: Pseudoflavor %d not found!",
@@ -675,9 +677,8 @@ gss_create(struct rpc_clnt *clnt, rpc_authflavor_t flavor)
goto err_free;
}
gss_auth->service = gss_pseudoflavor_to_service(gss_auth->mech, flavor);
- /* FIXME: Will go away once privacy support is merged in */
- if (gss_auth->service == RPC_GSS_SVC_PRIVACY)
- gss_auth->service = RPC_GSS_SVC_INTEGRITY;
+ if (gss_auth->service == 0)
+ goto err_put_mech;
INIT_LIST_HEAD(&gss_auth->upcalls);
spin_lock_init(&gss_auth->lock);
auth = &gss_auth->rpc_auth;
@@ -687,15 +688,18 @@ gss_create(struct rpc_clnt *clnt, rpc_authflavor_t flavor)
auth->au_flavor = flavor;
atomic_set(&auth->au_count, 1);
- if (rpcauth_init_credcache(auth, GSS_CRED_EXPIRE) < 0)
+ err = rpcauth_init_credcache(auth, GSS_CRED_EXPIRE);
+ if (err)
goto err_put_mech;
snprintf(gss_auth->path, sizeof(gss_auth->path), "%s/%s",
clnt->cl_pathname,
gss_auth->mech->gm_name);
gss_auth->dentry = rpc_mkpipe(gss_auth->path, clnt, &gss_upcall_ops, RPC_PIPE_WAIT_FOR_OPEN);
- if (IS_ERR(gss_auth->dentry))
+ if (IS_ERR(gss_auth->dentry)) {
+ err = PTR_ERR(gss_auth->dentry);
goto err_put_mech;
+ }
return auth;
err_put_mech:
@@ -704,7 +708,7 @@ err_free:
kfree(gss_auth);
out_dec:
module_put(THIS_MODULE);
- return NULL;
+ return ERR_PTR(err);
}
static void
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 02bc029..f17e615 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -97,12 +97,13 @@ rpc_setup_pipedir(struct rpc_clnt *clnt, char *dir_name)
* made to sleep too long.
*/
struct rpc_clnt *
-rpc_create_client(struct rpc_xprt *xprt, char *servname,
+rpc_new_client(struct rpc_xprt *xprt, char *servname,
struct rpc_program *program, u32 vers,
rpc_authflavor_t flavor)
{
struct rpc_version *version;
struct rpc_clnt *clnt = NULL;
+ struct rpc_auth *auth;
int err;
int len;
@@ -157,10 +158,11 @@ rpc_create_client(struct rpc_xprt *xprt, char *servname,
if (err < 0)
goto out_no_path;
- err = -ENOMEM;
- if (!rpcauth_create(flavor, clnt)) {
+ auth = rpcauth_create(flavor, clnt);
+ if (IS_ERR(auth)) {
printk(KERN_INFO "RPC: Couldn't create auth handle (flavor %u)\n",
flavor);
+ err = PTR_ERR(auth);
goto out_no_auth;
}
@@ -178,6 +180,37 @@ out_no_path:
kfree(clnt->cl_server);
kfree(clnt);
out_err:
+ xprt_destroy(xprt);
+ return ERR_PTR(err);
+}
+
+/**
+ * Create an RPC client
+ * @xprt - pointer to xprt struct
+ * @servname - name of server
+ * @info - rpc_program
+ * @version - rpc_program version
+ * @authflavor - rpc_auth flavour to use
+ *
+ * Creates an RPC client structure, then pings the server in order to
+ * determine if it is up, and if it supports this program and version.
+ *
+ * This function should never be called by asynchronous tasks such as
+ * the portmapper.
+ */
+struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname,
+ struct rpc_program *info, u32 version, rpc_authflavor_t authflavor)
+{
+ struct rpc_clnt *clnt;
+ int err;
+
+ clnt = rpc_new_client(xprt, servname, info, version, authflavor);
+ if (IS_ERR(clnt))
+ return clnt;
+ err = rpc_ping(clnt, RPC_TASK_SOFT|RPC_TASK_NOINTR);
+ if (err == 0)
+ return clnt;
+ rpc_shutdown_client(clnt);
return ERR_PTR(err);
}
@@ -208,6 +241,8 @@ rpc_clone_client(struct rpc_clnt *clnt)
rpc_init_rtt(&new->cl_rtt_default, clnt->cl_xprt->timeout.to_initval);
if (new->cl_auth)
atomic_inc(&new->cl_auth->au_count);
+ new->cl_pmap = &new->cl_pmap_default;
+ rpc_init_wait_queue(&new->cl_pmap_default.pm_bindwait, "bindwait");
return new;
out_no_clnt:
printk(KERN_INFO "RPC: out of memory in %s\n", __FUNCTION__);
@@ -296,6 +331,44 @@ rpc_release_client(struct rpc_clnt *clnt)
rpc_destroy_client(clnt);
}
+/**
+ * rpc_bind_new_program - bind a new RPC program to an existing client
+ * @old - old rpc_client
+ * @program - rpc program to set
+ * @vers - rpc program version
+ *
+ * Clones the rpc client and sets up a new RPC program. This is mainly
+ * of use for enabling different RPC programs to share the same transport.
+ * The Sun NFSv2/v3 ACL protocol can do this.
+ */
+struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *old,
+ struct rpc_program *program,
+ int vers)
+{
+ struct rpc_clnt *clnt;
+ struct rpc_version *version;
+ int err;
+
+ BUG_ON(vers >= program->nrvers || !program->version[vers]);
+ version = program->version[vers];
+ clnt = rpc_clone_client(old);
+ if (IS_ERR(clnt))
+ goto out;
+ clnt->cl_procinfo = version->procs;
+ clnt->cl_maxproc = version->nrprocs;
+ clnt->cl_protname = program->name;
+ clnt->cl_prog = program->number;
+ clnt->cl_vers = version->number;
+ clnt->cl_stats = program->stats;
+ err = rpc_ping(clnt, RPC_TASK_SOFT|RPC_TASK_NOINTR);
+ if (err != 0) {
+ rpc_shutdown_client(clnt);
+ clnt = ERR_PTR(err);
+ }
+out:
+ return clnt;
+}
+
/*
* Default callback for async RPC calls
*/
@@ -305,38 +378,41 @@ rpc_default_callback(struct rpc_task *task)
}
/*
- * Export the signal mask handling for aysnchronous code that
+ * Export the signal mask handling for synchronous code that
* sleeps on RPC calls
*/
+#define RPC_INTR_SIGNALS (sigmask(SIGINT) | sigmask(SIGQUIT) | sigmask(SIGKILL))
+static void rpc_save_sigmask(sigset_t *oldset, int intr)
+{
+ unsigned long sigallow = 0;
+ sigset_t sigmask;
+
+ /* Block all signals except those listed in sigallow */
+ if (intr)
+ sigallow |= RPC_INTR_SIGNALS;
+ siginitsetinv(&sigmask, sigallow);
+ sigprocmask(SIG_BLOCK, &sigmask, oldset);
+}
+
+static inline void rpc_task_sigmask(struct rpc_task *task, sigset_t *oldset)
+{
+ rpc_save_sigmask(oldset, !RPC_TASK_UNINTERRUPTIBLE(task));
+}
+
+static inline void rpc_restore_sigmask(sigset_t *oldset)
+{
+ sigprocmask(SIG_SETMASK, oldset, NULL);
+}
+
void rpc_clnt_sigmask(struct rpc_clnt *clnt, sigset_t *oldset)
{
- unsigned long sigallow = sigmask(SIGKILL);
- unsigned long irqflags;
-
- /* Turn off various signals */
- if (clnt->cl_intr) {
- struct k_sigaction *action = current->sighand->action;
- if (action[SIGINT-1].sa.sa_handler == SIG_DFL)
- sigallow |= sigmask(SIGINT);
- if (action[SIGQUIT-1].sa.sa_handler == SIG_DFL)
- sigallow |= sigmask(SIGQUIT);
- }
- spin_lock_irqsave(&current->sighand->siglock, irqflags);
- *oldset = current->blocked;
- siginitsetinv(&current->blocked, sigallow & ~oldset->sig[0]);
- recalc_sigpending();
- spin_unlock_irqrestore(&current->sighand->siglock, irqflags);
+ rpc_save_sigmask(oldset, clnt->cl_intr);
}
void rpc_clnt_sigunmask(struct rpc_clnt *clnt, sigset_t *oldset)
{
- unsigned long irqflags;
-
- spin_lock_irqsave(&current->sighand->siglock, irqflags);
- current->blocked = *oldset;
- recalc_sigpending();
- spin_unlock_irqrestore(&current->sighand->siglock, irqflags);
+ rpc_restore_sigmask(oldset);
}
/*
@@ -354,26 +430,26 @@ int rpc_call_sync(struct rpc_clnt *clnt, struct rpc_message *msg, int flags)
BUG_ON(flags & RPC_TASK_ASYNC);
- rpc_clnt_sigmask(clnt, &oldset);
-
status = -ENOMEM;
task = rpc_new_task(clnt, NULL, flags);
if (task == NULL)
goto out;
+ /* Mask signals on RPC calls _and_ GSS_AUTH upcalls */
+ rpc_task_sigmask(task, &oldset);
+
rpc_call_setup(task, msg, 0);
/* Set up the call info struct and execute the task */
- if (task->tk_status == 0)
+ if (task->tk_status == 0) {
status = rpc_execute(task);
- else {
+ } else {
status = task->tk_status;
rpc_release_task(task);
}
+ rpc_restore_sigmask(&oldset);
out:
- rpc_clnt_sigunmask(clnt, &oldset);
-
return status;
}
@@ -394,8 +470,6 @@ rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg, int flags,
flags |= RPC_TASK_ASYNC;
- rpc_clnt_sigmask(clnt, &oldset);
-
/* Create/initialize a new RPC task */
if (!callback)
callback = rpc_default_callback;
@@ -404,6 +478,9 @@ rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg, int flags,
goto out;
task->tk_calldata = data;
+ /* Mask signals on GSS_AUTH upcalls */
+ rpc_task_sigmask(task, &oldset);
+
rpc_call_setup(task, msg, 0);
/* Set up the call info struct and execute the task */
@@ -413,9 +490,8 @@ rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg, int flags,
else
rpc_release_task(task);
+ rpc_restore_sigmask(&oldset);
out:
- rpc_clnt_sigunmask(clnt, &oldset);
-
return status;
}
@@ -593,7 +669,7 @@ call_allocate(struct rpc_task *task)
return;
printk(KERN_INFO "RPC: buffer allocation failed for task %p\n", task);
- if (RPC_IS_ASYNC(task) || !(task->tk_client->cl_intr && signalled())) {
+ if (RPC_IS_ASYNC(task) || !signalled()) {
xprt_release(task);
task->tk_action = call_reserve;
rpc_delay(task, HZ>>4);
@@ -957,7 +1033,9 @@ call_header(struct rpc_task *task)
*p++ = htonl(clnt->cl_prog); /* program number */
*p++ = htonl(clnt->cl_vers); /* program version */
*p++ = htonl(task->tk_msg.rpc_proc->p_proc); /* procedure */
- return rpcauth_marshcred(task, p);
+ p = rpcauth_marshcred(task, p);
+ req->rq_slen = xdr_adjust_iovec(&req->rq_svec[0], p);
+ return p;
}
/*
@@ -986,10 +1064,11 @@ call_verify(struct rpc_task *task)
case RPC_AUTH_ERROR:
break;
case RPC_MISMATCH:
- printk(KERN_WARNING "%s: RPC call version mismatch!\n", __FUNCTION__);
- goto out_eio;
+ dprintk("%s: RPC call version mismatch!\n", __FUNCTION__);
+ error = -EPROTONOSUPPORT;
+ goto out_err;
default:
- printk(KERN_WARNING "%s: RPC call rejected, unknown error: %x\n", __FUNCTION__, n);
+ dprintk("%s: RPC call rejected, unknown error: %x\n", __FUNCTION__, n);
goto out_eio;
}
if (--len < 0)
@@ -1040,23 +1119,26 @@ call_verify(struct rpc_task *task)
case RPC_SUCCESS:
return p;
case RPC_PROG_UNAVAIL:
- printk(KERN_WARNING "RPC: call_verify: program %u is unsupported by server %s\n",
+ dprintk("RPC: call_verify: program %u is unsupported by server %s\n",
(unsigned int)task->tk_client->cl_prog,
task->tk_client->cl_server);
- goto out_eio;
+ error = -EPFNOSUPPORT;
+ goto out_err;
case RPC_PROG_MISMATCH:
- printk(KERN_WARNING "RPC: call_verify: program %u, version %u unsupported by server %s\n",
+ dprintk("RPC: call_verify: program %u, version %u unsupported by server %s\n",
(unsigned int)task->tk_client->cl_prog,
(unsigned int)task->tk_client->cl_vers,
task->tk_client->cl_server);
- goto out_eio;
+ error = -EPROTONOSUPPORT;
+ goto out_err;
case RPC_PROC_UNAVAIL:
- printk(KERN_WARNING "RPC: call_verify: proc %p unsupported by program %u, version %u on server %s\n",
+ dprintk("RPC: call_verify: proc %p unsupported by program %u, version %u on server %s\n",
task->tk_msg.rpc_proc,
task->tk_client->cl_prog,
task->tk_client->cl_vers,
task->tk_client->cl_server);
- goto out_eio;
+ error = -EOPNOTSUPP;
+ goto out_err;
case RPC_GARBAGE_ARGS:
dprintk("RPC: %4d %s: server saw garbage\n", task->tk_pid, __FUNCTION__);
break; /* retry */
@@ -1069,7 +1151,7 @@ out_retry:
task->tk_client->cl_stats->rpcgarbage++;
if (task->tk_garb_retry) {
task->tk_garb_retry--;
- dprintk(KERN_WARNING "RPC %s: retrying %4d\n", __FUNCTION__, task->tk_pid);
+ dprintk("RPC %s: retrying %4d\n", __FUNCTION__, task->tk_pid);
task->tk_action = call_bind;
return NULL;
}
@@ -1083,3 +1165,30 @@ out_overflow:
printk(KERN_WARNING "RPC %s: server reply was truncated.\n", __FUNCTION__);
goto out_retry;
}
+
+static int rpcproc_encode_null(void *rqstp, u32 *data, void *obj)
+{
+ return 0;
+}
+
+static int rpcproc_decode_null(void *rqstp, u32 *data, void *obj)
+{
+ return 0;
+}
+
+static struct rpc_procinfo rpcproc_null = {
+ .p_encode = rpcproc_encode_null,
+ .p_decode = rpcproc_decode_null,
+};
+
+int rpc_ping(struct rpc_clnt *clnt, int flags)
+{
+ struct rpc_message msg = {
+ .rpc_proc = &rpcproc_null,
+ };
+ int err;
+ msg.rpc_cred = authnull_ops.lookup_cred(NULL, NULL, 0);
+ err = rpc_call_sync(clnt, &msg, flags);
+ put_rpccred(msg.rpc_cred);
+ return err;
+}
diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c
index d0b1d2c..4e81f27 100644
--- a/net/sunrpc/pmap_clnt.c
+++ b/net/sunrpc/pmap_clnt.c
@@ -53,6 +53,9 @@ rpc_getport(struct rpc_task *task, struct rpc_clnt *clnt)
task->tk_pid, clnt->cl_server,
map->pm_prog, map->pm_vers, map->pm_prot);
+ /* Autobind on cloned rpc clients is discouraged */
+ BUG_ON(clnt->cl_parent != clnt);
+
spin_lock(&pmap_lock);
if (map->pm_binding) {
rpc_sleep_on(&map->pm_bindwait, task, NULL, NULL);
@@ -207,12 +210,10 @@ pmap_create(char *hostname, struct sockaddr_in *srvaddr, int proto)
xprt->addr.sin_port = htons(RPC_PMAP_PORT);
/* printk("pmap: create clnt\n"); */
- clnt = rpc_create_client(xprt, hostname,
+ clnt = rpc_new_client(xprt, hostname,
&pmap_program, RPC_PMAP_VERSION,
RPC_AUTH_UNIX);
- if (IS_ERR(clnt)) {
- xprt_destroy(xprt);
- } else {
+ if (!IS_ERR(clnt)) {
clnt->cl_softrtry = 1;
clnt->cl_chatty = 1;
clnt->cl_oneshot = 1;
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index c06614d..2d9eb7f 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -290,7 +290,7 @@ static void rpc_make_runnable(struct rpc_task *task)
return;
}
} else
- wake_up(&task->u.tk_wait.waitq);
+ wake_up_bit(&task->tk_runstate, RPC_TASK_QUEUED);
}
/*
@@ -555,6 +555,38 @@ __rpc_atrun(struct rpc_task *task)
}
/*
+ * Helper that calls task->tk_exit if it exists and then returns
+ * true if we should exit __rpc_execute.
+ */
+static inline int __rpc_do_exit(struct rpc_task *task)
+{
+ if (task->tk_exit != NULL) {
+ lock_kernel();
+ task->tk_exit(task);
+ unlock_kernel();
+ /* If tk_action is non-null, we should restart the call */
+ if (task->tk_action != NULL) {
+ if (!RPC_ASSASSINATED(task)) {
+ /* Release RPC slot and buffer memory */
+ xprt_release(task);
+ rpc_free(task);
+ return 0;
+ }
+ printk(KERN_ERR "RPC: dead task tried to walk away.\n");
+ }
+ }
+ return 1;
+}
+
+static int rpc_wait_bit_interruptible(void *word)
+{
+ if (signal_pending(current))
+ return -ERESTARTSYS;
+ schedule();
+ return 0;
+}
+
+/*
* This is the RPC `scheduler' (or rather, the finite state machine).
*/
static int __rpc_execute(struct rpc_task *task)
@@ -566,8 +598,7 @@ static int __rpc_execute(struct rpc_task *task)
BUG_ON(RPC_IS_QUEUED(task));
- restarted:
- while (1) {
+ for (;;) {
/*
* Garbage collection of pending timers...
*/
@@ -600,11 +631,12 @@ static int __rpc_execute(struct rpc_task *task)
* by someone else.
*/
if (!RPC_IS_QUEUED(task)) {
- if (!task->tk_action)
+ if (task->tk_action != NULL) {
+ lock_kernel();
+ task->tk_action(task);
+ unlock_kernel();
+ } else if (__rpc_do_exit(task))
break;
- lock_kernel();
- task->tk_action(task);
- unlock_kernel();
}
/*
@@ -624,44 +656,26 @@ static int __rpc_execute(struct rpc_task *task)
/* sync task: sleep here */
dprintk("RPC: %4d sync task going to sleep\n", task->tk_pid);
- if (RPC_TASK_UNINTERRUPTIBLE(task)) {
- __wait_event(task->u.tk_wait.waitq, !RPC_IS_QUEUED(task));
- } else {
- __wait_event_interruptible(task->u.tk_wait.waitq, !RPC_IS_QUEUED(task), status);
+ /* Note: Caller should be using rpc_clnt_sigmask() */
+ status = out_of_line_wait_on_bit(&task->tk_runstate,
+ RPC_TASK_QUEUED, rpc_wait_bit_interruptible,
+ TASK_INTERRUPTIBLE);
+ if (status == -ERESTARTSYS) {
/*
* When a sync task receives a signal, it exits with
* -ERESTARTSYS. In order to catch any callbacks that
* clean up after sleeping on some queue, we don't
* break the loop here, but go around once more.
*/
- if (status == -ERESTARTSYS) {
- dprintk("RPC: %4d got signal\n", task->tk_pid);
- task->tk_flags |= RPC_TASK_KILLED;
- rpc_exit(task, -ERESTARTSYS);
- rpc_wake_up_task(task);
- }
+ dprintk("RPC: %4d got signal\n", task->tk_pid);
+ task->tk_flags |= RPC_TASK_KILLED;
+ rpc_exit(task, -ERESTARTSYS);
+ rpc_wake_up_task(task);
}
rpc_set_running(task);
dprintk("RPC: %4d sync task resuming\n", task->tk_pid);
}
- if (task->tk_exit) {
- lock_kernel();
- task->tk_exit(task);
- unlock_kernel();
- /* If tk_action is non-null, the user wants us to restart */
- if (task->tk_action) {
- if (!RPC_ASSASSINATED(task)) {
- /* Release RPC slot and buffer memory */
- if (task->tk_rqstp)
- xprt_release(task);
- rpc_free(task);
- goto restarted;
- }
- printk(KERN_ERR "RPC: dead task tries to walk away.\n");
- }
- }
-
dprintk("RPC: %4d exit() = %d\n", task->tk_pid, task->tk_status);
status = task->tk_status;
@@ -759,8 +773,6 @@ void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, rpc_action call
/* Initialize workqueue for async tasks */
task->tk_workqueue = rpciod_workqueue;
- if (!RPC_IS_ASYNC(task))
- init_waitqueue_head(&task->u.tk_wait.waitq);
if (clnt) {
atomic_inc(&clnt->cl_users);
diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index d4f26bf..32e8acb 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -42,6 +42,7 @@ EXPORT_SYMBOL(rpc_release_task);
/* RPC client functions */
EXPORT_SYMBOL(rpc_create_client);
EXPORT_SYMBOL(rpc_clone_client);
+EXPORT_SYMBOL(rpc_bind_new_program);
EXPORT_SYMBOL(rpc_destroy_client);
EXPORT_SYMBOL(rpc_shutdown_client);
EXPORT_SYMBOL(rpc_release_client);
@@ -61,7 +62,6 @@ EXPORT_SYMBOL(rpc_mkpipe);
/* Client transport */
EXPORT_SYMBOL(xprt_create_proto);
-EXPORT_SYMBOL(xprt_destroy);
EXPORT_SYMBOL(xprt_set_timeout);
EXPORT_SYMBOL(xprt_udp_slot_table_entries);
EXPORT_SYMBOL(xprt_tcp_slot_table_entries);
@@ -129,6 +129,10 @@ EXPORT_SYMBOL(xdr_encode_netobj);
EXPORT_SYMBOL(xdr_encode_pages);
EXPORT_SYMBOL(xdr_inline_pages);
EXPORT_SYMBOL(xdr_shift_buf);
+EXPORT_SYMBOL(xdr_encode_word);
+EXPORT_SYMBOL(xdr_decode_word);
+EXPORT_SYMBOL(xdr_encode_array2);
+EXPORT_SYMBOL(xdr_decode_array2);
EXPORT_SYMBOL(xdr_buf_from_iov);
EXPORT_SYMBOL(xdr_buf_subsegment);
EXPORT_SYMBOL(xdr_buf_read_netobj);
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index bb2d99f..e9bd912 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -35,20 +35,24 @@ svc_create(struct svc_program *prog, unsigned int bufsize)
if (!(serv = (struct svc_serv *) kmalloc(sizeof(*serv), GFP_KERNEL)))
return NULL;
memset(serv, 0, sizeof(*serv));
+ serv->sv_name = prog->pg_name;
serv->sv_program = prog;
serv->sv_nrthreads = 1;
serv->sv_stats = prog->pg_stats;
serv->sv_bufsz = bufsize? bufsize : 4096;
- prog->pg_lovers = prog->pg_nvers-1;
xdrsize = 0;
- for (vers=0; vers<prog->pg_nvers ; vers++)
- if (prog->pg_vers[vers]) {
- prog->pg_hivers = vers;
- if (prog->pg_lovers > vers)
- prog->pg_lovers = vers;
- if (prog->pg_vers[vers]->vs_xdrsize > xdrsize)
- xdrsize = prog->pg_vers[vers]->vs_xdrsize;
- }
+ while (prog) {
+ prog->pg_lovers = prog->pg_nvers-1;
+ for (vers=0; vers<prog->pg_nvers ; vers++)
+ if (prog->pg_vers[vers]) {
+ prog->pg_hivers = vers;
+ if (prog->pg_lovers > vers)
+ prog->pg_lovers = vers;
+ if (prog->pg_vers[vers]->vs_xdrsize > xdrsize)
+ xdrsize = prog->pg_vers[vers]->vs_xdrsize;
+ }
+ prog = prog->pg_next;
+ }
serv->sv_xdrsize = xdrsize;
INIT_LIST_HEAD(&serv->sv_threads);
INIT_LIST_HEAD(&serv->sv_sockets);
@@ -56,8 +60,6 @@ svc_create(struct svc_program *prog, unsigned int bufsize)
INIT_LIST_HEAD(&serv->sv_permsocks);
spin_lock_init(&serv->sv_lock);
- serv->sv_name = prog->pg_name;
-
/* Remove any stale portmap registrations */
svc_register(serv, 0, 0);
@@ -281,6 +283,7 @@ svc_process(struct svc_serv *serv, struct svc_rqst *rqstp)
rqstp->rq_res.len = 0;
rqstp->rq_res.page_base = 0;
rqstp->rq_res.page_len = 0;
+ rqstp->rq_res.buflen = PAGE_SIZE;
rqstp->rq_res.tail[0].iov_len = 0;
/* tcp needs a space for the record length... */
if (rqstp->rq_prot == IPPROTO_TCP)
@@ -338,7 +341,10 @@ svc_process(struct svc_serv *serv, struct svc_rqst *rqstp)
goto sendit;
}
- if (prog != progp->pg_prog)
+ for (progp = serv->sv_program; progp; progp = progp->pg_next)
+ if (prog == progp->pg_prog)
+ break;
+ if (progp == NULL)
goto err_bad_prog;
if (vers >= progp->pg_nvers ||
@@ -451,11 +457,7 @@ err_bad_auth:
goto sendit;
err_bad_prog:
-#ifdef RPC_PARANOIA
- if (prog != 100227 || progp->pg_prog != 100003)
- printk("svc: unknown program %d (me %d)\n", prog, progp->pg_prog);
- /* else it is just a Solaris client seeing if ACLs are supported */
-#endif
+ dprintk("svc: unknown program %d\n", prog);
serv->sv_stats->rpcbadfmt++;
svc_putu32(resv, rpc_prog_unavail);
goto sendit;
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 67b9f035..8a4d9c1 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -176,21 +176,23 @@ xdr_inline_pages(struct xdr_buf *xdr, unsigned int offset,
xdr->buflen += len;
}
-void
+ssize_t
xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base,
skb_reader_t *desc,
skb_read_actor_t copy_actor)
{
struct page **ppage = xdr->pages;
unsigned int len, pglen = xdr->page_len;
+ ssize_t copied = 0;
int ret;
len = xdr->head[0].iov_len;
if (base < len) {
len -= base;
ret = copy_actor(desc, (char *)xdr->head[0].iov_base + base, len);
+ copied += ret;
if (ret != len || !desc->count)
- return;
+ goto out;
base = 0;
} else
base -= len;
@@ -210,6 +212,17 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base,
do {
char *kaddr;
+ /* ACL likes to be lazy in allocating pages - ACLs
+ * are small by default but can get huge. */
+ if (unlikely(*ppage == NULL)) {
+ *ppage = alloc_page(GFP_ATOMIC);
+ if (unlikely(*ppage == NULL)) {
+ if (copied == 0)
+ copied = -ENOMEM;
+ goto out;
+ }
+ }
+
len = PAGE_CACHE_SIZE;
kaddr = kmap_atomic(*ppage, KM_SKB_SUNRPC_DATA);
if (base) {
@@ -225,14 +238,17 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base,
}
flush_dcache_page(*ppage);
kunmap_atomic(kaddr, KM_SKB_SUNRPC_DATA);
+ copied += ret;
if (ret != len || !desc->count)
- return;
+ goto out;
ppage++;
} while ((pglen -= len) != 0);
copy_tail:
len = xdr->tail[0].iov_len;
if (base < len)
- copy_actor(desc, (char *)xdr->tail[0].iov_base + base, len - base);
+ copied += copy_actor(desc, (char *)xdr->tail[0].iov_base + base, len - base);
+out:
+ return copied;
}
@@ -616,12 +632,24 @@ xdr_shift_buf(struct xdr_buf *buf, size_t len)
void xdr_init_encode(struct xdr_stream *xdr, struct xdr_buf *buf, uint32_t *p)
{
struct kvec *iov = buf->head;
+ int scratch_len = buf->buflen - buf->page_len - buf->tail[0].iov_len;
+ BUG_ON(scratch_len < 0);
xdr->buf = buf;
xdr->iov = iov;
- xdr->end = (uint32_t *)((char *)iov->iov_base + iov->iov_len);
- buf->len = iov->iov_len = (char *)p - (char *)iov->iov_base;
- xdr->p = p;
+ xdr->p = (uint32_t *)((char *)iov->iov_base + iov->iov_len);
+ xdr->end = (uint32_t *)((char *)iov->iov_base + scratch_len);
+ BUG_ON(iov->iov_len > scratch_len);
+
+ if (p != xdr->p && p != NULL) {
+ size_t len;
+
+ BUG_ON(p < xdr->p || p > xdr->end);
+ len = (char *)p - (char *)xdr->p;
+ xdr->p = p;
+ buf->len += len;
+ iov->iov_len += len;
+ }
}
EXPORT_SYMBOL(xdr_init_encode);
@@ -859,8 +887,34 @@ out:
return status;
}
-static int
-read_u32_from_xdr_buf(struct xdr_buf *buf, int base, u32 *obj)
+/* obj is assumed to point to allocated memory of size at least len: */
+int
+write_bytes_to_xdr_buf(struct xdr_buf *buf, int base, void *obj, int len)
+{
+ struct xdr_buf subbuf;
+ int this_len;
+ int status;
+
+ status = xdr_buf_subsegment(buf, &subbuf, base, len);
+ if (status)
+ goto out;
+ this_len = min(len, (int)subbuf.head[0].iov_len);
+ memcpy(subbuf.head[0].iov_base, obj, this_len);
+ len -= this_len;
+ obj += this_len;
+ this_len = min(len, (int)subbuf.page_len);
+ if (this_len)
+ _copy_to_pages(subbuf.pages, subbuf.page_base, obj, this_len);
+ len -= this_len;
+ obj += this_len;
+ this_len = min(len, (int)subbuf.tail[0].iov_len);
+ memcpy(subbuf.tail[0].iov_base, obj, this_len);
+out:
+ return status;
+}
+
+int
+xdr_decode_word(struct xdr_buf *buf, int base, u32 *obj)
{
u32 raw;
int status;
@@ -872,6 +926,14 @@ read_u32_from_xdr_buf(struct xdr_buf *buf, int base, u32 *obj)
return 0;
}
+int
+xdr_encode_word(struct xdr_buf *buf, int base, u32 obj)
+{
+ u32 raw = htonl(obj);
+
+ return write_bytes_to_xdr_buf(buf, base, &raw, sizeof(obj));
+}
+
/* If the netobj starting offset bytes from the start of xdr_buf is contained
* entirely in the head or the tail, set object to point to it; otherwise
* try to find space for it at the end of the tail, copy it there, and
@@ -882,7 +944,7 @@ xdr_buf_read_netobj(struct xdr_buf *buf, struct xdr_netobj *obj, int offset)
u32 tail_offset = buf->head[0].iov_len + buf->page_len;
u32 obj_end_offset;
- if (read_u32_from_xdr_buf(buf, offset, &obj->len))
+ if (xdr_decode_word(buf, offset, &obj->len))
goto out;
obj_end_offset = offset + 4 + obj->len;
@@ -915,3 +977,219 @@ xdr_buf_read_netobj(struct xdr_buf *buf, struct xdr_netobj *obj, int offset)
out:
return -1;
}
+
+/* Returns 0 on success, or else a negative error code. */
+static int
+xdr_xcode_array2(struct xdr_buf *buf, unsigned int base,
+ struct xdr_array2_desc *desc, int encode)
+{
+ char *elem = NULL, *c;
+ unsigned int copied = 0, todo, avail_here;
+ struct page **ppages = NULL;
+ int err;
+
+ if (encode) {
+ if (xdr_encode_word(buf, base, desc->array_len) != 0)
+ return -EINVAL;
+ } else {
+ if (xdr_decode_word(buf, base, &desc->array_len) != 0 ||
+ (unsigned long) base + 4 + desc->array_len *
+ desc->elem_size > buf->len)
+ return -EINVAL;
+ }
+ base += 4;
+
+ if (!desc->xcode)
+ return 0;
+
+ todo = desc->array_len * desc->elem_size;
+
+ /* process head */
+ if (todo && base < buf->head->iov_len) {
+ c = buf->head->iov_base + base;
+ avail_here = min_t(unsigned int, todo,
+ buf->head->iov_len - base);
+ todo -= avail_here;
+
+ while (avail_here >= desc->elem_size) {
+ err = desc->xcode(desc, c);
+ if (err)
+ goto out;
+ c += desc->elem_size;
+ avail_here -= desc->elem_size;
+ }
+ if (avail_here) {
+ if (!elem) {
+ elem = kmalloc(desc->elem_size, GFP_KERNEL);
+ err = -ENOMEM;
+ if (!elem)
+ goto out;
+ }
+ if (encode) {
+ err = desc->xcode(desc, elem);
+ if (err)
+ goto out;
+ memcpy(c, elem, avail_here);
+ } else
+ memcpy(elem, c, avail_here);
+ copied = avail_here;
+ }
+ base = buf->head->iov_len; /* align to start of pages */
+ }
+
+ /* process pages array */
+ base -= buf->head->iov_len;
+ if (todo && base < buf->page_len) {
+ unsigned int avail_page;
+
+ avail_here = min(todo, buf->page_len - base);
+ todo -= avail_here;
+
+ base += buf->page_base;
+ ppages = buf->pages + (base >> PAGE_CACHE_SHIFT);
+ base &= ~PAGE_CACHE_MASK;
+ avail_page = min_t(unsigned int, PAGE_CACHE_SIZE - base,
+ avail_here);
+ c = kmap(*ppages) + base;
+
+ while (avail_here) {
+ avail_here -= avail_page;
+ if (copied || avail_page < desc->elem_size) {
+ unsigned int l = min(avail_page,
+ desc->elem_size - copied);
+ if (!elem) {
+ elem = kmalloc(desc->elem_size,
+ GFP_KERNEL);
+ err = -ENOMEM;
+ if (!elem)
+ goto out;
+ }
+ if (encode) {
+ if (!copied) {
+ err = desc->xcode(desc, elem);
+ if (err)
+ goto out;
+ }
+ memcpy(c, elem + copied, l);
+ copied += l;
+ if (copied == desc->elem_size)
+ copied = 0;
+ } else {
+ memcpy(elem + copied, c, l);
+ copied += l;
+ if (copied == desc->elem_size) {
+ err = desc->xcode(desc, elem);
+ if (err)
+ goto out;
+ copied = 0;
+ }
+ }
+ avail_page -= l;
+ c += l;
+ }
+ while (avail_page >= desc->elem_size) {
+ err = desc->xcode(desc, c);
+ if (err)
+ goto out;
+ c += desc->elem_size;
+ avail_page -= desc->elem_size;
+ }
+ if (avail_page) {
+ unsigned int l = min(avail_page,
+ desc->elem_size - copied);
+ if (!elem) {
+ elem = kmalloc(desc->elem_size,
+ GFP_KERNEL);
+ err = -ENOMEM;
+ if (!elem)
+ goto out;
+ }
+ if (encode) {
+ if (!copied) {
+ err = desc->xcode(desc, elem);
+ if (err)
+ goto out;
+ }
+ memcpy(c, elem + copied, l);
+ copied += l;
+ if (copied == desc->elem_size)
+ copied = 0;
+ } else {
+ memcpy(elem + copied, c, l);
+ copied += l;
+ if (copied == desc->elem_size) {
+ err = desc->xcode(desc, elem);
+ if (err)
+ goto out;
+ copied = 0;
+ }
+ }
+ }
+ if (avail_here) {
+ kunmap(*ppages);
+ ppages++;
+ c = kmap(*ppages);
+ }
+
+ avail_page = min(avail_here,
+ (unsigned int) PAGE_CACHE_SIZE);
+ }
+ base = buf->page_len; /* align to start of tail */
+ }
+
+ /* process tail */
+ base -= buf->page_len;
+ if (todo) {
+ c = buf->tail->iov_base + base;
+ if (copied) {
+ unsigned int l = desc->elem_size - copied;
+
+ if (encode)
+ memcpy(c, elem + copied, l);
+ else {
+ memcpy(elem + copied, c, l);
+ err = desc->xcode(desc, elem);
+ if (err)
+ goto out;
+ }
+ todo -= l;
+ c += l;
+ }
+ while (todo) {
+ err = desc->xcode(desc, c);
+ if (err)
+ goto out;
+ c += desc->elem_size;
+ todo -= desc->elem_size;
+ }
+ }
+ err = 0;
+
+out:
+ if (elem)
+ kfree(elem);
+ if (ppages)
+ kunmap(*ppages);
+ return err;
+}
+
+int
+xdr_decode_array2(struct xdr_buf *buf, unsigned int base,
+ struct xdr_array2_desc *desc)
+{
+ if (base >= buf->len)
+ return -EINVAL;
+
+ return xdr_xcode_array2(buf, base, desc, 0);
+}
+
+int
+xdr_encode_array2(struct xdr_buf *buf, unsigned int base,
+ struct xdr_array2_desc *desc)
+{
+ if ((unsigned long) base + 4 + desc->array_len * desc->elem_size >
+ buf->head->iov_len + buf->page_len + buf->tail->iov_len)
+ return -EINVAL;
+
+ return xdr_xcode_array2(buf, base, desc, 1);
+}
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index c74a6bb..eca9240 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -569,8 +569,11 @@ void xprt_connect(struct rpc_task *task)
if (xprt->sock != NULL)
schedule_delayed_work(&xprt->sock_connect,
RPC_REESTABLISH_TIMEOUT);
- else
+ else {
schedule_work(&xprt->sock_connect);
+ if (!RPC_IS_ASYNC(task))
+ flush_scheduled_work();
+ }
}
return;
out_write:
@@ -725,7 +728,8 @@ csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb)
goto no_checksum;
desc.csum = csum_partial(skb->data, desc.offset, skb->csum);
- xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_and_csum_bits);
+ if (xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_and_csum_bits) < 0)
+ return -1;
if (desc.offset != skb->len) {
unsigned int csum2;
csum2 = skb_checksum(skb, desc.offset, skb->len - desc.offset, 0);
@@ -737,7 +741,8 @@ csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb)
return -1;
return 0;
no_checksum:
- xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_bits);
+ if (xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_bits) < 0)
+ return -1;
if (desc.count)
return -1;
return 0;
@@ -821,10 +826,15 @@ tcp_copy_data(skb_reader_t *desc, void *p, size_t len)
{
if (len > desc->count)
len = desc->count;
- if (skb_copy_bits(desc->skb, desc->offset, p, len))
+ if (skb_copy_bits(desc->skb, desc->offset, p, len)) {
+ dprintk("RPC: failed to copy %zu bytes from skb. %zu bytes remain\n",
+ len, desc->count);
return 0;
+ }
desc->offset += len;
desc->count -= len;
+ dprintk("RPC: copied %zu bytes from skb. %zu bytes remain\n",
+ len, desc->count);
return len;
}
@@ -863,6 +873,8 @@ tcp_read_fraghdr(struct rpc_xprt *xprt, skb_reader_t *desc)
static void
tcp_check_recm(struct rpc_xprt *xprt)
{
+ dprintk("RPC: xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u, tcp_flags = %lx\n",
+ xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen, xprt->tcp_flags);
if (xprt->tcp_offset == xprt->tcp_reclen) {
xprt->tcp_flags |= XPRT_COPY_RECM;
xprt->tcp_offset = 0;
@@ -907,6 +919,7 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc)
struct rpc_rqst *req;
struct xdr_buf *rcvbuf;
size_t len;
+ ssize_t r;
/* Find and lock the request corresponding to this xid */
spin_lock(&xprt->sock_lock);
@@ -927,15 +940,40 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc)
len = xprt->tcp_reclen - xprt->tcp_offset;
memcpy(&my_desc, desc, sizeof(my_desc));
my_desc.count = len;
- xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied,
+ r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied,
&my_desc, tcp_copy_data);
- desc->count -= len;
- desc->offset += len;
+ desc->count -= r;
+ desc->offset += r;
} else
- xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied,
+ r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied,
desc, tcp_copy_data);
- xprt->tcp_copied += len;
- xprt->tcp_offset += len;
+
+ if (r > 0) {
+ xprt->tcp_copied += r;
+ xprt->tcp_offset += r;
+ }
+ if (r != len) {
+ /* Error when copying to the receive buffer,
+ * usually because we weren't able to allocate
+ * additional buffer pages. All we can do now
+ * is turn off XPRT_COPY_DATA, so the request
+ * will not receive any additional updates,
+ * and time out.
+ * Any remaining data from this record will
+ * be discarded.
+ */
+ xprt->tcp_flags &= ~XPRT_COPY_DATA;
+ dprintk("RPC: XID %08x truncated request\n",
+ ntohl(xprt->tcp_xid));
+ dprintk("RPC: xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u\n",
+ xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen);
+ goto out;
+ }
+
+ dprintk("RPC: XID %08x read %u bytes\n",
+ ntohl(xprt->tcp_xid), r);
+ dprintk("RPC: xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u\n",
+ xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen);
if (xprt->tcp_copied == req->rq_private_buf.buflen)
xprt->tcp_flags &= ~XPRT_COPY_DATA;
@@ -944,6 +982,7 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc)
xprt->tcp_flags &= ~XPRT_COPY_DATA;
}
+out:
if (!(xprt->tcp_flags & XPRT_COPY_DATA)) {
dprintk("RPC: %4d received reply complete\n",
req->rq_task->tk_pid);
@@ -967,6 +1006,7 @@ tcp_read_discard(struct rpc_xprt *xprt, skb_reader_t *desc)
desc->count -= len;
desc->offset += len;
xprt->tcp_offset += len;
+ dprintk("RPC: discarded %u bytes\n", len);
tcp_check_recm(xprt);
}
@@ -1064,8 +1104,7 @@ tcp_state_change(struct sock *sk)
case TCP_SYN_RECV:
break;
default:
- if (xprt_test_and_clear_connected(xprt))
- rpc_wake_up_status(&xprt->pending, -ENOTCONN);
+ xprt_disconnect(xprt);
break;
}
out:
@@ -1203,6 +1242,8 @@ xprt_transmit(struct rpc_task *task)
list_add_tail(&req->rq_list, &xprt->recv);
spin_unlock_bh(&xprt->sock_lock);
xprt_reset_majortimeo(req);
+ /* Turn off autodisconnect */
+ del_singleshot_timer_sync(&xprt->timer);
}
} else if (!req->rq_bytes_sent)
return;
@@ -1333,8 +1374,6 @@ xprt_reserve(struct rpc_task *task)
spin_lock(&xprt->xprt_lock);
do_xprt_reserve(task);
spin_unlock(&xprt->xprt_lock);
- if (task->tk_rqstp)
- del_timer_sync(&xprt->timer);
}
}
@@ -1649,6 +1688,10 @@ xprt_shutdown(struct rpc_xprt *xprt)
rpc_wake_up(&xprt->backlog);
wake_up(&xprt->cong_wait);
del_timer_sync(&xprt->timer);
+
+ /* synchronously wait for connect worker to finish */
+ cancel_delayed_work(&xprt->sock_connect);
+ flush_scheduled_work();
}
/*
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 0a42607..d65ed86 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -118,7 +118,6 @@ retry:
xfrm_policy_put_afinfo(afinfo);
return type;
}
-EXPORT_SYMBOL(xfrm_get_type);
int xfrm_dst_lookup(struct xfrm_dst **dst, struct flowi *fl,
unsigned short family)
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 2537f26..9d206c2 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1055,6 +1055,43 @@ int xfrm_state_mtu(struct xfrm_state *x, int mtu)
}
EXPORT_SYMBOL(xfrm_state_mtu);
+
+int xfrm_init_state(struct xfrm_state *x)
+{
+ struct xfrm_state_afinfo *afinfo;
+ int family = x->props.family;
+ int err;
+
+ err = -EAFNOSUPPORT;
+ afinfo = xfrm_state_get_afinfo(family);
+ if (!afinfo)
+ goto error;
+
+ err = 0;
+ if (afinfo->init_flags)
+ err = afinfo->init_flags(x);
+
+ xfrm_state_put_afinfo(afinfo);
+
+ if (err)
+ goto error;
+
+ err = -EPROTONOSUPPORT;
+ x->type = xfrm_get_type(x->id.proto, family);
+ if (x->type == NULL)
+ goto error;
+
+ err = x->type->init_state(x);
+ if (err)
+ goto error;
+
+ x->km.state = XFRM_STATE_VALID;
+
+error:
+ return err;
+}
+
+EXPORT_SYMBOL(xfrm_init_state);
void __init xfrm_state_init(void)
{
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 5ce8558..ecade48 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -249,17 +249,10 @@ static struct xfrm_state *xfrm_state_construct(struct xfrm_usersa_info *p,
if ((err = attach_encap_tmpl(&x->encap, xfrma[XFRMA_ENCAP-1])))
goto error;
- err = -ENOENT;
- x->type = xfrm_get_type(x->id.proto, x->props.family);
- if (x->type == NULL)
- goto error;
-
- err = x->type->init_state(x, NULL);
+ err = xfrm_init_state(x);
if (err)
goto error;
- x->curlft.add_time = (unsigned long) xtime.tv_sec;
- x->km.state = XFRM_STATE_VALID;
x->km.seq = p->seq;
return x;