diff options
Diffstat (limited to 'net')
176 files changed, 9606 insertions, 3538 deletions
diff --git a/net/802/tr.c b/net/802/tr.c index 85293cc..a755e88 100644 --- a/net/802/tr.c +++ b/net/802/tr.c @@ -47,12 +47,12 @@ static void rif_check_expire(unsigned long dummy); * Each RIF entry we learn is kept this way */ -struct rif_cache_s { +struct rif_cache { unsigned char addr[TR_ALEN]; int iface; - __u16 rcf; - __u16 rseg[8]; - struct rif_cache_s *next; + __be16 rcf; + __be16 rseg[8]; + struct rif_cache *next; unsigned long last_used; unsigned char local_ring; }; @@ -64,7 +64,7 @@ struct rif_cache_s { * up a lot. */ -static struct rif_cache_s *rif_table[RIF_TABLE_SIZE]; +static struct rif_cache *rif_table[RIF_TABLE_SIZE]; static DEFINE_SPINLOCK(rif_lock); @@ -249,7 +249,7 @@ void tr_source_route(struct sk_buff *skb,struct trh_hdr *trh,struct net_device * { int slack; unsigned int hash; - struct rif_cache_s *entry; + struct rif_cache *entry; unsigned char *olddata; static const unsigned char mcast_func_addr[] = {0xC0,0x00,0x00,0x04,0x00,0x00}; @@ -337,7 +337,7 @@ printk("source routing for %02X:%02X:%02X:%02X:%02X:%02X\n",trh->daddr[0], static void tr_add_rif_info(struct trh_hdr *trh, struct net_device *dev) { unsigned int hash, rii_p = 0; - struct rif_cache_s *entry; + struct rif_cache *entry; spin_lock_bh(&rif_lock); @@ -373,7 +373,7 @@ printk("adding rif_entry: addr:%02X:%02X:%02X:%02X:%02X:%02X rcf:%04X\n", * FIXME: We ought to keep some kind of cache size * limiting and adjust the timers to suit. */ - entry=kmalloc(sizeof(struct rif_cache_s),GFP_ATOMIC); + entry=kmalloc(sizeof(struct rif_cache),GFP_ATOMIC); if(!entry) { @@ -435,7 +435,7 @@ static void rif_check_expire(unsigned long dummy) spin_lock_bh(&rif_lock); for(i =0; i < RIF_TABLE_SIZE; i++) { - struct rif_cache_s *entry, **pentry; + struct rif_cache *entry, **pentry; pentry = rif_table+i; while((entry=*pentry) != NULL) { @@ -467,10 +467,10 @@ static void rif_check_expire(unsigned long dummy) #ifdef CONFIG_PROC_FS -static struct rif_cache_s *rif_get_idx(loff_t pos) +static struct rif_cache *rif_get_idx(loff_t pos) { int i; - struct rif_cache_s *entry; + struct rif_cache *entry; loff_t off = 0; for(i = 0; i < RIF_TABLE_SIZE; i++) @@ -493,7 +493,7 @@ static void *rif_seq_start(struct seq_file *seq, loff_t *pos) static void *rif_seq_next(struct seq_file *seq, void *v, loff_t *pos) { int i; - struct rif_cache_s *ent = v; + struct rif_cache *ent = v; ++*pos; @@ -522,7 +522,7 @@ static void rif_seq_stop(struct seq_file *seq, void *v) static int rif_seq_show(struct seq_file *seq, void *v) { int j, rcf_len, segment, brdgnmb; - struct rif_cache_s *entry = v; + struct rif_cache *entry = v; if (v == SEQ_START_TOKEN) seq_puts(seq, diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c index 54640c0..c34614e 100644 --- a/net/appletalk/aarp.c +++ b/net/appletalk/aarp.c @@ -35,6 +35,7 @@ #include <net/datalink.h> #include <net/psnap.h> #include <linux/atalk.h> +#include <linux/delay.h> #include <linux/init.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> @@ -462,8 +463,7 @@ void aarp_probe_network(struct atalk_iface *atif) aarp_send_probe(atif->dev, &atif->address); /* Defer 1/10th */ - current->state = TASK_INTERRUPTIBLE; - schedule_timeout(HZ / 10); + msleep(100); if (atif->status & ATIF_PROBE_FAIL) break; @@ -510,9 +510,8 @@ int aarp_proxy_probe_network(struct atalk_iface *atif, struct atalk_addr *sa) aarp_send_probe(atif->dev, sa); /* Defer 1/10th */ - current->state = TASK_INTERRUPTIBLE; write_unlock_bh(&aarp_lock); - schedule_timeout(HZ / 10); + msleep(100); write_lock_bh(&aarp_lock); if (entry->status & ATIF_PROBE_FAIL) @@ -565,7 +564,7 @@ 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_device.c b/net/bridge/br_device.c index d9b72fd..f564ee9 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -21,10 +21,7 @@ static struct net_device_stats *br_dev_get_stats(struct net_device *dev) { - struct net_bridge *br; - - br = dev->priv; - + struct net_bridge *br = netdev_priv(dev); return &br->statistics; } @@ -54,9 +51,11 @@ int br_dev_xmit(struct sk_buff *skb, struct net_device *dev) static int br_dev_open(struct net_device *dev) { - netif_start_queue(dev); + struct net_bridge *br = netdev_priv(dev); - br_stp_enable_bridge(dev->priv); + br_features_recompute(br); + netif_start_queue(dev); + br_stp_enable_bridge(br); return 0; } @@ -67,7 +66,7 @@ static void br_dev_set_multicast_list(struct net_device *dev) static int br_dev_stop(struct net_device *dev) { - br_stp_disable_bridge(dev->priv); + br_stp_disable_bridge(netdev_priv(dev)); netif_stop_queue(dev); @@ -76,7 +75,7 @@ static int br_dev_stop(struct net_device *dev) static int br_change_mtu(struct net_device *dev, int new_mtu) { - if ((new_mtu < 68) || new_mtu > br_min_mtu(dev->priv)) + if (new_mtu < 68 || new_mtu > br_min_mtu(netdev_priv(dev))) return -EINVAL; dev->mtu = new_mtu; 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_if.c b/net/bridge/br_if.c index 69872bf..91bb895 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -314,6 +314,28 @@ int br_min_mtu(const struct net_bridge *br) return mtu; } +/* + * Recomputes features using slave's features + */ +void br_features_recompute(struct net_bridge *br) +{ + struct net_bridge_port *p; + unsigned long features, checksum; + + features = NETIF_F_SG | NETIF_F_FRAGLIST + | NETIF_F_HIGHDMA | NETIF_F_TSO; + checksum = NETIF_F_IP_CSUM; /* least commmon subset */ + + list_for_each_entry(p, &br->port_list, list) { + if (!(p->dev->features + & (NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM))) + checksum = 0; + features &= p->dev->features; + } + + br->dev->features = features | checksum | NETIF_F_LLTX; +} + /* called with RTNL */ int br_add_if(struct net_bridge *br, struct net_device *dev) { @@ -368,6 +390,7 @@ int br_del_if(struct net_bridge *br, struct net_device *dev) spin_lock_bh(&br->lock); br_stp_recalculate_bridge_id(br); + br_features_recompute(br); spin_unlock_bh(&br->lock); return 0; diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 2b1cce4..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_rx(skb); - + netif_receive_skb(skb); return 0; } @@ -54,6 +50,9 @@ int br_handle_frame_finish(struct sk_buff *skb) struct net_bridge_fdb_entry *dst; int passedup = 0; + /* insert into forwarding database after filtering to avoid spoofing */ + br_fdb_update(p->br, p, eth_hdr(skb)->h_source); + if (br->dev->flags & IFF_PROMISC) { struct sk_buff *skb2; @@ -108,8 +107,7 @@ int br_handle_frame(struct net_bridge_port *p, struct sk_buff **pskb) if (!is_valid_ether_addr(eth_hdr(skb)->h_source)) goto err; - if (p->state == BR_STATE_LEARNING || - p->state == BR_STATE_FORWARDING) + if (p->state == BR_STATE_LEARNING) br_fdb_update(p->br, p, eth_hdr(skb)->h_source); if (p->br->stp_enabled && 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/bridge/br_notify.c b/net/bridge/br_notify.c index f8fb49e..917311c 100644 --- a/net/bridge/br_notify.c +++ b/net/bridge/br_notify.c @@ -65,6 +65,15 @@ static int br_device_event(struct notifier_block *unused, unsigned long event, v } break; + case NETDEV_FEAT_CHANGE: + if (br->dev->flags & IFF_UP) + br_features_recompute(br); + + /* could do recursive feature change notification + * but who would care?? + */ + break; + case NETDEV_DOWN: if (br->dev->flags & IFF_UP) br_stp_disable_port(p); diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 54d63f13..bdf95a7 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -174,6 +174,7 @@ extern int br_add_if(struct net_bridge *br, extern int br_del_if(struct net_bridge *br, struct net_device *dev); extern int br_min_mtu(const struct net_bridge *br); +extern void br_features_recompute(struct net_bridge *br); /* br_input.c */ extern int br_handle_frame_finish(struct sk_buff *skb); diff --git a/net/bridge/br_stp_bpdu.c b/net/bridge/br_stp_bpdu.c index b91a875..d071f1c 100644 --- a/net/bridge/br_stp_bpdu.c +++ b/net/bridge/br_stp_bpdu.c @@ -140,6 +140,9 @@ int br_stp_handle_bpdu(struct sk_buff *skb) struct net_bridge *br = p->br; unsigned char *buf; + /* insert into forwarding database after filtering to avoid spoofing */ + br_fdb_update(p->br, p, eth_hdr(skb)->h_source); + /* need at least the 802 and STP headers */ if (!pskb_may_pull(skb, sizeof(header)+1) || memcmp(skb->data, header, sizeof(header))) diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 18ebc66..c454014 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -859,8 +859,7 @@ static int translate_table(struct ebt_replace *repl, if (repl->valid_hooks & (1 << i)) if (check_chainloops(newinfo->hook_entry[i], cl_s, udc_cnt, i, newinfo->entries)) { - if (cl_s) - vfree(cl_s); + vfree(cl_s); return -EINVAL; } @@ -883,8 +882,7 @@ static int translate_table(struct ebt_replace *repl, EBT_ENTRY_ITERATE(newinfo->entries, newinfo->entries_size, ebt_cleanup_entry, &i); } - if (cl_s) - vfree(cl_s); + vfree(cl_s); return ret; } @@ -1030,8 +1028,7 @@ static int do_replace(void __user *user, unsigned int len) } vfree(table); - if (counterstmp) - vfree(counterstmp); + vfree(counterstmp); return ret; free_unlock: @@ -1040,8 +1037,7 @@ free_iterate: EBT_ENTRY_ITERATE(newinfo->entries, newinfo->entries_size, ebt_cleanup_entry, NULL); free_counterstmp: - if (counterstmp) - vfree(counterstmp); + vfree(counterstmp); /* can be initialized in translate_table() */ if (newinfo->chainstack) { for (i = 0; i < num_possible_cpus(); i++) @@ -1049,11 +1045,9 @@ free_counterstmp: vfree(newinfo->chainstack); } free_entries: - if (newinfo->entries) - vfree(newinfo->entries); + vfree(newinfo->entries); free_newinfo: - if (newinfo) - vfree(newinfo); + vfree(newinfo); return ret; } @@ -1213,8 +1207,7 @@ void ebt_unregister_table(struct ebt_table *table) down(&ebt_mutex); LIST_DELETE(&ebt_tables, table); up(&ebt_mutex); - if (table->private->entries) - vfree(table->private->entries); + vfree(table->private->entries); if (table->private->chainstack) { for (i = 0; i < num_possible_cpus(); i++) vfree(table->private->chainstack[i]); diff --git a/net/core/Makefile b/net/core/Makefile index 81f0324..5e0c56b 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -2,7 +2,8 @@ # Makefile for the Linux networking core. # -obj-y := sock.o skbuff.o iovec.o datagram.o stream.o scm.o gen_stats.o gen_estimator.o +obj-y := sock.o request_sock.o skbuff.o iovec.o datagram.o stream.o scm.o \ + gen_stats.o gen_estimator.o obj-$(CONFIG_SYSCTL) += sysctl_net_core.o diff --git a/net/core/dev.c b/net/core/dev.c index d4d9e26..7016e0c 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -115,18 +115,6 @@ #endif /* CONFIG_NET_RADIO */ #include <asm/current.h> -/* This define, if set, will randomly drop a packet when congestion - * is more than moderate. It helps fairness in the multi-interface - * case when one of them is a hog, but it kills performance for the - * single interface case so it is off now by default. - */ -#undef RAND_LIE - -/* Setting this will sample the queue lengths and thus congestion - * via a timer instead of as each packet is received. - */ -#undef OFFLINE_SAMPLE - /* * The list of packet types we will receive (as opposed to discard) * and the routines to invoke. @@ -159,11 +147,6 @@ static DEFINE_SPINLOCK(ptype_lock); static struct list_head ptype_base[16]; /* 16 way hashed list */ static struct list_head ptype_all; /* Taps */ -#ifdef OFFLINE_SAMPLE -static void sample_queue(unsigned long dummy); -static struct timer_list samp_timer = TIMER_INITIALIZER(sample_queue, 0, 0); -#endif - /* * The @dev_base list is protected by @dev_base_lock and the rtln * semaphore. @@ -215,7 +198,7 @@ static struct notifier_block *netdev_chain; * Device drivers call our routines to queue packets here. We empty the * queue in the local softnet handler. */ -DEFINE_PER_CPU(struct softnet_data, softnet_data) = { 0, }; +DEFINE_PER_CPU(struct softnet_data, softnet_data) = { NULL }; #ifdef CONFIG_SYSFS extern int netdev_sysfs_init(void); @@ -761,6 +744,18 @@ int dev_change_name(struct net_device *dev, char *newname) } /** + * netdev_features_change - device changes fatures + * @dev: device to cause notification + * + * Called to indicate a device has changed features. + */ +void netdev_features_change(struct net_device *dev) +{ + notifier_call_chain(&netdev_chain, NETDEV_FEAT_CHANGE, dev); +} +EXPORT_SYMBOL(netdev_features_change); + +/** * netdev_state_change - device changes state * @dev: device to cause notification * @@ -1351,71 +1346,13 @@ out: Receiver routines =======================================================================*/ -int netdev_max_backlog = 300; +int netdev_max_backlog = 1000; +int netdev_budget = 300; int weight_p = 64; /* old backlog weight */ -/* These numbers are selected based on intuition and some - * experimentatiom, if you have more scientific way of doing this - * please go ahead and fix things. - */ -int no_cong_thresh = 10; -int no_cong = 20; -int lo_cong = 100; -int mod_cong = 290; DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; -static void get_sample_stats(int cpu) -{ -#ifdef RAND_LIE - unsigned long rd; - int rq; -#endif - struct softnet_data *sd = &per_cpu(softnet_data, cpu); - int blog = sd->input_pkt_queue.qlen; - int avg_blog = sd->avg_blog; - - avg_blog = (avg_blog >> 1) + (blog >> 1); - - if (avg_blog > mod_cong) { - /* Above moderate congestion levels. */ - sd->cng_level = NET_RX_CN_HIGH; -#ifdef RAND_LIE - rd = net_random(); - rq = rd % netdev_max_backlog; - if (rq < avg_blog) /* unlucky bastard */ - sd->cng_level = NET_RX_DROP; -#endif - } else if (avg_blog > lo_cong) { - sd->cng_level = NET_RX_CN_MOD; -#ifdef RAND_LIE - rd = net_random(); - rq = rd % netdev_max_backlog; - if (rq < avg_blog) /* unlucky bastard */ - sd->cng_level = NET_RX_CN_HIGH; -#endif - } else if (avg_blog > no_cong) - sd->cng_level = NET_RX_CN_LOW; - else /* no congestion */ - sd->cng_level = NET_RX_SUCCESS; - - sd->avg_blog = avg_blog; -} - -#ifdef OFFLINE_SAMPLE -static void sample_queue(unsigned long dummy) -{ -/* 10 ms 0r 1ms -- i don't care -- JHS */ - int next_tick = 1; - int cpu = smp_processor_id(); - - get_sample_stats(cpu); - next_tick += jiffies; - mod_timer(&samp_timer, next_tick); -} -#endif - - /** * netif_rx - post buffer to the network code * @skb: buffer to post @@ -1436,7 +1373,6 @@ static void sample_queue(unsigned long dummy) int netif_rx(struct sk_buff *skb) { - int this_cpu; struct softnet_data *queue; unsigned long flags; @@ -1452,38 +1388,22 @@ int netif_rx(struct sk_buff *skb) * short when CPU is congested, but is still operating. */ local_irq_save(flags); - this_cpu = smp_processor_id(); queue = &__get_cpu_var(softnet_data); __get_cpu_var(netdev_rx_stat).total++; if (queue->input_pkt_queue.qlen <= netdev_max_backlog) { if (queue->input_pkt_queue.qlen) { - if (queue->throttle) - goto drop; - enqueue: dev_hold(skb->dev); __skb_queue_tail(&queue->input_pkt_queue, skb); -#ifndef OFFLINE_SAMPLE - get_sample_stats(this_cpu); -#endif local_irq_restore(flags); - return queue->cng_level; + return NET_RX_SUCCESS; } - if (queue->throttle) - queue->throttle = 0; - netif_rx_schedule(&queue->backlog_dev); goto enqueue; } - if (!queue->throttle) { - queue->throttle = 1; - __get_cpu_var(netdev_rx_stat).throttled++; - } - -drop: __get_cpu_var(netdev_rx_stat).dropped++; local_irq_restore(flags); @@ -1732,6 +1652,7 @@ static int process_backlog(struct net_device *backlog_dev, int *budget) struct softnet_data *queue = &__get_cpu_var(softnet_data); unsigned long start_time = jiffies; + backlog_dev->weight = weight_p; for (;;) { struct sk_buff *skb; struct net_device *dev; @@ -1767,8 +1688,6 @@ job_done: smp_mb__before_clear_bit(); netif_poll_enable(backlog_dev); - if (queue->throttle) - queue->throttle = 0; local_irq_enable(); return 0; } @@ -1777,8 +1696,7 @@ static void net_rx_action(struct softirq_action *h) { struct softnet_data *queue = &__get_cpu_var(softnet_data); unsigned long start_time = jiffies; - int budget = netdev_max_backlog; - + int budget = netdev_budget; local_irq_disable(); @@ -2042,15 +1960,9 @@ static int softnet_seq_show(struct seq_file *seq, void *v) struct netif_rx_stats *s = v; seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x %08x\n", - s->total, s->dropped, s->time_squeeze, s->throttled, - s->fastroute_hit, s->fastroute_success, s->fastroute_defer, - s->fastroute_deferred_out, -#if 0 - s->fastroute_latency_reduction -#else - s->cpu_collision -#endif - ); + s->total, s->dropped, s->time_squeeze, 0, + 0, 0, 0, 0, /* was fastroute */ + s->cpu_collision ); return 0; } @@ -3292,9 +3204,6 @@ static int __init net_dev_init(void) queue = &per_cpu(softnet_data, i); skb_queue_head_init(&queue->input_pkt_queue); - queue->throttle = 0; - queue->cng_level = 0; - queue->avg_blog = 10; /* arbitrary non-zero */ queue->completion_queue = NULL; INIT_LIST_HEAD(&queue->poll_list); set_bit(__LINK_STATE_START, &queue->backlog_dev.state); @@ -3303,11 +3212,6 @@ static int __init net_dev_init(void) atomic_set(&queue->backlog_dev.refcnt, 1); } -#ifdef OFFLINE_SAMPLE - samp_timer.expires = jiffies + (10 * HZ); - add_timer(&samp_timer); -#endif - dev_boot_phase = 0; open_softirq(NET_TX_SOFTIRQ, net_tx_action, NULL); diff --git a/net/core/ethtool.c b/net/core/ethtool.c index f05fde9..a3eeb88 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -29,7 +29,7 @@ u32 ethtool_op_get_link(struct net_device *dev) u32 ethtool_op_get_tx_csum(struct net_device *dev) { - return (dev->features & NETIF_F_IP_CSUM) != 0; + return (dev->features & (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM)) != 0; } int ethtool_op_set_tx_csum(struct net_device *dev, u32 data) @@ -42,6 +42,15 @@ int ethtool_op_set_tx_csum(struct net_device *dev, u32 data) return 0; } +int ethtool_op_set_tx_hw_csum(struct net_device *dev, u32 data) +{ + if (data) + dev->features |= NETIF_F_HW_CSUM; + else + dev->features &= ~NETIF_F_HW_CSUM; + + return 0; +} u32 ethtool_op_get_sg(struct net_device *dev) { return (dev->features & NETIF_F_SG) != 0; @@ -347,7 +356,7 @@ static int ethtool_set_coalesce(struct net_device *dev, void __user *useraddr) { struct ethtool_coalesce coalesce; - if (!dev->ethtool_ops->get_coalesce) + if (!dev->ethtool_ops->set_coalesce) return -EOPNOTSUPP; if (copy_from_user(&coalesce, useraddr, sizeof(coalesce))) @@ -682,6 +691,7 @@ int dev_ethtool(struct ifreq *ifr) void __user *useraddr = ifr->ifr_data; u32 ethcmd; int rc; + unsigned long old_features; /* * XXX: This can be pushed down into the ethtool_* handlers that @@ -703,6 +713,8 @@ int dev_ethtool(struct ifreq *ifr) if ((rc = dev->ethtool_ops->begin(dev)) < 0) return rc; + old_features = dev->features; + switch (ethcmd) { case ETHTOOL_GSET: rc = ethtool_get_settings(dev, useraddr); @@ -712,7 +724,6 @@ int dev_ethtool(struct ifreq *ifr) break; case ETHTOOL_GDRVINFO: rc = ethtool_get_drvinfo(dev, useraddr); - break; case ETHTOOL_GREGS: rc = ethtool_get_regs(dev, useraddr); @@ -801,6 +812,10 @@ int dev_ethtool(struct ifreq *ifr) if(dev->ethtool_ops->complete) dev->ethtool_ops->complete(dev); + + if (old_features != dev->features) + netdev_features_change(dev); + return rc; ioctl: @@ -817,3 +832,4 @@ EXPORT_SYMBOL(ethtool_op_get_tx_csum); EXPORT_SYMBOL(ethtool_op_set_sg); EXPORT_SYMBOL(ethtool_op_set_tso); EXPORT_SYMBOL(ethtool_op_set_tx_csum); +EXPORT_SYMBOL(ethtool_op_set_tx_hw_csum); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 43bdc52..851eb92 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -32,6 +32,7 @@ #include <net/sock.h> #include <linux/rtnetlink.h> #include <linux/random.h> +#include <linux/string.h> #define NEIGH_DEBUG 1 @@ -1276,9 +1277,14 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev, INIT_RCU_HEAD(&p->rcu_head); p->reachable_time = neigh_rand_reach_time(p->base_reachable_time); - if (dev && dev->neigh_setup && dev->neigh_setup(dev, p)) { - kfree(p); - return NULL; + if (dev) { + if (dev->neigh_setup && dev->neigh_setup(dev, p)) { + kfree(p); + return NULL; + } + + dev_hold(dev); + p->dev = dev; } p->sysctl_table = NULL; write_lock_bh(&tbl->lock); @@ -1309,6 +1315,8 @@ void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms) *p = parms->next; parms->dead = 1; write_unlock_bh(&tbl->lock); + if (parms->dev) + dev_put(parms->dev); call_rcu(&parms->rcu_head, neigh_rcu_free_parms); return; } @@ -1546,20 +1554,323 @@ out: return err; } +static int neightbl_fill_parms(struct sk_buff *skb, struct neigh_parms *parms) +{ + struct rtattr *nest = NULL; + + nest = RTA_NEST(skb, NDTA_PARMS); + + if (parms->dev) + RTA_PUT_U32(skb, NDTPA_IFINDEX, parms->dev->ifindex); + + RTA_PUT_U32(skb, NDTPA_REFCNT, atomic_read(&parms->refcnt)); + RTA_PUT_U32(skb, NDTPA_QUEUE_LEN, parms->queue_len); + RTA_PUT_U32(skb, NDTPA_PROXY_QLEN, parms->proxy_qlen); + RTA_PUT_U32(skb, NDTPA_APP_PROBES, parms->app_probes); + RTA_PUT_U32(skb, NDTPA_UCAST_PROBES, parms->ucast_probes); + RTA_PUT_U32(skb, NDTPA_MCAST_PROBES, parms->mcast_probes); + RTA_PUT_MSECS(skb, NDTPA_REACHABLE_TIME, parms->reachable_time); + RTA_PUT_MSECS(skb, NDTPA_BASE_REACHABLE_TIME, + parms->base_reachable_time); + RTA_PUT_MSECS(skb, NDTPA_GC_STALETIME, parms->gc_staletime); + RTA_PUT_MSECS(skb, NDTPA_DELAY_PROBE_TIME, parms->delay_probe_time); + RTA_PUT_MSECS(skb, NDTPA_RETRANS_TIME, parms->retrans_time); + RTA_PUT_MSECS(skb, NDTPA_ANYCAST_DELAY, parms->anycast_delay); + RTA_PUT_MSECS(skb, NDTPA_PROXY_DELAY, parms->proxy_delay); + RTA_PUT_MSECS(skb, NDTPA_LOCKTIME, parms->locktime); + + return RTA_NEST_END(skb, nest); + +rtattr_failure: + return RTA_NEST_CANCEL(skb, nest); +} + +static int neightbl_fill_info(struct neigh_table *tbl, struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct nlmsghdr *nlh; + struct ndtmsg *ndtmsg; + + nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg), + NLM_F_MULTI); + + ndtmsg = NLMSG_DATA(nlh); + + read_lock_bh(&tbl->lock); + ndtmsg->ndtm_family = tbl->family; + + RTA_PUT_STRING(skb, NDTA_NAME, tbl->id); + RTA_PUT_MSECS(skb, NDTA_GC_INTERVAL, tbl->gc_interval); + RTA_PUT_U32(skb, NDTA_THRESH1, tbl->gc_thresh1); + RTA_PUT_U32(skb, NDTA_THRESH2, tbl->gc_thresh2); + RTA_PUT_U32(skb, NDTA_THRESH3, tbl->gc_thresh3); + + { + unsigned long now = jiffies; + unsigned int flush_delta = now - tbl->last_flush; + unsigned int rand_delta = now - tbl->last_rand; + + struct ndt_config ndc = { + .ndtc_key_len = tbl->key_len, + .ndtc_entry_size = tbl->entry_size, + .ndtc_entries = atomic_read(&tbl->entries), + .ndtc_last_flush = jiffies_to_msecs(flush_delta), + .ndtc_last_rand = jiffies_to_msecs(rand_delta), + .ndtc_hash_rnd = tbl->hash_rnd, + .ndtc_hash_mask = tbl->hash_mask, + .ndtc_hash_chain_gc = tbl->hash_chain_gc, + .ndtc_proxy_qlen = tbl->proxy_queue.qlen, + }; + + RTA_PUT(skb, NDTA_CONFIG, sizeof(ndc), &ndc); + } + + { + int cpu; + struct ndt_stats ndst; + + memset(&ndst, 0, sizeof(ndst)); + + for (cpu = 0; cpu < NR_CPUS; cpu++) { + struct neigh_statistics *st; + + if (!cpu_possible(cpu)) + continue; + + st = per_cpu_ptr(tbl->stats, cpu); + ndst.ndts_allocs += st->allocs; + ndst.ndts_destroys += st->destroys; + ndst.ndts_hash_grows += st->hash_grows; + ndst.ndts_res_failed += st->res_failed; + ndst.ndts_lookups += st->lookups; + ndst.ndts_hits += st->hits; + ndst.ndts_rcv_probes_mcast += st->rcv_probes_mcast; + ndst.ndts_rcv_probes_ucast += st->rcv_probes_ucast; + ndst.ndts_periodic_gc_runs += st->periodic_gc_runs; + ndst.ndts_forced_gc_runs += st->forced_gc_runs; + } + + RTA_PUT(skb, NDTA_STATS, sizeof(ndst), &ndst); + } + + BUG_ON(tbl->parms.dev); + if (neightbl_fill_parms(skb, &tbl->parms) < 0) + goto rtattr_failure; + + read_unlock_bh(&tbl->lock); + return NLMSG_END(skb, nlh); + +rtattr_failure: + read_unlock_bh(&tbl->lock); + return NLMSG_CANCEL(skb, nlh); + +nlmsg_failure: + return -1; +} + +static int neightbl_fill_param_info(struct neigh_table *tbl, + struct neigh_parms *parms, + struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct ndtmsg *ndtmsg; + struct nlmsghdr *nlh; + + nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg), + NLM_F_MULTI); + + ndtmsg = NLMSG_DATA(nlh); + + read_lock_bh(&tbl->lock); + ndtmsg->ndtm_family = tbl->family; + RTA_PUT_STRING(skb, NDTA_NAME, tbl->id); + + if (neightbl_fill_parms(skb, parms) < 0) + goto rtattr_failure; + + read_unlock_bh(&tbl->lock); + return NLMSG_END(skb, nlh); + +rtattr_failure: + read_unlock_bh(&tbl->lock); + return NLMSG_CANCEL(skb, nlh); + +nlmsg_failure: + return -1; +} + +static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl, + int ifindex) +{ + struct neigh_parms *p; + + for (p = &tbl->parms; p; p = p->next) + if ((p->dev && p->dev->ifindex == ifindex) || + (!p->dev && !ifindex)) + return p; + + return NULL; +} + +int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) +{ + struct neigh_table *tbl; + struct ndtmsg *ndtmsg = NLMSG_DATA(nlh); + struct rtattr **tb = arg; + int err = -EINVAL; + + if (!tb[NDTA_NAME - 1] || !RTA_PAYLOAD(tb[NDTA_NAME - 1])) + return -EINVAL; + + read_lock(&neigh_tbl_lock); + for (tbl = neigh_tables; tbl; tbl = tbl->next) { + if (ndtmsg->ndtm_family && tbl->family != ndtmsg->ndtm_family) + continue; + + if (!rtattr_strcmp(tb[NDTA_NAME - 1], tbl->id)) + break; + } + + if (tbl == NULL) { + err = -ENOENT; + goto errout; + } + + /* + * We acquire tbl->lock to be nice to the periodic timers and + * make sure they always see a consistent set of values. + */ + write_lock_bh(&tbl->lock); + + if (tb[NDTA_THRESH1 - 1]) + tbl->gc_thresh1 = RTA_GET_U32(tb[NDTA_THRESH1 - 1]); + + if (tb[NDTA_THRESH2 - 1]) + tbl->gc_thresh2 = RTA_GET_U32(tb[NDTA_THRESH2 - 1]); + + if (tb[NDTA_THRESH3 - 1]) + tbl->gc_thresh3 = RTA_GET_U32(tb[NDTA_THRESH3 - 1]); + + if (tb[NDTA_GC_INTERVAL - 1]) + tbl->gc_interval = RTA_GET_MSECS(tb[NDTA_GC_INTERVAL - 1]); + + if (tb[NDTA_PARMS - 1]) { + struct rtattr *tbp[NDTPA_MAX]; + struct neigh_parms *p; + u32 ifindex = 0; + + if (rtattr_parse_nested(tbp, NDTPA_MAX, tb[NDTA_PARMS - 1]) < 0) + goto rtattr_failure; + + if (tbp[NDTPA_IFINDEX - 1]) + ifindex = RTA_GET_U32(tbp[NDTPA_IFINDEX - 1]); + + p = lookup_neigh_params(tbl, ifindex); + if (p == NULL) { + err = -ENOENT; + goto rtattr_failure; + } + + if (tbp[NDTPA_QUEUE_LEN - 1]) + p->queue_len = RTA_GET_U32(tbp[NDTPA_QUEUE_LEN - 1]); + + if (tbp[NDTPA_PROXY_QLEN - 1]) + p->proxy_qlen = RTA_GET_U32(tbp[NDTPA_PROXY_QLEN - 1]); + + if (tbp[NDTPA_APP_PROBES - 1]) + p->app_probes = RTA_GET_U32(tbp[NDTPA_APP_PROBES - 1]); + + if (tbp[NDTPA_UCAST_PROBES - 1]) + p->ucast_probes = + RTA_GET_U32(tbp[NDTPA_UCAST_PROBES - 1]); + + if (tbp[NDTPA_MCAST_PROBES - 1]) + p->mcast_probes = + RTA_GET_U32(tbp[NDTPA_MCAST_PROBES - 1]); + + if (tbp[NDTPA_BASE_REACHABLE_TIME - 1]) + p->base_reachable_time = + RTA_GET_MSECS(tbp[NDTPA_BASE_REACHABLE_TIME - 1]); + + if (tbp[NDTPA_GC_STALETIME - 1]) + p->gc_staletime = + RTA_GET_MSECS(tbp[NDTPA_GC_STALETIME - 1]); + + if (tbp[NDTPA_DELAY_PROBE_TIME - 1]) + p->delay_probe_time = + RTA_GET_MSECS(tbp[NDTPA_DELAY_PROBE_TIME - 1]); + + if (tbp[NDTPA_RETRANS_TIME - 1]) + p->retrans_time = + RTA_GET_MSECS(tbp[NDTPA_RETRANS_TIME - 1]); + + if (tbp[NDTPA_ANYCAST_DELAY - 1]) + p->anycast_delay = + RTA_GET_MSECS(tbp[NDTPA_ANYCAST_DELAY - 1]); + + if (tbp[NDTPA_PROXY_DELAY - 1]) + p->proxy_delay = + RTA_GET_MSECS(tbp[NDTPA_PROXY_DELAY - 1]); + + if (tbp[NDTPA_LOCKTIME - 1]) + p->locktime = RTA_GET_MSECS(tbp[NDTPA_LOCKTIME - 1]); + } + + err = 0; + +rtattr_failure: + write_unlock_bh(&tbl->lock); +errout: + read_unlock(&neigh_tbl_lock); + return err; +} + +int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) +{ + int idx, family; + int s_idx = cb->args[0]; + struct neigh_table *tbl; + + family = ((struct rtgenmsg *)NLMSG_DATA(cb->nlh))->rtgen_family; + + read_lock(&neigh_tbl_lock); + for (tbl = neigh_tables, idx = 0; tbl; tbl = tbl->next) { + struct neigh_parms *p; + + if (idx < s_idx || (family && tbl->family != family)) + continue; + + if (neightbl_fill_info(tbl, skb, cb) <= 0) + break; + + for (++idx, p = tbl->parms.next; p; p = p->next, idx++) { + if (idx < s_idx) + continue; + + if (neightbl_fill_param_info(tbl, p, skb, cb) <= 0) + goto out; + } + + } +out: + read_unlock(&neigh_tbl_lock); + cb->args[0] = idx; + + return skb->len; +} static int neigh_fill_info(struct sk_buff *skb, struct neighbour *n, - u32 pid, u32 seq, int event) + u32 pid, u32 seq, int event, unsigned int flags) { unsigned long now = jiffies; unsigned char *b = skb->tail; struct nda_cacheinfo ci; int locked = 0; u32 probes; - struct nlmsghdr *nlh = NLMSG_PUT(skb, pid, seq, event, - sizeof(struct ndmsg)); + struct nlmsghdr *nlh = NLMSG_NEW(skb, pid, seq, event, + sizeof(struct ndmsg), flags); struct ndmsg *ndm = NLMSG_DATA(nlh); - nlh->nlmsg_flags = pid ? NLM_F_MULTI : 0; ndm->ndm_family = n->ops->family; ndm->ndm_flags = n->flags; ndm->ndm_type = n->type; @@ -1609,7 +1920,8 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, continue; if (neigh_fill_info(skb, n, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, - RTM_NEWNEIGH) <= 0) { + RTM_NEWNEIGH, + NLM_F_MULTI) <= 0) { read_unlock_bh(&tbl->lock); rc = -1; goto out; @@ -2018,7 +2330,7 @@ void neigh_app_ns(struct neighbour *n) if (!skb) return; - if (neigh_fill_info(skb, n, 0, 0, RTM_GETNEIGH) < 0) { + if (neigh_fill_info(skb, n, 0, 0, RTM_GETNEIGH, 0) < 0) { kfree_skb(skb); return; } @@ -2037,7 +2349,7 @@ static void neigh_app_notify(struct neighbour *n) if (!skb) return; - if (neigh_fill_info(skb, n, 0, 0, RTM_NEWNEIGH) < 0) { + if (neigh_fill_info(skb, n, 0, 0, RTM_NEWNEIGH, 0) < 0) { kfree_skb(skb); return; } @@ -2281,7 +2593,7 @@ int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p, t->neigh_vars[17].extra1 = dev; } - dev_name = net_sysctl_strdup(dev_name_source); + dev_name = kstrdup(dev_name_source, GFP_KERNEL); if (!dev_name) { err = -ENOBUFS; goto free; @@ -2352,6 +2664,8 @@ EXPORT_SYMBOL(neigh_update); EXPORT_SYMBOL(neigh_update_hhs); EXPORT_SYMBOL(pneigh_enqueue); EXPORT_SYMBOL(pneigh_lookup); +EXPORT_SYMBOL(neightbl_dump_info); +EXPORT_SYMBOL(neightbl_set); #ifdef CONFIG_ARPD EXPORT_SYMBOL(neigh_app_ns); diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 060f703..e2137f3 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -21,6 +21,7 @@ #define to_net_dev(class) container_of(class, struct net_device, class_dev) static const char fmt_hex[] = "%#x\n"; +static const char fmt_long_hex[] = "%#lx\n"; static const char fmt_dec[] = "%d\n"; static const char fmt_ulong[] = "%lu\n"; @@ -91,7 +92,7 @@ static CLASS_DEVICE_ATTR(field, S_IRUGO, show_##field, NULL) \ NETDEVICE_ATTR(addr_len, fmt_dec); NETDEVICE_ATTR(iflink, fmt_dec); NETDEVICE_ATTR(ifindex, fmt_dec); -NETDEVICE_ATTR(features, fmt_hex); +NETDEVICE_ATTR(features, fmt_long_hex); NETDEVICE_ATTR(type, fmt_dec); /* use same locking rules as GIFHWADDR ioctl's */ @@ -184,6 +185,22 @@ static ssize_t store_tx_queue_len(struct class_device *dev, const char *buf, siz static CLASS_DEVICE_ATTR(tx_queue_len, S_IRUGO | S_IWUSR, show_tx_queue_len, store_tx_queue_len); +NETDEVICE_SHOW(weight, fmt_dec); + +static int change_weight(struct net_device *net, unsigned long new_weight) +{ + net->weight = new_weight; + return 0; +} + +static ssize_t store_weight(struct class_device *dev, const char *buf, size_t len) +{ + return netdev_store(dev, buf, len, change_weight); +} + +static CLASS_DEVICE_ATTR(weight, S_IRUGO | S_IWUSR, show_weight, + store_weight); + static struct class_device_attribute *net_class_attributes[] = { &class_device_attr_ifindex, @@ -193,6 +210,7 @@ static struct class_device_attribute *net_class_attributes[] = { &class_device_attr_features, &class_device_attr_mtu, &class_device_attr_flags, + &class_device_attr_weight, &class_device_attr_type, &class_device_attr_address, &class_device_attr_broadcast, 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/netpoll.c b/net/core/netpoll.c index a119696..c327c9e 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -130,19 +130,20 @@ static int checksum_udp(struct sk_buff *skb, struct udphdr *uh, */ static void poll_napi(struct netpoll *np) { + struct netpoll_info *npinfo = np->dev->npinfo; int budget = 16; if (test_bit(__LINK_STATE_RX_SCHED, &np->dev->state) && - np->poll_owner != smp_processor_id() && - spin_trylock(&np->poll_lock)) { - np->rx_flags |= NETPOLL_RX_DROP; + npinfo->poll_owner != smp_processor_id() && + spin_trylock(&npinfo->poll_lock)) { + npinfo->rx_flags |= NETPOLL_RX_DROP; atomic_inc(&trapped); np->dev->poll(np->dev, &budget); atomic_dec(&trapped); - np->rx_flags &= ~NETPOLL_RX_DROP; - spin_unlock(&np->poll_lock); + npinfo->rx_flags &= ~NETPOLL_RX_DROP; + spin_unlock(&npinfo->poll_lock); } } @@ -245,6 +246,7 @@ repeat: static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) { int status; + struct netpoll_info *npinfo; repeat: if(!np || !np->dev || !netif_running(np->dev)) { @@ -253,8 +255,9 @@ repeat: } /* avoid recursion */ - if(np->poll_owner == smp_processor_id() || - np->dev->xmit_lock_owner == smp_processor_id()) { + npinfo = np->dev->npinfo; + if (npinfo->poll_owner == smp_processor_id() || + np->dev->xmit_lock_owner == smp_processor_id()) { if (np->drop) np->drop(skb); else @@ -341,14 +344,22 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len) static void arp_reply(struct sk_buff *skb) { + struct netpoll_info *npinfo = skb->dev->npinfo; struct arphdr *arp; unsigned char *arp_ptr; int size, type = ARPOP_REPLY, ptype = ETH_P_ARP; u32 sip, tip; + unsigned long flags; struct sk_buff *send_skb; - struct netpoll *np = skb->dev->np; + struct netpoll *np = NULL; + + spin_lock_irqsave(&npinfo->rx_lock, flags); + if (npinfo->rx_np && npinfo->rx_np->dev == skb->dev) + np = npinfo->rx_np; + spin_unlock_irqrestore(&npinfo->rx_lock, flags); - if (!np) return; + if (!np) + return; /* No arp on this interface */ if (skb->dev->flags & IFF_NOARP) @@ -429,9 +440,9 @@ int __netpoll_rx(struct sk_buff *skb) int proto, len, ulen; struct iphdr *iph; struct udphdr *uh; - struct netpoll *np = skb->dev->np; + struct netpoll *np = skb->dev->npinfo->rx_np; - if (!np->rx_hook) + if (!np) goto out; if (skb->dev->type != ARPHRD_ETHER) goto out; @@ -611,9 +622,8 @@ int netpoll_setup(struct netpoll *np) { struct net_device *ndev = NULL; struct in_device *in_dev; - - np->poll_lock = SPIN_LOCK_UNLOCKED; - np->poll_owner = -1; + struct netpoll_info *npinfo; + unsigned long flags; if (np->dev_name) ndev = dev_get_by_name(np->dev_name); @@ -624,7 +634,17 @@ int netpoll_setup(struct netpoll *np) } np->dev = ndev; - ndev->np = np; + if (!ndev->npinfo) { + npinfo = kmalloc(sizeof(*npinfo), GFP_KERNEL); + if (!npinfo) + goto release; + + npinfo->rx_np = NULL; + npinfo->poll_lock = SPIN_LOCK_UNLOCKED; + npinfo->poll_owner = -1; + npinfo->rx_lock = SPIN_LOCK_UNLOCKED; + } else + npinfo = ndev->npinfo; if (!ndev->poll_controller) { printk(KERN_ERR "%s: %s doesn't support polling, aborting.\n", @@ -692,13 +712,20 @@ int netpoll_setup(struct netpoll *np) np->name, HIPQUAD(np->local_ip)); } - if(np->rx_hook) - np->rx_flags = NETPOLL_RX_ENABLED; + if (np->rx_hook) { + spin_lock_irqsave(&npinfo->rx_lock, flags); + npinfo->rx_flags |= NETPOLL_RX_ENABLED; + npinfo->rx_np = np; + spin_unlock_irqrestore(&npinfo->rx_lock, flags); + } + /* last thing to do is link it to the net device structure */ + ndev->npinfo = npinfo; return 0; release: - ndev->np = NULL; + if (!ndev->npinfo) + kfree(npinfo); np->dev = NULL; dev_put(ndev); return -1; @@ -706,9 +733,20 @@ int netpoll_setup(struct netpoll *np) void netpoll_cleanup(struct netpoll *np) { - if (np->dev) - np->dev->np = NULL; - dev_put(np->dev); + struct netpoll_info *npinfo; + unsigned long flags; + + if (np->dev) { + npinfo = np->dev->npinfo; + if (npinfo && npinfo->rx_np == np) { + spin_lock_irqsave(&npinfo->rx_lock, flags); + npinfo->rx_np = NULL; + npinfo->rx_flags &= ~NETPOLL_RX_ENABLED; + spin_unlock_irqrestore(&npinfo->rx_lock, flags); + } + dev_put(np->dev); + } + np->dev = NULL; } diff --git a/net/core/request_sock.c b/net/core/request_sock.c new file mode 100644 index 0000000..bb55675 --- /dev/null +++ b/net/core/request_sock.c @@ -0,0 +1,64 @@ +/* + * NET Generic infrastructure for Network protocols. + * + * Authors: Arnaldo Carvalho de Melo <acme@conectiva.com.br> + * + * From code originally in include/net/tcp.h + * + * 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. + */ + +#include <linux/module.h> +#include <linux/random.h> +#include <linux/slab.h> +#include <linux/string.h> + +#include <net/request_sock.h> + +/* + * Maximum number of SYN_RECV sockets in queue per LISTEN socket. + * One SYN_RECV socket costs about 80bytes on a 32bit machine. + * It would be better to replace it with a global counter for all sockets + * but then some measure against one socket starving all other sockets + * would be needed. + * + * It was 128 by default. Experiments with real servers show, that + * it is absolutely not enough even at 100conn/sec. 256 cures most + * of problems. This value is adjusted to 128 for very small machines + * (<=32Mb of memory) and to 1024 on normal or better ones (>=256Mb). + * Further increasing requires to change hash table size. + */ +int sysctl_max_syn_backlog = 256; +EXPORT_SYMBOL(sysctl_max_syn_backlog); + +int reqsk_queue_alloc(struct request_sock_queue *queue, + const int nr_table_entries) +{ + const int lopt_size = sizeof(struct listen_sock) + + nr_table_entries * sizeof(struct request_sock *); + struct listen_sock *lopt = kmalloc(lopt_size, GFP_KERNEL); + + if (lopt == NULL) + return -ENOMEM; + + memset(lopt, 0, lopt_size); + + for (lopt->max_qlen_log = 6; + (1 << lopt->max_qlen_log) < sysctl_max_syn_backlog; + lopt->max_qlen_log++); + + get_random_bytes(&lopt->hash_rnd, sizeof(lopt->hash_rnd)); + rwlock_init(&queue->syn_wait_lock); + queue->rskq_accept_head = queue->rskq_accept_head = NULL; + + write_lock_bh(&queue->syn_wait_lock); + queue->listen_opt = lopt; + write_unlock_bh(&queue->syn_wait_lock); + + return 0; +} + +EXPORT_SYMBOL(reqsk_queue_alloc); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 00caf4b..e013d83 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -100,6 +100,7 @@ static const int rtm_min[RTM_NR_FAMILIES] = [RTM_FAM(RTM_NEWPREFIX)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), [RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), [RTM_FAM(RTM_GETANYCAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), + [RTM_FAM(RTM_NEWNEIGHTBL)] = NLMSG_LENGTH(sizeof(struct ndtmsg)), }; static const int rta_max[RTM_NR_FAMILIES] = @@ -113,6 +114,7 @@ static const int rta_max[RTM_NR_FAMILIES] = [RTM_FAM(RTM_NEWTCLASS)] = TCA_MAX, [RTM_FAM(RTM_NEWTFILTER)] = TCA_MAX, [RTM_FAM(RTM_NEWACTION)] = TCAA_MAX, + [RTM_FAM(RTM_NEWNEIGHTBL)] = NDTA_MAX, }; void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data) @@ -176,14 +178,14 @@ rtattr_failure: static int rtnetlink_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, - int type, u32 pid, u32 seq, u32 change) + int type, u32 pid, u32 seq, u32 change, + unsigned int flags) { struct ifinfomsg *r; struct nlmsghdr *nlh; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, type, sizeof(*r)); - if (pid) nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW(skb, pid, seq, type, sizeof(*r), flags); r = NLMSG_DATA(nlh); r->ifi_family = AF_UNSPEC; r->ifi_type = dev->type; @@ -273,7 +275,10 @@ static int rtnetlink_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *c for (dev=dev_base, idx=0; dev; dev = dev->next, idx++) { if (idx < s_idx) continue; - if (rtnetlink_fill_ifinfo(skb, dev, RTM_NEWLINK, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, 0) <= 0) + if (rtnetlink_fill_ifinfo(skb, dev, RTM_NEWLINK, + NETLINK_CB(cb->skb).pid, + cb->nlh->nlmsg_seq, 0, + NLM_F_MULTI) <= 0) break; } read_unlock(&dev_base_lock); @@ -447,7 +452,7 @@ void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change) if (!skb) return; - if (rtnetlink_fill_ifinfo(skb, dev, type, 0, 0, change) < 0) { + if (rtnetlink_fill_ifinfo(skb, dev, type, current->pid, 0, change, 0) < 0) { kfree_skb(skb); return; } @@ -649,14 +654,16 @@ static void rtnetlink_rcv(struct sock *sk, int len) static struct rtnetlink_link link_rtnetlink_table[RTM_NR_MSGTYPES] = { - [RTM_GETLINK - RTM_BASE] = { .dumpit = rtnetlink_dump_ifinfo }, - [RTM_SETLINK - RTM_BASE] = { .doit = do_setlink }, - [RTM_GETADDR - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, - [RTM_GETROUTE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, - [RTM_NEWNEIGH - RTM_BASE] = { .doit = neigh_add }, - [RTM_DELNEIGH - RTM_BASE] = { .doit = neigh_delete }, - [RTM_GETNEIGH - RTM_BASE] = { .dumpit = neigh_dump_info }, - [RTM_GETRULE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_GETLINK - RTM_BASE] = { .dumpit = rtnetlink_dump_ifinfo }, + [RTM_SETLINK - RTM_BASE] = { .doit = do_setlink }, + [RTM_GETADDR - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_GETROUTE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_NEWNEIGH - RTM_BASE] = { .doit = neigh_add }, + [RTM_DELNEIGH - RTM_BASE] = { .doit = neigh_delete }, + [RTM_GETNEIGH - RTM_BASE] = { .dumpit = neigh_dump_info }, + [RTM_GETRULE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_GETNEIGHTBL - RTM_BASE] = { .dumpit = neightbl_dump_info }, + [RTM_SETNEIGHTBL - RTM_BASE] = { .doit = neightbl_set }, }; static int rtnetlink_event(struct notifier_block *this, unsigned long event, void *ptr) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index f65b3de..bb73b21 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); @@ -1506,6 +1500,159 @@ void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len) skb_split_no_header(skb, skb1, len, pos); } +/** + * skb_prepare_seq_read - Prepare a sequential read of skb data + * @skb: the buffer to read + * @from: lower offset of data to be read + * @to: upper offset of data to be read + * @st: state variable + * + * Initializes the specified state variable. Must be called before + * invoking skb_seq_read() for the first time. + */ +void skb_prepare_seq_read(struct sk_buff *skb, unsigned int from, + unsigned int to, struct skb_seq_state *st) +{ + st->lower_offset = from; + st->upper_offset = to; + st->root_skb = st->cur_skb = skb; + st->frag_idx = st->stepped_offset = 0; + st->frag_data = NULL; +} + +/** + * skb_seq_read - Sequentially read skb data + * @consumed: number of bytes consumed by the caller so far + * @data: destination pointer for data to be returned + * @st: state variable + * + * Reads a block of skb data at &consumed relative to the + * lower offset specified to skb_prepare_seq_read(). Assigns + * the head of the data block to &data and returns the length + * of the block or 0 if the end of the skb data or the upper + * offset has been reached. + * + * The caller is not required to consume all of the data + * returned, i.e. &consumed is typically set to the number + * of bytes already consumed and the next call to + * skb_seq_read() will return the remaining part of the block. + * + * Note: The size of each block of data returned can be arbitary, + * this limitation is the cost for zerocopy seqeuental + * reads of potentially non linear data. + * + * Note: Fragment lists within fragments are not implemented + * at the moment, state->root_skb could be replaced with + * a stack for this purpose. + */ +unsigned int skb_seq_read(unsigned int consumed, const u8 **data, + struct skb_seq_state *st) +{ + unsigned int block_limit, abs_offset = consumed + st->lower_offset; + skb_frag_t *frag; + + if (unlikely(abs_offset >= st->upper_offset)) + return 0; + +next_skb: + block_limit = skb_headlen(st->cur_skb); + + if (abs_offset < block_limit) { + *data = st->cur_skb->data + abs_offset; + return block_limit - abs_offset; + } + + if (st->frag_idx == 0 && !st->frag_data) + st->stepped_offset += skb_headlen(st->cur_skb); + + while (st->frag_idx < skb_shinfo(st->cur_skb)->nr_frags) { + frag = &skb_shinfo(st->cur_skb)->frags[st->frag_idx]; + block_limit = frag->size + st->stepped_offset; + + if (abs_offset < block_limit) { + if (!st->frag_data) + st->frag_data = kmap_skb_frag(frag); + + *data = (u8 *) st->frag_data + frag->page_offset + + (abs_offset - st->stepped_offset); + + return block_limit - abs_offset; + } + + if (st->frag_data) { + kunmap_skb_frag(st->frag_data); + st->frag_data = NULL; + } + + st->frag_idx++; + st->stepped_offset += frag->size; + } + + if (st->cur_skb->next) { + st->cur_skb = st->cur_skb->next; + st->frag_idx = 0; + goto next_skb; + } else if (st->root_skb == st->cur_skb && + skb_shinfo(st->root_skb)->frag_list) { + st->cur_skb = skb_shinfo(st->root_skb)->frag_list; + goto next_skb; + } + + return 0; +} + +/** + * skb_abort_seq_read - Abort a sequential read of skb data + * @st: state variable + * + * Must be called if skb_seq_read() was not called until it + * returned 0. + */ +void skb_abort_seq_read(struct skb_seq_state *st) +{ + if (st->frag_data) + kunmap_skb_frag(st->frag_data); +} + +#define TS_SKB_CB(state) ((struct skb_seq_state *) &((state)->cb)) + +static unsigned int skb_ts_get_next_block(unsigned int offset, const u8 **text, + struct ts_config *conf, + struct ts_state *state) +{ + return skb_seq_read(offset, text, TS_SKB_CB(state)); +} + +static void skb_ts_finish(struct ts_config *conf, struct ts_state *state) +{ + skb_abort_seq_read(TS_SKB_CB(state)); +} + +/** + * skb_find_text - Find a text pattern in skb data + * @skb: the buffer to look in + * @from: search offset + * @to: search limit + * @config: textsearch configuration + * @state: uninitialized textsearch state variable + * + * Finds a pattern in the skb data according to the specified + * textsearch configuration. Use textsearch_next() to retrieve + * subsequent occurrences of the pattern. Returns the offset + * to the first occurrence or UINT_MAX if no match was found. + */ +unsigned int skb_find_text(struct sk_buff *skb, unsigned int from, + unsigned int to, struct ts_config *config, + struct ts_state *state) +{ + config->get_next_block = skb_ts_get_next_block; + config->finish = skb_ts_finish; + + skb_prepare_seq_read(skb, from, to, TS_SKB_CB(state)); + + return textsearch_find(config, state); +} + void __init skb_init(void) { skbuff_head_cache = kmem_cache_create("skbuff_head_cache", @@ -1544,3 +1691,7 @@ EXPORT_SYMBOL(skb_queue_tail); EXPORT_SYMBOL(skb_unlink); EXPORT_SYMBOL(skb_append); EXPORT_SYMBOL(skb_split); +EXPORT_SYMBOL(skb_prepare_seq_read); +EXPORT_SYMBOL(skb_seq_read); +EXPORT_SYMBOL(skb_abort_seq_read); +EXPORT_SYMBOL(skb_find_text); diff --git a/net/core/sock.c b/net/core/sock.c index 96e00b0..a6ec3ad 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -118,6 +118,7 @@ #include <linux/netdevice.h> #include <net/protocol.h> #include <linux/skbuff.h> +#include <net/request_sock.h> #include <net/sock.h> #include <net/xfrm.h> #include <linux/ipsec.h> @@ -1363,6 +1364,7 @@ static LIST_HEAD(proto_list); int proto_register(struct proto *prot, int alloc_slab) { + char *request_sock_slab_name; int rc = -ENOBUFS; if (alloc_slab) { @@ -1374,6 +1376,25 @@ int proto_register(struct proto *prot, int alloc_slab) prot->name); goto out; } + + if (prot->rsk_prot != NULL) { + static const char mask[] = "request_sock_%s"; + + request_sock_slab_name = kmalloc(strlen(prot->name) + sizeof(mask) - 1, GFP_KERNEL); + if (request_sock_slab_name == NULL) + goto out_free_sock_slab; + + sprintf(request_sock_slab_name, mask, prot->name); + prot->rsk_prot->slab = kmem_cache_create(request_sock_slab_name, + prot->rsk_prot->obj_size, 0, + SLAB_HWCACHE_ALIGN, NULL, NULL); + + if (prot->rsk_prot->slab == NULL) { + printk(KERN_CRIT "%s: Can't create request sock SLAB cache!\n", + prot->name); + goto out_free_request_sock_slab_name; + } + } } write_lock(&proto_list_lock); @@ -1382,6 +1403,12 @@ int proto_register(struct proto *prot, int alloc_slab) rc = 0; out: return rc; +out_free_request_sock_slab_name: + kfree(request_sock_slab_name); +out_free_sock_slab: + kmem_cache_destroy(prot->slab); + prot->slab = NULL; + goto out; } EXPORT_SYMBOL(proto_register); @@ -1395,6 +1422,14 @@ void proto_unregister(struct proto *prot) prot->slab = NULL; } + if (prot->rsk_prot != NULL && prot->rsk_prot->slab != NULL) { + const char *name = kmem_cache_name(prot->rsk_prot->slab); + + kmem_cache_destroy(prot->rsk_prot->slab); + kfree(name); + prot->rsk_prot->slab = NULL; + } + list_del(&prot->node); write_unlock(&proto_list_lock); } diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index c8be646c..8f817ad 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -13,12 +13,8 @@ #ifdef CONFIG_SYSCTL extern int netdev_max_backlog; +extern int netdev_budget; extern int weight_p; -extern int no_cong_thresh; -extern int no_cong; -extern int lo_cong; -extern int mod_cong; -extern int netdev_fastroute; extern int net_msg_cost; extern int net_msg_burst; @@ -35,19 +31,6 @@ extern int sysctl_somaxconn; extern char sysctl_divert_version[]; #endif /* CONFIG_NET_DIVERT */ -/* - * This strdup() is used for creating copies of network - * device names to be handed over to sysctl. - */ - -char *net_sysctl_strdup(const char *s) -{ - char *rv = kmalloc(strlen(s)+1, GFP_KERNEL); - if (rv) - strcpy(rv, s); - return rv; -} - ctl_table core_table[] = { #ifdef CONFIG_NET { @@ -99,38 +82,6 @@ ctl_table core_table[] = { .proc_handler = &proc_dointvec }, { - .ctl_name = NET_CORE_NO_CONG_THRESH, - .procname = "no_cong_thresh", - .data = &no_cong_thresh, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { - .ctl_name = NET_CORE_NO_CONG, - .procname = "no_cong", - .data = &no_cong, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { - .ctl_name = NET_CORE_LO_CONG, - .procname = "lo_cong", - .data = &lo_cong, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { - .ctl_name = NET_CORE_MOD_CONG, - .procname = "mod_cong", - .data = &mod_cong, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { .ctl_name = NET_CORE_MSG_COST, .procname = "message_cost", .data = &net_msg_cost, @@ -174,9 +125,15 @@ ctl_table core_table[] = { .mode = 0644, .proc_handler = &proc_dointvec }, + { + .ctl_name = NET_CORE_BUDGET, + .procname = "netdev_budget", + .data = &netdev_budget, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, { .ctl_name = 0 } }; -EXPORT_SYMBOL(net_sysctl_strdup); - #endif diff --git a/net/core/wireless.c b/net/core/wireless.c index 750cc5d..b2fe378 100644 --- a/net/core/wireless.c +++ b/net/core/wireless.c @@ -2,7 +2,7 @@ * This file implement the Wireless Extensions APIs. * * Authors : Jean Tourrilhes - HPL - <jt@hpl.hp.com> - * Copyright (c) 1997-2004 Jean Tourrilhes, All Rights Reserved. + * Copyright (c) 1997-2005 Jean Tourrilhes, All Rights Reserved. * * (As all part of the Linux kernel, this file is GPL) */ @@ -187,6 +187,12 @@ static const struct iw_ioctl_description standard_ioctl[] = { .header_type = IW_HEADER_TYPE_ADDR, .flags = IW_DESCR_FLAG_DUMP, }, + [SIOCSIWMLME - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_mlme), + .max_tokens = sizeof(struct iw_mlme), + }, [SIOCGIWAPLIST - SIOCIWFIRST] = { .header_type = IW_HEADER_TYPE_POINT, .token_size = sizeof(struct sockaddr) + @@ -195,7 +201,10 @@ static const struct iw_ioctl_description standard_ioctl[] = { .flags = IW_DESCR_FLAG_NOMAX, }, [SIOCSIWSCAN - SIOCIWFIRST] = { - .header_type = IW_HEADER_TYPE_PARAM, + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = 0, + .max_tokens = sizeof(struct iw_scan_req), }, [SIOCGIWSCAN - SIOCIWFIRST] = { .header_type = IW_HEADER_TYPE_POINT, @@ -273,6 +282,42 @@ static const struct iw_ioctl_description standard_ioctl[] = { [SIOCGIWPOWER - SIOCIWFIRST] = { .header_type = IW_HEADER_TYPE_PARAM, }, + [SIOCSIWGENIE - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [SIOCGIWGENIE - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [SIOCSIWAUTH - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_PARAM, + }, + [SIOCGIWAUTH - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_PARAM, + }, + [SIOCSIWENCODEEXT - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_encode_ext), + .max_tokens = sizeof(struct iw_encode_ext) + + IW_ENCODING_TOKEN_MAX, + }, + [SIOCGIWENCODEEXT - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_encode_ext), + .max_tokens = sizeof(struct iw_encode_ext) + + IW_ENCODING_TOKEN_MAX, + }, + [SIOCSIWPMKSA - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_pmksa), + .max_tokens = sizeof(struct iw_pmksa), + }, }; static const int standard_ioctl_num = (sizeof(standard_ioctl) / sizeof(struct iw_ioctl_description)); @@ -299,6 +344,31 @@ static const struct iw_ioctl_description standard_event[] = { [IWEVEXPIRED - IWEVFIRST] = { .header_type = IW_HEADER_TYPE_ADDR, }, + [IWEVGENIE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [IWEVMICHAELMICFAILURE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = sizeof(struct iw_michaelmicfailure), + }, + [IWEVASSOCREQIE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [IWEVASSOCRESPIE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [IWEVPMKIDCAND - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = sizeof(struct iw_pmkid_cand), + }, }; static const int standard_event_num = (sizeof(standard_event) / sizeof(struct iw_ioctl_description)); diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index ee7bf46..00233ec 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -716,13 +716,13 @@ static int dn_dev_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *a } static int dn_dev_fill_ifaddr(struct sk_buff *skb, struct dn_ifaddr *ifa, - u32 pid, u32 seq, int event) + u32 pid, u32 seq, int event, unsigned int flags) { struct ifaddrmsg *ifm; struct nlmsghdr *nlh; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*ifm)); + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*ifm), flags); ifm = NLMSG_DATA(nlh); ifm->ifa_family = AF_DECnet; @@ -755,7 +755,7 @@ static void rtmsg_ifa(int event, struct dn_ifaddr *ifa) netlink_set_err(rtnl, 0, RTMGRP_DECnet_IFADDR, ENOBUFS); return; } - if (dn_dev_fill_ifaddr(skb, ifa, 0, 0, event) < 0) { + if (dn_dev_fill_ifaddr(skb, ifa, 0, 0, event, 0) < 0) { kfree_skb(skb); netlink_set_err(rtnl, 0, RTMGRP_DECnet_IFADDR, EINVAL); return; @@ -790,7 +790,8 @@ static int dn_dev_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb) if (dn_dev_fill_ifaddr(skb, ifa, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, - RTM_NEWADDR) <= 0) + RTM_NEWADDR, + NLM_F_MULTI) <= 0) goto done; } } diff --git a/net/decnet/dn_neigh.c b/net/decnet/dn_neigh.c index f6dfe96..f32dba9 100644 --- a/net/decnet/dn_neigh.c +++ b/net/decnet/dn_neigh.c @@ -101,7 +101,6 @@ struct neigh_table dn_neigh_table = { .id = "dn_neigh_cache", .parms ={ .tbl = &dn_neigh_table, - .entries = 0, .base_reachable_time = 30 * HZ, .retrans_time = 1 * HZ, .gc_staletime = 60 * HZ, diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 1e7b5c3..2399fa8 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -1465,7 +1465,8 @@ int dn_route_input(struct sk_buff *skb) return dn_route_input_slow(skb); } -static int dn_rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, int nowait) +static int dn_rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, + int event, int nowait, unsigned int flags) { struct dn_route *rt = (struct dn_route *)skb->dst; struct rtmsg *r; @@ -1473,9 +1474,8 @@ static int dn_rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, int unsigned char *b = skb->tail; struct rta_cacheinfo ci; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*r)); + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*r), flags); r = NLMSG_DATA(nlh); - nlh->nlmsg_flags = (nowait && pid) ? NLM_F_MULTI : 0; r->rtm_family = AF_DECnet; r->rtm_dst_len = 16; r->rtm_src_len = 0; @@ -1596,7 +1596,7 @@ int dn_cache_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, void *arg) NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid; - err = dn_rt_fill_info(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, RTM_NEWROUTE, 0); + err = dn_rt_fill_info(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, RTM_NEWROUTE, 0, 0); if (err == 0) goto out_free; @@ -1644,7 +1644,8 @@ int dn_cache_dump(struct sk_buff *skb, struct netlink_callback *cb) continue; skb->dst = dst_clone(&rt->u.dst); if (dn_rt_fill_info(skb, NETLINK_CB(cb->skb).pid, - cb->nlh->nlmsg_seq, RTM_NEWROUTE, 1) <= 0) { + cb->nlh->nlmsg_seq, RTM_NEWROUTE, + 1, NLM_F_MULTI) <= 0) { dst_release(xchg(&skb->dst, NULL)); rcu_read_unlock_bh(); goto done; diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c index 597587d..1060de7 100644 --- a/net/decnet/dn_rules.c +++ b/net/decnet/dn_rules.c @@ -342,14 +342,15 @@ static struct notifier_block dn_fib_rules_notifier = { .notifier_call = dn_fib_rules_event, }; -static int dn_fib_fill_rule(struct sk_buff *skb, struct dn_fib_rule *r, struct netlink_callback *cb) +static int dn_fib_fill_rule(struct sk_buff *skb, struct dn_fib_rule *r, + struct netlink_callback *cb, unsigned int flags) { struct rtmsg *rtm; struct nlmsghdr *nlh; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, NETLINK_CREDS(cb->skb)->pid, cb->nlh->nlmsg_seq, RTM_NEWRULE, sizeof(*rtm)); + nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWRULE, sizeof(*rtm), flags); rtm = NLMSG_DATA(nlh); rtm->rtm_family = AF_DECnet; rtm->rtm_dst_len = r->r_dst_len; @@ -394,7 +395,7 @@ int dn_fib_dump_rules(struct sk_buff *skb, struct netlink_callback *cb) for(r = dn_fib_rules, idx = 0; r; r = r->r_next, idx++) { if (idx < s_idx) continue; - if (dn_fib_fill_rule(skb, r, cb) < 0) + if (dn_fib_fill_rule(skb, r, cb, NLM_F_MULTI) < 0) break; } read_unlock(&dn_fib_rules_lock); diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c index dad5603..28ba5777a 100644 --- a/net/decnet/dn_table.c +++ b/net/decnet/dn_table.c @@ -270,13 +270,13 @@ static int dn_fib_nh_match(struct rtmsg *r, struct nlmsghdr *nlh, struct dn_kern static int dn_fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event, u8 tb_id, u8 type, u8 scope, void *dst, int dst_len, - struct dn_fib_info *fi) + struct dn_fib_info *fi, unsigned int flags) { struct rtmsg *rtm; struct nlmsghdr *nlh; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*rtm)); + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*rtm), flags); rtm = NLMSG_DATA(nlh); rtm->rtm_family = AF_DECnet; rtm->rtm_dst_len = dst_len; @@ -345,7 +345,7 @@ static void dn_rtmsg_fib(int event, struct dn_fib_node *f, int z, int tb_id, if (dn_fib_dump_info(skb, pid, nlh->nlmsg_seq, event, tb_id, f->fn_type, f->fn_scope, &f->fn_key, z, - DN_FIB_INFO(f)) < 0) { + DN_FIB_INFO(f), 0) < 0) { kfree_skb(skb); return; } @@ -377,7 +377,7 @@ static __inline__ int dn_hash_dump_bucket(struct sk_buff *skb, tb->n, (f->fn_state & DN_S_ZOMBIE) ? 0 : f->fn_type, f->fn_scope, &f->fn_key, dz->dz_order, - f->fn_info) < 0) { + f->fn_info, NLM_F_MULTI) < 0) { cb->args[3] = i; return -1; } diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 6d3e8b1..3470834 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -53,6 +53,44 @@ config IP_ADVANCED_ROUTER If unsure, say N here. +choice + prompt "Choose IP: FIB lookup algorithm (choose FIB_HASH if unsure)" + depends on IP_ADVANCED_ROUTER + 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 if you have a large + number of routes. + + LC-trie is a longest matching prefix lookup algorithm which + performs better than FIB_HASH for large routing tables. + But, it consumes more memory and is more complex. + + 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 + +# If the user does not enable advanced routing, he gets the safe +# default of the fib-hash algorithm. +config IP_FIB_HASH + bool + depends on !IP_ADVANCED_ROUTER + default y + config IP_MULTIPLE_TABLES bool "IP: policy routing" depends on IP_ADVANCED_ROUTER @@ -407,5 +445,113 @@ config IP_TCPDIAG config IP_TCPDIAG_IPV6 def_bool (IP_TCPDIAG=y && IPV6=y) || (IP_TCPDIAG=m && IPV6) +config TCP_CONG_ADVANCED + bool "TCP: advanced congestion control" + depends on INET + default y + ---help--- + Support for selection of various TCP congestion control + modules. + + Nearly all users can safely say no here, and a safe default + selection will be made (BIC-TCP with new Reno as a fallback). + + If unsure, say N. + +# TCP Reno is builtin (required as fallback) +menu "TCP congestion control" + depends on TCP_CONG_ADVANCED + +config TCP_CONG_BIC + tristate "Binary Increase Congestion (BIC) control" + depends on INET + default y + ---help--- + BIC-TCP is a sender-side only change that ensures a linear RTT + fairness under large windows while offering both scalability and + bounded TCP-friendliness. The protocol combines two schemes + called additive increase and binary search increase. When the + congestion window is large, additive increase with a large + increment ensures linear RTT fairness as well as good + scalability. Under small congestion windows, binary search + increase provides TCP friendliness. + See http://www.csc.ncsu.edu/faculty/rhee/export/bitcp/ + +config TCP_CONG_WESTWOOD + tristate "TCP Westwood+" + depends on INET + default m + ---help--- + TCP Westwood+ is a sender-side only modification of the TCP Reno + protocol stack that optimizes the performance of TCP congestion + control. It is based on end-to-end bandwidth estimation to set + congestion window and slow start threshold after a congestion + episode. Using this estimation, TCP Westwood+ adaptively sets a + slow start threshold and a congestion window which takes into + account the bandwidth used at the time congestion is experienced. + TCP Westwood+ significantly increases fairness wrt TCP Reno in + wired networks and throughput over wireless links. + +config TCP_CONG_HTCP + tristate "H-TCP" + depends on INET + default m + ---help--- + H-TCP is a send-side only modifications of the TCP Reno + protocol stack that optimizes the performance of TCP + congestion control for high speed network links. It uses a + modeswitch to change the alpha and beta parameters of TCP Reno + based on network conditions and in a way so as to be fair with + other Reno and H-TCP flows. + +config TCP_CONG_HSTCP + tristate "High Speed TCP" + depends on INET && EXPERIMENTAL + default n + ---help--- + Sally Floyd's High Speed TCP (RFC 3649) congestion control. + A modification to TCP's congestion control mechanism for use + with large congestion windows. A table indicates how much to + increase the congestion window by when an ACK is received. + For more detail see http://www.icir.org/floyd/hstcp.html + +config TCP_CONG_HYBLA + tristate "TCP-Hybla congestion control algorithm" + depends on INET && EXPERIMENTAL + default n + ---help--- + TCP-Hybla is a sender-side only change that eliminates penalization of + long-RTT, large-bandwidth connections, like when satellite legs are + involved, expecially when sharing a common bottleneck with normal + terrestrial connections. + +config TCP_CONG_VEGAS + tristate "TCP Vegas" + depends on INET && EXPERIMENTAL + default n + ---help--- + TCP Vegas is a sender-side only change to TCP that anticipates + the onset of congestion by estimating the bandwidth. TCP Vegas + adjusts the sending rate by modifying the congestion + window. TCP Vegas should provide less packet loss, but it is + not as aggressive as TCP Reno. + +config TCP_CONG_SCALABLE + tristate "Scalable TCP" + depends on INET && EXPERIMENTAL + default n + ---help--- + Scalable TCP is a sender-side only change to TCP which uses a + MIMD congestion control algorithm which has some nice scaling + properties, though is known to have fairness issues. + See http://www-lce.eng.cam.ac.uk/~ctk21/scalable/ + +endmenu + +config TCP_CONG_BIC + boolean + depends on !TCP_CONG_ADVANCED + default y + source "net/ipv4/ipvs/Kconfig" diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index 8b379627..5718cdb 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -5,10 +5,13 @@ obj-y := utils.o route.o inetpeer.o protocol.o \ ip_input.o ip_fragment.o ip_forward.o ip_options.o \ ip_output.o ip_sockglue.o \ - tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o tcp_minisocks.o \ + tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \ + tcp_minisocks.o tcp_cong.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 @@ -28,6 +31,13 @@ obj-$(CONFIG_NETFILTER) += netfilter/ obj-$(CONFIG_IP_VS) += ipvs/ obj-$(CONFIG_IP_TCPDIAG) += tcp_diag.o obj-$(CONFIG_IP_ROUTE_MULTIPATH_CACHED) += multipath.o +obj-$(CONFIG_TCP_CONG_BIC) += tcp_bic.o +obj-$(CONFIG_TCP_CONG_WESTWOOD) += tcp_westwood.o +obj-$(CONFIG_TCP_CONG_HSTCP) += tcp_highspeed.o +obj-$(CONFIG_TCP_CONG_HYBLA) += tcp_hybla.o +obj-$(CONFIG_TCP_CONG_HTCP) += tcp_htcp.o +obj-$(CONFIG_TCP_CONG_VEGAS) += tcp_vegas.o +obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \ xfrm4_output.o diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index b3cb49c..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(); @@ -1181,6 +1193,7 @@ EXPORT_SYMBOL(inet_stream_connect); EXPORT_SYMBOL(inet_stream_ops); EXPORT_SYMBOL(inet_unregister_protosw); EXPORT_SYMBOL(net_statistics); +EXPORT_SYMBOL(sysctl_ip_nonlocal_bind); #ifdef INET_REFCNT_DEBUG EXPORT_SYMBOL(inet_sock_nr); 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/devinet.c b/net/ipv4/devinet.c index 3cc9673..d8a10e3 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -233,11 +233,14 @@ int inet_addr_onlink(struct in_device *in_dev, u32 a, u32 b) static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, int destroy) { + struct in_ifaddr *promote = NULL; struct in_ifaddr *ifa1 = *ifap; ASSERT_RTNL(); - /* 1. Deleting primary ifaddr forces deletion all secondaries */ + /* 1. Deleting primary ifaddr forces deletion all secondaries + * unless alias promotion is set + **/ if (!(ifa1->ifa_flags & IFA_F_SECONDARY)) { struct in_ifaddr *ifa; @@ -251,11 +254,16 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, continue; } - *ifap1 = ifa->ifa_next; + if (!IN_DEV_PROMOTE_SECONDARIES(in_dev)) { + *ifap1 = ifa->ifa_next; - rtmsg_ifa(RTM_DELADDR, ifa); - notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa); - inet_free_ifa(ifa); + rtmsg_ifa(RTM_DELADDR, ifa); + notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa); + inet_free_ifa(ifa); + } else { + promote = ifa; + break; + } } } @@ -281,6 +289,13 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, if (!in_dev->ifa_list) inetdev_destroy(in_dev); } + + if (promote && IN_DEV_PROMOTE_SECONDARIES(in_dev)) { + /* not sure if we should send a delete notify first? */ + promote->ifa_flags &= ~IFA_F_SECONDARY; + rtmsg_ifa(RTM_NEWADDR, promote); + notifier_call_chain(&inetaddr_chain, NETDEV_UP, promote); + } } static int inet_insert_ifa(struct in_ifaddr *ifa) @@ -1015,14 +1030,13 @@ static struct notifier_block ip_netdev_notifier = { }; static int inet_fill_ifaddr(struct sk_buff *skb, struct in_ifaddr *ifa, - u32 pid, u32 seq, int event) + u32 pid, u32 seq, int event, unsigned int flags) { struct ifaddrmsg *ifm; struct nlmsghdr *nlh; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*ifm)); - if (pid) nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*ifm), flags); ifm = NLMSG_DATA(nlh); ifm->ifa_family = AF_INET; ifm->ifa_prefixlen = ifa->ifa_prefixlen; @@ -1075,7 +1089,7 @@ static int inet_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb) continue; if (inet_fill_ifaddr(skb, ifa, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, - RTM_NEWADDR) <= 0) { + RTM_NEWADDR, NLM_F_MULTI) <= 0) { rcu_read_unlock(); goto done; } @@ -1098,7 +1112,7 @@ static void rtmsg_ifa(int event, struct in_ifaddr* ifa) if (!skb) netlink_set_err(rtnl, 0, RTMGRP_IPV4_IFADDR, ENOBUFS); - else if (inet_fill_ifaddr(skb, ifa, 0, 0, event) < 0) { + else if (inet_fill_ifaddr(skb, ifa, current->pid, 0, event, 0) < 0) { kfree_skb(skb); netlink_set_err(rtnl, 0, RTMGRP_IPV4_IFADDR, EINVAL); } else { @@ -1384,6 +1398,15 @@ static struct devinet_sysctl_table { .proc_handler = &ipv4_doint_and_flush, .strategy = &ipv4_doint_and_flush_strategy, }, + { + .ctl_name = NET_IPV4_CONF_PROMOTE_SECONDARIES, + .procname = "promote_secondaries", + .data = &ipv4_devconf.promote_secondaries, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &ipv4_doint_and_flush, + .strategy = &ipv4_doint_and_flush_strategy, + }, }, .devinet_dev = { { @@ -1448,7 +1471,7 @@ static void devinet_sysctl_register(struct in_device *in_dev, * by sysctl and we wouldn't want anyone to change it under our feet * (see SIOCSIFNAME). */ - dev_name = net_sysctl_strdup(dev_name); + dev_name = kstrdup(dev_name, GFP_KERNEL); if (!dev_name) goto free; diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 053a883..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; @@ -478,7 +478,7 @@ static int __init esp4_init(void) { struct xfrm_decap_state decap; - if (sizeof(struct esp_decap_data) < + if (sizeof(struct esp_decap_data) > sizeof(decap.decap_data)) { extern void decap_data_too_small(void); 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_hash.c b/net/ipv4/fib_hash.c index 6506dcc..b10d6bb 100644 --- a/net/ipv4/fib_hash.c +++ b/net/ipv4/fib_hash.c @@ -703,7 +703,8 @@ fn_hash_dump_bucket(struct sk_buff *skb, struct netlink_callback *cb, &f->fn_key, fz->fz_order, fa->fa_tos, - fa->fa_info) < 0) { + fa->fa_info, + NLM_F_MULTI) < 0) { cb->args[3] = i; return -1; } diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h index ac4485f..b729d97 100644 --- a/net/ipv4/fib_lookup.h +++ b/net/ipv4/fib_lookup.h @@ -30,7 +30,8 @@ extern int fib_nh_match(struct rtmsg *r, struct nlmsghdr *, struct kern_rta *rta, struct fib_info *fi); extern int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event, u8 tb_id, u8 type, u8 scope, void *dst, - int dst_len, u8 tos, struct fib_info *fi); + int dst_len, u8 tos, struct fib_info *fi, + unsigned int); 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); diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index 39d0aad..0b298bb 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -367,13 +367,14 @@ static struct notifier_block fib_rules_notifier = { static __inline__ int inet_fill_rule(struct sk_buff *skb, struct fib_rule *r, - struct netlink_callback *cb) + struct netlink_callback *cb, + unsigned int flags) { struct rtmsg *rtm; struct nlmsghdr *nlh; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, NETLINK_CREDS(cb->skb)->pid, cb->nlh->nlmsg_seq, RTM_NEWRULE, sizeof(*rtm)); + nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWRULE, sizeof(*rtm), flags); rtm = NLMSG_DATA(nlh); rtm->rtm_family = AF_INET; rtm->rtm_dst_len = r->r_dst_len; @@ -422,7 +423,7 @@ int inet_dump_rules(struct sk_buff *skb, struct netlink_callback *cb) for (r=fib_rules, idx=0; r; r = r->r_next, idx++) { if (idx < s_idx) continue; - if (inet_fill_rule(skb, r, cb) < 0) + if (inet_fill_rule(skb, r, cb, NLM_F_MULTI) < 0) break; } read_unlock(&fib_rules_lock); diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 029362d..c886b28 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -276,7 +276,7 @@ void rtmsg_fib(int event, u32 key, struct fib_alias *fa, struct nlmsghdr *n, struct netlink_skb_parms *req) { struct sk_buff *skb; - u32 pid = req ? req->pid : 0; + u32 pid = req ? req->pid : n->nlmsg_pid; int size = NLMSG_SPACE(sizeof(struct rtmsg)+256); skb = alloc_skb(size, GFP_KERNEL); @@ -286,7 +286,7 @@ void rtmsg_fib(int event, u32 key, struct fib_alias *fa, if (fib_dump_info(skb, pid, n->nlmsg_seq, event, tb_id, fa->fa_type, fa->fa_scope, &key, z, fa->fa_tos, - fa->fa_info) < 0) { + fa->fa_info, 0) < 0) { kfree_skb(skb); return; } @@ -932,13 +932,13 @@ u32 __fib_res_prefsrc(struct fib_result *res) int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event, u8 tb_id, u8 type, u8 scope, void *dst, int dst_len, u8 tos, - struct fib_info *fi) + struct fib_info *fi, unsigned int flags) { struct rtmsg *rtm; struct nlmsghdr *nlh; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*rtm)); + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*rtm), flags); rtm = NLMSG_DATA(nlh); rtm->rtm_family = AF_INET; rtm->rtm_dst_len = dst_len; @@ -1035,7 +1035,7 @@ fib_convert_rtentry(int cmd, struct nlmsghdr *nl, struct rtmsg *rtm, } nl->nlmsg_flags = NLM_F_REQUEST; - nl->nlmsg_pid = 0; + nl->nlmsg_pid = current->pid; nl->nlmsg_seq = 0; nl->nlmsg_len = NLMSG_LENGTH(sizeof(*rtm)); if (cmd == SIOCDELRT) { 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/icmp.c b/net/ipv4/icmp.c index 85bf0d3..cb75948 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -207,6 +207,7 @@ int sysctl_icmp_ignore_bogus_error_responses; int sysctl_icmp_ratelimit = 1 * HZ; int sysctl_icmp_ratemask = 0x1818; +int sysctl_icmp_errors_use_inbound_ifaddr; /* * ICMP control array. This specifies what to do with each ICMP. @@ -511,8 +512,12 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, u32 info) */ saddr = iph->daddr; - if (!(rt->rt_flags & RTCF_LOCAL)) - saddr = 0; + if (!(rt->rt_flags & RTCF_LOCAL)) { + if (sysctl_icmp_errors_use_inbound_ifaddr) + saddr = inet_select_addr(skb_in->dev, 0, RT_SCOPE_LINK); + else + saddr = 0; + } tos = icmp_pointers[type].error ? ((iph->tos & IPTOS_TOS_MASK) | IPTOS_PREC_INTERNETCONTROL) : 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 daebd93..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 } @@ -490,6 +479,14 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*)) /* Partially cloned skb? */ if (skb_shared(frag)) goto slow_path; + + BUG_ON(frag->sk); + if (skb->sk) { + sock_hold(skb->sk); + frag->sk = skb->sk; + frag->destructor = sock_wfree; + skb->truesize -= frag->truesize; + } } /* Everything is OK. Generate! */ diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index 47012b9..f8b172f 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -360,14 +360,14 @@ int ip_recv_error(struct sock *sk, struct msghdr *msg, int len) err = copied; /* Reset and regenerate socket error */ - spin_lock_irq(&sk->sk_error_queue.lock); + spin_lock_bh(&sk->sk_error_queue.lock); sk->sk_err = 0; if ((skb2 = skb_peek(&sk->sk_error_queue)) != NULL) { sk->sk_err = SKB_EXT_ERR(skb2)->ee.ee_errno; - spin_unlock_irq(&sk->sk_error_queue.lock); + spin_unlock_bh(&sk->sk_error_queue.lock); sk->sk_error_report(sk); } else - spin_unlock_irq(&sk->sk_error_queue.lock); + spin_unlock_bh(&sk->sk_error_queue.lock); out_free_skb: kfree_skb(skb); 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/Makefile b/net/ipv4/ipvs/Makefile index a788461..30e85de 100644 --- a/net/ipv4/ipvs/Makefile +++ b/net/ipv4/ipvs/Makefile @@ -11,7 +11,7 @@ ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_AH) += ip_vs_proto_ah.o ip_vs-objs := ip_vs_conn.o ip_vs_core.o ip_vs_ctl.o ip_vs_sched.o \ ip_vs_xmit.o ip_vs_app.o ip_vs_sync.o \ - ip_vs_est.o ip_vs_proto.o ip_vs_proto_icmp.o \ + ip_vs_est.o ip_vs_proto.o \ $(ip_vs_proto-objs-y) diff --git a/net/ipv4/ipvs/ip_vs_proto.c b/net/ipv4/ipvs/ip_vs_proto.c index 253c462..867d4e9 100644 --- a/net/ipv4/ipvs/ip_vs_proto.c +++ b/net/ipv4/ipvs/ip_vs_proto.c @@ -216,9 +216,6 @@ int ip_vs_protocol_init(void) #ifdef CONFIG_IP_VS_PROTO_UDP REGISTER_PROTOCOL(&ip_vs_protocol_udp); #endif -#ifdef CONFIG_IP_VS_PROTO_ICMP - REGISTER_PROTOCOL(&ip_vs_protocol_icmp); -#endif #ifdef CONFIG_IP_VS_PROTO_AH REGISTER_PROTOCOL(&ip_vs_protocol_ah); #endif diff --git a/net/ipv4/ipvs/ip_vs_proto_icmp.c b/net/ipv4/ipvs/ip_vs_proto_icmp.c deleted file mode 100644 index 191e94a..0000000 --- a/net/ipv4/ipvs/ip_vs_proto_icmp.c +++ /dev/null @@ -1,182 +0,0 @@ -/* - * ip_vs_proto_icmp.c: ICMP load balancing support for IP Virtual Server - * - * Authors: Julian Anastasov <ja@ssi.bg>, March 2002 - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * version 2 as published by the Free Software Foundation; - * - */ - -#include <linux/module.h> -#include <linux/kernel.h> -#include <linux/icmp.h> -#include <linux/netfilter.h> -#include <linux/netfilter_ipv4.h> - -#include <net/ip_vs.h> - - -static int icmp_timeouts[1] = { 1*60*HZ }; - -static char * icmp_state_name_table[1] = { "ICMP" }; - -static struct ip_vs_conn * -icmp_conn_in_get(const struct sk_buff *skb, - struct ip_vs_protocol *pp, - const struct iphdr *iph, - unsigned int proto_off, - int inverse) -{ -#if 0 - struct ip_vs_conn *cp; - - if (likely(!inverse)) { - cp = ip_vs_conn_in_get(iph->protocol, - iph->saddr, 0, - iph->daddr, 0); - } else { - cp = ip_vs_conn_in_get(iph->protocol, - iph->daddr, 0, - iph->saddr, 0); - } - - return cp; - -#else - return NULL; -#endif -} - -static struct ip_vs_conn * -icmp_conn_out_get(const struct sk_buff *skb, - struct ip_vs_protocol *pp, - const struct iphdr *iph, - unsigned int proto_off, - int inverse) -{ -#if 0 - struct ip_vs_conn *cp; - - if (likely(!inverse)) { - cp = ip_vs_conn_out_get(iph->protocol, - iph->saddr, 0, - iph->daddr, 0); - } else { - cp = ip_vs_conn_out_get(IPPROTO_UDP, - iph->daddr, 0, - iph->saddr, 0); - } - - return cp; -#else - return NULL; -#endif -} - -static int -icmp_conn_schedule(struct sk_buff *skb, struct ip_vs_protocol *pp, - int *verdict, struct ip_vs_conn **cpp) -{ - *verdict = NF_ACCEPT; - return 0; -} - -static int -icmp_csum_check(struct sk_buff *skb, struct ip_vs_protocol *pp) -{ - if (!(skb->nh.iph->frag_off & __constant_htons(IP_OFFSET))) { - if (skb->ip_summed != CHECKSUM_UNNECESSARY) { - if (ip_vs_checksum_complete(skb, skb->nh.iph->ihl * 4)) { - IP_VS_DBG_RL_PKT(0, pp, skb, 0, "Failed checksum for"); - return 0; - } - } - } - return 1; -} - -static void -icmp_debug_packet(struct ip_vs_protocol *pp, - const struct sk_buff *skb, - int offset, - const char *msg) -{ - char buf[256]; - struct iphdr _iph, *ih; - - ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph); - if (ih == NULL) - sprintf(buf, "%s TRUNCATED", pp->name); - else if (ih->frag_off & __constant_htons(IP_OFFSET)) - sprintf(buf, "%s %u.%u.%u.%u->%u.%u.%u.%u frag", - pp->name, NIPQUAD(ih->saddr), - NIPQUAD(ih->daddr)); - else { - struct icmphdr _icmph, *ic; - - ic = skb_header_pointer(skb, offset + ih->ihl*4, - sizeof(_icmph), &_icmph); - if (ic == NULL) - sprintf(buf, "%s TRUNCATED to %u bytes\n", - pp->name, skb->len - offset); - else - sprintf(buf, "%s %u.%u.%u.%u->%u.%u.%u.%u T:%d C:%d", - pp->name, NIPQUAD(ih->saddr), - NIPQUAD(ih->daddr), - ic->type, ic->code); - } - printk(KERN_DEBUG "IPVS: %s: %s\n", msg, buf); -} - -static int -icmp_state_transition(struct ip_vs_conn *cp, int direction, - const struct sk_buff *skb, - struct ip_vs_protocol *pp) -{ - cp->timeout = pp->timeout_table[IP_VS_ICMP_S_NORMAL]; - return 1; -} - -static int -icmp_set_state_timeout(struct ip_vs_protocol *pp, char *sname, int to) -{ - int num; - char **names; - - num = IP_VS_ICMP_S_LAST; - names = icmp_state_name_table; - return ip_vs_set_state_timeout(pp->timeout_table, num, names, sname, to); -} - - -static void icmp_init(struct ip_vs_protocol *pp) -{ - pp->timeout_table = icmp_timeouts; -} - -static void icmp_exit(struct ip_vs_protocol *pp) -{ -} - -struct ip_vs_protocol ip_vs_protocol_icmp = { - .name = "ICMP", - .protocol = IPPROTO_ICMP, - .dont_defrag = 0, - .init = icmp_init, - .exit = icmp_exit, - .conn_schedule = icmp_conn_schedule, - .conn_in_get = icmp_conn_in_get, - .conn_out_get = icmp_conn_out_get, - .snat_handler = NULL, - .dnat_handler = NULL, - .csum_check = icmp_csum_check, - .state_transition = icmp_state_transition, - .register_app = NULL, - .unregister_app = NULL, - .app_conn_bind = NULL, - .debug_packet = icmp_debug_packet, - .timeout_change = NULL, - .set_state_timeout = icmp_set_state_timeout, -}; diff --git a/net/ipv4/ipvs/ip_vs_xmit.c b/net/ipv4/ipvs/ip_vs_xmit.c index faa6176..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, \ @@ -508,7 +507,6 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, rc = NF_ACCEPT; /* do not touch skb anymore */ atomic_inc(&cp->in_pkts); - __ip_vs_conn_put(cp); goto out; } diff --git a/net/ipv4/multipath_drr.c b/net/ipv4/multipath_drr.c index 9349686..c9cf872 100644 --- a/net/ipv4/multipath_drr.c +++ b/net/ipv4/multipath_drr.c @@ -31,6 +31,7 @@ #include <linux/igmp.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> +#include <linux/module.h> #include <linux/mroute.h> #include <linux/init.h> #include <net/ip.h> @@ -57,7 +58,6 @@ struct multipath_device { static struct multipath_device state[MULTIPATH_MAX_DEVICECANDIDATES]; static DEFINE_SPINLOCK(state_lock); -static struct rtable *last_selection = NULL; static int inline __multipath_findslot(void) { @@ -111,11 +111,6 @@ struct notifier_block drr_dev_notifier = { .notifier_call = drr_dev_event, }; -static void drr_remove(struct rtable *rt) -{ - if (last_selection == rt) - last_selection = NULL; -} static void drr_safe_inc(atomic_t *usecount) { @@ -144,14 +139,6 @@ static void drr_select_route(const struct flowi *flp, int devidx = -1; int cur_min_devidx = -1; - /* if necessary and possible utilize the old alternative */ - if ((flp->flags & FLOWI_FLAG_MULTIPATHOLDROUTE) != 0 && - last_selection != NULL) { - result = last_selection; - *rp = result; - return; - } - /* 1. make sure all alt. nexthops have the same GC related data */ /* 2. determine the new candidate to be returned */ result = NULL; @@ -229,12 +216,10 @@ static void drr_select_route(const struct flowi *flp, } *rp = result; - last_selection = result; } static struct ip_mp_alg_ops drr_ops = { .mp_alg_select_route = drr_select_route, - .mp_alg_remove = drr_remove, }; static int __init drr_init(void) @@ -244,7 +229,7 @@ static int __init drr_init(void) if (err) return err; - err = multipath_alg_register(&drr_ops, IP_MP_ALG_RR); + err = multipath_alg_register(&drr_ops, IP_MP_ALG_DRR); if (err) goto fail; @@ -263,3 +248,4 @@ static void __exit drr_exit(void) module_init(drr_init); module_exit(drr_exit); +MODULE_LICENSE("GPL"); diff --git a/net/ipv4/multipath_random.c b/net/ipv4/multipath_random.c index 805a16e..5249dbe 100644 --- a/net/ipv4/multipath_random.c +++ b/net/ipv4/multipath_random.c @@ -31,6 +31,7 @@ #include <linux/igmp.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> +#include <linux/module.h> #include <linux/mroute.h> #include <linux/init.h> #include <net/ip.h> @@ -126,3 +127,4 @@ static void __exit random_exit(void) module_init(random_init); module_exit(random_exit); +MODULE_LICENSE("GPL"); diff --git a/net/ipv4/multipath_rr.c b/net/ipv4/multipath_rr.c index 554a8256..b6cd287 100644 --- a/net/ipv4/multipath_rr.c +++ b/net/ipv4/multipath_rr.c @@ -31,6 +31,7 @@ #include <linux/igmp.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> +#include <linux/module.h> #include <linux/mroute.h> #include <linux/init.h> #include <net/ip.h> @@ -47,29 +48,12 @@ #include <net/checksum.h> #include <net/ip_mp_alg.h> -#define MULTIPATH_MAX_CANDIDATES 40 - -static struct rtable* last_used = NULL; - -static void rr_remove(struct rtable *rt) -{ - if (last_used == rt) - last_used = NULL; -} - static void rr_select_route(const struct flowi *flp, struct rtable *first, struct rtable **rp) { struct rtable *nh, *result, *min_use_cand = NULL; int min_use = -1; - /* if necessary and possible utilize the old alternative */ - if ((flp->flags & FLOWI_FLAG_MULTIPATHOLDROUTE) != 0 && - last_used != NULL) { - result = last_used; - goto out; - } - /* 1. make sure all alt. nexthops have the same GC related data * 2. determine the new candidate to be returned */ @@ -90,15 +74,12 @@ static void rr_select_route(const struct flowi *flp, if (!result) result = first; -out: - last_used = result; result->u.dst.__use++; *rp = result; } static struct ip_mp_alg_ops rr_ops = { .mp_alg_select_route = rr_select_route, - .mp_alg_remove = rr_remove, }; static int __init rr_init(void) @@ -113,3 +94,4 @@ static void __exit rr_exit(void) module_init(rr_init); module_exit(rr_exit); +MODULE_LICENSE("GPL"); diff --git a/net/ipv4/multipath_wrandom.c b/net/ipv4/multipath_wrandom.c index c3d2ca1..bd7d75b 100644 --- a/net/ipv4/multipath_wrandom.c +++ b/net/ipv4/multipath_wrandom.c @@ -31,6 +31,7 @@ #include <linux/igmp.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> +#include <linux/module.h> #include <linux/mroute.h> #include <linux/init.h> #include <net/ip.h> @@ -342,3 +343,4 @@ static void __exit wrandom_exit(void) module_init(wrandom_init); module_exit(wrandom_exit); +MODULE_LICENSE("GPL"); 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 28d9425..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); } } @@ -940,37 +936,16 @@ void ip_ct_refresh_acct(struct ip_conntrack *ct, struct sk_buff * ip_ct_gather_frags(struct sk_buff *skb, u_int32_t user) { - struct sock *sk = skb->sk; -#ifdef CONFIG_NETFILTER_DEBUG - unsigned int olddebug = skb->nf_debug; -#endif - - if (sk) { - sock_hold(sk); - skb_orphan(skb); - } + skb_orphan(skb); local_bh_disable(); skb = ip_defrag(skb, user); local_bh_enable(); - if (!skb) { - if (sk) - sock_put(sk); - return skb; - } - - if (sk) { - skb_set_owner_w(skb, sk); - sock_put(sk); + if (skb) { + ip_send_check(skb->nh.iph); + skb->nfcache |= NFC_ALTERED; } - - 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; } @@ -1009,7 +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); @@ -1021,7 +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; } @@ -1213,14 +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 46ca45f..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; @@ -256,6 +256,7 @@ static void *exp_seq_next(struct seq_file *s, void *v, loff_t *pos) { struct list_head *e = v; + ++*pos; e = e->next; if (e == &ip_conntrack_expect_list) @@ -266,7 +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) @@ -920,22 +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_queue.c b/net/ipv4/netfilter/ip_queue.c index e5746b6..eda1fba 100644 --- a/net/ipv4/netfilter/ip_queue.c +++ b/net/ipv4/netfilter/ip_queue.c @@ -3,6 +3,7 @@ * communicating with userspace via netlink. * * (C) 2000-2002 James Morris <jmorris@intercode.com.au> + * (C) 2003-2005 Netfilter Core Team <coreteam@netfilter.org> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -17,6 +18,7 @@ * 2005-01-10: Added /proc counter for dropped packets; fixed so * packets aren't delivered to user space if they're going * to be dropped. + * 2005-05-26: local_bh_{disable,enable} around nf_reinject (Harald Welte) * */ #include <linux/module.h> @@ -71,7 +73,15 @@ static DECLARE_MUTEX(ipqnl_sem); static void ipq_issue_verdict(struct ipq_queue_entry *entry, int verdict) { + /* TCP input path (and probably other bits) assume to be called + * from softirq context, not from syscall, like ipq_issue_verdict is + * called. TCP input path deadlocks with locks taken from timer + * softirq, e.g. We therefore emulate this by local_bh_disable() */ + + local_bh_disable(); nf_reinject(entry->skb, entry->info, verdict); + local_bh_enable(); + kfree(entry); } 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..9cde8c6 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; } @@ -338,7 +339,7 @@ target(struct sk_buff **pskb, * error messages (RELATED) and information requests (see below) */ if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP && (ctinfo == IP_CT_RELATED - || ctinfo == IP_CT_IS_REPLY+IP_CT_IS_REPLY)) + || ctinfo == IP_CT_RELATED+IP_CT_IS_REPLY)) return IPT_CONTINUE; /* ip_conntrack_icmp guarantees us that we only have ICMP_ECHO, @@ -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/netfilter/ipt_recent.c b/net/ipv4/netfilter/ipt_recent.c index 25ab9fa..2d44b07 100644 --- a/net/ipv4/netfilter/ipt_recent.c +++ b/net/ipv4/netfilter/ipt_recent.c @@ -223,7 +223,7 @@ static int ip_recent_ctrl(struct file *file, const char __user *input, unsigned curr_table->table[count].last_seen = 0; curr_table->table[count].addr = 0; curr_table->table[count].ttl = 0; - memset(curr_table->table[count].last_pkts,0,ip_pkt_list_tot*sizeof(u_int32_t)); + memset(curr_table->table[count].last_pkts,0,ip_pkt_list_tot*sizeof(unsigned long)); curr_table->table[count].oldest_pkt = 0; curr_table->table[count].time_pos = 0; curr_table->time_info[count].position = count; @@ -502,7 +502,7 @@ match(const struct sk_buff *skb, location = time_info[curr_table->time_pos].position; hash_table[r_list[location].hash_entry] = -1; hash_table[hash_result] = location; - memset(r_list[location].last_pkts,0,ip_pkt_list_tot*sizeof(u_int32_t)); + memset(r_list[location].last_pkts,0,ip_pkt_list_tot*sizeof(unsigned long)); r_list[location].time_pos = curr_table->time_pos; r_list[location].addr = addr; r_list[location].ttl = ttl; @@ -631,7 +631,7 @@ match(const struct sk_buff *skb, r_list[location].last_seen = 0; r_list[location].addr = 0; r_list[location].ttl = 0; - memset(r_list[location].last_pkts,0,ip_pkt_list_tot*sizeof(u_int32_t)); + memset(r_list[location].last_pkts,0,ip_pkt_list_tot*sizeof(unsigned long)); r_list[location].oldest_pkt = 0; ans = !info->invert; } @@ -734,10 +734,10 @@ checkentry(const char *tablename, memset(curr_table->table,0,sizeof(struct recent_ip_list)*ip_list_tot); #ifdef DEBUG if(debug) printk(KERN_INFO RECENT_NAME ": checkentry: Allocating %d for pkt_list.\n", - sizeof(u_int32_t)*ip_pkt_list_tot*ip_list_tot); + sizeof(unsigned long)*ip_pkt_list_tot*ip_list_tot); #endif - hold = vmalloc(sizeof(u_int32_t)*ip_pkt_list_tot*ip_list_tot); + hold = vmalloc(sizeof(unsigned long)*ip_pkt_list_tot*ip_list_tot); #ifdef DEBUG if(debug) printk(KERN_INFO RECENT_NAME ": checkentry: After pkt_list allocation.\n"); #endif diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 5b1ec58..d1835b1 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -259,7 +259,7 @@ int raw_rcv(struct sock *sk, struct sk_buff *skb) return 0; } -static int raw_send_hdrinc(struct sock *sk, void *from, int length, +static int raw_send_hdrinc(struct sock *sk, void *from, size_t length, struct rtable *rt, unsigned int flags) { @@ -298,7 +298,7 @@ static int raw_send_hdrinc(struct sock *sk, void *from, int length, goto error_fault; /* We don't modify invalid header */ - if (length >= sizeof(*iph) && iph->ihl * 4 <= length) { + if (length >= sizeof(*iph) && iph->ihl * 4U <= length) { if (!iph->saddr) iph->saddr = rt->rt_src; iph->check = 0; @@ -332,7 +332,7 @@ static void raw_probe_proto_opt(struct flowi *fl, struct msghdr *msg) u8 __user *type = NULL; u8 __user *code = NULL; int probed = 0; - int i; + unsigned int i; if (!msg->msg_iov) return; @@ -384,7 +384,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, int err; err = -EMSGSIZE; - if (len < 0 || len > 0xFFFF) + if (len > 0xFFFF) goto out; /* @@ -514,7 +514,10 @@ done: kfree(ipc.opt); ip_rt_put(rt); -out: return err < 0 ? err : len; +out: + if (err < 0) + return err; + return len; do_confirm: dst_confirm(&rt->u.dst); @@ -610,7 +613,10 @@ static int raw_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, copied = skb->len; done: skb_free_datagram(sk, skb); -out: return err ? err : copied; +out: + if (err) + return err; + return copied; } static int raw_init(struct sock *sk) @@ -691,11 +697,11 @@ static int raw_ioctl(struct sock *sk, int cmd, unsigned long arg) struct sk_buff *skb; int amount = 0; - spin_lock_irq(&sk->sk_receive_queue.lock); + spin_lock_bh(&sk->sk_receive_queue.lock); skb = skb_peek(&sk->sk_receive_queue); if (skb != NULL) amount = skb->len; - spin_unlock_irq(&sk->sk_receive_queue.lock); + spin_unlock_bh(&sk->sk_receive_queue.lock); return put_user(amount, (int __user *)arg); } diff --git a/net/ipv4/route.c b/net/ipv4/route.c index a682d28..80cf633 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1767,7 +1767,7 @@ static inline int ip_mkroute_input_def(struct sk_buff *skb, struct in_device *in_dev, u32 daddr, u32 saddr, u32 tos) { - struct rtable* rth; + struct rtable* rth = NULL; int err; unsigned hash; @@ -1794,7 +1794,7 @@ static inline int ip_mkroute_input(struct sk_buff *skb, u32 daddr, u32 saddr, u32 tos) { #ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED - struct rtable* rth; + struct rtable* rth = NULL; unsigned char hop, hopcount, lasthop; int err = -EINVAL; unsigned int hash; @@ -2239,7 +2239,7 @@ static inline int ip_mkroute_output_def(struct rtable **rp, struct net_device *dev_out, unsigned flags) { - struct rtable *rth; + struct rtable *rth = NULL; int err = __mkroute_output(&rth, res, fl, oldflp, dev_out, flags); unsigned hash; if (err == 0) { @@ -2267,7 +2267,7 @@ static inline int ip_mkroute_output(struct rtable** rp, unsigned char hop; unsigned hash; int err = -EINVAL; - struct rtable *rth; + struct rtable *rth = NULL; if (res->fi && res->fi->fib_nhs > 1) { unsigned char hopcount = res->fi->fib_nhs; @@ -2581,7 +2581,7 @@ int ip_route_output_key(struct rtable **rp, struct flowi *flp) } static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, - int nowait) + int nowait, unsigned int flags) { struct rtable *rt = (struct rtable*)skb->dst; struct rtmsg *r; @@ -2591,9 +2591,8 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, #ifdef CONFIG_IP_MROUTE struct rtattr *eptr; #endif - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*r)); + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*r), flags); r = NLMSG_DATA(nlh); - nlh->nlmsg_flags = (nowait && pid) ? NLM_F_MULTI : 0; r->rtm_family = AF_INET; r->rtm_dst_len = 32; r->rtm_src_len = 0; @@ -2744,7 +2743,7 @@ int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg) NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid; err = rt_fill_info(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, - RTM_NEWROUTE, 0); + RTM_NEWROUTE, 0, 0); if (!err) goto out_free; if (err < 0) { @@ -2781,8 +2780,8 @@ int ip_rt_dump(struct sk_buff *skb, struct netlink_callback *cb) continue; skb->dst = dst_clone(&rt->u.dst); if (rt_fill_info(skb, NETLINK_CB(cb->skb).pid, - cb->nlh->nlmsg_seq, - RTM_NEWROUTE, 1) <= 0) { + cb->nlh->nlmsg_seq, RTM_NEWROUTE, + 1, NLM_F_MULTI) <= 0) { dst_release(xchg(&skb->dst, NULL)); rcu_read_unlock_bh(); goto done; diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index e923d2f..72d0144 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -169,10 +169,10 @@ static inline int cookie_check(struct sk_buff *skb, __u32 cookie) return mssind < NUM_MSS ? msstab[mssind] + 1 : 0; } -extern struct or_calltable or_ipv4; +extern struct request_sock_ops tcp_request_sock_ops; static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst) { struct tcp_sock *tp = tcp_sk(sk); @@ -182,7 +182,7 @@ static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, if (child) tcp_acceptq_queue(sk, req, child); else - tcp_openreq_free(req); + reqsk_free(req); return child; } @@ -190,10 +190,12 @@ static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct ip_options *opt) { + struct inet_request_sock *ireq; + struct tcp_request_sock *treq; struct tcp_sock *tp = tcp_sk(sk); __u32 cookie = ntohl(skb->h.th->ack_seq) - 1; struct sock *ret = sk; - struct open_request *req; + struct request_sock *req; int mss; struct rtable *rt; __u8 rcv_wscale; @@ -209,19 +211,20 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV); - req = tcp_openreq_alloc(); ret = NULL; + req = reqsk_alloc(&tcp_request_sock_ops); /* for safety */ if (!req) goto out; - req->rcv_isn = htonl(skb->h.th->seq) - 1; - req->snt_isn = cookie; + ireq = inet_rsk(req); + treq = tcp_rsk(req); + treq->rcv_isn = htonl(skb->h.th->seq) - 1; + treq->snt_isn = cookie; req->mss = mss; - req->rmt_port = skb->h.th->source; - req->af.v4_req.loc_addr = skb->nh.iph->daddr; - req->af.v4_req.rmt_addr = skb->nh.iph->saddr; - req->class = &or_ipv4; /* for savety */ - req->af.v4_req.opt = NULL; + ireq->rmt_port = skb->h.th->source; + ireq->loc_addr = skb->nh.iph->daddr; + ireq->rmt_addr = skb->nh.iph->saddr; + ireq->opt = NULL; /* We throwed the options of the initial SYN away, so we hope * the ACK carries the same options again (see RFC1122 4.2.3.8) @@ -229,17 +232,15 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, if (opt && opt->optlen) { int opt_size = sizeof(struct ip_options) + opt->optlen; - req->af.v4_req.opt = kmalloc(opt_size, GFP_ATOMIC); - if (req->af.v4_req.opt) { - if (ip_options_echo(req->af.v4_req.opt, skb)) { - kfree(req->af.v4_req.opt); - req->af.v4_req.opt = NULL; - } + ireq->opt = kmalloc(opt_size, GFP_ATOMIC); + if (ireq->opt != NULL && ip_options_echo(ireq->opt, skb)) { + kfree(ireq->opt); + ireq->opt = NULL; } } - req->snd_wscale = req->rcv_wscale = req->tstamp_ok = 0; - req->wscale_ok = req->sack_ok = 0; + ireq->snd_wscale = ireq->rcv_wscale = ireq->tstamp_ok = 0; + ireq->wscale_ok = ireq->sack_ok = 0; req->expires = 0UL; req->retrans = 0; @@ -253,15 +254,15 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct flowi fl = { .nl_u = { .ip4_u = { .daddr = ((opt && opt->srr) ? opt->faddr : - req->af.v4_req.rmt_addr), - .saddr = req->af.v4_req.loc_addr, + ireq->rmt_addr), + .saddr = ireq->loc_addr, .tos = RT_CONN_FLAGS(sk) } }, .proto = IPPROTO_TCP, .uli_u = { .ports = { .sport = skb->h.th->dest, .dport = skb->h.th->source } } }; if (ip_route_output_key(&rt, &fl)) { - tcp_openreq_free(req); + reqsk_free(req); goto out; } } @@ -272,7 +273,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, &req->rcv_wnd, &req->window_clamp, 0, &rcv_wscale); /* BTW win scale with syncookies is 0 by definition */ - req->rcv_wscale = rcv_wscale; + ireq->rcv_wscale = rcv_wscale; ret = get_cookie_sock(sk, skb, req, &rt->u.dst); out: return ret; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 3aafb29..e328945 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -23,6 +23,7 @@ extern int sysctl_ip_nonlocal_bind; extern int sysctl_icmp_echo_ignore_all; extern int sysctl_icmp_echo_ignore_broadcasts; extern int sysctl_icmp_ignore_bogus_error_responses; +extern int sysctl_icmp_errors_use_inbound_ifaddr; /* From ip_fragment.c */ extern int sysctl_ipfrag_low_thresh; @@ -117,6 +118,45 @@ static int ipv4_sysctl_forward_strategy(ctl_table *table, return 1; } +static int proc_tcp_congestion_control(ctl_table *ctl, int write, struct file * filp, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + char val[TCP_CA_NAME_MAX]; + ctl_table tbl = { + .data = val, + .maxlen = TCP_CA_NAME_MAX, + }; + int ret; + + tcp_get_default_congestion_control(val); + + ret = proc_dostring(&tbl, write, filp, buffer, lenp, ppos); + if (write && ret == 0) + ret = tcp_set_default_congestion_control(val); + return ret; +} + +int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, int nlen, + void __user *oldval, size_t __user *oldlenp, + void __user *newval, size_t newlen, + void **context) +{ + char val[TCP_CA_NAME_MAX]; + ctl_table tbl = { + .data = val, + .maxlen = TCP_CA_NAME_MAX, + }; + int ret; + + tcp_get_default_congestion_control(val); + ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen, + context); + if (ret == 0 && newval && newlen) + ret = tcp_set_default_congestion_control(val); + return ret; +} + + ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_TCP_TIMESTAMPS, @@ -396,6 +436,14 @@ ctl_table ipv4_table[] = { .proc_handler = &proc_dointvec }, { + .ctl_name = NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR, + .procname = "icmp_errors_use_inbound_ifaddr", + .data = &sysctl_icmp_errors_use_inbound_ifaddr, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { .ctl_name = NET_IPV4_ROUTE, .procname = "route", .maxlen = 0, @@ -603,70 +651,6 @@ ctl_table ipv4_table[] = { .proc_handler = &proc_dointvec, }, { - .ctl_name = NET_TCP_WESTWOOD, - .procname = "tcp_westwood", - .data = &sysctl_tcp_westwood, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_VEGAS, - .procname = "tcp_vegas_cong_avoid", - .data = &sysctl_tcp_vegas_cong_avoid, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_VEGAS_ALPHA, - .procname = "tcp_vegas_alpha", - .data = &sysctl_tcp_vegas_alpha, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_VEGAS_BETA, - .procname = "tcp_vegas_beta", - .data = &sysctl_tcp_vegas_beta, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_VEGAS_GAMMA, - .procname = "tcp_vegas_gamma", - .data = &sysctl_tcp_vegas_gamma, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_BIC, - .procname = "tcp_bic", - .data = &sysctl_tcp_bic, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_BIC_FAST_CONVERGENCE, - .procname = "tcp_bic_fast_convergence", - .data = &sysctl_tcp_bic_fast_convergence, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = NET_TCP_BIC_LOW_WINDOW, - .procname = "tcp_bic_low_window", - .data = &sysctl_tcp_bic_low_window, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { .ctl_name = NET_TCP_MODERATE_RCVBUF, .procname = "tcp_moderate_rcvbuf", .data = &sysctl_tcp_moderate_rcvbuf, @@ -683,13 +667,14 @@ ctl_table ipv4_table[] = { .proc_handler = &proc_dointvec, }, { - .ctl_name = NET_TCP_BIC_BETA, - .procname = "tcp_bic_beta", - .data = &sysctl_tcp_bic_beta, - .maxlen = sizeof(int), + .ctl_name = NET_TCP_CONG_CONTROL, + .procname = "tcp_congestion_control", .mode = 0644, - .proc_handler = &proc_dointvec, + .maxlen = TCP_CA_NAME_MAX, + .proc_handler = &proc_tcp_congestion_control, + .strategy = &sysctl_tcp_congestion_control, }, + { .ctl_name = 0 } }; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index a037baf..882436d 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -271,7 +271,6 @@ int sysctl_tcp_fin_timeout = TCP_FIN_TIMEOUT; DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics); -kmem_cache_t *tcp_openreq_cachep; kmem_cache_t *tcp_bucket_cachep; kmem_cache_t *tcp_timewait_cachep; @@ -317,7 +316,7 @@ EXPORT_SYMBOL(tcp_enter_memory_pressure); static __inline__ unsigned int tcp_listen_poll(struct sock *sk, poll_table *wait) { - return tcp_sk(sk)->accept_queue ? (POLLIN | POLLRDNORM) : 0; + return !reqsk_queue_empty(&tcp_sk(sk)->accept_queue) ? (POLLIN | POLLRDNORM) : 0; } /* @@ -463,28 +462,15 @@ int tcp_listen_start(struct sock *sk) { struct inet_sock *inet = inet_sk(sk); struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt; + int rc = reqsk_queue_alloc(&tp->accept_queue, TCP_SYNQ_HSIZE); + + if (rc != 0) + return rc; sk->sk_max_ack_backlog = 0; sk->sk_ack_backlog = 0; - tp->accept_queue = tp->accept_queue_tail = NULL; - rwlock_init(&tp->syn_wait_lock); tcp_delack_init(tp); - lopt = kmalloc(sizeof(struct tcp_listen_opt), GFP_KERNEL); - if (!lopt) - return -ENOMEM; - - memset(lopt, 0, sizeof(struct tcp_listen_opt)); - for (lopt->max_qlen_log = 6; ; lopt->max_qlen_log++) - if ((1 << lopt->max_qlen_log) >= sysctl_max_syn_backlog) - break; - get_random_bytes(&lopt->hash_rnd, 4); - - write_lock_bh(&tp->syn_wait_lock); - tp->listen_opt = lopt; - write_unlock_bh(&tp->syn_wait_lock); - /* There is race window here: we announce ourselves listening, * but this transition is still not validated by get_port(). * It is OK, because this socket enters to hash table only @@ -501,10 +487,7 @@ int tcp_listen_start(struct sock *sk) } sk->sk_state = TCP_CLOSE; - write_lock_bh(&tp->syn_wait_lock); - tp->listen_opt = NULL; - write_unlock_bh(&tp->syn_wait_lock); - kfree(lopt); + reqsk_queue_destroy(&tp->accept_queue); return -EADDRINUSE; } @@ -516,25 +499,23 @@ int tcp_listen_start(struct sock *sk) static void tcp_listen_stop (struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; - struct open_request *acc_req = tp->accept_queue; - struct open_request *req; + struct listen_sock *lopt; + struct request_sock *acc_req; + struct request_sock *req; int i; tcp_delete_keepalive_timer(sk); /* make all the listen_opt local to us */ - write_lock_bh(&tp->syn_wait_lock); - tp->listen_opt = NULL; - write_unlock_bh(&tp->syn_wait_lock); - tp->accept_queue = tp->accept_queue_tail = NULL; + lopt = reqsk_queue_yank_listen_sk(&tp->accept_queue); + acc_req = reqsk_queue_yank_acceptq(&tp->accept_queue); if (lopt->qlen) { for (i = 0; i < TCP_SYNQ_HSIZE; i++) { while ((req = lopt->syn_table[i]) != NULL) { lopt->syn_table[i] = req->dl_next; lopt->qlen--; - tcp_openreq_free(req); + reqsk_free(req); /* Following specs, it would be better either to send FIN * (and enter FIN-WAIT-1, it is normal close) @@ -574,7 +555,7 @@ static void tcp_listen_stop (struct sock *sk) sock_put(child); sk_acceptq_removed(sk); - tcp_openreq_fastfree(req); + __reqsk_free(req); } BUG_TRAP(!sk->sk_ack_backlog); } @@ -1345,7 +1326,7 @@ int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, cleanup_rbuf(sk, copied); - if (tp->ucopy.task == user_recv) { + if (!sysctl_tcp_low_latency && tp->ucopy.task == user_recv) { /* Install new reader */ if (!user_recv && !(flags & (MSG_TRUNC | MSG_PEEK))) { user_recv = current; @@ -1868,11 +1849,11 @@ static int wait_for_connect(struct sock *sk, long timeo) prepare_to_wait_exclusive(sk->sk_sleep, &wait, TASK_INTERRUPTIBLE); release_sock(sk); - if (!tp->accept_queue) + if (reqsk_queue_empty(&tp->accept_queue)) timeo = schedule_timeout(timeo); lock_sock(sk); err = 0; - if (tp->accept_queue) + if (!reqsk_queue_empty(&tp->accept_queue)) break; err = -EINVAL; if (sk->sk_state != TCP_LISTEN) @@ -1895,7 +1876,6 @@ static int wait_for_connect(struct sock *sk, long timeo) struct sock *tcp_accept(struct sock *sk, int flags, int *err) { struct tcp_sock *tp = tcp_sk(sk); - struct open_request *req; struct sock *newsk; int error; @@ -1906,37 +1886,31 @@ struct sock *tcp_accept(struct sock *sk, int flags, int *err) */ error = -EINVAL; if (sk->sk_state != TCP_LISTEN) - goto out; + goto out_err; /* Find already established connection */ - if (!tp->accept_queue) { + if (reqsk_queue_empty(&tp->accept_queue)) { long timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); /* If this is a non blocking socket don't sleep */ error = -EAGAIN; if (!timeo) - goto out; + goto out_err; error = wait_for_connect(sk, timeo); if (error) - goto out; + goto out_err; } - req = tp->accept_queue; - if ((tp->accept_queue = req->dl_next) == NULL) - tp->accept_queue_tail = NULL; - - newsk = req->sk; - sk_acceptq_removed(sk); - tcp_openreq_fastfree(req); + newsk = reqsk_queue_get_child(&tp->accept_queue, sk); BUG_TRAP(newsk->sk_state != TCP_SYN_RECV); - release_sock(sk); - return newsk; - out: release_sock(sk); + return newsk; +out_err: + newsk = NULL; *err = error; - return NULL; + goto out; } /* @@ -1953,6 +1927,25 @@ int tcp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, return tp->af_specific->setsockopt(sk, level, optname, optval, optlen); + /* This is a string value all the others are int's */ + if (optname == TCP_CONGESTION) { + char name[TCP_CA_NAME_MAX]; + + if (optlen < 1) + return -EINVAL; + + val = strncpy_from_user(name, optval, + min(TCP_CA_NAME_MAX-1, optlen)); + if (val < 0) + return -EFAULT; + name[val] = 0; + + lock_sock(sk); + err = tcp_set_congestion_control(tp, name); + release_sock(sk); + return err; + } + if (optlen < sizeof(int)) return -EINVAL; @@ -2237,6 +2230,16 @@ int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, case TCP_QUICKACK: val = !tp->ack.pingpong; break; + + case TCP_CONGESTION: + if (get_user(len, optlen)) + return -EFAULT; + len = min_t(unsigned int, len, TCP_CA_NAME_MAX); + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, tp->ca_ops->name, len)) + return -EFAULT; + return 0; default: return -ENOPROTOOPT; }; @@ -2250,7 +2253,7 @@ int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, extern void __skb_cb_too_small_for_tcp(int, int); -extern void tcpdiag_init(void); +extern struct tcp_congestion_ops tcp_reno; static __initdata unsigned long thash_entries; static int __init set_thash_entries(char *str) @@ -2271,13 +2274,6 @@ void __init tcp_init(void) __skb_cb_too_small_for_tcp(sizeof(struct tcp_skb_cb), sizeof(skb->cb)); - tcp_openreq_cachep = kmem_cache_create("tcp_open_request", - sizeof(struct open_request), - 0, SLAB_HWCACHE_ALIGN, - NULL, NULL); - if (!tcp_openreq_cachep) - panic("tcp_init: Cannot alloc open_request cache."); - tcp_bucket_cachep = kmem_cache_create("tcp_bind_bucket", sizeof(struct tcp_bind_bucket), 0, SLAB_HWCACHE_ALIGN, @@ -2338,7 +2334,7 @@ void __init tcp_init(void) (tcp_bhash_size * sizeof(struct tcp_bind_hashbucket)); order++) ; - if (order > 4) { + if (order >= 4) { sysctl_local_port_range[0] = 32768; sysctl_local_port_range[1] = 61000; sysctl_tcp_max_tw_buckets = 180000; @@ -2366,6 +2362,8 @@ void __init tcp_init(void) printk(KERN_INFO "TCP: Hash tables configured " "(established %d bind %d)\n", tcp_ehash_size << 1, tcp_bhash_size); + + tcp_register_congestion_control(&tcp_reno); } EXPORT_SYMBOL(tcp_accept); @@ -2374,7 +2372,6 @@ EXPORT_SYMBOL(tcp_destroy_sock); EXPORT_SYMBOL(tcp_disconnect); EXPORT_SYMBOL(tcp_getsockopt); EXPORT_SYMBOL(tcp_ioctl); -EXPORT_SYMBOL(tcp_openreq_cachep); EXPORT_SYMBOL(tcp_poll); EXPORT_SYMBOL(tcp_read_sock); EXPORT_SYMBOL(tcp_recvmsg); diff --git a/net/ipv4/tcp_bic.c b/net/ipv4/tcp_bic.c new file mode 100644 index 0000000..ec38d45 --- /dev/null +++ b/net/ipv4/tcp_bic.c @@ -0,0 +1,331 @@ +/* + * Binary Increase Congestion control for TCP + * + * This is from the implementation of BICTCP in + * Lison-Xu, Kahaled Harfoush, and Injong Rhee. + * "Binary Increase Congestion Control for Fast, Long Distance + * Networks" in InfoComm 2004 + * Available from: + * http://www.csc.ncsu.edu/faculty/rhee/export/bitcp.pdf + * + * Unless BIC is enabled and congestion window is large + * this behaves the same as the original Reno. + */ + +#include <linux/config.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <net/tcp.h> + + +#define BICTCP_BETA_SCALE 1024 /* Scale factor beta calculation + * max_cwnd = snd_cwnd * beta + */ +#define BICTCP_B 4 /* + * In binary search, + * go to point (max+min)/N + */ + +static int fast_convergence = 1; +static int max_increment = 32; +static int low_window = 14; +static int beta = 819; /* = 819/1024 (BICTCP_BETA_SCALE) */ +static int low_utilization_threshold = 153; +static int low_utilization_period = 2; +static int initial_ssthresh = 100; +static int smooth_part = 20; + +module_param(fast_convergence, int, 0644); +MODULE_PARM_DESC(fast_convergence, "turn on/off fast convergence"); +module_param(max_increment, int, 0644); +MODULE_PARM_DESC(max_increment, "Limit on increment allowed during binary search"); +module_param(low_window, int, 0644); +MODULE_PARM_DESC(low_window, "lower bound on congestion window (for TCP friendliness)"); +module_param(beta, int, 0644); +MODULE_PARM_DESC(beta, "beta for multiplicative increase"); +module_param(low_utilization_threshold, int, 0644); +MODULE_PARM_DESC(low_utilization_threshold, "percent (scaled by 1024) for low utilization mode"); +module_param(low_utilization_period, int, 0644); +MODULE_PARM_DESC(low_utilization_period, "if average delay exceeds then goto to low utilization mode (seconds)"); +module_param(initial_ssthresh, int, 0644); +MODULE_PARM_DESC(initial_ssthresh, "initial value of slow start threshold"); +module_param(smooth_part, int, 0644); +MODULE_PARM_DESC(smooth_part, "log(B/(B*Smin))/log(B/(B-1))+B, # of RTT from Wmax-B to Wmax"); + + +/* BIC TCP Parameters */ +struct bictcp { + u32 cnt; /* increase cwnd by 1 after ACKs */ + u32 last_max_cwnd; /* last maximum snd_cwnd */ + u32 loss_cwnd; /* congestion window at last loss */ + u32 last_cwnd; /* the last snd_cwnd */ + u32 last_time; /* time when updated last_cwnd */ + u32 delay_min; /* min delay */ + u32 delay_max; /* max delay */ + u32 last_delay; + u8 low_utilization;/* 0: high; 1: low */ + u32 low_utilization_start; /* starting time of low utilization detection*/ + u32 epoch_start; /* beginning of an epoch */ +#define ACK_RATIO_SHIFT 4 + u32 delayed_ack; /* estimate the ratio of Packets/ACKs << 4 */ +}; + +static inline void bictcp_reset(struct bictcp *ca) +{ + ca->cnt = 0; + ca->last_max_cwnd = 0; + ca->loss_cwnd = 0; + ca->last_cwnd = 0; + ca->last_time = 0; + ca->delay_min = 0; + ca->delay_max = 0; + ca->last_delay = 0; + ca->low_utilization = 0; + ca->low_utilization_start = 0; + ca->epoch_start = 0; + ca->delayed_ack = 2 << ACK_RATIO_SHIFT; +} + +static void bictcp_init(struct tcp_sock *tp) +{ + bictcp_reset(tcp_ca(tp)); + if (initial_ssthresh) + tp->snd_ssthresh = initial_ssthresh; +} + +/* + * Compute congestion window to use. + */ +static inline void bictcp_update(struct bictcp *ca, u32 cwnd) +{ + if (ca->last_cwnd == cwnd && + (s32)(tcp_time_stamp - ca->last_time) <= HZ / 32) + return; + + ca->last_cwnd = cwnd; + ca->last_time = tcp_time_stamp; + + if (ca->epoch_start == 0) /* record the beginning of an epoch */ + ca->epoch_start = tcp_time_stamp; + + /* start off normal */ + if (cwnd <= low_window) { + ca->cnt = cwnd; + return; + } + + /* binary increase */ + if (cwnd < ca->last_max_cwnd) { + __u32 dist = (ca->last_max_cwnd - cwnd) + / BICTCP_B; + + if (dist > max_increment) + /* linear increase */ + ca->cnt = cwnd / max_increment; + else if (dist <= 1U) + /* binary search increase */ + ca->cnt = (cwnd * smooth_part) / BICTCP_B; + else + /* binary search increase */ + ca->cnt = cwnd / dist; + } else { + /* slow start AMD linear increase */ + if (cwnd < ca->last_max_cwnd + BICTCP_B) + /* slow start */ + ca->cnt = (cwnd * smooth_part) / BICTCP_B; + else if (cwnd < ca->last_max_cwnd + max_increment*(BICTCP_B-1)) + /* slow start */ + ca->cnt = (cwnd * (BICTCP_B-1)) + / cwnd-ca->last_max_cwnd; + else + /* linear increase */ + ca->cnt = cwnd / max_increment; + } + + /* if in slow start or link utilization is very low */ + if ( ca->loss_cwnd == 0 || + (cwnd > ca->loss_cwnd && ca->low_utilization)) { + if (ca->cnt > 20) /* increase cwnd 5% per RTT */ + ca->cnt = 20; + } + + ca->cnt = (ca->cnt << ACK_RATIO_SHIFT) / ca->delayed_ack; + if (ca->cnt == 0) /* cannot be zero */ + ca->cnt = 1; +} + + +/* Detect low utilization in congestion avoidance */ +static inline void bictcp_low_utilization(struct tcp_sock *tp, int flag) +{ + struct bictcp *ca = tcp_ca(tp); + u32 dist, delay; + + /* No time stamp */ + if (!(tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr) || + /* Discard delay samples right after fast recovery */ + tcp_time_stamp < ca->epoch_start + HZ || + /* this delay samples may not be accurate */ + flag == 0) { + ca->last_delay = 0; + goto notlow; + } + + delay = ca->last_delay<<3; /* use the same scale as tp->srtt*/ + ca->last_delay = tcp_time_stamp - tp->rx_opt.rcv_tsecr; + if (delay == 0) /* no previous delay sample */ + goto notlow; + + /* first time call or link delay decreases */ + if (ca->delay_min == 0 || ca->delay_min > delay) { + ca->delay_min = ca->delay_max = delay; + goto notlow; + } + + if (ca->delay_max < delay) + ca->delay_max = delay; + + /* utilization is low, if avg delay < dist*threshold + for checking_period time */ + dist = ca->delay_max - ca->delay_min; + if (dist <= ca->delay_min>>6 || + tp->srtt - ca->delay_min >= (dist*low_utilization_threshold)>>10) + goto notlow; + + if (ca->low_utilization_start == 0) { + ca->low_utilization = 0; + ca->low_utilization_start = tcp_time_stamp; + } else if ((s32)(tcp_time_stamp - ca->low_utilization_start) + > low_utilization_period*HZ) { + ca->low_utilization = 1; + } + + return; + + notlow: + ca->low_utilization = 0; + ca->low_utilization_start = 0; + +} + +static void bictcp_cong_avoid(struct tcp_sock *tp, u32 ack, + u32 seq_rtt, u32 in_flight, int data_acked) +{ + struct bictcp *ca = tcp_ca(tp); + + bictcp_low_utilization(tp, data_acked); + + if (in_flight < tp->snd_cwnd) + return; + + if (tp->snd_cwnd <= tp->snd_ssthresh) { + /* In "safe" area, increase. */ + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + } else { + bictcp_update(ca, tp->snd_cwnd); + + /* In dangerous area, increase slowly. + * In theory this is tp->snd_cwnd += 1 / tp->snd_cwnd + */ + if (tp->snd_cwnd_cnt >= ca->cnt) { + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + tp->snd_cwnd_cnt = 0; + } else + tp->snd_cwnd_cnt++; + } + +} + +/* + * behave like Reno until low_window is reached, + * then increase congestion window slowly + */ +static u32 bictcp_recalc_ssthresh(struct tcp_sock *tp) +{ + struct bictcp *ca = tcp_ca(tp); + + ca->epoch_start = 0; /* end of epoch */ + + /* in case of wrong delay_max*/ + if (ca->delay_min > 0 && ca->delay_max > ca->delay_min) + ca->delay_max = ca->delay_min + + ((ca->delay_max - ca->delay_min)* 90) / 100; + + /* Wmax and fast convergence */ + if (tp->snd_cwnd < ca->last_max_cwnd && fast_convergence) + ca->last_max_cwnd = (tp->snd_cwnd * (BICTCP_BETA_SCALE + beta)) + / (2 * BICTCP_BETA_SCALE); + else + ca->last_max_cwnd = tp->snd_cwnd; + + ca->loss_cwnd = tp->snd_cwnd; + + + if (tp->snd_cwnd <= low_window) + return max(tp->snd_cwnd >> 1U, 2U); + else + return max((tp->snd_cwnd * beta) / BICTCP_BETA_SCALE, 2U); +} + +static u32 bictcp_undo_cwnd(struct tcp_sock *tp) +{ + struct bictcp *ca = tcp_ca(tp); + + return max(tp->snd_cwnd, ca->last_max_cwnd); +} + +static u32 bictcp_min_cwnd(struct tcp_sock *tp) +{ + return tp->snd_ssthresh; +} + +static void bictcp_state(struct tcp_sock *tp, u8 new_state) +{ + if (new_state == TCP_CA_Loss) + bictcp_reset(tcp_ca(tp)); +} + +/* Track delayed acknowledgement ratio using sliding window + * ratio = (15*ratio + sample) / 16 + */ +static void bictcp_acked(struct tcp_sock *tp, u32 cnt) +{ + if (cnt > 0 && tp->ca_state == TCP_CA_Open) { + struct bictcp *ca = tcp_ca(tp); + cnt -= ca->delayed_ack >> ACK_RATIO_SHIFT; + ca->delayed_ack += cnt; + } +} + + +static struct tcp_congestion_ops bictcp = { + .init = bictcp_init, + .ssthresh = bictcp_recalc_ssthresh, + .cong_avoid = bictcp_cong_avoid, + .set_state = bictcp_state, + .undo_cwnd = bictcp_undo_cwnd, + .min_cwnd = bictcp_min_cwnd, + .pkts_acked = bictcp_acked, + .owner = THIS_MODULE, + .name = "bic", +}; + +static int __init bictcp_register(void) +{ + BUG_ON(sizeof(struct bictcp) > TCP_CA_PRIV_SIZE); + return tcp_register_congestion_control(&bictcp); +} + +static void __exit bictcp_unregister(void) +{ + tcp_unregister_congestion_control(&bictcp); +} + +module_init(bictcp_register); +module_exit(bictcp_unregister); + +MODULE_AUTHOR("Stephen Hemminger"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("BIC TCP"); diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c new file mode 100644 index 0000000..4970d10 --- /dev/null +++ b/net/ipv4/tcp_cong.c @@ -0,0 +1,237 @@ +/* + * Plugable TCP congestion control support and newReno + * congestion control. + * Based on ideas from I/O scheduler suport and Web100. + * + * Copyright (C) 2005 Stephen Hemminger <shemminger@osdl.org> + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/types.h> +#include <linux/list.h> +#include <net/tcp.h> + +static DEFINE_SPINLOCK(tcp_cong_list_lock); +static LIST_HEAD(tcp_cong_list); + +/* Simple linear search, don't expect many entries! */ +static struct tcp_congestion_ops *tcp_ca_find(const char *name) +{ + struct tcp_congestion_ops *e; + + list_for_each_entry_rcu(e, &tcp_cong_list, list) { + if (strcmp(e->name, name) == 0) + return e; + } + + return NULL; +} + +/* + * Attach new congestion control algorthim to the list + * of available options. + */ +int tcp_register_congestion_control(struct tcp_congestion_ops *ca) +{ + int ret = 0; + + /* all algorithms must implement ssthresh and cong_avoid ops */ + if (!ca->ssthresh || !ca->cong_avoid || !ca->min_cwnd) { + printk(KERN_ERR "TCP %s does not implement required ops\n", + ca->name); + return -EINVAL; + } + + spin_lock(&tcp_cong_list_lock); + if (tcp_ca_find(ca->name)) { + printk(KERN_NOTICE "TCP %s already registered\n", ca->name); + ret = -EEXIST; + } else { + list_add_rcu(&ca->list, &tcp_cong_list); + printk(KERN_INFO "TCP %s registered\n", ca->name); + } + spin_unlock(&tcp_cong_list_lock); + + return ret; +} +EXPORT_SYMBOL_GPL(tcp_register_congestion_control); + +/* + * Remove congestion control algorithm, called from + * the module's remove function. Module ref counts are used + * to ensure that this can't be done till all sockets using + * that method are closed. + */ +void tcp_unregister_congestion_control(struct tcp_congestion_ops *ca) +{ + spin_lock(&tcp_cong_list_lock); + list_del_rcu(&ca->list); + spin_unlock(&tcp_cong_list_lock); +} +EXPORT_SYMBOL_GPL(tcp_unregister_congestion_control); + +/* Assign choice of congestion control. */ +void tcp_init_congestion_control(struct tcp_sock *tp) +{ + struct tcp_congestion_ops *ca; + + if (tp->ca_ops != &tcp_init_congestion_ops) + return; + + rcu_read_lock(); + list_for_each_entry_rcu(ca, &tcp_cong_list, list) { + if (try_module_get(ca->owner)) { + tp->ca_ops = ca; + break; + } + + } + rcu_read_unlock(); + + if (tp->ca_ops->init) + tp->ca_ops->init(tp); +} + +/* Manage refcounts on socket close. */ +void tcp_cleanup_congestion_control(struct tcp_sock *tp) +{ + if (tp->ca_ops->release) + tp->ca_ops->release(tp); + module_put(tp->ca_ops->owner); +} + +/* Used by sysctl to change default congestion control */ +int tcp_set_default_congestion_control(const char *name) +{ + struct tcp_congestion_ops *ca; + int ret = -ENOENT; + + spin_lock(&tcp_cong_list_lock); + ca = tcp_ca_find(name); +#ifdef CONFIG_KMOD + if (!ca) { + spin_unlock(&tcp_cong_list_lock); + + request_module("tcp_%s", name); + spin_lock(&tcp_cong_list_lock); + ca = tcp_ca_find(name); + } +#endif + + if (ca) { + list_move(&ca->list, &tcp_cong_list); + ret = 0; + } + spin_unlock(&tcp_cong_list_lock); + + return ret; +} + +/* Get current default congestion control */ +void tcp_get_default_congestion_control(char *name) +{ + struct tcp_congestion_ops *ca; + /* We will always have reno... */ + BUG_ON(list_empty(&tcp_cong_list)); + + rcu_read_lock(); + ca = list_entry(tcp_cong_list.next, struct tcp_congestion_ops, list); + strncpy(name, ca->name, TCP_CA_NAME_MAX); + rcu_read_unlock(); +} + +/* Change congestion control for socket */ +int tcp_set_congestion_control(struct tcp_sock *tp, const char *name) +{ + struct tcp_congestion_ops *ca; + int err = 0; + + rcu_read_lock(); + ca = tcp_ca_find(name); + if (ca == tp->ca_ops) + goto out; + + if (!ca) + err = -ENOENT; + + else if (!try_module_get(ca->owner)) + err = -EBUSY; + + else { + tcp_cleanup_congestion_control(tp); + tp->ca_ops = ca; + if (tp->ca_ops->init) + tp->ca_ops->init(tp); + } + out: + rcu_read_unlock(); + return err; +} + +/* + * TCP Reno congestion control + * This is special case used for fallback as well. + */ +/* This is Jacobson's slow start and congestion avoidance. + * SIGCOMM '88, p. 328. + */ +void tcp_reno_cong_avoid(struct tcp_sock *tp, u32 ack, u32 rtt, u32 in_flight, + int flag) +{ + if (in_flight < tp->snd_cwnd) + return; + + if (tp->snd_cwnd <= tp->snd_ssthresh) { + /* In "safe" area, increase. */ + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + } else { + /* In dangerous area, increase slowly. + * In theory this is tp->snd_cwnd += 1 / tp->snd_cwnd + */ + if (tp->snd_cwnd_cnt >= tp->snd_cwnd) { + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + tp->snd_cwnd_cnt = 0; + } else + tp->snd_cwnd_cnt++; + } +} +EXPORT_SYMBOL_GPL(tcp_reno_cong_avoid); + +/* Slow start threshold is half the congestion window (min 2) */ +u32 tcp_reno_ssthresh(struct tcp_sock *tp) +{ + return max(tp->snd_cwnd >> 1U, 2U); +} +EXPORT_SYMBOL_GPL(tcp_reno_ssthresh); + +/* Lower bound on congestion window. */ +u32 tcp_reno_min_cwnd(struct tcp_sock *tp) +{ + return tp->snd_ssthresh/2; +} +EXPORT_SYMBOL_GPL(tcp_reno_min_cwnd); + +struct tcp_congestion_ops tcp_reno = { + .name = "reno", + .owner = THIS_MODULE, + .ssthresh = tcp_reno_ssthresh, + .cong_avoid = tcp_reno_cong_avoid, + .min_cwnd = tcp_reno_min_cwnd, +}; + +/* Initial congestion control used (until SYN) + * really reno under another name so we can tell difference + * during tcp_set_default_congestion_control + */ +struct tcp_congestion_ops tcp_init_congestion_ops = { + .name = "", + .owner = THIS_MODULE, + .ssthresh = tcp_reno_ssthresh, + .cong_avoid = tcp_reno_cong_avoid, + .min_cwnd = tcp_reno_min_cwnd, +}; +EXPORT_SYMBOL_GPL(tcp_init_congestion_ops); diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 8faa894..f66945c 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -42,15 +42,8 @@ struct tcpdiag_entry static struct sock *tcpnl; - #define TCPDIAG_PUT(skb, attrtype, attrlen) \ -({ int rtalen = RTA_LENGTH(attrlen); \ - struct rtattr *rta; \ - if (skb_tailroom(skb) < RTA_ALIGN(rtalen)) goto nlmsg_failure; \ - rta = (void*)__skb_put(skb, RTA_ALIGN(rtalen)); \ - rta->rta_type = attrtype; \ - rta->rta_len = rtalen; \ - RTA_DATA(rta); }) + RTA_DATA(__RTA_PUT(skb, attrtype, attrlen)) static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk, int ext, u32 pid, u32 seq, u16 nlmsg_flags) @@ -61,7 +54,6 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk, struct nlmsghdr *nlh; struct tcp_info *info = NULL; struct tcpdiag_meminfo *minfo = NULL; - struct tcpvegas_info *vinfo = NULL; unsigned char *b = skb->tail; nlh = NLMSG_PUT(skb, pid, seq, TCPDIAG_GETSOCK, sizeof(*r)); @@ -73,9 +65,11 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk, if (ext & (1<<(TCPDIAG_INFO-1))) info = TCPDIAG_PUT(skb, TCPDIAG_INFO, sizeof(*info)); - if ((tcp_is_westwood(tp) || tcp_is_vegas(tp)) - && (ext & (1<<(TCPDIAG_VEGASINFO-1)))) - vinfo = TCPDIAG_PUT(skb, TCPDIAG_VEGASINFO, sizeof(*vinfo)); + if (ext & (1<<(TCPDIAG_CONG-1))) { + size_t len = strlen(tp->ca_ops->name); + strcpy(TCPDIAG_PUT(skb, TCPDIAG_CONG, len+1), + tp->ca_ops->name); + } } r->tcpdiag_family = sk->sk_family; r->tcpdiag_state = sk->sk_state; @@ -166,23 +160,13 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk, if (info) tcp_get_info(sk, info); - if (vinfo) { - if (tcp_is_vegas(tp)) { - vinfo->tcpv_enabled = tp->vegas.doing_vegas_now; - vinfo->tcpv_rttcnt = tp->vegas.cntRTT; - vinfo->tcpv_rtt = jiffies_to_usecs(tp->vegas.baseRTT); - vinfo->tcpv_minrtt = jiffies_to_usecs(tp->vegas.minRTT); - } else { - vinfo->tcpv_enabled = 0; - vinfo->tcpv_rttcnt = 0; - vinfo->tcpv_rtt = jiffies_to_usecs(tp->westwood.rtt); - vinfo->tcpv_minrtt = jiffies_to_usecs(tp->westwood.rtt_min); - } - } + if (sk->sk_state < TCP_TIME_WAIT && tp->ca_ops->get_info) + tp->ca_ops->get_info(tp, ext, skb); nlh->nlmsg_len = skb->tail - b; return skb->len; +rtattr_failure: nlmsg_failure: skb_trim(skb, b - skb->data); return -1; @@ -455,9 +439,10 @@ static int tcpdiag_dump_sock(struct sk_buff *skb, struct sock *sk, } static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, - struct open_request *req, + struct request_sock *req, u32 pid, u32 seq) { + const struct inet_request_sock *ireq = inet_rsk(req); struct inet_sock *inet = inet_sk(sk); unsigned char *b = skb->tail; struct tcpdiagmsg *r; @@ -482,9 +467,9 @@ static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, tmo = 0; r->id.tcpdiag_sport = inet->sport; - r->id.tcpdiag_dport = req->rmt_port; - r->id.tcpdiag_src[0] = req->af.v4_req.loc_addr; - r->id.tcpdiag_dst[0] = req->af.v4_req.rmt_addr; + r->id.tcpdiag_dport = ireq->rmt_port; + r->id.tcpdiag_src[0] = ireq->loc_addr; + r->id.tcpdiag_dst[0] = ireq->rmt_addr; r->tcpdiag_expires = jiffies_to_msecs(tmo), r->tcpdiag_rqueue = 0; r->tcpdiag_wqueue = 0; @@ -493,9 +478,9 @@ static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, #ifdef CONFIG_IP_TCPDIAG_IPV6 if (r->tcpdiag_family == AF_INET6) { ipv6_addr_copy((struct in6_addr *)r->id.tcpdiag_src, - &req->af.v6_req.loc_addr); + &tcp6_rsk(req)->loc_addr); ipv6_addr_copy((struct in6_addr *)r->id.tcpdiag_dst, - &req->af.v6_req.rmt_addr); + &tcp6_rsk(req)->rmt_addr); } #endif nlh->nlmsg_len = skb->tail - b; @@ -513,7 +498,7 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, struct tcpdiag_entry entry; struct tcpdiagreq *r = NLMSG_DATA(cb->nlh); struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt; + struct listen_sock *lopt; struct rtattr *bc = NULL; struct inet_sock *inet = inet_sk(sk); int j, s_j; @@ -528,9 +513,9 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, entry.family = sk->sk_family; - read_lock_bh(&tp->syn_wait_lock); + read_lock_bh(&tp->accept_queue.syn_wait_lock); - lopt = tp->listen_opt; + lopt = tp->accept_queue.listen_opt; if (!lopt || !lopt->qlen) goto out; @@ -541,13 +526,15 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, } for (j = s_j; j < TCP_SYNQ_HSIZE; j++) { - struct open_request *req, *head = lopt->syn_table[j]; + struct request_sock *req, *head = lopt->syn_table[j]; reqnum = 0; for (req = head; req; reqnum++, req = req->dl_next) { + struct inet_request_sock *ireq = inet_rsk(req); + if (reqnum < s_reqnum) continue; - if (r->id.tcpdiag_dport != req->rmt_port && + if (r->id.tcpdiag_dport != ireq->rmt_port && r->id.tcpdiag_dport) continue; @@ -555,16 +542,16 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, entry.saddr = #ifdef CONFIG_IP_TCPDIAG_IPV6 (entry.family == AF_INET6) ? - req->af.v6_req.loc_addr.s6_addr32 : + tcp6_rsk(req)->loc_addr.s6_addr32 : #endif - &req->af.v4_req.loc_addr; + &ireq->loc_addr; entry.daddr = #ifdef CONFIG_IP_TCPDIAG_IPV6 (entry.family == AF_INET6) ? - req->af.v6_req.rmt_addr.s6_addr32 : + tcp6_rsk(req)->rmt_addr.s6_addr32 : #endif - &req->af.v4_req.rmt_addr; - entry.dport = ntohs(req->rmt_port); + &ireq->rmt_addr; + entry.dport = ntohs(ireq->rmt_port); if (!tcpdiag_bc_run(RTA_DATA(bc), RTA_PAYLOAD(bc), &entry)) @@ -585,7 +572,7 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, } out: - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); return err; } diff --git a/net/ipv4/tcp_highspeed.c b/net/ipv4/tcp_highspeed.c new file mode 100644 index 0000000..36c51f8 --- /dev/null +++ b/net/ipv4/tcp_highspeed.c @@ -0,0 +1,181 @@ +/* + * Sally Floyd's High Speed TCP (RFC 3649) congestion control + * + * See http://www.icir.org/floyd/hstcp.html + * + * John Heffner <jheffner@psc.edu> + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <net/tcp.h> + + +/* From AIMD tables from RFC 3649 appendix B, + * with fixed-point MD scaled <<8. + */ +static const struct hstcp_aimd_val { + unsigned int cwnd; + unsigned int md; +} hstcp_aimd_vals[] = { + { 38, 128, /* 0.50 */ }, + { 118, 112, /* 0.44 */ }, + { 221, 104, /* 0.41 */ }, + { 347, 98, /* 0.38 */ }, + { 495, 93, /* 0.37 */ }, + { 663, 89, /* 0.35 */ }, + { 851, 86, /* 0.34 */ }, + { 1058, 83, /* 0.33 */ }, + { 1284, 81, /* 0.32 */ }, + { 1529, 78, /* 0.31 */ }, + { 1793, 76, /* 0.30 */ }, + { 2076, 74, /* 0.29 */ }, + { 2378, 72, /* 0.28 */ }, + { 2699, 71, /* 0.28 */ }, + { 3039, 69, /* 0.27 */ }, + { 3399, 68, /* 0.27 */ }, + { 3778, 66, /* 0.26 */ }, + { 4177, 65, /* 0.26 */ }, + { 4596, 64, /* 0.25 */ }, + { 5036, 62, /* 0.25 */ }, + { 5497, 61, /* 0.24 */ }, + { 5979, 60, /* 0.24 */ }, + { 6483, 59, /* 0.23 */ }, + { 7009, 58, /* 0.23 */ }, + { 7558, 57, /* 0.22 */ }, + { 8130, 56, /* 0.22 */ }, + { 8726, 55, /* 0.22 */ }, + { 9346, 54, /* 0.21 */ }, + { 9991, 53, /* 0.21 */ }, + { 10661, 52, /* 0.21 */ }, + { 11358, 52, /* 0.20 */ }, + { 12082, 51, /* 0.20 */ }, + { 12834, 50, /* 0.20 */ }, + { 13614, 49, /* 0.19 */ }, + { 14424, 48, /* 0.19 */ }, + { 15265, 48, /* 0.19 */ }, + { 16137, 47, /* 0.19 */ }, + { 17042, 46, /* 0.18 */ }, + { 17981, 45, /* 0.18 */ }, + { 18955, 45, /* 0.18 */ }, + { 19965, 44, /* 0.17 */ }, + { 21013, 43, /* 0.17 */ }, + { 22101, 43, /* 0.17 */ }, + { 23230, 42, /* 0.17 */ }, + { 24402, 41, /* 0.16 */ }, + { 25618, 41, /* 0.16 */ }, + { 26881, 40, /* 0.16 */ }, + { 28193, 39, /* 0.16 */ }, + { 29557, 39, /* 0.15 */ }, + { 30975, 38, /* 0.15 */ }, + { 32450, 38, /* 0.15 */ }, + { 33986, 37, /* 0.15 */ }, + { 35586, 36, /* 0.14 */ }, + { 37253, 36, /* 0.14 */ }, + { 38992, 35, /* 0.14 */ }, + { 40808, 35, /* 0.14 */ }, + { 42707, 34, /* 0.13 */ }, + { 44694, 33, /* 0.13 */ }, + { 46776, 33, /* 0.13 */ }, + { 48961, 32, /* 0.13 */ }, + { 51258, 32, /* 0.13 */ }, + { 53677, 31, /* 0.12 */ }, + { 56230, 30, /* 0.12 */ }, + { 58932, 30, /* 0.12 */ }, + { 61799, 29, /* 0.12 */ }, + { 64851, 28, /* 0.11 */ }, + { 68113, 28, /* 0.11 */ }, + { 71617, 27, /* 0.11 */ }, + { 75401, 26, /* 0.10 */ }, + { 79517, 26, /* 0.10 */ }, + { 84035, 25, /* 0.10 */ }, + { 89053, 24, /* 0.10 */ }, +}; + +#define HSTCP_AIMD_MAX ARRAY_SIZE(hstcp_aimd_vals) + +struct hstcp { + u32 ai; +}; + +static void hstcp_init(struct tcp_sock *tp) +{ + struct hstcp *ca = tcp_ca(tp); + + ca->ai = 0; + + /* Ensure the MD arithmetic works. This is somewhat pedantic, + * since I don't think we will see a cwnd this large. :) */ + tp->snd_cwnd_clamp = min_t(u32, tp->snd_cwnd_clamp, 0xffffffff/128); +} + +static void hstcp_cong_avoid(struct tcp_sock *tp, u32 adk, u32 rtt, + u32 in_flight, int good) +{ + struct hstcp *ca = tcp_ca(tp); + + if (in_flight < tp->snd_cwnd) + return; + + if (tp->snd_cwnd <= tp->snd_ssthresh) { + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + } else { + /* Update AIMD parameters */ + if (tp->snd_cwnd > hstcp_aimd_vals[ca->ai].cwnd) { + while (tp->snd_cwnd > hstcp_aimd_vals[ca->ai].cwnd && + ca->ai < HSTCP_AIMD_MAX) + ca->ai++; + } else if (tp->snd_cwnd < hstcp_aimd_vals[ca->ai].cwnd) { + while (tp->snd_cwnd > hstcp_aimd_vals[ca->ai].cwnd && + ca->ai > 0) + ca->ai--; + } + + /* Do additive increase */ + if (tp->snd_cwnd < tp->snd_cwnd_clamp) { + tp->snd_cwnd_cnt += ca->ai; + if (tp->snd_cwnd_cnt >= tp->snd_cwnd) { + tp->snd_cwnd++; + tp->snd_cwnd_cnt -= tp->snd_cwnd; + } + } + } +} + +static u32 hstcp_ssthresh(struct tcp_sock *tp) +{ + struct hstcp *ca = tcp_ca(tp); + + /* Do multiplicative decrease */ + return max(tp->snd_cwnd - ((tp->snd_cwnd * hstcp_aimd_vals[ca->ai].md) >> 8), 2U); +} + + +static struct tcp_congestion_ops tcp_highspeed = { + .init = hstcp_init, + .ssthresh = hstcp_ssthresh, + .cong_avoid = hstcp_cong_avoid, + .min_cwnd = tcp_reno_min_cwnd, + + .owner = THIS_MODULE, + .name = "highspeed" +}; + +static int __init hstcp_register(void) +{ + BUG_ON(sizeof(struct hstcp) > TCP_CA_PRIV_SIZE); + return tcp_register_congestion_control(&tcp_highspeed); +} + +static void __exit hstcp_unregister(void) +{ + tcp_unregister_congestion_control(&tcp_highspeed); +} + +module_init(hstcp_register); +module_exit(hstcp_unregister); + +MODULE_AUTHOR("John Heffner"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("High Speed TCP"); diff --git a/net/ipv4/tcp_htcp.c b/net/ipv4/tcp_htcp.c new file mode 100644 index 0000000..4016827 --- /dev/null +++ b/net/ipv4/tcp_htcp.c @@ -0,0 +1,289 @@ +/* + * H-TCP congestion control. The algorithm is detailed in: + * R.N.Shorten, D.J.Leith: + * "H-TCP: TCP for high-speed and long-distance networks" + * Proc. PFLDnet, Argonne, 2004. + * http://www.hamilton.ie/net/htcp3.pdf + */ + +#include <linux/config.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <net/tcp.h> + +#define ALPHA_BASE (1<<7) /* 1.0 with shift << 7 */ +#define BETA_MIN (1<<6) /* 0.5 with shift << 7 */ +#define BETA_MAX 102 /* 0.8 with shift << 7 */ + +static int use_rtt_scaling = 1; +module_param(use_rtt_scaling, int, 0644); +MODULE_PARM_DESC(use_rtt_scaling, "turn on/off RTT scaling"); + +static int use_bandwidth_switch = 1; +module_param(use_bandwidth_switch, int, 0644); +MODULE_PARM_DESC(use_bandwidth_switch, "turn on/off bandwidth switcher"); + +struct htcp { + u16 alpha; /* Fixed point arith, << 7 */ + u8 beta; /* Fixed point arith, << 7 */ + u8 modeswitch; /* Delay modeswitch until we had at least one congestion event */ + u8 ccount; /* Number of RTTs since last congestion event */ + u8 undo_ccount; + u16 packetcount; + u32 minRTT; + u32 maxRTT; + u32 snd_cwnd_cnt2; + + u32 undo_maxRTT; + u32 undo_old_maxB; + + /* Bandwidth estimation */ + u32 minB; + u32 maxB; + u32 old_maxB; + u32 Bi; + u32 lasttime; +}; + +static inline void htcp_reset(struct htcp *ca) +{ + ca->undo_ccount = ca->ccount; + ca->undo_maxRTT = ca->maxRTT; + ca->undo_old_maxB = ca->old_maxB; + + ca->ccount = 0; + ca->snd_cwnd_cnt2 = 0; +} + +static u32 htcp_cwnd_undo(struct tcp_sock *tp) +{ + struct htcp *ca = tcp_ca(tp); + ca->ccount = ca->undo_ccount; + ca->maxRTT = ca->undo_maxRTT; + ca->old_maxB = ca->undo_old_maxB; + return max(tp->snd_cwnd, (tp->snd_ssthresh<<7)/ca->beta); +} + +static inline void measure_rtt(struct tcp_sock *tp) +{ + struct htcp *ca = tcp_ca(tp); + u32 srtt = tp->srtt>>3; + + /* keep track of minimum RTT seen so far, minRTT is zero at first */ + if (ca->minRTT > srtt || !ca->minRTT) + ca->minRTT = srtt; + + /* max RTT */ + if (tp->ca_state == TCP_CA_Open && tp->snd_ssthresh < 0xFFFF && ca->ccount > 3) { + if (ca->maxRTT < ca->minRTT) + ca->maxRTT = ca->minRTT; + if (ca->maxRTT < srtt && srtt <= ca->maxRTT+HZ/50) + ca->maxRTT = srtt; + } +} + +static void measure_achieved_throughput(struct tcp_sock *tp, u32 pkts_acked) +{ + struct htcp *ca = tcp_ca(tp); + u32 now = tcp_time_stamp; + + /* achieved throughput calculations */ + if (tp->ca_state != TCP_CA_Open && tp->ca_state != TCP_CA_Disorder) { + ca->packetcount = 0; + ca->lasttime = now; + return; + } + + ca->packetcount += pkts_acked; + + if (ca->packetcount >= tp->snd_cwnd - (ca->alpha>>7? : 1) + && now - ca->lasttime >= ca->minRTT + && ca->minRTT > 0) { + __u32 cur_Bi = ca->packetcount*HZ/(now - ca->lasttime); + if (ca->ccount <= 3) { + /* just after backoff */ + ca->minB = ca->maxB = ca->Bi = cur_Bi; + } else { + ca->Bi = (3*ca->Bi + cur_Bi)/4; + if (ca->Bi > ca->maxB) + ca->maxB = ca->Bi; + if (ca->minB > ca->maxB) + ca->minB = ca->maxB; + } + ca->packetcount = 0; + ca->lasttime = now; + } +} + +static inline void htcp_beta_update(struct htcp *ca, u32 minRTT, u32 maxRTT) +{ + if (use_bandwidth_switch) { + u32 maxB = ca->maxB; + u32 old_maxB = ca->old_maxB; + ca->old_maxB = ca->maxB; + + if (!between(5*maxB, 4*old_maxB, 6*old_maxB)) { + ca->beta = BETA_MIN; + ca->modeswitch = 0; + return; + } + } + + if (ca->modeswitch && minRTT > max(HZ/100, 1) && maxRTT) { + ca->beta = (minRTT<<7)/maxRTT; + if (ca->beta < BETA_MIN) + ca->beta = BETA_MIN; + else if (ca->beta > BETA_MAX) + ca->beta = BETA_MAX; + } else { + ca->beta = BETA_MIN; + ca->modeswitch = 1; + } +} + +static inline void htcp_alpha_update(struct htcp *ca) +{ + u32 minRTT = ca->minRTT; + u32 factor = 1; + u32 diff = ca->ccount * minRTT; /* time since last backoff */ + + if (diff > HZ) { + diff -= HZ; + factor = 1+ ( 10*diff + ((diff/2)*(diff/2)/HZ) )/HZ; + } + + if (use_rtt_scaling && minRTT) { + u32 scale = (HZ<<3)/(10*minRTT); + scale = min(max(scale, 1U<<2), 10U<<3); /* clamping ratio to interval [0.5,10]<<3 */ + factor = (factor<<3)/scale; + if (!factor) + factor = 1; + } + + ca->alpha = 2*factor*((1<<7)-ca->beta); + if (!ca->alpha) + ca->alpha = ALPHA_BASE; +} + +/* After we have the rtt data to calculate beta, we'd still prefer to wait one + * rtt before we adjust our beta to ensure we are working from a consistent + * data. + * + * This function should be called when we hit a congestion event since only at + * that point do we really have a real sense of maxRTT (the queues en route + * were getting just too full now). + */ +static void htcp_param_update(struct tcp_sock *tp) +{ + struct htcp *ca = tcp_ca(tp); + u32 minRTT = ca->minRTT; + u32 maxRTT = ca->maxRTT; + + htcp_beta_update(ca, minRTT, maxRTT); + htcp_alpha_update(ca); + + /* add slowly fading memory for maxRTT to accommodate routing changes etc */ + if (minRTT > 0 && maxRTT > minRTT) + ca->maxRTT = minRTT + ((maxRTT-minRTT)*95)/100; +} + +static u32 htcp_recalc_ssthresh(struct tcp_sock *tp) +{ + struct htcp *ca = tcp_ca(tp); + htcp_param_update(tp); + return max((tp->snd_cwnd * ca->beta) >> 7, 2U); +} + +static void htcp_cong_avoid(struct tcp_sock *tp, u32 ack, u32 rtt, + u32 in_flight, int data_acked) +{ + struct htcp *ca = tcp_ca(tp); + + if (in_flight < tp->snd_cwnd) + return; + + if (tp->snd_cwnd <= tp->snd_ssthresh) { + /* In "safe" area, increase. */ + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + } else { + measure_rtt(tp); + + /* keep track of number of round-trip times since last backoff event */ + if (ca->snd_cwnd_cnt2++ > tp->snd_cwnd) { + ca->ccount++; + ca->snd_cwnd_cnt2 = 0; + htcp_alpha_update(ca); + } + + /* In dangerous area, increase slowly. + * In theory this is tp->snd_cwnd += alpha / tp->snd_cwnd + */ + if ((tp->snd_cwnd_cnt++ * ca->alpha)>>7 >= tp->snd_cwnd) { + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + tp->snd_cwnd_cnt = 0; + ca->ccount++; + } + } +} + +/* Lower bound on congestion window. */ +static u32 htcp_min_cwnd(struct tcp_sock *tp) +{ + return tp->snd_ssthresh; +} + + +static void htcp_init(struct tcp_sock *tp) +{ + struct htcp *ca = tcp_ca(tp); + + memset(ca, 0, sizeof(struct htcp)); + ca->alpha = ALPHA_BASE; + ca->beta = BETA_MIN; +} + +static void htcp_state(struct tcp_sock *tp, u8 new_state) +{ + switch (new_state) { + case TCP_CA_CWR: + case TCP_CA_Recovery: + case TCP_CA_Loss: + htcp_reset(tcp_ca(tp)); + break; + } +} + +static struct tcp_congestion_ops htcp = { + .init = htcp_init, + .ssthresh = htcp_recalc_ssthresh, + .min_cwnd = htcp_min_cwnd, + .cong_avoid = htcp_cong_avoid, + .set_state = htcp_state, + .undo_cwnd = htcp_cwnd_undo, + .pkts_acked = measure_achieved_throughput, + .owner = THIS_MODULE, + .name = "htcp", +}; + +static int __init htcp_register(void) +{ + BUG_ON(sizeof(struct htcp) > TCP_CA_PRIV_SIZE); + BUILD_BUG_ON(BETA_MIN >= BETA_MAX); + if (!use_bandwidth_switch) + htcp.pkts_acked = NULL; + return tcp_register_congestion_control(&htcp); +} + +static void __exit htcp_unregister(void) +{ + tcp_unregister_congestion_control(&htcp); +} + +module_init(htcp_register); +module_exit(htcp_unregister); + +MODULE_AUTHOR("Baruch Even"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("H-TCP"); diff --git a/net/ipv4/tcp_hybla.c b/net/ipv4/tcp_hybla.c new file mode 100644 index 0000000..13a6634 --- /dev/null +++ b/net/ipv4/tcp_hybla.c @@ -0,0 +1,187 @@ +/* + * TCP HYBLA + * + * TCP-HYBLA Congestion control algorithm, based on: + * C.Caini, R.Firrincieli, "TCP-Hybla: A TCP Enhancement + * for Heterogeneous Networks", + * International Journal on satellite Communications, + * September 2004 + * Daniele Lacamera + * root at danielinux.net + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <net/tcp.h> + +/* Tcp Hybla structure. */ +struct hybla { + u8 hybla_en; + u32 snd_cwnd_cents; /* Keeps increment values when it is <1, <<7 */ + u32 rho; /* Rho parameter, integer part */ + u32 rho2; /* Rho * Rho, integer part */ + u32 rho_3ls; /* Rho parameter, <<3 */ + u32 rho2_7ls; /* Rho^2, <<7 */ + u32 minrtt; /* Minimum smoothed round trip time value seen */ +}; + +/* Hybla reference round trip time (default= 1/40 sec = 25 ms), + expressed in jiffies */ +static int rtt0 = 25; +module_param(rtt0, int, 0644); +MODULE_PARM_DESC(rtt0, "reference rout trip time (ms)"); + + +/* This is called to refresh values for hybla parameters */ +static inline void hybla_recalc_param (struct tcp_sock *tp) +{ + struct hybla *ca = tcp_ca(tp); + + ca->rho_3ls = max_t(u32, tp->srtt / msecs_to_jiffies(rtt0), 8); + ca->rho = ca->rho_3ls >> 3; + ca->rho2_7ls = (ca->rho_3ls * ca->rho_3ls) << 1; + ca->rho2 = ca->rho2_7ls >>7; +} + +static void hybla_init(struct tcp_sock *tp) +{ + struct hybla *ca = tcp_ca(tp); + + ca->rho = 0; + ca->rho2 = 0; + ca->rho_3ls = 0; + ca->rho2_7ls = 0; + ca->snd_cwnd_cents = 0; + ca->hybla_en = 1; + tp->snd_cwnd = 2; + tp->snd_cwnd_clamp = 65535; + + /* 1st Rho measurement based on initial srtt */ + hybla_recalc_param(tp); + + /* set minimum rtt as this is the 1st ever seen */ + ca->minrtt = tp->srtt; + tp->snd_cwnd = ca->rho; +} + +static void hybla_state(struct tcp_sock *tp, u8 ca_state) +{ + struct hybla *ca = tcp_ca(tp); + + ca->hybla_en = (ca_state == TCP_CA_Open); +} + +static inline u32 hybla_fraction(u32 odds) +{ + static const u32 fractions[] = { + 128, 139, 152, 165, 181, 197, 215, 234, + }; + + return (odds < ARRAY_SIZE(fractions)) ? fractions[odds] : 128; +} + +/* TCP Hybla main routine. + * This is the algorithm behavior: + * o Recalc Hybla parameters if min_rtt has changed + * o Give cwnd a new value based on the model proposed + * o remember increments <1 + */ +static void hybla_cong_avoid(struct tcp_sock *tp, u32 ack, u32 rtt, + u32 in_flight, int flag) +{ + struct hybla *ca = tcp_ca(tp); + u32 increment, odd, rho_fractions; + int is_slowstart = 0; + + /* Recalculate rho only if this srtt is the lowest */ + if (tp->srtt < ca->minrtt){ + hybla_recalc_param(tp); + ca->minrtt = tp->srtt; + } + + if (!ca->hybla_en) + return tcp_reno_cong_avoid(tp, ack, rtt, in_flight, flag); + + if (in_flight < tp->snd_cwnd) + return; + + if (ca->rho == 0) + hybla_recalc_param(tp); + + rho_fractions = ca->rho_3ls - (ca->rho << 3); + + if (tp->snd_cwnd < tp->snd_ssthresh) { + /* + * slow start + * INC = 2^RHO - 1 + * This is done by splitting the rho parameter + * into 2 parts: an integer part and a fraction part. + * Inrement<<7 is estimated by doing: + * [2^(int+fract)]<<7 + * that is equal to: + * (2^int) * [(2^fract) <<7] + * 2^int is straightly computed as 1<<int, + * while we will use hybla_slowstart_fraction_increment() to + * calculate 2^fract in a <<7 value. + */ + is_slowstart = 1; + increment = ((1 << ca->rho) * hybla_fraction(rho_fractions)) + - 128; + } else { + /* + * congestion avoidance + * INC = RHO^2 / W + * as long as increment is estimated as (rho<<7)/window + * it already is <<7 and we can easily count its fractions. + */ + increment = ca->rho2_7ls / tp->snd_cwnd; + if (increment < 128) + tp->snd_cwnd_cnt++; + } + + odd = increment % 128; + tp->snd_cwnd += increment >> 7; + ca->snd_cwnd_cents += odd; + + /* check when fractions goes >=128 and increase cwnd by 1. */ + while(ca->snd_cwnd_cents >= 128) { + tp->snd_cwnd++; + ca->snd_cwnd_cents -= 128; + tp->snd_cwnd_cnt = 0; + } + + /* clamp down slowstart cwnd to ssthresh value. */ + if (is_slowstart) + tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh); + + tp->snd_cwnd = min_t(u32, tp->snd_cwnd, tp->snd_cwnd_clamp); +} + +static struct tcp_congestion_ops tcp_hybla = { + .init = hybla_init, + .ssthresh = tcp_reno_ssthresh, + .min_cwnd = tcp_reno_min_cwnd, + .cong_avoid = hybla_cong_avoid, + .set_state = hybla_state, + + .owner = THIS_MODULE, + .name = "hybla" +}; + +static int __init hybla_register(void) +{ + BUG_ON(sizeof(struct hybla) > TCP_CA_PRIV_SIZE); + return tcp_register_congestion_control(&tcp_hybla); +} + +static void __exit hybla_unregister(void) +{ + tcp_unregister_congestion_control(&tcp_hybla); +} + +module_init(hybla_register); +module_exit(hybla_unregister); + +MODULE_AUTHOR("Daniele Lacamera"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("TCP Hybla"); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 79835a6..7bbbbc3 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -61,7 +61,6 @@ * Panu Kuhlberg: Experimental audit of TCP (re)transmission * engine. Lots of bugs are found. * Pasi Sarolahti: F-RTO for dealing with spurious RTOs - * Angelo Dell'Aera: TCP Westwood+ support */ #include <linux/config.h> @@ -88,23 +87,9 @@ int sysctl_tcp_rfc1337; int sysctl_tcp_max_orphans = NR_FILE; int sysctl_tcp_frto; int sysctl_tcp_nometrics_save; -int sysctl_tcp_westwood; -int sysctl_tcp_vegas_cong_avoid; int sysctl_tcp_moderate_rcvbuf = 1; -/* Default values of the Vegas variables, in fixed-point representation - * with V_PARAM_SHIFT bits to the right of the binary point. - */ -#define V_PARAM_SHIFT 1 -int sysctl_tcp_vegas_alpha = 1<<V_PARAM_SHIFT; -int sysctl_tcp_vegas_beta = 3<<V_PARAM_SHIFT; -int sysctl_tcp_vegas_gamma = 1<<V_PARAM_SHIFT; -int sysctl_tcp_bic = 1; -int sysctl_tcp_bic_fast_convergence = 1; -int sysctl_tcp_bic_low_window = 14; -int sysctl_tcp_bic_beta = 819; /* = 819/1024 (BICTCP_BETA_SCALE) */ - #define FLAG_DATA 0x01 /* Incoming frame contained data. */ #define FLAG_WIN_UPDATE 0x02 /* Incoming ACK was a window update. */ #define FLAG_DATA_ACKED 0x04 /* This ACK acknowledged new data. */ @@ -333,15 +318,6 @@ static void tcp_init_buffer_space(struct sock *sk) tp->snd_cwnd_stamp = tcp_time_stamp; } -static void init_bictcp(struct tcp_sock *tp) -{ - tp->bictcp.cnt = 0; - - tp->bictcp.last_max_cwnd = 0; - tp->bictcp.last_cwnd = 0; - tp->bictcp.last_stamp = 0; -} - /* 5. Recalculate window clamp after socket hit its memory bounds. */ static void tcp_clamp_window(struct sock *sk, struct tcp_sock *tp) { @@ -558,45 +534,6 @@ static void tcp_event_data_recv(struct sock *sk, struct tcp_sock *tp, struct sk_ tcp_grow_window(sk, tp, skb); } -/* When starting a new connection, pin down the current choice of - * congestion algorithm. - */ -void tcp_ca_init(struct tcp_sock *tp) -{ - if (sysctl_tcp_westwood) - tp->adv_cong = TCP_WESTWOOD; - else if (sysctl_tcp_bic) - tp->adv_cong = TCP_BIC; - else if (sysctl_tcp_vegas_cong_avoid) { - tp->adv_cong = TCP_VEGAS; - tp->vegas.baseRTT = 0x7fffffff; - tcp_vegas_enable(tp); - } -} - -/* Do RTT sampling needed for Vegas. - * Basically we: - * o min-filter RTT samples from within an RTT to get the current - * propagation delay + queuing delay (we are min-filtering to try to - * avoid the effects of delayed ACKs) - * o min-filter RTT samples from a much longer window (forever for now) - * to find the propagation delay (baseRTT) - */ -static inline void vegas_rtt_calc(struct tcp_sock *tp, __u32 rtt) -{ - __u32 vrtt = rtt + 1; /* Never allow zero rtt or baseRTT */ - - /* Filter to find propagation delay: */ - if (vrtt < tp->vegas.baseRTT) - tp->vegas.baseRTT = vrtt; - - /* Find the min RTT during the last RTT to find - * the current prop. delay + queuing delay: - */ - tp->vegas.minRTT = min(tp->vegas.minRTT, vrtt); - tp->vegas.cntRTT++; -} - /* Called to compute a smoothed rtt estimate. The data fed to this * routine either comes from timestamps, or from segments that were * known _not_ to have been retransmitted [see Karn/Partridge @@ -606,13 +543,10 @@ static inline void vegas_rtt_calc(struct tcp_sock *tp, __u32 rtt) * To save cycles in the RFC 1323 implementation it was better to break * it up into three procedures. -- erics */ -static void tcp_rtt_estimator(struct tcp_sock *tp, __u32 mrtt) +static void tcp_rtt_estimator(struct tcp_sock *tp, __u32 mrtt, u32 *usrtt) { long m = mrtt; /* RTT */ - if (tcp_vegas_enabled(tp)) - vegas_rtt_calc(tp, mrtt); - /* The following amusing code comes from Jacobson's * article in SIGCOMM '88. Note that rtt and mdev * are scaled versions of rtt and mean deviation. @@ -670,7 +604,8 @@ static void tcp_rtt_estimator(struct tcp_sock *tp, __u32 mrtt) tp->rtt_seq = tp->snd_nxt; } - tcp_westwood_update_rtt(tp, tp->srtt >> 3); + if (tp->ca_ops->rtt_sample) + tp->ca_ops->rtt_sample(tp, *usrtt); } /* Calculate rto without backoff. This is the second half of Van Jacobson's @@ -1185,8 +1120,8 @@ void tcp_enter_frto(struct sock *sk) tp->snd_una == tp->high_seq || (tp->ca_state == TCP_CA_Loss && !tp->retransmits)) { tp->prior_ssthresh = tcp_current_ssthresh(tp); - if (!tcp_westwood_ssthresh(tp)) - tp->snd_ssthresh = tcp_recalc_ssthresh(tp); + tp->snd_ssthresh = tp->ca_ops->ssthresh(tp); + tcp_ca_event(tp, CA_EVENT_FRTO); } /* Have to clear retransmission markers here to keep the bookkeeping @@ -1252,8 +1187,6 @@ static void tcp_enter_frto_loss(struct sock *sk) tcp_set_ca_state(tp, TCP_CA_Loss); tp->high_seq = tp->frto_highmark; TCP_ECN_queue_cwr(tp); - - init_bictcp(tp); } void tcp_clear_retrans(struct tcp_sock *tp) @@ -1283,7 +1216,8 @@ void tcp_enter_loss(struct sock *sk, int how) if (tp->ca_state <= TCP_CA_Disorder || tp->snd_una == tp->high_seq || (tp->ca_state == TCP_CA_Loss && !tp->retransmits)) { tp->prior_ssthresh = tcp_current_ssthresh(tp); - tp->snd_ssthresh = tcp_recalc_ssthresh(tp); + tp->snd_ssthresh = tp->ca_ops->ssthresh(tp); + tcp_ca_event(tp, CA_EVENT_LOSS); } tp->snd_cwnd = 1; tp->snd_cwnd_cnt = 0; @@ -1596,28 +1530,14 @@ static inline void tcp_moderate_cwnd(struct tcp_sock *tp) } /* Decrease cwnd each second ack. */ - static void tcp_cwnd_down(struct tcp_sock *tp) { int decr = tp->snd_cwnd_cnt + 1; - __u32 limit; - - /* - * TCP Westwood - * Here limit is evaluated as BWestimation*RTTmin (for obtaining it - * in packets we use mss_cache). If sysctl_tcp_westwood is off - * tcp_westwood_bw_rttmin() returns 0. In such case snd_ssthresh is - * still used as usual. It prevents other strange cases in which - * BWE*RTTmin could assume value 0. It should not happen but... - */ - - if (!(limit = tcp_westwood_bw_rttmin(tp))) - limit = tp->snd_ssthresh/2; tp->snd_cwnd_cnt = decr&1; decr >>= 1; - if (decr && tp->snd_cwnd > limit) + if (decr && tp->snd_cwnd > tp->ca_ops->min_cwnd(tp)) tp->snd_cwnd -= decr; tp->snd_cwnd = min(tp->snd_cwnd, tcp_packets_in_flight(tp)+1); @@ -1654,8 +1574,8 @@ static void DBGUNDO(struct sock *sk, struct tcp_sock *tp, const char *msg) static void tcp_undo_cwr(struct tcp_sock *tp, int undo) { if (tp->prior_ssthresh) { - if (tcp_is_bic(tp)) - tp->snd_cwnd = max(tp->snd_cwnd, tp->bictcp.last_max_cwnd); + if (tp->ca_ops->undo_cwnd) + tp->snd_cwnd = tp->ca_ops->undo_cwnd(tp); else tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh<<1); @@ -1767,11 +1687,9 @@ static int tcp_try_undo_loss(struct sock *sk, struct tcp_sock *tp) static inline void tcp_complete_cwr(struct tcp_sock *tp) { - if (tcp_westwood_cwnd(tp)) - tp->snd_ssthresh = tp->snd_cwnd; - else - tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh); + tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh); tp->snd_cwnd_stamp = tcp_time_stamp; + tcp_ca_event(tp, CA_EVENT_COMPLETE_CWR); } static void tcp_try_to_open(struct sock *sk, struct tcp_sock *tp, int flag) @@ -1946,7 +1864,7 @@ tcp_fastretrans_alert(struct sock *sk, u32 prior_snd_una, if (tp->ca_state < TCP_CA_CWR) { if (!(flag&FLAG_ECE)) tp->prior_ssthresh = tcp_current_ssthresh(tp); - tp->snd_ssthresh = tcp_recalc_ssthresh(tp); + tp->snd_ssthresh = tp->ca_ops->ssthresh(tp); TCP_ECN_queue_cwr(tp); } @@ -1963,7 +1881,7 @@ tcp_fastretrans_alert(struct sock *sk, u32 prior_snd_una, /* Read draft-ietf-tcplw-high-performance before mucking * with this code. (Superceeds RFC1323) */ -static void tcp_ack_saw_tstamp(struct tcp_sock *tp, int flag) +static void tcp_ack_saw_tstamp(struct tcp_sock *tp, u32 *usrtt, int flag) { __u32 seq_rtt; @@ -1983,13 +1901,13 @@ static void tcp_ack_saw_tstamp(struct tcp_sock *tp, int flag) * in window is lost... Voila. --ANK (010210) */ seq_rtt = tcp_time_stamp - tp->rx_opt.rcv_tsecr; - tcp_rtt_estimator(tp, seq_rtt); + tcp_rtt_estimator(tp, seq_rtt, usrtt); tcp_set_rto(tp); tp->backoff = 0; tcp_bound_rto(tp); } -static void tcp_ack_no_tstamp(struct tcp_sock *tp, u32 seq_rtt, int flag) +static void tcp_ack_no_tstamp(struct tcp_sock *tp, u32 seq_rtt, u32 *usrtt, int flag) { /* We don't have a timestamp. Can only use * packets that are not retransmitted to determine @@ -2003,338 +1921,29 @@ static void tcp_ack_no_tstamp(struct tcp_sock *tp, u32 seq_rtt, int flag) if (flag & FLAG_RETRANS_DATA_ACKED) return; - tcp_rtt_estimator(tp, seq_rtt); + tcp_rtt_estimator(tp, seq_rtt, usrtt); tcp_set_rto(tp); tp->backoff = 0; tcp_bound_rto(tp); } static inline void tcp_ack_update_rtt(struct tcp_sock *tp, - int flag, s32 seq_rtt) + int flag, s32 seq_rtt, u32 *usrtt) { /* Note that peer MAY send zero echo. In this case it is ignored. (rfc1323) */ if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr) - tcp_ack_saw_tstamp(tp, flag); + tcp_ack_saw_tstamp(tp, usrtt, flag); else if (seq_rtt >= 0) - tcp_ack_no_tstamp(tp, seq_rtt, flag); -} - -/* - * Compute congestion window to use. - * - * This is from the implementation of BICTCP in - * Lison-Xu, Kahaled Harfoush, and Injog Rhee. - * "Binary Increase Congestion Control for Fast, Long Distance - * Networks" in InfoComm 2004 - * Available from: - * http://www.csc.ncsu.edu/faculty/rhee/export/bitcp.pdf - * - * Unless BIC is enabled and congestion window is large - * this behaves the same as the original Reno. - */ -static inline __u32 bictcp_cwnd(struct tcp_sock *tp) -{ - /* orignal Reno behaviour */ - if (!tcp_is_bic(tp)) - return tp->snd_cwnd; - - if (tp->bictcp.last_cwnd == tp->snd_cwnd && - (s32)(tcp_time_stamp - tp->bictcp.last_stamp) <= (HZ>>5)) - return tp->bictcp.cnt; - - tp->bictcp.last_cwnd = tp->snd_cwnd; - tp->bictcp.last_stamp = tcp_time_stamp; - - /* start off normal */ - if (tp->snd_cwnd <= sysctl_tcp_bic_low_window) - tp->bictcp.cnt = tp->snd_cwnd; - - /* binary increase */ - else if (tp->snd_cwnd < tp->bictcp.last_max_cwnd) { - __u32 dist = (tp->bictcp.last_max_cwnd - tp->snd_cwnd) - / BICTCP_B; - - if (dist > BICTCP_MAX_INCREMENT) - /* linear increase */ - tp->bictcp.cnt = tp->snd_cwnd / BICTCP_MAX_INCREMENT; - else if (dist <= 1U) - /* binary search increase */ - tp->bictcp.cnt = tp->snd_cwnd * BICTCP_FUNC_OF_MIN_INCR - / BICTCP_B; - else - /* binary search increase */ - tp->bictcp.cnt = tp->snd_cwnd / dist; - } else { - /* slow start amd linear increase */ - if (tp->snd_cwnd < tp->bictcp.last_max_cwnd + BICTCP_B) - /* slow start */ - tp->bictcp.cnt = tp->snd_cwnd * BICTCP_FUNC_OF_MIN_INCR - / BICTCP_B; - else if (tp->snd_cwnd < tp->bictcp.last_max_cwnd - + BICTCP_MAX_INCREMENT*(BICTCP_B-1)) - /* slow start */ - tp->bictcp.cnt = tp->snd_cwnd * (BICTCP_B-1) - / (tp->snd_cwnd-tp->bictcp.last_max_cwnd); - else - /* linear increase */ - tp->bictcp.cnt = tp->snd_cwnd / BICTCP_MAX_INCREMENT; - } - return tp->bictcp.cnt; + tcp_ack_no_tstamp(tp, seq_rtt, usrtt, flag); } -/* This is Jacobson's slow start and congestion avoidance. - * SIGCOMM '88, p. 328. - */ -static inline void reno_cong_avoid(struct tcp_sock *tp) +static inline void tcp_cong_avoid(struct tcp_sock *tp, u32 ack, u32 rtt, + u32 in_flight, int good) { - if (tp->snd_cwnd <= tp->snd_ssthresh) { - /* In "safe" area, increase. */ - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; - } else { - /* In dangerous area, increase slowly. - * In theory this is tp->snd_cwnd += 1 / tp->snd_cwnd - */ - if (tp->snd_cwnd_cnt >= bictcp_cwnd(tp)) { - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; - tp->snd_cwnd_cnt=0; - } else - tp->snd_cwnd_cnt++; - } - tp->snd_cwnd_stamp = tcp_time_stamp; -} - -/* This is based on the congestion detection/avoidance scheme described in - * Lawrence S. Brakmo and Larry L. Peterson. - * "TCP Vegas: End to end congestion avoidance on a global internet." - * IEEE Journal on Selected Areas in Communication, 13(8):1465--1480, - * October 1995. Available from: - * ftp://ftp.cs.arizona.edu/xkernel/Papers/jsac.ps - * - * See http://www.cs.arizona.edu/xkernel/ for their implementation. - * The main aspects that distinguish this implementation from the - * Arizona Vegas implementation are: - * o We do not change the loss detection or recovery mechanisms of - * Linux in any way. Linux already recovers from losses quite well, - * using fine-grained timers, NewReno, and FACK. - * o To avoid the performance penalty imposed by increasing cwnd - * only every-other RTT during slow start, we increase during - * every RTT during slow start, just like Reno. - * o Largely to allow continuous cwnd growth during slow start, - * we use the rate at which ACKs come back as the "actual" - * rate, rather than the rate at which data is sent. - * o To speed convergence to the right rate, we set the cwnd - * to achieve the right ("actual") rate when we exit slow start. - * o To filter out the noise caused by delayed ACKs, we use the - * minimum RTT sample observed during the last RTT to calculate - * the actual rate. - * o When the sender re-starts from idle, it waits until it has - * received ACKs for an entire flight of new data before making - * a cwnd adjustment decision. The original Vegas implementation - * assumed senders never went idle. - */ -static void vegas_cong_avoid(struct tcp_sock *tp, u32 ack, u32 seq_rtt) -{ - /* The key players are v_beg_snd_una and v_beg_snd_nxt. - * - * These are so named because they represent the approximate values - * of snd_una and snd_nxt at the beginning of the current RTT. More - * precisely, they represent the amount of data sent during the RTT. - * At the end of the RTT, when we receive an ACK for v_beg_snd_nxt, - * we will calculate that (v_beg_snd_nxt - v_beg_snd_una) outstanding - * bytes of data have been ACKed during the course of the RTT, giving - * an "actual" rate of: - * - * (v_beg_snd_nxt - v_beg_snd_una) / (rtt duration) - * - * Unfortunately, v_beg_snd_una is not exactly equal to snd_una, - * because delayed ACKs can cover more than one segment, so they - * don't line up nicely with the boundaries of RTTs. - * - * Another unfortunate fact of life is that delayed ACKs delay the - * advance of the left edge of our send window, so that the number - * of bytes we send in an RTT is often less than our cwnd will allow. - * So we keep track of our cwnd separately, in v_beg_snd_cwnd. - */ - - if (after(ack, tp->vegas.beg_snd_nxt)) { - /* Do the Vegas once-per-RTT cwnd adjustment. */ - u32 old_wnd, old_snd_cwnd; - - - /* Here old_wnd is essentially the window of data that was - * sent during the previous RTT, and has all - * been acknowledged in the course of the RTT that ended - * with the ACK we just received. Likewise, old_snd_cwnd - * is the cwnd during the previous RTT. - */ - old_wnd = (tp->vegas.beg_snd_nxt - tp->vegas.beg_snd_una) / - tp->mss_cache_std; - old_snd_cwnd = tp->vegas.beg_snd_cwnd; - - /* Save the extent of the current window so we can use this - * at the end of the next RTT. - */ - tp->vegas.beg_snd_una = tp->vegas.beg_snd_nxt; - tp->vegas.beg_snd_nxt = tp->snd_nxt; - tp->vegas.beg_snd_cwnd = tp->snd_cwnd; - - /* Take into account the current RTT sample too, to - * decrease the impact of delayed acks. This double counts - * this sample since we count it for the next window as well, - * but that's not too awful, since we're taking the min, - * rather than averaging. - */ - vegas_rtt_calc(tp, seq_rtt); - - /* We do the Vegas calculations only if we got enough RTT - * samples that we can be reasonably sure that we got - * at least one RTT sample that wasn't from a delayed ACK. - * If we only had 2 samples total, - * then that means we're getting only 1 ACK per RTT, which - * means they're almost certainly delayed ACKs. - * If we have 3 samples, we should be OK. - */ - - if (tp->vegas.cntRTT <= 2) { - /* We don't have enough RTT samples to do the Vegas - * calculation, so we'll behave like Reno. - */ - if (tp->snd_cwnd > tp->snd_ssthresh) - tp->snd_cwnd++; - } else { - u32 rtt, target_cwnd, diff; - - /* We have enough RTT samples, so, using the Vegas - * algorithm, we determine if we should increase or - * decrease cwnd, and by how much. - */ - - /* Pluck out the RTT we are using for the Vegas - * calculations. This is the min RTT seen during the - * last RTT. Taking the min filters out the effects - * of delayed ACKs, at the cost of noticing congestion - * a bit later. - */ - rtt = tp->vegas.minRTT; - - /* Calculate the cwnd we should have, if we weren't - * going too fast. - * - * This is: - * (actual rate in segments) * baseRTT - * We keep it as a fixed point number with - * V_PARAM_SHIFT bits to the right of the binary point. - */ - target_cwnd = ((old_wnd * tp->vegas.baseRTT) - << V_PARAM_SHIFT) / rtt; - - /* Calculate the difference between the window we had, - * and the window we would like to have. This quantity - * is the "Diff" from the Arizona Vegas papers. - * - * Again, this is a fixed point number with - * V_PARAM_SHIFT bits to the right of the binary - * point. - */ - diff = (old_wnd << V_PARAM_SHIFT) - target_cwnd; - - if (tp->snd_cwnd < tp->snd_ssthresh) { - /* Slow start. */ - if (diff > sysctl_tcp_vegas_gamma) { - /* Going too fast. Time to slow down - * and switch to congestion avoidance. - */ - tp->snd_ssthresh = 2; - - /* Set cwnd to match the actual rate - * exactly: - * cwnd = (actual rate) * baseRTT - * Then we add 1 because the integer - * truncation robs us of full link - * utilization. - */ - tp->snd_cwnd = min(tp->snd_cwnd, - (target_cwnd >> - V_PARAM_SHIFT)+1); - - } - } else { - /* Congestion avoidance. */ - u32 next_snd_cwnd; - - /* Figure out where we would like cwnd - * to be. - */ - if (diff > sysctl_tcp_vegas_beta) { - /* The old window was too fast, so - * we slow down. - */ - next_snd_cwnd = old_snd_cwnd - 1; - } else if (diff < sysctl_tcp_vegas_alpha) { - /* We don't have enough extra packets - * in the network, so speed up. - */ - next_snd_cwnd = old_snd_cwnd + 1; - } else { - /* Sending just as fast as we - * should be. - */ - next_snd_cwnd = old_snd_cwnd; - } - - /* Adjust cwnd upward or downward, toward the - * desired value. - */ - if (next_snd_cwnd > tp->snd_cwnd) - tp->snd_cwnd++; - else if (next_snd_cwnd < tp->snd_cwnd) - tp->snd_cwnd--; - } - } - - /* Wipe the slate clean for the next RTT. */ - tp->vegas.cntRTT = 0; - tp->vegas.minRTT = 0x7fffffff; - } - - /* The following code is executed for every ack we receive, - * except for conditions checked in should_advance_cwnd() - * before the call to tcp_cong_avoid(). Mainly this means that - * we only execute this code if the ack actually acked some - * data. - */ - - /* If we are in slow start, increase our cwnd in response to this ACK. - * (If we are not in slow start then we are in congestion avoidance, - * and adjust our congestion window only once per RTT. See the code - * above.) - */ - if (tp->snd_cwnd <= tp->snd_ssthresh) - tp->snd_cwnd++; - - /* to keep cwnd from growing without bound */ - tp->snd_cwnd = min_t(u32, tp->snd_cwnd, tp->snd_cwnd_clamp); - - /* Make sure that we are never so timid as to reduce our cwnd below - * 2 MSS. - * - * Going below 2 MSS would risk huge delayed ACKs from our receiver. - */ - tp->snd_cwnd = max(tp->snd_cwnd, 2U); - + tp->ca_ops->cong_avoid(tp, ack, rtt, in_flight, good); tp->snd_cwnd_stamp = tcp_time_stamp; } -static inline void tcp_cong_avoid(struct tcp_sock *tp, u32 ack, u32 seq_rtt) -{ - if (tcp_vegas_enabled(tp)) - vegas_cong_avoid(tp, ack, seq_rtt); - else - reno_cong_avoid(tp); -} - /* Restart timer after forward progress on connection. * RFC2988 recommends to restart timer to now+rto. */ @@ -2415,13 +2024,18 @@ static int tcp_tso_acked(struct sock *sk, struct sk_buff *skb, /* Remove acknowledged frames from the retransmission queue. */ -static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) +static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p, s32 *seq_usrtt) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; __u32 now = tcp_time_stamp; int acked = 0; __s32 seq_rtt = -1; + struct timeval usnow; + u32 pkts_acked = 0; + + if (seq_usrtt) + do_gettimeofday(&usnow); while ((skb = skb_peek(&sk->sk_write_queue)) && skb != sk->sk_send_head) { @@ -2448,6 +2062,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) */ if (!(scb->flags & TCPCB_FLAG_SYN)) { acked |= FLAG_DATA_ACKED; + ++pkts_acked; } else { acked |= FLAG_SYN_ACKED; tp->retrans_stamp = 0; @@ -2461,6 +2076,10 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) seq_rtt = -1; } else if (seq_rtt < 0) seq_rtt = now - scb->when; + if (seq_usrtt) + *seq_usrtt = (usnow.tv_sec - skb->stamp.tv_sec) * 1000000 + + (usnow.tv_usec - skb->stamp.tv_usec); + if (sacked & TCPCB_SACKED_ACKED) tp->sacked_out -= tcp_skb_pcount(skb); if (sacked & TCPCB_LOST) @@ -2479,8 +2098,11 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) } if (acked&FLAG_ACKED) { - tcp_ack_update_rtt(tp, acked, seq_rtt); + tcp_ack_update_rtt(tp, acked, seq_rtt, seq_usrtt); tcp_ack_packets_out(sk, tp); + + if (tp->ca_ops->pkts_acked) + tp->ca_ops->pkts_acked(tp, pkts_acked); } #if FASTRETRANS_DEBUG > 0 @@ -2624,257 +2246,6 @@ static void tcp_process_frto(struct sock *sk, u32 prior_snd_una) tp->frto_counter = (tp->frto_counter + 1) % 3; } -/* - * TCP Westwood+ - */ - -/* - * @init_westwood - * This function initializes fields used in TCP Westwood+. We can't - * get no information about RTTmin at this time so we simply set it to - * TCP_WESTWOOD_INIT_RTT. This value was chosen to be too conservative - * since in this way we're sure it will be updated in a consistent - * way as soon as possible. It will reasonably happen within the first - * RTT period of the connection lifetime. - */ - -static void init_westwood(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - - tp->westwood.bw_ns_est = 0; - tp->westwood.bw_est = 0; - tp->westwood.accounted = 0; - tp->westwood.cumul_ack = 0; - tp->westwood.rtt_win_sx = tcp_time_stamp; - tp->westwood.rtt = TCP_WESTWOOD_INIT_RTT; - tp->westwood.rtt_min = TCP_WESTWOOD_INIT_RTT; - tp->westwood.snd_una = tp->snd_una; -} - -/* - * @westwood_do_filter - * Low-pass filter. Implemented using constant coeffients. - */ - -static inline __u32 westwood_do_filter(__u32 a, __u32 b) -{ - return (((7 * a) + b) >> 3); -} - -static void westwood_filter(struct sock *sk, __u32 delta) -{ - struct tcp_sock *tp = tcp_sk(sk); - - tp->westwood.bw_ns_est = - westwood_do_filter(tp->westwood.bw_ns_est, - tp->westwood.bk / delta); - tp->westwood.bw_est = - westwood_do_filter(tp->westwood.bw_est, - tp->westwood.bw_ns_est); -} - -/* - * @westwood_update_rttmin - * It is used to update RTTmin. In this case we MUST NOT use - * WESTWOOD_RTT_MIN minimum bound since we could be on a LAN! - */ - -static inline __u32 westwood_update_rttmin(const struct sock *sk) -{ - const struct tcp_sock *tp = tcp_sk(sk); - __u32 rttmin = tp->westwood.rtt_min; - - if (tp->westwood.rtt != 0 && - (tp->westwood.rtt < tp->westwood.rtt_min || !rttmin)) - rttmin = tp->westwood.rtt; - - return rttmin; -} - -/* - * @westwood_acked - * Evaluate increases for dk. - */ - -static inline __u32 westwood_acked(const struct sock *sk) -{ - const struct tcp_sock *tp = tcp_sk(sk); - - return tp->snd_una - tp->westwood.snd_una; -} - -/* - * @westwood_new_window - * It evaluates if we are receiving data inside the same RTT window as - * when we started. - * Return value: - * It returns 0 if we are still evaluating samples in the same RTT - * window, 1 if the sample has to be considered in the next window. - */ - -static int westwood_new_window(const struct sock *sk) -{ - const struct tcp_sock *tp = tcp_sk(sk); - __u32 left_bound; - __u32 rtt; - int ret = 0; - - left_bound = tp->westwood.rtt_win_sx; - rtt = max(tp->westwood.rtt, (u32) TCP_WESTWOOD_RTT_MIN); - - /* - * A RTT-window has passed. Be careful since if RTT is less than - * 50ms we don't filter but we continue 'building the sample'. - * This minimum limit was choosen since an estimation on small - * time intervals is better to avoid... - * Obvioulsy on a LAN we reasonably will always have - * right_bound = left_bound + WESTWOOD_RTT_MIN - */ - - if ((left_bound + rtt) < tcp_time_stamp) - ret = 1; - - return ret; -} - -/* - * @westwood_update_window - * It updates RTT evaluation window if it is the right moment to do - * it. If so it calls filter for evaluating bandwidth. - */ - -static void __westwood_update_window(struct sock *sk, __u32 now) -{ - struct tcp_sock *tp = tcp_sk(sk); - __u32 delta = now - tp->westwood.rtt_win_sx; - - if (delta) { - if (tp->westwood.rtt) - westwood_filter(sk, delta); - - tp->westwood.bk = 0; - tp->westwood.rtt_win_sx = tcp_time_stamp; - } -} - - -static void westwood_update_window(struct sock *sk, __u32 now) -{ - if (westwood_new_window(sk)) - __westwood_update_window(sk, now); -} - -/* - * @__tcp_westwood_fast_bw - * It is called when we are in fast path. In particular it is called when - * header prediction is successfull. In such case infact update is - * straight forward and doesn't need any particular care. - */ - -static void __tcp_westwood_fast_bw(struct sock *sk, struct sk_buff *skb) -{ - struct tcp_sock *tp = tcp_sk(sk); - - westwood_update_window(sk, tcp_time_stamp); - - tp->westwood.bk += westwood_acked(sk); - tp->westwood.snd_una = tp->snd_una; - tp->westwood.rtt_min = westwood_update_rttmin(sk); -} - -static inline void tcp_westwood_fast_bw(struct sock *sk, struct sk_buff *skb) -{ - if (tcp_is_westwood(tcp_sk(sk))) - __tcp_westwood_fast_bw(sk, skb); -} - - -/* - * @westwood_dupack_update - * It updates accounted and cumul_ack when receiving a dupack. - */ - -static void westwood_dupack_update(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - - tp->westwood.accounted += tp->mss_cache_std; - tp->westwood.cumul_ack = tp->mss_cache_std; -} - -static inline int westwood_may_change_cumul(struct tcp_sock *tp) -{ - return (tp->westwood.cumul_ack > tp->mss_cache_std); -} - -static inline void westwood_partial_update(struct tcp_sock *tp) -{ - tp->westwood.accounted -= tp->westwood.cumul_ack; - tp->westwood.cumul_ack = tp->mss_cache_std; -} - -static inline void westwood_complete_update(struct tcp_sock *tp) -{ - tp->westwood.cumul_ack -= tp->westwood.accounted; - tp->westwood.accounted = 0; -} - -/* - * @westwood_acked_count - * This function evaluates cumul_ack for evaluating dk in case of - * delayed or partial acks. - */ - -static inline __u32 westwood_acked_count(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - - tp->westwood.cumul_ack = westwood_acked(sk); - - /* If cumul_ack is 0 this is a dupack since it's not moving - * tp->snd_una. - */ - if (!(tp->westwood.cumul_ack)) - westwood_dupack_update(sk); - - if (westwood_may_change_cumul(tp)) { - /* Partial or delayed ack */ - if (tp->westwood.accounted >= tp->westwood.cumul_ack) - westwood_partial_update(tp); - else - westwood_complete_update(tp); - } - - tp->westwood.snd_una = tp->snd_una; - - return tp->westwood.cumul_ack; -} - - -/* - * @__tcp_westwood_slow_bw - * It is called when something is going wrong..even if there could - * be no problems! Infact a simple delayed packet may trigger a - * dupack. But we need to be careful in such case. - */ - -static void __tcp_westwood_slow_bw(struct sock *sk, struct sk_buff *skb) -{ - struct tcp_sock *tp = tcp_sk(sk); - - westwood_update_window(sk, tcp_time_stamp); - - tp->westwood.bk += westwood_acked_count(sk); - tp->westwood.rtt_min = westwood_update_rttmin(sk); -} - -static inline void tcp_westwood_slow_bw(struct sock *sk, struct sk_buff *skb) -{ - if (tcp_is_westwood(tcp_sk(sk))) - __tcp_westwood_slow_bw(sk, skb); -} - /* This routine deals with incoming acks, but not outgoing ones. */ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) { @@ -2884,6 +2255,7 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) u32 ack = TCP_SKB_CB(skb)->ack_seq; u32 prior_in_flight; s32 seq_rtt; + s32 seq_usrtt = 0; int prior_packets; /* If the ack is newer than sent or older than previous acks @@ -2902,9 +2274,10 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) */ tcp_update_wl(tp, ack, ack_seq); tp->snd_una = ack; - tcp_westwood_fast_bw(sk, skb); flag |= FLAG_WIN_UPDATE; + tcp_ca_event(tp, CA_EVENT_FAST_ACK); + NET_INC_STATS_BH(LINUX_MIB_TCPHPACKS); } else { if (ack_seq != TCP_SKB_CB(skb)->end_seq) @@ -2920,7 +2293,7 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) if (TCP_ECN_rcv_ecn_echo(tp, skb->h.th)) flag |= FLAG_ECE; - tcp_westwood_slow_bw(sk,skb); + tcp_ca_event(tp, CA_EVENT_SLOW_ACK); } /* We passed data and got it acked, remove any soft error @@ -2935,22 +2308,20 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) prior_in_flight = tcp_packets_in_flight(tp); /* See if we can take anything off of the retransmit queue. */ - flag |= tcp_clean_rtx_queue(sk, &seq_rtt); + flag |= tcp_clean_rtx_queue(sk, &seq_rtt, + tp->ca_ops->rtt_sample ? &seq_usrtt : NULL); if (tp->frto_counter) tcp_process_frto(sk, prior_snd_una); if (tcp_ack_is_dubious(tp, flag)) { /* Advanve CWND, if state allows this. */ - if ((flag & FLAG_DATA_ACKED) && - (tcp_vegas_enabled(tp) || prior_in_flight >= tp->snd_cwnd) && - tcp_may_raise_cwnd(tp, flag)) - tcp_cong_avoid(tp, ack, seq_rtt); + if ((flag & FLAG_DATA_ACKED) && tcp_may_raise_cwnd(tp, flag)) + tcp_cong_avoid(tp, ack, seq_rtt, prior_in_flight, 0); tcp_fastretrans_alert(sk, prior_snd_una, prior_packets, flag); } else { - if ((flag & FLAG_DATA_ACKED) && - (tcp_vegas_enabled(tp) || prior_in_flight >= tp->snd_cwnd)) - tcp_cong_avoid(tp, ack, seq_rtt); + if ((flag & FLAG_DATA_ACKED)) + tcp_cong_avoid(tp, ack, seq_rtt, prior_in_flight, 1); } if ((flag & FLAG_FORWARD_PROGRESS) || !(flag&FLAG_NOT_DUP)) @@ -4355,16 +3726,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, goto no_ack; } - if (eaten) { - if (tcp_in_quickack_mode(tp)) { - tcp_send_ack(sk); - } else { - tcp_send_delayed_ack(sk); - } - } else { - __tcp_ack_snd_check(sk, 0); - } - + __tcp_ack_snd_check(sk, 0); no_ack: if (eaten) __kfree_skb(skb); @@ -4561,6 +3923,8 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, tcp_init_metrics(sk); + tcp_init_congestion_control(tp); + /* Prevent spurious tcp_cwnd_restart() on first data * packet. */ @@ -4717,9 +4081,6 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, if(tp->af_specific->conn_request(sk, skb) < 0) return 1; - init_westwood(sk); - init_bictcp(tp); - /* Now we have several options: In theory there is * nothing else in the frame. KA9Q has an option to * send data with the syn, BSD accepts data with the @@ -4741,9 +4102,6 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, goto discard; case TCP_SYN_SENT: - init_westwood(sk); - init_bictcp(tp); - queued = tcp_rcv_synsent_state_process(sk, skb, th, len); if (queued >= 0) return queued; @@ -4825,7 +4183,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, */ if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr && !tp->srtt) - tcp_ack_saw_tstamp(tp, 0); + tcp_ack_saw_tstamp(tp, 0, 0); if (tp->rx_opt.tstamp_ok) tp->advmss -= TCPOLEN_TSTAMP_ALIGNED; @@ -4837,6 +4195,8 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, tcp_init_metrics(sk); + tcp_init_congestion_control(tp); + /* Prevent spurious tcp_cwnd_restart() on * first data packet. */ diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index dad98e4..ebf1123 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -36,7 +36,7 @@ * ACK bit. * Andi Kleen : Implemented fast path mtu discovery. * Fixed many serious bugs in the - * open_request handling and moved + * request_sock handling and moved * most of it into the af independent code. * Added tail drop and some other bugfixes. * Added new listen sematics. @@ -869,21 +869,23 @@ static __inline__ u32 tcp_v4_synq_hash(u32 raddr, u16 rport, u32 rnd) return (jhash_2words(raddr, (u32) rport, rnd) & (TCP_SYNQ_HSIZE - 1)); } -static struct open_request *tcp_v4_search_req(struct tcp_sock *tp, - struct open_request ***prevp, +static struct request_sock *tcp_v4_search_req(struct tcp_sock *tp, + struct request_sock ***prevp, __u16 rport, __u32 raddr, __u32 laddr) { - struct tcp_listen_opt *lopt = tp->listen_opt; - struct open_request *req, **prev; + struct listen_sock *lopt = tp->accept_queue.listen_opt; + struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v4_synq_hash(raddr, rport, lopt->hash_rnd)]; (req = *prev) != NULL; prev = &req->dl_next) { - if (req->rmt_port == rport && - req->af.v4_req.rmt_addr == raddr && - req->af.v4_req.loc_addr == laddr && - TCP_INET_FAMILY(req->class->family)) { + const struct inet_request_sock *ireq = inet_rsk(req); + + if (ireq->rmt_port == rport && + ireq->rmt_addr == raddr && + ireq->loc_addr == laddr && + TCP_INET_FAMILY(req->rsk_ops->family)) { BUG_TRAP(!req->sk); *prevp = prev; break; @@ -893,21 +895,13 @@ static struct open_request *tcp_v4_search_req(struct tcp_sock *tp, return req; } -static void tcp_v4_synq_add(struct sock *sk, struct open_request *req) +static void tcp_v4_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; - u32 h = tcp_v4_synq_hash(req->af.v4_req.rmt_addr, req->rmt_port, lopt->hash_rnd); - - req->expires = jiffies + TCP_TIMEOUT_INIT; - req->retrans = 0; - req->sk = NULL; - req->dl_next = lopt->syn_table[h]; - - write_lock(&tp->syn_wait_lock); - lopt->syn_table[h] = req; - write_unlock(&tp->syn_wait_lock); + struct listen_sock *lopt = tp->accept_queue.listen_opt; + u32 h = tcp_v4_synq_hash(inet_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); + reqsk_queue_hash_req(&tp->accept_queue, h, req, TCP_TIMEOUT_INIT); tcp_synq_added(sk); } @@ -1050,7 +1044,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info) } switch (sk->sk_state) { - struct open_request *req, **prev; + struct request_sock *req, **prev; case TCP_LISTEN: if (sock_owned_by_user(sk)) goto out; @@ -1065,7 +1059,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info) */ BUG_TRAP(!req->sk); - if (seq != req->snt_isn) { + if (seq != tcp_rsk(req)->snt_isn) { NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS); goto out; } @@ -1254,28 +1248,29 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb) tcp_tw_put(tw); } -static void tcp_v4_or_send_ack(struct sk_buff *skb, struct open_request *req) +static void tcp_v4_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req) { - tcp_v4_send_ack(skb, req->snt_isn + 1, req->rcv_isn + 1, req->rcv_wnd, + tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } static struct dst_entry* tcp_v4_route_req(struct sock *sk, - struct open_request *req) + struct request_sock *req) { struct rtable *rt; - struct ip_options *opt = req->af.v4_req.opt; + const struct inet_request_sock *ireq = inet_rsk(req); + struct ip_options *opt = inet_rsk(req)->opt; struct flowi fl = { .oif = sk->sk_bound_dev_if, .nl_u = { .ip4_u = { .daddr = ((opt && opt->srr) ? opt->faddr : - req->af.v4_req.rmt_addr), - .saddr = req->af.v4_req.loc_addr, + ireq->rmt_addr), + .saddr = ireq->loc_addr, .tos = RT_CONN_FLAGS(sk) } }, .proto = IPPROTO_TCP, .uli_u = { .ports = { .sport = inet_sk(sk)->sport, - .dport = req->rmt_port } } }; + .dport = ireq->rmt_port } } }; if (ip_route_output_flow(&rt, &fl, sk, 0)) { IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES); @@ -1291,12 +1286,13 @@ static struct dst_entry* tcp_v4_route_req(struct sock *sk, /* * Send a SYN-ACK after having received an ACK. - * This still operates on a open_request only, not on a big + * This still operates on a request_sock only, not on a big * socket. */ -static int tcp_v4_send_synack(struct sock *sk, struct open_request *req, +static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req, struct dst_entry *dst) { + const struct inet_request_sock *ireq = inet_rsk(req); int err = -1; struct sk_buff * skb; @@ -1310,14 +1306,14 @@ static int tcp_v4_send_synack(struct sock *sk, struct open_request *req, struct tcphdr *th = skb->h.th; th->check = tcp_v4_check(th, skb->len, - req->af.v4_req.loc_addr, - req->af.v4_req.rmt_addr, + ireq->loc_addr, + ireq->rmt_addr, csum_partial((char *)th, skb->len, skb->csum)); - err = ip_build_and_send_pkt(skb, sk, req->af.v4_req.loc_addr, - req->af.v4_req.rmt_addr, - req->af.v4_req.opt); + err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr, + ireq->rmt_addr, + ireq->opt); if (err == NET_XMIT_CN) err = 0; } @@ -1328,12 +1324,12 @@ out: } /* - * IPv4 open_request destructor. + * IPv4 request_sock destructor. */ -static void tcp_v4_or_free(struct open_request *req) +static void tcp_v4_reqsk_destructor(struct request_sock *req) { - if (req->af.v4_req.opt) - kfree(req->af.v4_req.opt); + if (inet_rsk(req)->opt) + kfree(inet_rsk(req)->opt); } static inline void syn_flood_warning(struct sk_buff *skb) @@ -1349,7 +1345,7 @@ static inline void syn_flood_warning(struct sk_buff *skb) } /* - * Save and compile IPv4 options into the open_request if needed. + * Save and compile IPv4 options into the request_sock if needed. */ static inline struct ip_options *tcp_v4_save_options(struct sock *sk, struct sk_buff *skb) @@ -1370,33 +1366,20 @@ static inline struct ip_options *tcp_v4_save_options(struct sock *sk, return dopt; } -/* - * Maximum number of SYN_RECV sockets in queue per LISTEN socket. - * One SYN_RECV socket costs about 80bytes on a 32bit machine. - * It would be better to replace it with a global counter for all sockets - * but then some measure against one socket starving all other sockets - * would be needed. - * - * It was 128 by default. Experiments with real servers show, that - * it is absolutely not enough even at 100conn/sec. 256 cures most - * of problems. This value is adjusted to 128 for very small machines - * (<=32Mb of memory) and to 1024 on normal or better ones (>=256Mb). - * Further increasing requires to change hash table size. - */ -int sysctl_max_syn_backlog = 256; - -struct or_calltable or_ipv4 = { +struct request_sock_ops tcp_request_sock_ops = { .family = PF_INET, + .obj_size = sizeof(struct tcp_request_sock), .rtx_syn_ack = tcp_v4_send_synack, - .send_ack = tcp_v4_or_send_ack, - .destructor = tcp_v4_or_free, + .send_ack = tcp_v4_reqsk_send_ack, + .destructor = tcp_v4_reqsk_destructor, .send_reset = tcp_v4_send_reset, }; int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) { + struct inet_request_sock *ireq; struct tcp_options_received tmp_opt; - struct open_request *req; + struct request_sock *req; __u32 saddr = skb->nh.iph->saddr; __u32 daddr = skb->nh.iph->daddr; __u32 isn = TCP_SKB_CB(skb)->when; @@ -1433,7 +1416,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1) goto drop; - req = tcp_openreq_alloc(); + req = reqsk_alloc(&tcp_request_sock_ops); if (!req) goto drop; @@ -1461,10 +1444,10 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) tcp_openreq_init(req, &tmp_opt, skb); - req->af.v4_req.loc_addr = daddr; - req->af.v4_req.rmt_addr = saddr; - req->af.v4_req.opt = tcp_v4_save_options(sk, skb); - req->class = &or_ipv4; + ireq = inet_rsk(req); + ireq->loc_addr = daddr; + ireq->rmt_addr = saddr; + ireq->opt = tcp_v4_save_options(sk, skb); if (!want_cookie) TCP_ECN_create_request(req, skb->h.th); @@ -1523,20 +1506,20 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) isn = tcp_v4_init_sequence(sk, skb); } - req->snt_isn = isn; + tcp_rsk(req)->snt_isn = isn; if (tcp_v4_send_synack(sk, req, dst)) goto drop_and_free; if (want_cookie) { - tcp_openreq_free(req); + reqsk_free(req); } else { tcp_v4_synq_add(sk, req); } return 0; drop_and_free: - tcp_openreq_free(req); + reqsk_free(req); drop: TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS); return 0; @@ -1548,9 +1531,10 @@ drop: * now create the new socket. */ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst) { + struct inet_request_sock *ireq; struct inet_sock *newinet; struct tcp_sock *newtp; struct sock *newsk; @@ -1570,11 +1554,12 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, newtp = tcp_sk(newsk); newinet = inet_sk(newsk); - newinet->daddr = req->af.v4_req.rmt_addr; - newinet->rcv_saddr = req->af.v4_req.loc_addr; - newinet->saddr = req->af.v4_req.loc_addr; - newinet->opt = req->af.v4_req.opt; - req->af.v4_req.opt = NULL; + ireq = inet_rsk(req); + newinet->daddr = ireq->rmt_addr; + newinet->rcv_saddr = ireq->loc_addr; + newinet->saddr = ireq->loc_addr; + newinet->opt = ireq->opt; + ireq->opt = NULL; newinet->mc_index = tcp_v4_iif(skb); newinet->mc_ttl = skb->nh.iph->ttl; newtp->ext_header_len = 0; @@ -1605,9 +1590,9 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) struct iphdr *iph = skb->nh.iph; struct tcp_sock *tp = tcp_sk(sk); struct sock *nsk; - struct open_request **prev; + struct request_sock **prev; /* Find possible connection requests. */ - struct open_request *req = tcp_v4_search_req(tp, &prev, th->source, + struct request_sock *req = tcp_v4_search_req(tp, &prev, th->source, iph->saddr, iph->daddr); if (req) return tcp_check_req(sk, skb, req, prev); @@ -2063,6 +2048,7 @@ static int tcp_v4_init_sock(struct sock *sk) tp->mss_cache_std = tp->mss_cache = 536; tp->reordering = sysctl_tcp_reordering; + tp->ca_ops = &tcp_init_congestion_ops; sk->sk_state = TCP_CLOSE; @@ -2085,6 +2071,8 @@ int tcp_v4_destroy_sock(struct sock *sk) tcp_clear_xmit_timers(sk); + tcp_cleanup_congestion_control(tp); + /* Cleanup up the write buffer. */ sk_stream_writequeue_purge(sk); @@ -2144,13 +2132,13 @@ static void *listening_get_next(struct seq_file *seq, void *cur) ++st->num; if (st->state == TCP_SEQ_STATE_OPENREQ) { - struct open_request *req = cur; + struct request_sock *req = cur; tp = tcp_sk(st->syn_wait_sk); req = req->dl_next; while (1) { while (req) { - if (req->class->family == st->family) { + if (req->rsk_ops->family == st->family) { cur = req; goto out; } @@ -2159,17 +2147,17 @@ static void *listening_get_next(struct seq_file *seq, void *cur) if (++st->sbucket >= TCP_SYNQ_HSIZE) break; get_req: - req = tp->listen_opt->syn_table[st->sbucket]; + req = tp->accept_queue.listen_opt->syn_table[st->sbucket]; } sk = sk_next(st->syn_wait_sk); st->state = TCP_SEQ_STATE_LISTENING; - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); } else { tp = tcp_sk(sk); - read_lock_bh(&tp->syn_wait_lock); - if (tp->listen_opt && tp->listen_opt->qlen) + read_lock_bh(&tp->accept_queue.syn_wait_lock); + if (reqsk_queue_len(&tp->accept_queue)) goto start_req; - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); sk = sk_next(sk); } get_sk: @@ -2179,8 +2167,8 @@ get_sk: goto out; } tp = tcp_sk(sk); - read_lock_bh(&tp->syn_wait_lock); - if (tp->listen_opt && tp->listen_opt->qlen) { + read_lock_bh(&tp->accept_queue.syn_wait_lock); + if (reqsk_queue_len(&tp->accept_queue)) { start_req: st->uid = sock_i_uid(sk); st->syn_wait_sk = sk; @@ -2188,7 +2176,7 @@ start_req: st->sbucket = 0; goto get_req; } - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); } if (++st->bucket < TCP_LHTABLE_SIZE) { sk = sk_head(&tcp_listening_hash[st->bucket]); @@ -2375,7 +2363,7 @@ static void tcp_seq_stop(struct seq_file *seq, void *v) case TCP_SEQ_STATE_OPENREQ: if (v) { struct tcp_sock *tp = tcp_sk(st->syn_wait_sk); - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); } case TCP_SEQ_STATE_LISTENING: if (v != SEQ_START_TOKEN) @@ -2451,18 +2439,19 @@ void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo) memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops)); } -static void get_openreq4(struct sock *sk, struct open_request *req, +static void get_openreq4(struct sock *sk, struct request_sock *req, char *tmpbuf, int i, int uid) { + const struct inet_request_sock *ireq = inet_rsk(req); int ttd = req->expires - jiffies; sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X" " %02X %08X:%08X %02X:%08lX %08X %5d %8d %u %d %p", i, - req->af.v4_req.loc_addr, + ireq->loc_addr, ntohs(inet_sk(sk)->sport), - req->af.v4_req.rmt_addr, - ntohs(req->rmt_port), + ireq->rmt_addr, + ntohs(ireq->rmt_port), TCP_SYN_RECV, 0, 0, /* could print option size, but that is af dependent. */ 1, /* timers active (only the expire timer) */ @@ -2618,6 +2607,7 @@ struct proto tcp_prot = { .sysctl_rmem = sysctl_tcp_rmem, .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp_sock), + .rsk_prot = &tcp_request_sock_ops, }; @@ -2660,7 +2650,6 @@ EXPORT_SYMBOL(tcp_proc_register); EXPORT_SYMBOL(tcp_proc_unregister); #endif EXPORT_SYMBOL(sysctl_local_port_range); -EXPORT_SYMBOL(sysctl_max_syn_backlog); EXPORT_SYMBOL(sysctl_tcp_low_latency); EXPORT_SYMBOL(sysctl_tcp_tw_reuse); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index eea1a17..f42a284 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -684,7 +684,7 @@ out: * Actually, we could lots of memory writes here. tp of listening * socket contains all necessary default parameters. */ -struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, struct sk_buff *skb) +struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, struct sk_buff *skb) { /* allocate the newsk from the same slab of the master sock, * if not, at sk_free time we'll try to free it from the wrong @@ -692,6 +692,8 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, struct sock *newsk = sk_alloc(PF_INET, GFP_ATOMIC, sk->sk_prot, 0); if(newsk != NULL) { + struct inet_request_sock *ireq = inet_rsk(req); + struct tcp_request_sock *treq = tcp_rsk(req); struct tcp_sock *newtp; struct sk_filter *filter; @@ -703,7 +705,7 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, tcp_sk(newsk)->bind_hash = NULL; /* Clone the TCP header template */ - inet_sk(newsk)->dport = req->rmt_port; + inet_sk(newsk)->dport = ireq->rmt_port; sock_lock_init(newsk); bh_lock_sock(newsk); @@ -739,14 +741,14 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, /* Now setup tcp_sock */ newtp = tcp_sk(newsk); newtp->pred_flags = 0; - newtp->rcv_nxt = req->rcv_isn + 1; - newtp->snd_nxt = req->snt_isn + 1; - newtp->snd_una = req->snt_isn + 1; - newtp->snd_sml = req->snt_isn + 1; + newtp->rcv_nxt = treq->rcv_isn + 1; + newtp->snd_nxt = treq->snt_isn + 1; + newtp->snd_una = treq->snt_isn + 1; + newtp->snd_sml = treq->snt_isn + 1; tcp_prequeue_init(newtp); - tcp_init_wl(newtp, req->snt_isn, req->rcv_isn); + tcp_init_wl(newtp, treq->snt_isn, treq->rcv_isn); newtp->retransmits = 0; newtp->backoff = 0; @@ -772,13 +774,15 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, newtp->frto_counter = 0; newtp->frto_highmark = 0; + newtp->ca_ops = &tcp_reno; + tcp_set_ca_state(newtp, TCP_CA_Open); tcp_init_xmit_timers(newsk); skb_queue_head_init(&newtp->out_of_order_queue); - newtp->rcv_wup = req->rcv_isn + 1; - newtp->write_seq = req->snt_isn + 1; + newtp->rcv_wup = treq->rcv_isn + 1; + newtp->write_seq = treq->snt_isn + 1; newtp->pushed_seq = newtp->write_seq; - newtp->copied_seq = req->rcv_isn + 1; + newtp->copied_seq = treq->rcv_isn + 1; newtp->rx_opt.saw_tstamp = 0; @@ -788,10 +792,8 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, newtp->probes_out = 0; newtp->rx_opt.num_sacks = 0; newtp->urg_data = 0; - newtp->listen_opt = NULL; - newtp->accept_queue = newtp->accept_queue_tail = NULL; - /* Deinitialize syn_wait_lock to trap illegal accesses. */ - memset(&newtp->syn_wait_lock, 0, sizeof(newtp->syn_wait_lock)); + /* Deinitialize accept_queue to trap illegal accesses. */ + memset(&newtp->accept_queue, 0, sizeof(newtp->accept_queue)); /* Back to base struct sock members. */ newsk->sk_err = 0; @@ -808,18 +810,18 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, newsk->sk_socket = NULL; newsk->sk_sleep = NULL; - newtp->rx_opt.tstamp_ok = req->tstamp_ok; - if((newtp->rx_opt.sack_ok = req->sack_ok) != 0) { + newtp->rx_opt.tstamp_ok = ireq->tstamp_ok; + if((newtp->rx_opt.sack_ok = ireq->sack_ok) != 0) { if (sysctl_tcp_fack) newtp->rx_opt.sack_ok |= 2; } newtp->window_clamp = req->window_clamp; newtp->rcv_ssthresh = req->rcv_wnd; newtp->rcv_wnd = req->rcv_wnd; - newtp->rx_opt.wscale_ok = req->wscale_ok; + newtp->rx_opt.wscale_ok = ireq->wscale_ok; if (newtp->rx_opt.wscale_ok) { - newtp->rx_opt.snd_wscale = req->snd_wscale; - newtp->rx_opt.rcv_wscale = req->rcv_wscale; + newtp->rx_opt.snd_wscale = ireq->snd_wscale; + newtp->rx_opt.rcv_wscale = ireq->rcv_wscale; } else { newtp->rx_opt.snd_wscale = newtp->rx_opt.rcv_wscale = 0; newtp->window_clamp = min(newtp->window_clamp, 65535U); @@ -842,8 +844,6 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, if (newtp->ecn_flags&TCP_ECN_OK) sock_set_flag(newsk, SOCK_NO_LARGESEND); - tcp_ca_init(newtp); - TCP_INC_STATS_BH(TCP_MIB_PASSIVEOPENS); } return newsk; @@ -851,12 +851,12 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, /* * Process an incoming packet for SYN_RECV sockets represented - * as an open_request. + * as a request_sock. */ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, - struct open_request *req, - struct open_request **prev) + struct request_sock *req, + struct request_sock **prev) { struct tcphdr *th = skb->h.th; struct tcp_sock *tp = tcp_sk(sk); @@ -881,7 +881,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, } /* Check for pure retransmitted SYN. */ - if (TCP_SKB_CB(skb)->seq == req->rcv_isn && + if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn && flg == TCP_FLAG_SYN && !paws_reject) { /* @@ -901,7 +901,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, * Enforce "SYN-ACK" according to figure 8, figure 6 * of RFC793, fixed by RFC1122. */ - req->class->rtx_syn_ack(sk, req, NULL); + req->rsk_ops->rtx_syn_ack(sk, req, NULL); return NULL; } @@ -959,7 +959,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, * Invalid ACK: reset will be sent by listening socket */ if ((flg & TCP_FLAG_ACK) && - (TCP_SKB_CB(skb)->ack_seq != req->snt_isn+1)) + (TCP_SKB_CB(skb)->ack_seq != tcp_rsk(req)->snt_isn + 1)) return sk; /* Also, it would be not so bad idea to check rcv_tsecr, which @@ -970,10 +970,10 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, /* RFC793: "first check sequence number". */ if (paws_reject || !tcp_in_window(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq, - req->rcv_isn+1, req->rcv_isn+1+req->rcv_wnd)) { + tcp_rsk(req)->rcv_isn + 1, tcp_rsk(req)->rcv_isn + 1 + req->rcv_wnd)) { /* Out of window: send ACK and drop. */ if (!(flg & TCP_FLAG_RST)) - req->class->send_ack(skb, req); + req->rsk_ops->send_ack(skb, req); if (paws_reject) NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED); return NULL; @@ -981,12 +981,12 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, /* In sequence, PAWS is OK. */ - if (tmp_opt.saw_tstamp && !after(TCP_SKB_CB(skb)->seq, req->rcv_isn+1)) + if (tmp_opt.saw_tstamp && !after(TCP_SKB_CB(skb)->seq, tcp_rsk(req)->rcv_isn + 1)) req->ts_recent = tmp_opt.rcv_tsval; - if (TCP_SKB_CB(skb)->seq == req->rcv_isn) { + if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn) { /* Truncate SYN, it is out of window starting - at req->rcv_isn+1. */ + at tcp_rsk(req)->rcv_isn + 1. */ flg &= ~TCP_FLAG_SYN; } @@ -1003,8 +1003,8 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, return NULL; /* If TCP_DEFER_ACCEPT is set, drop bare ACK. */ - if (tp->defer_accept && TCP_SKB_CB(skb)->end_seq == req->rcv_isn+1) { - req->acked = 1; + if (tp->defer_accept && TCP_SKB_CB(skb)->end_seq == tcp_rsk(req)->rcv_isn + 1) { + inet_rsk(req)->acked = 1; return NULL; } @@ -1026,14 +1026,14 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, listen_overflow: if (!sysctl_tcp_abort_on_overflow) { - req->acked = 1; + inet_rsk(req)->acked = 1; return NULL; } embryonic_reset: NET_INC_STATS_BH(LINUX_MIB_EMBRYONICRSTS); if (!(flg & TCP_FLAG_RST)) - req->class->send_reset(skb); + req->rsk_ops->send_reset(skb); tcp_synq_drop(sk, req, prev); return NULL; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index fa24e7a..0e17c24 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -111,8 +111,7 @@ static void tcp_cwnd_restart(struct tcp_sock *tp, struct dst_entry *dst) u32 restart_cwnd = tcp_init_cwnd(tp, dst); u32 cwnd = tp->snd_cwnd; - if (tcp_is_vegas(tp)) - tcp_vegas_enable(tp); + tcp_ca_event(tp, CA_EVENT_CWND_RESTART); tp->snd_ssthresh = tcp_current_ssthresh(tp); restart_cwnd = min(restart_cwnd, cwnd); @@ -280,6 +279,10 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb) #define SYSCTL_FLAG_WSCALE 0x2 #define SYSCTL_FLAG_SACK 0x4 + /* If congestion control is doing timestamping */ + if (tp->ca_ops->rtt_sample) + do_gettimeofday(&skb->stamp); + sysctl_flags = 0; if (tcb->flags & TCPCB_FLAG_SYN) { tcp_header_size = sizeof(struct tcphdr) + TCPOLEN_MSS; @@ -304,17 +307,8 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb) (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK)); } - /* - * If the connection is idle and we are restarting, - * then we don't want to do any Vegas calculations - * until we get fresh RTT samples. So when we - * restart, we reset our Vegas state to a clean - * slate. After we get acks for this flight of - * packets, _then_ we can make Vegas calculations - * again. - */ - if (tcp_is_vegas(tp) && tcp_packets_in_flight(tp) == 0) - tcp_vegas_enable(tp); + if (tcp_packets_in_flight(tp) == 0) + tcp_ca_event(tp, CA_EVENT_TX_START); th = (struct tcphdr *) skb_push(skb, tcp_header_size); skb->h.th = th; @@ -521,6 +515,7 @@ static int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len) * skbs, which it never sent before. --ANK */ TCP_SKB_CB(buff)->when = TCP_SKB_CB(skb)->when; + buff->stamp = skb->stamp; if (TCP_SKB_CB(skb)->sacked & TCPCB_LOST) { tp->lost_out -= tcp_skb_pcount(skb); @@ -1356,8 +1351,9 @@ int tcp_send_synack(struct sock *sk) * Prepare a SYN-ACK. */ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, - struct open_request *req) + struct request_sock *req) { + struct inet_request_sock *ireq = inet_rsk(req); struct tcp_sock *tp = tcp_sk(sk); struct tcphdr *th; int tcp_header_size; @@ -1373,47 +1369,47 @@ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, skb->dst = dst_clone(dst); tcp_header_size = (sizeof(struct tcphdr) + TCPOLEN_MSS + - (req->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0) + - (req->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) + + (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0) + + (ireq->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) + /* SACK_PERM is in the place of NOP NOP of TS */ - ((req->sack_ok && !req->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0)); + ((ireq->sack_ok && !ireq->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0)); skb->h.th = th = (struct tcphdr *) skb_push(skb, tcp_header_size); memset(th, 0, sizeof(struct tcphdr)); th->syn = 1; th->ack = 1; if (dst->dev->features&NETIF_F_TSO) - req->ecn_ok = 0; + ireq->ecn_ok = 0; TCP_ECN_make_synack(req, th); th->source = inet_sk(sk)->sport; - th->dest = req->rmt_port; - TCP_SKB_CB(skb)->seq = req->snt_isn; + th->dest = ireq->rmt_port; + TCP_SKB_CB(skb)->seq = tcp_rsk(req)->snt_isn; TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq + 1; TCP_SKB_CB(skb)->sacked = 0; skb_shinfo(skb)->tso_segs = 1; skb_shinfo(skb)->tso_size = 0; th->seq = htonl(TCP_SKB_CB(skb)->seq); - th->ack_seq = htonl(req->rcv_isn + 1); + th->ack_seq = htonl(tcp_rsk(req)->rcv_isn + 1); if (req->rcv_wnd == 0) { /* ignored for retransmitted syns */ __u8 rcv_wscale; /* Set this up on the first call only */ req->window_clamp = tp->window_clamp ? : dst_metric(dst, RTAX_WINDOW); /* tcp_full_space because it is guaranteed to be the first packet */ tcp_select_initial_window(tcp_full_space(sk), - dst_metric(dst, RTAX_ADVMSS) - (req->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), + dst_metric(dst, RTAX_ADVMSS) - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), &req->rcv_wnd, &req->window_clamp, - req->wscale_ok, + ireq->wscale_ok, &rcv_wscale); - req->rcv_wscale = rcv_wscale; + ireq->rcv_wscale = rcv_wscale; } /* RFC1323: The window in SYN & SYN/ACK segments is never scaled. */ th->window = htons(req->rcv_wnd); TCP_SKB_CB(skb)->when = tcp_time_stamp; - tcp_syn_build_options((__u32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), req->tstamp_ok, - req->sack_ok, req->wscale_ok, req->rcv_wscale, + tcp_syn_build_options((__u32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok, + ireq->sack_ok, ireq->wscale_ok, ireq->rcv_wscale, TCP_SKB_CB(skb)->when, req->ts_recent); @@ -1448,7 +1444,6 @@ static inline void tcp_connect_init(struct sock *sk) tp->window_clamp = dst_metric(dst, RTAX_WINDOW); tp->advmss = dst_metric(dst, RTAX_ADVMSS); tcp_initialize_rcv_mss(sk); - tcp_ca_init(tp); tcp_select_initial_window(tcp_full_space(sk), tp->advmss - (tp->rx_opt.ts_recent_stamp ? tp->tcp_header_len - sizeof(struct tcphdr) : 0), @@ -1502,7 +1497,6 @@ int tcp_connect(struct sock *sk) TCP_SKB_CB(buff)->end_seq = tp->write_seq; tp->snd_nxt = tp->write_seq; tp->pushed_seq = tp->write_seq; - tcp_ca_init(tp); /* Send it off. */ TCP_SKB_CB(buff)->when = tcp_time_stamp; diff --git a/net/ipv4/tcp_scalable.c b/net/ipv4/tcp_scalable.c new file mode 100644 index 0000000..70e108e --- /dev/null +++ b/net/ipv4/tcp_scalable.c @@ -0,0 +1,68 @@ +/* Tom Kelly's Scalable TCP + * + * See htt://www-lce.eng.cam.ac.uk/~ctk21/scalable/ + * + * John Heffner <jheffner@sc.edu> + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <net/tcp.h> + +/* These factors derived from the recommended values in the aer: + * .01 and and 7/8. We use 50 instead of 100 to account for + * delayed ack. + */ +#define TCP_SCALABLE_AI_CNT 50U +#define TCP_SCALABLE_MD_SCALE 3 + +static void tcp_scalable_cong_avoid(struct tcp_sock *tp, u32 ack, u32 rtt, + u32 in_flight, int flag) +{ + if (in_flight < tp->snd_cwnd) + return; + + if (tp->snd_cwnd <= tp->snd_ssthresh) { + tp->snd_cwnd++; + } else { + tp->snd_cwnd_cnt++; + if (tp->snd_cwnd_cnt > min(tp->snd_cwnd, TCP_SCALABLE_AI_CNT)){ + tp->snd_cwnd++; + tp->snd_cwnd_cnt = 0; + } + } + tp->snd_cwnd = min_t(u32, tp->snd_cwnd, tp->snd_cwnd_clamp); + tp->snd_cwnd_stamp = tcp_time_stamp; +} + +static u32 tcp_scalable_ssthresh(struct tcp_sock *tp) +{ + return max(tp->snd_cwnd - (tp->snd_cwnd>>TCP_SCALABLE_MD_SCALE), 2U); +} + + +static struct tcp_congestion_ops tcp_scalable = { + .ssthresh = tcp_scalable_ssthresh, + .cong_avoid = tcp_scalable_cong_avoid, + .min_cwnd = tcp_reno_min_cwnd, + + .owner = THIS_MODULE, + .name = "scalable", +}; + +static int __init tcp_scalable_register(void) +{ + return tcp_register_congestion_control(&tcp_scalable); +} + +static void __exit tcp_scalable_unregister(void) +{ + tcp_unregister_congestion_control(&tcp_scalable); +} + +module_init(tcp_scalable_register); +module_exit(tcp_scalable_unregister); + +MODULE_AUTHOR("John Heffner"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("Scalable TCP"); diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 799ebe0..b127b44 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -464,11 +464,11 @@ out_unlock: static void tcp_synack_timer(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; + struct listen_sock *lopt = tp->accept_queue.listen_opt; int max_retries = tp->syn_retries ? : sysctl_tcp_synack_retries; int thresh = max_retries; unsigned long now = jiffies; - struct open_request **reqp, *req; + struct request_sock **reqp, *req; int i, budget; if (lopt == NULL || lopt->qlen == 0) @@ -513,8 +513,8 @@ static void tcp_synack_timer(struct sock *sk) while ((req = *reqp) != NULL) { if (time_after_eq(now, req->expires)) { if ((req->retrans < thresh || - (req->acked && req->retrans < max_retries)) - && !req->class->rtx_syn_ack(sk, req, NULL)) { + (inet_rsk(req)->acked && req->retrans < max_retries)) + && !req->rsk_ops->rtx_syn_ack(sk, req, NULL)) { unsigned long timeo; if (req->retrans++ == 0) @@ -527,13 +527,9 @@ static void tcp_synack_timer(struct sock *sk) } /* Drop this request */ - write_lock(&tp->syn_wait_lock); - *reqp = req->dl_next; - write_unlock(&tp->syn_wait_lock); - lopt->qlen--; - if (req->retrans == 0) - lopt->qlen_young--; - tcp_openreq_free(req); + tcp_synq_unlink(tp, req, reqp); + reqsk_queue_removed(&tp->accept_queue, req); + reqsk_free(req); continue; } reqp = &req->dl_next; diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c new file mode 100644 index 0000000..9bd443d --- /dev/null +++ b/net/ipv4/tcp_vegas.c @@ -0,0 +1,411 @@ +/* + * TCP Vegas congestion control + * + * This is based on the congestion detection/avoidance scheme described in + * Lawrence S. Brakmo and Larry L. Peterson. + * "TCP Vegas: End to end congestion avoidance on a global internet." + * IEEE Journal on Selected Areas in Communication, 13(8):1465--1480, + * October 1995. Available from: + * ftp://ftp.cs.arizona.edu/xkernel/Papers/jsac.ps + * + * See http://www.cs.arizona.edu/xkernel/ for their implementation. + * The main aspects that distinguish this implementation from the + * Arizona Vegas implementation are: + * o We do not change the loss detection or recovery mechanisms of + * Linux in any way. Linux already recovers from losses quite well, + * using fine-grained timers, NewReno, and FACK. + * o To avoid the performance penalty imposed by increasing cwnd + * only every-other RTT during slow start, we increase during + * every RTT during slow start, just like Reno. + * o Largely to allow continuous cwnd growth during slow start, + * we use the rate at which ACKs come back as the "actual" + * rate, rather than the rate at which data is sent. + * o To speed convergence to the right rate, we set the cwnd + * to achieve the right ("actual") rate when we exit slow start. + * o To filter out the noise caused by delayed ACKs, we use the + * minimum RTT sample observed during the last RTT to calculate + * the actual rate. + * o When the sender re-starts from idle, it waits until it has + * received ACKs for an entire flight of new data before making + * a cwnd adjustment decision. The original Vegas implementation + * assumed senders never went idle. + */ + +#include <linux/config.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/tcp_diag.h> + +#include <net/tcp.h> + +/* Default values of the Vegas variables, in fixed-point representation + * with V_PARAM_SHIFT bits to the right of the binary point. + */ +#define V_PARAM_SHIFT 1 +static int alpha = 1<<V_PARAM_SHIFT; +static int beta = 3<<V_PARAM_SHIFT; +static int gamma = 1<<V_PARAM_SHIFT; + +module_param(alpha, int, 0644); +MODULE_PARM_DESC(alpha, "lower bound of packets in network (scale by 2)"); +module_param(beta, int, 0644); +MODULE_PARM_DESC(beta, "upper bound of packets in network (scale by 2)"); +module_param(gamma, int, 0644); +MODULE_PARM_DESC(gamma, "limit on increase (scale by 2)"); + + +/* Vegas variables */ +struct vegas { + u32 beg_snd_nxt; /* right edge during last RTT */ + u32 beg_snd_una; /* left edge during last RTT */ + u32 beg_snd_cwnd; /* saves the size of the cwnd */ + u8 doing_vegas_now;/* if true, do vegas for this RTT */ + u16 cntRTT; /* # of RTTs measured within last RTT */ + u32 minRTT; /* min of RTTs measured within last RTT (in usec) */ + u32 baseRTT; /* the min of all Vegas RTT measurements seen (in usec) */ +}; + +/* There are several situations when we must "re-start" Vegas: + * + * o when a connection is established + * o after an RTO + * o after fast recovery + * o when we send a packet and there is no outstanding + * unacknowledged data (restarting an idle connection) + * + * In these circumstances we cannot do a Vegas calculation at the + * end of the first RTT, because any calculation we do is using + * stale info -- both the saved cwnd and congestion feedback are + * stale. + * + * Instead we must wait until the completion of an RTT during + * which we actually receive ACKs. + */ +static inline void vegas_enable(struct tcp_sock *tp) +{ + struct vegas *vegas = tcp_ca(tp); + + /* Begin taking Vegas samples next time we send something. */ + vegas->doing_vegas_now = 1; + + /* Set the beginning of the next send window. */ + vegas->beg_snd_nxt = tp->snd_nxt; + + vegas->cntRTT = 0; + vegas->minRTT = 0x7fffffff; +} + +/* Stop taking Vegas samples for now. */ +static inline void vegas_disable(struct tcp_sock *tp) +{ + struct vegas *vegas = tcp_ca(tp); + + vegas->doing_vegas_now = 0; +} + +static void tcp_vegas_init(struct tcp_sock *tp) +{ + struct vegas *vegas = tcp_ca(tp); + + vegas->baseRTT = 0x7fffffff; + vegas_enable(tp); +} + +/* Do RTT sampling needed for Vegas. + * Basically we: + * o min-filter RTT samples from within an RTT to get the current + * propagation delay + queuing delay (we are min-filtering to try to + * avoid the effects of delayed ACKs) + * o min-filter RTT samples from a much longer window (forever for now) + * to find the propagation delay (baseRTT) + */ +static void tcp_vegas_rtt_calc(struct tcp_sock *tp, u32 usrtt) +{ + struct vegas *vegas = tcp_ca(tp); + u32 vrtt = usrtt + 1; /* Never allow zero rtt or baseRTT */ + + /* Filter to find propagation delay: */ + if (vrtt < vegas->baseRTT) + vegas->baseRTT = vrtt; + + /* Find the min RTT during the last RTT to find + * the current prop. delay + queuing delay: + */ + vegas->minRTT = min(vegas->minRTT, vrtt); + vegas->cntRTT++; +} + +static void tcp_vegas_state(struct tcp_sock *tp, u8 ca_state) +{ + + if (ca_state == TCP_CA_Open) + vegas_enable(tp); + else + vegas_disable(tp); +} + +/* + * If the connection is idle and we are restarting, + * then we don't want to do any Vegas calculations + * until we get fresh RTT samples. So when we + * restart, we reset our Vegas state to a clean + * slate. After we get acks for this flight of + * packets, _then_ we can make Vegas calculations + * again. + */ +static void tcp_vegas_cwnd_event(struct tcp_sock *tp, enum tcp_ca_event event) +{ + if (event == CA_EVENT_CWND_RESTART || + event == CA_EVENT_TX_START) + tcp_vegas_init(tp); +} + +static void tcp_vegas_cong_avoid(struct tcp_sock *tp, u32 ack, + u32 seq_rtt, u32 in_flight, int flag) +{ + struct vegas *vegas = tcp_ca(tp); + + if (!vegas->doing_vegas_now) + return tcp_reno_cong_avoid(tp, ack, seq_rtt, in_flight, flag); + + /* The key players are v_beg_snd_una and v_beg_snd_nxt. + * + * These are so named because they represent the approximate values + * of snd_una and snd_nxt at the beginning of the current RTT. More + * precisely, they represent the amount of data sent during the RTT. + * At the end of the RTT, when we receive an ACK for v_beg_snd_nxt, + * we will calculate that (v_beg_snd_nxt - v_beg_snd_una) outstanding + * bytes of data have been ACKed during the course of the RTT, giving + * an "actual" rate of: + * + * (v_beg_snd_nxt - v_beg_snd_una) / (rtt duration) + * + * Unfortunately, v_beg_snd_una is not exactly equal to snd_una, + * because delayed ACKs can cover more than one segment, so they + * don't line up nicely with the boundaries of RTTs. + * + * Another unfortunate fact of life is that delayed ACKs delay the + * advance of the left edge of our send window, so that the number + * of bytes we send in an RTT is often less than our cwnd will allow. + * So we keep track of our cwnd separately, in v_beg_snd_cwnd. + */ + + if (after(ack, vegas->beg_snd_nxt)) { + /* Do the Vegas once-per-RTT cwnd adjustment. */ + u32 old_wnd, old_snd_cwnd; + + + /* Here old_wnd is essentially the window of data that was + * sent during the previous RTT, and has all + * been acknowledged in the course of the RTT that ended + * with the ACK we just received. Likewise, old_snd_cwnd + * is the cwnd during the previous RTT. + */ + old_wnd = (vegas->beg_snd_nxt - vegas->beg_snd_una) / + tp->mss_cache; + old_snd_cwnd = vegas->beg_snd_cwnd; + + /* Save the extent of the current window so we can use this + * at the end of the next RTT. + */ + vegas->beg_snd_una = vegas->beg_snd_nxt; + vegas->beg_snd_nxt = tp->snd_nxt; + vegas->beg_snd_cwnd = tp->snd_cwnd; + + /* Take into account the current RTT sample too, to + * decrease the impact of delayed acks. This double counts + * this sample since we count it for the next window as well, + * but that's not too awful, since we're taking the min, + * rather than averaging. + */ + tcp_vegas_rtt_calc(tp, seq_rtt*1000); + + /* We do the Vegas calculations only if we got enough RTT + * samples that we can be reasonably sure that we got + * at least one RTT sample that wasn't from a delayed ACK. + * If we only had 2 samples total, + * then that means we're getting only 1 ACK per RTT, which + * means they're almost certainly delayed ACKs. + * If we have 3 samples, we should be OK. + */ + + if (vegas->cntRTT <= 2) { + /* We don't have enough RTT samples to do the Vegas + * calculation, so we'll behave like Reno. + */ + if (tp->snd_cwnd > tp->snd_ssthresh) + tp->snd_cwnd++; + } else { + u32 rtt, target_cwnd, diff; + + /* We have enough RTT samples, so, using the Vegas + * algorithm, we determine if we should increase or + * decrease cwnd, and by how much. + */ + + /* Pluck out the RTT we are using for the Vegas + * calculations. This is the min RTT seen during the + * last RTT. Taking the min filters out the effects + * of delayed ACKs, at the cost of noticing congestion + * a bit later. + */ + rtt = vegas->minRTT; + + /* Calculate the cwnd we should have, if we weren't + * going too fast. + * + * This is: + * (actual rate in segments) * baseRTT + * We keep it as a fixed point number with + * V_PARAM_SHIFT bits to the right of the binary point. + */ + target_cwnd = ((old_wnd * vegas->baseRTT) + << V_PARAM_SHIFT) / rtt; + + /* Calculate the difference between the window we had, + * and the window we would like to have. This quantity + * is the "Diff" from the Arizona Vegas papers. + * + * Again, this is a fixed point number with + * V_PARAM_SHIFT bits to the right of the binary + * point. + */ + diff = (old_wnd << V_PARAM_SHIFT) - target_cwnd; + + if (tp->snd_cwnd < tp->snd_ssthresh) { + /* Slow start. */ + if (diff > gamma) { + /* Going too fast. Time to slow down + * and switch to congestion avoidance. + */ + tp->snd_ssthresh = 2; + + /* Set cwnd to match the actual rate + * exactly: + * cwnd = (actual rate) * baseRTT + * Then we add 1 because the integer + * truncation robs us of full link + * utilization. + */ + tp->snd_cwnd = min(tp->snd_cwnd, + (target_cwnd >> + V_PARAM_SHIFT)+1); + + } + } else { + /* Congestion avoidance. */ + u32 next_snd_cwnd; + + /* Figure out where we would like cwnd + * to be. + */ + if (diff > beta) { + /* The old window was too fast, so + * we slow down. + */ + next_snd_cwnd = old_snd_cwnd - 1; + } else if (diff < alpha) { + /* We don't have enough extra packets + * in the network, so speed up. + */ + next_snd_cwnd = old_snd_cwnd + 1; + } else { + /* Sending just as fast as we + * should be. + */ + next_snd_cwnd = old_snd_cwnd; + } + + /* Adjust cwnd upward or downward, toward the + * desired value. + */ + if (next_snd_cwnd > tp->snd_cwnd) + tp->snd_cwnd++; + else if (next_snd_cwnd < tp->snd_cwnd) + tp->snd_cwnd--; + } + } + + /* Wipe the slate clean for the next RTT. */ + vegas->cntRTT = 0; + vegas->minRTT = 0x7fffffff; + } + + /* The following code is executed for every ack we receive, + * except for conditions checked in should_advance_cwnd() + * before the call to tcp_cong_avoid(). Mainly this means that + * we only execute this code if the ack actually acked some + * data. + */ + + /* If we are in slow start, increase our cwnd in response to this ACK. + * (If we are not in slow start then we are in congestion avoidance, + * and adjust our congestion window only once per RTT. See the code + * above.) + */ + if (tp->snd_cwnd <= tp->snd_ssthresh) + tp->snd_cwnd++; + + /* to keep cwnd from growing without bound */ + tp->snd_cwnd = min_t(u32, tp->snd_cwnd, tp->snd_cwnd_clamp); + + /* Make sure that we are never so timid as to reduce our cwnd below + * 2 MSS. + * + * Going below 2 MSS would risk huge delayed ACKs from our receiver. + */ + tp->snd_cwnd = max(tp->snd_cwnd, 2U); +} + +/* Extract info for Tcp socket info provided via netlink. */ +static void tcp_vegas_get_info(struct tcp_sock *tp, u32 ext, + struct sk_buff *skb) +{ + const struct vegas *ca = tcp_ca(tp); + if (ext & (1<<(TCPDIAG_VEGASINFO-1))) { + struct tcpvegas_info *info; + + info = RTA_DATA(__RTA_PUT(skb, TCPDIAG_VEGASINFO, + sizeof(*info))); + + info->tcpv_enabled = ca->doing_vegas_now; + info->tcpv_rttcnt = ca->cntRTT; + info->tcpv_rtt = ca->baseRTT; + info->tcpv_minrtt = ca->minRTT; + rtattr_failure: ; + } +} + +static struct tcp_congestion_ops tcp_vegas = { + .init = tcp_vegas_init, + .ssthresh = tcp_reno_ssthresh, + .cong_avoid = tcp_vegas_cong_avoid, + .min_cwnd = tcp_reno_min_cwnd, + .rtt_sample = tcp_vegas_rtt_calc, + .set_state = tcp_vegas_state, + .cwnd_event = tcp_vegas_cwnd_event, + .get_info = tcp_vegas_get_info, + + .owner = THIS_MODULE, + .name = "vegas", +}; + +static int __init tcp_vegas_register(void) +{ + BUG_ON(sizeof(struct vegas) > TCP_CA_PRIV_SIZE); + tcp_register_congestion_control(&tcp_vegas); + return 0; +} + +static void __exit tcp_vegas_unregister(void) +{ + tcp_unregister_congestion_control(&tcp_vegas); +} + +module_init(tcp_vegas_register); +module_exit(tcp_vegas_unregister); + +MODULE_AUTHOR("Stephen Hemminger"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("TCP Vegas"); diff --git a/net/ipv4/tcp_westwood.c b/net/ipv4/tcp_westwood.c new file mode 100644 index 0000000..ef82724 --- /dev/null +++ b/net/ipv4/tcp_westwood.c @@ -0,0 +1,259 @@ +/* + * TCP Westwood+ + * + * Angelo Dell'Aera: TCP Westwood+ support + */ + +#include <linux/config.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/tcp_diag.h> +#include <net/tcp.h> + +/* TCP Westwood structure */ +struct westwood { + u32 bw_ns_est; /* first bandwidth estimation..not too smoothed 8) */ + u32 bw_est; /* bandwidth estimate */ + u32 rtt_win_sx; /* here starts a new evaluation... */ + u32 bk; + u32 snd_una; /* used for evaluating the number of acked bytes */ + u32 cumul_ack; + u32 accounted; + u32 rtt; + u32 rtt_min; /* minimum observed RTT */ +}; + + +/* TCP Westwood functions and constants */ +#define TCP_WESTWOOD_RTT_MIN (HZ/20) /* 50ms */ +#define TCP_WESTWOOD_INIT_RTT (20*HZ) /* maybe too conservative?! */ + +/* + * @tcp_westwood_create + * This function initializes fields used in TCP Westwood+, + * it is called after the initial SYN, so the sequence numbers + * are correct but new passive connections we have no + * information about RTTmin at this time so we simply set it to + * TCP_WESTWOOD_INIT_RTT. This value was chosen to be too conservative + * since in this way we're sure it will be updated in a consistent + * way as soon as possible. It will reasonably happen within the first + * RTT period of the connection lifetime. + */ +static void tcp_westwood_init(struct tcp_sock *tp) +{ + struct westwood *w = tcp_ca(tp); + + w->bk = 0; + w->bw_ns_est = 0; + w->bw_est = 0; + w->accounted = 0; + w->cumul_ack = 0; + w->rtt_min = w->rtt = TCP_WESTWOOD_INIT_RTT; + w->rtt_win_sx = tcp_time_stamp; + w->snd_una = tp->snd_una; +} + +/* + * @westwood_do_filter + * Low-pass filter. Implemented using constant coefficients. + */ +static inline u32 westwood_do_filter(u32 a, u32 b) +{ + return (((7 * a) + b) >> 3); +} + +static inline void westwood_filter(struct westwood *w, u32 delta) +{ + w->bw_ns_est = westwood_do_filter(w->bw_ns_est, w->bk / delta); + w->bw_est = westwood_do_filter(w->bw_est, w->bw_ns_est); +} + +/* + * @westwood_pkts_acked + * Called after processing group of packets. + * but all westwood needs is the last sample of srtt. + */ +static void tcp_westwood_pkts_acked(struct tcp_sock *tp, u32 cnt) +{ + struct westwood *w = tcp_ca(tp); + if (cnt > 0) + w->rtt = tp->srtt >> 3; +} + +/* + * @westwood_update_window + * It updates RTT evaluation window if it is the right moment to do + * it. If so it calls filter for evaluating bandwidth. + */ +static void westwood_update_window(struct tcp_sock *tp) +{ + struct westwood *w = tcp_ca(tp); + s32 delta = tcp_time_stamp - w->rtt_win_sx; + + /* + * See if a RTT-window has passed. + * Be careful since if RTT is less than + * 50ms we don't filter but we continue 'building the sample'. + * This minimum limit was chosen since an estimation on small + * time intervals is better to avoid... + * Obviously on a LAN we reasonably will always have + * right_bound = left_bound + WESTWOOD_RTT_MIN + */ + if (w->rtt && delta > max_t(u32, w->rtt, TCP_WESTWOOD_RTT_MIN)) { + westwood_filter(w, delta); + + w->bk = 0; + w->rtt_win_sx = tcp_time_stamp; + } +} + +/* + * @westwood_fast_bw + * It is called when we are in fast path. In particular it is called when + * header prediction is successful. In such case in fact update is + * straight forward and doesn't need any particular care. + */ +static inline void westwood_fast_bw(struct tcp_sock *tp) +{ + struct westwood *w = tcp_ca(tp); + + westwood_update_window(tp); + + w->bk += tp->snd_una - w->snd_una; + w->snd_una = tp->snd_una; + w->rtt_min = min(w->rtt, w->rtt_min); +} + +/* + * @westwood_acked_count + * This function evaluates cumul_ack for evaluating bk in case of + * delayed or partial acks. + */ +static inline u32 westwood_acked_count(struct tcp_sock *tp) +{ + struct westwood *w = tcp_ca(tp); + + w->cumul_ack = tp->snd_una - w->snd_una; + + /* If cumul_ack is 0 this is a dupack since it's not moving + * tp->snd_una. + */ + if (!w->cumul_ack) { + w->accounted += tp->mss_cache; + w->cumul_ack = tp->mss_cache; + } + + if (w->cumul_ack > tp->mss_cache) { + /* Partial or delayed ack */ + if (w->accounted >= w->cumul_ack) { + w->accounted -= w->cumul_ack; + w->cumul_ack = tp->mss_cache; + } else { + w->cumul_ack -= w->accounted; + w->accounted = 0; + } + } + + w->snd_una = tp->snd_una; + + return w->cumul_ack; +} + +static inline u32 westwood_bw_rttmin(const struct tcp_sock *tp) +{ + struct westwood *w = tcp_ca(tp); + return max_t(u32, (w->bw_est * w->rtt_min) / tp->mss_cache, 2); +} + +/* + * TCP Westwood + * Here limit is evaluated as Bw estimation*RTTmin (for obtaining it + * in packets we use mss_cache). Rttmin is guaranteed to be >= 2 + * so avoids ever returning 0. + */ +static u32 tcp_westwood_cwnd_min(struct tcp_sock *tp) +{ + return westwood_bw_rttmin(tp); +} + +static void tcp_westwood_event(struct tcp_sock *tp, enum tcp_ca_event event) +{ + struct westwood *w = tcp_ca(tp); + + switch(event) { + case CA_EVENT_FAST_ACK: + westwood_fast_bw(tp); + break; + + case CA_EVENT_COMPLETE_CWR: + tp->snd_cwnd = tp->snd_ssthresh = westwood_bw_rttmin(tp); + break; + + case CA_EVENT_FRTO: + tp->snd_ssthresh = westwood_bw_rttmin(tp); + break; + + case CA_EVENT_SLOW_ACK: + westwood_update_window(tp); + w->bk += westwood_acked_count(tp); + w->rtt_min = min(w->rtt, w->rtt_min); + break; + + default: + /* don't care */ + break; + } +} + + +/* Extract info for Tcp socket info provided via netlink. */ +static void tcp_westwood_info(struct tcp_sock *tp, u32 ext, + struct sk_buff *skb) +{ + const struct westwood *ca = tcp_ca(tp); + if (ext & (1<<(TCPDIAG_VEGASINFO-1))) { + struct rtattr *rta; + struct tcpvegas_info *info; + + rta = __RTA_PUT(skb, TCPDIAG_VEGASINFO, sizeof(*info)); + info = RTA_DATA(rta); + info->tcpv_enabled = 1; + info->tcpv_rttcnt = 0; + info->tcpv_rtt = jiffies_to_usecs(ca->rtt); + info->tcpv_minrtt = jiffies_to_usecs(ca->rtt_min); + rtattr_failure: ; + } +} + + +static struct tcp_congestion_ops tcp_westwood = { + .init = tcp_westwood_init, + .ssthresh = tcp_reno_ssthresh, + .cong_avoid = tcp_reno_cong_avoid, + .min_cwnd = tcp_westwood_cwnd_min, + .cwnd_event = tcp_westwood_event, + .get_info = tcp_westwood_info, + .pkts_acked = tcp_westwood_pkts_acked, + + .owner = THIS_MODULE, + .name = "westwood" +}; + +static int __init tcp_westwood_register(void) +{ + BUG_ON(sizeof(struct westwood) > TCP_CA_PRIV_SIZE); + return tcp_register_congestion_control(&tcp_westwood); +} + +static void __exit tcp_westwood_unregister(void) +{ + tcp_unregister_congestion_control(&tcp_westwood); +} + +module_init(tcp_westwood_register); +module_exit(tcp_westwood_unregister); + +MODULE_AUTHOR("Stephen Hemminger, Angelo Dell'Aera"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("TCP Westwood+"); diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 4a6952e..7c24e64 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -738,7 +738,7 @@ int udp_ioctl(struct sock *sk, int cmd, unsigned long arg) unsigned long amount; amount = 0; - spin_lock_irq(&sk->sk_receive_queue.lock); + spin_lock_bh(&sk->sk_receive_queue.lock); skb = skb_peek(&sk->sk_receive_queue); if (skb != NULL) { /* @@ -748,7 +748,7 @@ int udp_ioctl(struct sock *sk, int cmd, unsigned long arg) */ amount = skb->len - sizeof(struct udphdr); } - spin_unlock_irq(&sk->sk_receive_queue.lock); + spin_unlock_bh(&sk->sk_receive_queue.lock); return put_user(amount, (int __user *)arg); } @@ -848,12 +848,12 @@ csum_copy_err: /* Clear queue. */ if (flags&MSG_PEEK) { int clear = 0; - spin_lock_irq(&sk->sk_receive_queue.lock); + spin_lock_bh(&sk->sk_receive_queue.lock); if (skb == skb_peek(&sk->sk_receive_queue)) { __skb_unlink(skb, &sk->sk_receive_queue); clear = 1; } - spin_unlock_irq(&sk->sk_receive_queue.lock); + spin_unlock_bh(&sk->sk_receive_queue.lock); if (clear) kfree_skb(skb); } @@ -1334,7 +1334,7 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait) struct sk_buff_head *rcvq = &sk->sk_receive_queue; struct sk_buff *skb; - spin_lock_irq(&rcvq->lock); + spin_lock_bh(&rcvq->lock); while ((skb = skb_peek(rcvq)) != NULL) { if (udp_checksum_complete(skb)) { UDP_INC_STATS_BH(UDP_MIB_INERRORS); @@ -1345,7 +1345,7 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait) break; } } - spin_unlock_irq(&rcvq->lock); + spin_unlock_bh(&rcvq->lock); /* nothing to see, move along */ if (skb == NULL) 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..e1fe360e 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 7744a25..a54d4ef 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -57,6 +57,7 @@ #endif #include <linux/delay.h> #include <linux/notifier.h> +#include <linux/string.h> #include <net/sock.h> #include <net/snmp.h> @@ -131,7 +132,7 @@ static void addrconf_leave_anycast(struct inet6_ifaddr *ifp); static int addrconf_ifdown(struct net_device *dev, int how); -static void addrconf_dad_start(struct inet6_ifaddr *ifp, int flags); +static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags); static void addrconf_dad_timer(unsigned long data); static void addrconf_dad_completed(struct inet6_ifaddr *ifp); static void addrconf_rs_timer(unsigned long data); @@ -372,6 +373,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev) ndev->regen_timer.data = (unsigned long) ndev; if ((dev->flags&IFF_LOOPBACK) || dev->type == ARPHRD_TUNNEL || + dev->type == ARPHRD_NONE || dev->type == ARPHRD_SIT) { printk(KERN_INFO "Disabled Privacy Extensions on device %p(%s)\n", @@ -491,7 +493,7 @@ void inet6_ifa_finish_destroy(struct inet6_ifaddr *ifp) static struct inet6_ifaddr * ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, int pfxlen, - int scope, unsigned flags) + int scope, u32 flags) { struct inet6_ifaddr *ifa = NULL; struct rt6_info *rt; @@ -694,7 +696,7 @@ 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; @@ -1319,7 +1321,7 @@ static int __ipv6_try_regen_rndid(struct inet6_dev *idev, struct in6_addr *tmpad static void addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev, - unsigned long expires, unsigned flags) + unsigned long expires, u32 flags) { struct in6_rtmsg rtmsg; @@ -1339,7 +1341,7 @@ 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 */ @@ -1356,7 +1358,7 @@ 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) @@ -1373,7 +1375,7 @@ 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) @@ -1466,7 +1468,7 @@ 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; @@ -2228,7 +2230,7 @@ out: /* * Duplicate Address Detection */ -static void addrconf_dad_start(struct inet6_ifaddr *ifp, int flags) +static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags) { struct inet6_dev *idev = ifp->idev; struct net_device *dev = idev->dev; @@ -2621,15 +2623,14 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) } static int inet6_fill_ifaddr(struct sk_buff *skb, struct inet6_ifaddr *ifa, - u32 pid, u32 seq, int event) + u32 pid, u32 seq, int event, unsigned int flags) { struct ifaddrmsg *ifm; struct nlmsghdr *nlh; struct ifa_cacheinfo ci; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*ifm)); - if (pid) nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*ifm), flags); ifm = NLMSG_DATA(nlh); ifm->ifa_family = AF_INET6; ifm->ifa_prefixlen = ifa->prefix_len; @@ -2671,15 +2672,14 @@ rtattr_failure: } static int inet6_fill_ifmcaddr(struct sk_buff *skb, struct ifmcaddr6 *ifmca, - u32 pid, u32 seq, int event) + u32 pid, u32 seq, int event, u16 flags) { struct ifaddrmsg *ifm; struct nlmsghdr *nlh; struct ifa_cacheinfo ci; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*ifm)); - if (pid) nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*ifm), flags); ifm = NLMSG_DATA(nlh); ifm->ifa_family = AF_INET6; ifm->ifa_prefixlen = 128; @@ -2708,15 +2708,14 @@ rtattr_failure: } static int inet6_fill_ifacaddr(struct sk_buff *skb, struct ifacaddr6 *ifaca, - u32 pid, u32 seq, int event) + u32 pid, u32 seq, int event, unsigned int flags) { struct ifaddrmsg *ifm; struct nlmsghdr *nlh; struct ifa_cacheinfo ci; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*ifm)); - if (pid) nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*ifm), flags); ifm = NLMSG_DATA(nlh); ifm->ifa_family = AF_INET6; ifm->ifa_prefixlen = 128; @@ -2785,7 +2784,8 @@ static int inet6_dump_addr(struct sk_buff *skb, struct netlink_callback *cb, continue; if ((err = inet6_fill_ifaddr(skb, ifa, NETLINK_CB(cb->skb).pid, - cb->nlh->nlmsg_seq, RTM_NEWADDR)) <= 0) + cb->nlh->nlmsg_seq, RTM_NEWADDR, + NLM_F_MULTI)) <= 0) goto done; } /* temp addr */ @@ -2796,7 +2796,8 @@ static int inet6_dump_addr(struct sk_buff *skb, struct netlink_callback *cb, continue; if ((err = inet6_fill_ifaddr(skb, ifa, NETLINK_CB(cb->skb).pid, - cb->nlh->nlmsg_seq, RTM_NEWADDR)) <= 0) + cb->nlh->nlmsg_seq, RTM_NEWADDR, + NLM_F_MULTI)) <= 0) goto done; } #endif @@ -2809,7 +2810,8 @@ static int inet6_dump_addr(struct sk_buff *skb, struct netlink_callback *cb, continue; if ((err = inet6_fill_ifmcaddr(skb, ifmca, NETLINK_CB(cb->skb).pid, - cb->nlh->nlmsg_seq, RTM_GETMULTICAST)) <= 0) + cb->nlh->nlmsg_seq, RTM_GETMULTICAST, + NLM_F_MULTI)) <= 0) goto done; } break; @@ -2821,7 +2823,8 @@ static int inet6_dump_addr(struct sk_buff *skb, struct netlink_callback *cb, continue; if ((err = inet6_fill_ifacaddr(skb, ifaca, NETLINK_CB(cb->skb).pid, - cb->nlh->nlmsg_seq, RTM_GETANYCAST)) <= 0) + cb->nlh->nlmsg_seq, RTM_GETANYCAST, + NLM_F_MULTI)) <= 0) goto done; } break; @@ -2871,7 +2874,7 @@ static void inet6_ifa_notify(int event, struct inet6_ifaddr *ifa) netlink_set_err(rtnl, 0, RTMGRP_IPV6_IFADDR, ENOBUFS); return; } - if (inet6_fill_ifaddr(skb, ifa, 0, 0, event) < 0) { + if (inet6_fill_ifaddr(skb, ifa, current->pid, 0, event, 0) < 0) { kfree_skb(skb); netlink_set_err(rtnl, 0, RTMGRP_IPV6_IFADDR, EINVAL); return; @@ -2906,7 +2909,7 @@ static void inline ipv6_store_devconf(struct ipv6_devconf *cnf, } static int inet6_fill_ifinfo(struct sk_buff *skb, struct inet6_dev *idev, - u32 pid, u32 seq, int event) + u32 pid, u32 seq, int event, unsigned int flags) { struct net_device *dev = idev->dev; __s32 *array = NULL; @@ -2917,8 +2920,7 @@ static int inet6_fill_ifinfo(struct sk_buff *skb, struct inet6_dev *idev, __u32 mtu = dev->mtu; struct ifla_cacheinfo ci; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*r)); - if (pid) nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*r), flags); r = NLMSG_DATA(nlh); r->ifi_family = AF_INET6; r->ifi_type = dev->type; @@ -2985,7 +2987,7 @@ static int inet6_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) if ((idev = in6_dev_get(dev)) == NULL) continue; err = inet6_fill_ifinfo(skb, idev, NETLINK_CB(cb->skb).pid, - cb->nlh->nlmsg_seq, RTM_NEWLINK); + cb->nlh->nlmsg_seq, RTM_NEWLINK, NLM_F_MULTI); in6_dev_put(idev); if (err <= 0) break; @@ -3007,7 +3009,7 @@ void inet6_ifinfo_notify(int event, struct inet6_dev *idev) netlink_set_err(rtnl, 0, RTMGRP_IPV6_IFINFO, ENOBUFS); return; } - if (inet6_fill_ifinfo(skb, idev, 0, 0, event) < 0) { + if (inet6_fill_ifinfo(skb, idev, current->pid, 0, event, 0) < 0) { kfree_skb(skb); netlink_set_err(rtnl, 0, RTMGRP_IPV6_IFINFO, EINVAL); return; @@ -3017,18 +3019,15 @@ void inet6_ifinfo_notify(int event, struct inet6_dev *idev) } static int inet6_fill_prefix(struct sk_buff *skb, struct inet6_dev *idev, - struct prefix_info *pinfo, u32 pid, u32 seq, int event) + struct prefix_info *pinfo, u32 pid, u32 seq, + int event, unsigned int flags) { struct prefixmsg *pmsg; struct nlmsghdr *nlh; unsigned char *b = skb->tail; struct prefix_cacheinfo ci; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*pmsg)); - - if (pid) - nlh->nlmsg_flags |= NLM_F_MULTI; - + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*pmsg), flags); pmsg = NLMSG_DATA(nlh); pmsg->prefix_family = AF_INET6; pmsg->prefix_ifindex = idev->dev->ifindex; @@ -3067,7 +3066,7 @@ static void inet6_prefix_notify(int event, struct inet6_dev *idev, netlink_set_err(rtnl, 0, RTMGRP_IPV6_PREFIX, ENOBUFS); return; } - if (inet6_fill_prefix(skb, idev, pinfo, 0, 0, event) < 0) { + if (inet6_fill_prefix(skb, idev, pinfo, current->pid, 0, event, 0) < 0) { kfree_skb(skb); netlink_set_err(rtnl, 0, RTMGRP_IPV6_PREFIX, EINVAL); return; @@ -3096,7 +3095,7 @@ 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); @@ -3106,7 +3105,7 @@ 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); @@ -3439,7 +3438,7 @@ static void addrconf_sysctl_register(struct inet6_dev *idev, struct ipv6_devconf * by sysctl and we wouldn't want anyone to change it under our feet * (see SIOCSIFNAME). */ - dev_name = net_sysctl_strdup(dev_name); + dev_name = kstrdup(dev_name, GFP_KERNEL); if (!dev_name) goto free; 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/datagram.c b/net/ipv6/datagram.c index 65b9375..5229365 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -353,14 +353,14 @@ int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len) err = copied; /* Reset and regenerate socket error */ - spin_lock_irq(&sk->sk_error_queue.lock); + spin_lock_bh(&sk->sk_error_queue.lock); sk->sk_err = 0; if ((skb2 = skb_peek(&sk->sk_error_queue)) != NULL) { sk->sk_err = SKB_EXT_ERR(skb2)->ee.ee_errno; - spin_unlock_irq(&sk->sk_error_queue.lock); + spin_unlock_bh(&sk->sk_error_queue.lock); sk->sk_error_report(sk); } else { - spin_unlock_irq(&sk->sk_error_queue.lock); + spin_unlock_bh(&sk->sk_error_queue.lock); } out_free_skb: 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/icmp.c b/net/ipv6/icmp.c index 8e0f569..ff3ec98 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -277,8 +277,8 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, { struct inet6_dev *idev = NULL; struct ipv6hdr *hdr = skb->nh.ipv6h; - struct sock *sk = icmpv6_socket->sk; - struct ipv6_pinfo *np = inet6_sk(sk); + struct sock *sk; + struct ipv6_pinfo *np; struct in6_addr *saddr = NULL; struct dst_entry *dst; struct icmp6hdr tmp_hdr; @@ -358,6 +358,9 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, if (icmpv6_xmit_lock()) return; + sk = icmpv6_socket->sk; + np = inet6_sk(sk); + if (!icmpv6_xrlim_allow(sk, type, &fl)) goto out; @@ -423,9 +426,9 @@ out: static void icmpv6_echo_reply(struct sk_buff *skb) { - struct sock *sk = icmpv6_socket->sk; + struct sock *sk; struct inet6_dev *idev; - struct ipv6_pinfo *np = inet6_sk(sk); + struct ipv6_pinfo *np; struct in6_addr *saddr = NULL; struct icmp6hdr *icmph = (struct icmp6hdr *) skb->h.raw; struct icmp6hdr tmp_hdr; @@ -454,6 +457,9 @@ static void icmpv6_echo_reply(struct sk_buff *skb) if (icmpv6_xmit_lock()) return; + sk = icmpv6_socket->sk; + np = inet6_sk(sk); + if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst)) fl.oif = np->mcast_oif; 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_flowlabel.c b/net/ipv6/ip6_flowlabel.c index a93f6dc..0e5f7499 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -535,10 +535,12 @@ release: if (err) goto done; - /* Do not check for fault */ - if (!freq.flr_label) - copy_to_user(&((struct in6_flowlabel_req __user *) optval)->flr_label, - &fl->label, sizeof(fl->label)); + if (!freq.flr_label) { + if (copy_to_user(&((struct in6_flowlabel_req __user *) optval)->flr_label, + &fl->label, sizeof(fl->label))) { + /* Intentionally ignore fault. */ + } + } sfl1->fl = fl; sfl1->next = np->ipv6_fl_list; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 0f07114..06e7cdae 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 } @@ -552,13 +549,17 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) skb_headroom(frag) < hlen) goto slow_path; - /* Correct socket ownership. */ - if (frag->sk == NULL) - goto slow_path; - /* Partially cloned skb? */ if (skb_shared(frag)) goto slow_path; + + BUG_ON(frag->sk); + if (skb->sk) { + sock_hold(skb->sk); + frag->sk = skb->sk; + frag->destructor = sock_wfree; + skb->truesize -= frag->truesize; + } } err = 0; @@ -1116,12 +1117,10 @@ int ip6_push_pending_frames(struct sock *sk) tail_skb = &(tmp_skb->next); skb->len += tmp_skb->len; skb->data_len += tmp_skb->len; -#if 0 /* Logically correct, but useless work, ip_fragment() will have to undo */ skb->truesize += tmp_skb->truesize; __sock_put(tmp_skb->sk); tmp_skb->destructor = NULL; tmp_skb->sk = NULL; -#endif } ipv6_addr_copy(final_dst, &fl->fl6_dst); diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 3b1c9fa..ba3b0c2 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -882,6 +882,7 @@ ip6ip6_tnl_change(struct ip6_tnl *t, struct ip6_tnl_parm *p) t->parms.hop_limit = p->hop_limit; t->parms.encap_limit = p->encap_limit; t->parms.flowinfo = p->flowinfo; + t->parms.link = p->link; ip6ip6_tnl_link_config(t); return 0; } 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/ipv6_syms.c b/net/ipv6/ipv6_syms.c index 2f4c91d..5ade5a5 100644 --- a/net/ipv6/ipv6_syms.c +++ b/net/ipv6/ipv6_syms.c @@ -37,5 +37,4 @@ EXPORT_SYMBOL(in6_dev_finish_destroy); EXPORT_SYMBOL(xfrm6_rcv); #endif EXPORT_SYMBOL(rt6_lookup); -EXPORT_SYMBOL(fl6_sock_lookup); EXPORT_SYMBOL(ipv6_push_nfrag_opts); 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/raw.c b/net/ipv6/raw.c index 617645b..e2b848e 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -434,12 +434,12 @@ csum_copy_err: /* Clear queue. */ if (flags&MSG_PEEK) { int clear = 0; - spin_lock_irq(&sk->sk_receive_queue.lock); + spin_lock_bh(&sk->sk_receive_queue.lock); if (skb == skb_peek(&sk->sk_receive_queue)) { __skb_unlink(skb, &sk->sk_receive_queue); clear = 1; } - spin_unlock_irq(&sk->sk_receive_queue.lock); + spin_unlock_bh(&sk->sk_receive_queue.lock); if (clear) kfree_skb(skb); } @@ -971,11 +971,11 @@ static int rawv6_ioctl(struct sock *sk, int cmd, unsigned long arg) struct sk_buff *skb; int amount = 0; - spin_lock_irq(&sk->sk_receive_queue.lock); + spin_lock_bh(&sk->sk_receive_queue.lock); skb = skb_peek(&sk->sk_receive_queue); if (skb != NULL) amount = skb->tail - skb->h.raw; - spin_unlock_irq(&sk->sk_receive_queue.lock); + spin_unlock_bh(&sk->sk_receive_queue.lock); return put_user(amount, (int __user *)arg); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 3bf8a02..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 @@ -1566,11 +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) + 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; @@ -1584,11 +1585,7 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, } } - if (!pid && in_nlh) { - pid = in_nlh->nlmsg_pid; - } - - nlh = NLMSG_PUT(skb, pid, seq, type, sizeof(*rtm)); + nlh = NLMSG_NEW(skb, pid, seq, type, sizeof(*rtm), flags); rtm = NLMSG_DATA(nlh); rtm->rtm_family = AF_INET6; rtm->rtm_dst_len = rt->rt6i_dst.plen; @@ -1674,7 +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); + prefix, NLM_F_MULTI); } static int fib6_dump_node(struct fib6_walker_t *w) @@ -1822,7 +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); + nlh->nlmsg_seq, 0, 0); if (err < 0) { err = -EMSGSIZE; goto out_free; @@ -1838,17 +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) { + 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/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 0f69e80..9dac7fd 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -65,7 +65,7 @@ #include <linux/seq_file.h> static void tcp_v6_send_reset(struct sk_buff *skb); -static void tcp_v6_or_send_ack(struct sk_buff *skb, struct open_request *req); +static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req); static void tcp_v6_send_check(struct sock *sk, struct tcphdr *th, int len, struct sk_buff *skb); @@ -394,24 +394,26 @@ static u32 tcp_v6_synq_hash(struct in6_addr *raddr, u16 rport, u32 rnd) return c & (TCP_SYNQ_HSIZE - 1); } -static struct open_request *tcp_v6_search_req(struct tcp_sock *tp, - struct open_request ***prevp, +static struct request_sock *tcp_v6_search_req(struct tcp_sock *tp, + struct request_sock ***prevp, __u16 rport, struct in6_addr *raddr, struct in6_addr *laddr, int iif) { - struct tcp_listen_opt *lopt = tp->listen_opt; - struct open_request *req, **prev; + struct listen_sock *lopt = tp->accept_queue.listen_opt; + struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v6_synq_hash(raddr, rport, lopt->hash_rnd)]; (req = *prev) != NULL; prev = &req->dl_next) { - if (req->rmt_port == rport && - req->class->family == AF_INET6 && - ipv6_addr_equal(&req->af.v6_req.rmt_addr, raddr) && - ipv6_addr_equal(&req->af.v6_req.loc_addr, laddr) && - (!req->af.v6_req.iif || req->af.v6_req.iif == iif)) { + const struct tcp6_request_sock *treq = tcp6_rsk(req); + + if (inet_rsk(req)->rmt_port == rport && + req->rsk_ops->family == AF_INET6 && + ipv6_addr_equal(&treq->rmt_addr, raddr) && + ipv6_addr_equal(&treq->loc_addr, laddr) && + (!treq->iif || treq->iif == iif)) { BUG_TRAP(req->sk == NULL); *prevp = prev; return req; @@ -906,9 +908,9 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, icmpv6_err_convert(type, code, &err); - /* Might be for an open_request */ + /* Might be for an request_sock */ switch (sk->sk_state) { - struct open_request *req, **prev; + struct request_sock *req, **prev; case TCP_LISTEN: if (sock_owned_by_user(sk)) goto out; @@ -923,7 +925,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, */ BUG_TRAP(req->sk == NULL); - if (seq != req->snt_isn) { + if (seq != tcp_rsk(req)->snt_isn) { NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS); goto out; } @@ -957,9 +959,10 @@ out: } -static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, +static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req, struct dst_entry *dst) { + struct tcp6_request_sock *treq = tcp6_rsk(req); struct ipv6_pinfo *np = inet6_sk(sk); struct sk_buff * skb; struct ipv6_txoptions *opt = NULL; @@ -969,19 +972,19 @@ static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, memset(&fl, 0, sizeof(fl)); fl.proto = IPPROTO_TCP; - ipv6_addr_copy(&fl.fl6_dst, &req->af.v6_req.rmt_addr); - ipv6_addr_copy(&fl.fl6_src, &req->af.v6_req.loc_addr); + ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); + ipv6_addr_copy(&fl.fl6_src, &treq->loc_addr); fl.fl6_flowlabel = 0; - fl.oif = req->af.v6_req.iif; - fl.fl_ip_dport = req->rmt_port; + fl.oif = treq->iif; + fl.fl_ip_dport = inet_rsk(req)->rmt_port; fl.fl_ip_sport = inet_sk(sk)->sport; if (dst == NULL) { opt = np->opt; if (opt == NULL && np->rxopt.bits.srcrt == 2 && - req->af.v6_req.pktopts) { - struct sk_buff *pktopts = req->af.v6_req.pktopts; + treq->pktopts) { + struct sk_buff *pktopts = treq->pktopts; struct inet6_skb_parm *rxopt = IP6CB(pktopts); if (rxopt->srcrt) opt = ipv6_invert_rthdr(sk, (struct ipv6_rt_hdr*)(pktopts->nh.raw + rxopt->srcrt)); @@ -1008,10 +1011,10 @@ static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, struct tcphdr *th = skb->h.th; th->check = tcp_v6_check(th, skb->len, - &req->af.v6_req.loc_addr, &req->af.v6_req.rmt_addr, + &treq->loc_addr, &treq->rmt_addr, csum_partial((char *)th, skb->len, skb->csum)); - ipv6_addr_copy(&fl.fl6_dst, &req->af.v6_req.rmt_addr); + ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); err = ip6_xmit(sk, skb, &fl, opt, 0); if (err == NET_XMIT_CN) err = 0; @@ -1024,17 +1027,18 @@ done: return err; } -static void tcp_v6_or_free(struct open_request *req) +static void tcp_v6_reqsk_destructor(struct request_sock *req) { - if (req->af.v6_req.pktopts) - kfree_skb(req->af.v6_req.pktopts); + if (tcp6_rsk(req)->pktopts) + kfree_skb(tcp6_rsk(req)->pktopts); } -static struct or_calltable or_ipv6 = { +static struct request_sock_ops tcp6_request_sock_ops = { .family = AF_INET6, + .obj_size = sizeof(struct tcp6_request_sock), .rtx_syn_ack = tcp_v6_send_synack, - .send_ack = tcp_v6_or_send_ack, - .destructor = tcp_v6_or_free, + .send_ack = tcp_v6_reqsk_send_ack, + .destructor = tcp_v6_reqsk_destructor, .send_reset = tcp_v6_send_reset }; @@ -1219,15 +1223,15 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) tcp_tw_put(tw); } -static void tcp_v6_or_send_ack(struct sk_buff *skb, struct open_request *req) +static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req) { - tcp_v6_send_ack(skb, req->snt_isn+1, req->rcv_isn+1, req->rcv_wnd, req->ts_recent); + tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) { - struct open_request *req, **prev; + struct request_sock *req, **prev; struct tcphdr *th = skb->h.th; struct tcp_sock *tp = tcp_sk(sk); struct sock *nsk; @@ -1260,21 +1264,13 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) return sk; } -static void tcp_v6_synq_add(struct sock *sk, struct open_request *req) +static void tcp_v6_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; - u32 h = tcp_v6_synq_hash(&req->af.v6_req.rmt_addr, req->rmt_port, lopt->hash_rnd); - - req->sk = NULL; - req->expires = jiffies + TCP_TIMEOUT_INIT; - req->retrans = 0; - req->dl_next = lopt->syn_table[h]; - - write_lock(&tp->syn_wait_lock); - lopt->syn_table[h] = req; - write_unlock(&tp->syn_wait_lock); + struct listen_sock *lopt = tp->accept_queue.listen_opt; + u32 h = tcp_v6_synq_hash(&tcp6_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); + reqsk_queue_hash_req(&tp->accept_queue, h, req, TCP_TIMEOUT_INIT); tcp_synq_added(sk); } @@ -1284,10 +1280,11 @@ static void tcp_v6_synq_add(struct sock *sk, struct open_request *req) */ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) { + struct tcp6_request_sock *treq; struct ipv6_pinfo *np = inet6_sk(sk); struct tcp_options_received tmp_opt; struct tcp_sock *tp = tcp_sk(sk); - struct open_request *req = NULL; + struct request_sock *req = NULL; __u32 isn = TCP_SKB_CB(skb)->when; if (skb->protocol == htons(ETH_P_IP)) @@ -1308,7 +1305,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1) goto drop; - req = tcp_openreq_alloc(); + req = reqsk_alloc(&tcp6_request_sock_ops); if (req == NULL) goto drop; @@ -1321,28 +1318,28 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) tmp_opt.tstamp_ok = tmp_opt.saw_tstamp; tcp_openreq_init(req, &tmp_opt, skb); - req->class = &or_ipv6; - ipv6_addr_copy(&req->af.v6_req.rmt_addr, &skb->nh.ipv6h->saddr); - ipv6_addr_copy(&req->af.v6_req.loc_addr, &skb->nh.ipv6h->daddr); + treq = tcp6_rsk(req); + ipv6_addr_copy(&treq->rmt_addr, &skb->nh.ipv6h->saddr); + ipv6_addr_copy(&treq->loc_addr, &skb->nh.ipv6h->daddr); TCP_ECN_create_request(req, skb->h.th); - req->af.v6_req.pktopts = NULL; + treq->pktopts = NULL; if (ipv6_opt_accepted(sk, skb) || np->rxopt.bits.rxinfo || np->rxopt.bits.rxhlim) { atomic_inc(&skb->users); - req->af.v6_req.pktopts = skb; + treq->pktopts = skb; } - req->af.v6_req.iif = sk->sk_bound_dev_if; + treq->iif = sk->sk_bound_dev_if; /* So that link locals have meaning */ if (!sk->sk_bound_dev_if && - ipv6_addr_type(&req->af.v6_req.rmt_addr) & IPV6_ADDR_LINKLOCAL) - req->af.v6_req.iif = tcp_v6_iif(skb); + ipv6_addr_type(&treq->rmt_addr) & IPV6_ADDR_LINKLOCAL) + treq->iif = tcp_v6_iif(skb); if (isn == 0) isn = tcp_v6_init_sequence(sk,skb); - req->snt_isn = isn; + tcp_rsk(req)->snt_isn = isn; if (tcp_v6_send_synack(sk, req, NULL)) goto drop; @@ -1353,16 +1350,17 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) drop: if (req) - tcp_openreq_free(req); + reqsk_free(req); TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS); return 0; /* don't send reset */ } static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst) { + struct tcp6_request_sock *treq = tcp6_rsk(req); struct ipv6_pinfo *newnp, *np = inet6_sk(sk); struct tcp6_sock *newtcp6sk; struct inet_sock *newinet; @@ -1426,10 +1424,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, goto out_overflow; if (np->rxopt.bits.srcrt == 2 && - opt == NULL && req->af.v6_req.pktopts) { - struct inet6_skb_parm *rxopt = IP6CB(req->af.v6_req.pktopts); + opt == NULL && treq->pktopts) { + struct inet6_skb_parm *rxopt = IP6CB(treq->pktopts); if (rxopt->srcrt) - opt = ipv6_invert_rthdr(sk, (struct ipv6_rt_hdr*)(req->af.v6_req.pktopts->nh.raw+rxopt->srcrt)); + opt = ipv6_invert_rthdr(sk, (struct ipv6_rt_hdr *)(treq->pktopts->nh.raw + rxopt->srcrt)); } if (dst == NULL) { @@ -1438,16 +1436,16 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, memset(&fl, 0, sizeof(fl)); fl.proto = IPPROTO_TCP; - ipv6_addr_copy(&fl.fl6_dst, &req->af.v6_req.rmt_addr); + ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); if (opt && opt->srcrt) { struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt; ipv6_addr_copy(&final, &fl.fl6_dst); ipv6_addr_copy(&fl.fl6_dst, rt0->addr); final_p = &final; } - ipv6_addr_copy(&fl.fl6_src, &req->af.v6_req.loc_addr); + ipv6_addr_copy(&fl.fl6_src, &treq->loc_addr); fl.oif = sk->sk_bound_dev_if; - fl.fl_ip_dport = req->rmt_port; + fl.fl_ip_dport = inet_rsk(req)->rmt_port; fl.fl_ip_sport = inet_sk(sk)->sport; if (ip6_dst_lookup(sk, &dst, &fl)) @@ -1482,10 +1480,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, memcpy(newnp, np, sizeof(struct ipv6_pinfo)); - ipv6_addr_copy(&newnp->daddr, &req->af.v6_req.rmt_addr); - ipv6_addr_copy(&newnp->saddr, &req->af.v6_req.loc_addr); - ipv6_addr_copy(&newnp->rcv_saddr, &req->af.v6_req.loc_addr); - newsk->sk_bound_dev_if = req->af.v6_req.iif; + ipv6_addr_copy(&newnp->daddr, &treq->rmt_addr); + ipv6_addr_copy(&newnp->saddr, &treq->loc_addr); + ipv6_addr_copy(&newnp->rcv_saddr, &treq->loc_addr); + newsk->sk_bound_dev_if = treq->iif; /* Now IPv6 options... @@ -1498,11 +1496,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, /* Clone pktoptions received with SYN */ newnp->pktoptions = NULL; - if (req->af.v6_req.pktopts) { - newnp->pktoptions = skb_clone(req->af.v6_req.pktopts, - GFP_ATOMIC); - kfree_skb(req->af.v6_req.pktopts); - req->af.v6_req.pktopts = NULL; + if (treq->pktopts != NULL) { + newnp->pktoptions = skb_clone(treq->pktopts, GFP_ATOMIC); + kfree_skb(treq->pktopts); + treq->pktopts = NULL; if (newnp->pktoptions) skb_set_owner_r(newnp->pktoptions, newsk); } @@ -2028,7 +2025,7 @@ static int tcp_v6_init_sock(struct sock *sk) sk->sk_state = TCP_CLOSE; tp->af_specific = &ipv6_specific; - + tp->ca_ops = &tcp_init_congestion_ops; sk->sk_write_space = sk_stream_write_space; sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); @@ -2050,7 +2047,7 @@ static int tcp_v6_destroy_sock(struct sock *sk) /* Proc filesystem TCPv6 sock list dumping. */ static void get_openreq6(struct seq_file *seq, - struct sock *sk, struct open_request *req, int i, int uid) + struct sock *sk, struct request_sock *req, int i, int uid) { struct in6_addr *dest, *src; int ttd = req->expires - jiffies; @@ -2058,8 +2055,8 @@ static void get_openreq6(struct seq_file *seq, if (ttd < 0) ttd = 0; - src = &req->af.v6_req.loc_addr; - dest = &req->af.v6_req.rmt_addr; + src = &tcp6_rsk(req)->loc_addr; + dest = &tcp6_rsk(req)->rmt_addr; seq_printf(seq, "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X " "%02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %p\n", @@ -2069,7 +2066,7 @@ static void get_openreq6(struct seq_file *seq, ntohs(inet_sk(sk)->sport), dest->s6_addr32[0], dest->s6_addr32[1], dest->s6_addr32[2], dest->s6_addr32[3], - ntohs(req->rmt_port), + ntohs(inet_rsk(req)->rmt_port), TCP_SYN_RECV, 0,0, /* could print option size, but that is af dependent. */ 1, /* timers active (only the expire timer) */ @@ -2239,6 +2236,7 @@ struct proto tcpv6_prot = { .sysctl_rmem = sysctl_tcp_rmem, .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp6_sock), + .rsk_prot = &tcp6_request_sock_ops, }; static struct inet6_protocol tcpv6_protocol = { diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index e251d0b..eff050a 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -300,12 +300,12 @@ csum_copy_err: /* Clear queue. */ if (flags&MSG_PEEK) { int clear = 0; - spin_lock_irq(&sk->sk_receive_queue.lock); + spin_lock_bh(&sk->sk_receive_queue.lock); if (skb == skb_peek(&sk->sk_receive_queue)) { __skb_unlink(skb, &sk->sk_receive_queue); clear = 1; } - spin_unlock_irq(&sk->sk_receive_queue.lock); + spin_unlock_bh(&sk->sk_receive_queue.lock); if (clear) kfree_skb(skb); } diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c index 601a148..6b98677 100644 --- a/net/ipv6/xfrm6_output.c +++ b/net/ipv6/xfrm6_output.c @@ -84,6 +84,7 @@ static int xfrm6_tunnel_check_size(struct sk_buff *skb) mtu = IPV6_MIN_MTU; if (skb->len > mtu) { + skb->dev = dst->dev; icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); ret = -EMSGSIZE; } diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 4429b1a..cf1d91e 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -113,6 +113,8 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int xdst = (struct xfrm_dst *)dst1; xdst->route = &rt->u.dst; + if (rt->rt6i_node) + xdst->route_cookie = rt->rt6i_node->fn_sernum; dst1->next = dst_prev; dst_prev = dst1; @@ -137,6 +139,8 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int dst_prev->child = &rt->u.dst; dst->path = &rt->u.dst; + if (rt->rt6i_node) + ((struct xfrm_dst *)dst)->path_cookie = rt->rt6i_node->fn_sernum; *dst_p = dst; dst = dst_prev; 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 ce980aa..4879743 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -656,13 +656,18 @@ static struct sk_buff * pfkey_xfrm_state2msg(struct xfrm_state *x, int add_keys, sa->sadb_sa_exttype = SADB_EXT_SA; sa->sadb_sa_spi = x->id.spi; sa->sadb_sa_replay = x->props.replay_window; - sa->sadb_sa_state = SADB_SASTATE_DYING; - if (x->km.state == XFRM_STATE_VALID && !x->km.dying) - sa->sadb_sa_state = SADB_SASTATE_MATURE; - else if (x->km.state == XFRM_STATE_ACQ) + switch (x->km.state) { + case XFRM_STATE_VALID: + sa->sadb_sa_state = x->km.dying ? + SADB_SASTATE_DYING : SADB_SASTATE_MATURE; + break; + case XFRM_STATE_ACQ: sa->sadb_sa_state = SADB_SASTATE_LARVAL; - else if (x->km.state == XFRM_STATE_EXPIRED) + break; + default: sa->sadb_sa_state = SADB_SASTATE_DEAD; + break; + } sa->sadb_sa_auth = 0; if (x->aalg) { struct xfrm_algo_desc *a = xfrm_aalg_get_byname(x->aalg->alg_name, 0); @@ -685,6 +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) { @@ -969,6 +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) { @@ -1091,17 +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: @@ -1240,13 +1243,78 @@ static int pfkey_acquire(struct sock *sk, struct sk_buff *skb, struct sadb_msg * return 0; } +static inline int event2poltype(int event) +{ + switch (event) { + case XFRM_MSG_DELPOLICY: + return SADB_X_SPDDELETE; + case XFRM_MSG_NEWPOLICY: + return SADB_X_SPDADD; + case XFRM_MSG_UPDPOLICY: + return SADB_X_SPDUPDATE; + case XFRM_MSG_POLEXPIRE: + // return SADB_X_SPDEXPIRE; + default: + printk("pfkey: Unknown policy event %d\n", event); + break; + } + + return 0; +} + +static inline int event2keytype(int event) +{ + switch (event) { + case XFRM_MSG_DELSA: + return SADB_DELETE; + case XFRM_MSG_NEWSA: + return SADB_ADD; + case XFRM_MSG_UPDSA: + return SADB_UPDATE; + case XFRM_MSG_EXPIRE: + return SADB_EXPIRE; + default: + printk("pfkey: Unknown SA event %d\n", event); + break; + } + + return 0; +} + +/* ADD/UPD/DEL */ +static int key_notify_sa(struct xfrm_state *x, struct km_event *c) +{ + struct sk_buff *skb; + struct sadb_msg *hdr; + int hsc = 3; + + if (c->event == XFRM_MSG_DELSA) + hsc = 0; + + skb = pfkey_xfrm_state2msg(x, 0, hsc); + + if (IS_ERR(skb)) + return PTR_ERR(skb); + + hdr = (struct sadb_msg *) skb->data; + hdr->sadb_msg_version = PF_KEY_V2; + hdr->sadb_msg_type = event2keytype(c->event); + hdr->sadb_msg_satype = pfkey_proto2satype(x->id.proto); + hdr->sadb_msg_errno = 0; + hdr->sadb_msg_reserved = 0; + hdr->sadb_msg_seq = c->seq; + hdr->sadb_msg_pid = c->pid; + + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL); + + return 0; +} static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; struct xfrm_state *x; int err; + struct km_event c; xfrm_probe_algs(); @@ -1254,6 +1322,7 @@ static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, if (IS_ERR(x)) return PTR_ERR(x); + xfrm_state_hold(x); if (hdr->sadb_msg_type == SADB_ADD) err = xfrm_state_add(x); else @@ -1262,30 +1331,26 @@ static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, if (err < 0) { x->km.state = XFRM_STATE_DEAD; xfrm_state_put(x); - return err; + goto out; } - out_skb = pfkey_xfrm_state2msg(x, 0, 3); - if (IS_ERR(out_skb)) - return PTR_ERR(out_skb); /* XXX Should we return 0 here ? */ - - out_hdr = (struct sadb_msg *) out_skb->data; - out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = hdr->sadb_msg_type; - out_hdr->sadb_msg_satype = pfkey_proto2satype(x->id.proto); - out_hdr->sadb_msg_errno = 0; - out_hdr->sadb_msg_reserved = 0; - out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; - out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); - - return 0; + if (hdr->sadb_msg_type == SADB_ADD) + c.event = XFRM_MSG_NEWSA; + else + c.event = XFRM_MSG_UPDSA; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + km_state_notify(x, &c); +out: + xfrm_state_put(x); + return err; } static int pfkey_delete(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { struct xfrm_state *x; + struct km_event c; + int err; if (!ext_hdrs[SADB_EXT_SA-1] || !present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1], @@ -1301,13 +1366,19 @@ static int pfkey_delete(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h return -EPERM; } - xfrm_state_delete(x); - xfrm_state_put(x); + err = xfrm_state_delete(x); + if (err < 0) { + xfrm_state_put(x); + return err; + } - pfkey_broadcast(skb_clone(skb, GFP_KERNEL), GFP_KERNEL, - BROADCAST_ALL, sk); + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + c.event = XFRM_MSG_DELSA; + km_state_notify(x, &c); + xfrm_state_put(x); - return 0; + return err; } static int pfkey_get(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) @@ -1445,28 +1516,42 @@ static int pfkey_register(struct sock *sk, struct sk_buff *skb, struct sadb_msg return 0; } +static int key_notify_sa_flush(struct km_event *c) +{ + struct sk_buff *skb; + struct sadb_msg *hdr; + + skb = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_ATOMIC); + if (!skb) + return -ENOBUFS; + hdr = (struct sadb_msg *) skb_put(skb, sizeof(struct sadb_msg)); + hdr->sadb_msg_satype = pfkey_proto2satype(c->data.proto); + hdr->sadb_msg_seq = c->seq; + hdr->sadb_msg_pid = c->pid; + hdr->sadb_msg_version = PF_KEY_V2; + hdr->sadb_msg_errno = (uint8_t) 0; + hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); + + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL); + + return 0; +} + static int pfkey_flush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { unsigned proto; - struct sk_buff *skb_out; - struct sadb_msg *hdr_out; + struct km_event c; proto = pfkey_satype2proto(hdr->sadb_msg_satype); if (proto == 0) return -EINVAL; - skb_out = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_KERNEL); - if (!skb_out) - return -ENOBUFS; - xfrm_state_flush(proto); - - hdr_out = (struct sadb_msg *) skb_put(skb_out, sizeof(struct sadb_msg)); - pfkey_hdr_dup(hdr_out, hdr); - hdr_out->sadb_msg_errno = (uint8_t) 0; - hdr_out->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); - - pfkey_broadcast(skb_out, GFP_KERNEL, BROADCAST_ALL, NULL); + c.data.proto = proto; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + c.event = XFRM_MSG_FLUSHSA; + km_state_notify(NULL, &c); return 0; } @@ -1859,6 +1944,35 @@ static void pfkey_xfrm_policy2msg(struct sk_buff *skb, struct xfrm_policy *xp, i hdr->sadb_msg_reserved = atomic_read(&xp->refcnt); } +static int key_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + struct sk_buff *out_skb; + struct sadb_msg *out_hdr; + int err; + + out_skb = pfkey_xfrm_policy2msg_prep(xp); + if (IS_ERR(out_skb)) { + err = PTR_ERR(out_skb); + goto out; + } + pfkey_xfrm_policy2msg(out_skb, xp, dir); + + out_hdr = (struct sadb_msg *) out_skb->data; + out_hdr->sadb_msg_version = PF_KEY_V2; + + if (c->data.byid && c->event == XFRM_MSG_DELPOLICY) + out_hdr->sadb_msg_type = SADB_X_SPDDELETE2; + else + out_hdr->sadb_msg_type = event2poltype(c->event); + out_hdr->sadb_msg_errno = 0; + out_hdr->sadb_msg_seq = c->seq; + out_hdr->sadb_msg_pid = c->pid; + pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, NULL); +out: + return 0; + +} + static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { int err; @@ -1866,8 +1980,7 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h struct sadb_address *sa; struct sadb_x_policy *pol; struct xfrm_policy *xp; - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; + struct km_event c; if (!present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1], ext_hdrs[SADB_EXT_ADDRESS_DST-1]) || @@ -1935,31 +2048,23 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h (err = parse_ipsecrequests(xp, pol)) < 0) goto out; - out_skb = pfkey_xfrm_policy2msg_prep(xp); - if (IS_ERR(out_skb)) { - err = PTR_ERR(out_skb); - goto out; - } - err = xfrm_policy_insert(pol->sadb_x_policy_dir-1, xp, hdr->sadb_msg_type != SADB_X_SPDUPDATE); if (err) { - kfree_skb(out_skb); - goto out; + kfree(xp); + return err; } - pfkey_xfrm_policy2msg(out_skb, xp, pol->sadb_x_policy_dir-1); + if (hdr->sadb_msg_type == SADB_X_SPDUPDATE) + c.event = XFRM_MSG_UPDPOLICY; + else + c.event = XFRM_MSG_NEWPOLICY; - xfrm_pol_put(xp); + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; - out_hdr = (struct sadb_msg *) out_skb->data; - out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = hdr->sadb_msg_type; - out_hdr->sadb_msg_satype = 0; - out_hdr->sadb_msg_errno = 0; - out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; - out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); + km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); + xfrm_pol_put(xp); return 0; out: @@ -1973,9 +2078,8 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg struct sadb_address *sa; struct sadb_x_policy *pol; struct xfrm_policy *xp; - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; struct xfrm_selector sel; + struct km_event c; if (!present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1], ext_hdrs[SADB_EXT_ADDRESS_DST-1]) || @@ -2010,25 +2114,40 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg err = 0; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + c.event = XFRM_MSG_DELPOLICY; + km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); + + xfrm_pol_put(xp); + return err; +} + +static int key_pol_get_resp(struct sock *sk, struct xfrm_policy *xp, struct sadb_msg *hdr, int dir) +{ + int err; + struct sk_buff *out_skb; + struct sadb_msg *out_hdr; + err = 0; + out_skb = pfkey_xfrm_policy2msg_prep(xp); if (IS_ERR(out_skb)) { err = PTR_ERR(out_skb); goto out; } - pfkey_xfrm_policy2msg(out_skb, xp, pol->sadb_x_policy_dir-1); + pfkey_xfrm_policy2msg(out_skb, xp, dir); out_hdr = (struct sadb_msg *) out_skb->data; out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = SADB_X_SPDDELETE; + out_hdr->sadb_msg_type = hdr->sadb_msg_type; out_hdr->sadb_msg_satype = 0; out_hdr->sadb_msg_errno = 0; out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); + pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ONE, sk); err = 0; out: - xfrm_pol_put(xp); return err; } @@ -2037,8 +2156,7 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h int err; struct sadb_x_policy *pol; struct xfrm_policy *xp; - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; + struct km_event c; if ((pol = ext_hdrs[SADB_X_EXT_POLICY-1]) == NULL) return -EINVAL; @@ -2050,24 +2168,16 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h err = 0; - out_skb = pfkey_xfrm_policy2msg_prep(xp); - if (IS_ERR(out_skb)) { - err = PTR_ERR(out_skb); - goto out; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + if (hdr->sadb_msg_type == SADB_X_SPDDELETE2) { + c.data.byid = 1; + c.event = XFRM_MSG_DELPOLICY; + km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); + } else { + err = key_pol_get_resp(sk, xp, hdr, pol->sadb_x_policy_dir-1); } - pfkey_xfrm_policy2msg(out_skb, xp, pol->sadb_x_policy_dir-1); - out_hdr = (struct sadb_msg *) out_skb->data; - out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = hdr->sadb_msg_type; - out_hdr->sadb_msg_satype = 0; - out_hdr->sadb_msg_errno = 0; - out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; - out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); - err = 0; - -out: xfrm_pol_put(xp); return err; } @@ -2102,22 +2212,34 @@ static int pfkey_spddump(struct sock *sk, struct sk_buff *skb, struct sadb_msg * return xfrm_policy_walk(dump_sp, &data); } -static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) +static int key_notify_policy_flush(struct km_event *c) { struct sk_buff *skb_out; - struct sadb_msg *hdr_out; + struct sadb_msg *hdr; - skb_out = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_KERNEL); + skb_out = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_ATOMIC); if (!skb_out) return -ENOBUFS; + hdr = (struct sadb_msg *) skb_put(skb_out, sizeof(struct sadb_msg)); + hdr->sadb_msg_seq = c->seq; + hdr->sadb_msg_pid = c->pid; + hdr->sadb_msg_version = PF_KEY_V2; + hdr->sadb_msg_errno = (uint8_t) 0; + hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); + pfkey_broadcast(skb_out, GFP_ATOMIC, BROADCAST_ALL, NULL); + return 0; - xfrm_policy_flush(); +} - hdr_out = (struct sadb_msg *) skb_put(skb_out, sizeof(struct sadb_msg)); - pfkey_hdr_dup(hdr_out, hdr); - hdr_out->sadb_msg_errno = (uint8_t) 0; - hdr_out->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); - pfkey_broadcast(skb_out, GFP_KERNEL, BROADCAST_ALL, NULL); +static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) +{ + struct km_event c; + + xfrm_policy_flush(); + c.event = XFRM_MSG_FLUSHPOLICY; + c.pid = hdr->sadb_msg_pid; + c.seq = hdr->sadb_msg_seq; + km_policy_notify(NULL, 0, &c); return 0; } @@ -2317,11 +2439,23 @@ static void dump_esp_combs(struct sk_buff *skb, struct xfrm_tmpl *t) } } -static int pfkey_send_notify(struct xfrm_state *x, int hard) +static int key_notify_policy_expire(struct xfrm_policy *xp, struct km_event *c) +{ + return 0; +} + +static int key_notify_sa_expire(struct xfrm_state *x, struct km_event *c) { struct sk_buff *out_skb; struct sadb_msg *out_hdr; - int hsc = (hard ? 2 : 1); + int hard; + int hsc; + + hard = c->data.hard; + if (hard) + hsc = 2; + else + hsc = 1; out_skb = pfkey_xfrm_state2msg(x, 0, hsc); if (IS_ERR(out_skb)) @@ -2340,6 +2474,44 @@ static int pfkey_send_notify(struct xfrm_state *x, int hard) return 0; } +static int pfkey_send_notify(struct xfrm_state *x, struct km_event *c) +{ + switch (c->event) { + case XFRM_MSG_EXPIRE: + return key_notify_sa_expire(x, c); + case XFRM_MSG_DELSA: + case XFRM_MSG_NEWSA: + case XFRM_MSG_UPDSA: + return key_notify_sa(x, c); + case XFRM_MSG_FLUSHSA: + return key_notify_sa_flush(c); + default: + printk("pfkey: Unknown SA event %d\n", c->event); + break; + } + + return 0; +} + +static int pfkey_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + switch (c->event) { + case XFRM_MSG_POLEXPIRE: + return key_notify_policy_expire(xp, c); + case XFRM_MSG_DELPOLICY: + case XFRM_MSG_NEWPOLICY: + case XFRM_MSG_UPDPOLICY: + return key_notify_policy(xp, dir, c); + case XFRM_MSG_FLUSHPOLICY: + return key_notify_policy_flush(c); + default: + printk("pfkey: Unknown policy event %d\n", c->event); + break; + } + + return 0; +} + static u32 get_acqseq(void) { u32 res; @@ -2856,6 +3028,7 @@ static struct xfrm_mgr pfkeyv2_mgr = .acquire = pfkey_send_acquire, .compile_policy = pfkey_compile_policy, .new_mapping = pfkey_send_new_mapping, + .notify_policy = pfkey_send_policy_notify, }; static void __exit ipsec_pfkey_exit(void) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 733bf52..70bcd47 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -735,11 +735,15 @@ static inline int do_one_broadcast(struct sock *sk, sock_hold(sk); if (p->skb2 == NULL) { - if (atomic_read(&p->skb->users) != 1) { + if (skb_shared(p->skb)) { p->skb2 = skb_clone(p->skb, p->allocation); } else { - p->skb2 = p->skb; - atomic_inc(&p->skb->users); + p->skb2 = skb_get(p->skb); + /* + * skb ownership may have been set when + * delivered to a previous socket. + */ + skb_orphan(p->skb2); } } if (p->skb2 == NULL) { @@ -785,11 +789,12 @@ int netlink_broadcast(struct sock *ssk, struct sk_buff *skb, u32 pid, sk_for_each_bound(sk, node, &nl_table[ssk->sk_protocol].mc_list) do_one_broadcast(sk, &info); + kfree_skb(skb); + netlink_unlock_table(); if (info.skb2) kfree_skb(info.skb2); - kfree_skb(skb); if (info.delivered) { if (info.congested && (allocation & __GFP_WAIT)) @@ -1090,8 +1095,7 @@ static int netlink_dump(struct sock *sk) return 0; } - nlh = __nlmsg_put(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, NLMSG_DONE, sizeof(int)); - nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW_ANSWER(skb, cb, NLMSG_DONE, sizeof(len), NLM_F_MULTI); memcpy(NLMSG_DATA(nlh), &len, sizeof(len)); skb_queue_tail(&sk->sk_receive_queue, skb); sk->sk_data_ready(sk, skb->len); @@ -1102,6 +1106,9 @@ static int netlink_dump(struct sock *sk) netlink_destroy_callback(cb); return 0; + +nlmsg_failure: + return -ENOBUFS; } int netlink_dump_start(struct sock *ssk, struct sk_buff *skb, @@ -1173,7 +1180,7 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err) } rep = __nlmsg_put(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, - NLMSG_ERROR, sizeof(struct nlmsgerr)); + NLMSG_ERROR, sizeof(struct nlmsgerr), 0); errmsg = NLMSG_DATA(rep); errmsg->error = err; memcpy(&errmsg->msg, nlh, err ? nlh->nlmsg_len : sizeof(struct nlmsghdr)); diff --git a/net/rxrpc/krxiod.c b/net/rxrpc/krxiod.c index 2b537f4..dada34a 100644 --- a/net/rxrpc/krxiod.c +++ b/net/rxrpc/krxiod.c @@ -138,7 +138,7 @@ static int rxrpc_krxiod(void *arg) _debug("### End Work"); - try_to_freeze(PF_FREEZE); + try_to_freeze(); /* discard pending signals */ rxrpc_discard_my_signals(); diff --git a/net/rxrpc/krxsecd.c b/net/rxrpc/krxsecd.c index 6020c89..1aadd02 100644 --- a/net/rxrpc/krxsecd.c +++ b/net/rxrpc/krxsecd.c @@ -107,7 +107,7 @@ static int rxrpc_krxsecd(void *arg) _debug("### End Inbound Calls"); - try_to_freeze(PF_FREEZE); + try_to_freeze(); /* discard pending signals */ rxrpc_discard_my_signals(); diff --git a/net/rxrpc/krxtimod.c b/net/rxrpc/krxtimod.c index 249c2b0..3ac81cd 100644 --- a/net/rxrpc/krxtimod.c +++ b/net/rxrpc/krxtimod.c @@ -90,7 +90,7 @@ static int krxtimod(void *arg) complete_and_exit(&krxtimod_dead, 0); } - try_to_freeze(PF_FREEZE); + try_to_freeze(); /* discard pending signals */ rxrpc_discard_my_signals(); diff --git a/net/sched/Kconfig b/net/sched/Kconfig index b094118..7bac249 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -405,7 +405,7 @@ config NET_EMATCH_STACK ---help--- Size of the local stack variable used while evaluating the tree of ematches. Limits the depth of the tree, i.e. the number of - encapsulated precedences. Every level requires 4 bytes of addtional + encapsulated precedences. Every level requires 4 bytes of additional stack space. config NET_EMATCH_CMP @@ -449,6 +449,19 @@ config NET_EMATCH_META To compile this code as a module, choose M here: the module will be called em_meta. +config NET_EMATCH_TEXT + tristate "Textsearch" + depends on NET_EMATCH + select TEXTSEARCH + select TEXTSEARCH_KMP + select TEXTSEARCH_FSM + ---help--- + Say Y here if you want to be ablt to classify packets based on + textsearch comparisons. + + To compile this code as a module, choose M here: the + module will be called em_text. + config NET_CLS_ACT bool "Packet ACTION" depends on EXPERIMENTAL && NET_CLS && NET_QOS diff --git a/net/sched/Makefile b/net/sched/Makefile index eb3fe58..8f58cec 100644 --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -40,3 +40,4 @@ obj-$(CONFIG_NET_EMATCH_CMP) += em_cmp.o obj-$(CONFIG_NET_EMATCH_NBYTE) += em_nbyte.o obj-$(CONFIG_NET_EMATCH_U32) += em_u32.o obj-$(CONFIG_NET_EMATCH_META) += em_meta.o +obj-$(CONFIG_NET_EMATCH_TEXT) += em_text.o diff --git a/net/sched/act_api.c b/net/sched/act_api.c index cafcb08..9594206 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -428,15 +428,15 @@ errout: static int tca_get_fill(struct sk_buff *skb, struct tc_action *a, u32 pid, u32 seq, - unsigned flags, int event, int bind, int ref) + u16 flags, int event, int bind, int ref) { struct tcamsg *t; struct nlmsghdr *nlh; unsigned char *b = skb->tail; struct rtattr *x; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*t)); - nlh->nlmsg_flags = flags; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*t), flags); + t = NLMSG_DATA(nlh); t->tca_family = AF_UNSPEC; @@ -669,7 +669,7 @@ err: } static int tcf_add_notify(struct tc_action *a, u32 pid, u32 seq, int event, - unsigned flags) + u16 flags) { struct tcamsg *t; struct nlmsghdr *nlh; @@ -684,8 +684,7 @@ static int tcf_add_notify(struct tc_action *a, u32 pid, u32 seq, int event, b = (unsigned char *)skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*t)); - nlh->nlmsg_flags = flags; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*t), flags); t = NLMSG_DATA(nlh); t->tca_family = AF_UNSPEC; @@ -881,7 +880,7 @@ static int __init tc_action_init(void) link_p[RTM_GETACTION-RTM_BASE].dumpit = tc_dump_action; } - printk("TC classifier action (bugs to netdev@oss.sgi.com cc " + printk("TC classifier action (bugs to netdev@vger.kernel.org cc " "hadi@cyberus.ca)\n"); return 0; } diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 56e66c3..1616bf5 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -322,14 +322,13 @@ errout: static int tcf_fill_node(struct sk_buff *skb, struct tcf_proto *tp, unsigned long fh, - u32 pid, u32 seq, unsigned flags, int event) + u32 pid, u32 seq, u16 flags, int event) { struct tcmsg *tcm; struct nlmsghdr *nlh; unsigned char *b = skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*tcm)); - nlh->nlmsg_flags = flags; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*tcm), flags); tcm = NLMSG_DATA(nlh); tcm->tcm_family = AF_UNSPEC; tcm->tcm_ifindex = tp->q->dev->ifindex; diff --git a/net/sched/cls_basic.c b/net/sched/cls_basic.c index 0d2d441..dfb300b 100644 --- a/net/sched/cls_basic.c +++ b/net/sched/cls_basic.c @@ -261,6 +261,9 @@ static int basic_dump(struct tcf_proto *tp, unsigned long fh, rta = (struct rtattr *) b; RTA_PUT(skb, TCA_OPTIONS, 0, NULL); + if (f->res.classid) + RTA_PUT(skb, TCA_BASIC_CLASSID, sizeof(u32), &f->res.classid); + if (tcf_exts_dump(skb, &f->exts, &basic_ext_map) < 0 || tcf_em_tree_dump(skb, &f->ematches, TCA_BASIC_EMATCHES) < 0) goto rtattr_failure; diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index f1eeaf6..48bb23c 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -32,7 +32,7 @@ * +-----------+ +-----------+ * | | * ---> meta_ops[INT][INDEV](...) | - * | | + * | | * ----------- | * V V * +-----------+ +-----------+ @@ -70,6 +70,7 @@ #include <net/dst.h> #include <net/route.h> #include <net/pkt_cls.h> +#include <net/sock.h> struct meta_obj { @@ -284,6 +285,214 @@ META_COLLECTOR(int_rtiif) } /************************************************************************** + * Socket Attributes + **************************************************************************/ + +#define SKIP_NONLOCAL(skb) \ + if (unlikely(skb->sk == NULL)) { \ + *err = -1; \ + return; \ + } + +META_COLLECTOR(int_sk_family) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_family; +} + +META_COLLECTOR(int_sk_state) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_state; +} + +META_COLLECTOR(int_sk_reuse) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_reuse; +} + +META_COLLECTOR(int_sk_bound_if) +{ + SKIP_NONLOCAL(skb); + /* No error if bound_dev_if is 0, legal userspace check */ + dst->value = skb->sk->sk_bound_dev_if; +} + +META_COLLECTOR(var_sk_bound_if) +{ + SKIP_NONLOCAL(skb); + + if (skb->sk->sk_bound_dev_if == 0) { + dst->value = (unsigned long) "any"; + dst->len = 3; + } else { + struct net_device *dev; + + dev = dev_get_by_index(skb->sk->sk_bound_dev_if); + *err = var_dev(dev, dst); + if (dev) + dev_put(dev); + } +} + +META_COLLECTOR(int_sk_refcnt) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_refcnt); +} + +META_COLLECTOR(int_sk_rcvbuf) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_rcvbuf; +} + +META_COLLECTOR(int_sk_shutdown) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_shutdown; +} + +META_COLLECTOR(int_sk_proto) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_protocol; +} + +META_COLLECTOR(int_sk_type) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_type; +} + +META_COLLECTOR(int_sk_rmem_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_rmem_alloc); +} + +META_COLLECTOR(int_sk_wmem_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_wmem_alloc); +} + +META_COLLECTOR(int_sk_omem_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_omem_alloc); +} + +META_COLLECTOR(int_sk_rcv_qlen) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_receive_queue.qlen; +} + +META_COLLECTOR(int_sk_snd_qlen) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_write_queue.qlen; +} + +META_COLLECTOR(int_sk_wmem_queued) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_wmem_queued; +} + +META_COLLECTOR(int_sk_fwd_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_forward_alloc; +} + +META_COLLECTOR(int_sk_sndbuf) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_sndbuf; +} + +META_COLLECTOR(int_sk_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_allocation; +} + +META_COLLECTOR(int_sk_route_caps) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_route_caps; +} + +META_COLLECTOR(int_sk_hashent) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_hashent; +} + +META_COLLECTOR(int_sk_lingertime) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_lingertime / HZ; +} + +META_COLLECTOR(int_sk_err_qlen) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_error_queue.qlen; +} + +META_COLLECTOR(int_sk_ack_bl) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_ack_backlog; +} + +META_COLLECTOR(int_sk_max_ack_bl) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_max_ack_backlog; +} + +META_COLLECTOR(int_sk_prio) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_priority; +} + +META_COLLECTOR(int_sk_rcvlowat) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_rcvlowat; +} + +META_COLLECTOR(int_sk_rcvtimeo) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_rcvtimeo / HZ; +} + +META_COLLECTOR(int_sk_sndtimeo) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_sndtimeo / HZ; +} + +META_COLLECTOR(int_sk_sendmsg_off) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_sndmsg_off; +} + +META_COLLECTOR(int_sk_write_pend) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_write_pending; +} + +/************************************************************************** * Meta value collectors assignment table **************************************************************************/ @@ -293,41 +502,75 @@ struct meta_ops struct meta_value *, struct meta_obj *, int *); }; +#define META_ID(name) TCF_META_ID_##name +#define META_FUNC(name) { .get = meta_##name } + /* Meta value operations table listing all meta value collectors and * assigns them to a type and meta id. */ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [TCF_META_TYPE_VAR] = { - [TCF_META_ID_DEV] = { .get = meta_var_dev }, - [TCF_META_ID_INDEV] = { .get = meta_var_indev }, - [TCF_META_ID_REALDEV] = { .get = meta_var_realdev } + [META_ID(DEV)] = META_FUNC(var_dev), + [META_ID(INDEV)] = META_FUNC(var_indev), + [META_ID(REALDEV)] = META_FUNC(var_realdev), + [META_ID(SK_BOUND_IF)] = META_FUNC(var_sk_bound_if), }, [TCF_META_TYPE_INT] = { - [TCF_META_ID_RANDOM] = { .get = meta_int_random }, - [TCF_META_ID_LOADAVG_0] = { .get = meta_int_loadavg_0 }, - [TCF_META_ID_LOADAVG_1] = { .get = meta_int_loadavg_1 }, - [TCF_META_ID_LOADAVG_2] = { .get = meta_int_loadavg_2 }, - [TCF_META_ID_DEV] = { .get = meta_int_dev }, - [TCF_META_ID_INDEV] = { .get = meta_int_indev }, - [TCF_META_ID_REALDEV] = { .get = meta_int_realdev }, - [TCF_META_ID_PRIORITY] = { .get = meta_int_priority }, - [TCF_META_ID_PROTOCOL] = { .get = meta_int_protocol }, - [TCF_META_ID_SECURITY] = { .get = meta_int_security }, - [TCF_META_ID_PKTTYPE] = { .get = meta_int_pkttype }, - [TCF_META_ID_PKTLEN] = { .get = meta_int_pktlen }, - [TCF_META_ID_DATALEN] = { .get = meta_int_datalen }, - [TCF_META_ID_MACLEN] = { .get = meta_int_maclen }, + [META_ID(RANDOM)] = META_FUNC(int_random), + [META_ID(LOADAVG_0)] = META_FUNC(int_loadavg_0), + [META_ID(LOADAVG_1)] = META_FUNC(int_loadavg_1), + [META_ID(LOADAVG_2)] = META_FUNC(int_loadavg_2), + [META_ID(DEV)] = META_FUNC(int_dev), + [META_ID(INDEV)] = META_FUNC(int_indev), + [META_ID(REALDEV)] = META_FUNC(int_realdev), + [META_ID(PRIORITY)] = META_FUNC(int_priority), + [META_ID(PROTOCOL)] = META_FUNC(int_protocol), + [META_ID(SECURITY)] = META_FUNC(int_security), + [META_ID(PKTTYPE)] = META_FUNC(int_pkttype), + [META_ID(PKTLEN)] = META_FUNC(int_pktlen), + [META_ID(DATALEN)] = META_FUNC(int_datalen), + [META_ID(MACLEN)] = META_FUNC(int_maclen), #ifdef CONFIG_NETFILTER - [TCF_META_ID_NFMARK] = { .get = meta_int_nfmark }, + [META_ID(NFMARK)] = META_FUNC(int_nfmark), #endif - [TCF_META_ID_TCINDEX] = { .get = meta_int_tcindex }, + [META_ID(TCINDEX)] = META_FUNC(int_tcindex), #ifdef CONFIG_NET_CLS_ACT - [TCF_META_ID_TCVERDICT] = { .get = meta_int_tcverd }, - [TCF_META_ID_TCCLASSID] = { .get = meta_int_tcclassid }, + [META_ID(TCVERDICT)] = META_FUNC(int_tcverd), + [META_ID(TCCLASSID)] = META_FUNC(int_tcclassid), #endif #ifdef CONFIG_NET_CLS_ROUTE - [TCF_META_ID_RTCLASSID] = { .get = meta_int_rtclassid }, + [META_ID(RTCLASSID)] = META_FUNC(int_rtclassid), #endif - [TCF_META_ID_RTIIF] = { .get = meta_int_rtiif } + [META_ID(RTIIF)] = META_FUNC(int_rtiif), + [META_ID(SK_FAMILY)] = META_FUNC(int_sk_family), + [META_ID(SK_STATE)] = META_FUNC(int_sk_state), + [META_ID(SK_REUSE)] = META_FUNC(int_sk_reuse), + [META_ID(SK_BOUND_IF)] = META_FUNC(int_sk_bound_if), + [META_ID(SK_REFCNT)] = META_FUNC(int_sk_refcnt), + [META_ID(SK_RCVBUF)] = META_FUNC(int_sk_rcvbuf), + [META_ID(SK_SNDBUF)] = META_FUNC(int_sk_sndbuf), + [META_ID(SK_SHUTDOWN)] = META_FUNC(int_sk_shutdown), + [META_ID(SK_PROTO)] = META_FUNC(int_sk_proto), + [META_ID(SK_TYPE)] = META_FUNC(int_sk_type), + [META_ID(SK_RMEM_ALLOC)] = META_FUNC(int_sk_rmem_alloc), + [META_ID(SK_WMEM_ALLOC)] = META_FUNC(int_sk_wmem_alloc), + [META_ID(SK_OMEM_ALLOC)] = META_FUNC(int_sk_omem_alloc), + [META_ID(SK_WMEM_QUEUED)] = META_FUNC(int_sk_wmem_queued), + [META_ID(SK_RCV_QLEN)] = META_FUNC(int_sk_rcv_qlen), + [META_ID(SK_SND_QLEN)] = META_FUNC(int_sk_snd_qlen), + [META_ID(SK_ERR_QLEN)] = META_FUNC(int_sk_err_qlen), + [META_ID(SK_FORWARD_ALLOCS)] = META_FUNC(int_sk_fwd_alloc), + [META_ID(SK_ALLOCS)] = META_FUNC(int_sk_alloc), + [META_ID(SK_ROUTE_CAPS)] = META_FUNC(int_sk_route_caps), + [META_ID(SK_HASHENT)] = META_FUNC(int_sk_hashent), + [META_ID(SK_LINGERTIME)] = META_FUNC(int_sk_lingertime), + [META_ID(SK_ACK_BACKLOG)] = META_FUNC(int_sk_ack_bl), + [META_ID(SK_MAX_ACK_BACKLOG)] = META_FUNC(int_sk_max_ack_bl), + [META_ID(SK_PRIO)] = META_FUNC(int_sk_prio), + [META_ID(SK_RCVLOWAT)] = META_FUNC(int_sk_rcvlowat), + [META_ID(SK_RCVTIMEO)] = META_FUNC(int_sk_rcvtimeo), + [META_ID(SK_SNDTIMEO)] = META_FUNC(int_sk_sndtimeo), + [META_ID(SK_SENDMSG_OFF)] = META_FUNC(int_sk_sendmsg_off), + [META_ID(SK_WRITE_PENDING)] = META_FUNC(int_sk_write_pend), } }; @@ -396,9 +639,9 @@ static int meta_int_compare(struct meta_obj *a, struct meta_obj *b) /* Let gcc optimize it, the unlikely is not really based on * some numbers but jump free code for mismatches seems * more logical. */ - if (unlikely(a == b)) + if (unlikely(a->value == b->value)) return 0; - else if (a < b) + else if (a->value < b->value) return -1; else return 1; diff --git a/net/sched/em_text.c b/net/sched/em_text.c new file mode 100644 index 0000000..873840d --- /dev/null +++ b/net/sched/em_text.c @@ -0,0 +1,157 @@ +/* + * net/sched/em_text.c Textsearch ematch + * + * 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. + * + * Authors: Thomas Graf <tgraf@suug.ch> + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/string.h> +#include <linux/skbuff.h> +#include <linux/textsearch.h> +#include <linux/tc_ematch/tc_em_text.h> +#include <net/pkt_cls.h> + +struct text_match +{ + u16 from_offset; + u16 to_offset; + u8 from_layer; + u8 to_layer; + struct ts_config *config; +}; + +#define EM_TEXT_PRIV(m) ((struct text_match *) (m)->data) + +static int em_text_match(struct sk_buff *skb, struct tcf_ematch *m, + struct tcf_pkt_info *info) +{ + struct text_match *tm = EM_TEXT_PRIV(m); + int from, to; + struct ts_state state; + + from = tcf_get_base_ptr(skb, tm->from_layer) - skb->data; + from += tm->from_offset; + + to = tcf_get_base_ptr(skb, tm->to_layer) - skb->data; + to += tm->to_offset; + + return skb_find_text(skb, from, to, tm->config, &state) != UINT_MAX; +} + +static int em_text_change(struct tcf_proto *tp, void *data, int len, + struct tcf_ematch *m) +{ + struct text_match *tm; + struct tcf_em_text *conf = data; + struct ts_config *ts_conf; + int flags = 0; + + printk("Configuring text: %s from %d:%d to %d:%d len %d\n", conf->algo, conf->from_offset, + conf->from_layer, conf->to_offset, conf->to_layer, conf->pattern_len); + + if (len < sizeof(*conf) || len < (sizeof(*conf) + conf->pattern_len)) + return -EINVAL; + + if (conf->from_layer > conf->to_layer) + return -EINVAL; + + if (conf->from_layer == conf->to_layer && + conf->from_offset > conf->to_offset) + return -EINVAL; + +retry: + ts_conf = textsearch_prepare(conf->algo, (u8 *) conf + sizeof(*conf), + conf->pattern_len, GFP_KERNEL, flags); + + if (flags & TS_AUTOLOAD) + rtnl_lock(); + + if (IS_ERR(ts_conf)) { + if (PTR_ERR(ts_conf) == -ENOENT && !(flags & TS_AUTOLOAD)) { + rtnl_unlock(); + flags |= TS_AUTOLOAD; + goto retry; + } else + return PTR_ERR(ts_conf); + } else if (flags & TS_AUTOLOAD) { + textsearch_destroy(ts_conf); + return -EAGAIN; + } + + tm = kmalloc(sizeof(*tm), GFP_KERNEL); + if (tm == NULL) { + textsearch_destroy(ts_conf); + return -ENOBUFS; + } + + tm->from_offset = conf->from_offset; + tm->to_offset = conf->to_offset; + tm->from_layer = conf->from_layer; + tm->to_layer = conf->to_layer; + tm->config = ts_conf; + + m->datalen = sizeof(*tm); + m->data = (unsigned long) tm; + + return 0; +} + +static void em_text_destroy(struct tcf_proto *tp, struct tcf_ematch *m) +{ + textsearch_destroy(EM_TEXT_PRIV(m)->config); +} + +static int em_text_dump(struct sk_buff *skb, struct tcf_ematch *m) +{ + struct text_match *tm = EM_TEXT_PRIV(m); + struct tcf_em_text conf; + + strncpy(conf.algo, tm->config->ops->name, sizeof(conf.algo) - 1); + conf.from_offset = tm->from_offset; + conf.to_offset = tm->to_offset; + conf.from_layer = tm->from_layer; + conf.to_layer = tm->to_layer; + conf.pattern_len = textsearch_get_pattern_len(tm->config); + conf.pad = 0; + + RTA_PUT_NOHDR(skb, sizeof(conf), &conf); + RTA_APPEND(skb, conf.pattern_len, textsearch_get_pattern(tm->config)); + return 0; + +rtattr_failure: + return -1; +} + +static struct tcf_ematch_ops em_text_ops = { + .kind = TCF_EM_TEXT, + .change = em_text_change, + .match = em_text_match, + .destroy = em_text_destroy, + .dump = em_text_dump, + .owner = THIS_MODULE, + .link = LIST_HEAD_INIT(em_text_ops.link) +}; + +static int __init init_em_text(void) +{ + return tcf_em_register(&em_text_ops); +} + +static void __exit exit_em_text(void) +{ + tcf_em_unregister(&em_text_ops); +} + +MODULE_LICENSE("GPL"); + +module_init(init_em_text); +module_exit(exit_em_text); diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 07977f8..97c1c75 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -760,15 +760,14 @@ graft: } static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid, - u32 pid, u32 seq, unsigned flags, int event) + u32 pid, u32 seq, u16 flags, int event) { struct tcmsg *tcm; struct nlmsghdr *nlh; unsigned char *b = skb->tail; struct gnet_dump d; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*tcm)); - nlh->nlmsg_flags = flags; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*tcm), flags); tcm = NLMSG_DATA(nlh); tcm->tcm_family = AF_UNSPEC; tcm->tcm_ifindex = q->dev->ifindex; @@ -997,7 +996,7 @@ out: static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q, unsigned long cl, - u32 pid, u32 seq, unsigned flags, int event) + u32 pid, u32 seq, u16 flags, int event) { struct tcmsg *tcm; struct nlmsghdr *nlh; @@ -1005,8 +1004,7 @@ static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q, struct gnet_dump d; struct Qdisc_class_ops *cl_ops = q->ops->cl_ops; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*tcm)); - nlh->nlmsg_flags = flags; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*tcm), flags); tcm = NLMSG_DATA(nlh); tcm->tcm_family = AF_UNSPEC; tcm->tcm_ifindex = q->dev->ifindex; diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c index 8a3db9d..13e0e7b 100644 --- a/net/sched/sch_dsmark.c +++ b/net/sched/sch_dsmark.c @@ -18,7 +18,7 @@ #include <asm/byteorder.h> -#if 1 /* control */ +#if 0 /* control */ #define DPRINTK(format,args...) printk(KERN_DEBUG format,##args) #else #define DPRINTK(format,args...) @@ -31,7 +31,7 @@ #endif -#define PRIV(sch) qdisc_priv(sch) +#define PRIV(sch) ((struct dsmark_qdisc_data *) qdisc_priv(sch)) /* @@ -55,145 +55,163 @@ struct dsmark_qdisc_data { struct Qdisc *q; struct tcf_proto *filter_list; - __u8 *mask; /* "owns" the array */ - __u8 *value; - __u16 indices; - __u32 default_index; /* index range is 0...0xffff */ + u8 *mask; /* "owns" the array */ + u8 *value; + u16 indices; + u32 default_index; /* index range is 0...0xffff */ int set_tc_index; }; +static inline int dsmark_valid_indices(u16 indices) +{ + while (indices != 1) { + if (indices & 1) + return 0; + indices >>= 1; + } + + return 1; +} -/* ------------------------- Class/flow operations ------------------------- */ +static inline int dsmark_valid_index(struct dsmark_qdisc_data *p, u16 index) +{ + return (index <= p->indices && index > 0); +} +/* ------------------------- Class/flow operations ------------------------- */ -static int dsmark_graft(struct Qdisc *sch,unsigned long arg, - struct Qdisc *new,struct Qdisc **old) +static int dsmark_graft(struct Qdisc *sch, unsigned long arg, + struct Qdisc *new, struct Qdisc **old) { struct dsmark_qdisc_data *p = PRIV(sch); - DPRINTK("dsmark_graft(sch %p,[qdisc %p],new %p,old %p)\n",sch,p,new, - old); - if (!new) - new = &noop_qdisc; + DPRINTK("dsmark_graft(sch %p,[qdisc %p],new %p,old %p)\n", + sch, p, new, old); + + if (new == NULL) { + new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); + if (new == NULL) + new = &noop_qdisc; + } + sch_tree_lock(sch); - *old = xchg(&p->q,new); - if (*old) - qdisc_reset(*old); + *old = xchg(&p->q, new); + qdisc_reset(*old); sch->q.qlen = 0; - sch_tree_unlock(sch); /* @@@ move up ? */ + sch_tree_unlock(sch); + return 0; } - static struct Qdisc *dsmark_leaf(struct Qdisc *sch, unsigned long arg) { - struct dsmark_qdisc_data *p = PRIV(sch); - - return p->q; + return PRIV(sch)->q; } - -static unsigned long dsmark_get(struct Qdisc *sch,u32 classid) +static unsigned long dsmark_get(struct Qdisc *sch, u32 classid) { - struct dsmark_qdisc_data *p __attribute__((unused)) = PRIV(sch); + DPRINTK("dsmark_get(sch %p,[qdisc %p],classid %x)\n", + sch, PRIV(sch), classid); - DPRINTK("dsmark_get(sch %p,[qdisc %p],classid %x)\n",sch,p,classid); - return TC_H_MIN(classid)+1; + return TC_H_MIN(classid) + 1; } - static unsigned long dsmark_bind_filter(struct Qdisc *sch, - unsigned long parent, u32 classid) + unsigned long parent, u32 classid) { - return dsmark_get(sch,classid); + return dsmark_get(sch, classid); } - static void dsmark_put(struct Qdisc *sch, unsigned long cl) { } - static int dsmark_change(struct Qdisc *sch, u32 classid, u32 parent, - struct rtattr **tca, unsigned long *arg) + struct rtattr **tca, unsigned long *arg) { struct dsmark_qdisc_data *p = PRIV(sch); struct rtattr *opt = tca[TCA_OPTIONS-1]; struct rtattr *tb[TCA_DSMARK_MAX]; + int err = -EINVAL; + u8 mask = 0; DPRINTK("dsmark_change(sch %p,[qdisc %p],classid %x,parent %x)," - "arg 0x%lx\n",sch,p,classid,parent,*arg); - if (*arg > p->indices) - return -ENOENT; - if (!opt || rtattr_parse_nested(tb, TCA_DSMARK_MAX, opt)) - return -EINVAL; - if (tb[TCA_DSMARK_MASK-1]) { - if (!RTA_PAYLOAD(tb[TCA_DSMARK_MASK-1])) - return -EINVAL; - p->mask[*arg-1] = *(__u8 *) RTA_DATA(tb[TCA_DSMARK_MASK-1]); - } - if (tb[TCA_DSMARK_VALUE-1]) { - if (!RTA_PAYLOAD(tb[TCA_DSMARK_VALUE-1])) - return -EINVAL; - p->value[*arg-1] = *(__u8 *) RTA_DATA(tb[TCA_DSMARK_VALUE-1]); + "arg 0x%lx\n", sch, p, classid, parent, *arg); + + if (!dsmark_valid_index(p, *arg)) { + err = -ENOENT; + goto rtattr_failure; } - return 0; -} + if (!opt || rtattr_parse_nested(tb, TCA_DSMARK_MAX, opt)) + goto rtattr_failure; + + if (tb[TCA_DSMARK_MASK-1]) + mask = RTA_GET_U8(tb[TCA_DSMARK_MASK-1]); + + if (tb[TCA_DSMARK_VALUE-1]) + p->value[*arg-1] = RTA_GET_U8(tb[TCA_DSMARK_VALUE-1]); + + if (tb[TCA_DSMARK_MASK-1]) + p->mask[*arg-1] = mask; -static int dsmark_delete(struct Qdisc *sch,unsigned long arg) + err = 0; + +rtattr_failure: + return err; +} + +static int dsmark_delete(struct Qdisc *sch, unsigned long arg) { struct dsmark_qdisc_data *p = PRIV(sch); - if (!arg || arg > p->indices) + if (!dsmark_valid_index(p, arg)) return -EINVAL; + p->mask[arg-1] = 0xff; p->value[arg-1] = 0; + return 0; } - static void dsmark_walk(struct Qdisc *sch,struct qdisc_walker *walker) { struct dsmark_qdisc_data *p = PRIV(sch); int i; - DPRINTK("dsmark_walk(sch %p,[qdisc %p],walker %p)\n",sch,p,walker); + DPRINTK("dsmark_walk(sch %p,[qdisc %p],walker %p)\n", sch, p, walker); + if (walker->stop) return; + for (i = 0; i < p->indices; i++) { if (p->mask[i] == 0xff && !p->value[i]) - continue; + goto ignore; if (walker->count >= walker->skip) { if (walker->fn(sch, i+1, walker) < 0) { walker->stop = 1; break; } } - walker->count++; +ignore: + walker->count++; } } - static struct tcf_proto **dsmark_find_tcf(struct Qdisc *sch,unsigned long cl) { - struct dsmark_qdisc_data *p = PRIV(sch); - - return &p->filter_list; + return &PRIV(sch)->filter_list; } - /* --------------------------- Qdisc operations ---------------------------- */ - static int dsmark_enqueue(struct sk_buff *skb,struct Qdisc *sch) { struct dsmark_qdisc_data *p = PRIV(sch); - struct tcf_result res; - int result; - int ret = NET_XMIT_POLICED; + int err; + + D2PRINTK("dsmark_enqueue(skb %p,sch %p,[qdisc %p])\n", skb, sch, p); - D2PRINTK("dsmark_enqueue(skb %p,sch %p,[qdisc %p])\n",skb,sch,p); if (p->set_tc_index) { /* FIXME: Safe with non-linear skbs? --RR */ switch (skb->protocol) { @@ -210,17 +228,21 @@ static int dsmark_enqueue(struct sk_buff *skb,struct Qdisc *sch) break; }; } - result = TC_POLICE_OK; /* be nice to gcc */ - if (TC_H_MAJ(skb->priority) == sch->handle) { + + if (TC_H_MAJ(skb->priority) == sch->handle) skb->tc_index = TC_H_MIN(skb->priority); - } else { - result = tc_classify(skb,p->filter_list,&res); - D2PRINTK("result %d class 0x%04x\n",result,res.classid); + else { + struct tcf_result res; + int result = tc_classify(skb, p->filter_list, &res); + + D2PRINTK("result %d class 0x%04x\n", result, res.classid); + switch (result) { #ifdef CONFIG_NET_CLS_POLICE case TC_POLICE_SHOT: kfree_skb(skb); - break; + sch->qstats.drops++; + return NET_XMIT_POLICED; #if 0 case TC_POLICE_RECLASSIFY: /* FIXME: what to do here ??? */ @@ -237,43 +259,45 @@ static int dsmark_enqueue(struct sk_buff *skb,struct Qdisc *sch) break; }; } - if ( -#ifdef CONFIG_NET_CLS_POLICE - result == TC_POLICE_SHOT || -#endif - ((ret = p->q->enqueue(skb,p->q)) != 0)) { + err = p->q->enqueue(skb,p->q); + if (err != NET_XMIT_SUCCESS) { sch->qstats.drops++; - return ret; + return err; } + sch->bstats.bytes += skb->len; sch->bstats.packets++; sch->q.qlen++; - return ret; -} + return NET_XMIT_SUCCESS; +} static struct sk_buff *dsmark_dequeue(struct Qdisc *sch) { struct dsmark_qdisc_data *p = PRIV(sch); struct sk_buff *skb; - int index; + u32 index; + + D2PRINTK("dsmark_dequeue(sch %p,[qdisc %p])\n", sch, p); - D2PRINTK("dsmark_dequeue(sch %p,[qdisc %p])\n",sch,p); skb = p->q->ops->dequeue(p->q); - if (!skb) + if (skb == NULL) return NULL; + sch->q.qlen--; - index = skb->tc_index & (p->indices-1); - D2PRINTK("index %d->%d\n",skb->tc_index,index); + + index = skb->tc_index & (p->indices - 1); + D2PRINTK("index %d->%d\n", skb->tc_index, index); + switch (skb->protocol) { case __constant_htons(ETH_P_IP): - ipv4_change_dsfield(skb->nh.iph, - p->mask[index],p->value[index]); + ipv4_change_dsfield(skb->nh.iph, p->mask[index], + p->value[index]); break; case __constant_htons(ETH_P_IPV6): - ipv6_change_dsfield(skb->nh.ipv6h, - p->mask[index],p->value[index]); + ipv6_change_dsfield(skb->nh.ipv6h, p->mask[index], + p->value[index]); break; default: /* @@ -287,152 +311,162 @@ static struct sk_buff *dsmark_dequeue(struct Qdisc *sch) htons(skb->protocol)); break; }; + return skb; } - static int dsmark_requeue(struct sk_buff *skb,struct Qdisc *sch) { - int ret; struct dsmark_qdisc_data *p = PRIV(sch); + int err; - D2PRINTK("dsmark_requeue(skb %p,sch %p,[qdisc %p])\n",skb,sch,p); - if ((ret = p->q->ops->requeue(skb, p->q)) == 0) { - sch->q.qlen++; - sch->qstats.requeues++; - return 0; + D2PRINTK("dsmark_requeue(skb %p,sch %p,[qdisc %p])\n", skb, sch, p); + + err = p->q->ops->requeue(skb, p->q); + if (err != NET_XMIT_SUCCESS) { + sch->qstats.drops++; + return err; } - sch->qstats.drops++; - return ret; -} + sch->q.qlen++; + sch->qstats.requeues++; + + return NET_XMIT_SUCCESS; +} static unsigned int dsmark_drop(struct Qdisc *sch) { struct dsmark_qdisc_data *p = PRIV(sch); unsigned int len; - DPRINTK("dsmark_reset(sch %p,[qdisc %p])\n",sch,p); - if (!p->q->ops->drop) - return 0; - if (!(len = p->q->ops->drop(p->q))) + DPRINTK("dsmark_reset(sch %p,[qdisc %p])\n", sch, p); + + if (p->q->ops->drop == NULL) return 0; - sch->q.qlen--; + + len = p->q->ops->drop(p->q); + if (len) + sch->q.qlen--; + return len; } - -static int dsmark_init(struct Qdisc *sch,struct rtattr *opt) +static int dsmark_init(struct Qdisc *sch, struct rtattr *opt) { struct dsmark_qdisc_data *p = PRIV(sch); struct rtattr *tb[TCA_DSMARK_MAX]; - __u16 tmp; - - DPRINTK("dsmark_init(sch %p,[qdisc %p],opt %p)\n",sch,p,opt); - if (!opt || - rtattr_parse(tb,TCA_DSMARK_MAX,RTA_DATA(opt),RTA_PAYLOAD(opt)) < 0 || - !tb[TCA_DSMARK_INDICES-1] || - RTA_PAYLOAD(tb[TCA_DSMARK_INDICES-1]) < sizeof(__u16)) - return -EINVAL; - p->indices = *(__u16 *) RTA_DATA(tb[TCA_DSMARK_INDICES-1]); - if (!p->indices) - return -EINVAL; - for (tmp = p->indices; tmp != 1; tmp >>= 1) { - if (tmp & 1) - return -EINVAL; - } - p->default_index = NO_DEFAULT_INDEX; - if (tb[TCA_DSMARK_DEFAULT_INDEX-1]) { - if (RTA_PAYLOAD(tb[TCA_DSMARK_DEFAULT_INDEX-1]) < sizeof(__u16)) - return -EINVAL; - p->default_index = - *(__u16 *) RTA_DATA(tb[TCA_DSMARK_DEFAULT_INDEX-1]); + int err = -EINVAL; + u32 default_index = NO_DEFAULT_INDEX; + u16 indices; + u8 *mask; + + DPRINTK("dsmark_init(sch %p,[qdisc %p],opt %p)\n", sch, p, opt); + + if (!opt || rtattr_parse_nested(tb, TCA_DSMARK_MAX, opt) < 0) + goto errout; + + indices = RTA_GET_U16(tb[TCA_DSMARK_INDICES-1]); + if (!indices || !dsmark_valid_indices(indices)) + goto errout; + + if (tb[TCA_DSMARK_DEFAULT_INDEX-1]) + default_index = RTA_GET_U16(tb[TCA_DSMARK_DEFAULT_INDEX-1]); + + mask = kmalloc(indices * 2, GFP_KERNEL); + if (mask == NULL) { + err = -ENOMEM; + goto errout; } - p->set_tc_index = !!tb[TCA_DSMARK_SET_TC_INDEX-1]; - p->mask = kmalloc(p->indices*2,GFP_KERNEL); - if (!p->mask) - return -ENOMEM; - p->value = p->mask+p->indices; - memset(p->mask,0xff,p->indices); - memset(p->value,0,p->indices); - if (!(p->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops))) + + p->mask = mask; + memset(p->mask, 0xff, indices); + + p->value = p->mask + indices; + memset(p->value, 0, indices); + + p->indices = indices; + p->default_index = default_index; + p->set_tc_index = RTA_GET_FLAG(tb[TCA_DSMARK_SET_TC_INDEX-1]); + + p->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); + if (p->q == NULL) p->q = &noop_qdisc; - DPRINTK("dsmark_init: qdisc %p\n",&p->q); - return 0; -} + DPRINTK("dsmark_init: qdisc %p\n", p->q); + + err = 0; +errout: +rtattr_failure: + return err; +} static void dsmark_reset(struct Qdisc *sch) { struct dsmark_qdisc_data *p = PRIV(sch); - DPRINTK("dsmark_reset(sch %p,[qdisc %p])\n",sch,p); + DPRINTK("dsmark_reset(sch %p,[qdisc %p])\n", sch, p); qdisc_reset(p->q); sch->q.qlen = 0; } - static void dsmark_destroy(struct Qdisc *sch) { struct dsmark_qdisc_data *p = PRIV(sch); struct tcf_proto *tp; - DPRINTK("dsmark_destroy(sch %p,[qdisc %p])\n",sch,p); + DPRINTK("dsmark_destroy(sch %p,[qdisc %p])\n", sch, p); + while (p->filter_list) { tp = p->filter_list; p->filter_list = tp->next; tcf_destroy(tp); } + qdisc_destroy(p->q); kfree(p->mask); } - static int dsmark_dump_class(struct Qdisc *sch, unsigned long cl, - struct sk_buff *skb, struct tcmsg *tcm) + struct sk_buff *skb, struct tcmsg *tcm) { struct dsmark_qdisc_data *p = PRIV(sch); - unsigned char *b = skb->tail; - struct rtattr *rta; + struct rtattr *opts = NULL; + + DPRINTK("dsmark_dump_class(sch %p,[qdisc %p],class %ld\n", sch, p, cl); - DPRINTK("dsmark_dump_class(sch %p,[qdisc %p],class %ld\n",sch,p,cl); - if (!cl || cl > p->indices) + if (!dsmark_valid_index(p, cl)) return -EINVAL; - tcm->tcm_handle = TC_H_MAKE(TC_H_MAJ(sch->handle),cl-1); - rta = (struct rtattr *) b; - RTA_PUT(skb,TCA_OPTIONS,0,NULL); - RTA_PUT(skb,TCA_DSMARK_MASK,1,&p->mask[cl-1]); - RTA_PUT(skb,TCA_DSMARK_VALUE,1,&p->value[cl-1]); - rta->rta_len = skb->tail-b; - return skb->len; + + tcm->tcm_handle = TC_H_MAKE(TC_H_MAJ(sch->handle), cl-1); + + opts = RTA_NEST(skb, TCA_OPTIONS); + RTA_PUT_U8(skb,TCA_DSMARK_MASK, p->mask[cl-1]); + RTA_PUT_U8(skb,TCA_DSMARK_VALUE, p->value[cl-1]); + + return RTA_NEST_END(skb, opts); rtattr_failure: - skb_trim(skb,b-skb->data); - return -1; + return RTA_NEST_CANCEL(skb, opts); } static int dsmark_dump(struct Qdisc *sch, struct sk_buff *skb) { struct dsmark_qdisc_data *p = PRIV(sch); - unsigned char *b = skb->tail; - struct rtattr *rta; + struct rtattr *opts = NULL; - rta = (struct rtattr *) b; - RTA_PUT(skb,TCA_OPTIONS,0,NULL); - RTA_PUT(skb,TCA_DSMARK_INDICES,sizeof(__u16),&p->indices); - if (p->default_index != NO_DEFAULT_INDEX) { - __u16 tmp = p->default_index; + opts = RTA_NEST(skb, TCA_OPTIONS); + RTA_PUT_U16(skb, TCA_DSMARK_INDICES, p->indices); + + if (p->default_index != NO_DEFAULT_INDEX) + RTA_PUT_U16(skb, TCA_DSMARK_DEFAULT_INDEX, p->default_index); - RTA_PUT(skb,TCA_DSMARK_DEFAULT_INDEX, sizeof(__u16), &tmp); - } if (p->set_tc_index) - RTA_PUT(skb, TCA_DSMARK_SET_TC_INDEX, 0, NULL); - rta->rta_len = skb->tail-b; - return skb->len; + RTA_PUT_FLAG(skb, TCA_DSMARK_SET_TC_INDEX); + + return RTA_NEST_END(skb, opts); rtattr_failure: - skb_trim(skb,b-skb->data); - return -1; + return RTA_NEST_CANCEL(skb, opts); } static struct Qdisc_class_ops dsmark_class_ops = { @@ -470,10 +504,13 @@ static int __init dsmark_module_init(void) { return register_qdisc(&dsmark_qdisc_ops); } + static void __exit dsmark_module_exit(void) { unregister_qdisc(&dsmark_qdisc_ops); } + module_init(dsmark_module_init) module_exit(dsmark_module_exit) + MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c index 4888305..033083b 100644 --- a/net/sched/sch_fifo.c +++ b/net/sched/sch_fifo.c @@ -11,131 +11,38 @@ #include <linux/config.h> #include <linux/module.h> -#include <asm/uaccess.h> -#include <asm/system.h> -#include <linux/bitops.h> #include <linux/types.h> #include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/string.h> -#include <linux/mm.h> -#include <linux/socket.h> -#include <linux/sockios.h> -#include <linux/in.h> #include <linux/errno.h> -#include <linux/interrupt.h> -#include <linux/if_ether.h> -#include <linux/inet.h> #include <linux/netdevice.h> -#include <linux/etherdevice.h> -#include <linux/notifier.h> -#include <net/ip.h> -#include <net/route.h> #include <linux/skbuff.h> -#include <net/sock.h> #include <net/pkt_sched.h> /* 1 band FIFO pseudo-"scheduler" */ struct fifo_sched_data { - unsigned limit; + u32 limit; }; -static int -bfifo_enqueue(struct sk_buff *skb, struct Qdisc* sch) +static int bfifo_enqueue(struct sk_buff *skb, struct Qdisc* sch) { struct fifo_sched_data *q = qdisc_priv(sch); - if (sch->qstats.backlog + skb->len <= q->limit) { - __skb_queue_tail(&sch->q, skb); - sch->qstats.backlog += skb->len; - sch->bstats.bytes += skb->len; - sch->bstats.packets++; - return 0; - } - sch->qstats.drops++; -#ifdef CONFIG_NET_CLS_POLICE - if (sch->reshape_fail==NULL || sch->reshape_fail(skb, sch)) -#endif - kfree_skb(skb); - return NET_XMIT_DROP; -} - -static int -bfifo_requeue(struct sk_buff *skb, struct Qdisc* sch) -{ - __skb_queue_head(&sch->q, skb); - sch->qstats.backlog += skb->len; - sch->qstats.requeues++; - return 0; -} - -static struct sk_buff * -bfifo_dequeue(struct Qdisc* sch) -{ - struct sk_buff *skb; + if (likely(sch->qstats.backlog + skb->len <= q->limit)) + return qdisc_enqueue_tail(skb, sch); - skb = __skb_dequeue(&sch->q); - if (skb) - sch->qstats.backlog -= skb->len; - return skb; + return qdisc_reshape_fail(skb, sch); } -static unsigned int -fifo_drop(struct Qdisc* sch) -{ - struct sk_buff *skb; - - skb = __skb_dequeue_tail(&sch->q); - if (skb) { - unsigned int len = skb->len; - sch->qstats.backlog -= len; - kfree_skb(skb); - return len; - } - return 0; -} - -static void -fifo_reset(struct Qdisc* sch) -{ - skb_queue_purge(&sch->q); - sch->qstats.backlog = 0; -} - -static int -pfifo_enqueue(struct sk_buff *skb, struct Qdisc* sch) +static int pfifo_enqueue(struct sk_buff *skb, struct Qdisc* sch) { struct fifo_sched_data *q = qdisc_priv(sch); - if (sch->q.qlen < q->limit) { - __skb_queue_tail(&sch->q, skb); - sch->bstats.bytes += skb->len; - sch->bstats.packets++; - return 0; - } - sch->qstats.drops++; -#ifdef CONFIG_NET_CLS_POLICE - if (sch->reshape_fail==NULL || sch->reshape_fail(skb, sch)) -#endif - kfree_skb(skb); - return NET_XMIT_DROP; -} - -static int -pfifo_requeue(struct sk_buff *skb, struct Qdisc* sch) -{ - __skb_queue_head(&sch->q, skb); - sch->qstats.requeues++; - return 0; -} - + if (likely(skb_queue_len(&sch->q) < q->limit)) + return qdisc_enqueue_tail(skb, sch); -static struct sk_buff * -pfifo_dequeue(struct Qdisc* sch) -{ - return __skb_dequeue(&sch->q); + return qdisc_reshape_fail(skb, sch); } static int fifo_init(struct Qdisc *sch, struct rtattr *opt) @@ -143,66 +50,59 @@ static int fifo_init(struct Qdisc *sch, struct rtattr *opt) struct fifo_sched_data *q = qdisc_priv(sch); if (opt == NULL) { - unsigned int limit = sch->dev->tx_queue_len ? : 1; + u32 limit = sch->dev->tx_queue_len ? : 1; if (sch->ops == &bfifo_qdisc_ops) - q->limit = limit*sch->dev->mtu; - else - q->limit = limit; + limit *= sch->dev->mtu; + + q->limit = limit; } else { struct tc_fifo_qopt *ctl = RTA_DATA(opt); - if (opt->rta_len < RTA_LENGTH(sizeof(*ctl))) + + if (RTA_PAYLOAD(opt) < sizeof(*ctl)) return -EINVAL; + q->limit = ctl->limit; } + return 0; } static int fifo_dump(struct Qdisc *sch, struct sk_buff *skb) { struct fifo_sched_data *q = qdisc_priv(sch); - unsigned char *b = skb->tail; - struct tc_fifo_qopt opt; + struct tc_fifo_qopt opt = { .limit = q->limit }; - opt.limit = q->limit; RTA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt); - return skb->len; rtattr_failure: - skb_trim(skb, b - skb->data); return -1; } struct Qdisc_ops pfifo_qdisc_ops = { - .next = NULL, - .cl_ops = NULL, .id = "pfifo", .priv_size = sizeof(struct fifo_sched_data), .enqueue = pfifo_enqueue, - .dequeue = pfifo_dequeue, - .requeue = pfifo_requeue, - .drop = fifo_drop, + .dequeue = qdisc_dequeue_head, + .requeue = qdisc_requeue, + .drop = qdisc_queue_drop, .init = fifo_init, - .reset = fifo_reset, - .destroy = NULL, + .reset = qdisc_reset_queue, .change = fifo_init, .dump = fifo_dump, .owner = THIS_MODULE, }; struct Qdisc_ops bfifo_qdisc_ops = { - .next = NULL, - .cl_ops = NULL, .id = "bfifo", .priv_size = sizeof(struct fifo_sched_data), .enqueue = bfifo_enqueue, - .dequeue = bfifo_dequeue, - .requeue = bfifo_requeue, - .drop = fifo_drop, + .dequeue = qdisc_dequeue_head, + .requeue = qdisc_requeue, + .drop = qdisc_queue_drop, .init = fifo_init, - .reset = fifo_reset, - .destroy = NULL, + .reset = qdisc_reset_queue, .change = fifo_init, .dump = fifo_dump, .owner = THIS_MODULE, diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 87e48a4..7683b34 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -243,31 +243,27 @@ static void dev_watchdog_down(struct net_device *dev) cheaper. */ -static int -noop_enqueue(struct sk_buff *skb, struct Qdisc * qdisc) +static int noop_enqueue(struct sk_buff *skb, struct Qdisc * qdisc) { kfree_skb(skb); return NET_XMIT_CN; } -static struct sk_buff * -noop_dequeue(struct Qdisc * qdisc) +static struct sk_buff *noop_dequeue(struct Qdisc * qdisc) { return NULL; } -static int -noop_requeue(struct sk_buff *skb, struct Qdisc* qdisc) +static int noop_requeue(struct sk_buff *skb, struct Qdisc* qdisc) { if (net_ratelimit()) - printk(KERN_DEBUG "%s deferred output. It is buggy.\n", skb->dev->name); + printk(KERN_DEBUG "%s deferred output. It is buggy.\n", + skb->dev->name); kfree_skb(skb); return NET_XMIT_CN; } struct Qdisc_ops noop_qdisc_ops = { - .next = NULL, - .cl_ops = NULL, .id = "noop", .priv_size = 0, .enqueue = noop_enqueue, @@ -285,8 +281,6 @@ struct Qdisc noop_qdisc = { }; static struct Qdisc_ops noqueue_qdisc_ops = { - .next = NULL, - .cl_ops = NULL, .id = "noqueue", .priv_size = 0, .enqueue = noop_enqueue, @@ -311,97 +305,87 @@ static const u8 prio2band[TC_PRIO_MAX+1] = generic prio+fifo combination. */ -static int -pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc* qdisc) +#define PFIFO_FAST_BANDS 3 + +static inline struct sk_buff_head *prio2list(struct sk_buff *skb, + struct Qdisc *qdisc) { struct sk_buff_head *list = qdisc_priv(qdisc); + return list + prio2band[skb->priority & TC_PRIO_MAX]; +} - list += prio2band[skb->priority&TC_PRIO_MAX]; +static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc* qdisc) +{ + struct sk_buff_head *list = prio2list(skb, qdisc); - if (list->qlen < qdisc->dev->tx_queue_len) { - __skb_queue_tail(list, skb); + if (skb_queue_len(list) < qdisc->dev->tx_queue_len) { qdisc->q.qlen++; - qdisc->bstats.bytes += skb->len; - qdisc->bstats.packets++; - return 0; + return __qdisc_enqueue_tail(skb, qdisc, list); } - qdisc->qstats.drops++; - kfree_skb(skb); - return NET_XMIT_DROP; + + return qdisc_drop(skb, qdisc); } -static struct sk_buff * -pfifo_fast_dequeue(struct Qdisc* qdisc) +static struct sk_buff *pfifo_fast_dequeue(struct Qdisc* qdisc) { int prio; struct sk_buff_head *list = qdisc_priv(qdisc); - struct sk_buff *skb; - for (prio = 0; prio < 3; prio++, list++) { - skb = __skb_dequeue(list); + for (prio = 0; prio < PFIFO_FAST_BANDS; prio++, list++) { + struct sk_buff *skb = __qdisc_dequeue_head(qdisc, list); if (skb) { qdisc->q.qlen--; return skb; } } + return NULL; } -static int -pfifo_fast_requeue(struct sk_buff *skb, struct Qdisc* qdisc) +static int pfifo_fast_requeue(struct sk_buff *skb, struct Qdisc* qdisc) { - struct sk_buff_head *list = qdisc_priv(qdisc); - - list += prio2band[skb->priority&TC_PRIO_MAX]; - - __skb_queue_head(list, skb); qdisc->q.qlen++; - qdisc->qstats.requeues++; - return 0; + return __qdisc_requeue(skb, qdisc, prio2list(skb, qdisc)); } -static void -pfifo_fast_reset(struct Qdisc* qdisc) +static void pfifo_fast_reset(struct Qdisc* qdisc) { int prio; struct sk_buff_head *list = qdisc_priv(qdisc); - for (prio=0; prio < 3; prio++) - skb_queue_purge(list+prio); + for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) + __qdisc_reset_queue(qdisc, list + prio); + + qdisc->qstats.backlog = 0; qdisc->q.qlen = 0; } static int pfifo_fast_dump(struct Qdisc *qdisc, struct sk_buff *skb) { - unsigned char *b = skb->tail; - struct tc_prio_qopt opt; + struct tc_prio_qopt opt = { .bands = PFIFO_FAST_BANDS }; - opt.bands = 3; memcpy(&opt.priomap, prio2band, TC_PRIO_MAX+1); RTA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt); return skb->len; rtattr_failure: - skb_trim(skb, b - skb->data); return -1; } static int pfifo_fast_init(struct Qdisc *qdisc, struct rtattr *opt) { - int i; + int prio; struct sk_buff_head *list = qdisc_priv(qdisc); - for (i=0; i<3; i++) - skb_queue_head_init(list+i); + for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) + skb_queue_head_init(list + prio); return 0; } static struct Qdisc_ops pfifo_fast_ops = { - .next = NULL, - .cl_ops = NULL, .id = "pfifo_fast", - .priv_size = 3 * sizeof(struct sk_buff_head), + .priv_size = PFIFO_FAST_BANDS * sizeof(struct sk_buff_head), .enqueue = pfifo_fast_enqueue, .dequeue = pfifo_fast_dequeue, .requeue = pfifo_fast_requeue, diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index e0c9fbe..bb9bf8d 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -53,7 +53,6 @@ struct netem_sched_data { struct Qdisc *qdisc; - struct sk_buff_head delayed; struct timer_list timer; u32 latency; @@ -63,11 +62,12 @@ struct netem_sched_data { u32 gap; u32 jitter; u32 duplicate; + u32 reorder; struct crndstate { unsigned long last; unsigned long rho; - } delay_cor, loss_cor, dup_cor; + } delay_cor, loss_cor, dup_cor, reorder_cor; struct disttable { u32 size; @@ -137,122 +137,68 @@ static long tabledist(unsigned long mu, long sigma, return x / NETEM_DIST_SCALE + (sigma / NETEM_DIST_SCALE) * t + mu; } -/* Put skb in the private delayed queue. */ -static int netem_delay(struct Qdisc *sch, struct sk_buff *skb) -{ - struct netem_sched_data *q = qdisc_priv(sch); - psched_tdiff_t td; - psched_time_t now; - - PSCHED_GET_TIME(now); - td = tabledist(q->latency, q->jitter, &q->delay_cor, q->delay_dist); - - /* Always queue at tail to keep packets in order */ - if (likely(q->delayed.qlen < q->limit)) { - struct netem_skb_cb *cb = (struct netem_skb_cb *)skb->cb; - - PSCHED_TADD2(now, td, cb->time_to_send); - - pr_debug("netem_delay: skb=%p now=%llu tosend=%llu\n", skb, - now, cb->time_to_send); - - __skb_queue_tail(&q->delayed, skb); - return NET_XMIT_SUCCESS; - } - - pr_debug("netem_delay: queue over limit %d\n", q->limit); - sch->qstats.overlimits++; - kfree_skb(skb); - return NET_XMIT_DROP; -} - /* - * Move a packet that is ready to send from the delay holding - * list to the underlying qdisc. + * Insert one skb into qdisc. + * Note: parent depends on return value to account for queue length. + * NET_XMIT_DROP: queue length didn't change. + * NET_XMIT_SUCCESS: one skb was queued. */ -static int netem_run(struct Qdisc *sch) -{ - struct netem_sched_data *q = qdisc_priv(sch); - struct sk_buff *skb; - psched_time_t now; - - PSCHED_GET_TIME(now); - - skb = skb_peek(&q->delayed); - if (skb) { - const struct netem_skb_cb *cb - = (const struct netem_skb_cb *)skb->cb; - long delay - = PSCHED_US2JIFFIE(PSCHED_TDIFF(cb->time_to_send, now)); - pr_debug("netem_run: skb=%p delay=%ld\n", skb, delay); - - /* if more time remaining? */ - if (delay > 0) { - mod_timer(&q->timer, jiffies + delay); - return 1; - } - - __skb_unlink(skb, &q->delayed); - - if (q->qdisc->enqueue(skb, q->qdisc)) { - sch->q.qlen--; - sch->qstats.drops++; - } - } - - return 0; -} - static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) { struct netem_sched_data *q = qdisc_priv(sch); + struct netem_skb_cb *cb = (struct netem_skb_cb *)skb->cb; + struct sk_buff *skb2; int ret; + int count = 1; pr_debug("netem_enqueue skb=%p\n", skb); + /* Random duplication */ + if (q->duplicate && q->duplicate >= get_crandom(&q->dup_cor)) + ++count; + /* Random packet drop 0 => none, ~0 => all */ - if (q->loss && q->loss >= get_crandom(&q->loss_cor)) { - pr_debug("netem_enqueue: random loss\n"); + if (q->loss && q->loss >= get_crandom(&q->loss_cor)) + --count; + + if (count == 0) { sch->qstats.drops++; kfree_skb(skb); - return 0; /* lie about loss so TCP doesn't know */ + return NET_XMIT_DROP; } - /* Random duplication */ - if (q->duplicate && q->duplicate >= get_crandom(&q->dup_cor)) { - struct sk_buff *skb2; - - skb2 = skb_clone(skb, GFP_ATOMIC); - if (skb2 && netem_delay(sch, skb2) == NET_XMIT_SUCCESS) { - struct Qdisc *qp; - - /* Since one packet can generate two packets in the - * queue, the parent's qlen accounting gets confused, - * so fix it. - */ - qp = qdisc_lookup(sch->dev, TC_H_MAJ(sch->parent)); - if (qp) - qp->q.qlen++; - - sch->q.qlen++; - sch->bstats.bytes += skb2->len; - sch->bstats.packets++; - } else - sch->qstats.drops++; + /* + * If we need to duplicate packet, then re-insert at top of the + * qdisc tree, since parent queuer expects that only one + * skb will be queued. + */ + if (count > 1 && (skb2 = skb_clone(skb, GFP_ATOMIC)) != NULL) { + struct Qdisc *rootq = sch->dev->qdisc; + u32 dupsave = q->duplicate; /* prevent duplicating a dup... */ + q->duplicate = 0; + + rootq->enqueue(skb2, rootq); + q->duplicate = dupsave; } - /* If doing simple delay then gap == 0 so all packets - * go into the delayed holding queue - * otherwise if doing out of order only "1 out of gap" - * packets will be delayed. - */ - if (q->counter < q->gap) { + if (q->gap == 0 /* not doing reordering */ + || q->counter < q->gap /* inside last reordering gap */ + || q->reorder < get_crandom(&q->reorder_cor)) { + psched_time_t now; + PSCHED_GET_TIME(now); + PSCHED_TADD2(now, tabledist(q->latency, q->jitter, + &q->delay_cor, q->delay_dist), + cb->time_to_send); ++q->counter; ret = q->qdisc->enqueue(skb, q->qdisc); } else { + /* + * Do re-ordering by putting one out of N packets at the front + * of the queue. + */ + PSCHED_GET_TIME(cb->time_to_send); q->counter = 0; - ret = netem_delay(sch, skb); - netem_run(sch); + ret = q->qdisc->ops->requeue(skb, q->qdisc); } if (likely(ret == NET_XMIT_SUCCESS)) { @@ -296,22 +242,33 @@ static struct sk_buff *netem_dequeue(struct Qdisc *sch) { struct netem_sched_data *q = qdisc_priv(sch); struct sk_buff *skb; - int pending; - - pending = netem_run(sch); skb = q->qdisc->dequeue(q->qdisc); if (skb) { - pr_debug("netem_dequeue: return skb=%p\n", skb); - sch->q.qlen--; - sch->flags &= ~TCQ_F_THROTTLED; - } - else if (pending) { - pr_debug("netem_dequeue: throttling\n"); + const struct netem_skb_cb *cb + = (const struct netem_skb_cb *)skb->cb; + psched_time_t now; + long delay; + + /* if more time remaining? */ + PSCHED_GET_TIME(now); + delay = PSCHED_US2JIFFIE(PSCHED_TDIFF(cb->time_to_send, now)); + pr_debug("netem_run: skb=%p delay=%ld\n", skb, delay); + if (delay <= 0) { + pr_debug("netem_dequeue: return skb=%p\n", skb); + sch->q.qlen--; + sch->flags &= ~TCQ_F_THROTTLED; + return skb; + } + + mod_timer(&q->timer, jiffies + delay); sch->flags |= TCQ_F_THROTTLED; - } - return skb; + if (q->qdisc->ops->requeue(skb, q->qdisc) != 0) + sch->qstats.drops++; + } + + return NULL; } static void netem_watchdog(unsigned long arg) @@ -328,8 +285,6 @@ static void netem_reset(struct Qdisc *sch) struct netem_sched_data *q = qdisc_priv(sch); qdisc_reset(q->qdisc); - skb_queue_purge(&q->delayed); - sch->q.qlen = 0; sch->flags &= ~TCQ_F_THROTTLED; del_timer_sync(&q->timer); @@ -397,6 +352,19 @@ static int get_correlation(struct Qdisc *sch, const struct rtattr *attr) return 0; } +static int get_reorder(struct Qdisc *sch, const struct rtattr *attr) +{ + struct netem_sched_data *q = qdisc_priv(sch); + const struct tc_netem_reorder *r = RTA_DATA(attr); + + if (RTA_PAYLOAD(attr) != sizeof(*r)) + return -EINVAL; + + q->reorder = r->probability; + init_crandom(&q->reorder_cor, r->correlation); + return 0; +} + static int netem_change(struct Qdisc *sch, struct rtattr *opt) { struct netem_sched_data *q = qdisc_priv(sch); @@ -417,9 +385,15 @@ static int netem_change(struct Qdisc *sch, struct rtattr *opt) q->jitter = qopt->jitter; q->limit = qopt->limit; q->gap = qopt->gap; + q->counter = 0; q->loss = qopt->loss; q->duplicate = qopt->duplicate; + /* for compatiablity with earlier versions. + * if gap is set, need to assume 100% probablity + */ + q->reorder = ~0; + /* Handle nested options after initial queue options. * Should have put all options in nested format but too late now. */ @@ -441,6 +415,11 @@ static int netem_change(struct Qdisc *sch, struct rtattr *opt) if (ret) return ret; } + if (tb[TCA_NETEM_REORDER-1]) { + ret = get_reorder(sch, tb[TCA_NETEM_REORDER-1]); + if (ret) + return ret; + } } @@ -455,11 +434,9 @@ static int netem_init(struct Qdisc *sch, struct rtattr *opt) if (!opt) return -EINVAL; - skb_queue_head_init(&q->delayed); init_timer(&q->timer); q->timer.function = netem_watchdog; q->timer.data = (unsigned long) sch; - q->counter = 0; q->qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); if (!q->qdisc) { @@ -491,6 +468,7 @@ static int netem_dump(struct Qdisc *sch, struct sk_buff *skb) struct rtattr *rta = (struct rtattr *) b; struct tc_netem_qopt qopt; struct tc_netem_corr cor; + struct tc_netem_reorder reorder; qopt.latency = q->latency; qopt.jitter = q->jitter; @@ -504,6 +482,11 @@ static int netem_dump(struct Qdisc *sch, struct sk_buff *skb) cor.loss_corr = q->loss_cor.rho; cor.dup_corr = q->dup_cor.rho; RTA_PUT(skb, TCA_NETEM_CORR, sizeof(cor), &cor); + + reorder.probability = q->reorder; + reorder.correlation = q->reorder_cor.rho; + RTA_PUT(skb, TCA_NETEM_REORDER, sizeof(reorder), &reorder); + rta->rta_len = skb->tail - b; return skb->len; 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 b719a77..339f7ac 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -178,6 +178,37 @@ int sctp_rcv(struct sk_buff *skb) asoc = __sctp_rcv_lookup(skb, &src, &dest, &transport); + if (!asoc) + ep = __sctp_rcv_lookup_endpoint(&dest); + + /* Retrieve the common input handling substructure. */ + rcvr = asoc ? &asoc->base : &ep->base; + sk = rcvr->sk; + + /* + * If a frame arrives on an interface and the receiving socket is + * bound to another interface, via SO_BINDTODEVICE, treat it as OOTB + */ + if (sk->sk_bound_dev_if && (sk->sk_bound_dev_if != af->skb_iif(skb))) + { + sock_put(sk); + if (asoc) { + sctp_association_put(asoc); + asoc = NULL; + } else { + sctp_endpoint_put(ep); + ep = NULL; + } + sk = sctp_get_ctl_sock(); + ep = sctp_sk(sk)->ep; + sctp_endpoint_hold(ep); + sock_hold(sk); + rcvr = &ep->base; + } + + if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf) + goto discard_release; + /* * RFC 2960, 8.4 - Handle "Out of the blue" Packets. * An SCTP packet is called an "out of the blue" (OOTB) @@ -187,22 +218,12 @@ int sctp_rcv(struct sk_buff *skb) * packet belongs. */ if (!asoc) { - ep = __sctp_rcv_lookup_endpoint(&dest); if (sctp_rcv_ootb(skb)) { SCTP_INC_STATS_BH(SCTP_MIB_OUTOFBLUES); goto discard_release; } } - /* Retrieve the common input handling substructure. */ - rcvr = asoc ? &asoc->base : &ep->base; - sk = rcvr->sk; - - if ((sk) && (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf)) { - goto discard_release; - } - - /* SCTP seems to always need a timestamp right now (FIXME) */ if (skb->stamp.tv_sec == 0) { do_gettimeofday(&skb->stamp); @@ -265,13 +286,11 @@ discard_it: discard_release: /* Release any structures we may be holding. */ - if (asoc) { - sock_put(asoc->base.sk); + sock_put(sk); + if (asoc) sctp_association_put(asoc); - } else { - sock_put(ep->base.sk); + else sctp_endpoint_put(ep); - } goto discard_it; } @@ -334,7 +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/ipv6.c b/net/sctp/ipv6.c index c9d9ea0..c7e42d1 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -812,26 +812,23 @@ static int sctp_inet6_bind_verify(struct sctp_sock *opt, union sctp_addr *addr) if (addr->sa.sa_family != AF_INET6) af = sctp_get_af_specific(addr->sa.sa_family); else { - struct sock *sk; int type = ipv6_addr_type(&addr->v6.sin6_addr); - sk = sctp_opt2sk(opt); + struct net_device *dev; + if (type & IPV6_ADDR_LINKLOCAL) { - /* Note: Behavior similar to af_inet6.c: - * 1) Overrides previous bound_dev_if - * 2) Destructive even if bind isn't successful. - */ - - if (addr->v6.sin6_scope_id) - sk->sk_bound_dev_if = addr->v6.sin6_scope_id; - if (!sk->sk_bound_dev_if) + if (!addr->v6.sin6_scope_id) + return 0; + dev = dev_get_by_index(addr->v6.sin6_scope_id); + if (!dev) return 0; + dev_put(dev); } af = opt->pf->af; } return af->available(addr, opt); } -/* Verify that the provided sockaddr looks bindable. Common verification, +/* Verify that the provided sockaddr looks sendable. Common verification, * has already been taken care of. */ static int sctp_inet6_send_verify(struct sctp_sock *opt, union sctp_addr *addr) @@ -842,19 +839,16 @@ static int sctp_inet6_send_verify(struct sctp_sock *opt, union sctp_addr *addr) if (addr->sa.sa_family != AF_INET6) af = sctp_get_af_specific(addr->sa.sa_family); else { - struct sock *sk; int type = ipv6_addr_type(&addr->v6.sin6_addr); - sk = sctp_opt2sk(opt); + struct net_device *dev; + if (type & IPV6_ADDR_LINKLOCAL) { - /* Note: Behavior similar to af_inet6.c: - * 1) Overrides previous bound_dev_if - * 2) Destructive even if bind isn't successful. - */ - - if (addr->v6.sin6_scope_id) - sk->sk_bound_dev_if = addr->v6.sin6_scope_id; - if (!sk->sk_bound_dev_if) + if (!addr->v6.sin6_scope_id) + return 0; + dev = dev_get_by_index(addr->v6.sin6_scope_id); + if (!dev) return 0; + dev_put(dev); } af = opt->pf->af; } diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index 1b2d4adc..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/proc.c b/net/sctp/proc.c index e42fd8c..98d49ec 100644 --- a/net/sctp/proc.c +++ b/net/sctp/proc.c @@ -132,14 +132,25 @@ void sctp_snmp_proc_exit(void) static void sctp_seq_dump_local_addrs(struct seq_file *seq, struct sctp_ep_common *epb) { struct list_head *pos; + struct sctp_association *asoc; struct sctp_sockaddr_entry *laddr; - union sctp_addr *addr; + struct sctp_transport *peer; + union sctp_addr *addr, *primary = NULL; struct sctp_af *af; + if (epb->type == SCTP_EP_TYPE_ASSOCIATION) { + asoc = sctp_assoc(epb); + peer = asoc->peer.primary_path; + primary = &peer->saddr; + } + list_for_each(pos, &epb->bind_addr.address_list) { laddr = list_entry(pos, struct sctp_sockaddr_entry, list); addr = (union sctp_addr *)&laddr->a; af = sctp_get_af_specific(addr->sa.sa_family); + if (primary && af->cmp_addr(addr, primary)) { + seq_printf(seq, "*"); + } af->seq_dump_addr(seq, addr); } } @@ -149,17 +160,54 @@ static void sctp_seq_dump_remote_addrs(struct seq_file *seq, struct sctp_associa { struct list_head *pos; struct sctp_transport *transport; - union sctp_addr *addr; + union sctp_addr *addr, *primary; struct sctp_af *af; + primary = &(assoc->peer.primary_addr); list_for_each(pos, &assoc->peer.transport_addr_list) { transport = list_entry(pos, struct sctp_transport, transports); addr = (union sctp_addr *)&transport->ipaddr; af = sctp_get_af_specific(addr->sa.sa_family); + if (af->cmp_addr(addr, primary)) { + seq_printf(seq, "*"); + } af->seq_dump_addr(seq, addr); } } +static void * sctp_eps_seq_start(struct seq_file *seq, loff_t *pos) +{ + if (*pos > sctp_ep_hashsize) + return NULL; + + if (*pos < 0) + *pos = 0; + + if (*pos == 0) + seq_printf(seq, " ENDPT SOCK STY SST HBKT LPORT UID INODE LADDRS\n"); + + ++*pos; + + return (void *)pos; +} + +static void sctp_eps_seq_stop(struct seq_file *seq, void *v) +{ + return; +} + + +static void * sctp_eps_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + if (*pos > sctp_ep_hashsize) + return NULL; + + ++*pos; + + return pos; +} + + /* Display sctp endpoints (/proc/net/sctp/eps). */ static int sctp_eps_seq_show(struct seq_file *seq, void *v) { @@ -167,38 +215,50 @@ static int sctp_eps_seq_show(struct seq_file *seq, void *v) struct sctp_ep_common *epb; struct sctp_endpoint *ep; struct sock *sk; - int hash; - - seq_printf(seq, " ENDPT SOCK STY SST HBKT LPORT LADDRS\n"); - for (hash = 0; hash < sctp_ep_hashsize; hash++) { - head = &sctp_ep_hashtable[hash]; - read_lock(&head->lock); - for (epb = head->chain; epb; epb = epb->next) { - ep = sctp_ep(epb); - sk = epb->sk; - seq_printf(seq, "%8p %8p %-3d %-3d %-4d %-5d ", ep, sk, - sctp_sk(sk)->type, sk->sk_state, hash, - epb->bind_addr.port); - sctp_seq_dump_local_addrs(seq, epb); - seq_printf(seq, "\n"); - } - read_unlock(&head->lock); + int hash = *(int *)v; + + if (hash > sctp_ep_hashsize) + return -ENOMEM; + + head = &sctp_ep_hashtable[hash-1]; + sctp_local_bh_disable(); + read_lock(&head->lock); + for (epb = head->chain; epb; epb = epb->next) { + ep = sctp_ep(epb); + sk = epb->sk; + seq_printf(seq, "%8p %8p %-3d %-3d %-4d %-5d %5d %5lu ", ep, sk, + sctp_sk(sk)->type, sk->sk_state, hash-1, + epb->bind_addr.port, + sock_i_uid(sk), sock_i_ino(sk)); + + sctp_seq_dump_local_addrs(seq, epb); + seq_printf(seq, "\n"); } + read_unlock(&head->lock); + sctp_local_bh_enable(); return 0; } +static struct seq_operations sctp_eps_ops = { + .start = sctp_eps_seq_start, + .next = sctp_eps_seq_next, + .stop = sctp_eps_seq_stop, + .show = sctp_eps_seq_show, +}; + + /* Initialize the seq file operations for 'eps' object. */ static int sctp_eps_seq_open(struct inode *inode, struct file *file) { - return single_open(file, sctp_eps_seq_show, NULL); + return seq_open(file, &sctp_eps_ops); } static struct file_operations sctp_eps_seq_fops = { .open = sctp_eps_seq_open, .read = seq_read, .llseek = seq_lseek, - .release = single_release, + .release = seq_release, }; /* Set up the proc fs entry for 'eps' object. */ @@ -221,6 +281,40 @@ void sctp_eps_proc_exit(void) remove_proc_entry("eps", proc_net_sctp); } + +static void * sctp_assocs_seq_start(struct seq_file *seq, loff_t *pos) +{ + if (*pos > sctp_assoc_hashsize) + return NULL; + + if (*pos < 0) + *pos = 0; + + if (*pos == 0) + seq_printf(seq, " ASSOC SOCK STY SST ST HBKT ASSOC-ID TX_QUEUE RX_QUEUE UID INODE LPORT " + "RPORT LADDRS <-> RADDRS\n"); + + ++*pos; + + return (void *)pos; +} + +static void sctp_assocs_seq_stop(struct seq_file *seq, void *v) +{ + return; +} + + +static void * sctp_assocs_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + if (*pos > sctp_assoc_hashsize) + return NULL; + + ++*pos; + + return pos; +} + /* Display sctp associations (/proc/net/sctp/assocs). */ static int sctp_assocs_seq_show(struct seq_file *seq, void *v) { @@ -228,43 +322,57 @@ static int sctp_assocs_seq_show(struct seq_file *seq, void *v) struct sctp_ep_common *epb; struct sctp_association *assoc; struct sock *sk; - int hash; - - seq_printf(seq, " ASSOC SOCK STY SST ST HBKT LPORT RPORT " - "LADDRS <-> RADDRS\n"); - for (hash = 0; hash < sctp_assoc_hashsize; hash++) { - head = &sctp_assoc_hashtable[hash]; - read_lock(&head->lock); - for (epb = head->chain; epb; epb = epb->next) { - assoc = sctp_assoc(epb); - sk = epb->sk; - seq_printf(seq, - "%8p %8p %-3d %-3d %-2d %-4d %-5d %-5d ", - assoc, sk, sctp_sk(sk)->type, sk->sk_state, - assoc->state, hash, epb->bind_addr.port, - assoc->peer.port); - sctp_seq_dump_local_addrs(seq, epb); - seq_printf(seq, "<-> "); - sctp_seq_dump_remote_addrs(seq, assoc); - seq_printf(seq, "\n"); - } - read_unlock(&head->lock); + int hash = *(int *)v; + + if (hash > sctp_assoc_hashsize) + return -ENOMEM; + + head = &sctp_assoc_hashtable[hash-1]; + sctp_local_bh_disable(); + read_lock(&head->lock); + for (epb = head->chain; epb; epb = epb->next) { + assoc = sctp_assoc(epb); + sk = epb->sk; + seq_printf(seq, + "%8p %8p %-3d %-3d %-2d %-4d %4d %8d %8d %7d %5lu %-5d %5d ", + assoc, sk, sctp_sk(sk)->type, sk->sk_state, + assoc->state, hash-1, assoc->assoc_id, + (sk->sk_rcvbuf - assoc->rwnd), + assoc->sndbuf_used, + sock_i_uid(sk), sock_i_ino(sk), + epb->bind_addr.port, + assoc->peer.port); + + seq_printf(seq, " "); + sctp_seq_dump_local_addrs(seq, epb); + seq_printf(seq, "<-> "); + sctp_seq_dump_remote_addrs(seq, assoc); + seq_printf(seq, "\n"); } + read_unlock(&head->lock); + sctp_local_bh_enable(); return 0; } +static struct seq_operations sctp_assoc_ops = { + .start = sctp_assocs_seq_start, + .next = sctp_assocs_seq_next, + .stop = sctp_assocs_seq_stop, + .show = sctp_assocs_seq_show, +}; + /* Initialize the seq file operations for 'assocs' object. */ static int sctp_assocs_seq_open(struct inode *inode, struct file *file) { - return single_open(file, sctp_assocs_seq_show, NULL); + return seq_open(file, &sctp_assoc_ops); } static struct file_operations sctp_assocs_seq_fops = { .open = sctp_assocs_seq_open, .read = seq_read, .llseek = seq_lseek, - .release = single_release, + .release = seq_release, }; /* Set up the proc fs entry for 'assocs' object. */ diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 2e1f9c3..5135e1a 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -378,10 +378,13 @@ static int sctp_v4_available(union sctp_addr *addr, struct sctp_sock *sp) { int ret = inet_addr_type(addr->v4.sin_addr.s_addr); - /* FIXME: ip_nonlocal_bind sysctl support. */ - if (addr->v4.sin_addr.s_addr != INADDR_ANY && ret != RTN_LOCAL) + if (addr->v4.sin_addr.s_addr != INADDR_ANY && + ret != RTN_LOCAL && + !sp->inet.freebind && + !sysctl_ip_nonlocal_bind) return 0; + return 1; } 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..86073df 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -92,6 +92,17 @@ static sctp_disposition_t sctp_sf_shut_8_4_5(const struct sctp_endpoint *ep, sctp_cmd_seq_t *commands); static struct sctp_sackhdr *sctp_sm_pull_sack(struct sctp_chunk *chunk); +static sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, + __u16 error, + const struct sctp_association *asoc, + struct sctp_transport *transport); + +static sctp_disposition_t sctp_sf_violation_chunklen( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands); /* Small helper function that checks if the chunk length * is of the appropriate length. The 'required_length' argument @@ -533,6 +544,9 @@ 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 +789,7 @@ 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 +1032,22 @@ 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 +2120,7 @@ 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 +2180,7 @@ 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 +2303,7 @@ 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 +2315,8 @@ 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 +2339,12 @@ 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) +static 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 +2353,7 @@ 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; } /* @@ -3672,7 +3698,8 @@ sctp_disposition_t sctp_sf_violation(const struct sctp_endpoint *ep, * * Generate an ABORT chunk and terminate the association. */ -sctp_disposition_t sctp_sf_violation_chunklen(const struct sctp_endpoint *ep, +static sctp_disposition_t sctp_sf_violation_chunklen( + const struct sctp_endpoint *ep, const struct sctp_association *asoc, const sctp_subtype_t type, void *arg, @@ -3805,6 +3832,10 @@ 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 +4620,7 @@ 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 +4634,59 @@ 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 +4696,26 @@ 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; + struct sctp_chunk *repl = NULL; + int attempts = asoc->init_err_counter + 1; - timeout = asoc->timeouts[timer]; - attempts = asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1; - repl = NULL; - - 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 +4724,6 @@ 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 0b338ec..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; + SCTP_DEBUG_PRINTK("%s - sk: %p, sockaddr: %p, addr_len: %d\n", + __FUNCTION__, sk, addr, addr_len); - 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; - } - - /* 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 @@ -4368,15 +4540,11 @@ static struct sk_buff *sctp_skb_recv_datagram(struct sock *sk, int flags, * However, this function was corrent in any case. 8) */ if (flags & MSG_PEEK) { - unsigned long cpu_flags; - - sctp_spin_lock_irqsave(&sk->sk_receive_queue.lock, - cpu_flags); + spin_lock_bh(&sk->sk_receive_queue.lock); skb = skb_peek(&sk->sk_receive_queue); if (skb) atomic_inc(&skb->users); - sctp_spin_unlock_irqrestore(&sk->sk_receive_queue.lock, - cpu_flags); + spin_unlock_bh(&sk->sk_receive_queue.lock); } else { skb = skb_dequeue(&sk->sk_receive_queue); } @@ -4600,8 +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; @@ -4686,6 +4853,7 @@ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, struct sctp_endpoint *newep = newsp->ep; struct sk_buff *skb, *tmp; struct sctp_ulpevent *event; + int flags = 0; /* Migrate socket buffer sizes and all the socket level options to the * new socket. @@ -4707,6 +4875,17 @@ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, sctp_sk(newsk)->bind_hash = pp; inet_sk(newsk)->num = inet_sk(oldsk)->num; + /* Copy the bind_addr list from the original endpoint to the new + * endpoint so that we can handle restarts properly + */ + if (assoc->peer.ipv4_address) + flags |= SCTP_ADDR4_PEERSUPP; + if (assoc->peer.ipv6_address) + flags |= SCTP_ADDR6_PEERSUPP; + sctp_bind_addr_copy(&newsp->ep->base.bind_addr, + &oldsp->ep->base.bind_addr, + SCTP_SCOPE_GLOBAL, GFP_KERNEL, flags); + /* Move any messages in the old socket's receive queue that are for the * peeled off association to the new socket's receive queue. */ 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/socket.c b/net/socket.c index cec0cb3..6f2a1788 100644 --- a/net/socket.c +++ b/net/socket.c @@ -81,6 +81,7 @@ #include <linux/syscalls.h> #include <linux/compat.h> #include <linux/kmod.h> +#include <linux/audit.h> #ifdef CONFIG_NET_RADIO #include <linux/wireless.h> /* Note : will define WIRELESS_EXT */ @@ -226,7 +227,7 @@ int move_addr_to_kernel(void __user *uaddr, int ulen, void *kaddr) return 0; if(copy_from_user(kaddr,uaddr,ulen)) return -EFAULT; - return 0; + return audit_sockaddr(ulen, kaddr); } /** @@ -382,9 +383,8 @@ int sock_map_fd(struct socket *sock) goto out; } - sprintf(name, "[%lu]", SOCK_INODE(sock)->i_ino); + this.len = sprintf(name, "[%lu]", SOCK_INODE(sock)->i_ino); this.name = name; - this.len = strlen(name); this.hash = SOCK_INODE(sock)->i_ino; file->f_dentry = d_alloc(sock_mnt->mnt_sb->s_root, &this); @@ -1906,7 +1906,11 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args) /* copy_from_user should be SMP safe. */ if (copy_from_user(a, args, nargs[call])) return -EFAULT; - + + err = audit_socketcall(nargs[call]/sizeof(unsigned long), a); + if (err) + return err; + a0=a[0]; a1=a[1]; 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(¤t->sighand->siglock, irqflags); - *oldset = current->blocked; - siginitsetinv(¤t->blocked, sigallow & ~oldset->sig[0]); - recalc_sigpending(); - spin_unlock_irqrestore(¤t->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(¤t->sighand->siglock, irqflags); - current->blocked = *oldset; - recalc_sigpending(); - spin_unlock_irqrestore(¤t->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..62a0734 100644 --- a/net/sunrpc/sunrpc_syms.c +++ b/net/sunrpc/sunrpc_syms.c @@ -41,7 +41,9 @@ EXPORT_SYMBOL(rpc_release_task); /* RPC client functions */ EXPORT_SYMBOL(rpc_create_client); +EXPORT_SYMBOL(rpc_new_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 +63,6 @@ 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 +130,10 @@ 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/svcauth_unix.c b/net/sunrpc/svcauth_unix.c index 2b99b40..d6baf6f 100644 --- a/net/sunrpc/svcauth_unix.c +++ b/net/sunrpc/svcauth_unix.c @@ -8,6 +8,7 @@ #include <linux/err.h> #include <linux/seq_file.h> #include <linux/hash.h> +#include <linux/string.h> #define RPCDBG_FACILITY RPCDBG_AUTH @@ -20,14 +21,6 @@ */ -static char *strdup(char *s) -{ - char *rv = kmalloc(strlen(s)+1, GFP_KERNEL); - if (rv) - strcpy(rv, s); - return rv; -} - struct unix_domain { struct auth_domain h; int addr_changes; @@ -55,7 +48,7 @@ struct auth_domain *unix_domain_find(char *name) if (new == NULL) return NULL; cache_init(&new->h.h); - new->h.name = strdup(name); + new->h.name = kstrdup(name, GFP_KERNEL); new->h.flavour = RPC_AUTH_UNIX; new->addr_changes = 0; new->h.h.expiry_time = NEVER; diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 0590703..56db8f1 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1185,8 +1185,8 @@ svc_recv(struct svc_serv *serv, struct svc_rqst *rqstp, long timeout) arg->page_len = (pages-2)*PAGE_SIZE; arg->len = (pages-1)*PAGE_SIZE; arg->tail[0].iov_len = 0; - - try_to_freeze(PF_FREEZE); + + try_to_freeze(); if (signalled()) return -EINTR; @@ -1227,7 +1227,7 @@ svc_recv(struct svc_serv *serv, struct svc_rqst *rqstp, long timeout) schedule_timeout(timeout); - try_to_freeze(PF_FREEZE); + try_to_freeze(); spin_lock_bh(&serv->sv_lock); remove_wait_queue(&rqstp->rq_wait, &wait); 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..269f217 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 %Zd 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 %Zu 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/unix/af_unix.c b/net/unix/af_unix.c index c478fc8..c420eba 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -770,33 +770,12 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) err = path_lookup(sunaddr->sun_path, LOOKUP_PARENT, &nd); if (err) goto out_mknod_parent; - /* - * Yucky last component or no last component at all? - * (foo/., foo/.., /////) - */ - err = -EEXIST; - if (nd.last_type != LAST_NORM) - goto out_mknod; - /* - * Lock the directory. - */ - down(&nd.dentry->d_inode->i_sem); - /* - * Do the final lookup. - */ - dentry = lookup_hash(&nd.last, nd.dentry); + + dentry = lookup_create(&nd, 0); err = PTR_ERR(dentry); if (IS_ERR(dentry)) goto out_mknod_unlock; - err = -ENOENT; - /* - * Special case - lookup gave negative, but... we had foo/bar/ - * From the vfs_mknod() POV we just have a negative dentry - - * all is fine. Let's be bastards - you had / on the end, you've - * been asking for (non-existent) directory. -ENOENT for you. - */ - if (nd.last.name[nd.last.len] && !dentry->d_inode) - goto out_mknod_dput; + /* * All right, let's create it. */ @@ -845,7 +824,6 @@ out_mknod_dput: dput(dentry); out_mknod_unlock: up(&nd.dentry->d_inode->i_sem); -out_mknod: path_release(&nd); out_mknod_parent: if (err==-EEXIST) diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index 2a24b24..04bec04 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -29,6 +29,10 @@ * 2000-11-14 Henner Eisen Closing datalink from NETDEV_GOING_DOWN * 2002-10-06 Arnaldo C. Melo Get rid of cli/sti, move proc stuff to * x25_proc.c, using seq_file + * 2005-04-02 Shaun Pereira Selective sub address matching + * with call user data + * 2005-04-15 Shaun Pereira Fast select with no restriction on + * response */ #include <linux/config.h> @@ -219,7 +223,8 @@ static void x25_insert_socket(struct sock *sk) * Note: if a listening socket has cud set it must only get calls * with matching cud. */ -static struct sock *x25_find_listener(struct x25_address *addr, struct x25_calluserdata *calluserdata) +static struct sock *x25_find_listener(struct x25_address *addr, + struct sk_buff *skb) { struct sock *s; struct sock *next_best; @@ -230,22 +235,23 @@ static struct sock *x25_find_listener(struct x25_address *addr, struct x25_callu sk_for_each(s, node, &x25_list) if ((!strcmp(addr->x25_addr, - x25_sk(s)->source_addr.x25_addr) || - !strcmp(addr->x25_addr, - null_x25_address.x25_addr)) && - s->sk_state == TCP_LISTEN) { - + x25_sk(s)->source_addr.x25_addr) || + !strcmp(addr->x25_addr, + null_x25_address.x25_addr)) && + s->sk_state == TCP_LISTEN) { /* * Found a listening socket, now check the incoming * call user data vs this sockets call user data */ - if (x25_check_calluserdata(&x25_sk(s)->calluserdata, calluserdata)) { - sock_hold(s); - goto found; - } - if (x25_sk(s)->calluserdata.cudlength == 0) { + if(skb->len > 0 && x25_sk(s)->cudmatchlength > 0) { + if((memcmp(x25_sk(s)->calluserdata.cuddata, + skb->data, + x25_sk(s)->cudmatchlength)) == 0) { + sock_hold(s); + goto found; + } + } else next_best = s; - } } if (next_best) { s = next_best; @@ -497,6 +503,9 @@ static int x25_create(struct socket *sock, int protocol) x25->t23 = sysctl_x25_clear_request_timeout; x25->t2 = sysctl_x25_ack_holdback_timeout; x25->state = X25_STATE_0; + x25->cudmatchlength = 0; + x25->accptapprv = X25_DENY_ACCPT_APPRV; /* normally no cud */ + /* on call accept */ x25->facilities.winsize_in = X25_DEFAULT_WINDOW_SIZE; x25->facilities.winsize_out = X25_DEFAULT_WINDOW_SIZE; @@ -545,6 +554,8 @@ static struct sock *x25_make_new(struct sock *osk) x25->t2 = ox25->t2; x25->facilities = ox25->facilities; x25->qbitincl = ox25->qbitincl; + x25->cudmatchlength = ox25->cudmatchlength; + x25->accptapprv = ox25->accptapprv; x25_init_timers(sk); out: @@ -822,7 +833,6 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb, struct x25_sock *makex25; struct x25_address source_addr, dest_addr; struct x25_facilities facilities; - struct x25_calluserdata calluserdata; int len, rc; /* @@ -845,19 +855,10 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb, skb_pull(skb,len); /* - * Incoming Call User Data. - */ - if (skb->len >= 0) { - memcpy(calluserdata.cuddata, skb->data, skb->len); - calluserdata.cudlength = skb->len; - } - - skb_push(skb,len); - - /* * Find a listener for the particular address/cud pair. */ - sk = x25_find_listener(&source_addr,&calluserdata); + sk = x25_find_listener(&source_addr,skb); + skb_push(skb,len); /* * We can't accept the Call Request. @@ -900,11 +901,23 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb, makex25->neighbour = nb; makex25->facilities = facilities; makex25->vc_facil_mask = x25_sk(sk)->vc_facil_mask; - makex25->calluserdata = calluserdata; - - x25_write_internal(make, X25_CALL_ACCEPTED); + /* ensure no reverse facil on accept */ + makex25->vc_facil_mask &= ~X25_MASK_REVERSE; + makex25->cudmatchlength = x25_sk(sk)->cudmatchlength; + + /* Normally all calls are accepted immediatly */ + if(makex25->accptapprv & X25_DENY_ACCPT_APPRV) { + x25_write_internal(make, X25_CALL_ACCEPTED); + makex25->state = X25_STATE_3; + } - makex25->state = X25_STATE_3; + /* + * Incoming Call User Data. + */ + if (skb->len >= 0) { + memcpy(makex25->calluserdata.cuddata, skb->data, skb->len); + makex25->calluserdata.cudlength = skb->len; + } sk->sk_ack_backlog++; @@ -1288,7 +1301,8 @@ static int x25_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) if (facilities.throughput < 0x03 || facilities.throughput > 0xDD) break; - if (facilities.reverse && facilities.reverse != 1) + if (facilities.reverse && + (facilities.reverse | 0x81)!= 0x81) break; x25->facilities = facilities; rc = 0; @@ -1325,6 +1339,44 @@ static int x25_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) break; } + case SIOCX25SCUDMATCHLEN: { + struct x25_subaddr sub_addr; + rc = -EINVAL; + if(sk->sk_state != TCP_CLOSE) + break; + rc = -EFAULT; + if (copy_from_user(&sub_addr, argp, + sizeof(sub_addr))) + break; + rc = -EINVAL; + if(sub_addr.cudmatchlength > X25_MAX_CUD_LEN) + break; + x25->cudmatchlength = sub_addr.cudmatchlength; + rc = 0; + break; + } + + case SIOCX25CALLACCPTAPPRV: { + rc = -EINVAL; + if (sk->sk_state != TCP_CLOSE) + break; + x25->accptapprv = X25_ALLOW_ACCPT_APPRV; + rc = 0; + break; + } + + case SIOCX25SENDCALLACCPT: { + rc = -EINVAL; + if (sk->sk_state != TCP_ESTABLISHED) + break; + if (x25->accptapprv) /* must call accptapprv above */ + break; + x25_write_internal(sk, X25_CALL_ACCEPTED); + x25->state = X25_STATE_3; + rc = 0; + break; + } + default: rc = dev_ioctl(cmd, argp); break; diff --git a/net/x25/x25_facilities.c b/net/x25/x25_facilities.c index a21bdb9..54278b9 100644 --- a/net/x25/x25_facilities.c +++ b/net/x25/x25_facilities.c @@ -17,6 +17,8 @@ * X.25 001 Split from x25_subr.c * mar/20/00 Daniela Squassoni Disabling/enabling of facilities * negotiation. + * apr/14/05 Shaun Pereira - Allow fast select with no restriction + * on response. */ #include <linux/kernel.h> @@ -43,9 +45,31 @@ int x25_parse_facilities(struct sk_buff *skb, case X25_FAC_CLASS_A: switch (*p) { case X25_FAC_REVERSE: - facilities->reverse = p[1] & 0x01; - *vc_fac_mask |= X25_MASK_REVERSE; - break; + if((p[1] & 0x81) == 0x81) { + facilities->reverse = p[1] & 0x81; + *vc_fac_mask |= X25_MASK_REVERSE; + break; + } + + if((p[1] & 0x01) == 0x01) { + facilities->reverse = p[1] & 0x01; + *vc_fac_mask |= X25_MASK_REVERSE; + break; + } + + if((p[1] & 0x80) == 0x80) { + facilities->reverse = p[1] & 0x80; + *vc_fac_mask |= X25_MASK_REVERSE; + break; + } + + if(p[1] == 0x00) { + facilities->reverse + = X25_DEFAULT_REVERSE; + *vc_fac_mask |= X25_MASK_REVERSE; + break; + } + case X25_FAC_THROUGHPUT: facilities->throughput = p[1]; *vc_fac_mask |= X25_MASK_THROUGHPUT; @@ -122,7 +146,7 @@ int x25_create_facilities(unsigned char *buffer, if (facilities->reverse && (facil_mask & X25_MASK_REVERSE)) { *p++ = X25_FAC_REVERSE; - *p++ = !!facilities->reverse; + *p++ = facilities->reverse; } if (facilities->throughput && (facil_mask & X25_MASK_THROUGHPUT)) { @@ -171,7 +195,7 @@ int x25_negotiate_facilities(struct sk_buff *skb, struct sock *sk, /* * They want reverse charging, we won't accept it. */ - if (theirs.reverse && ours->reverse) { + if ((theirs.reverse & 0x01 ) && (ours->reverse & 0x01)) { SOCK_DEBUG(sk, "X.25: rejecting reverse charging request"); return -1; } diff --git a/net/x25/x25_subr.c b/net/x25/x25_subr.c index 183fea3..7fd872ad 100644 --- a/net/x25/x25_subr.c +++ b/net/x25/x25_subr.c @@ -19,6 +19,8 @@ * mar/20/00 Daniela Squassoni Disabling/enabling of facilities * negotiation. * jun/24/01 Arnaldo C. Melo use skb_queue_purge, cleanups + * apr/04/15 Shaun Pereira Fast select with no + * restriction on response. */ #include <linux/kernel.h> @@ -127,8 +129,12 @@ void x25_write_internal(struct sock *sk, int frametype) len += 1 + X25_ADDR_LEN + X25_MAX_FAC_LEN + X25_MAX_CUD_LEN; break; - case X25_CALL_ACCEPTED: - len += 1 + X25_MAX_FAC_LEN + X25_MAX_CUD_LEN; + case X25_CALL_ACCEPTED: /* fast sel with no restr on resp */ + if(x25->facilities.reverse & 0x80) { + len += 1 + X25_MAX_FAC_LEN + X25_MAX_CUD_LEN; + } else { + len += 1 + X25_MAX_FAC_LEN; + } break; case X25_CLEAR_REQUEST: case X25_RESET_REQUEST: @@ -203,9 +209,16 @@ void x25_write_internal(struct sock *sk, int frametype) x25->vc_facil_mask); dptr = skb_put(skb, len); memcpy(dptr, facilities, len); - dptr = skb_put(skb, x25->calluserdata.cudlength); - memcpy(dptr, x25->calluserdata.cuddata, - x25->calluserdata.cudlength); + + /* fast select with no restriction on response + allows call user data. Userland must + ensure it is ours and not theirs */ + if(x25->facilities.reverse & 0x80) { + dptr = skb_put(skb, + x25->calluserdata.cudlength); + memcpy(dptr, x25->calluserdata.cuddata, + x25->calluserdata.cudlength); + } x25->calluserdata.cudlength = 0; break; @@ -354,21 +367,3 @@ void x25_check_rbuf(struct sock *sk) } } -/* - * Compare 2 calluserdata structures, used to find correct listening sockets - * when call user data is used. - */ -int x25_check_calluserdata(struct x25_calluserdata *ours, struct x25_calluserdata *theirs) -{ - int i; - if (ours->cudlength != theirs->cudlength) - return 0; - - for (i=0;i<ours->cudlength;i++) { - if (ours->cuddata[i] != theirs->cuddata[i]) { - return 0; - } - } - return 1; -} - diff --git a/net/xfrm/xfrm_algo.c b/net/xfrm/xfrm_algo.c index 080aae2..2f4531f 100644 --- a/net/xfrm/xfrm_algo.c +++ b/net/xfrm/xfrm_algo.c @@ -698,7 +698,7 @@ int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer) return -ENOMEM; if (skb1->sk) - skb_set_owner_w(skb, skb1->sk); + skb_set_owner_w(skb2, skb1->sk); /* Looking around. Are we still alive? * OK, link new skb, drop old one */ diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 55ed979..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) @@ -216,8 +215,8 @@ out: expired: read_unlock(&xp->lock); - km_policy_expired(xp, dir, 1); - xfrm_policy_delete(xp, dir); + if (!xfrm_policy_delete(xp, dir)) + km_policy_expired(xp, dir, 1); xfrm_pol_put(xp); } @@ -555,7 +554,7 @@ static struct xfrm_policy *__xfrm_policy_unlink(struct xfrm_policy *pol, return NULL; } -void xfrm_policy_delete(struct xfrm_policy *pol, int dir) +int xfrm_policy_delete(struct xfrm_policy *pol, int dir) { write_lock_bh(&xfrm_policy_lock); pol = __xfrm_policy_unlink(pol, dir); @@ -564,7 +563,9 @@ void xfrm_policy_delete(struct xfrm_policy *pol, int dir) if (dir < XFRM_POLICY_MAX) atomic_inc(&flow_cache_genid); xfrm_policy_kill(pol); + return 0; } + return -ENOENT; } int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol) @@ -1136,7 +1137,7 @@ int xfrm_bundle_ok(struct xfrm_dst *first, struct flowi *fl, int family) struct xfrm_dst *last; u32 mtu; - if (!dst_check(dst->path, 0) || + if (!dst_check(dst->path, ((struct xfrm_dst *)dst)->path_cookie) || (dst->dev && !netif_running(dst->dev))) return 0; @@ -1156,7 +1157,7 @@ int xfrm_bundle_ok(struct xfrm_dst *first, struct flowi *fl, int family) xdst->child_mtu_cached = mtu; } - if (!dst_check(xdst->route, 0)) + if (!dst_check(xdst->route, xdst->route_cookie)) return 0; mtu = dst_mtu(xdst->route); if (xdst->route_mtu_cached != mtu) { diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index d11747c..9d206c2 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -50,7 +50,7 @@ static DEFINE_SPINLOCK(xfrm_state_gc_lock); static int xfrm_state_gc_flush_bundles; -static void __xfrm_state_delete(struct xfrm_state *x); +static int __xfrm_state_delete(struct xfrm_state *x); static struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned short family); static void xfrm_state_put_afinfo(struct xfrm_state_afinfo *afinfo); @@ -154,6 +154,7 @@ static void xfrm_timer_handler(unsigned long data) next = tmo; } + x->km.dying = warn; if (warn) km_state_expired(x, 0); resched: @@ -169,9 +170,8 @@ expired: next = 2; goto resched; } - if (x->id.spi != 0) + if (!__xfrm_state_delete(x) && x->id.spi) km_state_expired(x, 1); - __xfrm_state_delete(x); out: spin_unlock(&x->lock); @@ -215,8 +215,10 @@ void __xfrm_state_destroy(struct xfrm_state *x) } EXPORT_SYMBOL(__xfrm_state_destroy); -static void __xfrm_state_delete(struct xfrm_state *x) +static int __xfrm_state_delete(struct xfrm_state *x) { + int err = -ESRCH; + if (x->km.state != XFRM_STATE_DEAD) { x->km.state = XFRM_STATE_DEAD; spin_lock(&xfrm_state_lock); @@ -245,14 +247,21 @@ static void __xfrm_state_delete(struct xfrm_state *x) * is what we are dropping here. */ atomic_dec(&x->refcnt); + err = 0; } + + return err; } -void xfrm_state_delete(struct xfrm_state *x) +int xfrm_state_delete(struct xfrm_state *x) { + int err; + spin_lock_bh(&x->lock); - __xfrm_state_delete(x); + err = __xfrm_state_delete(x); spin_unlock_bh(&x->lock); + + return err; } EXPORT_SYMBOL(xfrm_state_delete); @@ -557,16 +566,18 @@ int xfrm_state_check_expire(struct xfrm_state *x) if (x->curlft.bytes >= x->lft.hard_byte_limit || x->curlft.packets >= x->lft.hard_packet_limit) { - km_state_expired(x, 1); - if (!mod_timer(&x->timer, jiffies + XFRM_ACQ_EXPIRES*HZ)) + x->km.state = XFRM_STATE_EXPIRED; + if (!mod_timer(&x->timer, jiffies)) xfrm_state_hold(x); return -EINVAL; } if (!x->km.dying && (x->curlft.bytes >= x->lft.soft_byte_limit || - x->curlft.packets >= x->lft.soft_packet_limit)) + x->curlft.packets >= x->lft.soft_packet_limit)) { + x->km.dying = 1; km_state_expired(x, 0); + } return 0; } EXPORT_SYMBOL(xfrm_state_check_expire); @@ -796,34 +807,56 @@ EXPORT_SYMBOL(xfrm_replay_advance); static struct list_head xfrm_km_list = LIST_HEAD_INIT(xfrm_km_list); static DEFINE_RWLOCK(xfrm_km_lock); -static void km_state_expired(struct xfrm_state *x, int hard) +void km_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) { struct xfrm_mgr *km; - if (hard) - x->km.state = XFRM_STATE_EXPIRED; - else - x->km.dying = 1; + read_lock(&xfrm_km_lock); + list_for_each_entry(km, &xfrm_km_list, list) + if (km->notify_policy) + km->notify_policy(xp, dir, c); + read_unlock(&xfrm_km_lock); +} +void km_state_notify(struct xfrm_state *x, struct km_event *c) +{ + struct xfrm_mgr *km; read_lock(&xfrm_km_lock); list_for_each_entry(km, &xfrm_km_list, list) - km->notify(x, hard); + if (km->notify) + km->notify(x, c); read_unlock(&xfrm_km_lock); +} + +EXPORT_SYMBOL(km_policy_notify); +EXPORT_SYMBOL(km_state_notify); + +static void km_state_expired(struct xfrm_state *x, int hard) +{ + struct km_event c; + + c.data.hard = hard; + c.event = XFRM_MSG_EXPIRE; + km_state_notify(x, &c); if (hard) wake_up(&km_waitq); } +/* + * We send to all registered managers regardless of failure + * We are happy with one success +*/ static int km_query(struct xfrm_state *x, struct xfrm_tmpl *t, struct xfrm_policy *pol) { - int err = -EINVAL; + int err = -EINVAL, acqret; struct xfrm_mgr *km; read_lock(&xfrm_km_lock); list_for_each_entry(km, &xfrm_km_list, list) { - err = km->acquire(x, t, pol, XFRM_POLICY_OUT); - if (!err) - break; + acqret = km->acquire(x, t, pol, XFRM_POLICY_OUT); + if (!acqret) + err = acqret; } read_unlock(&xfrm_km_lock); return err; @@ -848,13 +881,11 @@ EXPORT_SYMBOL(km_new_mapping); void km_policy_expired(struct xfrm_policy *pol, int dir, int hard) { - struct xfrm_mgr *km; + struct km_event c; - read_lock(&xfrm_km_lock); - list_for_each_entry(km, &xfrm_km_list, list) - if (km->notify_policy) - km->notify_policy(pol, dir, hard); - read_unlock(&xfrm_km_lock); + c.data.hard = hard; + c.event = XFRM_MSG_POLEXPIRE; + km_policy_notify(pol, dir, &c); if (hard) wake_up(&km_waitq); @@ -1024,6 +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 5ddda2c..ecade48 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -34,14 +34,21 @@ static int verify_one_alg(struct rtattr **xfrma, enum xfrm_attr_type_t type) { struct rtattr *rt = xfrma[type - 1]; struct xfrm_algo *algp; + int len; if (!rt) return 0; - if ((rt->rta_len - sizeof(*rt)) < sizeof(*algp)) + len = (rt->rta_len - sizeof(*rt)) - sizeof(*algp); + if (len < 0) return -EINVAL; algp = RTA_DATA(rt); + + len -= (algp->alg_key_len + 7U) / 8; + if (len < 0) + return -EINVAL; + switch (type) { case XFRMA_ALG_AUTH: if (!algp->alg_key_len && @@ -162,6 +169,7 @@ static int attach_one_algo(struct xfrm_algo **algpp, u8 *props, struct rtattr *rta = u_arg; struct xfrm_algo *p, *ualg; struct xfrm_algo_desc *algo; + int len; if (!rta) return 0; @@ -173,11 +181,12 @@ static int attach_one_algo(struct xfrm_algo **algpp, u8 *props, return -ENOSYS; *props = algo->desc.sadb_alg_id; - p = kmalloc(sizeof(*ualg) + ualg->alg_key_len, GFP_KERNEL); + len = sizeof(*ualg) + (ualg->alg_key_len + 7U) / 8; + p = kmalloc(len, GFP_KERNEL); if (!p) return -ENOMEM; - memcpy(p, ualg, sizeof(*ualg) + ualg->alg_key_len); + memcpy(p, ualg, len); *algpp = p; return 0; } @@ -240,17 +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; @@ -268,6 +270,7 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) struct xfrm_usersa_info *p = NLMSG_DATA(nlh); struct xfrm_state *x; int err; + struct km_event c; err = verify_newsa_info(p, (struct rtattr **) xfrma); if (err) @@ -277,6 +280,7 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) if (!x) return err; + xfrm_state_hold(x); if (nlh->nlmsg_type == XFRM_MSG_NEWSA) err = xfrm_state_add(x); else @@ -285,14 +289,24 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) if (err < 0) { x->km.state = XFRM_STATE_DEAD; xfrm_state_put(x); + goto out; } + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + c.event = nlh->nlmsg_type; + + km_state_notify(x, &c); +out: + xfrm_state_put(x); return err; } static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) { struct xfrm_state *x; + int err; + struct km_event c; struct xfrm_usersa_id *p = NLMSG_DATA(nlh); x = xfrm_state_lookup(&p->daddr, p->spi, p->proto, p->family); @@ -304,10 +318,19 @@ static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) return -EPERM; } - xfrm_state_delete(x); + err = xfrm_state_delete(x); + if (err < 0) { + xfrm_state_put(x); + return err; + } + + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + c.event = nlh->nlmsg_type; + km_state_notify(x, &c); xfrm_state_put(x); - return 0; + return err; } static void copy_to_user_state(struct xfrm_state *x, struct xfrm_usersa_info *p) @@ -672,6 +695,7 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr { struct xfrm_userpolicy_info *p = NLMSG_DATA(nlh); struct xfrm_policy *xp; + struct km_event c; int err; int excl; @@ -683,6 +707,10 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr if (!xp) return err; + /* shouldnt excl be based on nlh flags?? + * Aha! this is anti-netlink really i.e more pfkey derived + * in netlink excl is a flag and you wouldnt need + * a type XFRM_MSG_UPDPOLICY - JHS */ excl = nlh->nlmsg_type == XFRM_MSG_NEWPOLICY; err = xfrm_policy_insert(p->dir, xp, excl); if (err) { @@ -690,6 +718,11 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr return err; } + c.event = nlh->nlmsg_type; + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_policy_notify(xp, p->dir, &c); + xfrm_pol_put(xp); return 0; @@ -807,6 +840,7 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr struct xfrm_policy *xp; struct xfrm_userpolicy_id *p; int err; + struct km_event c; int delete; p = NLMSG_DATA(nlh); @@ -834,6 +868,12 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr NETLINK_CB(skb).pid, MSG_DONTWAIT); } + } else { + c.data.byid = p->index; + c.event = nlh->nlmsg_type; + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_policy_notify(xp, p->dir, &c); } xfrm_pol_put(xp); @@ -843,15 +883,28 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr static int xfrm_flush_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) { + struct km_event c; struct xfrm_usersa_flush *p = NLMSG_DATA(nlh); xfrm_state_flush(p->proto); + c.data.proto = p->proto; + c.event = nlh->nlmsg_type; + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_state_notify(NULL, &c); + return 0; } static int xfrm_flush_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) { + struct km_event c; + xfrm_policy_flush(); + c.event = nlh->nlmsg_type; + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_policy_notify(NULL, 0, &c); return 0; } @@ -1060,15 +1113,16 @@ nlmsg_failure: return -1; } -static int xfrm_send_state_notify(struct xfrm_state *x, int hard) +static int xfrm_exp_state_notify(struct xfrm_state *x, struct km_event *c) { struct sk_buff *skb; + int len = NLMSG_LENGTH(sizeof(struct xfrm_user_expire)); - skb = alloc_skb(sizeof(struct xfrm_user_expire) + 16, GFP_ATOMIC); + skb = alloc_skb(len, GFP_ATOMIC); if (skb == NULL) return -ENOMEM; - if (build_expire(skb, x, hard) < 0) + if (build_expire(skb, x, c->data.hard) < 0) BUG(); NETLINK_CB(skb).dst_groups = XFRMGRP_EXPIRE; @@ -1076,6 +1130,131 @@ static int xfrm_send_state_notify(struct xfrm_state *x, int hard) return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_EXPIRE, GFP_ATOMIC); } +static int xfrm_notify_sa_flush(struct km_event *c) +{ + struct xfrm_usersa_flush *p; + struct nlmsghdr *nlh; + struct sk_buff *skb; + unsigned char *b; + int len = NLMSG_LENGTH(sizeof(struct xfrm_usersa_flush)); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + nlh = NLMSG_PUT(skb, c->pid, c->seq, + XFRM_MSG_FLUSHSA, sizeof(*p)); + nlh->nlmsg_flags = 0; + + p = NLMSG_DATA(nlh); + p->proto = c->data.proto; + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_SA, GFP_ATOMIC); + +nlmsg_failure: + kfree_skb(skb); + return -1; +} + +static int inline xfrm_sa_len(struct xfrm_state *x) +{ + int l = 0; + if (x->aalg) + l += RTA_SPACE(sizeof(*x->aalg) + (x->aalg->alg_key_len+7)/8); + if (x->ealg) + l += RTA_SPACE(sizeof(*x->ealg) + (x->ealg->alg_key_len+7)/8); + if (x->calg) + l += RTA_SPACE(sizeof(*x->calg)); + if (x->encap) + l += RTA_SPACE(sizeof(*x->encap)); + + return l; +} + +static int xfrm_notify_sa(struct xfrm_state *x, struct km_event *c) +{ + struct xfrm_usersa_info *p; + struct xfrm_usersa_id *id; + struct nlmsghdr *nlh; + struct sk_buff *skb; + unsigned char *b; + int len = xfrm_sa_len(x); + int headlen; + + headlen = sizeof(*p); + if (c->event == XFRM_MSG_DELSA) { + len += RTA_SPACE(headlen); + headlen = sizeof(*id); + } + len += NLMSG_SPACE(headlen); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, headlen); + nlh->nlmsg_flags = 0; + + p = NLMSG_DATA(nlh); + if (c->event == XFRM_MSG_DELSA) { + id = NLMSG_DATA(nlh); + memcpy(&id->daddr, &x->id.daddr, sizeof(id->daddr)); + id->spi = x->id.spi; + id->family = x->props.family; + id->proto = x->id.proto; + + p = RTA_DATA(__RTA_PUT(skb, XFRMA_SA, sizeof(*p))); + } + + copy_to_user_state(x, p); + + if (x->aalg) + RTA_PUT(skb, XFRMA_ALG_AUTH, + sizeof(*(x->aalg))+(x->aalg->alg_key_len+7)/8, x->aalg); + if (x->ealg) + RTA_PUT(skb, XFRMA_ALG_CRYPT, + sizeof(*(x->ealg))+(x->ealg->alg_key_len+7)/8, x->ealg); + if (x->calg) + RTA_PUT(skb, XFRMA_ALG_COMP, sizeof(*(x->calg)), x->calg); + + if (x->encap) + RTA_PUT(skb, XFRMA_ENCAP, sizeof(*x->encap), x->encap); + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_SA, GFP_ATOMIC); + +nlmsg_failure: +rtattr_failure: + kfree_skb(skb); + return -1; +} + +static int xfrm_send_state_notify(struct xfrm_state *x, struct km_event *c) +{ + + switch (c->event) { + case XFRM_MSG_EXPIRE: + return xfrm_exp_state_notify(x, c); + case XFRM_MSG_DELSA: + case XFRM_MSG_UPDSA: + case XFRM_MSG_NEWSA: + return xfrm_notify_sa(x, c); + case XFRM_MSG_FLUSHSA: + return xfrm_notify_sa_flush(c); + default: + printk("xfrm_user: Unknown SA event %d\n", c->event); + break; + } + + return 0; + +} + static int build_acquire(struct sk_buff *skb, struct xfrm_state *x, struct xfrm_tmpl *xt, struct xfrm_policy *xp, int dir) @@ -1209,7 +1388,7 @@ nlmsg_failure: return -1; } -static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, int hard) +static int xfrm_exp_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) { struct sk_buff *skb; size_t len; @@ -1220,7 +1399,7 @@ static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, int hard) if (skb == NULL) return -ENOMEM; - if (build_polexpire(skb, xp, dir, hard) < 0) + if (build_polexpire(skb, xp, dir, c->data.hard) < 0) BUG(); NETLINK_CB(skb).dst_groups = XFRMGRP_EXPIRE; @@ -1228,6 +1407,103 @@ static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, int hard) return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_EXPIRE, GFP_ATOMIC); } +static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + struct xfrm_userpolicy_info *p; + struct xfrm_userpolicy_id *id; + struct nlmsghdr *nlh; + struct sk_buff *skb; + unsigned char *b; + int len = RTA_SPACE(sizeof(struct xfrm_user_tmpl) * xp->xfrm_nr); + int headlen; + + headlen = sizeof(*p); + if (c->event == XFRM_MSG_DELPOLICY) { + len += RTA_SPACE(headlen); + headlen = sizeof(*id); + } + len += NLMSG_SPACE(headlen); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, headlen); + + p = NLMSG_DATA(nlh); + if (c->event == XFRM_MSG_DELPOLICY) { + id = NLMSG_DATA(nlh); + memset(id, 0, sizeof(*id)); + id->dir = dir; + if (c->data.byid) + id->index = xp->index; + else + memcpy(&id->sel, &xp->selector, sizeof(id->sel)); + + p = RTA_DATA(__RTA_PUT(skb, XFRMA_POLICY, sizeof(*p))); + } + + nlh->nlmsg_flags = 0; + + copy_to_user_policy(xp, p, dir); + if (copy_to_user_tmpl(xp, skb) < 0) + goto nlmsg_failure; + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_POLICY, GFP_ATOMIC); + +nlmsg_failure: +rtattr_failure: + kfree_skb(skb); + return -1; +} + +static int xfrm_notify_policy_flush(struct km_event *c) +{ + struct nlmsghdr *nlh; + struct sk_buff *skb; + unsigned char *b; + int len = NLMSG_LENGTH(0); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + + nlh = NLMSG_PUT(skb, c->pid, c->seq, XFRM_MSG_FLUSHPOLICY, 0); + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_POLICY, GFP_ATOMIC); + +nlmsg_failure: + kfree_skb(skb); + return -1; +} + +static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + + switch (c->event) { + case XFRM_MSG_NEWPOLICY: + case XFRM_MSG_UPDPOLICY: + case XFRM_MSG_DELPOLICY: + return xfrm_notify_policy(xp, dir, c); + case XFRM_MSG_FLUSHPOLICY: + return xfrm_notify_policy_flush(c); + case XFRM_MSG_POLEXPIRE: + return xfrm_exp_policy_notify(xp, dir, c); + default: + printk("xfrm_user: Unknown Policy event %d\n", c->event); + } + + return 0; + +} + static struct xfrm_mgr netlink_mgr = { .id = "netlink", .notify = xfrm_send_state_notify, |
