From 76620aafd66f0004829764940c5466144969cffc Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 16 Apr 2009 02:02:07 -0700 Subject: gro: New frags interface to avoid copying shinfo It turns out that copying a 16-byte area at ~800k times a second can be really expensive :) This patch redesigns the frags GRO interface to avoid copying that area twice. The two disciples of the frags interface have been converted. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/core/dev.c | 81 ++++++++++++++++++++++++++-------------------------------- 1 file changed, 36 insertions(+), 45 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 91d792d17e0..619fa141b8f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2519,16 +2519,10 @@ void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb) } EXPORT_SYMBOL(napi_reuse_skb); -struct sk_buff *napi_fraginfo_skb(struct napi_struct *napi, - struct napi_gro_fraginfo *info) +struct sk_buff *napi_get_frags(struct napi_struct *napi) { struct net_device *dev = napi->dev; struct sk_buff *skb = napi->skb; - struct ethhdr *eth; - skb_frag_t *frag; - int i; - - napi->skb = NULL; if (!skb) { skb = netdev_alloc_skb(dev, GRO_MAX_HEAD + NET_IP_ALIGN); @@ -2536,47 +2530,14 @@ struct sk_buff *napi_fraginfo_skb(struct napi_struct *napi, goto out; skb_reserve(skb, NET_IP_ALIGN); - } - - BUG_ON(info->nr_frags > MAX_SKB_FRAGS); - frag = &info->frags[info->nr_frags - 1]; - for (i = skb_shinfo(skb)->nr_frags; i < info->nr_frags; i++) { - skb_fill_page_desc(skb, i, frag->page, frag->page_offset, - frag->size); - frag++; + napi->skb = skb; } - skb_shinfo(skb)->nr_frags = info->nr_frags; - - skb->data_len = info->len; - skb->len += info->len; - skb->truesize += info->len; - - skb_reset_mac_header(skb); - skb_gro_reset_offset(skb); - - eth = skb_gro_header(skb, sizeof(*eth)); - if (!eth) { - napi_reuse_skb(napi, skb); - skb = NULL; - goto out; - } - - skb_gro_pull(skb, sizeof(*eth)); - - /* - * This works because the only protocols we care about don't require - * special handling. We'll fix it up properly at the end. - */ - skb->protocol = eth->h_proto; - - skb->ip_summed = info->ip_summed; - skb->csum = info->csum; out: return skb; } -EXPORT_SYMBOL(napi_fraginfo_skb); +EXPORT_SYMBOL(napi_get_frags); int napi_frags_finish(struct napi_struct *napi, struct sk_buff *skb, int ret) { @@ -2606,9 +2567,39 @@ int napi_frags_finish(struct napi_struct *napi, struct sk_buff *skb, int ret) } EXPORT_SYMBOL(napi_frags_finish); -int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info) +struct sk_buff *napi_frags_skb(struct napi_struct *napi) +{ + struct sk_buff *skb = napi->skb; + struct ethhdr *eth; + + napi->skb = NULL; + + skb_reset_mac_header(skb); + skb_gro_reset_offset(skb); + + eth = skb_gro_header(skb, sizeof(*eth)); + if (!eth) { + napi_reuse_skb(napi, skb); + skb = NULL; + goto out; + } + + skb_gro_pull(skb, sizeof(*eth)); + + /* + * This works because the only protocols we care about don't require + * special handling. We'll fix it up properly at the end. + */ + skb->protocol = eth->h_proto; + +out: + return skb; +} +EXPORT_SYMBOL(napi_frags_skb); + +int napi_gro_frags(struct napi_struct *napi) { - struct sk_buff *skb = napi_fraginfo_skb(napi, info); + struct sk_buff *skb = napi_frags_skb(napi); if (!skb) return NET_RX_DROP; @@ -2712,7 +2703,7 @@ void netif_napi_del(struct napi_struct *napi) struct sk_buff *skb, *next; list_del_init(&napi->dev_list); - kfree_skb(napi->skb); + napi_free_frags(napi); for (skb = napi->gro_list; skb; skb = next) { next = skb->next; -- cgit v1.2.3-70-g09d2 From edbd9e30306067c3a45c035eb95a6f49daaa2337 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 27 Apr 2009 05:44:29 -0700 Subject: gro: Fix handling of headers that extend over the tail The skb_gro_* code fails to handle the case where a header starts in the linear area but ends in the frags area. Since the goal of skb_gro_* is to optimise the case of completely non-linear packets, we can simply bail out if we have anything in the linear area. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- net/core/dev.c | 11 +++-------- 2 files changed, 4 insertions(+), 9 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 31167451d08..c9ef4191607 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1137,7 +1137,7 @@ static inline void skb_gro_reset_offset(struct sk_buff *skb) static inline void *skb_gro_mac_header(struct sk_buff *skb) { - return skb_mac_header(skb) < skb->data ? skb_mac_header(skb) : + return skb_headlen(skb) ? skb_mac_header(skb) : page_address(skb_shinfo(skb)->frags[0].page) + skb_shinfo(skb)->frags[0].page_offset; } diff --git a/net/core/dev.c b/net/core/dev.c index e48c08af76a..6785b067ad5 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2378,18 +2378,13 @@ void *skb_gro_header(struct sk_buff *skb, unsigned int hlen) unsigned int offset = skb_gro_offset(skb); hlen += offset; - if (hlen <= skb_headlen(skb)) - return skb->data + offset; - - if (unlikely(!skb_shinfo(skb)->nr_frags || - skb_shinfo(skb)->frags[0].size <= - hlen - skb_headlen(skb) || + if (unlikely(skb_headlen(skb) || + skb_shinfo(skb)->frags[0].size < hlen || PageHighMem(skb_shinfo(skb)->frags[0].page))) return pskb_may_pull(skb, hlen) ? skb->data + offset : NULL; return page_address(skb_shinfo(skb)->frags[0].page) + - skb_shinfo(skb)->frags[0].page_offset + - offset - skb_headlen(skb); + skb_shinfo(skb)->frags[0].page_offset + offset; } EXPORT_SYMBOL(skb_gro_header); -- cgit v1.2.3-70-g09d2 From 513de11bba246b7a67df4c314d9fc936b6a75d0e Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 3 May 2009 14:43:10 -0700 Subject: net: Avoid modulus in skb_tx_hash() for forwarding case. Based almost entirely upon a patch by Eric Dumazet. The common case is to have num-tx-queues <= num_rx_queues and even if num_tx_queues is larger it will not be significantly larger. Therefore, a subtraction loop is always going to be faster than modulus. Signed-off-by: David S. Miller --- net/core/dev.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 81442957c5c..3c8073fe970 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1735,8 +1735,12 @@ u16 skb_tx_hash(const struct net_device *dev, const struct sk_buff *skb) { u32 hash; - if (skb_rx_queue_recorded(skb)) - return skb_get_rx_queue(skb) % dev->real_num_tx_queues; + if (skb_rx_queue_recorded(skb)) { + hash = skb_get_rx_queue(skb); + while (unlikely (hash >= dev->real_num_tx_queues)) + hash -= dev->real_num_tx_queues; + return hash; + } if (skb->sk && skb->sk->sk_hash) hash = skb->sk->sk_hash; -- cgit v1.2.3-70-g09d2 From f001fde5eadd915f4858d22ed70d7040f48767cf Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 May 2009 02:48:28 +0000 Subject: net: introduce a list of device addresses dev_addr_list (v6) v5 -> v6 (current): -removed so far unused static functions -corrected dev_addr_del_multiple to call del instead of add v4 -> v5: -added device address type (suggested by davem) -removed refcounting (better to have simplier code then safe potentially few bytes) v3 -> v4: -changed kzalloc to kmalloc in __hw_addr_add_ii() -ASSERT_RTNL() avoided in dev_addr_flush() and dev_addr_init() v2 -> v3: -removed unnecessary rcu read locking -moved dev_addr_flush() calling to ensure no null dereference of dev_addr v1 -> v2: -added forgotten ASSERT_RTNL to dev_addr_init and dev_addr_flush -removed unnecessary rcu_read locking in dev_addr_init -use compare_ether_addr_64bits instead of compare_ether_addr -use L1_CACHE_BYTES as size for allocating struct netdev_hw_addr -use call_rcu instead of rcu_synchronize -moved is_etherdev_addr into __KERNEL__ ifdef This patch introduces a new list in struct net_device and brings a set of functions to handle the work with device address list. The list is a replacement for the original dev_addr field and because in some situations there is need to carry several device addresses with the net device. To be backward compatible, dev_addr is made to point to the first member of the list so original drivers sees no difference. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/etherdevice.h | 27 +++++ include/linux/netdevice.h | 37 ++++++- net/core/dev.c | 250 ++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 312 insertions(+), 2 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index a1f17abba7d..3d7a6687d24 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -182,6 +182,33 @@ static inline unsigned compare_ether_addr_64bits(const u8 addr1[6+2], return compare_ether_addr(addr1, addr2); #endif } + +/** + * is_etherdev_addr - Tell if given Ethernet address belongs to the device. + * @dev: Pointer to a device structure + * @addr: Pointer to a six-byte array containing the Ethernet address + * + * Compare passed address with all addresses of the device. Return true if the + * address if one of the device addresses. + * + * Note that this function calls compare_ether_addr_64bits() so take care of + * the right padding. + */ +static inline bool is_etherdev_addr(const struct net_device *dev, + const u8 addr[6 + 2]) +{ + struct netdev_hw_addr *ha; + int res = 1; + + rcu_read_lock(); + for_each_dev_addr(dev, ha) { + res = compare_ether_addr_64bits(addr, ha->addr); + if (!res) + break; + } + rcu_read_unlock(); + return !res; +} #endif /* __KERNEL__ */ /** diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ff42aba403c..02882e2ebd4 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -210,6 +210,16 @@ struct dev_addr_list #define dmi_users da_users #define dmi_gusers da_gusers +struct netdev_hw_addr { + struct list_head list; + unsigned char addr[MAX_ADDR_LEN]; + unsigned char type; +#define NETDEV_HW_ADDR_T_LAN 1 +#define NETDEV_HW_ADDR_T_SAN 2 +#define NETDEV_HW_ADDR_T_SLAVE 3 + struct rcu_head rcu_head; +}; + struct hh_cache { struct hh_cache *hh_next; /* Next entry */ @@ -784,8 +794,11 @@ struct net_device */ unsigned long last_rx; /* Time of last Rx */ /* Interface address info used in eth_type_trans() */ - unsigned char dev_addr[MAX_ADDR_LEN]; /* hw address, (before bcast - because most packets are unicast) */ + unsigned char *dev_addr; /* hw address, (before bcast + because most packets are + unicast) */ + + struct list_head dev_addr_list; /* list of device hw addresses */ unsigned char broadcast[MAX_ADDR_LEN]; /* hw bcast add */ @@ -1791,6 +1804,13 @@ static inline void netif_addr_unlock_bh(struct net_device *dev) spin_unlock_bh(&dev->addr_list_lock); } +/* + * dev_addr_list walker. Should be used only for read access. Call with + * rcu_read_lock held. + */ +#define for_each_dev_addr(dev, ha) \ + list_for_each_entry_rcu(ha, &dev->dev_addr_list, list) + /* These functions live elsewhere (drivers/net/net_init.c, but related) */ extern void ether_setup(struct net_device *dev); @@ -1803,6 +1823,19 @@ extern struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, alloc_netdev_mq(sizeof_priv, name, setup, 1) extern int register_netdev(struct net_device *dev); extern void unregister_netdev(struct net_device *dev); + +/* Functions used for device addresses handling */ +extern int dev_addr_add(struct net_device *dev, unsigned char *addr, + unsigned char addr_type); +extern int dev_addr_del(struct net_device *dev, unsigned char *addr, + unsigned char addr_type); +extern int dev_addr_add_multiple(struct net_device *to_dev, + struct net_device *from_dev, + unsigned char addr_type); +extern int dev_addr_del_multiple(struct net_device *to_dev, + struct net_device *from_dev, + unsigned char addr_type); + /* Functions used for secondary unicast and multicast support */ extern void dev_set_rx_mode(struct net_device *dev); extern void __dev_set_rx_mode(struct net_device *dev); diff --git a/net/core/dev.c b/net/core/dev.c index 3c8073fe970..637ea71b0a0 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3434,6 +3434,252 @@ void dev_set_rx_mode(struct net_device *dev) netif_addr_unlock_bh(dev); } +/* hw addresses list handling functions */ + +static int __hw_addr_add(struct list_head *list, unsigned char *addr, + int addr_len, unsigned char addr_type) +{ + struct netdev_hw_addr *ha; + int alloc_size; + + if (addr_len > MAX_ADDR_LEN) + return -EINVAL; + + alloc_size = sizeof(*ha); + if (alloc_size < L1_CACHE_BYTES) + alloc_size = L1_CACHE_BYTES; + ha = kmalloc(alloc_size, GFP_ATOMIC); + if (!ha) + return -ENOMEM; + memcpy(ha->addr, addr, addr_len); + ha->type = addr_type; + list_add_tail_rcu(&ha->list, list); + return 0; +} + +static void ha_rcu_free(struct rcu_head *head) +{ + struct netdev_hw_addr *ha; + + ha = container_of(head, struct netdev_hw_addr, rcu_head); + kfree(ha); +} + +static int __hw_addr_del_ii(struct list_head *list, unsigned char *addr, + int addr_len, unsigned char addr_type, + int ignore_index) +{ + struct netdev_hw_addr *ha; + int i = 0; + + list_for_each_entry(ha, list, list) { + if (i++ != ignore_index && + !memcmp(ha->addr, addr, addr_len) && + (ha->type == addr_type || !addr_type)) { + list_del_rcu(&ha->list); + call_rcu(&ha->rcu_head, ha_rcu_free); + return 0; + } + } + return -ENOENT; +} + +static int __hw_addr_add_multiple_ii(struct list_head *to_list, + struct list_head *from_list, + int addr_len, unsigned char addr_type, + int ignore_index) +{ + int err; + struct netdev_hw_addr *ha, *ha2; + unsigned char type; + + list_for_each_entry(ha, from_list, list) { + type = addr_type ? addr_type : ha->type; + err = __hw_addr_add(to_list, ha->addr, addr_len, type); + if (err) + goto unroll; + } + return 0; + +unroll: + list_for_each_entry(ha2, from_list, list) { + if (ha2 == ha) + break; + type = addr_type ? addr_type : ha2->type; + __hw_addr_del_ii(to_list, ha2->addr, addr_len, type, + ignore_index); + } + return err; +} + +static void __hw_addr_del_multiple_ii(struct list_head *to_list, + struct list_head *from_list, + int addr_len, unsigned char addr_type, + int ignore_index) +{ + struct netdev_hw_addr *ha; + unsigned char type; + + list_for_each_entry(ha, from_list, list) { + type = addr_type ? addr_type : ha->type; + __hw_addr_del_ii(to_list, ha->addr, addr_len, addr_type, + ignore_index); + } +} + +static void __hw_addr_flush(struct list_head *list) +{ + struct netdev_hw_addr *ha, *tmp; + + list_for_each_entry_safe(ha, tmp, list, list) { + list_del_rcu(&ha->list); + call_rcu(&ha->rcu_head, ha_rcu_free); + } +} + +/* Device addresses handling functions */ + +static void dev_addr_flush(struct net_device *dev) +{ + /* rtnl_mutex must be held here */ + + __hw_addr_flush(&dev->dev_addr_list); + dev->dev_addr = NULL; +} + +static int dev_addr_init(struct net_device *dev) +{ + unsigned char addr[MAX_ADDR_LEN]; + struct netdev_hw_addr *ha; + int err; + + /* rtnl_mutex must be held here */ + + INIT_LIST_HEAD(&dev->dev_addr_list); + memset(addr, 0, sizeof(*addr)); + err = __hw_addr_add(&dev->dev_addr_list, addr, sizeof(*addr), + NETDEV_HW_ADDR_T_LAN); + if (!err) { + /* + * Get the first (previously created) address from the list + * and set dev_addr pointer to this location. + */ + ha = list_first_entry(&dev->dev_addr_list, + struct netdev_hw_addr, list); + dev->dev_addr = ha->addr; + } + return err; +} + +/** + * dev_addr_add - Add a device address + * @dev: device + * @addr: address to add + * @addr_type: address type + * + * Add a device address to the device or increase the reference count if + * it already exists. + * + * The caller must hold the rtnl_mutex. + */ +int dev_addr_add(struct net_device *dev, unsigned char *addr, + unsigned char addr_type) +{ + int err; + + ASSERT_RTNL(); + + err = __hw_addr_add(&dev->dev_addr_list, addr, dev->addr_len, + addr_type); + if (!err) + call_netdevice_notifiers(NETDEV_CHANGEADDR, dev); + return err; +} +EXPORT_SYMBOL(dev_addr_add); + +/** + * dev_addr_del - Release a device address. + * @dev: device + * @addr: address to delete + * @addr_type: address type + * + * Release reference to a device address and remove it from the device + * if the reference count drops to zero. + * + * The caller must hold the rtnl_mutex. + */ +int dev_addr_del(struct net_device *dev, unsigned char *addr, + unsigned char addr_type) +{ + int err; + + ASSERT_RTNL(); + + err = __hw_addr_del_ii(&dev->dev_addr_list, addr, dev->addr_len, + addr_type, 0); + if (!err) + call_netdevice_notifiers(NETDEV_CHANGEADDR, dev); + return err; +} +EXPORT_SYMBOL(dev_addr_del); + +/** + * dev_addr_add_multiple - Add device addresses from another device + * @to_dev: device to which addresses will be added + * @from_dev: device from which addresses will be added + * @addr_type: address type - 0 means type will be used from from_dev + * + * Add device addresses of the one device to another. + ** + * The caller must hold the rtnl_mutex. + */ +int dev_addr_add_multiple(struct net_device *to_dev, + struct net_device *from_dev, + unsigned char addr_type) +{ + int err; + + ASSERT_RTNL(); + + if (from_dev->addr_len != to_dev->addr_len) + return -EINVAL; + err = __hw_addr_add_multiple_ii(&to_dev->dev_addr_list, + &from_dev->dev_addr_list, + to_dev->addr_len, addr_type, 0); + if (!err) + call_netdevice_notifiers(NETDEV_CHANGEADDR, to_dev); + return err; +} +EXPORT_SYMBOL(dev_addr_add_multiple); + +/** + * dev_addr_del_multiple - Delete device addresses by another device + * @to_dev: device where the addresses will be deleted + * @from_dev: device by which addresses the addresses will be deleted + * @addr_type: address type - 0 means type will used from from_dev + * + * Deletes addresses in to device by the list of addresses in from device. + * + * The caller must hold the rtnl_mutex. + */ +int dev_addr_del_multiple(struct net_device *to_dev, + struct net_device *from_dev, + unsigned char addr_type) +{ + ASSERT_RTNL(); + + if (from_dev->addr_len != to_dev->addr_len) + return -EINVAL; + __hw_addr_del_multiple_ii(&to_dev->dev_addr_list, + &from_dev->dev_addr_list, + to_dev->addr_len, addr_type, 0); + call_netdevice_notifiers(NETDEV_CHANGEADDR, to_dev); + return 0; +} +EXPORT_SYMBOL(dev_addr_del_multiple); + +/* unicast and multicast addresses handling functions */ + int __dev_addr_delete(struct dev_addr_list **list, int *count, void *addr, int alen, int glbl) { @@ -4776,6 +5022,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, dev->gso_max_size = GSO_MAX_SIZE; + dev_addr_init(dev); netdev_init_queues(dev); INIT_LIST_HEAD(&dev->napi_list); @@ -4801,6 +5048,9 @@ void free_netdev(struct net_device *dev) kfree(dev->_tx); + /* Flush device addresses */ + dev_addr_flush(dev); + list_for_each_entry_safe(p, n, &dev->napi_list, dev_list) netif_napi_del(p); -- cgit v1.2.3-70-g09d2 From ab9c73ccb52f40576ce017528d542eda3c6ae766 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 8 May 2009 13:30:17 +0000 Subject: net: check retval of dev_addr_init() Add missed checking of dev_addr_init return value in alloc_netdev_mq. Signed-off-by: Jiri Pirko net/core/dev.c | 15 ++++++++++++--- 1 files changed, 12 insertions(+), 3 deletions(-) Signed-off-by: David S. Miller --- net/core/dev.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 637ea71b0a0..14dd725aaab 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5007,13 +5007,16 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, if (!tx) { printk(KERN_ERR "alloc_netdev: Unable to allocate " "tx qdiscs.\n"); - kfree(p); - return NULL; + goto free_p; } dev = (struct net_device *) (((long)p + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST); dev->padded = (char *)dev - (char *)p; + + if (dev_addr_init(dev)) + goto free_tx; + dev_net_set(dev, &init_net); dev->_tx = tx; @@ -5022,13 +5025,19 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, dev->gso_max_size = GSO_MAX_SIZE; - dev_addr_init(dev); netdev_init_queues(dev); INIT_LIST_HEAD(&dev->napi_list); setup(dev); strcpy(dev->name, name); return dev; + +free_tx: + kfree(tx); + +free_p: + kfree(p); + return NULL; } EXPORT_SYMBOL(alloc_netdev_mq); -- cgit v1.2.3-70-g09d2 From 7004bf252c53da18f6b55103e0c92f777f846806 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 18 May 2009 00:34:33 +0000 Subject: net: add tx_packets/tx_bytes/tx_dropped counters in struct netdev_queue offsetof(struct net_device, features)=0x44 offsetof(struct net_device, stats.tx_packets)=0x54 offsetof(struct net_device, stats.tx_bytes)=0x5c offsetof(struct net_device, stats.tx_dropped)=0x6c Network drivers that touch dev->stats.tx_packets/stats.tx_bytes in their tx path can slow down SMP operations, since they dirty a cache line that should stay shared (dev->features is needed in rx and tx paths) We could move away stats field in net_device but it wont help that much. (Two cache lines dirtied in tx path, we can do one only) Better solution is to add tx_packets/tx_bytes/tx_dropped in struct netdev_queue because this structure is already touched in tx path and counters updates will then be free (no increase in size) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 3 +++ net/core/dev.c | 23 ++++++++++++++++++++--- 2 files changed, 23 insertions(+), 3 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index cd547d04a8c..f8574e76b74 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -474,6 +474,9 @@ struct netdev_queue { * please use this field instead of dev->trans_start */ unsigned long trans_start; + unsigned long tx_bytes; + unsigned long tx_packets; + unsigned long tx_dropped; } ____cacheline_aligned_in_smp; diff --git a/net/core/dev.c b/net/core/dev.c index 14dd725aaab..6d3630d1627 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4943,13 +4943,30 @@ void netdev_run_todo(void) * the internal statistics structure is used. */ const struct net_device_stats *dev_get_stats(struct net_device *dev) - { +{ const struct net_device_ops *ops = dev->netdev_ops; if (ops->ndo_get_stats) return ops->ndo_get_stats(dev); - else - return &dev->stats; + else { + unsigned long tx_bytes = 0, tx_packets = 0, tx_dropped = 0; + struct net_device_stats *stats = &dev->stats; + unsigned int i; + struct netdev_queue *txq; + + for (i = 0; i < dev->num_tx_queues; i++) { + txq = netdev_get_tx_queue(dev, i); + tx_bytes += txq->tx_bytes; + tx_packets += txq->tx_packets; + tx_dropped += txq->tx_dropped; + } + if (tx_bytes || tx_packets || tx_dropped) { + stats->tx_bytes = tx_bytes; + stats->tx_packets = tx_packets; + stats->tx_dropped = tx_dropped; + } + return stats; + } } EXPORT_SYMBOL(dev_get_stats); -- cgit v1.2.3-70-g09d2 From 93f154b594fe47e4a7e5358b309add449a046cd3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 18 May 2009 22:19:19 -0700 Subject: net: release dst entry in dev_hard_start_xmit() One point of contention in high network loads is the dst_release() performed when a transmited skb is freed. This is because NIC tx completion calls dev_kree_skb() long after original call to dev_queue_xmit(skb). CPU cache is cold and the atomic op in dst_release() stalls. On SMP, this is quite visible if one CPU is 100% handling softirqs for a network device, since dst_clone() is done by other cpus, involving cache line ping pongs. It seems right place to release dst is in dev_hard_start_xmit(), for most devices but ones that are virtual, and some exceptions. David Miller suggested to define a new device flag, set in alloc_netdev_mq() (so that most devices set it at init time), and carefuly unset in devices which dont want a NULL skb->dst in their ndo_start_xmit(). List of devices that must clear this flag is : - loopback device, because it calls netif_rx() and quoting Patrick : "ip_route_input() doesn't accept loopback addresses, so loopback packets already need to have a dst_entry attached." - appletalk/ipddp.c : needs skb->dst in its xmit function - And all devices that call again dev_queue_xmit() from their xmit function (as some classifiers need skb->dst) : bonding, vlan, macvlan, eql, ifb, hdlc_fr Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/appletalk/ipddp.c | 1 + drivers/net/bonding/bond_main.c | 1 + drivers/net/eql.c | 1 + drivers/net/ifb.c | 1 + drivers/net/loopback.c | 1 + drivers/net/macvlan.c | 1 + drivers/net/wan/hdlc_fr.c | 1 + include/linux/if.h | 3 +++ net/8021q/vlan_dev.c | 1 + net/core/dev.c | 9 +++++++++ 10 files changed, 20 insertions(+) (limited to 'net/core/dev.c') diff --git a/drivers/net/appletalk/ipddp.c b/drivers/net/appletalk/ipddp.c index da64ba88d7f..f939e92fcf8 100644 --- a/drivers/net/appletalk/ipddp.c +++ b/drivers/net/appletalk/ipddp.c @@ -71,6 +71,7 @@ static struct net_device * __init ipddp_init(void) if (!dev) return ERR_PTR(-ENOMEM); + dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; strcpy(dev->name, "ipddp%d"); if (version_printed++ == 0) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 96d7689995c..92a9d69c565 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -5148,6 +5148,7 @@ int bond_create(char *name, struct bond_params *params) goto out_rtnl; } + bond_dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; if (!name) { res = dev_alloc_name(bond_dev, "bond%d"); if (res < 0) diff --git a/drivers/net/eql.c b/drivers/net/eql.c index 5210bb1027c..19b7dd98394 100644 --- a/drivers/net/eql.c +++ b/drivers/net/eql.c @@ -194,6 +194,7 @@ static void __init eql_setup(struct net_device *dev) dev->type = ARPHRD_SLIP; dev->tx_queue_len = 5; /* Hands them off fast */ + dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; } static int eql_open(struct net_device *dev) diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c index 60a26300193..96713ef0629 100644 --- a/drivers/net/ifb.c +++ b/drivers/net/ifb.c @@ -156,6 +156,7 @@ static void ifb_setup(struct net_device *dev) dev->flags |= IFF_NOARP; dev->flags &= ~IFF_MULTICAST; + dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; random_ether_addr(dev->dev_addr); } diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index 6f71157bea8..da472c68748 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -170,6 +170,7 @@ static void loopback_setup(struct net_device *dev) dev->tx_queue_len = 0; dev->type = ARPHRD_LOOPBACK; /* 0x0001*/ dev->flags = IFF_LOOPBACK; + dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; dev->features = NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_TSO | NETIF_F_NO_CSUM diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 329cd50d0e2..d5334b41e4b 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -414,6 +414,7 @@ static void macvlan_setup(struct net_device *dev) { ether_setup(dev); + dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; dev->netdev_ops = &macvlan_netdev_ops; dev->destructor = free_netdev; dev->header_ops = &macvlan_hard_header_ops, diff --git a/drivers/net/wan/hdlc_fr.c b/drivers/net/wan/hdlc_fr.c index 80053010109..bfa0161a02d 100644 --- a/drivers/net/wan/hdlc_fr.c +++ b/drivers/net/wan/hdlc_fr.c @@ -1054,6 +1054,7 @@ static void pvc_setup(struct net_device *dev) dev->flags = IFF_POINTOPOINT; dev->hard_header_len = 10; dev->addr_len = 2; + dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; } static const struct net_device_ops pvc_ops = { diff --git a/include/linux/if.h b/include/linux/if.h index 1108f3e099e..b9a6229f3be 100644 --- a/include/linux/if.h +++ b/include/linux/if.h @@ -67,6 +67,9 @@ #define IFF_ISATAP 0x80 /* ISATAP interface (RFC4214) */ #define IFF_MASTER_ARPMON 0x100 /* bonding master, ARP mon in use */ #define IFF_WAN_HDLC 0x200 /* WAN HDLC device */ +#define IFF_XMIT_DST_RELEASE 0x400 /* dev_hard_start_xmit() is allowed to + * release skb->dst + */ #define IF_GET_IFACE 0x0001 /* for querying only */ #define IF_GET_PROTO 0x0002 diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 8faacee6863..ff7572ac548 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -739,6 +739,7 @@ void vlan_setup(struct net_device *dev) ether_setup(dev); dev->priv_flags |= IFF_802_1Q_VLAN; + dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; dev->tx_queue_len = 0; dev->netdev_ops = &vlan_netdev_ops; diff --git a/net/core/dev.c b/net/core/dev.c index 6d3630d1627..92ebeca2990 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1688,6 +1688,14 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, goto gso; } + /* + * If device doesnt need skb->dst, release it right now while + * its hot in this cpu cache + */ + if ((dev->priv_flags & IFF_XMIT_DST_RELEASE) && skb->dst) { + dst_release(skb->dst); + skb->dst = NULL; + } rc = ops->ndo_start_xmit(skb, dev); /* * TODO: if skb_orphan() was called by @@ -5045,6 +5053,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, netdev_init_queues(dev); INIT_LIST_HEAD(&dev->napi_list); + dev->priv_flags = IFF_XMIT_DST_RELEASE; setup(dev); strcpy(dev->name, name); return dev; -- cgit v1.2.3-70-g09d2 From 4ea7e38696c7e798c47ebbecadfd392f23f814f9 Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Thu, 21 May 2009 07:36:08 +0000 Subject: dropmon: add ability to detect when hardware dropsrxpackets Patch to add the ability to detect drops in hardware interfaces via dropwatch. Adds a tracepoint to net_rx_action to signal everytime a napi instance is polled. The dropmon code then periodically checks to see if the rx_frames counter has changed, and if so, adds a drop notification to the netlink protocol, using the reserved all-0's vector to indicate the drop location was in hardware, rather than somewhere in the code. Signed-off-by: Neil Horman include/linux/net_dropmon.h | 8 ++ include/trace/napi.h | 11 +++ net/core/dev.c | 5 + net/core/drop_monitor.c | 124 ++++++++++++++++++++++++++++++++++++++++++-- net/core/net-traces.c | 4 + net/core/netpoll.c | 2 6 files changed, 149 insertions(+), 5 deletions(-) Signed-off-by: David S. Miller --- include/linux/net_dropmon.h | 8 +++ include/trace/napi.h | 11 ++++ net/core/dev.c | 5 +- net/core/drop_monitor.c | 124 ++++++++++++++++++++++++++++++++++++++++++-- net/core/net-traces.c | 4 ++ net/core/netpoll.c | 2 + 6 files changed, 149 insertions(+), 5 deletions(-) create mode 100644 include/trace/napi.h (limited to 'net/core/dev.c') diff --git a/include/linux/net_dropmon.h b/include/linux/net_dropmon.h index 0217fb81a63..e8a8b5c50ed 100644 --- a/include/linux/net_dropmon.h +++ b/include/linux/net_dropmon.h @@ -2,12 +2,20 @@ #define __NET_DROPMON_H #include +#include struct net_dm_drop_point { __u8 pc[8]; __u32 count; }; +#define is_drop_point_hw(x) do {\ + int ____i, ____j;\ + for (____i = 0; ____i < 8; i ____i++)\ + ____j |= x[____i];\ + ____j;\ +} while (0) + #define NET_DM_CFG_VERSION 0 #define NET_DM_CFG_ALERT_COUNT 1 #define NET_DM_CFG_ALERT_DELAY 2 diff --git a/include/trace/napi.h b/include/trace/napi.h new file mode 100644 index 00000000000..a8989c4547e --- /dev/null +++ b/include/trace/napi.h @@ -0,0 +1,11 @@ +#ifndef _TRACE_NAPI_H_ +#define _TRACE_NAPI_H_ + +#include +#include + +DECLARE_TRACE(napi_poll, + TP_PROTO(struct napi_struct *napi), + TP_ARGS(napi)); + +#endif diff --git a/net/core/dev.c b/net/core/dev.c index 92ebeca2990..3942266d1f6 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -126,6 +126,7 @@ #include #include #include +#include #include "net-sysfs.h" @@ -2771,8 +2772,10 @@ static void net_rx_action(struct softirq_action *h) * accidently calling ->poll() when NAPI is not scheduled. */ work = 0; - if (test_bit(NAPI_STATE_SCHED, &n->state)) + if (test_bit(NAPI_STATE_SCHED, &n->state)) { work = n->poll(n, weight); + trace_napi_poll(n); + } WARN_ON_ONCE(work > weight); diff --git a/net/core/drop_monitor.c b/net/core/drop_monitor.c index 2797b711a97..a6c2ac2828f 100644 --- a/net/core/drop_monitor.c +++ b/net/core/drop_monitor.c @@ -22,8 +22,10 @@ #include #include #include +#include #include +#include #include @@ -38,7 +40,8 @@ static void send_dm_alert(struct work_struct *unused); * and the work handle that will send up * netlink alerts */ -struct sock *dm_sock; +static int trace_state = TRACE_OFF; +static spinlock_t trace_state_lock = SPIN_LOCK_UNLOCKED; struct per_cpu_dm_data { struct work_struct dm_alert_work; @@ -47,6 +50,13 @@ struct per_cpu_dm_data { struct timer_list send_timer; }; +struct dm_hw_stat_delta { + struct net_device *dev; + struct list_head list; + struct rcu_head rcu; + unsigned long last_drop_val; +}; + static struct genl_family net_drop_monitor_family = { .id = GENL_ID_GENERATE, .hdrsize = 0, @@ -59,7 +69,8 @@ static DEFINE_PER_CPU(struct per_cpu_dm_data, dm_cpu_data); static int dm_hit_limit = 64; static int dm_delay = 1; - +static unsigned long dm_hw_check_delta = 2*HZ; +static LIST_HEAD(hw_stats_list); static void reset_per_cpu_data(struct per_cpu_dm_data *data) { @@ -115,7 +126,7 @@ static void sched_send_work(unsigned long unused) schedule_work(&data->dm_alert_work); } -static void trace_kfree_skb_hit(struct sk_buff *skb, void *location) +static void trace_drop_common(struct sk_buff *skb, void *location) { struct net_dm_alert_msg *msg; struct nlmsghdr *nlh; @@ -159,24 +170,80 @@ out: return; } +static void trace_kfree_skb_hit(struct sk_buff *skb, void *location) +{ + trace_drop_common(skb, location); +} + +static void trace_napi_poll_hit(struct napi_struct *napi) +{ + struct dm_hw_stat_delta *new_stat; + + /* + * Ratelimit our check time to dm_hw_check_delta jiffies + */ + if (!time_after(jiffies, napi->dev->last_rx + dm_hw_check_delta)) + return; + + rcu_read_lock(); + list_for_each_entry_rcu(new_stat, &hw_stats_list, list) { + if ((new_stat->dev == napi->dev) && + (napi->dev->stats.rx_dropped != new_stat->last_drop_val)) { + trace_drop_common(NULL, NULL); + new_stat->last_drop_val = napi->dev->stats.rx_dropped; + break; + } + } + rcu_read_unlock(); +} + + +static void free_dm_hw_stat(struct rcu_head *head) +{ + struct dm_hw_stat_delta *n; + n = container_of(head, struct dm_hw_stat_delta, rcu); + kfree(n); +} + static int set_all_monitor_traces(int state) { int rc = 0; + struct dm_hw_stat_delta *new_stat = NULL; + struct dm_hw_stat_delta *temp; + + spin_lock(&trace_state_lock); switch (state) { case TRACE_ON: rc |= register_trace_kfree_skb(trace_kfree_skb_hit); + rc |= register_trace_napi_poll(trace_napi_poll_hit); break; case TRACE_OFF: rc |= unregister_trace_kfree_skb(trace_kfree_skb_hit); + rc |= unregister_trace_napi_poll(trace_napi_poll_hit); tracepoint_synchronize_unregister(); + + /* + * Clean the device list + */ + list_for_each_entry_safe(new_stat, temp, &hw_stats_list, list) { + if (new_stat->dev == NULL) { + list_del_rcu(&new_stat->list); + call_rcu(&new_stat->rcu, free_dm_hw_stat); + } + } break; default: rc = 1; break; } + if (!rc) + trace_state = state; + + spin_unlock(&trace_state_lock); + if (rc) return -EINPROGRESS; return rc; @@ -204,6 +271,44 @@ static int net_dm_cmd_trace(struct sk_buff *skb, return -ENOTSUPP; } +static int dropmon_net_event(struct notifier_block *ev_block, + unsigned long event, void *ptr) +{ + struct net_device *dev = ptr; + struct dm_hw_stat_delta *new_stat = NULL; + struct dm_hw_stat_delta *tmp; + + switch (event) { + case NETDEV_REGISTER: + new_stat = kzalloc(sizeof(struct dm_hw_stat_delta), GFP_KERNEL); + + if (!new_stat) + goto out; + + new_stat->dev = dev; + INIT_RCU_HEAD(&new_stat->rcu); + spin_lock(&trace_state_lock); + list_add_rcu(&new_stat->list, &hw_stats_list); + spin_unlock(&trace_state_lock); + break; + case NETDEV_UNREGISTER: + spin_lock(&trace_state_lock); + list_for_each_entry_safe(new_stat, tmp, &hw_stats_list, list) { + if (new_stat->dev == dev) { + new_stat->dev = NULL; + if (trace_state == TRACE_OFF) { + list_del_rcu(&new_stat->list); + call_rcu(&new_stat->rcu, free_dm_hw_stat); + break; + } + } + } + spin_unlock(&trace_state_lock); + break; + } +out: + return NOTIFY_DONE; +} static struct genl_ops dropmon_ops[] = { { @@ -220,6 +325,10 @@ static struct genl_ops dropmon_ops[] = { }, }; +static struct notifier_block dropmon_net_notifier = { + .notifier_call = dropmon_net_event +}; + static int __init init_net_drop_monitor(void) { int cpu; @@ -243,12 +352,18 @@ static int __init init_net_drop_monitor(void) ret = genl_register_ops(&net_drop_monitor_family, &dropmon_ops[i]); if (ret) { - printk(KERN_CRIT "failed to register operation %d\n", + printk(KERN_CRIT "Failed to register operation %d\n", dropmon_ops[i].cmd); goto out_unreg; } } + rc = register_netdevice_notifier(&dropmon_net_notifier); + if (rc < 0) { + printk(KERN_CRIT "Failed to register netdevice notifier\n"); + goto out_unreg; + } + rc = 0; for_each_present_cpu(cpu) { @@ -259,6 +374,7 @@ static int __init init_net_drop_monitor(void) data->send_timer.data = cpu; data->send_timer.function = sched_send_work; } + goto out; out_unreg: diff --git a/net/core/net-traces.c b/net/core/net-traces.c index c8fb45665e4..b07b25bd2cd 100644 --- a/net/core/net-traces.c +++ b/net/core/net-traces.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -27,3 +28,6 @@ DEFINE_TRACE(kfree_skb); EXPORT_TRACEPOINT_SYMBOL_GPL(kfree_skb); + +DEFINE_TRACE(napi_poll); +EXPORT_TRACEPOINT_SYMBOL_GPL(napi_poll); diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 64f51eec657..00b14e2c50e 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -24,6 +24,7 @@ #include #include #include +#include /* * We maintain a small pool of fully-sized skbs, to make sure the @@ -137,6 +138,7 @@ static int poll_one_napi(struct netpoll_info *npinfo, set_bit(NAPI_STATE_NPSVC, &napi->state); work = napi->poll(napi, budget); + trace_napi_poll(napi->dev); clear_bit(NAPI_STATE_NPSVC, &napi->state); atomic_dec(&trapped); -- cgit v1.2.3-70-g09d2 From e3804cbebb67887879102925961d41b503f7fbe3 Mon Sep 17 00:00:00 2001 From: Alexander Beregalov Date: Mon, 25 May 2009 01:53:53 -0700 Subject: net: remove COMPAT_NET_DEV_OPS All drivers are already converted to new net_device_ops API and nobody uses old API anymore. Signed-off-by: Alexander Beregalov Signed-off-by: David S. Miller --- drivers/net/Kconfig | 9 --------- include/linux/netdevice.h | 38 ----------------------------------- net/802/fddi.c | 4 ---- net/802/hippi.c | 5 ----- net/8021q/vlan_dev.c | 1 - net/appletalk/dev.c | 11 ----------- net/core/dev.c | 50 ----------------------------------------------- net/ethernet/eth.c | 5 ----- 8 files changed, 123 deletions(-) (limited to 'net/core/dev.c') diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 0fb446e047e..efa659f0fb5 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -25,15 +25,6 @@ menuconfig NETDEVICES # that for each of the symbols. if NETDEVICES -config COMPAT_NET_DEV_OPS - default y - bool "Enable older network device API compatibility" - ---help--- - This option enables kernel compatibility with older network devices - that do not use net_device_ops interface. - - If unsure, say Y. - config IFB tristate "Intermediate Functional Block support" depends on NET_CLS_ACT diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index f8574e76b74..ae3c2099a04 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -901,44 +901,6 @@ struct net_device /* max exchange id for FCoE LRO by ddp */ unsigned int fcoe_ddp_xid; #endif - -#ifdef CONFIG_COMPAT_NET_DEV_OPS - struct { - int (*init)(struct net_device *dev); - void (*uninit)(struct net_device *dev); - int (*open)(struct net_device *dev); - int (*stop)(struct net_device *dev); - int (*hard_start_xmit) (struct sk_buff *skb, - struct net_device *dev); - u16 (*select_queue)(struct net_device *dev, - struct sk_buff *skb); - void (*change_rx_flags)(struct net_device *dev, - int flags); - void (*set_rx_mode)(struct net_device *dev); - void (*set_multicast_list)(struct net_device *dev); - int (*set_mac_address)(struct net_device *dev, - void *addr); - int (*validate_addr)(struct net_device *dev); - int (*do_ioctl)(struct net_device *dev, - struct ifreq *ifr, int cmd); - int (*set_config)(struct net_device *dev, - struct ifmap *map); - int (*change_mtu)(struct net_device *dev, int new_mtu); - int (*neigh_setup)(struct net_device *dev, - struct neigh_parms *); - void (*tx_timeout) (struct net_device *dev); - struct net_device_stats* (*get_stats)(struct net_device *dev); - void (*vlan_rx_register)(struct net_device *dev, - struct vlan_group *grp); - void (*vlan_rx_add_vid)(struct net_device *dev, - unsigned short vid); - void (*vlan_rx_kill_vid)(struct net_device *dev, - unsigned short vid); -#ifdef CONFIG_NET_POLL_CONTROLLER - void (*poll_controller)(struct net_device *dev); -#endif - }; -#endif }; #define to_net_dev(d) container_of(d, struct net_device, dev) diff --git a/net/802/fddi.c b/net/802/fddi.c index 539e6064e6d..3ef0ab0a543 100644 --- a/net/802/fddi.c +++ b/net/802/fddi.c @@ -185,10 +185,6 @@ static const struct header_ops fddi_header_ops = { static void fddi_setup(struct net_device *dev) { dev->header_ops = &fddi_header_ops; -#ifdef CONFIG_COMPAT_NET_DEV_OPS - dev->change_mtu = fddi_change_mtu, -#endif - dev->type = ARPHRD_FDDI; dev->hard_header_len = FDDI_K_SNAP_HLEN+3; /* Assume 802.2 SNAP hdr len + 3 pad bytes */ dev->mtu = FDDI_K_SNAP_DLEN; /* Assume max payload of 802.2 SNAP frame */ diff --git a/net/802/hippi.c b/net/802/hippi.c index 313b9ebf92e..cd3e8e92952 100644 --- a/net/802/hippi.c +++ b/net/802/hippi.c @@ -193,11 +193,6 @@ static const struct header_ops hippi_header_ops = { static void hippi_setup(struct net_device *dev) { -#ifdef CONFIG_COMPAT_NET_DEV_OPS - dev->change_mtu = hippi_change_mtu; - dev->set_mac_address = hippi_mac_addr; - dev->neigh_setup = hippi_neigh_setup_dev; -#endif dev->header_ops = &hippi_header_ops; /* diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index ff7572ac548..1e2ad4c7c59 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -644,7 +644,6 @@ static int vlan_dev_init(struct net_device *dev) dev->hard_header_len = real_dev->hard_header_len + VLAN_HLEN; dev->netdev_ops = &vlan_netdev_ops; } - netdev_resync_ops(dev); if (is_vlan_dev(real_dev)) subclass = 1; diff --git a/net/appletalk/dev.c b/net/appletalk/dev.c index 72277d70c98..6c8016f6186 100644 --- a/net/appletalk/dev.c +++ b/net/appletalk/dev.c @@ -9,21 +9,10 @@ #include #include -#ifdef CONFIG_COMPAT_NET_DEV_OPS -static int ltalk_change_mtu(struct net_device *dev, int mtu) -{ - return -EINVAL; -} -#endif - static void ltalk_setup(struct net_device *dev) { /* Fill in the fields of the device structure with localtalk-generic values. */ -#ifdef CONFIG_COMPAT_NET_DEV_OPS - dev->change_mtu = ltalk_change_mtu; -#endif - dev->type = ARPHRD_LOCALTLK; dev->hard_header_len = LTALK_HLEN; dev->mtu = LTALK_MTU; diff --git a/net/core/dev.c b/net/core/dev.c index 3942266d1f6..241613f6dd2 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4580,39 +4580,6 @@ unsigned long netdev_fix_features(unsigned long features, const char *name) } EXPORT_SYMBOL(netdev_fix_features); -/* Some devices need to (re-)set their netdev_ops inside - * ->init() or similar. If that happens, we have to setup - * the compat pointers again. - */ -void netdev_resync_ops(struct net_device *dev) -{ -#ifdef CONFIG_COMPAT_NET_DEV_OPS - const struct net_device_ops *ops = dev->netdev_ops; - - dev->init = ops->ndo_init; - dev->uninit = ops->ndo_uninit; - dev->open = ops->ndo_open; - dev->change_rx_flags = ops->ndo_change_rx_flags; - dev->set_rx_mode = ops->ndo_set_rx_mode; - dev->set_multicast_list = ops->ndo_set_multicast_list; - dev->set_mac_address = ops->ndo_set_mac_address; - dev->validate_addr = ops->ndo_validate_addr; - dev->do_ioctl = ops->ndo_do_ioctl; - dev->set_config = ops->ndo_set_config; - dev->change_mtu = ops->ndo_change_mtu; - dev->neigh_setup = ops->ndo_neigh_setup; - dev->tx_timeout = ops->ndo_tx_timeout; - dev->get_stats = ops->ndo_get_stats; - dev->vlan_rx_register = ops->ndo_vlan_rx_register; - dev->vlan_rx_add_vid = ops->ndo_vlan_rx_add_vid; - dev->vlan_rx_kill_vid = ops->ndo_vlan_rx_kill_vid; -#ifdef CONFIG_NET_POLL_CONTROLLER - dev->poll_controller = ops->ndo_poll_controller; -#endif -#endif -} -EXPORT_SYMBOL(netdev_resync_ops); - /** * register_netdevice - register a network device * @dev: device to register @@ -4652,23 +4619,6 @@ int register_netdevice(struct net_device *dev) dev->iflink = -1; -#ifdef CONFIG_COMPAT_NET_DEV_OPS - /* Netdevice_ops API compatibility support. - * This is temporary until all network devices are converted. - */ - if (dev->netdev_ops) { - netdev_resync_ops(dev); - } else { - char drivername[64]; - pr_info("%s (%s): not using net_device_ops yet\n", - dev->name, netdev_drivername(dev, drivername, 64)); - - /* This works only because net_device_ops and the - compatibility structure are the same. */ - dev->netdev_ops = (void *) &(dev->init); - } -#endif - /* Init, if this function is available */ if (dev->netdev_ops->ndo_init) { ret = dev->netdev_ops->ndo_init(dev); diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c index 280352aba40..5a883affecd 100644 --- a/net/ethernet/eth.c +++ b/net/ethernet/eth.c @@ -337,11 +337,6 @@ const struct header_ops eth_header_ops ____cacheline_aligned = { void ether_setup(struct net_device *dev) { dev->header_ops = ð_header_ops; -#ifdef CONFIG_COMPAT_NET_DEV_OPS - dev->change_mtu = eth_change_mtu; - dev->set_mac_address = eth_mac_addr; - dev->validate_addr = eth_validate_addr; -#endif dev->type = ARPHRD_ETHER; dev->hard_header_len = ETH_HLEN; dev->mtu = ETH_DATA_LEN; -- cgit v1.2.3-70-g09d2 From 08baf561083bc27a953aa087dd8a664bb2b88e8e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 25 May 2009 22:58:01 -0700 Subject: net: txq_trans_update() helper We would like to get rid of netdev->trans_start = jiffies; that about all net drivers have to use in their start_xmit() function, and use txq->trans_start instead. This can be done generically in core network, as suggested by David. Some devices, (particularly loopback) dont need trans_start update, because they dont have transmit watchdog. We could add a new device flag, or rely on fact that txq->tran_start can be updated is txq->xmit_lock_owner is different than -1. Use a helper function to hide our choice. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 6 ++++++ net/core/dev.c | 3 +++ net/core/netpoll.c | 5 ++++- net/core/pktgen.c | 1 + net/sched/sch_teql.c | 1 + 5 files changed, 15 insertions(+), 1 deletion(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ae3c2099a04..586b71f0358 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1674,6 +1674,12 @@ static inline void __netif_tx_unlock_bh(struct netdev_queue *txq) spin_unlock_bh(&txq->_xmit_lock); } +static inline void txq_trans_update(struct netdev_queue *txq) +{ + if (txq->xmit_lock_owner != -1) + txq->trans_start = jiffies; +} + /** * netif_tx_lock - grab network device transmit lock * @dev: network device diff --git a/net/core/dev.c b/net/core/dev.c index 241613f6dd2..5eb3e48ab31 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1698,6 +1698,8 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, skb->dst = NULL; } rc = ops->ndo_start_xmit(skb, dev); + if (rc == 0) + txq_trans_update(txq); /* * TODO: if skb_orphan() was called by * dev->hard_start_xmit() (for example, the unmodified @@ -1727,6 +1729,7 @@ gso: skb->next = nskb; return rc; } + txq_trans_update(txq); if (unlikely(netif_tx_queue_stopped(txq) && skb->next)) return NETDEV_TX_BUSY; } while (skb->next); diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 67b4f3e3d4a..7ab31a7576a 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -302,8 +302,11 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) for (tries = jiffies_to_usecs(1)/USEC_PER_POLL; tries > 0; --tries) { if (__netif_tx_trylock(txq)) { - if (!netif_tx_queue_stopped(txq)) + if (!netif_tx_queue_stopped(txq)) { status = ops->ndo_start_xmit(skb, dev); + if (status == NETDEV_TX_OK) + txq_trans_update(txq); + } __netif_tx_unlock(txq); if (status == NETDEV_TX_OK) diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 0666a827bc6..b8ccd3c88d6 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -3438,6 +3438,7 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev) retry_now: ret = (*xmit)(pkt_dev->skb, odev); if (likely(ret == NETDEV_TX_OK)) { + txq_trans_update(txq); pkt_dev->last_ok = 1; pkt_dev->sofar++; pkt_dev->seq_num++; diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index 428a5ef5b94..a886496bdc3 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -308,6 +308,7 @@ restart: if (!netif_tx_queue_stopped(slave_txq) && !netif_tx_queue_frozen(slave_txq) && slave_ops->ndo_start_xmit(skb, slave) == 0) { + txq_trans_update(slave_txq); __netif_tx_unlock(slave_txq); master->slaves = NEXT_SLAVE(q); netif_wake_queue(dev); -- cgit v1.2.3-70-g09d2 From 78a478d0efd9e86e5345b436e130497b4e5846e8 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 26 May 2009 18:50:21 +0000 Subject: gro: Inline skb_gro_header and cache frag0 virtual address The function skb_gro_header is called four times per packet which quickly adds up at 10Gb/s. This patch inlines it to allow better optimisations. Some architectures perform multiplication for page_address, which is done by each skb_gro_header invocation. This patch caches that value in skb->cb to avoid the unnecessary multiplications. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/netdevice.h | 22 +++++++++++++++------- net/core/dev.c | 27 ++++++++++++--------------- 2 files changed, 27 insertions(+), 22 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 586b71f0358..61890ed0bcf 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1008,6 +1008,9 @@ void netif_napi_add(struct net_device *dev, struct napi_struct *napi, void netif_napi_del(struct napi_struct *napi); struct napi_gro_cb { + /* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */ + void *frag0; + /* This indicates where we are processing relative to skb->data. */ int data_offset; @@ -1107,9 +1110,9 @@ extern int dev_restart(struct net_device *dev); #ifdef CONFIG_NETPOLL_TRAP extern int netpoll_trap(void); #endif -extern void *skb_gro_header(struct sk_buff *skb, unsigned int hlen); extern int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb); +extern void skb_gro_reset_offset(struct sk_buff *skb); static inline unsigned int skb_gro_offset(const struct sk_buff *skb) { @@ -1126,23 +1129,28 @@ static inline void skb_gro_pull(struct sk_buff *skb, unsigned int len) NAPI_GRO_CB(skb)->data_offset += len; } -static inline void skb_gro_reset_offset(struct sk_buff *skb) +static inline void *skb_gro_header(struct sk_buff *skb, unsigned int hlen) { - NAPI_GRO_CB(skb)->data_offset = 0; + unsigned int offset = skb_gro_offset(skb); + + hlen += offset; + if (!NAPI_GRO_CB(skb)->frag0 || + unlikely(skb_shinfo(skb)->frags[0].size + skb_headlen(skb) < hlen)) + return pskb_may_pull(skb, hlen) ? skb->data + offset : NULL; + + return NAPI_GRO_CB(skb)->frag0 + offset; } static inline void *skb_gro_mac_header(struct sk_buff *skb) { return skb_headlen(skb) ? skb_mac_header(skb) : - page_address(skb_shinfo(skb)->frags[0].page) + - skb_shinfo(skb)->frags[0].page_offset; + NAPI_GRO_CB(skb)->frag0; } static inline void *skb_gro_network_header(struct sk_buff *skb) { return skb_headlen(skb) ? skb_network_header(skb) : - page_address(skb_shinfo(skb)->frags[0].page) + - skb_shinfo(skb)->frags[0].page_offset + skb_network_offset(skb); + NAPI_GRO_CB(skb)->frag0 + skb_network_offset(skb); } static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev, diff --git a/net/core/dev.c b/net/core/dev.c index 5eb3e48ab31..bdb1a738193 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2390,21 +2390,6 @@ void napi_gro_flush(struct napi_struct *napi) } EXPORT_SYMBOL(napi_gro_flush); -void *skb_gro_header(struct sk_buff *skb, unsigned int hlen) -{ - unsigned int offset = skb_gro_offset(skb); - - hlen += offset; - if (unlikely(skb_headlen(skb) || - skb_shinfo(skb)->frags[0].size < hlen || - PageHighMem(skb_shinfo(skb)->frags[0].page))) - return pskb_may_pull(skb, hlen) ? skb->data + offset : NULL; - - return page_address(skb_shinfo(skb)->frags[0].page) + - skb_shinfo(skb)->frags[0].page_offset + offset; -} -EXPORT_SYMBOL(skb_gro_header); - int dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { struct sk_buff **pp = NULL; @@ -2520,6 +2505,18 @@ int napi_skb_finish(int ret, struct sk_buff *skb) } EXPORT_SYMBOL(napi_skb_finish); +void skb_gro_reset_offset(struct sk_buff *skb) +{ + NAPI_GRO_CB(skb)->data_offset = 0; + NAPI_GRO_CB(skb)->frag0 = NULL; + + if (!skb_headlen(skb) && !PageHighMem(skb_shinfo(skb)->frags[0].page)) + NAPI_GRO_CB(skb)->frag0 = + page_address(skb_shinfo(skb)->frags[0].page) + + skb_shinfo(skb)->frags[0].page_offset; +} +EXPORT_SYMBOL(skb_gro_reset_offset); + int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { skb_gro_reset_offset(skb); -- cgit v1.2.3-70-g09d2 From 78d3fd0b7de844a6dad56e9620fc9d2271b32ab9 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 26 May 2009 18:50:23 +0000 Subject: gro: Only use skb_gro_header for completely non-linear packets Currently skb_gro_header is used for packets which put the hardware header in skb->data with the rest in frags. Since the drivers that need this optimisation all provide completely non-linear packets, we can gain extra optimisations by only performing the frag0 optimisation for completely non-linear packets. In particular, we can simply test frag0 (instead of skb_headlen) to see whether the optimisation is in force. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/netdevice.h | 11 ++++++----- net/core/dev.c | 3 ++- 2 files changed, 8 insertions(+), 6 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 61890ed0bcf..d1afc3b6485 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1135,22 +1135,23 @@ static inline void *skb_gro_header(struct sk_buff *skb, unsigned int hlen) hlen += offset; if (!NAPI_GRO_CB(skb)->frag0 || - unlikely(skb_shinfo(skb)->frags[0].size + skb_headlen(skb) < hlen)) + unlikely(skb_shinfo(skb)->frags[0].size < hlen)) { + NAPI_GRO_CB(skb)->frag0 = NULL; return pskb_may_pull(skb, hlen) ? skb->data + offset : NULL; + } return NAPI_GRO_CB(skb)->frag0 + offset; } static inline void *skb_gro_mac_header(struct sk_buff *skb) { - return skb_headlen(skb) ? skb_mac_header(skb) : - NAPI_GRO_CB(skb)->frag0; + return NAPI_GRO_CB(skb)->frag0 ?: skb_mac_header(skb); } static inline void *skb_gro_network_header(struct sk_buff *skb) { - return skb_headlen(skb) ? skb_network_header(skb) : - NAPI_GRO_CB(skb)->frag0 + skb_network_offset(skb); + return (NAPI_GRO_CB(skb)->frag0 ?: skb->data) + + skb_network_offset(skb); } static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev, diff --git a/net/core/dev.c b/net/core/dev.c index bdb1a738193..f9d90c56b6f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2510,7 +2510,8 @@ void skb_gro_reset_offset(struct sk_buff *skb) NAPI_GRO_CB(skb)->data_offset = 0; NAPI_GRO_CB(skb)->frag0 = NULL; - if (!skb_headlen(skb) && !PageHighMem(skb_shinfo(skb)->frags[0].page)) + if (skb->mac_header == skb->tail && + !PageHighMem(skb_shinfo(skb)->frags[0].page)) NAPI_GRO_CB(skb)->frag0 = page_address(skb_shinfo(skb)->frags[0].page) + skb_shinfo(skb)->frags[0].page_offset; -- cgit v1.2.3-70-g09d2 From 7489594cb249aeb178287c9a43a9e4f366044259 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 26 May 2009 18:50:27 +0000 Subject: gro: Optimise length comparison in skb_gro_header By caching frag0_len, we can avoid checking both frag0 and the length separately in skb_gro_header. This helps as skb_gro_header is called four times per packet which amounts to a few million times at 10Gb/s. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/netdevice.h | 7 +++++-- net/core/dev.c | 5 ++++- 2 files changed, 9 insertions(+), 3 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d1afc3b6485..2e44a049be0 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1011,6 +1011,9 @@ struct napi_gro_cb { /* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */ void *frag0; + /* Length of frag0. */ + unsigned int frag0_len; + /* This indicates where we are processing relative to skb->data. */ int data_offset; @@ -1134,9 +1137,9 @@ static inline void *skb_gro_header(struct sk_buff *skb, unsigned int hlen) unsigned int offset = skb_gro_offset(skb); hlen += offset; - if (!NAPI_GRO_CB(skb)->frag0 || - unlikely(skb_shinfo(skb)->frags[0].size < hlen)) { + if (NAPI_GRO_CB(skb)->frag0_len < hlen) { NAPI_GRO_CB(skb)->frag0 = NULL; + NAPI_GRO_CB(skb)->frag0_len = 0; return pskb_may_pull(skb, hlen) ? skb->data + offset : NULL; } diff --git a/net/core/dev.c b/net/core/dev.c index f9d90c56b6f..b1722a2d1fb 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2509,12 +2509,15 @@ void skb_gro_reset_offset(struct sk_buff *skb) { NAPI_GRO_CB(skb)->data_offset = 0; NAPI_GRO_CB(skb)->frag0 = NULL; + NAPI_GRO_CB(skb)->frag0_len = 0; if (skb->mac_header == skb->tail && - !PageHighMem(skb_shinfo(skb)->frags[0].page)) + !PageHighMem(skb_shinfo(skb)->frags[0].page)) { NAPI_GRO_CB(skb)->frag0 = page_address(skb_shinfo(skb)->frags[0].page) + skb_shinfo(skb)->frags[0].page_offset; + NAPI_GRO_CB(skb)->frag0_len = skb_shinfo(skb)->frags[0].size; + } } EXPORT_SYMBOL(skb_gro_reset_offset); -- cgit v1.2.3-70-g09d2 From a5b1cf288d4200506ab62fbb86cc81ace948a306 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 26 May 2009 18:50:28 +0000 Subject: gro: Avoid unnecessary comparison after skb_gro_header For the overwhelming majority of cases, skb_gro_header's return value cannot be NULL. Yet we must check it because of its current form. This patch splits it up into multiple functions in order to avoid this. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/netdevice.h | 23 ++++++++++++++--------- net/core/dev.c | 17 ++++++++++++----- net/ipv4/af_inet.c | 13 ++++++++++--- net/ipv4/tcp.c | 22 ++++++++++++++++------ net/ipv6/af_inet6.c | 13 ++++++++++--- 5 files changed, 62 insertions(+), 26 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 2e44a049be0..371ece521e5 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1132,18 +1132,23 @@ static inline void skb_gro_pull(struct sk_buff *skb, unsigned int len) NAPI_GRO_CB(skb)->data_offset += len; } -static inline void *skb_gro_header(struct sk_buff *skb, unsigned int hlen) +static inline void *skb_gro_header_fast(struct sk_buff *skb, + unsigned int offset) { - unsigned int offset = skb_gro_offset(skb); + return NAPI_GRO_CB(skb)->frag0 + offset; +} - hlen += offset; - if (NAPI_GRO_CB(skb)->frag0_len < hlen) { - NAPI_GRO_CB(skb)->frag0 = NULL; - NAPI_GRO_CB(skb)->frag0_len = 0; - return pskb_may_pull(skb, hlen) ? skb->data + offset : NULL; - } +static inline int skb_gro_header_hard(struct sk_buff *skb, unsigned int hlen) +{ + return NAPI_GRO_CB(skb)->frag0_len < hlen; +} - return NAPI_GRO_CB(skb)->frag0 + offset; +static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen, + unsigned int offset) +{ + NAPI_GRO_CB(skb)->frag0 = NULL; + NAPI_GRO_CB(skb)->frag0_len = 0; + return pskb_may_pull(skb, hlen) ? skb->data + offset : NULL; } static inline void *skb_gro_mac_header(struct sk_buff *skb) diff --git a/net/core/dev.c b/net/core/dev.c index b1722a2d1fb..cd29e613bc5 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2590,17 +2590,24 @@ struct sk_buff *napi_frags_skb(struct napi_struct *napi) { struct sk_buff *skb = napi->skb; struct ethhdr *eth; + unsigned int hlen; + unsigned int off; napi->skb = NULL; skb_reset_mac_header(skb); skb_gro_reset_offset(skb); - eth = skb_gro_header(skb, sizeof(*eth)); - if (!eth) { - napi_reuse_skb(napi, skb); - skb = NULL; - goto out; + off = skb_gro_offset(skb); + hlen = off + sizeof(*eth); + eth = skb_gro_header_fast(skb, off); + if (skb_gro_header_hard(skb, hlen)) { + eth = skb_gro_header_slow(skb, hlen, off); + if (unlikely(!eth)) { + napi_reuse_skb(napi, skb); + skb = NULL; + goto out; + } } skb_gro_pull(skb, sizeof(*eth)); diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 170689681aa..644cc553531 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1246,13 +1246,20 @@ static struct sk_buff **inet_gro_receive(struct sk_buff **head, struct sk_buff **pp = NULL; struct sk_buff *p; struct iphdr *iph; + unsigned int hlen; + unsigned int off; int flush = 1; int proto; int id; - iph = skb_gro_header(skb, sizeof(*iph)); - if (unlikely(!iph)) - goto out; + off = skb_gro_offset(skb); + hlen = off + sizeof(*iph); + iph = skb_gro_header_fast(skb, off); + if (skb_gro_header_hard(skb, hlen)) { + iph = skb_gro_header_slow(skb, hlen, off); + if (unlikely(!iph)) + goto out; + } proto = iph->protocol & (MAX_INET_PROTOS - 1); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 68342d43189..c3dcec5efea 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2518,20 +2518,30 @@ struct sk_buff **tcp_gro_receive(struct sk_buff **head, struct sk_buff *skb) unsigned int thlen; unsigned int flags; unsigned int mss = 1; + unsigned int hlen; + unsigned int off; int flush = 1; int i; - th = skb_gro_header(skb, sizeof(*th)); - if (unlikely(!th)) - goto out; + off = skb_gro_offset(skb); + hlen = off + sizeof(*th); + th = skb_gro_header_fast(skb, off); + if (skb_gro_header_hard(skb, hlen)) { + th = skb_gro_header_slow(skb, hlen, off); + if (unlikely(!th)) + goto out; + } thlen = th->doff * 4; if (thlen < sizeof(*th)) goto out; - th = skb_gro_header(skb, thlen); - if (unlikely(!th)) - goto out; + hlen = off + thlen; + if (skb_gro_header_hard(skb, hlen)) { + th = skb_gro_header_slow(skb, hlen, off); + if (unlikely(!th)) + goto out; + } skb_gro_pull(skb, thlen); diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 61f55386a23..b6215be0963 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -817,13 +817,20 @@ static struct sk_buff **ipv6_gro_receive(struct sk_buff **head, struct sk_buff *p; struct ipv6hdr *iph; unsigned int nlen; + unsigned int hlen; + unsigned int off; int flush = 1; int proto; __wsum csum; - iph = skb_gro_header(skb, sizeof(*iph)); - if (unlikely(!iph)) - goto out; + off = skb_gro_offset(skb); + hlen = off + sizeof(*iph); + iph = skb_gro_header_fast(skb, off); + if (skb_gro_header_hard(skb, hlen)) { + iph = skb_gro_header_slow(skb, hlen, off); + if (unlikely(!iph)) + goto out; + } skb_gro_pull(skb, sizeof(*iph)); skb_set_transport_header(skb, skb_gro_offset(skb)); -- cgit v1.2.3-70-g09d2 From cb18978cbf454c236db5e4191a12ef71eef9b3a0 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 26 May 2009 18:50:31 +0000 Subject: gro: Open-code final pskb_may_pull As we know the only packets which need the final pskb_may_pull are completely non-linear, and have all the required bits in frag0, we can perform a straight memcpy instead of going through pskb_may_pull and doing skb_copy_bits. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/core/dev.c | 23 +++++++++++++++++++---- 1 file changed, 19 insertions(+), 4 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index cd29e613bc5..ed4550fd9ec 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2452,10 +2452,25 @@ int dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb) ret = GRO_HELD; pull: - if (unlikely(!pskb_may_pull(skb, skb_gro_offset(skb)))) { - if (napi->gro_list == skb) - napi->gro_list = skb->next; - ret = GRO_DROP; + if (skb_headlen(skb) < skb_gro_offset(skb)) { + int grow = skb_gro_offset(skb) - skb_headlen(skb); + + BUG_ON(skb->end - skb->tail < grow); + + memcpy(skb_tail_pointer(skb), NAPI_GRO_CB(skb)->frag0, grow); + + skb->tail += grow; + skb->data_len -= grow; + + skb_shinfo(skb)->frags[0].page_offset += grow; + skb_shinfo(skb)->frags[0].size -= grow; + + if (unlikely(!skb_shinfo(skb)->frags[0].size)) { + put_page(skb_shinfo(skb)->frags[0].page); + memmove(skb_shinfo(skb)->frags, + skb_shinfo(skb)->frags + 1, + --skb_shinfo(skb)->nr_frags); + } } ok: -- cgit v1.2.3-70-g09d2 From 1ce8e7b57b3a4527ef83da1c5c7bd8a6b9d87b56 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 27 May 2009 04:42:37 +0000 Subject: net: ALIGN/PTR_ALIGN cleanup in alloc_netdev_mq()/netdev_priv() Use ALIGN() and PTR_ALIGN() macros instead of handcoding them. Get rid of NETDEV_ALIGN_CONST ugly define Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 +---- net/core/dev.c | 9 ++++----- net/mac80211/main.c | 8 ++------ 3 files changed, 7 insertions(+), 15 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 371ece521e5..14efce33c00 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -905,7 +905,6 @@ struct net_device #define to_net_dev(d) container_of(d, struct net_device, dev) #define NETDEV_ALIGN 32 -#define NETDEV_ALIGN_CONST (NETDEV_ALIGN - 1) static inline struct netdev_queue *netdev_get_tx_queue(const struct net_device *dev, @@ -976,9 +975,7 @@ static inline bool netdev_uses_trailer_tags(struct net_device *dev) */ static inline void *netdev_priv(const struct net_device *dev) { - return (char *)dev + ((sizeof(struct net_device) - + NETDEV_ALIGN_CONST) - & ~NETDEV_ALIGN_CONST); + return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN); } /* Set the sysfs physical device reference for the network logical device diff --git a/net/core/dev.c b/net/core/dev.c index ed4550fd9ec..32ceee17896 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4988,18 +4988,18 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, struct netdev_queue *tx; struct net_device *dev; size_t alloc_size; - void *p; + struct net_device *p; BUG_ON(strlen(name) >= sizeof(dev->name)); alloc_size = sizeof(struct net_device); if (sizeof_priv) { /* ensure 32-byte alignment of private area */ - alloc_size = (alloc_size + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST; + alloc_size = ALIGN(alloc_size, NETDEV_ALIGN); alloc_size += sizeof_priv; } /* ensure 32-byte alignment of whole construct */ - alloc_size += NETDEV_ALIGN_CONST; + alloc_size += NETDEV_ALIGN - 1; p = kzalloc(alloc_size, GFP_KERNEL); if (!p) { @@ -5014,8 +5014,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, goto free_p; } - dev = (struct net_device *) - (((long)p + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST); + dev = PTR_ALIGN(p, NETDEV_ALIGN); dev->padded = (char *)dev - (char *)p; if (dev_addr_init(dev)) diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 6b7e92eaab4..e37770ced53 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -735,9 +735,7 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, * +-------------------------+ * */ - priv_size = ((sizeof(struct ieee80211_local) + - NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST) + - priv_data_len; + priv_size = ALIGN(sizeof(*local), NETDEV_ALIGN) + priv_data_len; wiphy = wiphy_new(&mac80211_config_ops, priv_size); @@ -754,9 +752,7 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, local->hw.wiphy = wiphy; - local->hw.priv = (char *)local + - ((sizeof(struct ieee80211_local) + - NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST); + local->hw.priv = (char *)local + ALIGN(sizeof(*local), NETDEV_ALIGN); BUG_ON(!ops->tx); BUG_ON(!ops->start); -- cgit v1.2.3-70-g09d2 From ccffad25b5136958d4769ed6de5e87992dd9c65c Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 22 May 2009 23:22:17 +0000 Subject: net: convert unicast addr list This patch converts unicast address list to standard list_head using previously introduced struct netdev_hw_addr. It also relaxes the locking. Original spinlock (still used for multicast addresses) is not needed and is no longer used for a protection of this list. All reading and writing takes place under rtnl (with no changes). I also removed a possibility to specify the length of the address while adding or deleting unicast address. It's always dev->addr_len. The convertion touched especially e1000 and ixgbe codes when the change is not so trivial. Signed-off-by: Jiri Pirko drivers/net/bnx2.c | 13 +-- drivers/net/e1000/e1000_main.c | 24 +++-- drivers/net/ixgbe/ixgbe_common.c | 14 ++-- drivers/net/ixgbe/ixgbe_common.h | 4 +- drivers/net/ixgbe/ixgbe_main.c | 6 +- drivers/net/ixgbe/ixgbe_type.h | 4 +- drivers/net/macvlan.c | 11 +- drivers/net/mv643xx_eth.c | 11 +- drivers/net/niu.c | 7 +- drivers/net/virtio_net.c | 7 +- drivers/s390/net/qeth_l2_main.c | 6 +- drivers/scsi/fcoe/fcoe.c | 16 ++-- include/linux/netdevice.h | 18 ++-- net/8021q/vlan.c | 4 +- net/8021q/vlan_dev.c | 10 +- net/core/dev.c | 195 +++++++++++++++++++++++++++----------- net/dsa/slave.c | 10 +- net/packet/af_packet.c | 4 +- 18 files changed, 227 insertions(+), 137 deletions(-) Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 13 ++- drivers/net/e1000/e1000_main.c | 24 +++-- drivers/net/ixgbe/ixgbe_common.c | 14 +-- drivers/net/ixgbe/ixgbe_common.h | 4 +- drivers/net/ixgbe/ixgbe_main.c | 6 +- drivers/net/ixgbe/ixgbe_type.h | 4 +- drivers/net/macvlan.c | 11 ++- drivers/net/mv643xx_eth.c | 11 ++- drivers/net/niu.c | 7 +- drivers/net/virtio_net.c | 7 +- drivers/s390/net/qeth_l2_main.c | 6 +- drivers/scsi/fcoe/fcoe.c | 16 ++-- include/linux/netdevice.h | 18 ++-- net/8021q/vlan.c | 4 +- net/8021q/vlan_dev.c | 10 +- net/core/dev.c | 195 +++++++++++++++++++++++++++------------ net/dsa/slave.c | 10 +- net/packet/af_packet.c | 4 +- 18 files changed, 227 insertions(+), 137 deletions(-) (limited to 'net/core/dev.c') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 83ee0f53f2d..f53017250e0 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -48,6 +48,7 @@ #include #include #include +#include #include "bnx2.h" #include "bnx2_fw.h" @@ -3310,7 +3311,7 @@ bnx2_set_rx_mode(struct net_device *dev) { struct bnx2 *bp = netdev_priv(dev); u32 rx_mode, sort_mode; - struct dev_addr_list *uc_ptr; + struct netdev_hw_addr *ha; int i; if (!netif_running(dev)) @@ -3369,21 +3370,19 @@ bnx2_set_rx_mode(struct net_device *dev) sort_mode |= BNX2_RPM_SORT_USER0_MC_HSH_EN; } - uc_ptr = NULL; if (dev->uc_count > BNX2_MAX_UNICAST_ADDRESSES) { rx_mode |= BNX2_EMAC_RX_MODE_PROMISCUOUS; sort_mode |= BNX2_RPM_SORT_USER0_PROM_EN | BNX2_RPM_SORT_USER0_PROM_VLAN; } else if (!(dev->flags & IFF_PROMISC)) { - uc_ptr = dev->uc_list; - /* Add all entries into to the match filter list */ - for (i = 0; i < dev->uc_count; i++) { - bnx2_set_mac_addr(bp, uc_ptr->da_addr, + i = 0; + list_for_each_entry(ha, &dev->uc_list, list) { + bnx2_set_mac_addr(bp, ha->addr, i + BNX2_START_UNICAST_ADDRESS_INDEX); sort_mode |= (1 << (i + BNX2_START_UNICAST_ADDRESS_INDEX)); - uc_ptr = uc_ptr->next; + i++; } } diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 79fe1ee3da5..74667e52143 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -2330,7 +2330,8 @@ static void e1000_set_rx_mode(struct net_device *netdev) { struct e1000_adapter *adapter = netdev_priv(netdev); struct e1000_hw *hw = &adapter->hw; - struct dev_addr_list *uc_ptr; + struct netdev_hw_addr *ha; + bool use_uc = false; struct dev_addr_list *mc_ptr; u32 rctl; u32 hash_value; @@ -2369,12 +2370,11 @@ static void e1000_set_rx_mode(struct net_device *netdev) rctl |= E1000_RCTL_VFE; } - uc_ptr = NULL; if (netdev->uc_count > rar_entries - 1) { rctl |= E1000_RCTL_UPE; } else if (!(netdev->flags & IFF_PROMISC)) { rctl &= ~E1000_RCTL_UPE; - uc_ptr = netdev->uc_list; + use_uc = true; } ew32(RCTL, rctl); @@ -2392,13 +2392,20 @@ static void e1000_set_rx_mode(struct net_device *netdev) * if there are not 14 addresses, go ahead and clear the filters * -- with 82571 controllers only 0-13 entries are filled here */ + i = 1; + if (use_uc) + list_for_each_entry(ha, &netdev->uc_list, list) { + if (i == rar_entries) + break; + e1000_rar_set(hw, ha->addr, i++); + } + + WARN_ON(i == rar_entries); + mc_ptr = netdev->mc_list; - for (i = 1; i < rar_entries; i++) { - if (uc_ptr) { - e1000_rar_set(hw, uc_ptr->da_addr, i); - uc_ptr = uc_ptr->next; - } else if (mc_ptr) { + for (; i < rar_entries; i++) { + if (mc_ptr) { e1000_rar_set(hw, mc_ptr->da_addr, i); mc_ptr = mc_ptr->next; } else { @@ -2408,7 +2415,6 @@ static void e1000_set_rx_mode(struct net_device *netdev) E1000_WRITE_FLUSH(); } } - WARN_ON(uc_ptr != NULL); /* load any remaining addresses into the hash table */ diff --git a/drivers/net/ixgbe/ixgbe_common.c b/drivers/net/ixgbe/ixgbe_common.c index 0cc3c47cb45..6f79409270a 100644 --- a/drivers/net/ixgbe/ixgbe_common.c +++ b/drivers/net/ixgbe/ixgbe_common.c @@ -28,6 +28,8 @@ #include #include #include +#include +#include #include "ixgbe.h" #include "ixgbe_common.h" @@ -1356,15 +1358,14 @@ static void ixgbe_add_uc_addr(struct ixgbe_hw *hw, u8 *addr, u32 vmdq) * Drivers using secondary unicast addresses must set user_set_promisc when * manually putting the device into promiscuous mode. **/ -s32 ixgbe_update_uc_addr_list_generic(struct ixgbe_hw *hw, u8 *addr_list, - u32 addr_count, ixgbe_mc_addr_itr next) +s32 ixgbe_update_uc_addr_list_generic(struct ixgbe_hw *hw, + struct list_head *uc_list) { - u8 *addr; u32 i; u32 old_promisc_setting = hw->addr_ctrl.overflow_promisc; u32 uc_addr_in_use; u32 fctrl; - u32 vmdq; + struct netdev_hw_addr *ha; /* * Clear accounting of old secondary address list, @@ -1382,10 +1383,9 @@ s32 ixgbe_update_uc_addr_list_generic(struct ixgbe_hw *hw, u8 *addr_list, } /* Add the new addresses */ - for (i = 0; i < addr_count; i++) { + list_for_each_entry(ha, uc_list, list) { hw_dbg(hw, " Adding the secondary addresses:\n"); - addr = next(hw, &addr_list, &vmdq); - ixgbe_add_uc_addr(hw, addr, vmdq); + ixgbe_add_uc_addr(hw, ha->addr, 0); } if (hw->addr_ctrl.overflow_promisc) { diff --git a/drivers/net/ixgbe/ixgbe_common.h b/drivers/net/ixgbe/ixgbe_common.h index dd260890ad0..b2a4b2c99c4 100644 --- a/drivers/net/ixgbe/ixgbe_common.h +++ b/drivers/net/ixgbe/ixgbe_common.h @@ -59,8 +59,8 @@ s32 ixgbe_init_rx_addrs_generic(struct ixgbe_hw *hw); s32 ixgbe_update_mc_addr_list_generic(struct ixgbe_hw *hw, u8 *mc_addr_list, u32 mc_addr_count, ixgbe_mc_addr_itr func); -s32 ixgbe_update_uc_addr_list_generic(struct ixgbe_hw *hw, u8 *addr_list, - u32 addr_count, ixgbe_mc_addr_itr func); +s32 ixgbe_update_uc_addr_list_generic(struct ixgbe_hw *hw, + struct list_head *uc_list); s32 ixgbe_enable_mc_generic(struct ixgbe_hw *hw); s32 ixgbe_disable_mc_generic(struct ixgbe_hw *hw); s32 ixgbe_enable_rx_dma_generic(struct ixgbe_hw *hw, u32 regval); diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c index 924aa5ed02c..de70a2df9ae 100644 --- a/drivers/net/ixgbe/ixgbe_main.c +++ b/drivers/net/ixgbe/ixgbe_main.c @@ -2181,11 +2181,7 @@ static void ixgbe_set_rx_mode(struct net_device *netdev) IXGBE_WRITE_REG(hw, IXGBE_VLNCTRL, vlnctrl); /* reprogram secondary unicast list */ - addr_count = netdev->uc_count; - if (addr_count) - addr_list = netdev->uc_list->dmi_addr; - hw->mac.ops.update_uc_addr_list(hw, addr_list, addr_count, - ixgbe_addr_list_itr); + hw->mac.ops.update_uc_addr_list(hw, &netdev->uc_list); /* reprogram multicast list */ addr_count = netdev->mc_count; diff --git a/drivers/net/ixgbe/ixgbe_type.h b/drivers/net/ixgbe/ixgbe_type.h index df1f7034c28..a8a8243d8fd 100644 --- a/drivers/net/ixgbe/ixgbe_type.h +++ b/drivers/net/ixgbe/ixgbe_type.h @@ -30,6 +30,7 @@ #include #include +#include /* Vendor ID */ #define IXGBE_INTEL_VENDOR_ID 0x8086 @@ -2223,8 +2224,7 @@ struct ixgbe_mac_operations { s32 (*set_vmdq)(struct ixgbe_hw *, u32, u32); s32 (*clear_vmdq)(struct ixgbe_hw *, u32, u32); s32 (*init_rx_addrs)(struct ixgbe_hw *); - s32 (*update_uc_addr_list)(struct ixgbe_hw *, u8 *, u32, - ixgbe_mc_addr_itr); + s32 (*update_uc_addr_list)(struct ixgbe_hw *, struct list_head *); s32 (*update_mc_addr_list)(struct ixgbe_hw *, u8 *, u32, ixgbe_mc_addr_itr); s32 (*enable_mc)(struct ixgbe_hw *); diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index d5334b41e4b..021d9941c29 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -232,7 +232,7 @@ static int macvlan_open(struct net_device *dev) if (macvlan_addr_busy(vlan->port, dev->dev_addr)) goto out; - err = dev_unicast_add(lowerdev, dev->dev_addr, ETH_ALEN); + err = dev_unicast_add(lowerdev, dev->dev_addr); if (err < 0) goto out; if (dev->flags & IFF_ALLMULTI) { @@ -244,7 +244,7 @@ static int macvlan_open(struct net_device *dev) return 0; del_unicast: - dev_unicast_delete(lowerdev, dev->dev_addr, ETH_ALEN); + dev_unicast_delete(lowerdev, dev->dev_addr); out: return err; } @@ -258,7 +258,7 @@ static int macvlan_stop(struct net_device *dev) if (dev->flags & IFF_ALLMULTI) dev_set_allmulti(lowerdev, -1); - dev_unicast_delete(lowerdev, dev->dev_addr, ETH_ALEN); + dev_unicast_delete(lowerdev, dev->dev_addr); macvlan_hash_del(vlan); return 0; @@ -282,10 +282,11 @@ static int macvlan_set_mac_address(struct net_device *dev, void *p) if (macvlan_addr_busy(vlan->port, addr->sa_data)) return -EBUSY; - if ((err = dev_unicast_add(lowerdev, addr->sa_data, ETH_ALEN))) + err = dev_unicast_add(lowerdev, addr->sa_data); + if (err) return err; - dev_unicast_delete(lowerdev, dev->dev_addr, ETH_ALEN); + dev_unicast_delete(lowerdev, dev->dev_addr); macvlan_hash_change_addr(vlan, addr->sa_data); } diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index 1361ddc8d31..b4e18a58cb1 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -55,6 +55,7 @@ #include #include #include +#include static char mv643xx_eth_driver_name[] = "mv643xx_eth"; static char mv643xx_eth_driver_version[] = "1.4"; @@ -1721,20 +1722,20 @@ static void uc_addr_set(struct mv643xx_eth_private *mp, unsigned char *addr) static u32 uc_addr_filter_mask(struct net_device *dev) { - struct dev_addr_list *uc_ptr; + struct netdev_hw_addr *ha; u32 nibbles; if (dev->flags & IFF_PROMISC) return 0; nibbles = 1 << (dev->dev_addr[5] & 0x0f); - for (uc_ptr = dev->uc_list; uc_ptr != NULL; uc_ptr = uc_ptr->next) { - if (memcmp(dev->dev_addr, uc_ptr->da_addr, 5)) + list_for_each_entry(ha, &dev->uc_list, list) { + if (memcmp(dev->dev_addr, ha->addr, 5)) return 0; - if ((dev->dev_addr[5] ^ uc_ptr->da_addr[5]) & 0xf0) + if ((dev->dev_addr[5] ^ ha->addr[5]) & 0xf0) return 0; - nibbles |= 1 << (uc_ptr->da_addr[5] & 0x0f); + nibbles |= 1 << (ha->addr[5] & 0x0f); } return nibbles; diff --git a/drivers/net/niu.c b/drivers/net/niu.c index edac3a0b02d..fa61a12c5e1 100644 --- a/drivers/net/niu.c +++ b/drivers/net/niu.c @@ -22,6 +22,7 @@ #include #include #include +#include #include @@ -6362,6 +6363,7 @@ static void niu_set_rx_mode(struct net_device *dev) struct niu *np = netdev_priv(dev); int i, alt_cnt, err; struct dev_addr_list *addr; + struct netdev_hw_addr *ha; unsigned long flags; u16 hash[16] = { 0, }; @@ -6383,9 +6385,8 @@ static void niu_set_rx_mode(struct net_device *dev) if (alt_cnt) { int index = 0; - for (addr = dev->uc_list; addr; addr = addr->next) { - err = niu_set_alt_mac(np, index, - addr->da_addr); + list_for_each_entry(ha, &dev->uc_list, list) { + err = niu_set_alt_mac(np, index, ha->addr); if (err) printk(KERN_WARNING PFX "%s: Error %d " "adding alt mac %d\n", diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 6cc5bcd34fb..0c9ca67f66e 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -680,6 +680,7 @@ static void virtnet_set_rx_mode(struct net_device *dev) u8 promisc, allmulti; struct virtio_net_ctrl_mac *mac_data; struct dev_addr_list *addr; + struct netdev_hw_addr *ha; void *buf; int i; @@ -718,9 +719,9 @@ static void virtnet_set_rx_mode(struct net_device *dev) /* Store the unicast list and count in the front of the buffer */ mac_data->entries = dev->uc_count; - addr = dev->uc_list; - for (i = 0; i < dev->uc_count; i++, addr = addr->next) - memcpy(&mac_data->macs[i][0], addr->da_addr, ETH_ALEN); + i = 0; + list_for_each_entry(ha, &dev->uc_list, list) + memcpy(&mac_data->macs[i++][0], ha->addr, ETH_ALEN); sg_set_buf(&sg[0], mac_data, sizeof(mac_data->entries) + (dev->uc_count * ETH_ALEN)); diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index 9ca6bab7c9b..ecd3d06c0d5 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -19,6 +19,7 @@ #include #include #include +#include #include "qeth_core.h" @@ -640,6 +641,7 @@ static void qeth_l2_set_multicast_list(struct net_device *dev) { struct qeth_card *card = dev->ml_priv; struct dev_addr_list *dm; + struct netdev_hw_addr *ha; if (card->info.type == QETH_CARD_TYPE_OSN) return ; @@ -653,8 +655,8 @@ static void qeth_l2_set_multicast_list(struct net_device *dev) for (dm = dev->mc_list; dm; dm = dm->next) qeth_l2_add_mc(card, dm->da_addr, 0); - for (dm = dev->uc_list; dm; dm = dm->next) - qeth_l2_add_mc(card, dm->da_addr, 1); + list_for_each_entry(ha, &dev->uc_list, list) + qeth_l2_add_mc(card, ha->addr, 1); spin_unlock_bh(&card->mclock); if (!qeth_adp_supported(card, IPA_SETADP_SET_PROMISC_MODE)) diff --git a/drivers/scsi/fcoe/fcoe.c b/drivers/scsi/fcoe/fcoe.c index ce33f107b0a..f791348871f 100644 --- a/drivers/scsi/fcoe/fcoe.c +++ b/drivers/scsi/fcoe/fcoe.c @@ -182,8 +182,8 @@ static void fcoe_update_src_mac(struct fcoe_ctlr *fip, u8 *old, u8 *new) fc = fcoe_from_ctlr(fip); rtnl_lock(); if (!is_zero_ether_addr(old)) - dev_unicast_delete(fc->real_dev, old, ETH_ALEN); - dev_unicast_add(fc->real_dev, new, ETH_ALEN); + dev_unicast_delete(fc->real_dev, old); + dev_unicast_add(fc->real_dev, new); rtnl_unlock(); } @@ -233,13 +233,11 @@ void fcoe_netdev_cleanup(struct fcoe_softc *fc) /* Delete secondary MAC addresses */ rtnl_lock(); memcpy(flogi_maddr, (u8[6]) FC_FCOE_FLOGI_MAC, ETH_ALEN); - dev_unicast_delete(fc->real_dev, flogi_maddr, ETH_ALEN); + dev_unicast_delete(fc->real_dev, flogi_maddr); if (!is_zero_ether_addr(fc->ctlr.data_src_addr)) - dev_unicast_delete(fc->real_dev, - fc->ctlr.data_src_addr, ETH_ALEN); + dev_unicast_delete(fc->real_dev, fc->ctlr.data_src_addr); if (fc->ctlr.spma) - dev_unicast_delete(fc->real_dev, - fc->ctlr.ctl_src_addr, ETH_ALEN); + dev_unicast_delete(fc->real_dev, fc->ctlr.ctl_src_addr); dev_mc_delete(fc->real_dev, FIP_ALL_ENODE_MACS, ETH_ALEN, 0); rtnl_unlock(); } @@ -347,9 +345,9 @@ static int fcoe_netdev_config(struct fc_lport *lp, struct net_device *netdev) */ rtnl_lock(); memcpy(flogi_maddr, (u8[6]) FC_FCOE_FLOGI_MAC, ETH_ALEN); - dev_unicast_add(fc->real_dev, flogi_maddr, ETH_ALEN); + dev_unicast_add(fc->real_dev, flogi_maddr); if (fc->ctlr.spma) - dev_unicast_add(fc->real_dev, fc->ctlr.ctl_src_addr, ETH_ALEN); + dev_unicast_add(fc->real_dev, fc->ctlr.ctl_src_addr); dev_mc_add(fc->real_dev, FIP_ALL_ENODE_MACS, ETH_ALEN, 0); rtnl_unlock(); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 1eaf5ae14fe..bbfabf3012b 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -215,9 +215,12 @@ struct netdev_hw_addr { struct list_head list; unsigned char addr[MAX_ADDR_LEN]; unsigned char type; -#define NETDEV_HW_ADDR_T_LAN 1 -#define NETDEV_HW_ADDR_T_SAN 2 -#define NETDEV_HW_ADDR_T_SLAVE 3 +#define NETDEV_HW_ADDR_T_LAN 1 +#define NETDEV_HW_ADDR_T_SAN 2 +#define NETDEV_HW_ADDR_T_SLAVE 3 +#define NETDEV_HW_ADDR_T_UNICAST 4 + int refcount; + bool synced; struct rcu_head rcu_head; }; @@ -773,10 +776,11 @@ struct net_device unsigned char addr_len; /* hardware address length */ unsigned short dev_id; /* for shared network cards */ - spinlock_t addr_list_lock; - struct dev_addr_list *uc_list; /* Secondary unicast mac addresses */ + struct list_head uc_list; /* Secondary unicast mac + addresses */ int uc_count; /* Number of installed ucasts */ int uc_promisc; + spinlock_t addr_list_lock; struct dev_addr_list *mc_list; /* Multicast mac addresses */ int mc_count; /* Number of installed mcasts */ unsigned int promiscuity; @@ -1836,8 +1840,8 @@ extern int dev_addr_del_multiple(struct net_device *to_dev, /* Functions used for secondary unicast and multicast support */ extern void dev_set_rx_mode(struct net_device *dev); extern void __dev_set_rx_mode(struct net_device *dev); -extern int dev_unicast_delete(struct net_device *dev, void *addr, int alen); -extern int dev_unicast_add(struct net_device *dev, void *addr, int alen); +extern int dev_unicast_delete(struct net_device *dev, void *addr); +extern int dev_unicast_add(struct net_device *dev, void *addr); extern int dev_unicast_sync(struct net_device *to, struct net_device *from); extern void dev_unicast_unsync(struct net_device *to, struct net_device *from); extern int dev_mc_delete(struct net_device *dev, void *addr, int alen, int all); diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index d1e10546eb8..714e1c3536b 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -378,13 +378,13 @@ static void vlan_sync_address(struct net_device *dev, * the new address */ if (compare_ether_addr(vlandev->dev_addr, vlan->real_dev_addr) && !compare_ether_addr(vlandev->dev_addr, dev->dev_addr)) - dev_unicast_delete(dev, vlandev->dev_addr, ETH_ALEN); + dev_unicast_delete(dev, vlandev->dev_addr); /* vlan address was equal to the old address and is different from * the new address */ if (!compare_ether_addr(vlandev->dev_addr, vlan->real_dev_addr) && compare_ether_addr(vlandev->dev_addr, dev->dev_addr)) - dev_unicast_add(dev, vlandev->dev_addr, ETH_ALEN); + dev_unicast_add(dev, vlandev->dev_addr); memcpy(vlan->real_dev_addr, dev->dev_addr, ETH_ALEN); } diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 1e2ad4c7c59..96bad8f233e 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -441,7 +441,7 @@ static int vlan_dev_open(struct net_device *dev) return -ENETDOWN; if (compare_ether_addr(dev->dev_addr, real_dev->dev_addr)) { - err = dev_unicast_add(real_dev, dev->dev_addr, ETH_ALEN); + err = dev_unicast_add(real_dev, dev->dev_addr); if (err < 0) goto out; } @@ -470,7 +470,7 @@ clear_allmulti: dev_set_allmulti(real_dev, -1); del_unicast: if (compare_ether_addr(dev->dev_addr, real_dev->dev_addr)) - dev_unicast_delete(real_dev, dev->dev_addr, ETH_ALEN); + dev_unicast_delete(real_dev, dev->dev_addr); out: netif_carrier_off(dev); return err; @@ -492,7 +492,7 @@ static int vlan_dev_stop(struct net_device *dev) dev_set_promiscuity(real_dev, -1); if (compare_ether_addr(dev->dev_addr, real_dev->dev_addr)) - dev_unicast_delete(real_dev, dev->dev_addr, dev->addr_len); + dev_unicast_delete(real_dev, dev->dev_addr); netif_carrier_off(dev); return 0; @@ -511,13 +511,13 @@ static int vlan_dev_set_mac_address(struct net_device *dev, void *p) goto out; if (compare_ether_addr(addr->sa_data, real_dev->dev_addr)) { - err = dev_unicast_add(real_dev, addr->sa_data, ETH_ALEN); + err = dev_unicast_add(real_dev, addr->sa_data); if (err < 0) return err; } if (compare_ether_addr(dev->dev_addr, real_dev->dev_addr)) - dev_unicast_delete(real_dev, dev->dev_addr, ETH_ALEN); + dev_unicast_delete(real_dev, dev->dev_addr); out: memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN); diff --git a/net/core/dev.c b/net/core/dev.c index 32ceee17896..e2fcc5f1017 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3473,8 +3473,9 @@ void dev_set_rx_mode(struct net_device *dev) /* hw addresses list handling functions */ -static int __hw_addr_add(struct list_head *list, unsigned char *addr, - int addr_len, unsigned char addr_type) +static int __hw_addr_add(struct list_head *list, int *delta, + unsigned char *addr, int addr_len, + unsigned char addr_type) { struct netdev_hw_addr *ha; int alloc_size; @@ -3482,6 +3483,15 @@ static int __hw_addr_add(struct list_head *list, unsigned char *addr, if (addr_len > MAX_ADDR_LEN) return -EINVAL; + list_for_each_entry(ha, list, list) { + if (!memcmp(ha->addr, addr, addr_len) && + ha->type == addr_type) { + ha->refcount++; + return 0; + } + } + + alloc_size = sizeof(*ha); if (alloc_size < L1_CACHE_BYTES) alloc_size = L1_CACHE_BYTES; @@ -3490,7 +3500,11 @@ static int __hw_addr_add(struct list_head *list, unsigned char *addr, return -ENOMEM; memcpy(ha->addr, addr, addr_len); ha->type = addr_type; + ha->refcount = 1; + ha->synced = false; list_add_tail_rcu(&ha->list, list); + if (delta) + (*delta)++; return 0; } @@ -3502,29 +3516,30 @@ static void ha_rcu_free(struct rcu_head *head) kfree(ha); } -static int __hw_addr_del_ii(struct list_head *list, unsigned char *addr, - int addr_len, unsigned char addr_type, - int ignore_index) +static int __hw_addr_del(struct list_head *list, int *delta, + unsigned char *addr, int addr_len, + unsigned char addr_type) { struct netdev_hw_addr *ha; - int i = 0; list_for_each_entry(ha, list, list) { - if (i++ != ignore_index && - !memcmp(ha->addr, addr, addr_len) && + if (!memcmp(ha->addr, addr, addr_len) && (ha->type == addr_type || !addr_type)) { + if (--ha->refcount) + return 0; list_del_rcu(&ha->list); call_rcu(&ha->rcu_head, ha_rcu_free); + if (delta) + (*delta)--; return 0; } } return -ENOENT; } -static int __hw_addr_add_multiple_ii(struct list_head *to_list, - struct list_head *from_list, - int addr_len, unsigned char addr_type, - int ignore_index) +static int __hw_addr_add_multiple(struct list_head *to_list, int *to_delta, + struct list_head *from_list, int addr_len, + unsigned char addr_type) { int err; struct netdev_hw_addr *ha, *ha2; @@ -3532,7 +3547,8 @@ static int __hw_addr_add_multiple_ii(struct list_head *to_list, list_for_each_entry(ha, from_list, list) { type = addr_type ? addr_type : ha->type; - err = __hw_addr_add(to_list, ha->addr, addr_len, type); + err = __hw_addr_add(to_list, to_delta, ha->addr, + addr_len, type); if (err) goto unroll; } @@ -3543,27 +3559,69 @@ unroll: if (ha2 == ha) break; type = addr_type ? addr_type : ha2->type; - __hw_addr_del_ii(to_list, ha2->addr, addr_len, type, - ignore_index); + __hw_addr_del(to_list, to_delta, ha2->addr, + addr_len, type); } return err; } -static void __hw_addr_del_multiple_ii(struct list_head *to_list, - struct list_head *from_list, - int addr_len, unsigned char addr_type, - int ignore_index) +static void __hw_addr_del_multiple(struct list_head *to_list, int *to_delta, + struct list_head *from_list, int addr_len, + unsigned char addr_type) { struct netdev_hw_addr *ha; unsigned char type; list_for_each_entry(ha, from_list, list) { type = addr_type ? addr_type : ha->type; - __hw_addr_del_ii(to_list, ha->addr, addr_len, addr_type, - ignore_index); + __hw_addr_del(to_list, to_delta, ha->addr, + addr_len, addr_type); + } +} + +static int __hw_addr_sync(struct list_head *to_list, int *to_delta, + struct list_head *from_list, int *from_delta, + int addr_len) +{ + int err = 0; + struct netdev_hw_addr *ha, *tmp; + + list_for_each_entry_safe(ha, tmp, from_list, list) { + if (!ha->synced) { + err = __hw_addr_add(to_list, to_delta, ha->addr, + addr_len, ha->type); + if (err) + break; + ha->synced = true; + ha->refcount++; + } else if (ha->refcount == 1) { + __hw_addr_del(to_list, to_delta, ha->addr, + addr_len, ha->type); + __hw_addr_del(from_list, from_delta, ha->addr, + addr_len, ha->type); + } } + return err; } +static void __hw_addr_unsync(struct list_head *to_list, int *to_delta, + struct list_head *from_list, int *from_delta, + int addr_len) +{ + struct netdev_hw_addr *ha, *tmp; + + list_for_each_entry_safe(ha, tmp, from_list, list) { + if (ha->synced) { + __hw_addr_del(to_list, to_delta, ha->addr, + addr_len, ha->type); + ha->synced = false; + __hw_addr_del(from_list, from_delta, ha->addr, + addr_len, ha->type); + } + } +} + + static void __hw_addr_flush(struct list_head *list) { struct netdev_hw_addr *ha, *tmp; @@ -3594,7 +3652,7 @@ static int dev_addr_init(struct net_device *dev) INIT_LIST_HEAD(&dev->dev_addr_list); memset(addr, 0, sizeof(*addr)); - err = __hw_addr_add(&dev->dev_addr_list, addr, sizeof(*addr), + err = __hw_addr_add(&dev->dev_addr_list, NULL, addr, sizeof(*addr), NETDEV_HW_ADDR_T_LAN); if (!err) { /* @@ -3626,7 +3684,7 @@ int dev_addr_add(struct net_device *dev, unsigned char *addr, ASSERT_RTNL(); - err = __hw_addr_add(&dev->dev_addr_list, addr, dev->addr_len, + err = __hw_addr_add(&dev->dev_addr_list, NULL, addr, dev->addr_len, addr_type); if (!err) call_netdevice_notifiers(NETDEV_CHANGEADDR, dev); @@ -3649,11 +3707,20 @@ int dev_addr_del(struct net_device *dev, unsigned char *addr, unsigned char addr_type) { int err; + struct netdev_hw_addr *ha; ASSERT_RTNL(); - err = __hw_addr_del_ii(&dev->dev_addr_list, addr, dev->addr_len, - addr_type, 0); + /* + * We can not remove the first address from the list because + * dev->dev_addr points to that. + */ + ha = list_first_entry(&dev->dev_addr_list, struct netdev_hw_addr, list); + if (ha->addr == dev->dev_addr && ha->refcount == 1) + return -ENOENT; + + err = __hw_addr_del(&dev->dev_addr_list, NULL, addr, dev->addr_len, + addr_type); if (!err) call_netdevice_notifiers(NETDEV_CHANGEADDR, dev); return err; @@ -3680,9 +3747,9 @@ int dev_addr_add_multiple(struct net_device *to_dev, if (from_dev->addr_len != to_dev->addr_len) return -EINVAL; - err = __hw_addr_add_multiple_ii(&to_dev->dev_addr_list, - &from_dev->dev_addr_list, - to_dev->addr_len, addr_type, 0); + err = __hw_addr_add_multiple(&to_dev->dev_addr_list, NULL, + &from_dev->dev_addr_list, + to_dev->addr_len, addr_type); if (!err) call_netdevice_notifiers(NETDEV_CHANGEADDR, to_dev); return err; @@ -3707,9 +3774,9 @@ int dev_addr_del_multiple(struct net_device *to_dev, if (from_dev->addr_len != to_dev->addr_len) return -EINVAL; - __hw_addr_del_multiple_ii(&to_dev->dev_addr_list, - &from_dev->dev_addr_list, - to_dev->addr_len, addr_type, 0); + __hw_addr_del_multiple(&to_dev->dev_addr_list, NULL, + &from_dev->dev_addr_list, + to_dev->addr_len, addr_type); call_netdevice_notifiers(NETDEV_CHANGEADDR, to_dev); return 0; } @@ -3779,24 +3846,22 @@ int __dev_addr_add(struct dev_addr_list **list, int *count, * dev_unicast_delete - Release secondary unicast address. * @dev: device * @addr: address to delete - * @alen: length of @addr * * Release reference to a secondary unicast address and remove it * from the device if the reference count drops to zero. * * The caller must hold the rtnl_mutex. */ -int dev_unicast_delete(struct net_device *dev, void *addr, int alen) +int dev_unicast_delete(struct net_device *dev, void *addr) { int err; ASSERT_RTNL(); - netif_addr_lock_bh(dev); - err = __dev_addr_delete(&dev->uc_list, &dev->uc_count, addr, alen, 0); + err = __hw_addr_del(&dev->uc_list, &dev->uc_count, addr, + dev->addr_len, NETDEV_HW_ADDR_T_UNICAST); if (!err) __dev_set_rx_mode(dev); - netif_addr_unlock_bh(dev); return err; } EXPORT_SYMBOL(dev_unicast_delete); @@ -3805,24 +3870,22 @@ EXPORT_SYMBOL(dev_unicast_delete); * dev_unicast_add - add a secondary unicast address * @dev: device * @addr: address to add - * @alen: length of @addr * * Add a secondary unicast address to the device or increase * the reference count if it already exists. * * The caller must hold the rtnl_mutex. */ -int dev_unicast_add(struct net_device *dev, void *addr, int alen) +int dev_unicast_add(struct net_device *dev, void *addr) { int err; ASSERT_RTNL(); - netif_addr_lock_bh(dev); - err = __dev_addr_add(&dev->uc_list, &dev->uc_count, addr, alen, 0); + err = __hw_addr_add(&dev->uc_list, &dev->uc_count, addr, + dev->addr_len, NETDEV_HW_ADDR_T_UNICAST); if (!err) __dev_set_rx_mode(dev); - netif_addr_unlock_bh(dev); return err; } EXPORT_SYMBOL(dev_unicast_add); @@ -3879,8 +3942,7 @@ void __dev_addr_unsync(struct dev_addr_list **to, int *to_count, * @from: source device * * Add newly added addresses to the destination device and release - * addresses that have no users left. The source device must be - * locked by netif_tx_lock_bh. + * addresses that have no users left. * * This function is intended to be called from the dev->set_rx_mode * function of layered software devices. @@ -3889,12 +3951,15 @@ int dev_unicast_sync(struct net_device *to, struct net_device *from) { int err = 0; - netif_addr_lock_bh(to); - err = __dev_addr_sync(&to->uc_list, &to->uc_count, - &from->uc_list, &from->uc_count); + ASSERT_RTNL(); + + if (to->addr_len != from->addr_len) + return -EINVAL; + + err = __hw_addr_sync(&to->uc_list, &to->uc_count, + &from->uc_list, &from->uc_count, to->addr_len); if (!err) __dev_set_rx_mode(to); - netif_addr_unlock_bh(to); return err; } EXPORT_SYMBOL(dev_unicast_sync); @@ -3910,18 +3975,33 @@ EXPORT_SYMBOL(dev_unicast_sync); */ void dev_unicast_unsync(struct net_device *to, struct net_device *from) { - netif_addr_lock_bh(from); - netif_addr_lock(to); + ASSERT_RTNL(); - __dev_addr_unsync(&to->uc_list, &to->uc_count, - &from->uc_list, &from->uc_count); - __dev_set_rx_mode(to); + if (to->addr_len != from->addr_len) + return; - netif_addr_unlock(to); - netif_addr_unlock_bh(from); + __hw_addr_unsync(&to->uc_list, &to->uc_count, + &from->uc_list, &from->uc_count, to->addr_len); + __dev_set_rx_mode(to); } EXPORT_SYMBOL(dev_unicast_unsync); +static void dev_unicast_flush(struct net_device *dev) +{ + /* rtnl_mutex must be held here */ + + __hw_addr_flush(&dev->uc_list); + dev->uc_count = 0; +} + +static void dev_unicast_init(struct net_device *dev) +{ + /* rtnl_mutex must be held here */ + + INIT_LIST_HEAD(&dev->uc_list); +} + + static void __dev_addr_discard(struct dev_addr_list **list) { struct dev_addr_list *tmp; @@ -3940,9 +4020,6 @@ static void dev_addr_discard(struct net_device *dev) { netif_addr_lock_bh(dev); - __dev_addr_discard(&dev->uc_list); - dev->uc_count = 0; - __dev_addr_discard(&dev->mc_list); dev->mc_count = 0; @@ -4535,6 +4612,7 @@ static void rollback_registered(struct net_device *dev) /* * Flush the unicast and multicast chains */ + dev_unicast_flush(dev); dev_addr_discard(dev); if (dev->netdev_ops->ndo_uninit) @@ -5020,6 +5098,8 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, if (dev_addr_init(dev)) goto free_tx; + dev_unicast_init(dev); + dev_net_set(dev, &init_net); dev->_tx = tx; @@ -5223,6 +5303,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char /* * Flush the unicast and multicast chains */ + dev_unicast_flush(dev); dev_addr_discard(dev); netdev_unregister_kobject(dev); diff --git a/net/dsa/slave.c b/net/dsa/slave.c index ed131181215..2175e6d5cc8 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -67,7 +67,7 @@ static int dsa_slave_open(struct net_device *dev) return -ENETDOWN; if (compare_ether_addr(dev->dev_addr, master->dev_addr)) { - err = dev_unicast_add(master, dev->dev_addr, ETH_ALEN); + err = dev_unicast_add(master, dev->dev_addr); if (err < 0) goto out; } @@ -90,7 +90,7 @@ clear_allmulti: dev_set_allmulti(master, -1); del_unicast: if (compare_ether_addr(dev->dev_addr, master->dev_addr)) - dev_unicast_delete(master, dev->dev_addr, ETH_ALEN); + dev_unicast_delete(master, dev->dev_addr); out: return err; } @@ -108,7 +108,7 @@ static int dsa_slave_close(struct net_device *dev) dev_set_promiscuity(master, -1); if (compare_ether_addr(dev->dev_addr, master->dev_addr)) - dev_unicast_delete(master, dev->dev_addr, ETH_ALEN); + dev_unicast_delete(master, dev->dev_addr); return 0; } @@ -147,13 +147,13 @@ static int dsa_slave_set_mac_address(struct net_device *dev, void *a) goto out; if (compare_ether_addr(addr->sa_data, master->dev_addr)) { - err = dev_unicast_add(master, addr->sa_data, ETH_ALEN); + err = dev_unicast_add(master, addr->sa_data); if (err < 0) return err; } if (compare_ether_addr(dev->dev_addr, master->dev_addr)) - dev_unicast_delete(master, dev->dev_addr, ETH_ALEN); + dev_unicast_delete(master, dev->dev_addr); out: memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN); diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index c7c5d524967..6da9f38ef5c 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1582,9 +1582,9 @@ static int packet_dev_mc(struct net_device *dev, struct packet_mclist *i, break; case PACKET_MR_UNICAST: if (what > 0) - return dev_unicast_add(dev, i->addr, i->alen); + return dev_unicast_add(dev, i->addr); else - return dev_unicast_delete(dev, i->addr, i->alen); + return dev_unicast_delete(dev, i->addr); break; default:; } -- cgit v1.2.3-70-g09d2 From adf30907d63893e4208dfe3f5c88ae12bc2f25d5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 2 Jun 2009 05:19:30 +0000 Subject: net: skb->dst accessors Define three accessors to get/set dst attached to a skb struct dst_entry *skb_dst(const struct sk_buff *skb) void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst) void skb_dst_drop(struct sk_buff *skb) This one should replace occurrences of : dst_release(skb->dst) skb->dst = NULL; Delete skb->dst field Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/infiniband/ulp/ipoib/ipoib_cm.c | 4 +- drivers/infiniband/ulp/ipoib/ipoib_main.c | 30 ++++++------- drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 10 ++--- drivers/net/pppol2tp.c | 11 +++-- drivers/s390/net/qeth_core_main.c | 4 +- drivers/s390/net/qeth_l3_main.c | 8 ++-- include/linux/skbuff.h | 13 +++++- include/net/dst.h | 12 ++++-- include/net/inet6_hashtables.h | 2 +- include/net/inet_hashtables.h | 2 +- include/net/ip6_route.h | 2 +- include/net/xfrm.h | 4 +- net/atm/br2684.c | 2 +- net/atm/clip.c | 14 +++--- net/bridge/br_netfilter.c | 18 ++++---- net/core/dev.c | 7 ++- net/core/neighbour.c | 11 +++-- net/core/skbuff.c | 4 +- net/dccp/ipv4.c | 4 +- net/dccp/ipv6.c | 8 ++-- net/dccp/output.c | 2 +- net/decnet/af_decnet.c | 6 ++- net/decnet/dn_neigh.c | 8 ++-- net/decnet/dn_nsp_out.c | 6 +-- net/decnet/dn_route.c | 25 +++++------ net/ipv4/arp.c | 2 +- net/ipv4/icmp.c | 10 ++--- net/ipv4/igmp.c | 4 +- net/ipv4/ip_forward.c | 4 +- net/ipv4/ip_fragment.c | 2 +- net/ipv4/ip_gre.c | 23 +++++----- net/ipv4/ip_input.c | 6 +-- net/ipv4/ip_options.c | 6 +-- net/ipv4/ip_output.c | 20 ++++----- net/ipv4/ipip.c | 13 +++--- net/ipv4/ipmr.c | 13 +++--- net/ipv4/netfilter.c | 28 ++++++------ net/ipv4/netfilter/ipt_REJECT.c | 7 ++- net/ipv4/netfilter/nf_nat_standalone.c | 7 ++- net/ipv4/raw.c | 2 +- net/ipv4/route.c | 14 +++--- net/ipv4/tcp_ipv4.c | 4 +- net/ipv4/tcp_output.c | 2 +- net/ipv4/udp.c | 4 +- net/ipv4/xfrm4_input.c | 2 +- net/ipv4/xfrm4_mode_tunnel.c | 4 +- net/ipv4/xfrm4_output.c | 6 +-- net/ipv6/exthdrs.c | 40 ++++++++--------- net/ipv6/inet6_connection_sock.c | 2 +- net/ipv6/ip6_input.c | 12 +++--- net/ipv6/ip6_output.c | 60 +++++++++++++------------- net/ipv6/ip6_tunnel.c | 26 ++++++----- net/ipv6/ip6mr.c | 13 +++--- net/ipv6/mcast.c | 17 +++++--- net/ipv6/ndisc.c | 4 +- net/ipv6/netfilter.c | 16 ++++--- net/ipv6/netfilter/ip6t_REJECT.c | 2 +- net/ipv6/raw.c | 2 +- net/ipv6/reassembly.c | 22 +++++----- net/ipv6/route.c | 12 +++--- net/ipv6/sit.c | 21 +++++---- net/ipv6/tcp_ipv6.c | 8 ++-- net/ipv6/udp.c | 7 ++- net/ipv6/xfrm6_mode_tunnel.c | 4 +- net/ipv6/xfrm6_output.c | 4 +- net/netfilter/ipvs/ip_vs_xmit.c | 48 ++++++++++----------- net/netfilter/nf_conntrack_proto_gre.c | 2 +- net/netfilter/xt_TCPMSS.c | 6 +-- net/netfilter/xt_policy.c | 2 +- net/netfilter/xt_realm.c | 2 +- net/packet/af_packet.c | 6 +-- net/sched/cls_flow.c | 8 ++-- net/sched/cls_route.c | 2 +- net/sched/em_meta.c | 4 +- net/sched/sch_sfq.c | 2 +- net/sched/sch_teql.c | 6 +-- net/sctp/output.c | 6 +-- net/sunrpc/xprtsock.c | 2 +- net/xfrm/xfrm_input.c | 3 +- net/xfrm/xfrm_output.c | 21 ++++----- net/xfrm/xfrm_policy.c | 8 +++- security/selinux/hooks.c | 2 +- security/selinux/xfrm.c | 2 +- 83 files changed, 414 insertions(+), 390 deletions(-) (limited to 'net/core/dev.c') diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index 4248c313936..181b1f32325 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -1394,8 +1394,8 @@ void ipoib_cm_skb_too_long(struct net_device *dev, struct sk_buff *skb, struct ipoib_dev_priv *priv = netdev_priv(dev); int e = skb_queue_empty(&priv->cm.skb_queue); - if (skb->dst) - skb->dst->ops->update_pmtu(skb->dst, mtu); + if (skb_dst(skb)) + skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); skb_queue_tail(&priv->cm.skb_queue, skb); if (e) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 69c6304cc94..e319d91f60a 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -561,7 +561,7 @@ static void neigh_add_path(struct sk_buff *skb, struct net_device *dev) struct ipoib_neigh *neigh; unsigned long flags; - neigh = ipoib_neigh_alloc(skb->dst->neighbour, skb->dev); + neigh = ipoib_neigh_alloc(skb_dst(skb)->neighbour, skb->dev); if (!neigh) { ++dev->stats.tx_dropped; dev_kfree_skb_any(skb); @@ -570,9 +570,9 @@ static void neigh_add_path(struct sk_buff *skb, struct net_device *dev) spin_lock_irqsave(&priv->lock, flags); - path = __path_find(dev, skb->dst->neighbour->ha + 4); + path = __path_find(dev, skb_dst(skb)->neighbour->ha + 4); if (!path) { - path = path_rec_create(dev, skb->dst->neighbour->ha + 4); + path = path_rec_create(dev, skb_dst(skb)->neighbour->ha + 4); if (!path) goto err_path; @@ -605,7 +605,7 @@ static void neigh_add_path(struct sk_buff *skb, struct net_device *dev) goto err_drop; } } else - ipoib_send(dev, skb, path->ah, IPOIB_QPN(skb->dst->neighbour->ha)); + ipoib_send(dev, skb, path->ah, IPOIB_QPN(skb_dst(skb)->neighbour->ha)); } else { neigh->ah = NULL; @@ -635,15 +635,15 @@ static void ipoib_path_lookup(struct sk_buff *skb, struct net_device *dev) struct ipoib_dev_priv *priv = netdev_priv(skb->dev); /* Look up path record for unicasts */ - if (skb->dst->neighbour->ha[4] != 0xff) { + if (skb_dst(skb)->neighbour->ha[4] != 0xff) { neigh_add_path(skb, dev); return; } /* Add in the P_Key for multicasts */ - skb->dst->neighbour->ha[8] = (priv->pkey >> 8) & 0xff; - skb->dst->neighbour->ha[9] = priv->pkey & 0xff; - ipoib_mcast_send(dev, skb->dst->neighbour->ha + 4, skb); + skb_dst(skb)->neighbour->ha[8] = (priv->pkey >> 8) & 0xff; + skb_dst(skb)->neighbour->ha[9] = priv->pkey & 0xff; + ipoib_mcast_send(dev, skb_dst(skb)->neighbour->ha + 4, skb); } static void unicast_arp_send(struct sk_buff *skb, struct net_device *dev, @@ -708,16 +708,16 @@ static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev) struct ipoib_neigh *neigh; unsigned long flags; - if (likely(skb->dst && skb->dst->neighbour)) { - if (unlikely(!*to_ipoib_neigh(skb->dst->neighbour))) { + if (likely(skb_dst(skb) && skb_dst(skb)->neighbour)) { + if (unlikely(!*to_ipoib_neigh(skb_dst(skb)->neighbour))) { ipoib_path_lookup(skb, dev); return NETDEV_TX_OK; } - neigh = *to_ipoib_neigh(skb->dst->neighbour); + neigh = *to_ipoib_neigh(skb_dst(skb)->neighbour); if (unlikely((memcmp(&neigh->dgid.raw, - skb->dst->neighbour->ha + 4, + skb_dst(skb)->neighbour->ha + 4, sizeof(union ib_gid))) || (neigh->dev != dev))) { spin_lock_irqsave(&priv->lock, flags); @@ -743,7 +743,7 @@ static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev) return NETDEV_TX_OK; } } else if (neigh->ah) { - ipoib_send(dev, skb, neigh->ah, IPOIB_QPN(skb->dst->neighbour->ha)); + ipoib_send(dev, skb, neigh->ah, IPOIB_QPN(skb_dst(skb)->neighbour->ha)); return NETDEV_TX_OK; } @@ -772,7 +772,7 @@ static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev) if ((be16_to_cpup((__be16 *) skb->data) != ETH_P_ARP) && (be16_to_cpup((__be16 *) skb->data) != ETH_P_RARP)) { ipoib_warn(priv, "Unicast, no %s: type %04x, QPN %06x %pI6\n", - skb->dst ? "neigh" : "dst", + skb_dst(skb) ? "neigh" : "dst", be16_to_cpup((__be16 *) skb->data), IPOIB_QPN(phdr->hwaddr), phdr->hwaddr + 4); @@ -817,7 +817,7 @@ static int ipoib_hard_header(struct sk_buff *skb, * destination address onto the front of the skb so we can * figure out where to send the packet later. */ - if ((!skb->dst || !skb->dst->neighbour) && daddr) { + if ((!skb_dst(skb) || !skb_dst(skb)->neighbour) && daddr) { struct ipoib_pseudoheader *phdr = (struct ipoib_pseudoheader *) skb_push(skb, sizeof *phdr); memcpy(phdr->hwaddr, daddr, INFINIBAND_ALEN); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 425e31112ed..a0e97532e71 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -261,7 +261,7 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, skb->dev = dev; - if (!skb->dst || !skb->dst->neighbour) { + if (!skb_dst(skb) || !skb_dst(skb)->neighbour) { /* put pseudoheader back on for next time */ skb_push(skb, sizeof (struct ipoib_pseudoheader)); } @@ -707,10 +707,10 @@ void ipoib_mcast_send(struct net_device *dev, void *mgid, struct sk_buff *skb) out: if (mcast && mcast->ah) { - if (skb->dst && - skb->dst->neighbour && - !*to_ipoib_neigh(skb->dst->neighbour)) { - struct ipoib_neigh *neigh = ipoib_neigh_alloc(skb->dst->neighbour, + if (skb_dst(skb) && + skb_dst(skb)->neighbour && + !*to_ipoib_neigh(skb_dst(skb)->neighbour)) { + struct ipoib_neigh *neigh = ipoib_neigh_alloc(skb_dst(skb)->neighbour, skb->dev); if (neigh) { diff --git a/drivers/net/pppol2tp.c b/drivers/net/pppol2tp.c index 5981debcde5..e7935d09c89 100644 --- a/drivers/net/pppol2tp.c +++ b/drivers/net/pppol2tp.c @@ -433,8 +433,7 @@ static void pppol2tp_recv_dequeue_skb(struct pppol2tp_session *session, struct s * to the inner packet either */ secpath_reset(skb); - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); nf_reset(skb); po = pppox_sk(session_sock); @@ -976,7 +975,7 @@ static int pppol2tp_sendmsg(struct kiocb *iocb, struct socket *sock, struct msgh /* Calculate UDP checksum if configured to do so */ if (sk_tun->sk_no_check == UDP_CSUM_NOXMIT) skb->ip_summed = CHECKSUM_NONE; - else if (!(skb->dst->dev->features & NETIF_F_V4_CSUM)) { + else if (!(skb_dst(skb)->dev->features & NETIF_F_V4_CSUM)) { skb->ip_summed = CHECKSUM_COMPLETE; csum = skb_checksum(skb, 0, udp_len, 0); uh->check = csum_tcpudp_magic(inet->saddr, inet->daddr, @@ -1172,14 +1171,14 @@ static int pppol2tp_xmit(struct ppp_channel *chan, struct sk_buff *skb) nf_reset(skb); /* Get routing info from the tunnel socket */ - dst_release(skb->dst); - skb->dst = dst_clone(__sk_dst_get(sk_tun)); + skb_dst_drop(skb); + skb_dst_set(skb, dst_clone(__sk_dst_get(sk_tun))); pppol2tp_skb_set_owner_w(skb, sk_tun); /* Calculate UDP checksum if configured to do so */ if (sk_tun->sk_no_check == UDP_CSUM_NOXMIT) skb->ip_summed = CHECKSUM_NONE; - else if (!(skb->dst->dev->features & NETIF_F_V4_CSUM)) { + else if (!(skb_dst(skb)->dev->features & NETIF_F_V4_CSUM)) { skb->ip_summed = CHECKSUM_COMPLETE; csum = skb_checksum(skb, 0, udp_len, 0); uh->check = csum_tcpudp_magic(inet->saddr, inet->daddr, diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 2994aa1ed46..74c49d9a8db 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -2937,8 +2937,8 @@ int qeth_get_cast_type(struct qeth_card *card, struct sk_buff *skb) if (card->info.type == QETH_CARD_TYPE_OSN) return cast_type; - if (skb->dst && skb->dst->neighbour) { - cast_type = skb->dst->neighbour->type; + if (skb_dst(skb) && skb_dst(skb)->neighbour) { + cast_type = skb_dst(skb)->neighbour->type; if ((cast_type == RTN_BROADCAST) || (cast_type == RTN_MULTICAST) || (cast_type == RTN_ANYCAST)) diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index cb64b0b534a..6f2386e9d6e 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -2549,9 +2549,9 @@ static void qeth_l3_fill_header(struct qeth_card *card, struct qeth_hdr *hdr, /* IPv4 */ hdr->hdr.l3.flags = qeth_l3_get_qeth_hdr_flags4(cast_type); memset(hdr->hdr.l3.dest_addr, 0, 12); - if ((skb->dst) && (skb->dst->neighbour)) { + if ((skb_dst(skb)) && (skb_dst(skb)->neighbour)) { *((u32 *) (&hdr->hdr.l3.dest_addr[12])) = - *((u32 *) skb->dst->neighbour->primary_key); + *((u32 *) skb_dst(skb)->neighbour->primary_key); } else { /* fill in destination address used in ip header */ *((u32 *) (&hdr->hdr.l3.dest_addr[12])) = @@ -2562,9 +2562,9 @@ static void qeth_l3_fill_header(struct qeth_card *card, struct qeth_hdr *hdr, hdr->hdr.l3.flags = qeth_l3_get_qeth_hdr_flags6(cast_type); if (card->info.type == QETH_CARD_TYPE_IQD) hdr->hdr.l3.flags &= ~QETH_HDR_PASSTHRU; - if ((skb->dst) && (skb->dst->neighbour)) { + if ((skb_dst(skb)) && (skb_dst(skb)->neighbour)) { memcpy(hdr->hdr.l3.dest_addr, - skb->dst->neighbour->primary_key, 16); + skb_dst(skb)->neighbour->primary_key, 16); } else { /* fill in destination address used in ip header */ memcpy(hdr->hdr.l3.dest_addr, diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index a3ae3c52583..9ef6eb20247 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -323,7 +323,6 @@ struct sk_buff { struct net_device *dev; union { - struct dst_entry *dst; unsigned long _skb_dst; }; #ifdef CONFIG_XFRM @@ -426,9 +425,19 @@ extern void skb_dma_unmap(struct device *dev, struct sk_buff *skb, enum dma_data_direction dir); #endif +static inline struct dst_entry *skb_dst(const struct sk_buff *skb) +{ + return (struct dst_entry *)skb->_skb_dst; +} + +static inline void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst) +{ + skb->_skb_dst = (unsigned long)dst; +} + static inline struct rtable *skb_rtable(const struct sk_buff *skb) { - return (struct rtable *)skb->_skb_dst; + return (struct rtable *)skb_dst(skb); } extern void kfree_skb(struct sk_buff *skb); diff --git a/include/net/dst.h b/include/net/dst.h index 6be3b082a07..7fc409c19b3 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -195,6 +195,12 @@ struct dst_entry * dst_clone(struct dst_entry * dst) } extern void dst_release(struct dst_entry *dst); +static inline void skb_dst_drop(struct sk_buff *skb) +{ + if (skb->_skb_dst) + dst_release(skb_dst(skb)); + skb->_skb_dst = 0UL; +} /* Children define the path of the packet through the * Linux networking. Thus, destinations are stackable. @@ -246,7 +252,7 @@ static inline void dst_negative_advice(struct dst_entry **dst_p) static inline void dst_link_failure(struct sk_buff *skb) { - struct dst_entry * dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); if (dst && dst->ops && dst->ops->link_failure) dst->ops->link_failure(skb); } @@ -265,13 +271,13 @@ static inline void dst_set_expires(struct dst_entry *dst, int timeout) /* Output packet to network from transport. */ static inline int dst_output(struct sk_buff *skb) { - return skb->dst->output(skb); + return skb_dst(skb)->output(skb); } /* Input packet from network to transport. */ static inline int dst_input(struct sk_buff *skb) { - return skb->dst->input(skb); + return skb_dst(skb)->input(skb); } static inline struct dst_entry *dst_check(struct dst_entry *dst, u32 cookie) diff --git a/include/net/inet6_hashtables.h b/include/net/inet6_hashtables.h index f74665d7bea..22c73a77cd9 100644 --- a/include/net/inet6_hashtables.h +++ b/include/net/inet6_hashtables.h @@ -100,7 +100,7 @@ static inline struct sock *__inet6_lookup_skb(struct inet_hashinfo *hashinfo, if (unlikely(sk = skb_steal_sock(skb))) return sk; - else return __inet6_lookup(dev_net(skb->dst->dev), hashinfo, + else return __inet6_lookup(dev_net(skb_dst(skb)->dev), hashinfo, &ipv6_hdr(skb)->saddr, sport, &ipv6_hdr(skb)->daddr, ntohs(dport), inet6_iif(skb)); diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index a44e2248b2e..d522dcf3031 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -385,7 +385,7 @@ static inline struct sock *__inet_lookup_skb(struct inet_hashinfo *hashinfo, if (unlikely(sk = skb_steal_sock(skb))) return sk; else - return __inet_lookup(dev_net(skb->dst->dev), hashinfo, + return __inet_lookup(dev_net(skb_dst(skb)->dev), hashinfo, iph->saddr, sport, iph->daddr, dport, inet_iif(skb)); } diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 5f53db7e4e5..0e1b8aebaff 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -142,7 +142,7 @@ static inline void ip6_dst_store(struct sock *sk, struct dst_entry *dst, static inline int ipv6_unicast_destination(struct sk_buff *skb) { - struct rt6_info *rt = (struct rt6_info *) skb->dst; + struct rt6_info *rt = (struct rt6_info *) skb_dst(skb); return rt->rt6i_flags & RTF_LOCAL; } diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 2e9f5c0018a..736bca45088 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -994,7 +994,7 @@ static inline int __xfrm_policy_check2(struct sock *sk, int dir, return __xfrm_policy_check(sk, ndir, skb, family); return (!net->xfrm.policy_count[dir] && !skb->sp) || - (skb->dst->flags & DST_NOPOLICY) || + (skb_dst(skb)->flags & DST_NOPOLICY) || __xfrm_policy_check(sk, ndir, skb, family); } @@ -1048,7 +1048,7 @@ static inline int xfrm_route_forward(struct sk_buff *skb, unsigned short family) struct net *net = dev_net(skb->dev); return !net->xfrm.policy_count[XFRM_POLICY_OUT] || - (skb->dst->flags & DST_NOXFRM) || + (skb_dst(skb)->flags & DST_NOXFRM) || __xfrm_route_forward(skb, family); } diff --git a/net/atm/br2684.c b/net/atm/br2684.c index bfa8fa9894f..2912665fc58 100644 --- a/net/atm/br2684.c +++ b/net/atm/br2684.c @@ -228,7 +228,7 @@ static int br2684_start_xmit(struct sk_buff *skb, struct net_device *dev) struct br2684_dev *brdev = BRPRIV(dev); struct br2684_vcc *brvcc; - pr_debug("br2684_start_xmit, skb->dst=%p\n", skb->dst); + pr_debug("br2684_start_xmit, skb_dst(skb)=%p\n", skb_dst(skb)); read_lock(&devs_lock); brvcc = pick_outgoing_vcc(skb, brdev); if (brvcc == NULL) { diff --git a/net/atm/clip.c b/net/atm/clip.c index fb7623c080f..e65a3b1477f 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -369,16 +369,16 @@ static int clip_start_xmit(struct sk_buff *skb, struct net_device *dev) unsigned long flags; pr_debug("clip_start_xmit (skb %p)\n", skb); - if (!skb->dst) { - printk(KERN_ERR "clip_start_xmit: skb->dst == NULL\n"); + if (!skb_dst(skb)) { + printk(KERN_ERR "clip_start_xmit: skb_dst(skb) == NULL\n"); dev_kfree_skb(skb); dev->stats.tx_dropped++; return 0; } - if (!skb->dst->neighbour) { + if (!skb_dst(skb)->neighbour) { #if 0 - skb->dst->neighbour = clip_find_neighbour(skb->dst, 1); - if (!skb->dst->neighbour) { + skb_dst(skb)->neighbour = clip_find_neighbour(skb_dst(skb), 1); + if (!skb_dst(skb)->neighbour) { dev_kfree_skb(skb); /* lost that one */ dev->stats.tx_dropped++; return 0; @@ -389,7 +389,7 @@ static int clip_start_xmit(struct sk_buff *skb, struct net_device *dev) dev->stats.tx_dropped++; return 0; } - entry = NEIGH2ENTRY(skb->dst->neighbour); + entry = NEIGH2ENTRY(skb_dst(skb)->neighbour); if (!entry->vccs) { if (time_after(jiffies, entry->expires)) { /* should be resolved */ @@ -406,7 +406,7 @@ static int clip_start_xmit(struct sk_buff *skb, struct net_device *dev) } pr_debug("neigh %p, vccs %p\n", entry, entry->vccs); ATM_SKB(skb)->vcc = vcc = entry->vccs->vcc; - pr_debug("using neighbour %p, vcc %p\n", skb->dst->neighbour, vcc); + pr_debug("using neighbour %p, vcc %p\n", skb_dst(skb)->neighbour, vcc); if (entry->vccs->encap) { void *here; diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index e0ceb66a9ec..d22f611e400 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -242,7 +242,7 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb) return 0; } dst_hold(&rt->u.dst); - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); skb->dev = nf_bridge->physindev; nf_bridge_push_encap_header(skb); @@ -322,7 +322,7 @@ static int br_nf_pre_routing_finish_bridge(struct sk_buff *skb) skb->dev = bridge_parent(skb->dev); if (skb->dev) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); nf_bridge_pull_encap_header(skb); @@ -375,7 +375,7 @@ static int br_nf_pre_routing_finish(struct sk_buff *skb) /* - Bridged-and-DNAT'ed traffic doesn't * require ip_forwarding. */ if (((struct dst_entry *)rt)->dev == dev) { - skb->dst = (struct dst_entry *)rt; + skb_dst_set(skb, (struct dst_entry *)rt); goto bridged_dnat; } /* we are sure that forwarding is disabled, so printing @@ -389,7 +389,7 @@ free_skb: kfree_skb(skb); return 0; } else { - if (skb->dst->dev == dev) { + if (skb_dst(skb)->dev == dev) { bridged_dnat: /* Tell br_nf_local_out this is a * bridged frame */ @@ -412,7 +412,7 @@ bridged_dnat: return 0; } dst_hold(&rt->u.dst); - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); } skb->dev = nf_bridge->physindev; @@ -633,10 +633,8 @@ static unsigned int br_nf_local_in(unsigned int hook, struct sk_buff *skb, { struct rtable *rt = skb_rtable(skb); - if (rt && rt == bridge_parent_rtable(in)) { - dst_release(&rt->u.dst); - skb->dst = NULL; - } + if (rt && rt == bridge_parent_rtable(in)) + skb_dst_drop(skb); return NF_ACCEPT; } @@ -851,7 +849,7 @@ static unsigned int br_nf_post_routing(unsigned int hook, struct sk_buff *skb, return NF_ACCEPT; #ifdef CONFIG_NETFILTER_DEBUG - if (skb->dst == NULL) { + if (skb_dst(skb) == NULL) { printk(KERN_INFO "br_netfilter post_routing: skb->dst == NULL\n"); goto print_error; } diff --git a/net/core/dev.c b/net/core/dev.c index e2fcc5f1017..34b49a6a22f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1693,10 +1693,9 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, * If device doesnt need skb->dst, release it right now while * its hot in this cpu cache */ - if ((dev->priv_flags & IFF_XMIT_DST_RELEASE) && skb->dst) { - dst_release(skb->dst); - skb->dst = NULL; - } + if (dev->priv_flags & IFF_XMIT_DST_RELEASE) + skb_dst_drop(skb); + rc = ops->ndo_start_xmit(skb, dev); if (rc == 0) txq_trans_update(txq); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index a1cbce7fdae..c54229befcf 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1088,8 +1088,8 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new, struct neighbour *n1 = neigh; write_unlock_bh(&neigh->lock); /* On shaper/eql skb->dst->neighbour != neigh :( */ - if (skb->dst && skb->dst->neighbour) - n1 = skb->dst->neighbour; + if (skb_dst(skb) && skb_dst(skb)->neighbour) + n1 = skb_dst(skb)->neighbour; n1->output(skb); write_lock_bh(&neigh->lock); } @@ -1182,7 +1182,7 @@ EXPORT_SYMBOL(neigh_compat_output); int neigh_resolve_output(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct neighbour *neigh; int rc = 0; @@ -1229,7 +1229,7 @@ EXPORT_SYMBOL(neigh_resolve_output); int neigh_connected_output(struct sk_buff *skb) { int err; - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct neighbour *neigh = dst->neighbour; struct net_device *dev = neigh->dev; @@ -1298,8 +1298,7 @@ void pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p, if (time_before(tbl->proxy_timer.expires, sched_next)) sched_next = tbl->proxy_timer.expires; } - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); dev_hold(skb->dev); __skb_queue_tail(&tbl->proxy_queue, skb); mod_timer(&tbl->proxy_timer, sched_next); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 8e815e685f2..6adf19ec95c 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -381,7 +381,7 @@ static void kfree_skbmem(struct sk_buff *skb) static void skb_release_head_state(struct sk_buff *skb) { - dst_release(skb->dst); + skb_dst_drop(skb); #ifdef CONFIG_XFRM secpath_put(skb->sp); #endif @@ -521,7 +521,7 @@ static void __copy_skb_header(struct sk_buff *new, const struct sk_buff *old) new->transport_header = old->transport_header; new->network_header = old->network_header; new->mac_header = old->mac_header; - new->dst = dst_clone(old->dst); + skb_dst_set(new, dst_clone(skb_dst(old))); #ifdef CONFIG_XFRM new->sp = secpath_get(old->sp); #endif diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 2cf48ba0dbb..a0a36c9e6cc 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -507,7 +507,7 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) const struct iphdr *rxiph; struct sk_buff *skb; struct dst_entry *dst; - struct net *net = dev_net(rxskb->dst->dev); + struct net *net = dev_net(skb_dst(rxskb)->dev); struct sock *ctl_sk = net->dccp.v4_ctl_sk; /* Never send a reset in response to a reset. */ @@ -528,7 +528,7 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) rxiph = ip_hdr(rxskb); dccp_hdr(skb)->dccph_checksum = dccp_v4_csum_finish(skb, rxiph->saddr, rxiph->daddr); - skb->dst = dst_clone(dst); + skb_dst_set(skb, dst_clone(dst)); bh_lock_sock(ctl_sk); err = ip_build_and_send_pkt(skb, ctl_sk, diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index b963f35c65f..05ea7440d9e 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -314,8 +314,9 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) struct ipv6hdr *rxip6h; struct sk_buff *skb; struct flowi fl; - struct net *net = dev_net(rxskb->dst->dev); + struct net *net = dev_net(skb_dst(rxskb)->dev); struct sock *ctl_sk = net->dccp.v6_ctl_sk; + struct dst_entry *dst; if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET) return; @@ -342,8 +343,9 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) security_skb_classify_flow(rxskb, &fl); /* sk = NULL, but it is safe for now. RST socket required. */ - if (!ip6_dst_lookup(ctl_sk, &skb->dst, &fl)) { - if (xfrm_lookup(net, &skb->dst, &fl, NULL, 0) >= 0) { + if (!ip6_dst_lookup(ctl_sk, &dst, &fl)) { + if (xfrm_lookup(net, &dst, &fl, NULL, 0) >= 0) { + skb_dst_set(skb, dst); ip6_xmit(ctl_sk, skb, &fl, NULL, 0); DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS); DCCP_INC_STATS_BH(DCCP_MIB_OUTRSTS); diff --git a/net/dccp/output.c b/net/dccp/output.c index 36bcc00654d..c0e88c16d08 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -350,7 +350,7 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, /* Reserve space for headers. */ skb_reserve(skb, sk->sk_prot->max_header); - skb->dst = dst_clone(dst); + skb_dst_set(skb, dst_clone(dst)); dreq = dccp_rsk(req); if (inet_rsk(req)->acked) /* increase ISS upon retransmission */ diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index bccb3887773..a5e3a593e47 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -1075,6 +1075,7 @@ static int dn_accept(struct socket *sock, struct socket *newsock, int flags) int err = 0; unsigned char type; long timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + struct dst_entry *dst; lock_sock(sk); @@ -1102,8 +1103,9 @@ static int dn_accept(struct socket *sock, struct socket *newsock, int flags) } release_sock(sk); - dst_release(xchg(&newsk->sk_dst_cache, skb->dst)); - skb->dst = NULL; + dst = skb_dst(skb); + dst_release(xchg(&newsk->sk_dst_cache, dst)); + skb_dst_set(skb, NULL); DN_SK(newsk)->state = DN_CR; DN_SK(newsk)->addrrem = cb->src_port; diff --git a/net/decnet/dn_neigh.c b/net/decnet/dn_neigh.c index 05b5aa05e50..923786bd6d0 100644 --- a/net/decnet/dn_neigh.c +++ b/net/decnet/dn_neigh.c @@ -204,7 +204,7 @@ static void dn_short_error_report(struct neighbour *neigh, struct sk_buff *skb) static int dn_neigh_output_packet(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct dn_route *rt = (struct dn_route *)dst; struct neighbour *neigh = dst->neighbour; struct net_device *dev = neigh->dev; @@ -224,7 +224,7 @@ static int dn_neigh_output_packet(struct sk_buff *skb) static int dn_long_output(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct neighbour *neigh = dst->neighbour; struct net_device *dev = neigh->dev; int headroom = dev->hard_header_len + sizeof(struct dn_long_packet) + 3; @@ -270,7 +270,7 @@ static int dn_long_output(struct sk_buff *skb) static int dn_short_output(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct neighbour *neigh = dst->neighbour; struct net_device *dev = neigh->dev; int headroom = dev->hard_header_len + sizeof(struct dn_short_packet) + 2; @@ -313,7 +313,7 @@ static int dn_short_output(struct sk_buff *skb) */ static int dn_phase3_output(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct neighbour *neigh = dst->neighbour; struct net_device *dev = neigh->dev; int headroom = dev->hard_header_len + sizeof(struct dn_short_packet) + 2; diff --git a/net/decnet/dn_nsp_out.c b/net/decnet/dn_nsp_out.c index da04f459337..a65e929ce76 100644 --- a/net/decnet/dn_nsp_out.c +++ b/net/decnet/dn_nsp_out.c @@ -85,7 +85,7 @@ static void dn_nsp_send(struct sk_buff *skb) dst = sk_dst_check(sk, 0); if (dst) { try_again: - skb->dst = dst; + skb_dst_set(skb, dst); dst_output(skb); return; } @@ -582,7 +582,7 @@ static __inline__ void dn_nsp_do_disc(struct sock *sk, unsigned char msgflg, * to be able to send disc packets out which have no socket * associations. */ - skb->dst = dst_clone(dst); + skb_dst_set(skb, dst_clone(dst)); dst_output(skb); } @@ -611,7 +611,7 @@ void dn_nsp_return_disc(struct sk_buff *skb, unsigned char msgflg, int ddl = 0; gfp_t gfp = GFP_ATOMIC; - dn_nsp_do_disc(NULL, msgflg, reason, gfp, skb->dst, ddl, + dn_nsp_do_disc(NULL, msgflg, reason, gfp, skb_dst(skb), ddl, NULL, cb->src_port, cb->dst_port); } diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 0cc4394117d..1d6ca8a98dc 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -678,7 +678,7 @@ out: static int dn_output(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct dn_route *rt = (struct dn_route *)dst; struct net_device *dev = dst->dev; struct dn_skb_cb *cb = DN_SKB_CB(skb); @@ -717,7 +717,7 @@ error: static int dn_forward(struct sk_buff *skb) { struct dn_skb_cb *cb = DN_SKB_CB(skb); - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct dn_dev *dn_db = dst->dev->dn_ptr; struct dn_route *rt; struct neighbour *neigh = dst->neighbour; @@ -730,7 +730,7 @@ static int dn_forward(struct sk_buff *skb) goto drop; /* Ensure that we have enough space for headers */ - rt = (struct dn_route *)skb->dst; + rt = (struct dn_route *)skb_dst(skb); header_len = dn_db->use_long ? 21 : 6; if (skb_cow(skb, LL_RESERVED_SPACE(rt->u.dst.dev)+header_len)) goto drop; @@ -1392,7 +1392,8 @@ make_route: goto e_neighbour; hash = dn_hash(rt->fl.fld_src, rt->fl.fld_dst); - dn_insert_route(rt, hash, (struct dn_route **)&skb->dst); + dn_insert_route(rt, hash, &rt); + skb_dst_set(skb, &rt->u.dst); done: if (neigh) @@ -1424,7 +1425,7 @@ static int dn_route_input(struct sk_buff *skb) struct dn_skb_cb *cb = DN_SKB_CB(skb); unsigned hash = dn_hash(cb->src, cb->dst); - if (skb->dst) + if (skb_dst(skb)) return 0; rcu_read_lock(); @@ -1437,7 +1438,7 @@ static int dn_route_input(struct sk_buff *skb) (rt->fl.iif == cb->iif)) { dst_use(&rt->u.dst, jiffies); rcu_read_unlock(); - skb->dst = (struct dst_entry *)rt; + skb_dst_set(skb, (struct dst_entry *)rt); return 0; } } @@ -1449,7 +1450,7 @@ static int dn_route_input(struct sk_buff *skb) 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 dn_route *rt = (struct dn_route *)skb_dst(skb); struct rtmsg *r; struct nlmsghdr *nlh; unsigned char *b = skb_tail_pointer(skb); @@ -1554,7 +1555,7 @@ static int dn_cache_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, void err = dn_route_input(skb); local_bh_enable(); memset(cb, 0, sizeof(struct dn_skb_cb)); - rt = (struct dn_route *)skb->dst; + rt = (struct dn_route *)skb_dst(skb); if (!err && -rt->u.dst.error) err = rt->u.dst.error; } else { @@ -1570,7 +1571,7 @@ static int dn_cache_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, void skb->dev = NULL; if (err) goto out_free; - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); if (rtm->rtm_flags & RTM_F_NOTIFY) rt->rt_flags |= RTCF_NOTIFY; @@ -1622,15 +1623,15 @@ int dn_cache_dump(struct sk_buff *skb, struct netlink_callback *cb) rt = rcu_dereference(rt->u.dst.dn_next), idx++) { if (idx < s_idx) continue; - skb->dst = dst_clone(&rt->u.dst); + skb_dst_set(skb, dst_clone(&rt->u.dst)); if (dn_rt_fill_info(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, RTM_NEWROUTE, 1, NLM_F_MULTI) <= 0) { - dst_release(xchg(&skb->dst, NULL)); + skb_dst_drop(skb); rcu_read_unlock_bh(); goto done; } - dst_release(xchg(&skb->dst, NULL)); + skb_dst_drop(skb); } rcu_read_unlock_bh(); } diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 816494f271a..8a3881e28ac 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -468,7 +468,7 @@ int arp_find(unsigned char *haddr, struct sk_buff *skb) __be32 paddr; struct neighbour *n; - if (!skb->dst) { + if (!skb_dst(skb)) { printk(KERN_DEBUG "arp_find is called with dst==NULL\n"); kfree_skb(skb); return 1; diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 94f75efae93..97c410e8438 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -591,13 +591,13 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) goto relookup_failed; /* Ugh! */ - odst = skb_in->dst; + odst = skb_dst(skb_in); err = ip_route_input(skb_in, fl.fl4_dst, fl.fl4_src, RT_TOS(tos), rt2->u.dst.dev); dst_release(&rt2->u.dst); rt2 = skb_rtable(skb_in); - skb_in->dst = odst; + skb_dst_set(skb_in, odst); } if (err) @@ -659,7 +659,7 @@ static void icmp_unreach(struct sk_buff *skb) u32 info = 0; struct net *net; - net = dev_net(skb->dst->dev); + net = dev_net(skb_dst(skb)->dev); /* * Incomplete header ? @@ -822,7 +822,7 @@ static void icmp_echo(struct sk_buff *skb) { struct net *net; - net = dev_net(skb->dst->dev); + net = dev_net(skb_dst(skb)->dev); if (!net->ipv4.sysctl_icmp_echo_ignore_all) { struct icmp_bxm icmp_param; @@ -873,7 +873,7 @@ static void icmp_timestamp(struct sk_buff *skb) out: return; out_err: - ICMP_INC_STATS_BH(dev_net(skb->dst->dev), ICMP_MIB_INERRORS); + ICMP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), ICMP_MIB_INERRORS); goto out; } diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index afabd2758b6..01b4284ed69 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -311,7 +311,7 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, int size) return NULL; } - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); skb->dev = dev; skb_reserve(skb, LL_RESERVED_SPACE(dev)); @@ -659,7 +659,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc, return -1; } - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); skb_reserve(skb, LL_RESERVED_SPACE(dev)); diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index 0761cd9bbd1..a2991bc8e32 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -42,7 +42,7 @@ static int ip_forward_finish(struct sk_buff *skb) { struct ip_options * opt = &(IPCB(skb)->opt); - IP_INC_STATS_BH(dev_net(skb->dst->dev), IPSTATS_MIB_OUTFORWDATAGRAMS); + IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTFORWDATAGRAMS); if (unlikely(opt->optlen)) ip_forward_options(skb); @@ -123,7 +123,7 @@ sr_failed: too_many_hops: /* Tell the sender its packet died... */ - IP_INC_STATS_BH(dev_net(skb->dst->dev), IPSTATS_MIB_INHDRERRORS); + IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_INHDRERRORS); icmp_send(skb, ICMP_TIME_EXCEEDED, ICMP_EXC_TTL, 0); drop: kfree_skb(skb); diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index 7985346653b..1f1b82475ea 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -573,7 +573,7 @@ int ip_defrag(struct sk_buff *skb, u32 user) struct ipq *qp; struct net *net; - net = skb->dev ? dev_net(skb->dev) : dev_net(skb->dst->dev); + net = skb->dev ? dev_net(skb->dev) : dev_net(skb_dst(skb)->dev); IP_INC_STATS_BH(net, IPSTATS_MIB_REASMREQDS); /* Start by cleaning up the memory. */ diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 85ddad45a91..44e2a3d2359 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -643,8 +643,7 @@ static int ipgre_rcv(struct sk_buff *skb) stats->rx_packets++; stats->rx_bytes += len; skb->dev = tunnel->dev; - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); nf_reset(skb); skb_reset_network_header(skb); @@ -698,7 +697,7 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if ((dst = tiph->daddr) == 0) { /* NBMA tunnel */ - if (skb->dst == NULL) { + if (skb_dst(skb) == NULL) { stats->tx_fifo_errors++; goto tx_error; } @@ -712,7 +711,7 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) else if (skb->protocol == htons(ETH_P_IPV6)) { struct in6_addr *addr6; int addr_type; - struct neighbour *neigh = skb->dst->neighbour; + struct neighbour *neigh = skb_dst(skb)->neighbour; if (neigh == NULL) goto tx_error; @@ -766,10 +765,10 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (df) mtu = dst_mtu(&rt->u.dst) - dev->hard_header_len - tunnel->hlen; else - mtu = skb->dst ? dst_mtu(skb->dst) : dev->mtu; + mtu = skb_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu; - if (skb->dst) - skb->dst->ops->update_pmtu(skb->dst, mtu); + if (skb_dst(skb)) + skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); if (skb->protocol == htons(ETH_P_IP)) { df |= (old_iph->frag_off&htons(IP_DF)); @@ -783,14 +782,14 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) } #ifdef CONFIG_IPV6 else if (skb->protocol == htons(ETH_P_IPV6)) { - struct rt6_info *rt6 = (struct rt6_info *)skb->dst; + struct rt6_info *rt6 = (struct rt6_info *)skb_dst(skb); - if (rt6 && mtu < dst_mtu(skb->dst) && mtu >= IPV6_MIN_MTU) { + if (rt6 && mtu < dst_mtu(skb_dst(skb)) && mtu >= IPV6_MIN_MTU) { if ((tunnel->parms.iph.daddr && !ipv4_is_multicast(tunnel->parms.iph.daddr)) || rt6->rt6i_dst.plen == 128) { rt6->rt6i_flags |= RTF_MODIFIED; - skb->dst->metrics[RTAX_MTU-1] = mtu; + skb_dst(skb)->metrics[RTAX_MTU-1] = mtu; } } @@ -837,8 +836,8 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED | IPSKB_REROUTED); - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* * Push down and install the IPIP header. diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index cea784b0aa4..490ce20faf3 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -329,7 +329,7 @@ static int ip_rcv_finish(struct sk_buff *skb) * Initialise the virtual path cache for the packet. It describes * how the packet travels inside Linux networking. */ - if (skb->dst == NULL) { + if (skb_dst(skb) == NULL) { int err = ip_route_input(skb, iph->daddr, iph->saddr, iph->tos, skb->dev); if (unlikely(err)) { @@ -344,9 +344,9 @@ static int ip_rcv_finish(struct sk_buff *skb) } #ifdef CONFIG_NET_CLS_ROUTE - if (unlikely(skb->dst->tclassid)) { + if (unlikely(skb_dst(skb)->tclassid)) { struct ip_rt_acct *st = per_cpu_ptr(ip_rt_acct, smp_processor_id()); - u32 idx = skb->dst->tclassid; + u32 idx = skb_dst(skb)->tclassid; st[idx&0xFF].o_packets++; st[idx&0xFF].o_bytes += skb->len; st[(idx>>16)&0xFF].i_packets++; diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index 7e1074ffdbd..94bf105ef3c 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -143,7 +143,7 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb) __be32 addr; memcpy(&addr, sptr+soffset-1, 4); - if (inet_addr_type(dev_net(skb->dst->dev), addr) != RTN_LOCAL) { + if (inet_addr_type(dev_net(skb_dst(skb)->dev), addr) != RTN_LOCAL) { dopt->ts_needtime = 1; soffset += 8; } @@ -624,12 +624,12 @@ int ip_options_rcv_srr(struct sk_buff *skb) memcpy(&nexthop, &optptr[srrptr-1], 4); rt = skb_rtable(skb); - skb->dst = NULL; + skb_dst_set(skb, NULL); err = ip_route_input(skb, nexthop, iph->saddr, iph->tos, skb->dev); rt2 = skb_rtable(skb); if (err || (rt2->rt_type != RTN_UNICAST && rt2->rt_type != RTN_LOCAL)) { ip_rt_put(rt2); - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); return -EINVAL; } ip_rt_put(rt); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 8d845ebfcca..3d6167fb2d9 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -95,7 +95,7 @@ int __ip_local_out(struct sk_buff *skb) iph->tot_len = htons(skb->len); ip_send_check(iph); - return nf_hook(PF_INET, NF_INET_LOCAL_OUT, skb, NULL, skb->dst->dev, + return nf_hook(PF_INET, NF_INET_LOCAL_OUT, skb, NULL, skb_dst(skb)->dev, dst_output); } @@ -118,7 +118,7 @@ static int ip_dev_loopback_xmit(struct sk_buff *newskb) __skb_pull(newskb, skb_network_offset(newskb)); newskb->pkt_type = PACKET_LOOPBACK; newskb->ip_summed = CHECKSUM_UNNECESSARY; - WARN_ON(!newskb->dst); + WARN_ON(!skb_dst(newskb)); netif_rx(newskb); return 0; } @@ -176,7 +176,7 @@ EXPORT_SYMBOL_GPL(ip_build_and_send_pkt); static inline int ip_finish_output2(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct rtable *rt = (struct rtable *)dst; struct net_device *dev = dst->dev; unsigned int hh_len = LL_RESERVED_SPACE(dev); @@ -217,14 +217,14 @@ static inline int ip_skb_dst_mtu(struct sk_buff *skb) struct inet_sock *inet = skb->sk ? inet_sk(skb->sk) : NULL; return (inet && inet->pmtudisc == IP_PMTUDISC_PROBE) ? - skb->dst->dev->mtu : dst_mtu(skb->dst); + skb_dst(skb)->dev->mtu : dst_mtu(skb_dst(skb)); } static int ip_finish_output(struct sk_buff *skb) { #if defined(CONFIG_NETFILTER) && defined(CONFIG_XFRM) /* Policy lookup after SNAT yielded a new policy */ - if (skb->dst->xfrm != NULL) { + if (skb_dst(skb)->xfrm != NULL) { IPCB(skb)->flags |= IPSKB_REROUTED; return dst_output(skb); } @@ -296,7 +296,7 @@ int ip_mc_output(struct sk_buff *skb) int ip_output(struct sk_buff *skb) { - struct net_device *dev = skb->dst->dev; + struct net_device *dev = skb_dst(skb)->dev; IP_UPD_PO_STATS(dev_net(dev), IPSTATS_MIB_OUT, skb->len); @@ -355,7 +355,7 @@ int ip_queue_xmit(struct sk_buff *skb, int ipfragok) } sk_setup_caps(sk, &rt->u.dst); } - skb->dst = dst_clone(&rt->u.dst); + skb_dst_set(skb, dst_clone(&rt->u.dst)); packet_routed: if (opt && opt->is_strictroute && rt->rt_dst != rt->rt_gateway) @@ -401,8 +401,8 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) to->pkt_type = from->pkt_type; to->priority = from->priority; to->protocol = from->protocol; - dst_release(to->dst); - to->dst = dst_clone(from->dst); + skb_dst_drop(to); + skb_dst_set(to, dst_clone(skb_dst(from))); to->dev = from->dev; to->mark = from->mark; @@ -1294,7 +1294,7 @@ int ip_push_pending_frames(struct sock *sk) * on dst refcount */ inet->cork.dst = NULL; - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); if (iph->protocol == IPPROTO_ICMP) icmp_out_count(net, ((struct icmphdr *) diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 0c6e7bf18a4..93e2b787da2 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -370,8 +370,7 @@ static int ipip_rcv(struct sk_buff *skb) tunnel->dev->stats.rx_packets++; tunnel->dev->stats.rx_bytes += skb->len; skb->dev = tunnel->dev; - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); nf_reset(skb); ipip_ecn_decapsulate(iph, skb); netif_rx(skb); @@ -447,15 +446,15 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (tiph->frag_off) mtu = dst_mtu(&rt->u.dst) - sizeof(struct iphdr); else - mtu = skb->dst ? dst_mtu(skb->dst) : dev->mtu; + mtu = skb_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu; if (mtu < 68) { stats->collisions++; ip_rt_put(rt); goto tx_error; } - if (skb->dst) - skb->dst->ops->update_pmtu(skb->dst, mtu); + if (skb_dst(skb)) + skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); df |= (old_iph->frag_off&htons(IP_DF)); @@ -502,8 +501,8 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED | IPSKB_REROUTED); - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* * Push down and install the IPIP header. diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 69dd058283e..ffd98610446 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -651,7 +651,7 @@ static int ipmr_cache_report(struct net *net, ip_hdr(skb)->protocol = 0; /* Flag to the kernel this is a route add */ msg = (struct igmpmsg *)skb_network_header(skb); msg->im_vif = vifi; - skb->dst = dst_clone(pkt->dst); + skb_dst_set(skb, dst_clone(skb_dst(pkt))); /* * Add our header @@ -1201,7 +1201,7 @@ static void ip_encap(struct sk_buff *skb, __be32 saddr, __be32 daddr) iph->protocol = IPPROTO_IPIP; iph->ihl = 5; iph->tot_len = htons(skb->len); - ip_select_ident(iph, skb->dst, NULL); + ip_select_ident(iph, skb_dst(skb), NULL); ip_send_check(iph); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); @@ -1212,7 +1212,7 @@ static inline int ipmr_forward_finish(struct sk_buff *skb) { struct ip_options * opt = &(IPCB(skb)->opt); - IP_INC_STATS_BH(dev_net(skb->dst->dev), IPSTATS_MIB_OUTFORWDATAGRAMS); + IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTFORWDATAGRAMS); if (unlikely(opt->optlen)) ip_forward_options(skb); @@ -1290,8 +1290,8 @@ static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi) vif->pkt_out++; vif->bytes_out += skb->len; - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); ip_decrease_ttl(ip_hdr(skb)); /* FIXME: forward and output firewalls used to be called here. @@ -1543,8 +1543,7 @@ static int __pim_rcv(struct sk_buff *skb, unsigned int pimlen) skb->protocol = htons(ETH_P_IP); skb->ip_summed = 0; skb->pkt_type = PACKET_HOST; - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); reg_dev->stats.rx_bytes += skb->len; reg_dev->stats.rx_packets++; nf_reset(skb); diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index fdf6811c31a..1725dc0ef68 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -12,7 +12,7 @@ /* route_me_harder function, used by iptable_nat, iptable_mangle + ip_queue */ int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type) { - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); const struct iphdr *iph = ip_hdr(skb); struct rtable *rt; struct flowi fl = {}; @@ -41,8 +41,8 @@ int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type) return -1; /* Drop old route. */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); } else { /* non-local src, find valid iif to satisfy * rp-filter when calling ip_route_input. */ @@ -50,7 +50,7 @@ int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type) if (ip_route_output_key(net, &rt, &fl) != 0) return -1; - odst = skb->dst; + odst = skb_dst(skb); if (ip_route_input(skb, iph->daddr, iph->saddr, RT_TOS(iph->tos), rt->u.dst.dev) != 0) { dst_release(&rt->u.dst); @@ -60,18 +60,22 @@ int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type) dst_release(odst); } - if (skb->dst->error) + if (skb_dst(skb)->error) return -1; #ifdef CONFIG_XFRM if (!(IPCB(skb)->flags & IPSKB_XFRM_TRANSFORMED) && - xfrm_decode_session(skb, &fl, AF_INET) == 0) - if (xfrm_lookup(net, &skb->dst, &fl, skb->sk, 0)) + xfrm_decode_session(skb, &fl, AF_INET) == 0) { + struct dst_entry *dst = skb_dst(skb); + skb_dst_set(skb, NULL); + if (xfrm_lookup(net, &dst, &fl, skb->sk, 0)) return -1; + skb_dst_set(skb, dst); + } #endif /* Change in oif may mean change in hh_len. */ - hh_len = skb->dst->dev->hard_header_len; + hh_len = skb_dst(skb)->dev->hard_header_len; if (skb_headroom(skb) < hh_len && pskb_expand_head(skb, hh_len - skb_headroom(skb), 0, GFP_ATOMIC)) return -1; @@ -92,7 +96,7 @@ int ip_xfrm_me_harder(struct sk_buff *skb) if (xfrm_decode_session(skb, &fl, AF_INET) < 0) return -1; - dst = skb->dst; + dst = skb_dst(skb); if (dst->xfrm) dst = ((struct xfrm_dst *)dst)->route; dst_hold(dst); @@ -100,11 +104,11 @@ int ip_xfrm_me_harder(struct sk_buff *skb) if (xfrm_lookup(dev_net(dst->dev), &dst, &fl, skb->sk, 0) < 0) return -1; - dst_release(skb->dst); - skb->dst = dst; + skb_dst_drop(skb); + skb_dst_set(skb, dst); /* Change in oif may mean change in hh_len. */ - hh_len = skb->dst->dev->hard_header_len; + hh_len = skb_dst(skb)->dev->hard_header_len; if (skb_headroom(skb) < hh_len && pskb_expand_head(skb, hh_len - skb_headroom(skb), 0, GFP_ATOMIC)) return -1; diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c index 0b4b6e0ff2b..c93ae44bff2 100644 --- a/net/ipv4/netfilter/ipt_REJECT.c +++ b/net/ipv4/netfilter/ipt_REJECT.c @@ -108,17 +108,16 @@ static void send_reset(struct sk_buff *oldskb, int hook) addr_type = RTN_LOCAL; /* ip_route_me_harder expects skb->dst to be set */ - dst_hold(oldskb->dst); - nskb->dst = oldskb->dst; + skb_dst_set(nskb, dst_clone(skb_dst(oldskb))); if (ip_route_me_harder(nskb, addr_type)) goto free_nskb; - niph->ttl = dst_metric(nskb->dst, RTAX_HOPLIMIT); + niph->ttl = dst_metric(skb_dst(nskb), RTAX_HOPLIMIT); nskb->ip_summed = CHECKSUM_NONE; /* "Never happens" */ - if (nskb->len > dst_mtu(nskb->dst)) + if (nskb->len > dst_mtu(skb_dst(nskb))) goto free_nskb; nf_ct_attach(nskb, oldskb); diff --git a/net/ipv4/netfilter/nf_nat_standalone.c b/net/ipv4/netfilter/nf_nat_standalone.c index b7dd695691a..5567bd0d075 100644 --- a/net/ipv4/netfilter/nf_nat_standalone.c +++ b/net/ipv4/netfilter/nf_nat_standalone.c @@ -167,10 +167,9 @@ nf_nat_in(unsigned int hooknum, ret = nf_nat_fn(hooknum, skb, in, out, okfn); if (ret != NF_DROP && ret != NF_STOLEN && - daddr != ip_hdr(skb)->daddr) { - dst_release(skb->dst); - skb->dst = NULL; - } + daddr != ip_hdr(skb)->daddr) + skb_dst_drop(skb); + return ret; } diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index f774651f0a4..3dc9171a272 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -343,7 +343,7 @@ static int raw_send_hdrinc(struct sock *sk, void *from, size_t length, skb->priority = sk->sk_priority; skb->mark = sk->sk_mark; - skb->dst = dst_clone(&rt->u.dst); + skb_dst_set(skb, dst_clone(&rt->u.dst)); skb_reset_network_header(skb); iph = ip_hdr(skb); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index f20060ac2f0..a849bb15d86 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1118,7 +1118,7 @@ restart: if (rp) *rp = rth; else - skb->dst = &rth->u.dst; + skb_dst_set(skb, &rth->u.dst); return 0; } @@ -1217,7 +1217,7 @@ restart: if (rp) *rp = rt; else - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); return 0; } @@ -2251,7 +2251,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr, dst_use(&rth->u.dst, jiffies); RT_CACHE_STAT_INC(in_hit); rcu_read_unlock(); - skb->dst = &rth->u.dst; + skb_dst_set(skb, &rth->u.dst); return 0; } RT_CACHE_STAT_INC(in_hlist_search); @@ -2934,7 +2934,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void if (err) goto errout_free; - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); if (rtm->rtm_flags & RTM_F_NOTIFY) rt->rt_flags |= RTCF_NOTIFY; @@ -2975,15 +2975,15 @@ int ip_rt_dump(struct sk_buff *skb, struct netlink_callback *cb) continue; if (rt_is_expired(rt)) continue; - skb->dst = dst_clone(&rt->u.dst); + skb_dst_set(skb, dst_clone(&rt->u.dst)); if (rt_fill_info(net, skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, RTM_NEWROUTE, 1, NLM_F_MULTI) <= 0) { - dst_release(xchg(&skb->dst, NULL)); + skb_dst_drop(skb); rcu_read_unlock_bh(); goto done; } - dst_release(xchg(&skb->dst, NULL)); + skb_dst_drop(skb); } rcu_read_unlock_bh(); } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 319c8852644..5a1ca2698c8 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -590,7 +590,7 @@ static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb) arg.csumoffset = offsetof(struct tcphdr, check) / 2; arg.flags = (sk && inet_sk(sk)->transparent) ? IP_REPLY_ARG_NOSRCCHECK : 0; - net = dev_net(skb->dst->dev); + net = dev_net(skb_dst(skb)->dev); ip_send_reply(net->ipv4.tcp_sock, skb, &arg, arg.iov[0].iov_len); @@ -617,7 +617,7 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, ]; } rep; struct ip_reply_arg arg; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); memset(&rep.th, 0, sizeof(struct tcphdr)); memset(&arg, 0, sizeof(arg)); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 79c39dc9b01..416fc4c2e7e 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2202,7 +2202,7 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst, /* Reserve space for headers. */ skb_reserve(skb, MAX_TCP_HEADER); - skb->dst = dst_clone(dst); + skb_dst_set(skb, dst_clone(dst)); mss = dst_metric(dst, RTAX_ADVMSS); if (tp->rx_opt.user_mss && tp->rx_opt.user_mss < mss) diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 7a1d1ce22e6..8f4158d7c9a 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -328,7 +328,7 @@ static inline struct sock *__udp4_lib_lookup_skb(struct sk_buff *skb, if (unlikely(sk = skb_steal_sock(skb))) return sk; else - return __udp4_lib_lookup(dev_net(skb->dst->dev), iph->saddr, sport, + return __udp4_lib_lookup(dev_net(skb_dst(skb)->dev), iph->saddr, sport, iph->daddr, dport, inet_iif(skb), udptable); } @@ -1237,7 +1237,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, struct sock *sk; struct udphdr *uh; unsigned short ulen; - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb_rtable(skb); __be32 saddr, daddr; struct net *net = dev_net(skb->dev); diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c index 4ec2162a437..f9f922a0ba8 100644 --- a/net/ipv4/xfrm4_input.c +++ b/net/ipv4/xfrm4_input.c @@ -23,7 +23,7 @@ int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb) static inline int xfrm4_rcv_encap_finish(struct sk_buff *skb) { - if (skb->dst == NULL) { + if (skb_dst(skb) == NULL) { const struct iphdr *iph = ip_hdr(skb); if (ip_route_input(skb, iph->daddr, iph->saddr, iph->tos, diff --git a/net/ipv4/xfrm4_mode_tunnel.c b/net/ipv4/xfrm4_mode_tunnel.c index 7135279f3f8..3444f3b34ec 100644 --- a/net/ipv4/xfrm4_mode_tunnel.c +++ b/net/ipv4/xfrm4_mode_tunnel.c @@ -28,7 +28,7 @@ static inline void ipip_ecn_decapsulate(struct sk_buff *skb) */ static int xfrm4_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct iphdr *top_iph; int flags; @@ -41,7 +41,7 @@ static int xfrm4_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) top_iph->ihl = 5; top_iph->version = 4; - top_iph->protocol = xfrm_af2proto(skb->dst->ops->family); + top_iph->protocol = xfrm_af2proto(skb_dst(skb)->ops->family); /* DS disclosed */ top_iph->tos = INET_ECN_encapsulate(XFRM_MODE_SKB_CB(skb)->tos, diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c index 8c3180adddb..c908bd99bcb 100644 --- a/net/ipv4/xfrm4_output.c +++ b/net/ipv4/xfrm4_output.c @@ -29,7 +29,7 @@ static int xfrm4_tunnel_check_size(struct sk_buff *skb) if (!(ip_hdr(skb)->frag_off & htons(IP_DF)) || skb->local_df) goto out; - dst = skb->dst; + dst = skb_dst(skb); mtu = dst_mtu(dst); if (skb->len > mtu) { icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); @@ -72,7 +72,7 @@ EXPORT_SYMBOL(xfrm4_prepare_output); static int xfrm4_output_finish(struct sk_buff *skb) { #ifdef CONFIG_NETFILTER - if (!skb->dst->xfrm) { + if (!skb_dst(skb)->xfrm) { IPCB(skb)->flags |= IPSKB_REROUTED; return dst_output(skb); } @@ -87,6 +87,6 @@ static int xfrm4_output_finish(struct sk_buff *skb) int xfrm4_output(struct sk_buff *skb) { return NF_HOOK_COND(PF_INET, NF_INET_POST_ROUTING, skb, - NULL, skb->dst->dev, xfrm4_output_finish, + NULL, skb_dst(skb)->dev, xfrm4_output_finish, !(IPCB(skb)->flags & IPSKB_REROUTED)); } diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 1c7f400a3cf..4aae658e550 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -277,7 +277,7 @@ static int ipv6_destopt_rcv(struct sk_buff *skb) if (!pskb_may_pull(skb, skb_transport_offset(skb) + 8) || !pskb_may_pull(skb, (skb_transport_offset(skb) + ((skb_transport_header(skb)[1] + 1) << 3)))) { - IP6_INC_STATS_BH(dev_net(skb->dst->dev), ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); kfree_skb(skb); return -1; @@ -288,7 +288,7 @@ static int ipv6_destopt_rcv(struct sk_buff *skb) dstbuf = opt->dst1; #endif - dst = dst_clone(skb->dst); + dst = dst_clone(skb_dst(skb)); if (ip6_parse_tlv(tlvprocdestopt_lst, skb)) { dst_release(dst); skb->transport_header += (skb_transport_header(skb)[1] + 1) << 3; @@ -333,7 +333,7 @@ static int ipv6_rthdr_rcv(struct sk_buff *skb) if (!pskb_may_pull(skb, skb_transport_offset(skb) + 8) || !pskb_may_pull(skb, (skb_transport_offset(skb) + ((skb_transport_header(skb)[1] + 1) << 3)))) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); kfree_skb(skb); return -1; @@ -343,7 +343,7 @@ static int ipv6_rthdr_rcv(struct sk_buff *skb) if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr) || skb->pkt_type != PACKET_HOST) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; @@ -358,7 +358,7 @@ looped_back: * processed by own */ if (!addr) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; @@ -384,7 +384,7 @@ looped_back: goto unknown_rh; /* Silently discard invalid RTH type 2 */ if (hdr->hdrlen != 2 || hdr->segments_left != 1) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); kfree_skb(skb); return -1; @@ -403,7 +403,7 @@ looped_back: n = hdr->hdrlen >> 1; if (hdr->segments_left > n) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, ((&hdr->segments_left) - @@ -417,7 +417,7 @@ looped_back: if (skb_cloned(skb)) { /* the copy is a forwarded packet */ if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); return -1; @@ -440,13 +440,13 @@ looped_back: if (xfrm6_input_addr(skb, (xfrm_address_t *)addr, (xfrm_address_t *)&ipv6_hdr(skb)->saddr, IPPROTO_ROUTING) < 0) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; } - if (!ipv6_chk_home_addr(dev_net(skb->dst->dev), addr)) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + if (!ipv6_chk_home_addr(dev_net(skb_dst(skb)->dev), addr)) { + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; @@ -458,7 +458,7 @@ looped_back: } if (ipv6_addr_is_multicast(addr)) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; @@ -468,17 +468,17 @@ looped_back: ipv6_addr_copy(addr, &ipv6_hdr(skb)->daddr); ipv6_addr_copy(&ipv6_hdr(skb)->daddr, &daddr); - dst_release(xchg(&skb->dst, NULL)); + skb_dst_drop(skb); ip6_route_input(skb); - if (skb->dst->error) { + if (skb_dst(skb)->error) { skb_push(skb, skb->data - skb_network_header(skb)); dst_input(skb); return -1; } - if (skb->dst->dev->flags&IFF_LOOPBACK) { + if (skb_dst(skb)->dev->flags&IFF_LOOPBACK) { if (ipv6_hdr(skb)->hop_limit <= 1) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, 0, skb->dev); @@ -494,7 +494,7 @@ looped_back: return -1; unknown_rh: - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->type) - skb_network_header(skb)); return -1; @@ -552,11 +552,11 @@ void ipv6_exthdrs_exit(void) **********************************/ /* - * Note: we cannot rely on skb->dst before we assign it in ip6_route_input(). + * Note: we cannot rely on skb_dst(skb) before we assign it in ip6_route_input(). */ static inline struct inet6_dev *ipv6_skb_idev(struct sk_buff *skb) { - return skb->dst ? ip6_dst_idev(skb->dst) : __in6_dev_get(skb->dev); + return skb_dst(skb) ? ip6_dst_idev(skb_dst(skb)) : __in6_dev_get(skb->dev); } /* Router Alert as of RFC 2711 */ @@ -581,7 +581,7 @@ static int ipv6_hop_jumbo(struct sk_buff *skb, int optoff) { const unsigned char *nh = skb_network_header(skb); u32 pkt_len; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); if (nh[optoff + 1] != 4 || (optoff & 3) != 2) { LIMIT_NETDEBUG(KERN_DEBUG "ipv6_hop_jumbo: wrong jumbo opt length/alignment %d\n", diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index 3c3732d50c1..cc4797dd832 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -228,7 +228,7 @@ int inet6_csk_xmit(struct sk_buff *skb, int ipfragok) __inet6_csk_dst_store(sk, dst, NULL, NULL); } - skb->dst = dst_clone(dst); + skb_dst_set(skb, dst_clone(dst)); /* Restore final destination back after routing done */ ipv6_addr_copy(&fl.fl6_dst, &np->daddr); diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index bc1a920c34a..c3a07d75b5f 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -48,7 +48,7 @@ inline int ip6_rcv_finish( struct sk_buff *skb) { - if (skb->dst == NULL) + if (skb_dst(skb) == NULL) ip6_route_input(skb); return dst_input(skb); @@ -91,7 +91,7 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt * arrived via the sending interface (ethX), because of the * nature of scoping architecture. --yoshfuji */ - IP6CB(skb)->iif = skb->dst ? ip6_dst_idev(skb->dst)->dev->ifindex : dev->ifindex; + IP6CB(skb)->iif = skb_dst(skb) ? ip6_dst_idev(skb_dst(skb))->dev->ifindex : dev->ifindex; if (unlikely(!pskb_may_pull(skb, sizeof(*hdr)))) goto err; @@ -161,7 +161,7 @@ static int ip6_input_finish(struct sk_buff *skb) int nexthdr, raw; u8 hash; struct inet6_dev *idev; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); /* * Parse extension headers @@ -169,7 +169,7 @@ static int ip6_input_finish(struct sk_buff *skb) rcu_read_lock(); resubmit: - idev = ip6_dst_idev(skb->dst); + idev = ip6_dst_idev(skb_dst(skb)); if (!pskb_pull(skb, skb_transport_offset(skb))) goto discard; nhoff = IP6CB(skb)->nhoff; @@ -242,8 +242,8 @@ int ip6_mc_input(struct sk_buff *skb) struct ipv6hdr *hdr; int deliver; - IP6_UPD_PO_STATS_BH(dev_net(skb->dst->dev), - ip6_dst_idev(skb->dst), IPSTATS_MIB_INMCAST, + IP6_UPD_PO_STATS_BH(dev_net(skb_dst(skb)->dev), + ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INMCAST, skb->len); hdr = ipv6_hdr(skb); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 735a2bf4b5f..c8dc8e5a822 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -78,7 +78,7 @@ int __ip6_local_out(struct sk_buff *skb) len = 0; ipv6_hdr(skb)->payload_len = htons(len); - return nf_hook(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, skb->dst->dev, + return nf_hook(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, skb_dst(skb)->dev, dst_output); } @@ -96,7 +96,7 @@ EXPORT_SYMBOL_GPL(ip6_local_out); static int ip6_output_finish(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); if (dst->hh) return neigh_hh_output(dst->hh, skb); @@ -117,7 +117,7 @@ static int ip6_dev_loopback_xmit(struct sk_buff *newskb) __skb_pull(newskb, skb_network_offset(newskb)); newskb->pkt_type = PACKET_LOOPBACK; newskb->ip_summed = CHECKSUM_UNNECESSARY; - WARN_ON(!newskb->dst); + WARN_ON(!skb_dst(newskb)); netif_rx(newskb); return 0; @@ -126,7 +126,7 @@ static int ip6_dev_loopback_xmit(struct sk_buff *newskb) static int ip6_output2(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct net_device *dev = dst->dev; skb->protocol = htons(ETH_P_IPV6); @@ -134,7 +134,7 @@ static int ip6_output2(struct sk_buff *skb) if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr)) { struct ipv6_pinfo* np = skb->sk ? inet6_sk(skb->sk) : NULL; - struct inet6_dev *idev = ip6_dst_idev(skb->dst); + struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb)); if (!(dev->flags & IFF_LOOPBACK) && (!np || np->mc_loop) && ((mroute6_socket(dev_net(dev)) && @@ -172,21 +172,21 @@ static inline int ip6_skb_dst_mtu(struct sk_buff *skb) struct ipv6_pinfo *np = skb->sk ? inet6_sk(skb->sk) : NULL; return (np && np->pmtudisc == IPV6_PMTUDISC_PROBE) ? - skb->dst->dev->mtu : dst_mtu(skb->dst); + skb_dst(skb)->dev->mtu : dst_mtu(skb_dst(skb)); } int ip6_output(struct sk_buff *skb) { - struct inet6_dev *idev = ip6_dst_idev(skb->dst); + struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb)); if (unlikely(idev->cnf.disable_ipv6)) { - IP6_INC_STATS(dev_net(skb->dst->dev), idev, + IP6_INC_STATS(dev_net(skb_dst(skb)->dev), idev, IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); return 0; } if ((skb->len > ip6_skb_dst_mtu(skb) && !skb_is_gso(skb)) || - dst_allfrag(skb->dst)) + dst_allfrag(skb_dst(skb))) return ip6_fragment(skb, ip6_output2); else return ip6_output2(skb); @@ -202,7 +202,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, struct net *net = sock_net(sk); struct ipv6_pinfo *np = inet6_sk(sk); struct in6_addr *first_hop = &fl->fl6_dst; - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct ipv6hdr *hdr; u8 proto = fl->proto; int seg_len = skb->len; @@ -222,7 +222,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, if (skb_headroom(skb) < head_room) { struct sk_buff *skb2 = skb_realloc_headroom(skb, head_room); if (skb2 == NULL) { - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); return -ENOBUFS; @@ -276,7 +276,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, mtu = dst_mtu(dst); if ((skb->len <= mtu) || skb->local_df || skb_is_gso(skb)) { - IP6_UPD_PO_STATS(net, ip6_dst_idev(skb->dst), + IP6_UPD_PO_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUT, skb->len); return NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, dst->dev, dst_output); @@ -286,7 +286,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, printk(KERN_DEBUG "IPv6: sending pkt_too_big to self\n"); skb->dev = dst->dev; icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), IPSTATS_MIB_FRAGFAILS); + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return -EMSGSIZE; } @@ -416,7 +416,7 @@ static inline int ip6_forward_finish(struct sk_buff *skb) int ip6_forward(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct ipv6hdr *hdr = ipv6_hdr(skb); struct inet6_skb_parm *opt = IP6CB(skb); struct net *net = dev_net(dst->dev); @@ -485,7 +485,7 @@ int ip6_forward(struct sk_buff *skb) IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_INDISCARDS); goto drop; } - dst = skb->dst; + dst = skb_dst(skb); /* IPv6 specs say nothing about it, but it is clear that we cannot send redirects to source routed frames. @@ -566,8 +566,8 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) to->pkt_type = from->pkt_type; to->priority = from->priority; to->protocol = from->protocol; - dst_release(to->dst); - to->dst = dst_clone(from->dst); + skb_dst_drop(to); + skb_dst_set(to, dst_clone(skb_dst(from))); to->dev = from->dev; to->mark = from->mark; @@ -624,7 +624,7 @@ int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr) static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) { struct sk_buff *frag; - struct rt6_info *rt = (struct rt6_info*)skb->dst; + struct rt6_info *rt = (struct rt6_info*)skb_dst(skb); struct ipv6_pinfo *np = skb->sk ? inet6_sk(skb->sk) : NULL; struct ipv6hdr *tmp_hdr; struct frag_hdr *fh; @@ -632,7 +632,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) __be32 frag_id = 0; int ptr, offset = 0, err=0; u8 *prevhdr, nexthdr = 0; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); hlen = ip6_find_1stfragopt(skb, &prevhdr); nexthdr = *prevhdr; @@ -644,9 +644,9 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) * check should be redundant, but it's free.) */ if (!skb->local_df) { - skb->dev = skb->dst->dev; + skb->dev = skb_dst(skb)->dev; icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return -EMSGSIZE; @@ -696,7 +696,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) *prevhdr = NEXTHDR_FRAGMENT; tmp_hdr = kmemdup(skb_network_header(skb), hlen, GFP_ATOMIC); if (!tmp_hdr) { - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); return -ENOMEM; } @@ -809,7 +809,7 @@ slow_path: if ((frag = alloc_skb(len+hlen+sizeof(struct frag_hdr)+LL_ALLOCATED_SPACE(rt->u.dst.dev), GFP_ATOMIC)) == NULL) { NETDEBUG(KERN_INFO "IPv6: frag: no memory for new fragment!\n"); - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); err = -ENOMEM; goto fail; @@ -873,16 +873,16 @@ slow_path: if (err) goto fail; - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGCREATES); } - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGOKS); kfree_skb(skb); return err; fail: - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return err; @@ -1516,10 +1516,10 @@ int ip6_push_pending_frames(struct sock *sk) skb->priority = sk->sk_priority; skb->mark = sk->sk_mark; - skb->dst = dst_clone(&rt->u.dst); + skb_dst_set(skb, dst_clone(&rt->u.dst)); IP6_UPD_PO_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len); if (proto == IPPROTO_ICMPV6) { - struct inet6_dev *idev = ip6_dst_idev(skb->dst); + struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb)); ICMP6MSGOUT_INC_STATS_BH(net, idev, icmp6_hdr(skb)->icmp6_type); ICMP6_INC_STATS_BH(net, idev, ICMP6_MIB_OUTMSGS); @@ -1545,8 +1545,8 @@ void ip6_flush_pending_frames(struct sock *sk) struct sk_buff *skb; while ((skb = __skb_dequeue_tail(&sk->sk_write_queue)) != NULL) { - if (skb->dst) - IP6_INC_STATS(sock_net(sk), ip6_dst_idev(skb->dst), + if (skb_dst(skb)) + IP6_INC_STATS(sock_net(sk), ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); } diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index af256d47fd3..404d16a97d5 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -532,8 +532,8 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (!skb2) return 0; - dst_release(skb2->dst); - skb2->dst = NULL; + skb_dst_drop(skb2); + skb_pull(skb2, offset); skb_reset_network_header(skb2); eiph = ip_hdr(skb2); @@ -560,21 +560,21 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, ip_rt_put(rt); goto out; } - skb2->dst = (struct dst_entry *)rt; + skb_dst_set(skb2, (struct dst_entry *)rt); } else { ip_rt_put(rt); if (ip_route_input(skb2, eiph->daddr, eiph->saddr, eiph->tos, skb2->dev) || - skb2->dst->dev->type != ARPHRD_TUNNEL) + skb_dst(skb2)->dev->type != ARPHRD_TUNNEL) goto out; } /* change mtu on this route */ if (rel_type == ICMP_DEST_UNREACH && rel_code == ICMP_FRAG_NEEDED) { - if (rel_info > dst_mtu(skb2->dst)) + if (rel_info > dst_mtu(skb_dst(skb2))) goto out; - skb2->dst->ops->update_pmtu(skb2->dst, rel_info); + skb_dst(skb2)->ops->update_pmtu(skb_dst(skb2), rel_info); } icmp_send(skb2, rel_type, rel_code, htonl(rel_info)); @@ -606,8 +606,7 @@ ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (!skb2) return 0; - dst_release(skb2->dst); - skb2->dst = NULL; + skb_dst_drop(skb2); skb_pull(skb2, offset); skb_reset_network_header(skb2); @@ -720,8 +719,7 @@ static int ip6_tnl_rcv(struct sk_buff *skb, __u16 protocol, skb->pkt_type = PACKET_HOST; memset(skb->cb, 0, sizeof(struct inet6_skb_parm)); skb->dev = t->dev; - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); nf_reset(skb); dscp_ecn_decapsulate(t, ipv6h, skb); @@ -885,8 +883,8 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, } if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; - if (skb->dst) - skb->dst->ops->update_pmtu(skb->dst, mtu); + if (skb_dst(skb)) + skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); if (skb->len > mtu) { *pmtu = mtu; err = -EMSGSIZE; @@ -910,8 +908,8 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, kfree_skb(skb); skb = new_skb; } - dst_release(skb->dst); - skb->dst = dst_clone(dst); + skb_dst_drop(skb); + skb_dst_set(skb, dst_clone(dst)); skb->transport_header = skb->network_header; diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 228be551e9c..a35d8fc55b0 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -398,10 +398,9 @@ static int pim6_rcv(struct sk_buff *skb) skb->protocol = htons(ETH_P_IPV6); skb->ip_summed = 0; skb->pkt_type = PACKET_HOST; - dst_release(skb->dst); + skb_dst_drop(skb); reg_dev->stats.rx_bytes += skb->len; reg_dev->stats.rx_packets++; - skb->dst = NULL; nf_reset(skb); netif_rx(skb); dev_put(reg_dev); @@ -849,7 +848,7 @@ static int ip6mr_cache_report(struct net *net, struct sk_buff *pkt, mifi_t mifi, ipv6_addr_copy(&msg->im6_src, &ipv6_hdr(pkt)->saddr); ipv6_addr_copy(&msg->im6_dst, &ipv6_hdr(pkt)->daddr); - skb->dst = dst_clone(pkt->dst); + skb_dst_set(skb, dst_clone(skb_dst(pkt))); skb->ip_summed = CHECKSUM_UNNECESSARY; } @@ -1487,7 +1486,7 @@ int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg) static inline int ip6mr_forward2_finish(struct sk_buff *skb) { - IP6_INC_STATS_BH(dev_net(skb->dst->dev), ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTFORWDATAGRAMS); return dst_output(skb); } @@ -1532,8 +1531,8 @@ static int ip6mr_forward2(struct sk_buff *skb, struct mfc6_cache *c, int vifi) if (!dst) goto out_free; - dst_release(skb->dst); - skb->dst = dst; + skb_dst_drop(skb); + skb_dst_set(skb, dst); /* * RFC1584 teaches, that DVMRP/PIM router must deliver packets locally @@ -1722,7 +1721,7 @@ int ip6mr_get_route(struct net *net, { int err; struct mfc6_cache *cache; - struct rt6_info *rt = (struct rt6_info *)skb->dst; + struct rt6_info *rt = (struct rt6_info *)skb_dst(skb); read_lock(&mrt_lock); cache = ip6mr_cache_find(net, &rt->rt6i_src.addr, &rt->rt6i_dst.addr); diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 4b48819a5b8..4b264ed40a8 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1448,6 +1448,7 @@ static void mld_sendpack(struct sk_buff *skb) struct net *net = dev_net(skb->dev); int err; struct flowi fl; + struct dst_entry *dst; IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len); @@ -1459,9 +1460,9 @@ static void mld_sendpack(struct sk_buff *skb) IPPROTO_ICMPV6, csum_partial(skb_transport_header(skb), mldlen, 0)); - skb->dst = icmp6_dst_alloc(skb->dev, NULL, &ipv6_hdr(skb)->daddr); + dst = icmp6_dst_alloc(skb->dev, NULL, &ipv6_hdr(skb)->daddr); - if (!skb->dst) { + if (!dst) { err = -ENOMEM; goto err_out; } @@ -1470,7 +1471,8 @@ static void mld_sendpack(struct sk_buff *skb) &ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->dev->ifindex); - err = xfrm_lookup(net, &skb->dst, &fl, NULL, 0); + err = xfrm_lookup(net, &dst, &fl, NULL, 0); + skb_dst_set(skb, dst); if (err) goto err_out; @@ -1775,6 +1777,7 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) IPV6_TLV_ROUTERALERT, 2, 0, 0, IPV6_TLV_PADN, 0 }; struct flowi fl; + struct dst_entry *dst; if (type == ICMPV6_MGM_REDUCTION) snd_addr = &in6addr_linklocal_allrouters; @@ -1828,8 +1831,8 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) idev = in6_dev_get(skb->dev); - skb->dst = icmp6_dst_alloc(skb->dev, NULL, &ipv6_hdr(skb)->daddr); - if (!skb->dst) { + dst = icmp6_dst_alloc(skb->dev, NULL, &ipv6_hdr(skb)->daddr); + if (!dst) { err = -ENOMEM; goto err_out; } @@ -1838,11 +1841,11 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) &ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->dev->ifindex); - err = xfrm_lookup(net, &skb->dst, &fl, NULL, 0); + err = xfrm_lookup(net, &dst, &fl, NULL, 0); if (err) goto err_out; - + skb_dst_set(skb, dst); err = NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, skb->dev, dst_output); out: diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 1d13d996498..9eb68e92cc1 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -530,7 +530,7 @@ void ndisc_send_skb(struct sk_buff *skb, return; } - skb->dst = dst; + skb_dst_set(skb, dst); idev = in6_dev_get(dst->dev); IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len); @@ -1612,7 +1612,7 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, len, IPPROTO_ICMPV6, csum_partial(icmph, len, 0)); - buff->dst = dst; + skb_dst_set(buff, dst); idev = in6_dev_get(dst->dev); IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len); err = NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, buff, NULL, dst->dev, diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 834cea69fb5..d5ed92b1434 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -12,7 +12,7 @@ int ip6_route_me_harder(struct sk_buff *skb) { - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); struct ipv6hdr *iph = ipv6_hdr(skb); struct dst_entry *dst; struct flowi fl = { @@ -28,9 +28,15 @@ int ip6_route_me_harder(struct sk_buff *skb) #ifdef CONFIG_XFRM if (!(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) && - xfrm_decode_session(skb, &fl, AF_INET6) == 0) - if (xfrm_lookup(net, &skb->dst, &fl, skb->sk, 0)) + xfrm_decode_session(skb, &fl, AF_INET6) == 0) { + struct dst_entry *dst2 = skb_dst(skb); + + if (xfrm_lookup(net, &dst2, &fl, skb->sk, 0)) { + skb_dst_set(skb, NULL); return -1; + } + skb_dst_set(skb, dst2); + } #endif if (dst->error) { @@ -41,9 +47,9 @@ int ip6_route_me_harder(struct sk_buff *skb) } /* Drop old route. */ - dst_release(skb->dst); + skb_dst_drop(skb); - skb->dst = dst; + skb_dst_set(skb, dst); return 0; } EXPORT_SYMBOL(ip6_route_me_harder); diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c index 5a2d0a41694..5a7f00cd15c 100644 --- a/net/ipv6/netfilter/ip6t_REJECT.c +++ b/net/ipv6/netfilter/ip6t_REJECT.c @@ -112,7 +112,7 @@ static void send_reset(struct net *net, struct sk_buff *oldskb) return; } - nskb->dst = dst; + skb_dst_set(nskb, dst); skb_reserve(nskb, hh_len + dst->header_len); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index e99307fba0b..36a090d87a3 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -625,7 +625,7 @@ static int rawv6_send_hdrinc(struct sock *sk, void *from, int length, skb->priority = sk->sk_priority; skb->mark = sk->sk_mark; - skb->dst = dst_clone(&rt->u.dst); + skb_dst_set(skb, dst_clone(&rt->u.dst)); skb_put(skb, length); skb_reset_network_header(skb); diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index e9ac7a12f59..54a387d31e1 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -267,7 +267,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, struct sk_buff *prev, *next; struct net_device *dev; int offset, end; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); if (fq->q.last_in & INET_FRAG_COMPLETE) goto err; @@ -277,7 +277,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, ((u8 *)(fhdr + 1) - (u8 *)(ipv6_hdr(skb) + 1))); if ((unsigned int)end > IPV6_MAXPLEN) { - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, ((u8 *)&fhdr->frag_off - @@ -310,7 +310,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, /* RFC2460 says always send parameter problem in * this case. -DaveM */ - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, offsetof(struct ipv6hdr, payload_len)); @@ -434,7 +434,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, return -1; err: - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_REASMFAILS); kfree_skb(skb); return -1; @@ -576,9 +576,9 @@ static int ipv6_frag_rcv(struct sk_buff *skb) struct frag_hdr *fhdr; struct frag_queue *fq; struct ipv6hdr *hdr = ipv6_hdr(skb); - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMREQDS); + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_REASMREQDS); /* Jumbo payload inhibits frag. header */ if (hdr->payload_len==0) @@ -595,17 +595,17 @@ static int ipv6_frag_rcv(struct sk_buff *skb) /* It is not a fragmented frame */ skb->transport_header += sizeof(struct frag_hdr); IP6_INC_STATS_BH(net, - ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMOKS); + ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_REASMOKS); IP6CB(skb)->nhoff = (u8 *)fhdr - skb_network_header(skb); return 1; } if (atomic_read(&net->ipv6.frags.mem) > net->ipv6.frags.high_thresh) - ip6_evictor(net, ip6_dst_idev(skb->dst)); + ip6_evictor(net, ip6_dst_idev(skb_dst(skb))); if ((fq = fq_find(net, fhdr->identification, &hdr->saddr, &hdr->daddr, - ip6_dst_idev(skb->dst))) != NULL) { + ip6_dst_idev(skb_dst(skb)))) != NULL) { int ret; spin_lock(&fq->q.lock); @@ -617,12 +617,12 @@ static int ipv6_frag_rcv(struct sk_buff *skb) return ret; } - IP6_INC_STATS_BH(net, ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMFAILS); + IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_REASMFAILS); kfree_skb(skb); return -1; fail_hdr: - IP6_INC_STATS(net, ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, skb_network_header_len(skb)); return -1; } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 032a5ec391c..658293ea05b 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -800,7 +800,7 @@ void ip6_route_input(struct sk_buff *skb) if (rt6_need_strict(&iph->daddr) && skb->dev->type != ARPHRD_PIMREG) flags |= RT6_LOOKUP_F_IFACE; - skb->dst = fib6_rule_lookup(net, &fl, flags, ip6_pol_route_input); + skb_dst_set(skb, fib6_rule_lookup(net, &fl, flags, ip6_pol_route_input)); } static struct rt6_info *ip6_pol_route_output(struct net *net, struct fib6_table *table, @@ -911,7 +911,7 @@ static void ip6_link_failure(struct sk_buff *skb) icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0, skb->dev); - rt = (struct rt6_info *) skb->dst; + rt = (struct rt6_info *) skb_dst(skb); if (rt) { if (rt->rt6i_flags&RTF_CACHE) { dst_set_expires(&rt->u.dst, 0); @@ -1868,7 +1868,7 @@ int ipv6_route_ioctl(struct net *net, unsigned int cmd, void __user *arg) static int ip6_pkt_drop(struct sk_buff *skb, int code, int ipstats_mib_noroutes) { int type; - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); switch (ipstats_mib_noroutes) { case IPSTATS_MIB_INNOROUTES: type = ipv6_addr_type(&ipv6_hdr(skb)->daddr); @@ -1895,7 +1895,7 @@ static int ip6_pkt_discard(struct sk_buff *skb) static int ip6_pkt_discard_out(struct sk_buff *skb) { - skb->dev = skb->dst->dev; + skb->dev = skb_dst(skb)->dev; return ip6_pkt_drop(skb, ICMPV6_NOROUTE, IPSTATS_MIB_OUTNOROUTES); } @@ -1908,7 +1908,7 @@ static int ip6_pkt_prohibit(struct sk_buff *skb) static int ip6_pkt_prohibit_out(struct sk_buff *skb) { - skb->dev = skb->dst->dev; + skb->dev = skb_dst(skb)->dev; return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED, IPSTATS_MIB_OUTNOROUTES); } @@ -2366,7 +2366,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr)); rt = (struct rt6_info*) ip6_route_output(net, NULL, &fl); - skb->dst = &rt->u.dst; + skb_dst_set(skb, &rt->u.dst); err = rt6_fill_node(net, skb, rt, &fl.fl6_dst, &fl.fl6_src, iif, RTM_NEWROUTE, NETLINK_CB(in_skb).pid, diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index b3a59bd40f0..68e52308e55 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -575,8 +575,7 @@ static int ipip6_rcv(struct sk_buff *skb) tunnel->dev->stats.rx_packets++; tunnel->dev->stats.rx_bytes += skb->len; skb->dev = tunnel->dev; - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); nf_reset(skb); ipip6_ecn_decapsulate(iph, skb); netif_rx(skb); @@ -638,8 +637,8 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (dev->priv_flags & IFF_ISATAP) { struct neighbour *neigh = NULL; - if (skb->dst) - neigh = skb->dst->neighbour; + if (skb_dst(skb)) + neigh = skb_dst(skb)->neighbour; if (neigh == NULL) { if (net_ratelimit()) @@ -663,8 +662,8 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (!dst) { struct neighbour *neigh = NULL; - if (skb->dst) - neigh = skb->dst->neighbour; + if (skb_dst(skb)) + neigh = skb_dst(skb)->neighbour; if (neigh == NULL) { if (net_ratelimit()) @@ -714,7 +713,7 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (tiph->frag_off) mtu = dst_mtu(&rt->u.dst) - sizeof(struct iphdr); else - mtu = skb->dst ? dst_mtu(skb->dst) : dev->mtu; + mtu = skb_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu; if (mtu < 68) { stats->collisions++; @@ -723,8 +722,8 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) } if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; - if (tunnel->parms.iph.daddr && skb->dst) - skb->dst->ops->update_pmtu(skb->dst, mtu); + if (tunnel->parms.iph.daddr && skb_dst(skb)) + skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); if (skb->len > mtu) { icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, dev); @@ -768,8 +767,8 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) skb_reset_network_header(skb); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); IPCB(skb)->flags = 0; - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* * Push down and install the IPIP header. diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index ea37741062a..53b6a4192b1 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -981,9 +981,10 @@ static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win, struct tcphdr *th = tcp_hdr(skb), *t1; struct sk_buff *buff; struct flowi fl; - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); struct sock *ctl_sk = net->ipv6.tcp_sk; unsigned int tot_len = sizeof(struct tcphdr); + struct dst_entry *dst; __be32 *topt; if (ts) @@ -1052,8 +1053,9 @@ static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win, * Underlying function will use this to retrieve the network * namespace */ - if (!ip6_dst_lookup(ctl_sk, &buff->dst, &fl)) { - if (xfrm_lookup(net, &buff->dst, &fl, NULL, 0) >= 0) { + if (!ip6_dst_lookup(ctl_sk, &dst, &fl)) { + if (xfrm_lookup(net, &dst, &fl, NULL, 0) >= 0) { + skb_dst_set(buff, dst); ip6_xmit(ctl_sk, buff, &fl, NULL, 0); TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS); if (rst) diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 8905712cfbb..fc333d85472 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -177,10 +177,9 @@ static struct sock *__udp6_lib_lookup_skb(struct sk_buff *skb, if (unlikely(sk = skb_steal_sock(skb))) return sk; - else - return __udp6_lib_lookup(dev_net(skb->dst->dev), &iph->saddr, sport, - &iph->daddr, dport, inet6_iif(skb), - udptable); + return __udp6_lib_lookup(dev_net(skb_dst(skb)->dev), &iph->saddr, sport, + &iph->daddr, dport, inet6_iif(skb), + udptable); } /* diff --git a/net/ipv6/xfrm6_mode_tunnel.c b/net/ipv6/xfrm6_mode_tunnel.c index e20529b4c82..3927832227b 100644 --- a/net/ipv6/xfrm6_mode_tunnel.c +++ b/net/ipv6/xfrm6_mode_tunnel.c @@ -31,7 +31,7 @@ static inline void ipip6_ecn_decapsulate(struct sk_buff *skb) */ static int xfrm6_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct ipv6hdr *top_iph; int dsfield; @@ -45,7 +45,7 @@ static int xfrm6_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) memcpy(top_iph->flow_lbl, XFRM_MODE_SKB_CB(skb)->flow_lbl, sizeof(top_iph->flow_lbl)); - top_iph->nexthdr = xfrm_af2proto(skb->dst->ops->family); + top_iph->nexthdr = xfrm_af2proto(skb_dst(skb)->ops->family); dsfield = XFRM_MODE_SKB_CB(skb)->tos; dsfield = INET_ECN_encapsulate(dsfield, dsfield); diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c index 5ee5a031bc9..c4f4eef032a 100644 --- a/net/ipv6/xfrm6_output.c +++ b/net/ipv6/xfrm6_output.c @@ -30,7 +30,7 @@ EXPORT_SYMBOL(xfrm6_find_1stfragopt); static int xfrm6_tunnel_check_size(struct sk_buff *skb) { int mtu, ret = 0; - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); mtu = dst_mtu(dst); if (mtu < IPV6_MIN_MTU) @@ -90,6 +90,6 @@ static int xfrm6_output_finish(struct sk_buff *skb) int xfrm6_output(struct sk_buff *skb) { - return NF_HOOK(PF_INET6, NF_INET_POST_ROUTING, skb, NULL, skb->dst->dev, + return NF_HOOK(PF_INET6, NF_INET_POST_ROUTING, skb, NULL, skb_dst(skb)->dev, xfrm6_output_finish); } diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 425ab144f15..5874657af7f 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -260,8 +260,8 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_send_check(ip_hdr(skb)); /* drop old route */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; @@ -324,8 +324,8 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, } /* drop old route */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; @@ -388,8 +388,8 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, goto tx_error_put; /* drop old route */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* mangle the packet */ if (pp->dnat_handler && !pp->dnat_handler(skb, pp, cp)) @@ -465,8 +465,8 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, goto tx_error_put; /* drop old route */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* mangle the packet */ if (pp->dnat_handler && !pp->dnat_handler(skb, pp, cp)) @@ -553,8 +553,8 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, IP_VS_DBG_RL("ip_vs_tunnel_xmit(): mtu less than 68\n"); goto tx_error; } - if (skb->dst) - skb->dst->ops->update_pmtu(skb->dst, mtu); + if (skb_dst(skb)) + skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); df |= (old_iph->frag_off & htons(IP_DF)); @@ -596,8 +596,8 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); /* drop old route */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* * Push down and install the IPIP header. @@ -665,8 +665,8 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, IP_VS_DBG_RL("ip_vs_tunnel_xmit_v6(): mtu less than 1280\n"); goto tx_error; } - if (skb->dst) - skb->dst->ops->update_pmtu(skb->dst, mtu); + if (skb_dst(skb)) + skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); if (mtu < ntohs(old_iph->payload_len) + sizeof(struct ipv6hdr)) { icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); @@ -702,8 +702,8 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); /* drop old route */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* * Push down and install the IPIP header. @@ -775,8 +775,8 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_send_check(ip_hdr(skb)); /* drop old route */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; @@ -828,8 +828,8 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, } /* drop old route */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; @@ -900,8 +900,8 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, goto tx_error_put; /* drop the old route when skb is not shared */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); ip_vs_nat_icmp(skb, pp, cp, 0); @@ -975,8 +975,8 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, goto tx_error_put; /* drop the old route when skb is not shared */ - dst_release(skb->dst); - skb->dst = &rt->u.dst; + skb_dst_drop(skb); + skb_dst_set(skb, &rt->u.dst); ip_vs_nat_icmp_v6(skb, pp, cp, 0); diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index 117b80112fc..a6d6ec320fb 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -176,7 +176,7 @@ static bool gre_invert_tuple(struct nf_conntrack_tuple *tuple, static bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, struct nf_conntrack_tuple *tuple) { - struct net *net = dev_net(skb->dev ? skb->dev : skb->dst->dev); + struct net *net = dev_net(skb->dev ? skb->dev : skb_dst(skb)->dev); const struct gre_hdr_pptp *pgrehdr; struct gre_hdr_pptp _pgrehdr; __be16 srckey; diff --git a/net/netfilter/xt_TCPMSS.c b/net/netfilter/xt_TCPMSS.c index 4f3b1f80879..eda64c1cb1e 100644 --- a/net/netfilter/xt_TCPMSS.c +++ b/net/netfilter/xt_TCPMSS.c @@ -73,11 +73,11 @@ tcpmss_mangle_packet(struct sk_buff *skb, } if (info->mss == XT_TCPMSS_CLAMP_PMTU) { - if (dst_mtu(skb->dst) <= minlen) { + if (dst_mtu(skb_dst(skb)) <= minlen) { if (net_ratelimit()) printk(KERN_ERR "xt_TCPMSS: " "unknown or invalid path-MTU (%u)\n", - dst_mtu(skb->dst)); + dst_mtu(skb_dst(skb))); return -1; } if (in_mtu <= minlen) { @@ -86,7 +86,7 @@ tcpmss_mangle_packet(struct sk_buff *skb, "invalid path-MTU (%u)\n", in_mtu); return -1; } - newmss = min(dst_mtu(skb->dst), in_mtu) - minlen; + newmss = min(dst_mtu(skb_dst(skb)), in_mtu) - minlen; } else newmss = info->mss; diff --git a/net/netfilter/xt_policy.c b/net/netfilter/xt_policy.c index 328bd20ddd2..4cbfebda8fa 100644 --- a/net/netfilter/xt_policy.c +++ b/net/netfilter/xt_policy.c @@ -86,7 +86,7 @@ match_policy_out(const struct sk_buff *skb, const struct xt_policy_info *info, unsigned short family) { const struct xt_policy_elem *e; - const struct dst_entry *dst = skb->dst; + const struct dst_entry *dst = skb_dst(skb); int strict = info->flags & XT_POLICY_MATCH_STRICT; int i, pos; diff --git a/net/netfilter/xt_realm.c b/net/netfilter/xt_realm.c index 67419287bc7..484d1689bfd 100644 --- a/net/netfilter/xt_realm.c +++ b/net/netfilter/xt_realm.c @@ -25,7 +25,7 @@ static bool realm_mt(const struct sk_buff *skb, const struct xt_match_param *par) { const struct xt_realm_info *info = par->matchinfo; - const struct dst_entry *dst = skb->dst; + const struct dst_entry *dst = skb_dst(skb); return (info->id == (dst->tclassid & info->mask)) ^ info->invert; } diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 6da9f38ef5c..4f76e5552d8 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -372,8 +372,7 @@ static int packet_rcv_spkt(struct sk_buff *skb, struct net_device *dev, struct goto oom; /* drop any routing info */ - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); /* drop conntrack reference */ nf_reset(skb); @@ -621,8 +620,7 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet skb_set_owner_r(skb, sk); skb->dev = NULL; - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); /* drop conntrack reference */ nf_reset(skb); diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c index 0ef4e3065bc..9402a7fd378 100644 --- a/net/sched/cls_flow.c +++ b/net/sched/cls_flow.c @@ -84,7 +84,7 @@ static u32 flow_get_dst(const struct sk_buff *skb) case htons(ETH_P_IPV6): return ntohl(ipv6_hdr(skb)->daddr.s6_addr32[3]); default: - return addr_fold(skb->dst) ^ (__force u16)skb->protocol; + return addr_fold(skb_dst(skb)) ^ (__force u16)skb->protocol; } } @@ -163,7 +163,7 @@ static u32 flow_get_proto_dst(const struct sk_buff *skb) break; } default: - res = addr_fold(skb->dst) ^ (__force u16)skb->protocol; + res = addr_fold(skb_dst(skb)) ^ (__force u16)skb->protocol; } return res; @@ -251,8 +251,8 @@ fallback: static u32 flow_get_rtclassid(const struct sk_buff *skb) { #ifdef CONFIG_NET_CLS_ROUTE - if (skb->dst) - return skb->dst->tclassid; + if (skb_dst(skb)) + return skb_dst(skb)->tclassid; #endif return 0; } diff --git a/net/sched/cls_route.c b/net/sched/cls_route.c index bdf1f4172ee..dd872d5383e 100644 --- a/net/sched/cls_route.c +++ b/net/sched/cls_route.c @@ -137,7 +137,7 @@ static int route4_classify(struct sk_buff *skb, struct tcf_proto *tp, u32 id, h; int iif, dont_cache = 0; - if ((dst = skb->dst) == NULL) + if ((dst = skb_dst(skb)) == NULL) goto failure; id = dst->tclassid; diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index b6b588bed4e..266151ae85a 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -246,11 +246,11 @@ META_COLLECTOR(int_tcindex) META_COLLECTOR(int_rtclassid) { - if (unlikely(skb->dst == NULL)) + if (unlikely(skb_dst(skb) == NULL)) *err = -1; else #ifdef CONFIG_NET_CLS_ROUTE - dst->value = skb->dst->tclassid; + dst->value = skb_dst(skb)->tclassid; #else dst->value = 0; #endif diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 33133d27b53..8706920a6d4 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -149,7 +149,7 @@ static unsigned sfq_hash(struct sfq_sched_data *q, struct sk_buff *skb) break; } default: - h = (unsigned long)skb->dst ^ skb->protocol; + h = (unsigned long)skb_dst(skb) ^ skb->protocol; h2 = (unsigned long)skb->sk; } diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index a886496bdc3..cb1cb1e76b9 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -222,7 +222,7 @@ __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device * { struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, 0); struct teql_sched_data *q = qdisc_priv(dev_queue->qdisc); - struct neighbour *mn = skb->dst->neighbour; + struct neighbour *mn = skb_dst(skb)->neighbour; struct neighbour *n = q->ncache; if (mn->tbl == NULL) @@ -262,8 +262,8 @@ static inline int teql_resolve(struct sk_buff *skb, return -ENODEV; if (dev->header_ops == NULL || - skb->dst == NULL || - skb->dst->neighbour == NULL) + skb_dst(skb) == NULL || + skb_dst(skb)->neighbour == NULL) return 0; return __teql_resolve(skb, skb_res, dev); } diff --git a/net/sctp/output.c b/net/sctp/output.c index f0c91df59d4..b7641144451 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -405,10 +405,10 @@ int sctp_packet_transmit(struct sctp_packet *packet) sctp_assoc_sync_pmtu(asoc); } } - nskb->dst = dst_clone(tp->dst); - if (!nskb->dst) + dst = dst_clone(tp->dst); + skb_dst_set(nskb, dst); + if (dst) goto no_route; - dst = nskb->dst; /* Build the SCTP header. */ sh = (struct sctphdr *)skb_push(nskb, sizeof(struct sctphdr)); diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index e1859614601..6c2d6158655 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -918,7 +918,7 @@ static void xs_udp_data_ready(struct sock *sk, int len) UDPX_INC_STATS_BH(sk, UDP_MIB_INDATAGRAMS); /* Something worked... */ - dst_confirm(skb->dst); + dst_confirm(skb_dst(skb)); xprt_adjust_cwnd(task, copied); xprt_update_rtt(task); diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index b4a13178fb4..e0009c17d80 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -251,8 +251,7 @@ resume: nf_reset(skb); if (decaps) { - dst_release(skb->dst); - skb->dst = NULL; + skb_dst_drop(skb); netif_rx(skb); return 0; } else { diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c index c235597ba8d..b9fe13138c0 100644 --- a/net/xfrm/xfrm_output.c +++ b/net/xfrm/xfrm_output.c @@ -22,7 +22,7 @@ static int xfrm_output2(struct sk_buff *skb); static int xfrm_state_check_space(struct xfrm_state *x, struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); int nhead = dst->header_len + LL_RESERVED_SPACE(dst->dev) - skb_headroom(skb); int ntail = dst->dev->needed_tailroom - skb_tailroom(skb); @@ -39,7 +39,7 @@ static int xfrm_state_check_space(struct xfrm_state *x, struct sk_buff *skb) static int xfrm_output_one(struct sk_buff *skb, int err) { - struct dst_entry *dst = skb->dst; + struct dst_entry *dst = skb_dst(skb); struct xfrm_state *x = dst->xfrm; struct net *net = xs_net(x); @@ -94,12 +94,13 @@ resume: goto error_nolock; } - if (!(skb->dst = dst_pop(dst))) { + dst = dst_pop(dst); + if (!dst) { XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTERROR); err = -EHOSTUNREACH; goto error_nolock; } - dst = skb->dst; + skb_dst_set(skb, dst); x = dst->xfrm; } while (x && !(x->outer_mode->flags & XFRM_MODE_FLAG_TUNNEL)); @@ -119,16 +120,16 @@ int xfrm_output_resume(struct sk_buff *skb, int err) while (likely((err = xfrm_output_one(skb, err)) == 0)) { nf_reset(skb); - err = skb->dst->ops->local_out(skb); + err = skb_dst(skb)->ops->local_out(skb); if (unlikely(err != 1)) goto out; - if (!skb->dst->xfrm) + if (!skb_dst(skb)->xfrm) return dst_output(skb); - err = nf_hook(skb->dst->ops->family, + err = nf_hook(skb_dst(skb)->ops->family, NF_INET_POST_ROUTING, skb, - NULL, skb->dst->dev, xfrm_output2); + NULL, skb_dst(skb)->dev, xfrm_output2); if (unlikely(err != 1)) goto out; } @@ -179,7 +180,7 @@ static int xfrm_output_gso(struct sk_buff *skb) int xfrm_output(struct sk_buff *skb) { - struct net *net = dev_net(skb->dst->dev); + struct net *net = dev_net(skb_dst(skb)->dev); int err; if (skb_is_gso(skb)) @@ -202,7 +203,7 @@ int xfrm_inner_extract_output(struct xfrm_state *x, struct sk_buff *skb) struct xfrm_mode *inner_mode; if (x->sel.family == AF_UNSPEC) inner_mode = xfrm_ip2inner_mode(x, - xfrm_af2proto(skb->dst->ops->family)); + xfrm_af2proto(skb_dst(skb)->ops->family)); else inner_mode = x->inner_mode; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 9c068ab3a83..cb81ca35b0d 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -2027,6 +2027,8 @@ int __xfrm_route_forward(struct sk_buff *skb, unsigned short family) { struct net *net = dev_net(skb->dev); struct flowi fl; + struct dst_entry *dst; + int res; if (xfrm_decode_session(skb, &fl, family) < 0) { /* XXX: we should have something like FWDHDRERROR here. */ @@ -2034,7 +2036,11 @@ int __xfrm_route_forward(struct sk_buff *skb, unsigned short family) return 0; } - return xfrm_lookup(net, &skb->dst, &fl, NULL, 0) == 0; + dst = skb_dst(skb); + + res = xfrm_lookup(net, &dst, &fl, NULL, 0) == 0; + skb_dst_set(skb, dst); + return res; } EXPORT_SYMBOL(__xfrm_route_forward); diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 2fcad7c33ea..4bfc6153ad4 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -4503,7 +4503,7 @@ static unsigned int selinux_ip_postroute(struct sk_buff *skb, int ifindex, * when the packet is on it's final way out. * NOTE: there appear to be some IPv6 multicast cases where skb->dst * is NULL, in this case go ahead and apply access control. */ - if (skb->dst != NULL && skb->dst->xfrm != NULL) + if (skb_dst(skb) != NULL && skb_dst(skb)->xfrm != NULL) return NF_ACCEPT; #endif secmark_active = selinux_secmark_enabled(); diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c index c0eb72013d6..72b18452e1a 100644 --- a/security/selinux/xfrm.c +++ b/security/selinux/xfrm.c @@ -447,7 +447,7 @@ int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, struct dst_entry *dst; int rc = 0; - dst = skb->dst; + dst = skb_dst(skb); if (dst) { struct dst_entry *dst_test; -- cgit v1.2.3-70-g09d2 From 3b8bcfd5d31ea0fec58681d035544ace707d2536 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sat, 30 May 2009 01:39:53 +0200 Subject: net: introduce pre-up netdev notifier NETDEV_UP is called after the device is set UP, but sometimes it is useful to be able to veto the device UP. Introduce a new NETDEV_PRE_UP notifier that can be used for exactly this. The first use case will be cfg80211 denying interfaces to be set UP if the device is known to be rfkill'ed. Signed-off-by: Johannes Berg Acked-by: David S. Miller Signed-off-by: John W. Linville --- include/linux/notifier.h | 1 + net/core/dev.c | 7 ++++++- 2 files changed, 7 insertions(+), 1 deletion(-) (limited to 'net/core/dev.c') diff --git a/include/linux/notifier.h b/include/linux/notifier.h index b86fa2ffca0..81bc252dc8a 100644 --- a/include/linux/notifier.h +++ b/include/linux/notifier.h @@ -198,6 +198,7 @@ static inline int notifier_to_errno(int ret) #define NETDEV_CHANGENAME 0x000A #define NETDEV_FEAT_CHANGE 0x000B #define NETDEV_BONDING_FAILOVER 0x000C +#define NETDEV_PRE_UP 0x000D #define SYS_DOWN 0x0001 /* Notify of system down */ #define SYS_RESTART SYS_DOWN diff --git a/net/core/dev.c b/net/core/dev.c index 34b49a6a22f..1f38401fc02 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1048,7 +1048,7 @@ void dev_load(struct net *net, const char *name) int dev_open(struct net_device *dev) { const struct net_device_ops *ops = dev->netdev_ops; - int ret = 0; + int ret; ASSERT_RTNL(); @@ -1065,6 +1065,11 @@ int dev_open(struct net_device *dev) if (!netif_device_present(dev)) return -ENODEV; + ret = call_netdevice_notifiers(NETDEV_PRE_UP, dev); + ret = notifier_to_errno(ret); + if (ret) + return ret; + /* * Call device private open method */ -- cgit v1.2.3-70-g09d2 From 4cf704fbea96075942bd033fd75aa4e76ae1c8a1 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 9 Jun 2009 00:18:51 -0700 Subject: net/core/dev.c: Use frag list abstraction interfaces. Signed-off-by: David S. Miller --- net/core/dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 1f38401fc02..4913089c91d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1820,7 +1820,7 @@ int dev_queue_xmit(struct sk_buff *skb) if (netif_needs_gso(dev, skb)) goto gso; - if (skb_shinfo(skb)->frag_list && + if (skb_has_frags(skb) && !(dev->features & NETIF_F_FRAGLIST) && __skb_linearize(skb)) goto out_kfree_skb; @@ -2407,7 +2407,7 @@ int dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb) if (!(skb->dev->features & NETIF_F_GRO)) goto normal; - if (skb_is_gso(skb) || skb_shinfo(skb)->frag_list) + if (skb_is_gso(skb) || skb_has_frags(skb)) goto normal; rcu_read_lock(); -- cgit v1.2.3-70-g09d2 From 0c27922e4933ceb86644f4a9b1af212ffe5aad75 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 8 Jun 2009 03:49:24 +0000 Subject: net: dev_addr_init() fix commit f001fde5eadd915f4858d22ed70d7040f48767cf (net: introduce a list of device addresses dev_addr_list (v6)) added one regression Vegard Nossum found in its testings. With kmemcheck help, Vegard found some uninitialized memory was read and reported to user, potentialy leaking kernel data. ( thread can be found on http://lkml.org/lkml/2009/5/30/177 ) dev_addr_init() incorrectly uses sizeof() operator. We were initializing one byte instead of MAX_ADDR_LEN bytes. Reported-by: Vegard Nossum Signed-off-by: Eric Dumazet Acked-by: Jiri Pirko Signed-off-by: David S. Miller --- net/core/dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 4913089c91d..81b392ef511 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3655,8 +3655,8 @@ static int dev_addr_init(struct net_device *dev) /* rtnl_mutex must be held here */ INIT_LIST_HEAD(&dev->dev_addr_list); - memset(addr, 0, sizeof(*addr)); - err = __hw_addr_add(&dev->dev_addr_list, NULL, addr, sizeof(*addr), + memset(addr, 0, sizeof(addr)); + err = __hw_addr_add(&dev->dev_addr_list, NULL, addr, sizeof(addr), NETDEV_HW_ADDR_T_LAN); if (!err) { /* -- cgit v1.2.3-70-g09d2 From fcb94e422479da52ed90bab230c59617a0462416 Mon Sep 17 00:00:00 2001 From: Sergey Lapin Date: Mon, 8 Jun 2009 12:18:47 +0000 Subject: Add constants for the ieee 802.15.4 stack IEEE 802.15.4 stack requires several constants to be defined/adjusted. Signed-off-by: Dmitry Eremin-Solenikov Signed-off-by: Sergey Lapin Signed-off-by: David S. Miller --- include/linux/if_arp.h | 2 ++ include/linux/if_ether.h | 1 + include/linux/socket.h | 4 +++- net/core/dev.c | 6 ++++-- net/core/sock.c | 3 +++ 5 files changed, 13 insertions(+), 3 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/if_arp.h b/include/linux/if_arp.h index 5ff89809a58..b554300ef8b 100644 --- a/include/linux/if_arp.h +++ b/include/linux/if_arp.h @@ -86,6 +86,8 @@ #define ARPHRD_IEEE80211 801 /* IEEE 802.11 */ #define ARPHRD_IEEE80211_PRISM 802 /* IEEE 802.11 + Prism2 header */ #define ARPHRD_IEEE80211_RADIOTAP 803 /* IEEE 802.11 + radiotap header */ +#define ARPHRD_IEEE802154 804 +#define ARPHRD_IEEE802154_PHY 805 #define ARPHRD_PHONET 820 /* PhoNet media type */ #define ARPHRD_PHONET_PIPE 821 /* PhoNet pipe header */ diff --git a/include/linux/if_ether.h b/include/linux/if_ether.h index cfe4fe1b713..11a60e4f0a6 100644 --- a/include/linux/if_ether.h +++ b/include/linux/if_ether.h @@ -106,6 +106,7 @@ #define ETH_P_DSA 0x001B /* Distributed Switch Arch. */ #define ETH_P_TRAILER 0x001C /* Trailer switch tagging */ #define ETH_P_PHONET 0x00F5 /* Nokia Phonet frames */ +#define ETH_P_IEEE802154 0x00F6 /* IEEE802.15.4 frame */ /* * This is an Ethernet frame header. diff --git a/include/linux/socket.h b/include/linux/socket.h index d2310cb45d2..3b461dffe24 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -194,7 +194,8 @@ struct ucred { #define AF_RXRPC 33 /* RxRPC sockets */ #define AF_ISDN 34 /* mISDN sockets */ #define AF_PHONET 35 /* Phonet sockets */ -#define AF_MAX 36 /* For now.. */ +#define AF_IEEE802154 36 /* IEEE802154 sockets */ +#define AF_MAX 37 /* For now.. */ /* Protocol families, same as address families. */ #define PF_UNSPEC AF_UNSPEC @@ -233,6 +234,7 @@ struct ucred { #define PF_RXRPC AF_RXRPC #define PF_ISDN AF_ISDN #define PF_PHONET AF_PHONET +#define PF_IEEE802154 AF_IEEE802154 #define PF_MAX AF_MAX /* Maximum queue length specifiable by listen. */ diff --git a/net/core/dev.c b/net/core/dev.c index 81b392ef511..11560e3258b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -269,7 +269,8 @@ static const unsigned short netdev_lock_type[] = ARPHRD_IRDA, ARPHRD_FCPP, ARPHRD_FCAL, ARPHRD_FCPL, ARPHRD_FCFABRIC, ARPHRD_IEEE802_TR, ARPHRD_IEEE80211, ARPHRD_IEEE80211_PRISM, ARPHRD_IEEE80211_RADIOTAP, ARPHRD_PHONET, - ARPHRD_PHONET_PIPE, ARPHRD_VOID, ARPHRD_NONE}; + ARPHRD_PHONET_PIPE, ARPHRD_IEEE802154, ARPHRD_IEEE802154_PHY, + ARPHRD_VOID, ARPHRD_NONE}; static const char *netdev_lock_name[] = {"_xmit_NETROM", "_xmit_ETHER", "_xmit_EETHER", "_xmit_AX25", @@ -286,7 +287,8 @@ static const char *netdev_lock_name[] = "_xmit_IRDA", "_xmit_FCPP", "_xmit_FCAL", "_xmit_FCPL", "_xmit_FCFABRIC", "_xmit_IEEE802_TR", "_xmit_IEEE80211", "_xmit_IEEE80211_PRISM", "_xmit_IEEE80211_RADIOTAP", "_xmit_PHONET", - "_xmit_PHONET_PIPE", "_xmit_VOID", "_xmit_NONE"}; + "_xmit_PHONET_PIPE", "_xmit_IEEE802154", "_xmit_IEEE802154_PHY", + "_xmit_VOID", "_xmit_NONE"}; static struct lock_class_key netdev_xmit_lock_key[ARRAY_SIZE(netdev_lock_type)]; static struct lock_class_key netdev_addr_lock_key[ARRAY_SIZE(netdev_lock_type)]; diff --git a/net/core/sock.c b/net/core/sock.c index 58dec9dff99..04e35eb2e73 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -155,6 +155,7 @@ static const char *af_family_key_strings[AF_MAX+1] = { "sk_lock-27" , "sk_lock-28" , "sk_lock-AF_CAN" , "sk_lock-AF_TIPC" , "sk_lock-AF_BLUETOOTH", "sk_lock-IUCV" , "sk_lock-AF_RXRPC" , "sk_lock-AF_ISDN" , "sk_lock-AF_PHONET" , + "sk_lock-AF_IEEE802154", "sk_lock-AF_MAX" }; static const char *af_family_slock_key_strings[AF_MAX+1] = { @@ -170,6 +171,7 @@ static const char *af_family_slock_key_strings[AF_MAX+1] = { "slock-27" , "slock-28" , "slock-AF_CAN" , "slock-AF_TIPC" , "slock-AF_BLUETOOTH", "slock-AF_IUCV" , "slock-AF_RXRPC" , "slock-AF_ISDN" , "slock-AF_PHONET" , + "slock-AF_IEEE802154", "slock-AF_MAX" }; static const char *af_family_clock_key_strings[AF_MAX+1] = { @@ -185,6 +187,7 @@ static const char *af_family_clock_key_strings[AF_MAX+1] = { "clock-27" , "clock-28" , "clock-AF_CAN" , "clock-AF_TIPC" , "clock-AF_BLUETOOTH", "clock-AF_IUCV" , "clock-AF_RXRPC" , "clock-AF_ISDN" , "clock-AF_PHONET" , + "clock-AF_IEEE802154", "clock-AF_MAX" }; -- cgit v1.2.3-70-g09d2 From 746e6ad23cd6fec2edce056e014a0eabeffa838c Mon Sep 17 00:00:00 2001 From: John Dykstra Date: Thu, 11 Jun 2009 20:57:21 -0700 Subject: [PATCH] net core: Some interface flags not returned by SIOCGIFFLAGS Commit b00055aacdb172c05067612278ba27265fcd05ce " [NET] core: add RFC2863 operstate" defined new interface flag values. Its documentation specified that these flags could be accessed from user space via SIOCGIFFLAGS. However, this does not work because the new flags do not fit in that ioctl's argument width. Change the documentation to match the code's behavior. Also change the source to explicitly show the truncation. This _should_ have no effect on executable code, and did not with gcc 4.2.4 generating x86 code. A new ioctl could be defined to return all interface flags to user space. However, since this has been broken for three years with no one complaining, there doesn't seem much need. They are still accessible via netlink. Reported-by: "Fredrik Arnerup" Signed-off-by: John Dykstra Signed-off-by: David S. Miller --- Documentation/networking/operstates.txt | 3 --- net/core/dev.c | 2 +- 2 files changed, 1 insertion(+), 4 deletions(-) (limited to 'net/core/dev.c') diff --git a/Documentation/networking/operstates.txt b/Documentation/networking/operstates.txt index c9074f9b78b..1a77a3cfae5 100644 --- a/Documentation/networking/operstates.txt +++ b/Documentation/networking/operstates.txt @@ -38,9 +38,6 @@ ifinfomsg::if_flags & IFF_LOWER_UP: ifinfomsg::if_flags & IFF_DORMANT: Driver has signaled netif_dormant_on() -These interface flags can also be queried without netlink using the -SIOCGIFFLAGS ioctl. - TLV IFLA_OPERSTATE contains RFC2863 state of the interface in numeric representation: diff --git a/net/core/dev.c b/net/core/dev.c index 11560e3258b..a09bf658970 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4209,7 +4209,7 @@ static int dev_ifsioc_locked(struct net *net, struct ifreq *ifr, unsigned int cm switch (cmd) { case SIOCGIFFLAGS: /* Get interface flags */ - ifr->ifr_flags = dev_get_flags(dev); + ifr->ifr_flags = (short) dev_get_flags(dev); return 0; case SIOCGIFMETRIC: /* Get the metric on the interface -- cgit v1.2.3-70-g09d2 From da6782927de809d9d427bd4bd6a4024243e41f13 Mon Sep 17 00:00:00 2001 From: Michał Mirosław Date: Fri, 5 Jun 2009 05:35:28 +0000 Subject: bridge: Simplify interface for ATM LANE MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch changes FDB entry check for ATM LANE bridge integration. There's no point in holding a FDB entry around SKB building. br_fdb_get()/br_fdb_put() pair are changed into single br_fdb_test_addr() hook that checks if the addr has FDB entry pointing to other port to the one the request arrived on. FDB entry refcounting is removed as it's not used anywhere else. Signed-off-by: Michał Mirosław Acked-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/atm/lec.c | 17 ++++------------- net/bridge/br.c | 10 ++++++---- net/bridge/br_fdb.c | 43 +++++++++++++++++++++---------------------- net/bridge/br_private.h | 12 ++++-------- net/core/dev.c | 20 +++++++++----------- 5 files changed, 44 insertions(+), 58 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/atm/lec.c b/net/atm/lec.c index 199b6bb79f4..75b9d59553f 100644 --- a/net/atm/lec.c +++ b/net/atm/lec.c @@ -34,7 +34,6 @@ /* Proxy LEC knows about bridging */ #if defined(CONFIG_BRIDGE) || defined(CONFIG_BRIDGE_MODULE) -#include #include "../bridge/br_private.h" static unsigned char bridge_ula_lec[] = { 0x01, 0x80, 0xc2, 0x00, 0x00 }; @@ -518,18 +517,14 @@ static int lec_atm_send(struct atm_vcc *vcc, struct sk_buff *skb) case l_should_bridge: #if defined(CONFIG_BRIDGE) || defined(CONFIG_BRIDGE_MODULE) { - struct net_bridge_fdb_entry *f; - pr_debug("%s: bridge zeppelin asks about %pM\n", dev->name, mesg->content.proxy.mac_addr); - if (br_fdb_get_hook == NULL || dev->br_port == NULL) + if (br_fdb_test_addr_hook == NULL) break; - f = br_fdb_get_hook(dev->br_port->br, - mesg->content.proxy.mac_addr); - if (f != NULL && f->dst->dev != dev - && f->dst->state == BR_STATE_FORWARDING) { + if (br_fdb_test_addr_hook(dev, + mesg->content.proxy.mac_addr)) { /* hit from bridge table, send LE_ARP_RESPONSE */ struct sk_buff *skb2; struct sock *sk; @@ -540,10 +535,8 @@ static int lec_atm_send(struct atm_vcc *vcc, struct sk_buff *skb) skb2 = alloc_skb(sizeof(struct atmlec_msg), GFP_ATOMIC); - if (skb2 == NULL) { - br_fdb_put_hook(f); + if (skb2 == NULL) break; - } skb2->len = sizeof(struct atmlec_msg); skb_copy_to_linear_data(skb2, mesg, sizeof(*mesg)); @@ -552,8 +545,6 @@ static int lec_atm_send(struct atm_vcc *vcc, struct sk_buff *skb) skb_queue_tail(&sk->sk_receive_queue, skb2); sk->sk_data_ready(sk, skb2->len); } - if (f != NULL) - br_fdb_put_hook(f); } #endif /* defined(CONFIG_BRIDGE) || defined(CONFIG_BRIDGE_MODULE) */ break; diff --git a/net/bridge/br.c b/net/bridge/br.c index 4d2c1f1cb52..9aac5213105 100644 --- a/net/bridge/br.c +++ b/net/bridge/br.c @@ -65,8 +65,9 @@ static int __init br_init(void) brioctl_set(br_ioctl_deviceless_stub); br_handle_frame_hook = br_handle_frame; - br_fdb_get_hook = br_fdb_get; - br_fdb_put_hook = br_fdb_put; +#if defined(CONFIG_ATM_LANE) || defined(CONFIG_ATM_LANE_MODULE) + br_fdb_test_addr_hook = br_fdb_test_addr; +#endif return 0; err_out4: @@ -95,8 +96,9 @@ static void __exit br_deinit(void) synchronize_net(); br_netfilter_fini(); - br_fdb_get_hook = NULL; - br_fdb_put_hook = NULL; +#if defined(CONFIG_ATM_LANE) || defined(CONFIG_ATM_LANE_MODULE) + br_fdb_test_addr_hook = NULL; +#endif br_handle_frame_hook = NULL; br_fdb_fini(); diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index cb3e97b93ae..57bf05c353b 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -71,10 +71,17 @@ static inline int br_mac_hash(const unsigned char *mac) return jhash_1word(key, fdb_salt) & (BR_HASH_SIZE - 1); } +static void fdb_rcu_free(struct rcu_head *head) +{ + struct net_bridge_fdb_entry *ent + = container_of(head, struct net_bridge_fdb_entry, rcu); + kmem_cache_free(br_fdb_cache, ent); +} + static inline void fdb_delete(struct net_bridge_fdb_entry *f) { hlist_del_rcu(&f->hlist); - br_fdb_put(f); + call_rcu(&f->rcu, fdb_rcu_free); } void br_fdb_changeaddr(struct net_bridge_port *p, const unsigned char *newaddr) @@ -226,33 +233,26 @@ struct net_bridge_fdb_entry *__br_fdb_get(struct net_bridge *br, return NULL; } -/* Interface used by ATM hook that keeps a ref count */ -struct net_bridge_fdb_entry *br_fdb_get(struct net_bridge *br, - unsigned char *addr) +#if defined(CONFIG_ATM_LANE) || defined(CONFIG_ATM_LANE_MODULE) +/* Interface used by ATM LANE hook to test + * if an addr is on some other bridge port */ +int br_fdb_test_addr(struct net_device *dev, unsigned char *addr) { struct net_bridge_fdb_entry *fdb; + int ret; + + if (!dev->br_port) + return 0; rcu_read_lock(); - fdb = __br_fdb_get(br, addr); - if (fdb && !atomic_inc_not_zero(&fdb->use_count)) - fdb = NULL; + fdb = __br_fdb_get(dev->br_port->br, addr); + ret = fdb && fdb->dst->dev != dev && + fdb->dst->state == BR_STATE_FORWARDING; rcu_read_unlock(); - return fdb; -} - -static void fdb_rcu_free(struct rcu_head *head) -{ - struct net_bridge_fdb_entry *ent - = container_of(head, struct net_bridge_fdb_entry, rcu); - kmem_cache_free(br_fdb_cache, ent); -} -/* Set entry up for deletion with RCU */ -void br_fdb_put(struct net_bridge_fdb_entry *ent) -{ - if (atomic_dec_and_test(&ent->use_count)) - call_rcu(&ent->rcu, fdb_rcu_free); + return ret; } +#endif /* CONFIG_ATM_LANE */ /* * Fill buffer with forwarding table records in @@ -326,7 +326,6 @@ static struct net_bridge_fdb_entry *fdb_create(struct hlist_head *head, fdb = kmem_cache_alloc(br_fdb_cache, GFP_ATOMIC); if (fdb) { memcpy(fdb->addr.addr, addr, ETH_ALEN); - atomic_set(&fdb->use_count, 1); hlist_add_head_rcu(&fdb->hlist, head); fdb->dst = source; diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index b6c3b71974d..d5b5537272b 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -51,7 +51,6 @@ struct net_bridge_fdb_entry struct net_bridge_port *dst; struct rcu_head rcu; - atomic_t use_count; unsigned long ageing_timer; mac_addr addr; unsigned char is_local; @@ -154,9 +153,7 @@ extern void br_fdb_delete_by_port(struct net_bridge *br, const struct net_bridge_port *p, int do_all); extern struct net_bridge_fdb_entry *__br_fdb_get(struct net_bridge *br, const unsigned char *addr); -extern struct net_bridge_fdb_entry *br_fdb_get(struct net_bridge *br, - unsigned char *addr); -extern void br_fdb_put(struct net_bridge_fdb_entry *ent); +extern int br_fdb_test_addr(struct net_device *dev, unsigned char *addr); extern int br_fdb_fillbuf(struct net_bridge *br, void *buf, unsigned long count, unsigned long off); extern int br_fdb_insert(struct net_bridge *br, @@ -242,10 +239,9 @@ extern void br_stp_port_timer_init(struct net_bridge_port *p); extern unsigned long br_timer_value(const struct timer_list *timer); /* br.c */ -extern struct net_bridge_fdb_entry *(*br_fdb_get_hook)(struct net_bridge *br, - unsigned char *addr); -extern void (*br_fdb_put_hook)(struct net_bridge_fdb_entry *ent); - +#if defined(CONFIG_ATM_LANE) || defined(CONFIG_ATM_LANE_MODULE) +extern int (*br_fdb_test_addr_hook)(struct net_device *dev, unsigned char *addr); +#endif /* br_netlink.c */ extern int br_netlink_init(void); diff --git a/net/core/dev.c b/net/core/dev.c index a09bf658970..ea00e36f48e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2071,11 +2071,13 @@ static inline int deliver_skb(struct sk_buff *skb, } #if defined(CONFIG_BRIDGE) || defined (CONFIG_BRIDGE_MODULE) -/* These hooks defined here for ATM */ -struct net_bridge; -struct net_bridge_fdb_entry *(*br_fdb_get_hook)(struct net_bridge *br, - unsigned char *addr); -void (*br_fdb_put_hook)(struct net_bridge_fdb_entry *ent) __read_mostly; + +#if defined(CONFIG_ATM_LANE) || defined(CONFIG_ATM_LANE_MODULE) +/* This hook is defined here for ATM LANE */ +int (*br_fdb_test_addr_hook)(struct net_device *dev, + unsigned char *addr) __read_mostly; +EXPORT_SYMBOL(br_fdb_test_addr_hook); +#endif /* * If bridge module is loaded call bridging hook. @@ -2083,6 +2085,8 @@ void (*br_fdb_put_hook)(struct net_bridge_fdb_entry *ent) __read_mostly; */ struct sk_buff *(*br_handle_frame_hook)(struct net_bridge_port *p, struct sk_buff *skb) __read_mostly; +EXPORT_SYMBOL(br_handle_frame_hook); + static inline struct sk_buff *handle_bridge(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, struct net_device *orig_dev) @@ -5665,12 +5669,6 @@ EXPORT_SYMBOL(net_enable_timestamp); EXPORT_SYMBOL(net_disable_timestamp); EXPORT_SYMBOL(dev_get_flags); -#if defined(CONFIG_BRIDGE) || defined(CONFIG_BRIDGE_MODULE) -EXPORT_SYMBOL(br_handle_frame_hook); -EXPORT_SYMBOL(br_fdb_get_hook); -EXPORT_SYMBOL(br_fdb_put_hook); -#endif - EXPORT_SYMBOL(dev_load); EXPORT_PER_CPU_SYMBOL(softnet_data); -- cgit v1.2.3-70-g09d2 From 31278e71471399beaff9280737e52b47db4dc345 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 17 Jun 2009 01:12:19 +0000 Subject: net: group address list and its count This patch is inspired by patch recently posted by Johannes Berg. Basically what my patch does is to group list and a count of addresses into newly introduced structure netdev_hw_addr_list. This brings us two benefits: 1) struct net_device becames a bit nicer. 2) in the future there will be a possibility to operate with lists independently on netdevices (with exporting right functions). I wanted to introduce this patch before I'll post a multicast lists conversion. Signed-off-by: Jiri Pirko drivers/net/bnx2.c | 4 +- drivers/net/e1000/e1000_main.c | 4 +- drivers/net/ixgbe/ixgbe_main.c | 6 +- drivers/net/mv643xx_eth.c | 2 +- drivers/net/niu.c | 4 +- drivers/net/virtio_net.c | 10 ++-- drivers/s390/net/qeth_l2_main.c | 2 +- include/linux/netdevice.h | 17 +++-- net/core/dev.c | 130 ++++++++++++++++++-------------------- 9 files changed, 89 insertions(+), 90 deletions(-) Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 4 +- drivers/net/e1000/e1000_main.c | 4 +- drivers/net/ixgbe/ixgbe_main.c | 6 +- drivers/net/mv643xx_eth.c | 2 +- drivers/net/niu.c | 4 +- drivers/net/virtio_net.c | 10 ++-- drivers/s390/net/qeth_l2_main.c | 2 +- include/linux/netdevice.h | 17 ++++-- net/core/dev.c | 130 +++++++++++++++++++--------------------- 9 files changed, 89 insertions(+), 90 deletions(-) (limited to 'net/core/dev.c') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 7e3738112c4..38f1c3375d7 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -3552,14 +3552,14 @@ bnx2_set_rx_mode(struct net_device *dev) sort_mode |= BNX2_RPM_SORT_USER0_MC_HSH_EN; } - if (dev->uc_count > BNX2_MAX_UNICAST_ADDRESSES) { + if (dev->uc.count > BNX2_MAX_UNICAST_ADDRESSES) { rx_mode |= BNX2_EMAC_RX_MODE_PROMISCUOUS; sort_mode |= BNX2_RPM_SORT_USER0_PROM_EN | BNX2_RPM_SORT_USER0_PROM_VLAN; } else if (!(dev->flags & IFF_PROMISC)) { /* Add all entries into to the match filter list */ i = 0; - list_for_each_entry(ha, &dev->uc_list, list) { + list_for_each_entry(ha, &dev->uc.list, list) { bnx2_set_mac_addr(bp, ha->addr, i + BNX2_START_UNICAST_ADDRESS_INDEX); sort_mode |= (1 << diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 8d36743c814..5e3356f8eb5 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -2370,7 +2370,7 @@ static void e1000_set_rx_mode(struct net_device *netdev) rctl |= E1000_RCTL_VFE; } - if (netdev->uc_count > rar_entries - 1) { + if (netdev->uc.count > rar_entries - 1) { rctl |= E1000_RCTL_UPE; } else if (!(netdev->flags & IFF_PROMISC)) { rctl &= ~E1000_RCTL_UPE; @@ -2394,7 +2394,7 @@ static void e1000_set_rx_mode(struct net_device *netdev) */ i = 1; if (use_uc) - list_for_each_entry(ha, &netdev->uc_list, list) { + list_for_each_entry(ha, &netdev->uc.list, list) { if (i == rar_entries) break; e1000_rar_set(hw, ha->addr, i++); diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c index a551a96ce67..e756e220db3 100644 --- a/drivers/net/ixgbe/ixgbe_main.c +++ b/drivers/net/ixgbe/ixgbe_main.c @@ -2321,7 +2321,7 @@ static void ixgbe_set_rx_mode(struct net_device *netdev) IXGBE_WRITE_REG(hw, IXGBE_VLNCTRL, vlnctrl); /* reprogram secondary unicast list */ - hw->mac.ops.update_uc_addr_list(hw, &netdev->uc_list); + hw->mac.ops.update_uc_addr_list(hw, &netdev->uc.list); /* reprogram multicast list */ addr_count = netdev->mc_count; @@ -5261,7 +5261,7 @@ static int ixgbe_ioctl(struct net_device *netdev, struct ifreq *req, int cmd) /** * ixgbe_add_sanmac_netdev - Add the SAN MAC address to the corresponding - * netdev->dev_addr_list + * netdev->dev_addrs * @netdev: network interface device structure * * Returns non-zero on failure @@ -5282,7 +5282,7 @@ static int ixgbe_add_sanmac_netdev(struct net_device *dev) /** * ixgbe_del_sanmac_netdev - Removes the SAN MAC address to the corresponding - * netdev->dev_addr_list + * netdev->dev_addrs * @netdev: network interface device structure * * Returns non-zero on failure diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index b4e18a58cb1..745ae8b4a2e 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -1729,7 +1729,7 @@ static u32 uc_addr_filter_mask(struct net_device *dev) return 0; nibbles = 1 << (dev->dev_addr[5] & 0x0f); - list_for_each_entry(ha, &dev->uc_list, list) { + list_for_each_entry(ha, &dev->uc.list, list) { if (memcmp(dev->dev_addr, ha->addr, 5)) return 0; if ((dev->dev_addr[5] ^ ha->addr[5]) & 0xf0) diff --git a/drivers/net/niu.c b/drivers/net/niu.c index fa61a12c5e1..d2146d4a10f 100644 --- a/drivers/net/niu.c +++ b/drivers/net/niu.c @@ -6376,7 +6376,7 @@ static void niu_set_rx_mode(struct net_device *dev) if ((dev->flags & IFF_ALLMULTI) || (dev->mc_count > 0)) np->flags |= NIU_FLAGS_MCAST; - alt_cnt = dev->uc_count; + alt_cnt = dev->uc.count; if (alt_cnt > niu_num_alt_addr(np)) { alt_cnt = 0; np->flags |= NIU_FLAGS_PROMISC; @@ -6385,7 +6385,7 @@ static void niu_set_rx_mode(struct net_device *dev) if (alt_cnt) { int index = 0; - list_for_each_entry(ha, &dev->uc_list, list) { + list_for_each_entry(ha, &dev->uc.list, list) { err = niu_set_alt_mac(np, index, ha->addr); if (err) printk(KERN_WARNING PFX "%s: Error %d " diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 52198f6797a..2a6e81d5b57 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -709,7 +709,7 @@ static void virtnet_set_rx_mode(struct net_device *dev) allmulti ? "en" : "dis"); /* MAC filter - use one buffer for both lists */ - mac_data = buf = kzalloc(((dev->uc_count + dev->mc_count) * ETH_ALEN) + + mac_data = buf = kzalloc(((dev->uc.count + dev->mc_count) * ETH_ALEN) + (2 * sizeof(mac_data->entries)), GFP_ATOMIC); if (!buf) { dev_warn(&dev->dev, "No memory for MAC address buffer\n"); @@ -719,16 +719,16 @@ static void virtnet_set_rx_mode(struct net_device *dev) sg_init_table(sg, 2); /* Store the unicast list and count in the front of the buffer */ - mac_data->entries = dev->uc_count; + mac_data->entries = dev->uc.count; i = 0; - list_for_each_entry(ha, &dev->uc_list, list) + list_for_each_entry(ha, &dev->uc.list, list) memcpy(&mac_data->macs[i++][0], ha->addr, ETH_ALEN); sg_set_buf(&sg[0], mac_data, - sizeof(mac_data->entries) + (dev->uc_count * ETH_ALEN)); + sizeof(mac_data->entries) + (dev->uc.count * ETH_ALEN)); /* multicast list and count fill the end */ - mac_data = (void *)&mac_data->macs[dev->uc_count][0]; + mac_data = (void *)&mac_data->macs[dev->uc.count][0]; mac_data->entries = dev->mc_count; addr = dev->mc_list; diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index ecd3d06c0d5..3607d107f49 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -655,7 +655,7 @@ static void qeth_l2_set_multicast_list(struct net_device *dev) for (dm = dev->mc_list; dm; dm = dm->next) qeth_l2_add_mc(card, dm->da_addr, 0); - list_for_each_entry(ha, &dev->uc_list, list) + list_for_each_entry(ha, &dev->uc.list, list) qeth_l2_add_mc(card, ha->addr, 1); spin_unlock_bh(&card->mclock); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 9ea8d6dfe54..d4a4d986779 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -224,6 +224,11 @@ struct netdev_hw_addr { struct rcu_head rcu_head; }; +struct netdev_hw_addr_list { + struct list_head list; + int count; +}; + struct hh_cache { struct hh_cache *hh_next; /* Next entry */ @@ -776,9 +781,8 @@ struct net_device unsigned char addr_len; /* hardware address length */ unsigned short dev_id; /* for shared network cards */ - struct list_head uc_list; /* Secondary unicast mac - addresses */ - int uc_count; /* Number of installed ucasts */ + struct netdev_hw_addr_list uc; /* Secondary unicast + mac addresses */ int uc_promisc; spinlock_t addr_list_lock; struct dev_addr_list *mc_list; /* Multicast mac addresses */ @@ -810,7 +814,8 @@ struct net_device because most packets are unicast) */ - struct list_head dev_addr_list; /* list of device hw addresses */ + struct netdev_hw_addr_list dev_addrs; /* list of device + hw addresses */ unsigned char broadcast[MAX_ADDR_LEN]; /* hw bcast add */ @@ -1806,11 +1811,11 @@ static inline void netif_addr_unlock_bh(struct net_device *dev) } /* - * dev_addr_list walker. Should be used only for read access. Call with + * dev_addrs walker. Should be used only for read access. Call with * rcu_read_lock held. */ #define for_each_dev_addr(dev, ha) \ - list_for_each_entry_rcu(ha, &dev->dev_addr_list, list) + list_for_each_entry_rcu(ha, &dev->dev_addrs.list, list) /* These functions live elsewhere (drivers/net/net_init.c, but related) */ diff --git a/net/core/dev.c b/net/core/dev.c index 576a61574a9..baf2dc13a34 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3461,10 +3461,10 @@ void __dev_set_rx_mode(struct net_device *dev) /* Unicast addresses changes may only happen under the rtnl, * therefore calling __dev_set_promiscuity here is safe. */ - if (dev->uc_count > 0 && !dev->uc_promisc) { + if (dev->uc.count > 0 && !dev->uc_promisc) { __dev_set_promiscuity(dev, 1); dev->uc_promisc = 1; - } else if (dev->uc_count == 0 && dev->uc_promisc) { + } else if (dev->uc.count == 0 && dev->uc_promisc) { __dev_set_promiscuity(dev, -1); dev->uc_promisc = 0; } @@ -3483,9 +3483,8 @@ void dev_set_rx_mode(struct net_device *dev) /* hw addresses list handling functions */ -static int __hw_addr_add(struct list_head *list, int *delta, - unsigned char *addr, int addr_len, - unsigned char addr_type) +static int __hw_addr_add(struct netdev_hw_addr_list *list, unsigned char *addr, + int addr_len, unsigned char addr_type) { struct netdev_hw_addr *ha; int alloc_size; @@ -3493,7 +3492,7 @@ static int __hw_addr_add(struct list_head *list, int *delta, if (addr_len > MAX_ADDR_LEN) return -EINVAL; - list_for_each_entry(ha, list, list) { + list_for_each_entry(ha, &list->list, list) { if (!memcmp(ha->addr, addr, addr_len) && ha->type == addr_type) { ha->refcount++; @@ -3512,9 +3511,8 @@ static int __hw_addr_add(struct list_head *list, int *delta, ha->type = addr_type; ha->refcount = 1; ha->synced = false; - list_add_tail_rcu(&ha->list, list); - if (delta) - (*delta)++; + list_add_tail_rcu(&ha->list, &list->list); + list->count++; return 0; } @@ -3526,120 +3524,121 @@ static void ha_rcu_free(struct rcu_head *head) kfree(ha); } -static int __hw_addr_del(struct list_head *list, int *delta, - unsigned char *addr, int addr_len, - unsigned char addr_type) +static int __hw_addr_del(struct netdev_hw_addr_list *list, unsigned char *addr, + int addr_len, unsigned char addr_type) { struct netdev_hw_addr *ha; - list_for_each_entry(ha, list, list) { + list_for_each_entry(ha, &list->list, list) { if (!memcmp(ha->addr, addr, addr_len) && (ha->type == addr_type || !addr_type)) { if (--ha->refcount) return 0; list_del_rcu(&ha->list); call_rcu(&ha->rcu_head, ha_rcu_free); - if (delta) - (*delta)--; + list->count--; return 0; } } return -ENOENT; } -static int __hw_addr_add_multiple(struct list_head *to_list, int *to_delta, - struct list_head *from_list, int addr_len, +static int __hw_addr_add_multiple(struct netdev_hw_addr_list *to_list, + struct netdev_hw_addr_list *from_list, + int addr_len, unsigned char addr_type) { int err; struct netdev_hw_addr *ha, *ha2; unsigned char type; - list_for_each_entry(ha, from_list, list) { + list_for_each_entry(ha, &from_list->list, list) { type = addr_type ? addr_type : ha->type; - err = __hw_addr_add(to_list, to_delta, ha->addr, - addr_len, type); + err = __hw_addr_add(to_list, ha->addr, addr_len, type); if (err) goto unroll; } return 0; unroll: - list_for_each_entry(ha2, from_list, list) { + list_for_each_entry(ha2, &from_list->list, list) { if (ha2 == ha) break; type = addr_type ? addr_type : ha2->type; - __hw_addr_del(to_list, to_delta, ha2->addr, - addr_len, type); + __hw_addr_del(to_list, ha2->addr, addr_len, type); } return err; } -static void __hw_addr_del_multiple(struct list_head *to_list, int *to_delta, - struct list_head *from_list, int addr_len, +static void __hw_addr_del_multiple(struct netdev_hw_addr_list *to_list, + struct netdev_hw_addr_list *from_list, + int addr_len, unsigned char addr_type) { struct netdev_hw_addr *ha; unsigned char type; - list_for_each_entry(ha, from_list, list) { + list_for_each_entry(ha, &from_list->list, list) { type = addr_type ? addr_type : ha->type; - __hw_addr_del(to_list, to_delta, ha->addr, - addr_len, addr_type); + __hw_addr_del(to_list, ha->addr, addr_len, addr_type); } } -static int __hw_addr_sync(struct list_head *to_list, int *to_delta, - struct list_head *from_list, int *from_delta, +static int __hw_addr_sync(struct netdev_hw_addr_list *to_list, + struct netdev_hw_addr_list *from_list, int addr_len) { int err = 0; struct netdev_hw_addr *ha, *tmp; - list_for_each_entry_safe(ha, tmp, from_list, list) { + list_for_each_entry_safe(ha, tmp, &from_list->list, list) { if (!ha->synced) { - err = __hw_addr_add(to_list, to_delta, ha->addr, + err = __hw_addr_add(to_list, ha->addr, addr_len, ha->type); if (err) break; ha->synced = true; ha->refcount++; } else if (ha->refcount == 1) { - __hw_addr_del(to_list, to_delta, ha->addr, - addr_len, ha->type); - __hw_addr_del(from_list, from_delta, ha->addr, - addr_len, ha->type); + __hw_addr_del(to_list, ha->addr, addr_len, ha->type); + __hw_addr_del(from_list, ha->addr, addr_len, ha->type); } } return err; } -static void __hw_addr_unsync(struct list_head *to_list, int *to_delta, - struct list_head *from_list, int *from_delta, +static void __hw_addr_unsync(struct netdev_hw_addr_list *to_list, + struct netdev_hw_addr_list *from_list, int addr_len) { struct netdev_hw_addr *ha, *tmp; - list_for_each_entry_safe(ha, tmp, from_list, list) { + list_for_each_entry_safe(ha, tmp, &from_list->list, list) { if (ha->synced) { - __hw_addr_del(to_list, to_delta, ha->addr, + __hw_addr_del(to_list, ha->addr, addr_len, ha->type); ha->synced = false; - __hw_addr_del(from_list, from_delta, ha->addr, + __hw_addr_del(from_list, ha->addr, addr_len, ha->type); } } } - -static void __hw_addr_flush(struct list_head *list) +static void __hw_addr_flush(struct netdev_hw_addr_list *list) { struct netdev_hw_addr *ha, *tmp; - list_for_each_entry_safe(ha, tmp, list, list) { + list_for_each_entry_safe(ha, tmp, &list->list, list) { list_del_rcu(&ha->list); call_rcu(&ha->rcu_head, ha_rcu_free); } + list->count = 0; +} + +static void __hw_addr_init(struct netdev_hw_addr_list *list) +{ + INIT_LIST_HEAD(&list->list); + list->count = 0; } /* Device addresses handling functions */ @@ -3648,7 +3647,7 @@ static void dev_addr_flush(struct net_device *dev) { /* rtnl_mutex must be held here */ - __hw_addr_flush(&dev->dev_addr_list); + __hw_addr_flush(&dev->dev_addrs); dev->dev_addr = NULL; } @@ -3660,16 +3659,16 @@ static int dev_addr_init(struct net_device *dev) /* rtnl_mutex must be held here */ - INIT_LIST_HEAD(&dev->dev_addr_list); + __hw_addr_init(&dev->dev_addrs); memset(addr, 0, sizeof(addr)); - err = __hw_addr_add(&dev->dev_addr_list, NULL, addr, sizeof(addr), + err = __hw_addr_add(&dev->dev_addrs, addr, sizeof(addr), NETDEV_HW_ADDR_T_LAN); if (!err) { /* * Get the first (previously created) address from the list * and set dev_addr pointer to this location. */ - ha = list_first_entry(&dev->dev_addr_list, + ha = list_first_entry(&dev->dev_addrs.list, struct netdev_hw_addr, list); dev->dev_addr = ha->addr; } @@ -3694,8 +3693,7 @@ int dev_addr_add(struct net_device *dev, unsigned char *addr, ASSERT_RTNL(); - err = __hw_addr_add(&dev->dev_addr_list, NULL, addr, dev->addr_len, - addr_type); + err = __hw_addr_add(&dev->dev_addrs, addr, dev->addr_len, addr_type); if (!err) call_netdevice_notifiers(NETDEV_CHANGEADDR, dev); return err; @@ -3725,11 +3723,12 @@ int dev_addr_del(struct net_device *dev, unsigned char *addr, * We can not remove the first address from the list because * dev->dev_addr points to that. */ - ha = list_first_entry(&dev->dev_addr_list, struct netdev_hw_addr, list); + ha = list_first_entry(&dev->dev_addrs.list, + struct netdev_hw_addr, list); if (ha->addr == dev->dev_addr && ha->refcount == 1) return -ENOENT; - err = __hw_addr_del(&dev->dev_addr_list, NULL, addr, dev->addr_len, + err = __hw_addr_del(&dev->dev_addrs, addr, dev->addr_len, addr_type); if (!err) call_netdevice_notifiers(NETDEV_CHANGEADDR, dev); @@ -3757,8 +3756,7 @@ int dev_addr_add_multiple(struct net_device *to_dev, if (from_dev->addr_len != to_dev->addr_len) return -EINVAL; - err = __hw_addr_add_multiple(&to_dev->dev_addr_list, NULL, - &from_dev->dev_addr_list, + err = __hw_addr_add_multiple(&to_dev->dev_addrs, &from_dev->dev_addrs, to_dev->addr_len, addr_type); if (!err) call_netdevice_notifiers(NETDEV_CHANGEADDR, to_dev); @@ -3784,15 +3782,14 @@ int dev_addr_del_multiple(struct net_device *to_dev, if (from_dev->addr_len != to_dev->addr_len) return -EINVAL; - __hw_addr_del_multiple(&to_dev->dev_addr_list, NULL, - &from_dev->dev_addr_list, + __hw_addr_del_multiple(&to_dev->dev_addrs, &from_dev->dev_addrs, to_dev->addr_len, addr_type); call_netdevice_notifiers(NETDEV_CHANGEADDR, to_dev); return 0; } EXPORT_SYMBOL(dev_addr_del_multiple); -/* unicast and multicast addresses handling functions */ +/* multicast addresses handling functions */ int __dev_addr_delete(struct dev_addr_list **list, int *count, void *addr, int alen, int glbl) @@ -3868,8 +3865,8 @@ int dev_unicast_delete(struct net_device *dev, void *addr) ASSERT_RTNL(); - err = __hw_addr_del(&dev->uc_list, &dev->uc_count, addr, - dev->addr_len, NETDEV_HW_ADDR_T_UNICAST); + err = __hw_addr_del(&dev->uc, addr, dev->addr_len, + NETDEV_HW_ADDR_T_UNICAST); if (!err) __dev_set_rx_mode(dev); return err; @@ -3892,8 +3889,8 @@ int dev_unicast_add(struct net_device *dev, void *addr) ASSERT_RTNL(); - err = __hw_addr_add(&dev->uc_list, &dev->uc_count, addr, - dev->addr_len, NETDEV_HW_ADDR_T_UNICAST); + err = __hw_addr_add(&dev->uc, addr, dev->addr_len, + NETDEV_HW_ADDR_T_UNICAST); if (!err) __dev_set_rx_mode(dev); return err; @@ -3966,8 +3963,7 @@ int dev_unicast_sync(struct net_device *to, struct net_device *from) if (to->addr_len != from->addr_len) return -EINVAL; - err = __hw_addr_sync(&to->uc_list, &to->uc_count, - &from->uc_list, &from->uc_count, to->addr_len); + err = __hw_addr_sync(&to->uc, &from->uc, to->addr_len); if (!err) __dev_set_rx_mode(to); return err; @@ -3990,8 +3986,7 @@ void dev_unicast_unsync(struct net_device *to, struct net_device *from) if (to->addr_len != from->addr_len) return; - __hw_addr_unsync(&to->uc_list, &to->uc_count, - &from->uc_list, &from->uc_count, to->addr_len); + __hw_addr_unsync(&to->uc, &from->uc, to->addr_len); __dev_set_rx_mode(to); } EXPORT_SYMBOL(dev_unicast_unsync); @@ -4000,15 +3995,14 @@ static void dev_unicast_flush(struct net_device *dev) { /* rtnl_mutex must be held here */ - __hw_addr_flush(&dev->uc_list); - dev->uc_count = 0; + __hw_addr_flush(&dev->uc); } static void dev_unicast_init(struct net_device *dev) { /* rtnl_mutex must be held here */ - INIT_LIST_HEAD(&dev->uc_list); + __hw_addr_init(&dev->uc); } -- cgit v1.2.3-70-g09d2