From 0a9627f2649a02bea165cfd529d7bcb625c2fcad Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Tue, 16 Mar 2010 08:03:29 +0000 Subject: rps: Receive Packet Steering This patch implements software receive side packet steering (RPS). RPS distributes the load of received packet processing across multiple CPUs. Problem statement: Protocol processing done in the NAPI context for received packets is serialized per device queue and becomes a bottleneck under high packet load. This substantially limits pps that can be achieved on a single queue NIC and provides no scaling with multiple cores. This solution queues packets early on in the receive path on the backlog queues of other CPUs. This allows protocol processing (e.g. IP and TCP) to be performed on packets in parallel. For each device (or each receive queue in a multi-queue device) a mask of CPUs is set to indicate the CPUs that can process packets. A CPU is selected on a per packet basis by hashing contents of the packet header (e.g. the TCP or UDP 4-tuple) and using the result to index into the CPU mask. The IPI mechanism is used to raise networking receive softirqs between CPUs. This effectively emulates in software what a multi-queue NIC can provide, but is generic requiring no device support. Many devices now provide a hash over the 4-tuple on a per packet basis (e.g. the Toeplitz hash). This patch allow drivers to set the HW reported hash in an skb field, and that value in turn is used to index into the RPS maps. Using the HW generated hash can avoid cache misses on the packet when steering it to a remote CPU. The CPU mask is set on a per device and per queue basis in the sysfs variable /sys/class/net//queues/rx-/rps_cpus. This is a set of canonical bit maps for receive queues in the device (numbered by ). If a device does not support multi-queue, a single variable is used for the device (rx-0). Generally, we have found this technique increases pps capabilities of a single queue device with good CPU utilization. Optimal settings for the CPU mask seem to depend on architectures and cache hierarcy. Below are some results running 500 instances of netperf TCP_RR test with 1 byte req. and resp. Results show cumulative transaction rate and system CPU utilization. e1000e on 8 core Intel Without RPS: 108K tps at 33% CPU With RPS: 311K tps at 64% CPU forcedeth on 16 core AMD Without RPS: 156K tps at 15% CPU With RPS: 404K tps at 49% CPU bnx2x on 16 core AMD Without RPS 567K tps at 61% CPU (4 HW RX queues) Without RPS 738K tps at 96% CPU (8 HW RX queues) With RPS: 854K tps at 76% CPU (4 HW RX queues) Caveats: - The benefits of this patch are dependent on architecture and cache hierarchy. Tuning the masks to get best performance is probably necessary. - This patch adds overhead in the path for processing a single packet. In a lightly loaded server this overhead may eliminate the advantages of increased parallelism, and possibly cause some relative performance degradation. We have found that masks that are cache aware (share same caches with the interrupting CPU) mitigate much of this. - The RPS masks can be changed dynamically, however whenever the mask is changed this introduces the possibility of generating out of order packets. It's probably best not change the masks too frequently. Signed-off-by: Tom Herbert include/linux/netdevice.h | 32 ++++- include/linux/skbuff.h | 3 + net/core/dev.c | 335 +++++++++++++++++++++++++++++++++++++-------- net/core/net-sysfs.c | 225 ++++++++++++++++++++++++++++++- net/core/skbuff.c | 2 + 5 files changed, 538 insertions(+), 59 deletions(-) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 335 +++++++++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 279 insertions(+), 56 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index bcc490cc9452..17b168671501 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1931,7 +1931,7 @@ out_kfree_skb: return rc; } -static u32 skb_tx_hashrnd; +static u32 hashrnd __read_mostly; u16 skb_tx_hash(const struct net_device *dev, const struct sk_buff *skb) { @@ -1949,7 +1949,7 @@ u16 skb_tx_hash(const struct net_device *dev, const struct sk_buff *skb) else hash = skb->protocol; - hash = jhash_1word(hash, skb_tx_hashrnd); + hash = jhash_1word(hash, hashrnd); return (u16) (((u64) hash * dev->real_num_tx_queues) >> 32); } @@ -1959,10 +1959,9 @@ static inline u16 dev_cap_txqueue(struct net_device *dev, u16 queue_index) { if (unlikely(queue_index >= dev->real_num_tx_queues)) { if (net_ratelimit()) { - WARN(1, "%s selects TX queue %d, but " + netdev_warn(dev, "selects TX queue %d, but " "real number of TX queues is %d\n", - dev->name, queue_index, - dev->real_num_tx_queues); + queue_index, dev->real_num_tx_queues); } return 0; } @@ -2175,6 +2174,172 @@ int weight_p __read_mostly = 64; /* old backlog weight */ DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; +/* + * get_rps_cpu is called from netif_receive_skb and returns the target + * CPU from the RPS map of the receiving queue for a given skb. + */ +static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb) +{ + struct ipv6hdr *ip6; + struct iphdr *ip; + struct netdev_rx_queue *rxqueue; + struct rps_map *map; + int cpu = -1; + u8 ip_proto; + u32 addr1, addr2, ports, ihl; + + rcu_read_lock(); + + if (skb_rx_queue_recorded(skb)) { + u16 index = skb_get_rx_queue(skb); + if (unlikely(index >= dev->num_rx_queues)) { + if (net_ratelimit()) { + netdev_warn(dev, "received packet on queue " + "%u, but number of RX queues is %u\n", + index, dev->num_rx_queues); + } + goto done; + } + rxqueue = dev->_rx + index; + } else + rxqueue = dev->_rx; + + if (!rxqueue->rps_map) + goto done; + + if (skb->rxhash) + goto got_hash; /* Skip hash computation on packet header */ + + switch (skb->protocol) { + case __constant_htons(ETH_P_IP): + if (!pskb_may_pull(skb, sizeof(*ip))) + goto done; + + ip = (struct iphdr *) skb->data; + ip_proto = ip->protocol; + addr1 = ip->saddr; + addr2 = ip->daddr; + ihl = ip->ihl; + break; + case __constant_htons(ETH_P_IPV6): + if (!pskb_may_pull(skb, sizeof(*ip6))) + goto done; + + ip6 = (struct ipv6hdr *) skb->data; + ip_proto = ip6->nexthdr; + addr1 = ip6->saddr.s6_addr32[3]; + addr2 = ip6->daddr.s6_addr32[3]; + ihl = (40 >> 2); + break; + default: + goto done; + } + ports = 0; + switch (ip_proto) { + case IPPROTO_TCP: + case IPPROTO_UDP: + case IPPROTO_DCCP: + case IPPROTO_ESP: + case IPPROTO_AH: + case IPPROTO_SCTP: + case IPPROTO_UDPLITE: + if (pskb_may_pull(skb, (ihl * 4) + 4)) + ports = *((u32 *) (skb->data + (ihl * 4))); + break; + + default: + break; + } + + skb->rxhash = jhash_3words(addr1, addr2, ports, hashrnd); + if (!skb->rxhash) + skb->rxhash = 1; + +got_hash: + map = rcu_dereference(rxqueue->rps_map); + if (map) { + u16 tcpu = map->cpus[((u64) skb->rxhash * map->len) >> 32]; + + if (cpu_online(tcpu)) { + cpu = tcpu; + goto done; + } + } + +done: + rcu_read_unlock(); + return cpu; +} + +/* + * This structure holds the per-CPU mask of CPUs for which IPIs are scheduled + * to be sent to kick remote softirq processing. There are two masks since + * the sending of IPIs must be done with interrupts enabled. The select field + * indicates the current mask that enqueue_backlog uses to schedule IPIs. + * select is flipped before net_rps_action is called while still under lock, + * net_rps_action then uses the non-selected mask to send the IPIs and clears + * it without conflicting with enqueue_backlog operation. + */ +struct rps_remote_softirq_cpus { + cpumask_t mask[2]; + int select; +}; +static DEFINE_PER_CPU(struct rps_remote_softirq_cpus, rps_remote_softirq_cpus); + +/* Called from hardirq (IPI) context */ +static void trigger_softirq(void *data) +{ + struct softnet_data *queue = data; + __napi_schedule(&queue->backlog); + __get_cpu_var(netdev_rx_stat).received_rps++; +} + +/* + * enqueue_to_backlog is called to queue an skb to a per CPU backlog + * queue (may be a remote CPU queue). + */ +static int enqueue_to_backlog(struct sk_buff *skb, int cpu) +{ + struct softnet_data *queue; + unsigned long flags; + + queue = &per_cpu(softnet_data, cpu); + + local_irq_save(flags); + __get_cpu_var(netdev_rx_stat).total++; + + spin_lock(&queue->input_pkt_queue.lock); + if (queue->input_pkt_queue.qlen <= netdev_max_backlog) { + if (queue->input_pkt_queue.qlen) { +enqueue: + __skb_queue_tail(&queue->input_pkt_queue, skb); + spin_unlock_irqrestore(&queue->input_pkt_queue.lock, + flags); + return NET_RX_SUCCESS; + } + + /* Schedule NAPI for backlog device */ + if (napi_schedule_prep(&queue->backlog)) { + if (cpu != smp_processor_id()) { + struct rps_remote_softirq_cpus *rcpus = + &__get_cpu_var(rps_remote_softirq_cpus); + + cpu_set(cpu, rcpus->mask[rcpus->select]); + __raise_softirq_irqoff(NET_RX_SOFTIRQ); + } else + __napi_schedule(&queue->backlog); + } + goto enqueue; + } + + spin_unlock(&queue->input_pkt_queue.lock); + + __get_cpu_var(netdev_rx_stat).dropped++; + local_irq_restore(flags); + + kfree_skb(skb); + return NET_RX_DROP; +} /** * netif_rx - post buffer to the network code @@ -2193,8 +2358,7 @@ DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; int netif_rx(struct sk_buff *skb) { - struct softnet_data *queue; - unsigned long flags; + int cpu; /* if netpoll wants it, pretend we never saw it */ if (netpoll_rx(skb)) @@ -2203,31 +2367,11 @@ int netif_rx(struct sk_buff *skb) if (!skb->tstamp.tv64) net_timestamp(skb); - /* - * The code is rearranged so that the path is the most - * short when CPU is congested, but is still operating. - */ - local_irq_save(flags); - queue = &__get_cpu_var(softnet_data); - - __get_cpu_var(netdev_rx_stat).total++; - if (queue->input_pkt_queue.qlen <= netdev_max_backlog) { - if (queue->input_pkt_queue.qlen) { -enqueue: - __skb_queue_tail(&queue->input_pkt_queue, skb); - local_irq_restore(flags); - return NET_RX_SUCCESS; - } - - napi_schedule(&queue->backlog); - goto enqueue; - } - - __get_cpu_var(netdev_rx_stat).dropped++; - local_irq_restore(flags); + cpu = get_rps_cpu(skb->dev, skb); + if (cpu < 0) + cpu = smp_processor_id(); - kfree_skb(skb); - return NET_RX_DROP; + return enqueue_to_backlog(skb, cpu); } EXPORT_SYMBOL(netif_rx); @@ -2464,22 +2608,7 @@ void netif_nit_deliver(struct sk_buff *skb) rcu_read_unlock(); } -/** - * netif_receive_skb - process receive buffer from network - * @skb: buffer to process - * - * netif_receive_skb() is the main receive data processing function. - * It always succeeds. The buffer may be dropped during processing - * for congestion control or by the protocol layers. - * - * This function may only be called from softirq context and interrupts - * should be enabled. - * - * Return values (usually ignored): - * NET_RX_SUCCESS: no congestion - * NET_RX_DROP: packet was dropped - */ -int netif_receive_skb(struct sk_buff *skb) +int __netif_receive_skb(struct sk_buff *skb) { struct packet_type *ptype, *pt_prev; struct net_device *orig_dev; @@ -2588,6 +2717,33 @@ out: rcu_read_unlock(); return ret; } + +/** + * netif_receive_skb - process receive buffer from network + * @skb: buffer to process + * + * netif_receive_skb() is the main receive data processing function. + * It always succeeds. The buffer may be dropped during processing + * for congestion control or by the protocol layers. + * + * This function may only be called from softirq context and interrupts + * should be enabled. + * + * Return values (usually ignored): + * NET_RX_SUCCESS: no congestion + * NET_RX_DROP: packet was dropped + */ +int netif_receive_skb(struct sk_buff *skb) +{ + int cpu; + + cpu = get_rps_cpu(skb->dev, skb); + + if (cpu < 0) + return __netif_receive_skb(skb); + else + return enqueue_to_backlog(skb, cpu); +} EXPORT_SYMBOL(netif_receive_skb); /* Network device is going away, flush any packets still pending */ @@ -2914,16 +3070,16 @@ static int process_backlog(struct napi_struct *napi, int quota) do { struct sk_buff *skb; - local_irq_disable(); + spin_lock_irq(&queue->input_pkt_queue.lock); skb = __skb_dequeue(&queue->input_pkt_queue); if (!skb) { __napi_complete(napi); - local_irq_enable(); + spin_unlock_irq(&queue->input_pkt_queue.lock); break; } - local_irq_enable(); + spin_unlock_irq(&queue->input_pkt_queue.lock); - netif_receive_skb(skb); + __netif_receive_skb(skb); } while (++work < quota && jiffies == start_time); return work; @@ -3012,6 +3168,22 @@ void netif_napi_del(struct napi_struct *napi) } EXPORT_SYMBOL(netif_napi_del); +/* + * net_rps_action sends any pending IPI's for rps. This is only called from + * softirq and interrupts must be enabled. + */ +static void net_rps_action(cpumask_t *mask) +{ + int cpu; + + /* Send pending IPI's to kick RPS processing on remote cpus. */ + for_each_cpu_mask_nr(cpu, *mask) { + struct softnet_data *queue = &per_cpu(softnet_data, cpu); + if (cpu_online(cpu)) + __smp_call_function_single(cpu, &queue->csd, 0); + } + cpus_clear(*mask); +} static void net_rx_action(struct softirq_action *h) { @@ -3019,6 +3191,8 @@ static void net_rx_action(struct softirq_action *h) unsigned long time_limit = jiffies + 2; int budget = netdev_budget; void *have; + int select; + struct rps_remote_softirq_cpus *rcpus; local_irq_disable(); @@ -3081,8 +3255,14 @@ static void net_rx_action(struct softirq_action *h) netpoll_poll_unlock(have); } out: + rcpus = &__get_cpu_var(rps_remote_softirq_cpus); + select = rcpus->select; + rcpus->select ^= 1; + local_irq_enable(); + net_rps_action(&rcpus->mask[select]); + #ifdef CONFIG_NET_DMA /* * There may not be any more sk_buffs coming right now, so push @@ -3327,10 +3507,10 @@ static int softnet_seq_show(struct seq_file *seq, void *v) { struct netif_rx_stats *s = v; - seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x %08x\n", + seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x %08x %08x\n", s->total, s->dropped, s->time_squeeze, 0, 0, 0, 0, 0, /* was fastroute */ - s->cpu_collision); + s->cpu_collision, s->received_rps); return 0; } @@ -5067,6 +5247,23 @@ int register_netdevice(struct net_device *dev) dev->iflink = -1; + if (!dev->num_rx_queues) { + /* + * Allocate a single RX queue if driver never called + * alloc_netdev_mq + */ + + dev->_rx = kzalloc(sizeof(struct netdev_rx_queue), GFP_KERNEL); + if (!dev->_rx) { + ret = -ENOMEM; + goto out; + } + + dev->_rx->first = dev->_rx; + atomic_set(&dev->_rx->count, 1); + dev->num_rx_queues = 1; + } + /* Init, if this function is available */ if (dev->netdev_ops->ndo_init) { ret = dev->netdev_ops->ndo_init(dev); @@ -5424,9 +5621,11 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, void (*setup)(struct net_device *), unsigned int queue_count) { struct netdev_queue *tx; + struct netdev_rx_queue *rx; struct net_device *dev; size_t alloc_size; struct net_device *p; + int i; BUG_ON(strlen(name) >= sizeof(dev->name)); @@ -5452,11 +5651,27 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, goto free_p; } + rx = kcalloc(queue_count, sizeof(struct netdev_rx_queue), GFP_KERNEL); + if (!rx) { + printk(KERN_ERR "alloc_netdev: Unable to allocate " + "rx queues.\n"); + goto free_tx; + } + + atomic_set(&rx->count, queue_count); + + /* + * Set a pointer to first element in the array which holds the + * reference count. + */ + for (i = 0; i < queue_count; i++) + rx[i].first = rx; + dev = PTR_ALIGN(p, NETDEV_ALIGN); dev->padded = (char *)dev - (char *)p; if (dev_addr_init(dev)) - goto free_tx; + goto free_rx; dev_unicast_init(dev); @@ -5466,6 +5681,9 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, dev->num_tx_queues = queue_count; dev->real_num_tx_queues = queue_count; + dev->_rx = rx; + dev->num_rx_queues = queue_count; + dev->gso_max_size = GSO_MAX_SIZE; netdev_init_queues(dev); @@ -5480,9 +5698,10 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, strcpy(dev->name, name); return dev; +free_rx: + kfree(rx); free_tx: kfree(tx); - free_p: kfree(p); return NULL; @@ -5985,6 +6204,10 @@ static int __init net_dev_init(void) queue->completion_queue = NULL; INIT_LIST_HEAD(&queue->poll_list); + queue->csd.func = trigger_softirq; + queue->csd.info = queue; + queue->csd.flags = 0; + queue->backlog.poll = process_backlog; queue->backlog.weight = weight_p; queue->backlog.gro_list = NULL; @@ -6023,7 +6246,7 @@ subsys_initcall(net_dev_init); static int __init initialize_hashrnd(void) { - get_random_bytes(&skb_tx_hashrnd, sizeof(skb_tx_hashrnd)); + get_random_bytes(&hashrnd, sizeof(hashrnd)); return 0; } -- cgit v1.2.3 From 1e94d72feab025b8f7c55d07020602f82f3a97dd Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Thu, 18 Mar 2010 17:45:44 -0700 Subject: rps: Fixed build with CONFIG_SMP not enabled. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 ++ net/core/dev.c | 24 ++++++++++++++++++++++++ 2 files changed, 26 insertions(+) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index de1a52bcb9e0..726ecd1af535 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1341,7 +1341,9 @@ struct softnet_data { struct sk_buff *completion_queue; /* Elements below can be accessed between CPUs for RPS */ +#ifdef CONFIG_SMP struct call_single_data csd ____cacheline_aligned_in_smp; +#endif struct sk_buff_head input_pkt_queue; struct napi_struct backlog; }; diff --git a/net/core/dev.c b/net/core/dev.c index 17b168671501..1a7e1d1d5ad9 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2174,6 +2174,7 @@ int weight_p __read_mostly = 64; /* old backlog weight */ DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; +#ifdef CONFIG_SMP /* * get_rps_cpu is called from netif_receive_skb and returns the target * CPU from the RPS map of the receiving queue for a given skb. @@ -2293,6 +2294,7 @@ static void trigger_softirq(void *data) __napi_schedule(&queue->backlog); __get_cpu_var(netdev_rx_stat).received_rps++; } +#endif /* CONFIG_SMP */ /* * enqueue_to_backlog is called to queue an skb to a per CPU backlog @@ -2320,6 +2322,7 @@ enqueue: /* Schedule NAPI for backlog device */ if (napi_schedule_prep(&queue->backlog)) { +#ifdef CONFIG_SMP if (cpu != smp_processor_id()) { struct rps_remote_softirq_cpus *rcpus = &__get_cpu_var(rps_remote_softirq_cpus); @@ -2328,6 +2331,9 @@ enqueue: __raise_softirq_irqoff(NET_RX_SOFTIRQ); } else __napi_schedule(&queue->backlog); +#else + __napi_schedule(&queue->backlog); +#endif } goto enqueue; } @@ -2367,9 +2373,13 @@ int netif_rx(struct sk_buff *skb) if (!skb->tstamp.tv64) net_timestamp(skb); +#ifdef CONFIG_SMP cpu = get_rps_cpu(skb->dev, skb); if (cpu < 0) cpu = smp_processor_id(); +#else + cpu = smp_processor_id(); +#endif return enqueue_to_backlog(skb, cpu); } @@ -2735,6 +2745,7 @@ out: */ int netif_receive_skb(struct sk_buff *skb) { +#ifdef CONFIG_SMP int cpu; cpu = get_rps_cpu(skb->dev, skb); @@ -2743,6 +2754,9 @@ int netif_receive_skb(struct sk_buff *skb) return __netif_receive_skb(skb); else return enqueue_to_backlog(skb, cpu); +#else + return __netif_receive_skb(skb); +#endif } EXPORT_SYMBOL(netif_receive_skb); @@ -3168,6 +3182,7 @@ void netif_napi_del(struct napi_struct *napi) } EXPORT_SYMBOL(netif_napi_del); +#ifdef CONFIG_SMP /* * net_rps_action sends any pending IPI's for rps. This is only called from * softirq and interrupts must be enabled. @@ -3184,6 +3199,7 @@ static void net_rps_action(cpumask_t *mask) } cpus_clear(*mask); } +#endif static void net_rx_action(struct softirq_action *h) { @@ -3191,8 +3207,10 @@ static void net_rx_action(struct softirq_action *h) unsigned long time_limit = jiffies + 2; int budget = netdev_budget; void *have; +#ifdef CONFIG_SMP int select; struct rps_remote_softirq_cpus *rcpus; +#endif local_irq_disable(); @@ -3255,6 +3273,7 @@ static void net_rx_action(struct softirq_action *h) netpoll_poll_unlock(have); } out: +#ifdef CONFIG_SMP rcpus = &__get_cpu_var(rps_remote_softirq_cpus); select = rcpus->select; rcpus->select ^= 1; @@ -3262,6 +3281,9 @@ out: local_irq_enable(); net_rps_action(&rcpus->mask[select]); +#else + local_irq_enable(); +#endif #ifdef CONFIG_NET_DMA /* @@ -6204,9 +6226,11 @@ static int __init net_dev_init(void) queue->completion_queue = NULL; INIT_LIST_HEAD(&queue->poll_list); +#ifdef CONFIG_SMP queue->csd.func = trigger_softirq; queue->csd.info = queue; queue->csd.flags = 0; +#endif queue->backlog.poll = process_backlog; queue->backlog.weight = weight_p; -- cgit v1.2.3 From 3ca5b4042ecae5e73c59de62e4ac0db31c10e0f8 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 10 Mar 2010 10:29:35 +0000 Subject: bonding: check return value of nofitier when changing type This patch adds the possibility to refuse the bonding type change for other subsystems (such as for example bridge, vlan, etc.) Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 11 +++++++++-- include/linux/netdevice.h | 2 +- net/core/dev.c | 4 ++-- 3 files changed, 12 insertions(+), 5 deletions(-) (limited to 'net/core/dev.c') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 7eeb18751d67..cbe9e353d46a 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1480,8 +1480,15 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) bond_dev->name, bond_dev->type, slave_dev->type); - netdev_bonding_change(bond_dev, - NETDEV_PRE_TYPE_CHANGE); + res = netdev_bonding_change(bond_dev, + NETDEV_PRE_TYPE_CHANGE); + res = notifier_to_errno(res); + if (res) { + pr_err("%s: refused to change device type\n", + bond_dev->name); + res = -EBUSY; + goto err_undo_flags; + } if (slave_dev->type != ARPHRD_ETHER) bond_setup_by_slave(bond_dev, slave_dev); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 726ecd1af535..813bed723f58 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2005,7 +2005,7 @@ extern void __dev_addr_unsync(struct dev_addr_list **to, int *to_count, struct extern int dev_set_promiscuity(struct net_device *dev, int inc); extern int dev_set_allmulti(struct net_device *dev, int inc); extern void netdev_state_change(struct net_device *dev); -extern void netdev_bonding_change(struct net_device *dev, +extern int netdev_bonding_change(struct net_device *dev, unsigned long event); extern void netdev_features_change(struct net_device *dev); /* Load a device via the kmod */ diff --git a/net/core/dev.c b/net/core/dev.c index 1a7e1d1d5ad9..d1f027c41e73 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1084,9 +1084,9 @@ void netdev_state_change(struct net_device *dev) } EXPORT_SYMBOL(netdev_state_change); -void netdev_bonding_change(struct net_device *dev, unsigned long event) +int netdev_bonding_change(struct net_device *dev, unsigned long event) { - call_netdevice_notifiers(event, dev); + return call_netdevice_notifiers(event, dev); } EXPORT_SYMBOL(netdev_bonding_change); -- cgit v1.2.3 From 32a806c194ea112cfab00f558482dd97bee5e44e Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 19 Mar 2010 04:00:23 +0000 Subject: bonding: flush unicast and multicast lists when changing type After the type change, addresses in unicast and multicast lists wouldn't make sense, not to mention possible different lenghts. So flush both lists here. Note "dev_addr_discard" will be very soon replaced by "dev_mc_flush" (once mc_list conversion will be done). Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 4 ++++ include/linux/netdevice.h | 2 ++ net/core/dev.c | 6 ++++-- 3 files changed, 10 insertions(+), 2 deletions(-) (limited to 'net/core/dev.c') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index cbe9e353d46a..c2aceaab0143 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1490,6 +1490,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) goto err_undo_flags; } + /* Flush unicast and multicast addresses */ + dev_unicast_flush(bond_dev); + dev_addr_discard(bond_dev); + if (slave_dev->type != ARPHRD_ETHER) bond_setup_by_slave(bond_dev, slave_dev); else diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 9fc6ee8e7508..c96c41e08e37 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1994,10 +1994,12 @@ 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 void dev_unicast_flush(struct net_device *dev); extern int dev_mc_delete(struct net_device *dev, void *addr, int alen, int all); extern int dev_mc_add(struct net_device *dev, void *addr, int alen, int newonly); extern int dev_mc_sync(struct net_device *to, struct net_device *from); extern void dev_mc_unsync(struct net_device *to, struct net_device *from); +extern void dev_addr_discard(struct net_device *dev); extern int __dev_addr_delete(struct dev_addr_list **list, int *count, void *addr, int alen, int all); extern int __dev_addr_add(struct dev_addr_list **list, int *count, void *addr, int alen, int newonly); extern int __dev_addr_sync(struct dev_addr_list **to, int *to_count, struct dev_addr_list **from, int *from_count); diff --git a/net/core/dev.c b/net/core/dev.c index c0e260870c0a..fe2a754238a9 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4457,12 +4457,13 @@ void dev_unicast_unsync(struct net_device *to, struct net_device *from) } EXPORT_SYMBOL(dev_unicast_unsync); -static void dev_unicast_flush(struct net_device *dev) +void dev_unicast_flush(struct net_device *dev) { netif_addr_lock_bh(dev); __hw_addr_flush(&dev->uc); netif_addr_unlock_bh(dev); } +EXPORT_SYMBOL(dev_unicast_flush); static void dev_unicast_init(struct net_device *dev) { @@ -4484,7 +4485,7 @@ static void __dev_addr_discard(struct dev_addr_list **list) } } -static void dev_addr_discard(struct net_device *dev) +void dev_addr_discard(struct net_device *dev) { netif_addr_lock_bh(dev); @@ -4493,6 +4494,7 @@ static void dev_addr_discard(struct net_device *dev) netif_addr_unlock_bh(dev); } +EXPORT_SYMBOL(dev_addr_discard); /** * dev_get_flags - get flags reported to userspace -- cgit v1.2.3 From 283f2fe87e980d8af5ad8aa63751e7e3258ee05a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Mar 2010 13:37:40 +0000 Subject: net: speedup netdev_set_master() We currently force a synchronize_net() in netdev_set_master() This seems necessary only when a slave had a master and we dismantle it. In the other case ("ifenslave bond0 ethO"), we dont need this long delay. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index fe2a754238a9..2d01f18f303a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3757,11 +3757,10 @@ int netdev_set_master(struct net_device *slave, struct net_device *master) slave->master = master; - synchronize_net(); - - if (old) + if (old) { + synchronize_net(); dev_put(old); - + } if (master) slave->flags |= IFF_SLAVE; else -- cgit v1.2.3 From 99fe3c391d50d381687fd84ed0ab22d57079e41f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Mar 2010 11:27:25 +0000 Subject: net: dev_getfirstbyhwtype() optimization Use RCU to avoid RTNL use in dev_getfirstbyhwtype() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 2d01f18f303a..a03aab45e84f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -772,14 +772,17 @@ EXPORT_SYMBOL(__dev_getfirstbyhwtype); struct net_device *dev_getfirstbyhwtype(struct net *net, unsigned short type) { - struct net_device *dev; + struct net_device *dev, *ret = NULL; - rtnl_lock(); - dev = __dev_getfirstbyhwtype(net, type); - if (dev) - dev_hold(dev); - rtnl_unlock(); - return dev; + rcu_read_lock(); + for_each_netdev_rcu(net, dev) + if (dev->type == type) { + dev_hold(dev); + ret = dev; + break; + } + rcu_read_unlock(); + return ret; } EXPORT_SYMBOL(dev_getfirstbyhwtype); -- cgit v1.2.3 From e51d739ab79110c43ca03daf3ddb3c52dadd38b7 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Tue, 23 Mar 2010 13:39:19 +0000 Subject: net: Fix locking in flush_backlog Need to take spinlocks when dequeuing from input_pkt_queue in flush_backlog. Also, flush_backlog can now be called directly from netdev_run_todo. Signed-off-by: Tom Herbert Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index a03aab45e84f..5e3dc28cbf5a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2766,17 +2766,19 @@ int netif_receive_skb(struct sk_buff *skb) EXPORT_SYMBOL(netif_receive_skb); /* Network device is going away, flush any packets still pending */ -static void flush_backlog(void *arg) +static void flush_backlog(struct net_device *dev, int cpu) { - struct net_device *dev = arg; - struct softnet_data *queue = &__get_cpu_var(softnet_data); + struct softnet_data *queue = &per_cpu(softnet_data, cpu); struct sk_buff *skb, *tmp; + unsigned long flags; + spin_lock_irqsave(&queue->input_pkt_queue.lock, flags); skb_queue_walk_safe(&queue->input_pkt_queue, skb, tmp) if (skb->dev == dev) { __skb_unlink(skb, &queue->input_pkt_queue); kfree_skb(skb); } + spin_unlock_irqrestore(&queue->input_pkt_queue.lock, flags); } static int napi_gro_complete(struct sk_buff *skb) @@ -5545,6 +5547,7 @@ void netdev_run_todo(void) while (!list_empty(&list)) { struct net_device *dev = list_first_entry(&list, struct net_device, todo_list); + int i; list_del(&dev->todo_list); if (unlikely(dev->reg_state != NETREG_UNREGISTERING)) { @@ -5556,7 +5559,8 @@ void netdev_run_todo(void) dev->reg_state = NETREG_UNREGISTERED; - on_each_cpu(flush_backlog, dev, 1); + for_each_online_cpu(i) + flush_backlog(dev, i); netdev_wait_allrefs(dev); -- cgit v1.2.3 From df3345457a7a174dfb5872a070af80d456985038 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 24 Mar 2010 19:13:54 +0000 Subject: rps: add CONFIG_RPS RPS currently depends on SMP and SYSFS Adding a CONFIG_RPS makes sense in case this requirement changes in the future. This patch saves about 1500 bytes of kernel text in case SMP is on but SYSFS is off. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 ++++ net/Kconfig | 5 +++++ net/core/dev.c | 29 +++++++++++++++++++---------- 3 files changed, 28 insertions(+), 10 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c96c41e08e37..53c272f2a734 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -531,6 +531,7 @@ struct netdev_queue { unsigned long tx_dropped; } ____cacheline_aligned_in_smp; +#ifdef CONFIG_RPS /* * This structure holds an RPS map which can be of variable length. The * map is an array of CPUs. @@ -549,6 +550,7 @@ struct netdev_rx_queue { struct netdev_rx_queue *first; atomic_t count; } ____cacheline_aligned_in_smp; +#endif /* * This structure defines the management hooks for network devices. @@ -897,12 +899,14 @@ struct net_device { unsigned char broadcast[MAX_ADDR_LEN]; /* hw bcast add */ +#ifdef CONFIG_RPS struct kset *queues_kset; struct netdev_rx_queue *_rx; /* Number of RX queues allocated at alloc_netdev_mq() time */ unsigned int num_rx_queues; +#endif struct netdev_queue rx_queue; diff --git a/net/Kconfig b/net/Kconfig index 041c35edb763..68514644ce91 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -203,6 +203,11 @@ source "net/ieee802154/Kconfig" source "net/sched/Kconfig" source "net/dcb/Kconfig" +config RPS + boolean + depends on SMP && SYSFS + default y + menu "Network testing" config NET_PKTGEN diff --git a/net/core/dev.c b/net/core/dev.c index 5e3dc28cbf5a..bcb3ed26af1c 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2177,7 +2177,7 @@ int weight_p __read_mostly = 64; /* old backlog weight */ DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; -#ifdef CONFIG_SMP +#ifdef CONFIG_RPS /* * get_rps_cpu is called from netif_receive_skb and returns the target * CPU from the RPS map of the receiving queue for a given skb. @@ -2325,7 +2325,7 @@ enqueue: /* Schedule NAPI for backlog device */ if (napi_schedule_prep(&queue->backlog)) { -#ifdef CONFIG_SMP +#ifdef CONFIG_RPS if (cpu != smp_processor_id()) { struct rps_remote_softirq_cpus *rcpus = &__get_cpu_var(rps_remote_softirq_cpus); @@ -2376,7 +2376,7 @@ int netif_rx(struct sk_buff *skb) if (!skb->tstamp.tv64) net_timestamp(skb); -#ifdef CONFIG_SMP +#ifdef CONFIG_RPS cpu = get_rps_cpu(skb->dev, skb); if (cpu < 0) cpu = smp_processor_id(); @@ -2750,7 +2750,7 @@ out: */ int netif_receive_skb(struct sk_buff *skb) { -#ifdef CONFIG_SMP +#ifdef CONFIG_RPS int cpu; cpu = get_rps_cpu(skb->dev, skb); @@ -3189,7 +3189,7 @@ void netif_napi_del(struct napi_struct *napi) } EXPORT_SYMBOL(netif_napi_del); -#ifdef CONFIG_SMP +#ifdef CONFIG_RPS /* * net_rps_action sends any pending IPI's for rps. This is only called from * softirq and interrupts must be enabled. @@ -3214,7 +3214,7 @@ static void net_rx_action(struct softirq_action *h) unsigned long time_limit = jiffies + 2; int budget = netdev_budget; void *have; -#ifdef CONFIG_SMP +#ifdef CONFIG_RPS int select; struct rps_remote_softirq_cpus *rcpus; #endif @@ -3280,7 +3280,7 @@ static void net_rx_action(struct softirq_action *h) netpoll_poll_unlock(have); } out: -#ifdef CONFIG_SMP +#ifdef CONFIG_RPS rcpus = &__get_cpu_var(rps_remote_softirq_cpus); select = rcpus->select; rcpus->select ^= 1; @@ -5277,6 +5277,7 @@ int register_netdevice(struct net_device *dev) dev->iflink = -1; +#ifdef CONFIG_RPS if (!dev->num_rx_queues) { /* * Allocate a single RX queue if driver never called @@ -5293,7 +5294,7 @@ int register_netdevice(struct net_device *dev) atomic_set(&dev->_rx->count, 1); dev->num_rx_queues = 1; } - +#endif /* Init, if this function is available */ if (dev->netdev_ops->ndo_init) { ret = dev->netdev_ops->ndo_init(dev); @@ -5653,11 +5654,13 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, void (*setup)(struct net_device *), unsigned int queue_count) { struct netdev_queue *tx; - struct netdev_rx_queue *rx; struct net_device *dev; size_t alloc_size; struct net_device *p; +#ifdef CONFIG_RPS + struct netdev_rx_queue *rx; int i; +#endif BUG_ON(strlen(name) >= sizeof(dev->name)); @@ -5683,6 +5686,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, goto free_p; } +#ifdef CONFIG_RPS rx = kcalloc(queue_count, sizeof(struct netdev_rx_queue), GFP_KERNEL); if (!rx) { printk(KERN_ERR "alloc_netdev: Unable to allocate " @@ -5698,6 +5702,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, */ for (i = 0; i < queue_count; i++) rx[i].first = rx; +#endif dev = PTR_ALIGN(p, NETDEV_ALIGN); dev->padded = (char *)dev - (char *)p; @@ -5713,8 +5718,10 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, dev->num_tx_queues = queue_count; dev->real_num_tx_queues = queue_count; +#ifdef CONFIG_RPS dev->_rx = rx; dev->num_rx_queues = queue_count; +#endif dev->gso_max_size = GSO_MAX_SIZE; @@ -5731,8 +5738,10 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, return dev; free_rx: +#ifdef CONFIG_RPS kfree(rx); free_tx: +#endif kfree(tx); free_p: kfree(p); @@ -6236,7 +6245,7 @@ static int __init net_dev_init(void) queue->completion_queue = NULL; INIT_LIST_HEAD(&queue->poll_list); -#ifdef CONFIG_SMP +#ifdef CONFIG_RPS queue->csd.func = trigger_softirq; queue->csd.info = queue; queue->csd.flags = 0; -- cgit v1.2.3 From 10f744d205dde72a0016dbdb11e239da8269958b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 28 Mar 2010 23:07:20 -0700 Subject: net: __netif_receive_skb should be static Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index bcb3ed26af1c..887aa84fcd46 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2621,7 +2621,7 @@ void netif_nit_deliver(struct sk_buff *skb) rcu_read_unlock(); } -int __netif_receive_skb(struct sk_buff *skb) +static int __netif_receive_skb(struct sk_buff *skb) { struct packet_type *ptype, *pt_prev; struct net_device *orig_dev; -- cgit v1.2.3 From 152102c7f2bf191690f1069bae292ea3925adf14 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 30 Mar 2010 20:16:22 +0000 Subject: rps: keep the old behavior on SMP without rps keep the old behavior on SMP without rps RPS introduces a lock operation to per cpu variable input_pkt_queue on SMP whenever rps is enabled or not. On SMP without RPS, this lock isn't needed at all. Signed-off-by: Changli Gao ---- net/core/dev.c | 42 ++++++++++++++++++++++++++++-------------- 1 file changed, 28 insertions(+), 14 deletions(-) Signed-off-by: David S. Miller --- net/core/dev.c | 42 ++++++++++++++++++++++++++++-------------- 1 file changed, 28 insertions(+), 14 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 887aa84fcd46..427cd53c118d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -206,6 +206,20 @@ static inline struct hlist_head *dev_index_hash(struct net *net, int ifindex) return &net->dev_index_head[ifindex & (NETDEV_HASHENTRIES - 1)]; } +static inline void rps_lock(struct softnet_data *queue) +{ +#ifdef CONFIG_RPS + spin_lock(&queue->input_pkt_queue.lock); +#endif +} + +static inline void rps_unlock(struct softnet_data *queue) +{ +#ifdef CONFIG_RPS + spin_unlock(&queue->input_pkt_queue.lock); +#endif +} + /* Device list insertion */ static int list_netdevice(struct net_device *dev) { @@ -2313,13 +2327,13 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu) local_irq_save(flags); __get_cpu_var(netdev_rx_stat).total++; - spin_lock(&queue->input_pkt_queue.lock); + rps_lock(queue); if (queue->input_pkt_queue.qlen <= netdev_max_backlog) { if (queue->input_pkt_queue.qlen) { enqueue: __skb_queue_tail(&queue->input_pkt_queue, skb); - spin_unlock_irqrestore(&queue->input_pkt_queue.lock, - flags); + rps_unlock(queue); + local_irq_restore(flags); return NET_RX_SUCCESS; } @@ -2341,7 +2355,7 @@ enqueue: goto enqueue; } - spin_unlock(&queue->input_pkt_queue.lock); + rps_unlock(queue); __get_cpu_var(netdev_rx_stat).dropped++; local_irq_restore(flags); @@ -2766,19 +2780,19 @@ int netif_receive_skb(struct sk_buff *skb) EXPORT_SYMBOL(netif_receive_skb); /* Network device is going away, flush any packets still pending */ -static void flush_backlog(struct net_device *dev, int cpu) +static void flush_backlog(void *arg) { - struct softnet_data *queue = &per_cpu(softnet_data, cpu); + struct net_device *dev = arg; + struct softnet_data *queue = &__get_cpu_var(softnet_data); struct sk_buff *skb, *tmp; - unsigned long flags; - spin_lock_irqsave(&queue->input_pkt_queue.lock, flags); + rps_lock(queue); skb_queue_walk_safe(&queue->input_pkt_queue, skb, tmp) if (skb->dev == dev) { __skb_unlink(skb, &queue->input_pkt_queue); kfree_skb(skb); } - spin_unlock_irqrestore(&queue->input_pkt_queue.lock, flags); + rps_unlock(queue); } static int napi_gro_complete(struct sk_buff *skb) @@ -3091,14 +3105,16 @@ static int process_backlog(struct napi_struct *napi, int quota) do { struct sk_buff *skb; - spin_lock_irq(&queue->input_pkt_queue.lock); + local_irq_disable(); + rps_lock(queue); skb = __skb_dequeue(&queue->input_pkt_queue); if (!skb) { __napi_complete(napi); spin_unlock_irq(&queue->input_pkt_queue.lock); break; } - spin_unlock_irq(&queue->input_pkt_queue.lock); + rps_unlock(queue); + local_irq_enable(); __netif_receive_skb(skb); } while (++work < quota && jiffies == start_time); @@ -5548,7 +5564,6 @@ void netdev_run_todo(void) while (!list_empty(&list)) { struct net_device *dev = list_first_entry(&list, struct net_device, todo_list); - int i; list_del(&dev->todo_list); if (unlikely(dev->reg_state != NETREG_UNREGISTERING)) { @@ -5560,8 +5575,7 @@ void netdev_run_todo(void) dev->reg_state = NETREG_UNREGISTERED; - for_each_online_cpu(i) - flush_backlog(dev, i); + on_each_cpu(flush_backlog, dev, 1); netdev_wait_allrefs(dev); -- cgit v1.2.3 From 5acbbd428db47b12f137a8a2aa96b3c0a96b744e Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Tue, 30 Mar 2010 22:35:50 +0000 Subject: net: change illegal_highdma to use dma_mask Robert Hancock pointed out two problems about NETIF_F_HIGHDMA: -Many drivers only set the flag when they detect they can use 64-bit DMA, since otherwise they could receive DMA addresses that they can't handle (which on platforms without IOMMU/SWIOTLB support is fatal). This means that if 64-bit support isn't available, even buffers located below 4GB will get copied unnecessarily. -Some drivers set the flag even though they can't actually handle 64-bit DMA, which would mean that on platforms without IOMMU/SWIOTLB they would get a DMA mapping error if the memory they received happened to be located above 4GB. http://lkml.org/lkml/2010/3/3/530 We can use the dma_mask if we need bouncing or not here. Then we can safely fix drivers that misuse NETIF_F_HIGHDMA. Signed-off-by: FUJITA Tomonori Signed-off-by: David S. Miller --- net/core/dev.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 427cd53c118d..e19cdae49fef 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -129,6 +129,7 @@ #include #include #include +#include #include "net-sysfs.h" @@ -1804,14 +1805,21 @@ static inline int illegal_highdma(struct net_device *dev, struct sk_buff *skb) { #ifdef CONFIG_HIGHMEM int i; + if (!(dev->features & NETIF_F_HIGHDMA)) { + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) + if (PageHighMem(skb_shinfo(skb)->frags[i].page)) + return 1; + } - if (dev->features & NETIF_F_HIGHDMA) - return 0; - - for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) - if (PageHighMem(skb_shinfo(skb)->frags[i].page)) - return 1; + if (PCI_DMA_BUS_IS_PHYS) { + struct device *pdev = dev->dev.parent; + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + dma_addr_t addr = page_to_phys(skb_shinfo(skb)->frags[i].page); + if (!pdev->dma_mask || addr + PAGE_SIZE - 1 > *pdev->dma_mask) + return 1; + } + } #endif return 0; } -- cgit v1.2.3 From 9092c658bab215b2752fa59d2a36c05b74d1e9e9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 2 Apr 2010 13:34:49 -0700 Subject: net: illegal_highdma() fix Followup to commit 5acbbd428db47b12f137a8a2aa96b3c0a96b744e (net: change illegal_highdma to use dma_mask) If dev->dev.parent is NULL, we should not try to dereference it. Dont force inline illegal_highdma() as its pretty big now. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index e19cdae49fef..c6b52068d5ec 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1801,7 +1801,7 @@ EXPORT_SYMBOL(netdev_rx_csum_fault); * 2. No high memory really exists on this machine. */ -static inline int illegal_highdma(struct net_device *dev, struct sk_buff *skb) +static int illegal_highdma(struct net_device *dev, struct sk_buff *skb) { #ifdef CONFIG_HIGHMEM int i; @@ -1814,6 +1814,8 @@ static inline int illegal_highdma(struct net_device *dev, struct sk_buff *skb) if (PCI_DMA_BUS_IS_PHYS) { struct device *pdev = dev->dev.parent; + if (!pdev) + return 0; for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { dma_addr_t addr = page_to_phys(skb_shinfo(skb)->frags[i].page); if (!pdev->dma_mask || addr + PAGE_SIZE - 1 > *pdev->dma_mask) -- cgit v1.2.3 From a748ee2426817a95b1f03012d8f339c45c722ae1 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 1 Apr 2010 21:22:09 +0000 Subject: net: move address list functions to a separate file +little renaming of unicast functions to be smooth with multicast ones Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 2 +- drivers/net/macvlan.c | 10 +- drivers/scsi/fcoe/fcoe.c | 14 +- include/linux/netdevice.h | 15 +- net/8021q/vlan.c | 4 +- net/8021q/vlan_dev.c | 14 +- net/core/Makefile | 3 +- net/core/dev.c | 430 +----------------------------------- net/core/dev_addr_lists.c | 478 ++++++++++++++++++++++++++++++++++++++++ net/dsa/slave.c | 14 +- net/packet/af_packet.c | 4 +- 11 files changed, 524 insertions(+), 464 deletions(-) create mode 100644 net/core/dev_addr_lists.c (limited to 'net/core/dev.c') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index f8ab60b914c4..d6ae63b2cf00 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1491,7 +1491,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) } /* Flush unicast and multicast addresses */ - dev_unicast_flush(bond_dev); + dev_uc_flush(bond_dev); dev_addr_discard(bond_dev); if (slave_dev->type != ARPHRD_ETHER) diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 445e73c343ba..9a939d828b47 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -282,7 +282,7 @@ 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); + err = dev_uc_add(lowerdev, dev->dev_addr); if (err < 0) goto out; if (dev->flags & IFF_ALLMULTI) { @@ -294,7 +294,7 @@ static int macvlan_open(struct net_device *dev) return 0; del_unicast: - dev_unicast_delete(lowerdev, dev->dev_addr); + dev_uc_del(lowerdev, dev->dev_addr); out: return err; } @@ -308,7 +308,7 @@ 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); + dev_uc_del(lowerdev, dev->dev_addr); macvlan_hash_del(vlan); return 0; @@ -332,11 +332,11 @@ static int macvlan_set_mac_address(struct net_device *dev, void *p) if (macvlan_addr_busy(vlan->port, addr->sa_data)) return -EBUSY; - err = dev_unicast_add(lowerdev, addr->sa_data); + err = dev_uc_add(lowerdev, addr->sa_data); if (err) return err; - dev_unicast_delete(lowerdev, dev->dev_addr); + dev_uc_del(lowerdev, dev->dev_addr); macvlan_hash_change_addr(vlan, addr->sa_data); } diff --git a/drivers/scsi/fcoe/fcoe.c b/drivers/scsi/fcoe/fcoe.c index 2f47ae7cce91..de33e38a4059 100644 --- a/drivers/scsi/fcoe/fcoe.c +++ b/drivers/scsi/fcoe/fcoe.c @@ -308,9 +308,9 @@ static int fcoe_interface_setup(struct fcoe_interface *fcoe, * for multiple unicast MACs. */ memcpy(flogi_maddr, (u8[6]) FC_FCOE_FLOGI_MAC, ETH_ALEN); - dev_unicast_add(netdev, flogi_maddr); + dev_uc_add(netdev, flogi_maddr); if (fip->spma) - dev_unicast_add(netdev, fip->ctl_src_addr); + dev_uc_add(netdev, fip->ctl_src_addr); dev_mc_add(netdev, FIP_ALL_ENODE_MACS, ETH_ALEN, 0); /* @@ -394,9 +394,9 @@ void fcoe_interface_cleanup(struct fcoe_interface *fcoe) /* Delete secondary MAC addresses */ memcpy(flogi_maddr, (u8[6]) FC_FCOE_FLOGI_MAC, ETH_ALEN); - dev_unicast_delete(netdev, flogi_maddr); + dev_uc_del(netdev, flogi_maddr); if (fip->spma) - dev_unicast_delete(netdev, fip->ctl_src_addr); + dev_uc_del(netdev, fip->ctl_src_addr); dev_mc_delete(netdev, FIP_ALL_ENODE_MACS, ETH_ALEN, 0); /* Tell the LLD we are done w/ FCoE */ @@ -490,9 +490,9 @@ static void fcoe_update_src_mac(struct fc_lport *lport, u8 *addr) rtnl_lock(); if (!is_zero_ether_addr(port->data_src_addr)) - dev_unicast_delete(fcoe->netdev, port->data_src_addr); + dev_uc_del(fcoe->netdev, port->data_src_addr); if (!is_zero_ether_addr(addr)) - dev_unicast_add(fcoe->netdev, addr); + dev_uc_add(fcoe->netdev, addr); memcpy(port->data_src_addr, addr, ETH_ALEN); rtnl_unlock(); } @@ -819,7 +819,7 @@ static void fcoe_if_destroy(struct fc_lport *lport) rtnl_lock(); if (!is_zero_ether_addr(port->data_src_addr)) - dev_unicast_delete(netdev, port->data_src_addr); + dev_uc_del(netdev, port->data_src_addr); rtnl_unlock(); /* receives may not be stopped until after this */ diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index b5670ab5107c..60f0c83192fe 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1991,15 +1991,20 @@ extern int dev_addr_add_multiple(struct net_device *to_dev, extern int dev_addr_del_multiple(struct net_device *to_dev, struct net_device *from_dev, unsigned char addr_type); +extern void dev_addr_flush(struct net_device *dev); +extern int dev_addr_init(struct net_device *dev); + +/* Functions used for unicast addresses handling */ +extern int dev_uc_add(struct net_device *dev, unsigned char *addr); +extern int dev_uc_del(struct net_device *dev, unsigned char *addr); +extern int dev_uc_sync(struct net_device *to, struct net_device *from); +extern void dev_uc_unsync(struct net_device *to, struct net_device *from); +extern void dev_uc_flush(struct net_device *dev); +extern void dev_uc_init(struct net_device *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); -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 void dev_unicast_flush(struct net_device *dev); extern int dev_mc_delete(struct net_device *dev, void *addr, int alen, int all); extern int dev_mc_add(struct net_device *dev, void *addr, int alen, int newonly); extern int dev_mc_sync(struct net_device *to, struct net_device *from); diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index c39a5f41169c..bd33f02013ec 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -356,13 +356,13 @@ 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); + dev_uc_del(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); + dev_uc_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 9e83272fc5b0..7f4d247237e4 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -461,7 +461,7 @@ 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); + err = dev_uc_add(real_dev, dev->dev_addr); if (err < 0) goto out; } @@ -490,7 +490,7 @@ 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); + dev_uc_del(real_dev, dev->dev_addr); out: netif_carrier_off(dev); return err; @@ -505,14 +505,14 @@ static int vlan_dev_stop(struct net_device *dev) vlan_gvrp_request_leave(dev); dev_mc_unsync(real_dev, dev); - dev_unicast_unsync(real_dev, dev); + dev_uc_unsync(real_dev, dev); if (dev->flags & IFF_ALLMULTI) dev_set_allmulti(real_dev, -1); if (dev->flags & IFF_PROMISC) 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_uc_del(real_dev, dev->dev_addr); netif_carrier_off(dev); return 0; @@ -531,13 +531,13 @@ 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); + err = dev_uc_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); + dev_uc_del(real_dev, dev->dev_addr); out: memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN); @@ -654,7 +654,7 @@ static void vlan_dev_change_rx_flags(struct net_device *dev, int change) static void vlan_dev_set_rx_mode(struct net_device *vlan_dev) { dev_mc_sync(vlan_dev_info(vlan_dev)->real_dev, vlan_dev); - dev_unicast_sync(vlan_dev_info(vlan_dev)->real_dev, vlan_dev); + dev_uc_sync(vlan_dev_info(vlan_dev)->real_dev, vlan_dev); } /* diff --git a/net/core/Makefile b/net/core/Makefile index 08791ac3e05a..0a899f1aadb9 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -8,7 +8,8 @@ obj-y := sock.o request_sock.o skbuff.o iovec.o datagram.o stream.o scm.o \ obj-$(CONFIG_SYSCTL) += sysctl_net_core.o obj-y += dev.o ethtool.o dev_mcast.o dst.o netevent.o \ - neighbour.o rtnetlink.o utils.o link_watch.o filter.o + neighbour.o rtnetlink.o utils.o link_watch.o filter.o \ + dev_addr_lists.o obj-$(CONFIG_XFRM) += flow.o obj-y += net-sysfs.o diff --git a/net/core/dev.c b/net/core/dev.c index c6b52068d5ec..949c62dba719 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3968,314 +3968,6 @@ 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 netdev_hw_addr_list *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; - - list_for_each_entry(ha, &list->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; - ha = kmalloc(alloc_size, GFP_ATOMIC); - if (!ha) - 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->list); - list->count++; - 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(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) { - 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); - list->count--; - return 0; - } - } - return -ENOENT; -} - -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) { - 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, list) { - if (ha2 == ha) - break; - type = addr_type ? addr_type : ha2->type; - __hw_addr_del(to_list, ha2->addr, addr_len, type); - } - return err; -} - -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) { - type = addr_type ? addr_type : ha->type; - __hw_addr_del(to_list, ha->addr, addr_len, addr_type); - } -} - -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) { - if (!ha->synced) { - 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, 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 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) { - if (ha->synced) { - __hw_addr_del(to_list, ha->addr, - addr_len, ha->type); - ha->synced = false; - __hw_addr_del(from_list, ha->addr, - addr_len, ha->type); - } - } -} - -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) { - 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 */ - -static void dev_addr_flush(struct net_device *dev) -{ - /* rtnl_mutex must be held here */ - - __hw_addr_flush(&dev->dev_addrs); - 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 */ - - __hw_addr_init(&dev->dev_addrs); - memset(addr, 0, 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_addrs.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_addrs, 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; - struct netdev_hw_addr *ha; - - ASSERT_RTNL(); - - /* - * We can not remove the first address from the list because - * dev->dev_addr points to that. - */ - 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_addrs, addr, dev->addr_len, - addr_type); - 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(&to_dev->dev_addrs, &from_dev->dev_addrs, - to_dev->addr_len, addr_type); - 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(&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); - /* multicast addresses handling functions */ int __dev_addr_delete(struct dev_addr_list **list, int *count, @@ -4336,57 +4028,6 @@ int __dev_addr_add(struct dev_addr_list **list, int *count, return 0; } -/** - * dev_unicast_delete - Release secondary unicast address. - * @dev: device - * @addr: address to delete - * - * 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 err; - - ASSERT_RTNL(); - - netif_addr_lock_bh(dev); - err = __hw_addr_del(&dev->uc, 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); - -/** - * dev_unicast_add - add a secondary unicast address - * @dev: device - * @addr: address to add - * - * 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 err; - - ASSERT_RTNL(); - - netif_addr_lock_bh(dev); - err = __hw_addr_add(&dev->uc, 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); int __dev_addr_sync(struct dev_addr_list **to, int *to_count, struct dev_addr_list **from, int *from_count) @@ -4436,71 +4077,6 @@ void __dev_addr_unsync(struct dev_addr_list **to, int *to_count, } EXPORT_SYMBOL_GPL(__dev_addr_unsync); -/** - * dev_unicast_sync - Synchronize device's unicast list to another device - * @to: destination device - * @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. - * - * This function is intended to be called from the dev->set_rx_mode - * function of layered software devices. - */ -int dev_unicast_sync(struct net_device *to, struct net_device *from) -{ - int err = 0; - - if (to->addr_len != from->addr_len) - return -EINVAL; - - netif_addr_lock_bh(to); - err = __hw_addr_sync(&to->uc, &from->uc, to->addr_len); - if (!err) - __dev_set_rx_mode(to); - netif_addr_unlock_bh(to); - return err; -} -EXPORT_SYMBOL(dev_unicast_sync); - -/** - * dev_unicast_unsync - Remove synchronized addresses from the destination device - * @to: destination device - * @from: source device - * - * Remove all addresses that were added to the destination device by - * dev_unicast_sync(). This function is intended to be called from the - * dev->stop function of layered software devices. - */ -void dev_unicast_unsync(struct net_device *to, struct net_device *from) -{ - if (to->addr_len != from->addr_len) - return; - - netif_addr_lock_bh(from); - netif_addr_lock(to); - __hw_addr_unsync(&to->uc, &from->uc, to->addr_len); - __dev_set_rx_mode(to); - netif_addr_unlock(to); - netif_addr_unlock_bh(from); -} -EXPORT_SYMBOL(dev_unicast_unsync); - -void dev_unicast_flush(struct net_device *dev) -{ - netif_addr_lock_bh(dev); - __hw_addr_flush(&dev->uc); - netif_addr_unlock_bh(dev); -} -EXPORT_SYMBOL(dev_unicast_flush); - -static void dev_unicast_init(struct net_device *dev) -{ - __hw_addr_init(&dev->uc); -} - - static void __dev_addr_discard(struct dev_addr_list **list) { struct dev_addr_list *tmp; @@ -5153,7 +4729,7 @@ static void rollback_registered_many(struct list_head *head) /* * Flush the unicast and multicast chains */ - dev_unicast_flush(dev); + dev_uc_flush(dev); dev_addr_discard(dev); if (dev->netdev_ops->ndo_uninit) @@ -5734,7 +5310,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, if (dev_addr_init(dev)) goto free_rx; - dev_unicast_init(dev); + dev_uc_init(dev); dev_net_set(dev, &init_net); @@ -5968,7 +5544,7 @@ 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_uc_flush(dev); dev_addr_discard(dev); netdev_unregister_kobject(dev); diff --git a/net/core/dev_addr_lists.c b/net/core/dev_addr_lists.c new file mode 100644 index 000000000000..7e52b6d18add --- /dev/null +++ b/net/core/dev_addr_lists.c @@ -0,0 +1,478 @@ +/* + * net/core/dev_addr_lists.c - Functions for handling net device lists + * Copyright (c) 2010 Jiri Pirko + * + * This file contains functions for working with unicast, multicast and device + * addresses lists. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include + +/* + * General list handling functions + */ + +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; + + if (addr_len > MAX_ADDR_LEN) + return -EINVAL; + + list_for_each_entry(ha, &list->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; + ha = kmalloc(alloc_size, GFP_ATOMIC); + if (!ha) + 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->list); + list->count++; + 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(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) { + 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); + list->count--; + return 0; + } + } + return -ENOENT; +} + +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) { + 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, list) { + if (ha2 == ha) + break; + type = addr_type ? addr_type : ha2->type; + __hw_addr_del(to_list, ha2->addr, addr_len, type); + } + return err; +} + +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) { + type = addr_type ? addr_type : ha->type; + __hw_addr_del(to_list, ha->addr, addr_len, addr_type); + } +} + +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) { + if (!ha->synced) { + 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, 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 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) { + if (ha->synced) { + __hw_addr_del(to_list, ha->addr, + addr_len, ha->type); + ha->synced = false; + __hw_addr_del(from_list, ha->addr, + addr_len, ha->type); + } + } +} + +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) { + 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 + */ + +/** + * dev_addr_flush - Flush device address list + * @dev: device + * + * Flush device address list and reset ->dev_addr. + * + * The caller must hold the rtnl_mutex. + */ +void dev_addr_flush(struct net_device *dev) +{ + /* rtnl_mutex must be held here */ + + __hw_addr_flush(&dev->dev_addrs); + dev->dev_addr = NULL; +} +EXPORT_SYMBOL(dev_addr_flush); + +/** + * dev_addr_init - Init device address list + * @dev: device + * + * Init device address list and create the first element, + * used by ->dev_addr. + * + * The caller must hold the rtnl_mutex. + */ +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 */ + + __hw_addr_init(&dev->dev_addrs); + memset(addr, 0, 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_addrs.list, + struct netdev_hw_addr, list); + dev->dev_addr = ha->addr; + } + return err; +} +EXPORT_SYMBOL(dev_addr_init); + +/** + * 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_addrs, 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; + struct netdev_hw_addr *ha; + + ASSERT_RTNL(); + + /* + * We can not remove the first address from the list because + * dev->dev_addr points to that. + */ + 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_addrs, addr, dev->addr_len, + addr_type); + 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(&to_dev->dev_addrs, &from_dev->dev_addrs, + to_dev->addr_len, addr_type); + 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(&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 list handling functions + */ + +/** + * dev_uc_add - Add a secondary unicast address + * @dev: device + * @addr: address to add + * + * Add a secondary unicast address to the device or increase + * the reference count if it already exists. + */ +int dev_uc_add(struct net_device *dev, unsigned char *addr) +{ + int err; + + netif_addr_lock_bh(dev); + err = __hw_addr_add(&dev->uc, 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_uc_add); + +/** + * dev_uc_del - Release secondary unicast address. + * @dev: device + * @addr: address to delete + * + * Release reference to a secondary unicast address and remove it + * from the device if the reference count drops to zero. + */ +int dev_uc_del(struct net_device *dev, unsigned char *addr) +{ + int err; + + netif_addr_lock_bh(dev); + err = __hw_addr_del(&dev->uc, 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_uc_del); + +/** + * dev_uc_sync - Synchronize device's unicast list to another device + * @to: destination device + * @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. + * + * This function is intended to be called from the dev->set_rx_mode + * function of layered software devices. + */ +int dev_uc_sync(struct net_device *to, struct net_device *from) +{ + int err = 0; + + if (to->addr_len != from->addr_len) + return -EINVAL; + + netif_addr_lock_bh(to); + err = __hw_addr_sync(&to->uc, &from->uc, to->addr_len); + if (!err) + __dev_set_rx_mode(to); + netif_addr_unlock_bh(to); + return err; +} +EXPORT_SYMBOL(dev_uc_sync); + +/** + * dev_uc_unsync - Remove synchronized addresses from the destination device + * @to: destination device + * @from: source device + * + * Remove all addresses that were added to the destination device by + * dev_uc_sync(). This function is intended to be called from the + * dev->stop function of layered software devices. + */ +void dev_uc_unsync(struct net_device *to, struct net_device *from) +{ + if (to->addr_len != from->addr_len) + return; + + netif_addr_lock_bh(from); + netif_addr_lock(to); + __hw_addr_unsync(&to->uc, &from->uc, to->addr_len); + __dev_set_rx_mode(to); + netif_addr_unlock(to); + netif_addr_unlock_bh(from); +} +EXPORT_SYMBOL(dev_uc_unsync); + +/** + * dev_uc_flush - Flush unicast addresses + * @dev: device + * + * Flush unicast addresses. + */ +void dev_uc_flush(struct net_device *dev) +{ + netif_addr_lock_bh(dev); + __hw_addr_flush(&dev->uc); + netif_addr_unlock_bh(dev); +} +EXPORT_SYMBOL(dev_uc_flush); + +/** + * dev_uc_flush - Init unicast address list + * @dev: device + * + * Init unicast address list. + */ +void dev_uc_init(struct net_device *dev) +{ + __hw_addr_init(&dev->uc); +} +EXPORT_SYMBOL(dev_uc_init); + +/* + * Multicast list handling functions + */ + +/* To be filled here */ diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 2175e6d5cc8d..8fdca56bb08f 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); + err = dev_uc_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); + dev_uc_del(master, dev->dev_addr); out: return err; } @@ -101,14 +101,14 @@ static int dsa_slave_close(struct net_device *dev) struct net_device *master = p->parent->dst->master_netdev; dev_mc_unsync(master, dev); - dev_unicast_unsync(master, dev); + dev_uc_unsync(master, dev); if (dev->flags & IFF_ALLMULTI) dev_set_allmulti(master, -1); if (dev->flags & IFF_PROMISC) dev_set_promiscuity(master, -1); if (compare_ether_addr(dev->dev_addr, master->dev_addr)) - dev_unicast_delete(master, dev->dev_addr); + dev_uc_del(master, dev->dev_addr); return 0; } @@ -130,7 +130,7 @@ static void dsa_slave_set_rx_mode(struct net_device *dev) struct net_device *master = p->parent->dst->master_netdev; dev_mc_sync(master, dev); - dev_unicast_sync(master, dev); + dev_uc_sync(master, dev); } static int dsa_slave_set_mac_address(struct net_device *dev, void *a) @@ -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); + err = dev_uc_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); + dev_uc_del(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 1612d417d10c..48c1e0ae565f 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1705,9 +1705,9 @@ static int packet_dev_mc(struct net_device *dev, struct packet_mclist *i, if (i->alen != dev->addr_len) return -EINVAL; if (what > 0) - return dev_unicast_add(dev, i->addr); + return dev_uc_add(dev, i->addr); else - return dev_unicast_delete(dev, i->addr); + return dev_uc_del(dev, i->addr); break; default: break; -- cgit v1.2.3 From 22bedad3ce112d5ca1eaf043d4990fa2ed698c87 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 1 Apr 2010 21:22:57 +0000 Subject: net: convert multicast list to list_head Converts the list and the core manipulating with it to be the same as uc_list. +uses two functions for adding/removing mc address (normal and "global" variant) instead of a function parameter. +removes dev_mcast.c completely. +exposes netdev_hw_addr_list_* macros along with __hw_addr_* functions for manipulation with lists on a sandbox (used in bonding and 80211 drivers) Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/infiniband/hw/nes/nes_nic.c | 7 +- drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 9 +- drivers/media/dvb/dvb-core/dvb_net.c | 10 +- drivers/net/3c505.c | 7 +- drivers/net/3c523.c | 7 +- drivers/net/3c527.c | 6 +- drivers/net/7990.c | 6 +- drivers/net/8139cp.c | 6 +- drivers/net/8139too.c | 6 +- drivers/net/82596.c | 6 +- drivers/net/a2065.c | 6 +- drivers/net/amd8111e.c | 6 +- drivers/net/arm/am79c961a.c | 6 +- drivers/net/arm/at91_ether.c | 6 +- drivers/net/arm/ixp4xx_eth.c | 8 +- drivers/net/arm/ks8695net.c | 10 +- drivers/net/at1700.c | 6 +- drivers/net/atl1c/atl1c_main.c | 6 +- drivers/net/atl1e/atl1e_main.c | 6 +- drivers/net/atlx/atl2.c | 6 +- drivers/net/atlx/atlx.c | 6 +- drivers/net/atp.c | 6 +- drivers/net/au1000_eth.c | 6 +- drivers/net/b44.c | 6 +- drivers/net/bcm63xx_enet.c | 6 +- drivers/net/benet/be_cmds.c | 6 +- drivers/net/bfin_mac.c | 6 +- drivers/net/bmac.c | 12 +- drivers/net/bnx2.c | 5 +- drivers/net/bnx2x_main.c | 18 +- drivers/net/bonding/bond_main.c | 157 +++++-------- drivers/net/bonding/bonding.h | 2 +- drivers/net/cassini.c | 12 +- drivers/net/chelsio/pm3393.c | 7 +- drivers/net/cpmac.c | 16 +- drivers/net/cris/eth_v10.c | 6 +- drivers/net/cxgb3/xgmac.c | 8 +- drivers/net/davinci_emac.c | 7 +- drivers/net/declance.c | 6 +- drivers/net/defxx.c | 6 +- drivers/net/depca.c | 6 +- drivers/net/dl2k.c | 6 +- drivers/net/dm9000.c | 6 +- drivers/net/e100.c | 6 +- drivers/net/e1000/e1000_main.c | 7 +- drivers/net/e1000e/netdev.c | 7 +- drivers/net/eepro.c | 6 +- drivers/net/eexpress.c | 6 +- drivers/net/ehea/ehea_main.c | 6 +- drivers/net/enic/enic_main.c | 6 +- drivers/net/epic100.c | 6 +- drivers/net/ethoc.c | 6 +- drivers/net/ewrk3.c | 6 +- drivers/net/fealnx.c | 6 +- drivers/net/fec.c | 10 +- drivers/net/fec_mpc52xx.c | 6 +- drivers/net/forcedeth.c | 10 +- drivers/net/fs_enet/mac-fcc.c | 6 +- drivers/net/fs_enet/mac-fec.c | 6 +- drivers/net/fs_enet/mac-scc.c | 6 +- drivers/net/gianfar.c | 9 +- drivers/net/greth.c | 6 +- drivers/net/hamachi.c | 8 +- drivers/net/hp100.c | 6 +- drivers/net/ibm_newemac/core.c | 11 +- drivers/net/ibmlana.c | 8 +- drivers/net/ibmveth.c | 6 +- drivers/net/igb/igb_main.c | 6 +- drivers/net/igbvf/netdev.c | 6 +- drivers/net/ioc3-eth.c | 6 +- drivers/net/ipg.c | 6 +- drivers/net/iseries_veth.c | 6 +- drivers/net/ixgb/ixgb_main.c | 6 +- drivers/net/ixgbe/ixgbe_common.c | 6 +- drivers/net/ixgbevf/vf.c | 6 +- drivers/net/jme.c | 6 +- drivers/net/korina.c | 6 +- drivers/net/ks8851.c | 6 +- drivers/net/ks8851_mll.c | 12 +- drivers/net/ksz884x.c | 11 +- drivers/net/lib82596.c | 6 +- drivers/net/lib8390.c | 6 +- drivers/net/ll_temac_main.c | 16 +- drivers/net/lp486e.c | 6 +- drivers/net/macb.c | 6 +- drivers/net/mace.c | 6 +- drivers/net/macmace.c | 6 +- drivers/net/mlx4/en_netdev.c | 6 +- drivers/net/mv643xx_eth.c | 6 +- drivers/net/myri10ge/myri10ge.c | 8 +- drivers/net/natsemi.c | 6 +- drivers/net/netxen/netxen_nic_hw.c | 12 +- drivers/net/ni52.c | 6 +- drivers/net/niu.c | 5 +- drivers/net/octeon/octeon_mgmt.c | 6 +- drivers/net/pci-skeleton.c | 6 +- drivers/net/pcmcia/axnet_cs.c | 6 +- drivers/net/pcmcia/fmvj18x_cs.c | 6 +- drivers/net/pcmcia/nmclan_cs.c | 6 +- drivers/net/pcmcia/smc91c92_cs.c | 8 +- drivers/net/pcmcia/xirc2ps_cs.c | 6 +- drivers/net/pcnet32.c | 6 +- drivers/net/ps3_gelic_net.c | 6 +- drivers/net/qlcnic/qlcnic_hw.c | 6 +- drivers/net/qlge/qlge_main.c | 6 +- drivers/net/r6040.c | 10 +- drivers/net/r8169.c | 6 +- drivers/net/s2io.c | 8 +- drivers/net/sb1250-mac.c | 6 +- drivers/net/sc92031.c | 6 +- drivers/net/sfc/efx.c | 6 +- drivers/net/sis190.c | 6 +- drivers/net/sis900.c | 10 +- drivers/net/skfp/skfddi.c | 12 +- drivers/net/skge.c | 12 +- drivers/net/sky2.c | 6 +- drivers/net/smc911x.c | 13 +- drivers/net/smc9194.c | 13 +- drivers/net/smc91x.c | 8 +- drivers/net/smsc911x.c | 6 +- drivers/net/smsc9420.c | 6 +- drivers/net/sonic.c | 6 +- drivers/net/spider_net.c | 6 +- drivers/net/starfire.c | 10 +- drivers/net/stmmac/dwmac100.c | 6 +- drivers/net/stmmac/dwmac1000_core.c | 6 +- drivers/net/sun3_82586.c | 6 +- drivers/net/sunbmac.c | 6 +- drivers/net/sundance.c | 6 +- drivers/net/sungem.c | 6 +- drivers/net/sunhme.c | 12 +- drivers/net/sunlance.c | 6 +- drivers/net/sunqe.c | 6 +- drivers/net/sunvnet.c | 8 +- drivers/net/tc35815.c | 6 +- drivers/net/tehuti.c | 6 +- drivers/net/tg3.c | 6 +- drivers/net/tlan.c | 8 +- drivers/net/tokenring/3c359.c | 12 +- drivers/net/tokenring/ibmtr.c | 12 +- drivers/net/tokenring/lanstreamer.c | 12 +- drivers/net/tokenring/olympic.c | 12 +- drivers/net/tokenring/tms380tr.c | 12 +- drivers/net/tsi108_eth.c | 6 +- drivers/net/tulip/de2104x.c | 12 +- drivers/net/tulip/de4x5.c | 10 +- drivers/net/tulip/dmfe.c | 12 +- drivers/net/tulip/tulip_core.c | 27 ++- drivers/net/tulip/uli526x.c | 6 +- drivers/net/tulip/winbond-840.c | 12 +- drivers/net/typhoon.c | 6 +- drivers/net/ucc_geth.c | 8 +- drivers/net/usb/asix.c | 16 +- drivers/net/usb/catc.c | 6 +- drivers/net/usb/dm9601.c | 6 +- drivers/net/usb/mcs7830.c | 6 +- drivers/net/usb/smsc75xx.c | 6 +- drivers/net/usb/smsc95xx.c | 6 +- drivers/net/via-rhine.c | 6 +- drivers/net/via-velocity.c | 6 +- drivers/net/virtio_net.c | 5 +- drivers/net/vmxnet3/vmxnet3_drv.c | 6 +- drivers/net/vxge/vxge-main.c | 6 +- drivers/net/wireless/adm8211.c | 12 +- drivers/net/wireless/ath/ar9170/main.c | 14 +- drivers/net/wireless/ath/ath5k/base.c | 17 +- drivers/net/wireless/libertas/main.c | 12 +- drivers/net/wireless/libertas_tf/main.c | 14 +- drivers/net/wireless/mwl8k.c | 24 +- drivers/net/wireless/orinoco/hw.c | 6 +- drivers/net/wireless/orinoco/hw.h | 1 - drivers/net/wireless/ray_cs.c | 12 +- drivers/net/wireless/rndis_wlan.c | 6 +- drivers/net/wireless/rtl818x/rtl8180_dev.c | 6 +- drivers/net/wireless/rtl818x/rtl8187_dev.c | 4 +- drivers/net/wireless/wl12xx/wl1271_main.c | 25 +- drivers/net/wireless/zd1201.c | 6 +- drivers/net/wireless/zd1211rw/zd_mac.c | 13 +- drivers/net/yellowfin.c | 12 +- drivers/s390/net/qeth_l2_main.c | 5 +- drivers/scsi/fcoe/fcoe.c | 4 +- drivers/staging/arlan/arlan-main.c | 9 +- drivers/staging/et131x/et131x_netdev.c | 6 +- drivers/staging/slicoss/slicoss.c | 6 +- drivers/staging/vt6655/device_main.c | 6 +- drivers/staging/vt6656/main_usb.c | 6 +- drivers/staging/wavelan/wavelan.c | 10 +- drivers/staging/wavelan/wavelan_cs.c | 12 +- drivers/staging/winbond/wbusb.c | 6 +- drivers/staging/wlags49_h2/wl_netdev.c | 12 +- include/linux/netdevice.h | 82 +++---- include/net/mac80211.h | 2 +- net/802/garp.c | 4 +- net/appletalk/ddp.c | 2 +- net/bluetooth/bnep/netdev.c | 8 +- net/core/Makefile | 5 +- net/core/dev.c | 145 +----------- net/core/dev_addr_lists.c | 305 +++++++++++++++++++++++-- net/core/dev_mcast.c | 232 ------------------- net/decnet/dn_dev.c | 12 +- net/ipv4/igmp.c | 4 +- net/ipv4/netfilter/ipt_CLUSTERIP.c | 4 +- net/ipv6/mcast.c | 4 +- net/mac80211/driver-ops.h | 8 +- net/mac80211/ieee80211_i.h | 3 +- net/mac80211/iface.c | 6 +- net/mac80211/main.c | 2 +- net/packet/af_packet.c | 4 +- 208 files changed, 1137 insertions(+), 1327 deletions(-) delete mode 100644 net/core/dev_mcast.c (limited to 'net/core/dev.c') diff --git a/drivers/infiniband/hw/nes/nes_nic.c b/drivers/infiniband/hw/nes/nes_nic.c index 91fdde382e82..ce9ef6bc865c 100644 --- a/drivers/infiniband/hw/nes/nes_nic.c +++ b/drivers/infiniband/hw/nes/nes_nic.c @@ -876,7 +876,7 @@ static void nes_netdev_set_multicast_list(struct net_device *netdev) if (!mc_all_on) { char *addrs; int i; - struct dev_mc_list *mcaddr; + struct netdev_hw_addr *ha; addrs = kmalloc(ETH_ALEN * mc_count, GFP_ATOMIC); if (!addrs) { @@ -884,9 +884,8 @@ static void nes_netdev_set_multicast_list(struct net_device *netdev) goto unlock; } i = 0; - netdev_for_each_mc_addr(mcaddr, netdev) - memcpy(get_addr(addrs, i++), - mcaddr->dmi_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, netdev) + memcpy(get_addr(addrs, i++), ha->addr, ETH_ALEN); perfect_filter_register_address = NES_IDX_PERFECT_FILTER_LOW + pft_entries_preallocated * 0x8; diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 19eba3c877cb..c8a0f7dab5b0 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -783,7 +783,7 @@ void ipoib_mcast_restart_task(struct work_struct *work) struct ipoib_dev_priv *priv = container_of(work, struct ipoib_dev_priv, restart_task); struct net_device *dev = priv->dev; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; struct ipoib_mcast *mcast, *tmcast; LIST_HEAD(remove_list); unsigned long flags; @@ -808,14 +808,13 @@ void ipoib_mcast_restart_task(struct work_struct *work) clear_bit(IPOIB_MCAST_FLAG_FOUND, &mcast->flags); /* Mark all of the entries that are found or don't exist */ - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { union ib_gid mgid; - if (!ipoib_mcast_addr_is_valid(mclist->dmi_addr, - dev->broadcast)) + if (!ipoib_mcast_addr_is_valid(ha->addr, dev->broadcast)) continue; - memcpy(mgid.raw, mclist->dmi_addr + 4, sizeof mgid); + memcpy(mgid.raw, ha->addr + 4, sizeof mgid); mcast = __ipoib_mcast_find(dev, &mgid); if (!mcast || test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) { diff --git a/drivers/media/dvb/dvb-core/dvb_net.c b/drivers/media/dvb/dvb-core/dvb_net.c index dba1c84058b7..cccea412088b 100644 --- a/drivers/media/dvb/dvb-core/dvb_net.c +++ b/drivers/media/dvb/dvb-core/dvb_net.c @@ -1109,14 +1109,14 @@ static int dvb_net_feed_stop(struct net_device *dev) } -static int dvb_set_mc_filter (struct net_device *dev, struct dev_mc_list *mc) +static int dvb_set_mc_filter(struct net_device *dev, unsigned char *addr) { struct dvb_net_priv *priv = netdev_priv(dev); if (priv->multi_num == DVB_NET_MULTICAST_MAX) return -ENOMEM; - memcpy(priv->multi_macs[priv->multi_num], mc->dmi_addr, 6); + memcpy(priv->multi_macs[priv->multi_num], addr, ETH_ALEN); priv->multi_num++; return 0; @@ -1140,7 +1140,7 @@ static void wq_set_multicast_list (struct work_struct *work) dprintk("%s: allmulti mode\n", dev->name); priv->rx_mode = RX_MODE_ALL_MULTI; } else if (!netdev_mc_empty(dev)) { - struct dev_mc_list *mc; + struct netdev_hw_addr *ha; dprintk("%s: set_mc_list, %d entries\n", dev->name, netdev_mc_count(dev)); @@ -1148,8 +1148,8 @@ static void wq_set_multicast_list (struct work_struct *work) priv->rx_mode = RX_MODE_MULTI; priv->multi_num = 0; - netdev_for_each_mc_addr(mc, dev) - dvb_set_mc_filter(dev, mc); + netdev_for_each_mc_addr(ha, dev) + dvb_set_mc_filter(dev, ha->addr); } netif_addr_unlock_bh(dev); diff --git a/drivers/net/3c505.c b/drivers/net/3c505.c index 04b5bba19021..81c8b31e629f 100644 --- a/drivers/net/3c505.c +++ b/drivers/net/3c505.c @@ -1216,7 +1216,7 @@ static int elp_close(struct net_device *dev) static void elp_set_mc_list(struct net_device *dev) { elp_device *adapter = netdev_priv(dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int i; unsigned long flags; @@ -1231,8 +1231,9 @@ static void elp_set_mc_list(struct net_device *dev) adapter->tx_pcb.command = CMD_LOAD_MULTICAST_LIST; adapter->tx_pcb.length = 6 * netdev_mc_count(dev); i = 0; - netdev_for_each_mc_addr(dmi, dev) - memcpy(adapter->tx_pcb.data.multicast[i++], dmi->dmi_addr, 6); + netdev_for_each_mc_addr(ha, dev) + memcpy(adapter->tx_pcb.data.multicast[i++], + ha->addr, 6); adapter->got[CMD_LOAD_MULTICAST_LIST] = 0; if (!send_pcb(dev, &adapter->tx_pcb)) pr_err("%s: couldn't send set_multicast command\n", dev->name); diff --git a/drivers/net/3c523.c b/drivers/net/3c523.c index beed4fa10c6e..966cb12e1e0a 100644 --- a/drivers/net/3c523.c +++ b/drivers/net/3c523.c @@ -625,7 +625,7 @@ static int init586(struct net_device *dev) volatile struct iasetup_cmd_struct *ias_cmd; volatile struct tdr_cmd_struct *tdr_cmd; volatile struct mcsetup_cmd_struct *mc_cmd; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int num_addrs = netdev_mc_count(dev); ptr = (void *) ((char *) p->scb + sizeof(struct scb_struct)); @@ -788,8 +788,9 @@ static int init586(struct net_device *dev) mc_cmd->cmd_link = 0xffff; mc_cmd->mc_cnt = num_addrs * 6; i = 0; - netdev_for_each_mc_addr(dmi, dev) - memcpy((char *) mc_cmd->mc_list[i++], dmi->dmi_addr, 6); + netdev_for_each_mc_addr(ha, dev) + memcpy((char *) mc_cmd->mc_list[i++], + ha->addr, 6); p->scb->cbl_offset = make16(mc_cmd); p->scb->cmd = CUC_START; elmc_id_attn586(); diff --git a/drivers/net/3c527.c b/drivers/net/3c527.c index 5c07b147ec99..38395dfa4963 100644 --- a/drivers/net/3c527.c +++ b/drivers/net/3c527.c @@ -1533,7 +1533,7 @@ static void do_mc32_set_multicast_list(struct net_device *dev, int retry) { unsigned char block[62]; unsigned char *bp; - struct dev_mc_list *dmc; + struct netdev_hw_addr *ha; if(retry==0) lp->mc_list_valid = 0; @@ -1543,8 +1543,8 @@ static void do_mc32_set_multicast_list(struct net_device *dev, int retry) block[0]=netdev_mc_count(dev); bp=block+2; - netdev_for_each_mc_addr(dmc, dev) { - memcpy(bp, dmc->dmi_addr, 6); + netdev_for_each_mc_addr(ha, dev) { + memcpy(bp, ha->addr, 6); bp+=6; } if(mc32_command_nowait(dev, 2, block, diff --git a/drivers/net/7990.c b/drivers/net/7990.c index 4e9a5a20b6a6..818837d8ffe7 100644 --- a/drivers/net/7990.c +++ b/drivers/net/7990.c @@ -595,7 +595,7 @@ static void lance_load_multicast (struct net_device *dev) struct lance_private *lp = netdev_priv(dev); volatile struct lance_init_block *ib = lp->init_block; volatile u16 *mcast_table = (u16 *)&ib->filter; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *addrs; u32 crc; @@ -610,8 +610,8 @@ static void lance_load_multicast (struct net_device *dev) ib->filter [1] = 0; /* Add addresses */ - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; /* multicast address? */ if (!(*addrs & 1)) diff --git a/drivers/net/8139cp.c b/drivers/net/8139cp.c index 3d4406b16658..e4e2aef11017 100644 --- a/drivers/net/8139cp.c +++ b/drivers/net/8139cp.c @@ -909,11 +909,11 @@ static void __cp_set_rx_mode (struct net_device *dev) rx_mode = AcceptBroadcast | AcceptMulticast | AcceptMyPhys; mc_filter[1] = mc_filter[0] = 0xffffffff; } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; rx_mode = AcceptBroadcast | AcceptMyPhys; mc_filter[1] = mc_filter[0] = 0; - netdev_for_each_mc_addr(mclist, dev) { - int bit_nr = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, dev) { + int bit_nr = ether_crc(ETH_ALEN, ha->addr) >> 26; mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 31); rx_mode |= AcceptMulticast; diff --git a/drivers/net/8139too.c b/drivers/net/8139too.c index b4efc913978b..f61784c3c4a2 100644 --- a/drivers/net/8139too.c +++ b/drivers/net/8139too.c @@ -2502,11 +2502,11 @@ static void __set_rx_mode (struct net_device *dev) rx_mode = AcceptBroadcast | AcceptMulticast | AcceptMyPhys; mc_filter[1] = mc_filter[0] = 0xffffffff; } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; rx_mode = AcceptBroadcast | AcceptMyPhys; mc_filter[1] = mc_filter[0] = 0; - netdev_for_each_mc_addr(mclist, dev) { - int bit_nr = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, dev) { + int bit_nr = ether_crc(ETH_ALEN, ha->addr) >> 26; mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 31); rx_mode |= AcceptMulticast; diff --git a/drivers/net/82596.c b/drivers/net/82596.c index f94d17d78bb0..3a28b1f451d9 100644 --- a/drivers/net/82596.c +++ b/drivers/net/82596.c @@ -1542,7 +1542,7 @@ static void set_multicast_list(struct net_device *dev) } if (!netdev_mc_empty(dev)) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; unsigned char *cp; struct mc_cmd *cmd; @@ -1552,10 +1552,10 @@ static void set_multicast_list(struct net_device *dev) cmd->cmd.command = CmdMulticastList; cmd->mc_cnt = cnt * ETH_ALEN; cp = cmd->mc_addrs; - netdev_for_each_mc_addr(dmi, dev) { + netdev_for_each_mc_addr(ha, dev) { if (!cnt--) break; - memcpy(cp, dmi->dmi_addr, ETH_ALEN); + memcpy(cp, ha->addr, ETH_ALEN); if (i596_debug > 1) DEB(DEB_MULTI,printk(KERN_INFO "%s: Adding address %pM\n", dev->name, cp)); diff --git a/drivers/net/a2065.c b/drivers/net/a2065.c index bd4d829eca12..7cce643793c3 100644 --- a/drivers/net/a2065.c +++ b/drivers/net/a2065.c @@ -603,7 +603,7 @@ static void lance_load_multicast (struct net_device *dev) struct lance_private *lp = netdev_priv(dev); volatile struct lance_init_block *ib = lp->init_block; volatile u16 *mcast_table = (u16 *)&ib->filter; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *addrs; u32 crc; @@ -618,8 +618,8 @@ static void lance_load_multicast (struct net_device *dev) ib->filter [1] = 0; /* Add addresses */ - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; /* multicast address? */ if (!(*addrs & 1)) diff --git a/drivers/net/amd8111e.c b/drivers/net/amd8111e.c index b8a59d255b49..d002c764a26c 100644 --- a/drivers/net/amd8111e.c +++ b/drivers/net/amd8111e.c @@ -1377,7 +1377,7 @@ list to the device. */ static void amd8111e_set_multicast_list(struct net_device *dev) { - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; struct amd8111e_priv *lp = netdev_priv(dev); u32 mc_filter[2] ; int bit_num; @@ -1408,8 +1408,8 @@ static void amd8111e_set_multicast_list(struct net_device *dev) /* load all the multicast addresses in the logic filter */ lp->options |= OPTION_MULTICAST_ENABLE; mc_filter[1] = mc_filter[0] = 0; - netdev_for_each_mc_addr(mc_ptr, dev) { - bit_num = (ether_crc_le(ETH_ALEN, mc_ptr->dmi_addr) >> 26) & 0x3f; + netdev_for_each_mc_addr(ha, dev) { + bit_num = (ether_crc_le(ETH_ALEN, ha->addr) >> 26) & 0x3f; mc_filter[bit_num >> 5] |= 1 << (bit_num & 31); } amd8111e_writeq(*(u64*)mc_filter,lp->mmio+ LADRF); diff --git a/drivers/net/arm/am79c961a.c b/drivers/net/arm/am79c961a.c index f1f58c5e27bf..a4b5b08276f8 100644 --- a/drivers/net/arm/am79c961a.c +++ b/drivers/net/arm/am79c961a.c @@ -383,12 +383,12 @@ static void am79c961_setmulticastlist (struct net_device *dev) } else if (dev->flags & IFF_ALLMULTI) { memset(multi_hash, 0xff, sizeof(multi_hash)); } else { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; memset(multi_hash, 0x00, sizeof(multi_hash)); - netdev_for_each_mc_addr(dmi, dev) - am79c961_mc_hash(dmi->dmi_addr, multi_hash); + netdev_for_each_mc_addr(ha, dev) + am79c961_mc_hash(ha->addr, multi_hash); } spin_lock_irqsave(&priv->chip_lock, flags); diff --git a/drivers/net/arm/at91_ether.c b/drivers/net/arm/at91_ether.c index 8b23d5a175bf..f31e8b6cbf73 100644 --- a/drivers/net/arm/at91_ether.c +++ b/drivers/net/arm/at91_ether.c @@ -556,14 +556,14 @@ static int hash_get_index(__u8 *addr) */ static void at91ether_sethashtable(struct net_device *dev) { - struct dev_mc_list *curr; + struct netdev_hw_addr *ha; unsigned long mc_filter[2]; unsigned int bitnr; mc_filter[0] = mc_filter[1] = 0; - netdev_for_each_mc_addr(curr, dev) { - bitnr = hash_get_index(curr->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + bitnr = hash_get_index(ha->addr); mc_filter[bitnr >> 5] |= 1 << (bitnr & 31); } diff --git a/drivers/net/arm/ixp4xx_eth.c b/drivers/net/arm/ixp4xx_eth.c index 6e2ae1d06df1..f9d168775d06 100644 --- a/drivers/net/arm/ixp4xx_eth.c +++ b/drivers/net/arm/ixp4xx_eth.c @@ -735,7 +735,7 @@ static int eth_xmit(struct sk_buff *skb, struct net_device *dev) static void eth_set_mcast_list(struct net_device *dev) { struct port *port = netdev_priv(dev); - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; u8 diffs[ETH_ALEN], *addr; int i; @@ -748,11 +748,11 @@ static void eth_set_mcast_list(struct net_device *dev) memset(diffs, 0, ETH_ALEN); addr = NULL; - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { if (!addr) - addr = mclist->dmi_addr; /* first MAC address */ + addr = ha->addr; /* first MAC address */ for (i = 0; i < ETH_ALEN; i++) - diffs[i] |= addr[i] ^ mclist->dmi_addr[i]; + diffs[i] |= addr[i] ^ ha->addr[i]; } for (i = 0; i < ETH_ALEN; i++) { diff --git a/drivers/net/arm/ks8695net.c b/drivers/net/arm/ks8695net.c index e7810b74f396..6ec245c6394b 100644 --- a/drivers/net/arm/ks8695net.c +++ b/drivers/net/arm/ks8695net.c @@ -331,16 +331,16 @@ ks8695_init_partial_multicast(struct ks8695_priv *ksp, { u32 low, high; int i; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; i = 0; - netdev_for_each_mc_addr(dmi, ndev) { + netdev_for_each_mc_addr(ha, ndev) { /* Ran out of space in chip? */ BUG_ON(i == KS8695_NR_ADDRESSES); - low = (dmi->dmi_addr[2] << 24) | (dmi->dmi_addr[3] << 16) | - (dmi->dmi_addr[4] << 8) | (dmi->dmi_addr[5]); - high = (dmi->dmi_addr[0] << 8) | (dmi->dmi_addr[1]); + low = (ha->addr[2] << 24) | (ha->addr[3] << 16) | + (ha->addr[4] << 8) | (ha->addr[5]); + high = (ha->addr[0] << 8) | (ha->addr[1]); ks8695_writereg(ksp, KS8695_AAL_(i), low); ks8695_writereg(ksp, KS8695_AAH_(i), AAH_E | high); diff --git a/drivers/net/at1700.c b/drivers/net/at1700.c index 309843ab8869..e2a549a60e25 100644 --- a/drivers/net/at1700.c +++ b/drivers/net/at1700.c @@ -848,12 +848,12 @@ set_rx_mode(struct net_device *dev) memset(mc_filter, 0x00, sizeof(mc_filter)); outb(1, ioaddr + RX_MODE); /* Ignore almost all multicasts. */ } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { unsigned int bit = - ether_crc_le(ETH_ALEN, mclist->dmi_addr) >> 26; + ether_crc_le(ETH_ALEN, ha->addr) >> 26; mc_filter[bit >> 3] |= (1 << bit); } outb(0x02, ioaddr + RX_MODE); /* Use normal mode. */ diff --git a/drivers/net/atl1c/atl1c_main.c b/drivers/net/atl1c/atl1c_main.c index a5508e1b261b..3d7051135c3a 100644 --- a/drivers/net/atl1c/atl1c_main.c +++ b/drivers/net/atl1c/atl1c_main.c @@ -354,7 +354,7 @@ static void atl1c_set_multi(struct net_device *netdev) { struct atl1c_adapter *adapter = netdev_priv(netdev); struct atl1c_hw *hw = &adapter->hw; - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u32 mac_ctrl_data; u32 hash_value; @@ -377,8 +377,8 @@ static void atl1c_set_multi(struct net_device *netdev) AT_WRITE_REG_ARRAY(hw, REG_RX_HASH_TABLE, 1, 0); /* comoute mc addresses' hash value ,and put it into hash table */ - netdev_for_each_mc_addr(mc_ptr, netdev) { - hash_value = atl1c_hash_mc_addr(hw, mc_ptr->dmi_addr); + netdev_for_each_mc_addr(ha, netdev) { + hash_value = atl1c_hash_mc_addr(hw, ha->addr); atl1c_hash_set(hw, hash_value); } } diff --git a/drivers/net/atl1e/atl1e_main.c b/drivers/net/atl1e/atl1e_main.c index 7231b5779122..b6605d433e91 100644 --- a/drivers/net/atl1e/atl1e_main.c +++ b/drivers/net/atl1e/atl1e_main.c @@ -284,7 +284,7 @@ static void atl1e_set_multi(struct net_device *netdev) { struct atl1e_adapter *adapter = netdev_priv(netdev); struct atl1e_hw *hw = &adapter->hw; - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u32 mac_ctrl_data = 0; u32 hash_value; @@ -307,8 +307,8 @@ static void atl1e_set_multi(struct net_device *netdev) AT_WRITE_REG_ARRAY(hw, REG_RX_HASH_TABLE, 1, 0); /* comoute mc addresses' hash value ,and put it into hash table */ - netdev_for_each_mc_addr(mc_ptr, netdev) { - hash_value = atl1e_hash_mc_addr(hw, mc_ptr->dmi_addr); + netdev_for_each_mc_addr(ha, netdev) { + hash_value = atl1e_hash_mc_addr(hw, ha->addr); atl1e_hash_set(hw, hash_value); } } diff --git a/drivers/net/atlx/atl2.c b/drivers/net/atlx/atl2.c index 199f2c9ce749..078d9d1b427c 100644 --- a/drivers/net/atlx/atl2.c +++ b/drivers/net/atlx/atl2.c @@ -135,7 +135,7 @@ static void atl2_set_multi(struct net_device *netdev) { struct atl2_adapter *adapter = netdev_priv(netdev); struct atl2_hw *hw = &adapter->hw; - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u32 rctl; u32 hash_value; @@ -157,8 +157,8 @@ static void atl2_set_multi(struct net_device *netdev) ATL2_WRITE_REG_ARRAY(hw, REG_RX_HASH_TABLE, 1, 0); /* comoute mc addresses' hash value ,and put it into hash table */ - netdev_for_each_mc_addr(mc_ptr, netdev) { - hash_value = atl2_hash_mc_addr(hw, mc_ptr->dmi_addr); + netdev_for_each_mc_addr(ha, netdev) { + hash_value = atl2_hash_mc_addr(hw, ha->addr); atl2_hash_set(hw, hash_value); } } diff --git a/drivers/net/atlx/atlx.c b/drivers/net/atlx/atlx.c index 72f3306352e2..f979ea2d6d3c 100644 --- a/drivers/net/atlx/atlx.c +++ b/drivers/net/atlx/atlx.c @@ -123,7 +123,7 @@ static void atlx_set_multi(struct net_device *netdev) { struct atlx_adapter *adapter = netdev_priv(netdev); struct atlx_hw *hw = &adapter->hw; - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u32 rctl; u32 hash_value; @@ -144,8 +144,8 @@ static void atlx_set_multi(struct net_device *netdev) iowrite32(0, (hw->hw_addr + REG_RX_HASH_TABLE) + (1 << 2)); /* compute mc addresses' hash value ,and put it into hash table */ - netdev_for_each_mc_addr(mc_ptr, netdev) { - hash_value = atlx_hash_mc_addr(hw, mc_ptr->dmi_addr); + netdev_for_each_mc_addr(ha, netdev) { + hash_value = atlx_hash_mc_addr(hw, ha->addr); atlx_hash_set(hw, hash_value); } } diff --git a/drivers/net/atp.c b/drivers/net/atp.c index 6ad16205dc17..0d730c8329d4 100644 --- a/drivers/net/atp.c +++ b/drivers/net/atp.c @@ -883,11 +883,11 @@ static void set_rx_mode_8012(struct net_device *dev) memset(mc_filter, 0xff, sizeof(mc_filter)); new_mode = CMR2h_Normal; } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { - int filterbit = ether_crc_le(ETH_ALEN, mclist->dmi_addr) & 0x3f; + netdev_for_each_mc_addr(ha, dev) { + int filterbit = ether_crc_le(ETH_ALEN, ha->addr) & 0x3f; mc_filter[filterbit >> 5] |= 1 << (filterbit & 31); } new_mode = CMR2h_Normal; diff --git a/drivers/net/au1000_eth.c b/drivers/net/au1000_eth.c index 4da191b87b0d..29631593cc23 100644 --- a/drivers/net/au1000_eth.c +++ b/drivers/net/au1000_eth.c @@ -957,12 +957,12 @@ static void au1000_multicast_list(struct net_device *dev) aup->mac->control &= ~MAC_PROMISCUOUS; printk(KERN_INFO "%s: Pass all multicast\n", dev->name); } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; u32 mc_filter[2]; /* Multicast hash filter */ mc_filter[1] = mc_filter[0] = 0; - netdev_for_each_mc_addr(mclist, dev) - set_bit(ether_crc(ETH_ALEN, mclist->dmi_addr)>>26, + netdev_for_each_mc_addr(ha, dev) + set_bit(ether_crc(ETH_ALEN, ha->addr)>>26, (long *)mc_filter); aup->mac->multi_hash_high = mc_filter[1]; aup->mac->multi_hash_low = mc_filter[0]; diff --git a/drivers/net/b44.c b/drivers/net/b44.c index 332c60356285..b2c5fd7b63af 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -1680,15 +1680,15 @@ static struct net_device_stats *b44_get_stats(struct net_device *dev) static int __b44_load_mcast(struct b44 *bp, struct net_device *dev) { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; int i, num_ents; num_ents = min_t(int, netdev_mc_count(dev), B44_MCAST_TABLE_SIZE); i = 0; - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { if (i == num_ents) break; - __b44_cam_write(bp, mclist->dmi_addr, i++ + 1); + __b44_cam_write(bp, ha->addr, i++ + 1); } return i+1; } diff --git a/drivers/net/bcm63xx_enet.c b/drivers/net/bcm63xx_enet.c index 37eb8021de1c..51733404478e 100644 --- a/drivers/net/bcm63xx_enet.c +++ b/drivers/net/bcm63xx_enet.c @@ -603,7 +603,7 @@ static int bcm_enet_set_mac_address(struct net_device *dev, void *p) static void bcm_enet_set_multicast_list(struct net_device *dev) { struct bcm_enet_priv *priv; - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; u32 val; int i; @@ -631,14 +631,14 @@ static void bcm_enet_set_multicast_list(struct net_device *dev) } i = 0; - netdev_for_each_mc_addr(mc_list, dev) { + netdev_for_each_mc_addr(ha, dev) { u8 *dmi_addr; u32 tmp; if (i == 3) break; /* update perfect match registers */ - dmi_addr = mc_list->dmi_addr; + dmi_addr = ha->addr; tmp = (dmi_addr[2] << 24) | (dmi_addr[3] << 16) | (dmi_addr[4] << 8) | dmi_addr[5]; enet_writel(priv, tmp, ENET_PML_REG(i + 1)); diff --git a/drivers/net/benet/be_cmds.c b/drivers/net/benet/be_cmds.c index 9f53d9e86e2b..61a9afdb83f4 100644 --- a/drivers/net/benet/be_cmds.c +++ b/drivers/net/benet/be_cmds.c @@ -1159,13 +1159,13 @@ int be_cmd_multicast_set(struct be_adapter *adapter, u32 if_id, req->interface_id = if_id; if (netdev) { int i; - struct dev_mc_list *mc; + struct netdev_hw_addr *ha; req->num_mac = cpu_to_le16(netdev_mc_count(netdev)); i = 0; - netdev_for_each_mc_addr(mc, netdev) - memcpy(req->mac[i].byte, mc->dmi_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, netdev) + memcpy(req->mac[i].byte, ha->addr, ETH_ALEN); } else { req->promiscuous = 1; } diff --git a/drivers/net/bfin_mac.c b/drivers/net/bfin_mac.c index 587f93cf03f6..c488cea8f455 100644 --- a/drivers/net/bfin_mac.c +++ b/drivers/net/bfin_mac.c @@ -812,14 +812,14 @@ static void bfin_mac_timeout(struct net_device *dev) static void bfin_mac_multicast_hash(struct net_device *dev) { u32 emac_hashhi, emac_hashlo; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *addrs; u32 crc; emac_hashhi = emac_hashlo = 0; - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; /* skip non-multicast addresses */ if (!(*addrs & 1)) diff --git a/drivers/net/bmac.c b/drivers/net/bmac.c index 119468e76323..1245e9835765 100644 --- a/drivers/net/bmac.c +++ b/drivers/net/bmac.c @@ -971,7 +971,7 @@ bmac_remove_multi(struct net_device *dev, */ static void bmac_set_multicast(struct net_device *dev) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; struct bmac_data *bp = netdev_priv(dev); int num_addrs = netdev_mc_count(dev); unsigned short rx_cfg; @@ -1000,8 +1000,8 @@ static void bmac_set_multicast(struct net_device *dev) rx_cfg = bmac_rx_on(dev, 0, 0); XXDEBUG(("bmac: multi disabled, rx_cfg=%#08x\n", rx_cfg)); } else { - netdev_for_each_mc_addr(dmi, dev) - bmac_addhash(bp, dmi->dmi_addr); + netdev_for_each_mc_addr(ha, dev) + bmac_addhash(bp, ha->addr); bmac_update_hash_table_mask(dev, bp); rx_cfg = bmac_rx_on(dev, 1, 0); XXDEBUG(("bmac: multi enabled, rx_cfg=%#08x\n", rx_cfg)); @@ -1015,7 +1015,7 @@ static void bmac_set_multicast(struct net_device *dev) static void bmac_set_multicast(struct net_device *dev) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *addrs; int i; unsigned short rx_cfg; @@ -1039,8 +1039,8 @@ static void bmac_set_multicast(struct net_device *dev) for(i = 0; i < 4; i++) hash_table[i] = 0; - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; if(!(*addrs & 1)) continue; diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 381887ba677c..0b69ffb7951d 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -3544,7 +3544,6 @@ bnx2_set_rx_mode(struct net_device *dev) } else { /* Accept one or more multicast(s). */ - struct dev_mc_list *mclist; u32 mc_filter[NUM_MC_HASH_REGISTERS]; u32 regidx; u32 bit; @@ -3552,8 +3551,8 @@ bnx2_set_rx_mode(struct net_device *dev) memset(mc_filter, 0, 4 * NUM_MC_HASH_REGISTERS); - netdev_for_each_mc_addr(mclist, dev) { - crc = ether_crc_le(ETH_ALEN, mclist->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + crc = ether_crc_le(ETH_ALEN, ha->addr); bit = crc & 0xff; regidx = (bit & 0xe0) >> 5; bit &= 0x1f; diff --git a/drivers/net/bnx2x_main.c b/drivers/net/bnx2x_main.c index f4ea99d06c77..fa9275c2ef5c 100644 --- a/drivers/net/bnx2x_main.c +++ b/drivers/net/bnx2x_main.c @@ -11496,21 +11496,21 @@ static void bnx2x_set_rx_mode(struct net_device *dev) else { /* some multicasts */ if (CHIP_IS_E1(bp)) { int i, old, offset; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; struct mac_configuration_cmd *config = bnx2x_sp(bp, mcast_config); i = 0; - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { config->config_table[i]. cam_entry.msb_mac_addr = - swab16(*(u16 *)&mclist->dmi_addr[0]); + swab16(*(u16 *)&ha->addr[0]); config->config_table[i]. cam_entry.middle_mac_addr = - swab16(*(u16 *)&mclist->dmi_addr[2]); + swab16(*(u16 *)&ha->addr[2]); config->config_table[i]. cam_entry.lsb_mac_addr = - swab16(*(u16 *)&mclist->dmi_addr[4]); + swab16(*(u16 *)&ha->addr[4]); config->config_table[i].cam_entry.flags = cpu_to_le16(port); config->config_table[i]. @@ -11564,18 +11564,18 @@ static void bnx2x_set_rx_mode(struct net_device *dev) 0); } else { /* E1H */ /* Accept one or more multicasts */ - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; u32 mc_filter[MC_HASH_SIZE]; u32 crc, bit, regidx; int i; memset(mc_filter, 0, 4 * MC_HASH_SIZE); - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { DP(NETIF_MSG_IFUP, "Adding mcast MAC: %pM\n", - mclist->dmi_addr); + ha->addr); - crc = crc32c_le(0, mclist->dmi_addr, ETH_ALEN); + crc = crc32c_le(0, ha->addr, ETH_ALEN); bit = (crc >> 24) & 0xff; regidx = bit >> 5; bit &= 0x1f; diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index d6ae63b2cf00..22682f1c8473 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -761,32 +761,6 @@ static int bond_check_dev_link(struct bonding *bond, /*----------------------------- Multicast list ------------------------------*/ -/* - * Returns 0 if dmi1 and dmi2 are the same, non-0 otherwise - */ -static inline int bond_is_dmi_same(const struct dev_mc_list *dmi1, - const struct dev_mc_list *dmi2) -{ - return memcmp(dmi1->dmi_addr, dmi2->dmi_addr, dmi1->dmi_addrlen) == 0 && - dmi1->dmi_addrlen == dmi2->dmi_addrlen; -} - -/* - * returns dmi entry if found, NULL otherwise - */ -static struct dev_mc_list *bond_mc_list_find_dmi(struct dev_mc_list *dmi, - struct dev_mc_list *mc_list) -{ - struct dev_mc_list *idmi; - - for (idmi = mc_list; idmi; idmi = idmi->next) { - if (bond_is_dmi_same(dmi, idmi)) - return idmi; - } - - return NULL; -} - /* * Push the promiscuity flag down to appropriate slaves */ @@ -839,18 +813,18 @@ static int bond_set_allmulti(struct bonding *bond, int inc) * Add a Multicast address to slaves * according to mode */ -static void bond_mc_add(struct bonding *bond, void *addr, int alen) +static void bond_mc_add(struct bonding *bond, void *addr) { if (USES_PRIMARY(bond->params.mode)) { /* write lock already acquired */ if (bond->curr_active_slave) - dev_mc_add(bond->curr_active_slave->dev, addr, alen, 0); + dev_mc_add(bond->curr_active_slave->dev, addr); } else { struct slave *slave; int i; bond_for_each_slave(bond, slave, i) - dev_mc_add(slave->dev, addr, alen, 0); + dev_mc_add(slave->dev, addr); } } @@ -858,18 +832,17 @@ static void bond_mc_add(struct bonding *bond, void *addr, int alen) * Remove a multicast address from slave * according to mode */ -static void bond_mc_delete(struct bonding *bond, void *addr, int alen) +static void bond_mc_del(struct bonding *bond, void *addr) { if (USES_PRIMARY(bond->params.mode)) { /* write lock already acquired */ if (bond->curr_active_slave) - dev_mc_delete(bond->curr_active_slave->dev, addr, - alen, 0); + dev_mc_del(bond->curr_active_slave->dev, addr); } else { struct slave *slave; int i; bond_for_each_slave(bond, slave, i) { - dev_mc_delete(slave->dev, addr, alen, 0); + dev_mc_del(slave->dev, addr); } } } @@ -895,50 +868,6 @@ static void bond_resend_igmp_join_requests(struct bonding *bond) rcu_read_unlock(); } -/* - * Totally destroys the mc_list in bond - */ -static void bond_mc_list_destroy(struct bonding *bond) -{ - struct dev_mc_list *dmi; - - dmi = bond->mc_list; - while (dmi) { - bond->mc_list = dmi->next; - kfree(dmi); - dmi = bond->mc_list; - } - - bond->mc_list = NULL; -} - -/* - * Copy all the Multicast addresses from src to the bonding device dst - */ -static int bond_mc_list_copy(struct dev_mc_list *mc_list, struct bonding *bond, - gfp_t gfp_flag) -{ - struct dev_mc_list *dmi, *new_dmi; - - for (dmi = mc_list; dmi; dmi = dmi->next) { - new_dmi = kmalloc(sizeof(struct dev_mc_list), gfp_flag); - - if (!new_dmi) { - /* FIXME: Potential memory leak !!! */ - return -ENOMEM; - } - - new_dmi->next = bond->mc_list; - bond->mc_list = new_dmi; - new_dmi->dmi_addrlen = dmi->dmi_addrlen; - memcpy(new_dmi->dmi_addr, dmi->dmi_addr, dmi->dmi_addrlen); - new_dmi->dmi_users = dmi->dmi_users; - new_dmi->dmi_gusers = dmi->dmi_gusers; - } - - return 0; -} - /* * flush all members of flush->mc_list from device dev->mc_list */ @@ -946,16 +875,16 @@ static void bond_mc_list_flush(struct net_device *bond_dev, struct net_device *slave_dev) { struct bonding *bond = netdev_priv(bond_dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; - for (dmi = bond_dev->mc_list; dmi; dmi = dmi->next) - dev_mc_delete(slave_dev, dmi->dmi_addr, dmi->dmi_addrlen, 0); + netdev_for_each_mc_addr(ha, bond_dev) + dev_mc_del(slave_dev, ha->addr); if (bond->params.mode == BOND_MODE_8023AD) { /* del lacpdu mc addr from mc list */ u8 lacpdu_multicast[ETH_ALEN] = MULTICAST_LACPDU_ADDR; - dev_mc_delete(slave_dev, lacpdu_multicast, ETH_ALEN, 0); + dev_mc_del(slave_dev, lacpdu_multicast); } } @@ -969,7 +898,7 @@ static void bond_mc_list_flush(struct net_device *bond_dev, static void bond_mc_swap(struct bonding *bond, struct slave *new_active, struct slave *old_active) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; if (!USES_PRIMARY(bond->params.mode)) /* nothing to do - mc list is already up-to-date on @@ -984,9 +913,8 @@ static void bond_mc_swap(struct bonding *bond, struct slave *new_active, if (bond->dev->flags & IFF_ALLMULTI) dev_set_allmulti(old_active->dev, -1); - for (dmi = bond->dev->mc_list; dmi; dmi = dmi->next) - dev_mc_delete(old_active->dev, dmi->dmi_addr, - dmi->dmi_addrlen, 0); + netdev_for_each_mc_addr(ha, bond->dev) + dev_mc_del(old_active->dev, ha->addr); } if (new_active) { @@ -997,9 +925,8 @@ static void bond_mc_swap(struct bonding *bond, struct slave *new_active, if (bond->dev->flags & IFF_ALLMULTI) dev_set_allmulti(new_active->dev, 1); - for (dmi = bond->dev->mc_list; dmi; dmi = dmi->next) - dev_mc_add(new_active->dev, dmi->dmi_addr, - dmi->dmi_addrlen, 0); + netdev_for_each_mc_addr(ha, bond->dev) + dev_mc_add(new_active->dev, ha->addr); bond_resend_igmp_join_requests(bond); } } @@ -1406,7 +1333,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) struct bonding *bond = netdev_priv(bond_dev); const struct net_device_ops *slave_ops = slave_dev->netdev_ops; struct slave *new_slave = NULL; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; struct sockaddr addr; int link_reporting; int old_features = bond_dev->features; @@ -1492,7 +1419,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) /* Flush unicast and multicast addresses */ dev_uc_flush(bond_dev); - dev_addr_discard(bond_dev); + dev_mc_flush(bond_dev); if (slave_dev->type != ARPHRD_ETHER) bond_setup_by_slave(bond_dev, slave_dev); @@ -1601,9 +1528,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) netif_addr_lock_bh(bond_dev); /* upload master's mc_list to new slave */ - for (dmi = bond_dev->mc_list; dmi; dmi = dmi->next) - dev_mc_add(slave_dev, dmi->dmi_addr, - dmi->dmi_addrlen, 0); + netdev_for_each_mc_addr(ha, bond_dev) + dev_mc_add(slave_dev, ha->addr); netif_addr_unlock_bh(bond_dev); } @@ -1611,7 +1537,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) /* add lacpdu mc addr to mc list */ u8 lacpdu_multicast[ETH_ALEN] = MULTICAST_LACPDU_ADDR; - dev_mc_add(slave_dev, lacpdu_multicast, ETH_ALEN, 0); + dev_mc_add(slave_dev, lacpdu_multicast); } bond_add_vlans_on_slave(bond, slave_dev); @@ -3913,10 +3839,24 @@ static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd return res; } +static bool bond_addr_in_mc_list(unsigned char *addr, + struct netdev_hw_addr_list *list, + int addrlen) +{ + struct netdev_hw_addr *ha; + + netdev_hw_addr_list_for_each(ha, list) + if (!memcmp(ha->addr, addr, addrlen)) + return true; + + return false; +} + static void bond_set_multicast_list(struct net_device *bond_dev) { struct bonding *bond = netdev_priv(bond_dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; + bool found; /* * Do promisc before checking multicast_mode @@ -3951,20 +3891,25 @@ static void bond_set_multicast_list(struct net_device *bond_dev) bond->flags = bond_dev->flags; /* looking for addresses to add to slaves' mc list */ - for (dmi = bond_dev->mc_list; dmi; dmi = dmi->next) { - if (!bond_mc_list_find_dmi(dmi, bond->mc_list)) - bond_mc_add(bond, dmi->dmi_addr, dmi->dmi_addrlen); + netdev_for_each_mc_addr(ha, bond_dev) { + found = bond_addr_in_mc_list(ha->addr, &bond->mc_list, + bond_dev->addr_len); + if (!found) + bond_mc_add(bond, ha->addr); } /* looking for addresses to delete from slaves' list */ - for (dmi = bond->mc_list; dmi; dmi = dmi->next) { - if (!bond_mc_list_find_dmi(dmi, bond_dev->mc_list)) - bond_mc_delete(bond, dmi->dmi_addr, dmi->dmi_addrlen); + netdev_hw_addr_list_for_each(ha, &bond->mc_list) { + found = bond_addr_in_mc_list(ha->addr, &bond_dev->mc, + bond_dev->addr_len); + if (!found) + bond_mc_del(bond, ha->addr); } /* save master's multicast list */ - bond_mc_list_destroy(bond); - bond_mc_list_copy(bond_dev->mc_list, bond, GFP_ATOMIC); + __hw_addr_flush(&bond->mc_list); + __hw_addr_add_multiple(&bond->mc_list, &bond_dev->mc, + bond_dev->addr_len, NETDEV_HW_ADDR_T_MULTICAST); read_unlock(&bond->lock); } @@ -4534,9 +4479,7 @@ static void bond_uninit(struct net_device *bond_dev) if (bond->wq) destroy_workqueue(bond->wq); - netif_addr_lock_bh(bond_dev); - bond_mc_list_destroy(bond); - netif_addr_unlock_bh(bond_dev); + __hw_addr_flush(&bond->mc_list); } /*------------------------- Module initialization ---------------------------*/ @@ -4908,6 +4851,8 @@ static int bond_init(struct net_device *bond_dev) list_add_tail(&bond->bond_list, &bn->dev_list); bond_prepare_sysfs_group(bond); + + __hw_addr_init(&bond->mc_list); return 0; } diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h index 257a7a4dfce9..2aa336720591 100644 --- a/drivers/net/bonding/bonding.h +++ b/drivers/net/bonding/bonding.h @@ -202,7 +202,7 @@ struct bonding { char proc_file_name[IFNAMSIZ]; #endif /* CONFIG_PROC_FS */ struct list_head bond_list; - struct dev_mc_list *mc_list; + struct netdev_hw_addr_list mc_list; int (*xmit_hash_policy)(struct sk_buff *, int); __be32 master_ip; u16 flags; diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c index 9bd155e4111c..bd857a20a755 100644 --- a/drivers/net/cassini.c +++ b/drivers/net/cassini.c @@ -2957,20 +2957,20 @@ static void cas_process_mc_list(struct cas *cp) { u16 hash_table[16]; u32 crc; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int i = 1; memset(hash_table, 0, sizeof(hash_table)); - netdev_for_each_mc_addr(dmi, cp->dev) { + netdev_for_each_mc_addr(ha, cp->dev) { if (i <= CAS_MC_EXACT_MATCH_SIZE) { /* use the alternate mac address registers for the * first 15 multicast addresses */ - writel((dmi->dmi_addr[4] << 8) | dmi->dmi_addr[5], + writel((ha->addr[4] << 8) | ha->addr[5], cp->regs + REG_MAC_ADDRN(i*3 + 0)); - writel((dmi->dmi_addr[2] << 8) | dmi->dmi_addr[3], + writel((ha->addr[2] << 8) | ha->addr[3], cp->regs + REG_MAC_ADDRN(i*3 + 1)); - writel((dmi->dmi_addr[0] << 8) | dmi->dmi_addr[1], + writel((ha->addr[0] << 8) | ha->addr[1], cp->regs + REG_MAC_ADDRN(i*3 + 2)); i++; } @@ -2978,7 +2978,7 @@ static void cas_process_mc_list(struct cas *cp) /* use hw hash table for the next series of * multicast addresses */ - crc = ether_crc_le(ETH_ALEN, dmi->dmi_addr); + crc = ether_crc_le(ETH_ALEN, ha->addr); crc >>= 24; hash_table[crc >> 4] |= 1 << (15 - (crc & 0xf)); } diff --git a/drivers/net/chelsio/pm3393.c b/drivers/net/chelsio/pm3393.c index a6eb30a6e2b9..13fd9628db1d 100644 --- a/drivers/net/chelsio/pm3393.c +++ b/drivers/net/chelsio/pm3393.c @@ -376,12 +376,13 @@ static int pm3393_set_rx_mode(struct cmac *cmac, struct t1_rx_mode *rm) rx_mode |= SUNI1x10GEXP_BITMSK_RXXG_MHASH_EN; } else if (t1_rx_mode_mc_cnt(rm)) { /* Accept one or more multicast(s). */ - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int bit; u16 mc_filter[4] = { 0, }; - netdev_for_each_mc_addr(dmi, t1_get_netdev(rm)) { - bit = (ether_crc(ETH_ALEN, dmi->dmi_addr) >> 23) & 0x3f; /* bit[23:28] */ + netdev_for_each_mc_addr(ha, t1_get_netdev(rm)) { + /* bit[23:28] */ + bit = (ether_crc(ETH_ALEN, ha->addr) >> 23) & 0x3f; mc_filter[bit >> 4] |= 1 << (bit & 0xf); } pmwrite(cmac, SUNI1x10GEXP_REG_RXXG_MULTICAST_HASH_LOW, mc_filter[0]); diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c index 60777fd90b33..bdfff784645c 100644 --- a/drivers/net/cpmac.c +++ b/drivers/net/cpmac.c @@ -328,7 +328,7 @@ static int cpmac_config(struct net_device *dev, struct ifmap *map) static void cpmac_set_multicast_list(struct net_device *dev) { - struct dev_mc_list *iter; + struct netdev_hw_addr *ha; u8 tmp; u32 mbp, bit, hash[2] = { 0, }; struct cpmac_priv *priv = netdev_priv(dev); @@ -348,19 +348,19 @@ static void cpmac_set_multicast_list(struct net_device *dev) * cpmac uses some strange mac address hashing * (not crc32) */ - netdev_for_each_mc_addr(iter, dev) { + netdev_for_each_mc_addr(ha, dev) { bit = 0; - tmp = iter->dmi_addr[0]; + tmp = ha->addr[0]; bit ^= (tmp >> 2) ^ (tmp << 4); - tmp = iter->dmi_addr[1]; + tmp = ha->addr[1]; bit ^= (tmp >> 4) ^ (tmp << 2); - tmp = iter->dmi_addr[2]; + tmp = ha->addr[2]; bit ^= (tmp >> 6) ^ tmp; - tmp = iter->dmi_addr[3]; + tmp = ha->addr[3]; bit ^= (tmp >> 2) ^ (tmp << 4); - tmp = iter->dmi_addr[4]; + tmp = ha->addr[4]; bit ^= (tmp >> 4) ^ (tmp << 2); - tmp = iter->dmi_addr[5]; + tmp = ha->addr[5]; bit ^= (tmp >> 6) ^ tmp; bit &= 0x3f; hash[bit / 32] |= 1 << (bit % 32); diff --git a/drivers/net/cris/eth_v10.c b/drivers/net/cris/eth_v10.c index dd24aadb778c..59110bc119a8 100644 --- a/drivers/net/cris/eth_v10.c +++ b/drivers/net/cris/eth_v10.c @@ -1596,16 +1596,16 @@ set_multicast_list(struct net_device *dev) } else { /* MC mode, receive normal and MC packets */ char hash_ix; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *baddr; lo_bits = 0x00000000ul; hi_bits = 0x00000000ul; - netdev_for_each_mc_addr(dmi, dev) { + netdev_for_each_mc_addr(ha, dev) { /* Calculate the hash index for the GA registers */ hash_ix = 0; - baddr = dmi->dmi_addr; + baddr = ha->addr; hash_ix ^= (*baddr) & 0x3f; hash_ix ^= ((*baddr) >> 6) & 0x03; ++baddr; diff --git a/drivers/net/cxgb3/xgmac.c b/drivers/net/cxgb3/xgmac.c index c142a2132e9f..3af19a550372 100644 --- a/drivers/net/cxgb3/xgmac.c +++ b/drivers/net/cxgb3/xgmac.c @@ -311,16 +311,16 @@ int t3_mac_set_rx_mode(struct cmac *mac, struct net_device *dev) if (dev->flags & IFF_ALLMULTI) hash_lo = hash_hi = 0xffffffff; else { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int exact_addr_idx = mac->nucast; hash_lo = hash_hi = 0; - netdev_for_each_mc_addr(dmi, dev) + netdev_for_each_mc_addr(ha, dev) if (exact_addr_idx < EXACT_ADDR_FILTERS) set_addr_filter(mac, exact_addr_idx++, - dmi->dmi_addr); + ha->addr); else { - int hash = hash_hw_addr(dmi->dmi_addr); + int hash = hash_hw_addr(ha->addr); if (hash < 32) hash_lo |= (1 << hash); diff --git a/drivers/net/davinci_emac.c b/drivers/net/davinci_emac.c index 2b8edd2efbf6..1f9df5c6a75a 100644 --- a/drivers/net/davinci_emac.c +++ b/drivers/net/davinci_emac.c @@ -952,13 +952,14 @@ static void emac_dev_mcast_set(struct net_device *ndev) emac_add_mcast(priv, EMAC_ALL_MULTI_SET, NULL); } if (!netdev_mc_empty(ndev)) { - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; + mbp_enable = (mbp_enable | EMAC_MBP_RXMCAST); emac_add_mcast(priv, EMAC_ALL_MULTI_CLR, NULL); /* program multicast address list into EMAC hardware */ - netdev_for_each_mc_addr(mc_ptr, ndev) { + netdev_for_each_mc_addr(ha, ndev) { emac_add_mcast(priv, EMAC_MULTICAST_ADD, - (u8 *) mc_ptr->dmi_addr); + (u8 *) ha->addr); } } else { mbp_enable = (mbp_enable & ~EMAC_MBP_RXMCAST); diff --git a/drivers/net/declance.c b/drivers/net/declance.c index 8cf3cc6f20e2..fb3f0984c289 100644 --- a/drivers/net/declance.c +++ b/drivers/net/declance.c @@ -940,7 +940,7 @@ static void lance_load_multicast(struct net_device *dev) { struct lance_private *lp = netdev_priv(dev); volatile u16 *ib = (volatile u16 *)dev->mem_start; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *addrs; u32 crc; @@ -959,8 +959,8 @@ static void lance_load_multicast(struct net_device *dev) *lib_ptr(ib, filter[3], lp->type) = 0; /* Add addresses */ - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; /* multicast address? */ if (!(*addrs & 1)) diff --git a/drivers/net/defxx.c b/drivers/net/defxx.c index ed53a8d45f4e..e5667c55844e 100644 --- a/drivers/net/defxx.c +++ b/drivers/net/defxx.c @@ -2195,7 +2195,7 @@ static void dfx_ctl_set_multicast_list(struct net_device *dev) { DFX_board_t *bp = netdev_priv(dev); int i; /* used as index in for loop */ - struct dev_mc_list *dmi; /* ptr to multicast addr entry */ + struct netdev_hw_addr *ha; /* Enable LLC frame promiscuous mode, if necessary */ @@ -2241,9 +2241,9 @@ static void dfx_ctl_set_multicast_list(struct net_device *dev) /* Copy addresses to multicast address table, then update adapter CAM */ i = 0; - netdev_for_each_mc_addr(dmi, dev) + netdev_for_each_mc_addr(ha, dev) memcpy(&bp->mc_table[i++ * FDDI_K_ALEN], - dmi->dmi_addr, FDDI_K_ALEN); + ha->addr, FDDI_K_ALEN); if (dfx_ctl_update_cam(bp) != DFX_K_SUCCESS) { diff --git a/drivers/net/depca.c b/drivers/net/depca.c index 744c1928dfca..a88300a0d1e8 100644 --- a/drivers/net/depca.c +++ b/drivers/net/depca.c @@ -1272,7 +1272,7 @@ static void set_multicast_list(struct net_device *dev) static void SetMulticastFilter(struct net_device *dev) { struct depca_private *lp = netdev_priv(dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *addrs; int i, j, bit, byte; u16 hashcode; @@ -1287,8 +1287,8 @@ static void SetMulticastFilter(struct net_device *dev) lp->init_block.mcast_table[i] = 0; } /* Add multicast addresses */ - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; if ((*addrs & 0x01) == 1) { /* multicast address? */ crc = ether_crc(ETH_ALEN, addrs); hashcode = (crc & 1); /* hashcode is 6 LSb of CRC ... */ diff --git a/drivers/net/dl2k.c b/drivers/net/dl2k.c index b05bad829827..6579225dbd91 100644 --- a/drivers/net/dl2k.c +++ b/drivers/net/dl2k.c @@ -1132,14 +1132,14 @@ set_multicast (struct net_device *dev) /* Receive broadcast and multicast frames */ rx_mode = ReceiveBroadcast | ReceiveMulticast | ReceiveUnicast; } else if (!netdev_mc_empty(dev)) { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; /* Receive broadcast frames and multicast frames filtering by Hashtable */ rx_mode = ReceiveBroadcast | ReceiveMulticastHash | ReceiveUnicast; - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { int bit, index = 0; - int crc = ether_crc_le (ETH_ALEN, mclist->dmi_addr); + int crc = ether_crc_le(ETH_ALEN, ha->addr); /* The inverted high significant 6 bits of CRC are used as an index to hashtable */ for (bit = 0; bit < 6; bit++) diff --git a/drivers/net/dm9000.c b/drivers/net/dm9000.c index 1c67f1138ca7..989f2beb123b 100644 --- a/drivers/net/dm9000.c +++ b/drivers/net/dm9000.c @@ -724,7 +724,7 @@ static void dm9000_hash_table(struct net_device *dev) { board_info_t *db = netdev_priv(dev); - struct dev_mc_list *mcptr; + struct netdev_hw_addr *ha; int i, oft; u32 hash_val; u16 hash_table[4]; @@ -752,8 +752,8 @@ dm9000_hash_table(struct net_device *dev) rcr |= RCR_ALL; /* the multicast address in Hash Table : 64 bits */ - netdev_for_each_mc_addr(mcptr, dev) { - hash_val = ether_crc_le(6, mcptr->dmi_addr) & 0x3f; + netdev_for_each_mc_addr(ha, dev) { + hash_val = ether_crc_le(6, ha->addr) & 0x3f; hash_table[hash_val / 16] |= (u16) 1 << (hash_val % 16); } diff --git a/drivers/net/e100.c b/drivers/net/e100.c index c0cd57656681..3e8d0005540f 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -1545,16 +1545,16 @@ static int e100_hw_init(struct nic *nic) static void e100_multi(struct nic *nic, struct cb *cb, struct sk_buff *skb) { struct net_device *netdev = nic->netdev; - struct dev_mc_list *list; + struct netdev_hw_addr *ha; u16 i, count = min(netdev_mc_count(netdev), E100_MAX_MULTICAST_ADDRS); cb->command = cpu_to_le16(cb_multi); cb->u.multi.count = cpu_to_le16(count * ETH_ALEN); i = 0; - netdev_for_each_mc_addr(list, netdev) { + netdev_for_each_mc_addr(ha, netdev) { if (i == count) break; - memcpy(&cb->u.multi.addr[i++ * ETH_ALEN], &list->dmi_addr, + memcpy(&cb->u.multi.addr[i++ * ETH_ALEN], &ha->addr, ETH_ALEN); } } diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 8be6faee43e6..41330349b07a 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -2101,7 +2101,6 @@ static void e1000_set_rx_mode(struct net_device *netdev) struct e1000_hw *hw = &adapter->hw; struct netdev_hw_addr *ha; bool use_uc = false; - struct dev_addr_list *mc_ptr; u32 rctl; u32 hash_value; int i, rar_entries = E1000_RAR_ENTRIES; @@ -2161,17 +2160,17 @@ static void e1000_set_rx_mode(struct net_device *netdev) WARN_ON(i == rar_entries); - netdev_for_each_mc_addr(mc_ptr, netdev) { + netdev_for_each_mc_addr(ha, netdev) { if (i == rar_entries) { /* load any remaining addresses into the hash table */ u32 hash_reg, hash_bit, mta; - hash_value = e1000_hash_mc_addr(hw, mc_ptr->da_addr); + hash_value = e1000_hash_mc_addr(hw, ha->addr); hash_reg = (hash_value >> 5) & 0x7F; hash_bit = hash_value & 0x1F; mta = (1 << hash_bit); mcarray[hash_reg] |= mta; } else { - e1000_rar_set(hw, mc_ptr->da_addr, i++); + e1000_rar_set(hw, ha->addr, i++); } } diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c index 5304959ae1f7..02f7d20f3c80 100644 --- a/drivers/net/e1000e/netdev.c +++ b/drivers/net/e1000e/netdev.c @@ -2567,7 +2567,7 @@ static void e1000_set_multi(struct net_device *netdev) { struct e1000_adapter *adapter = netdev_priv(netdev); struct e1000_hw *hw = &adapter->hw; - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u8 *mta_list; u32 rctl; int i; @@ -2599,9 +2599,8 @@ static void e1000_set_multi(struct net_device *netdev) /* prepare a packed array of only addresses. */ i = 0; - netdev_for_each_mc_addr(mc_ptr, netdev) - memcpy(mta_list + (i++ * ETH_ALEN), - mc_ptr->dmi_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, netdev) + memcpy(mta_list + (i++ * ETH_ALEN), ha->addr, ETH_ALEN); e1000_update_mc_addr_list(hw, mta_list, i); kfree(mta_list); diff --git a/drivers/net/eepro.c b/drivers/net/eepro.c index 15d6266b80c3..b01e6997403b 100644 --- a/drivers/net/eepro.c +++ b/drivers/net/eepro.c @@ -1287,7 +1287,7 @@ set_multicast_list(struct net_device *dev) struct eepro_local *lp = netdev_priv(dev); short ioaddr = dev->base_addr; unsigned short mode; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int mc_count = netdev_mc_count(dev); if (dev->flags&(IFF_ALLMULTI|IFF_PROMISC) || mc_count > 63) @@ -1332,8 +1332,8 @@ set_multicast_list(struct net_device *dev) outw(0, ioaddr + IO_PORT); outw(6 * (mc_count + 1), ioaddr + IO_PORT); - netdev_for_each_mc_addr(dmi, dev) { - eaddrs = (unsigned short *) dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + eaddrs = (unsigned short *) ha->addr; outw(*eaddrs++, ioaddr + IO_PORT); outw(*eaddrs++, ioaddr + IO_PORT); outw(*eaddrs++, ioaddr + IO_PORT); diff --git a/drivers/net/eexpress.c b/drivers/net/eexpress.c index 7013dc8a6cbc..b3882fd8db6d 100644 --- a/drivers/net/eexpress.c +++ b/drivers/net/eexpress.c @@ -1576,7 +1576,7 @@ static void eexp_hw_init586(struct net_device *dev) static void eexp_setup_filter(struct net_device *dev) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; unsigned short ioaddr = dev->base_addr; int count = netdev_mc_count(dev); int i; @@ -1589,8 +1589,8 @@ static void eexp_setup_filter(struct net_device *dev) outw(CONF_NR_MULTICAST & ~31, ioaddr+SM_PTR); outw(6*count, ioaddr+SHADOW(CONF_NR_MULTICAST)); i = 0; - netdev_for_each_mc_addr(dmi, dev) { - unsigned short *data = (unsigned short *) dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + unsigned short *data = (unsigned short *) ha->addr; if (i == count) break; diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index b004eaba3d7b..b97411aaa77c 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -1966,7 +1966,7 @@ static void ehea_add_multicast_entry(struct ehea_port *port, u8 *mc_mac_addr) static void ehea_set_multicast_list(struct net_device *dev) { struct ehea_port *port = netdev_priv(dev); - struct dev_mc_list *k_mcl_entry; + struct netdev_hw_addr *ha; int ret; if (dev->flags & IFF_PROMISC) { @@ -1997,8 +1997,8 @@ static void ehea_set_multicast_list(struct net_device *dev) goto out; } - netdev_for_each_mc_addr(k_mcl_entry, dev) - ehea_add_multicast_entry(port, k_mcl_entry->dmi_addr); + netdev_for_each_mc_addr(ha, dev) + ehea_add_multicast_entry(port, ha->addr); } out: diff --git a/drivers/net/enic/enic_main.c b/drivers/net/enic/enic_main.c index 6d70c349c954..1232887c243d 100644 --- a/drivers/net/enic/enic_main.c +++ b/drivers/net/enic/enic_main.c @@ -822,7 +822,7 @@ static int enic_set_mac_addr(struct net_device *netdev, char *addr) static void enic_set_multicast_list(struct net_device *netdev) { struct enic *enic = netdev_priv(netdev); - struct dev_mc_list *list; + struct netdev_hw_addr *ha; int directed = 1; int multicast = (netdev->flags & IFF_MULTICAST) ? 1 : 0; int broadcast = (netdev->flags & IFF_BROADCAST) ? 1 : 0; @@ -852,10 +852,10 @@ static void enic_set_multicast_list(struct net_device *netdev) */ i = 0; - netdev_for_each_mc_addr(list, netdev) { + netdev_for_each_mc_addr(ha, netdev) { if (i == mc_count) break; - memcpy(mc_addr[i++], list->dmi_addr, ETH_ALEN); + memcpy(mc_addr[i++], ha->addr, ETH_ALEN); } for (i = 0; i < enic->mc_count; i++) { diff --git a/drivers/net/epic100.c b/drivers/net/epic100.c index 39c271b6be44..f6584a1ad3bc 100644 --- a/drivers/net/epic100.c +++ b/drivers/net/epic100.c @@ -1400,12 +1400,12 @@ static void set_rx_mode(struct net_device *dev) outl(0x0004, ioaddr + RxCtrl); return; } else { /* Never executed, for now. */ - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { unsigned int bit_nr = - ether_crc_le(ETH_ALEN, mclist->dmi_addr) & 0x3f; + ether_crc_le(ETH_ALEN, ha->addr) & 0x3f; mc_filter[bit_nr >> 3] |= (1 << bit_nr); } } diff --git a/drivers/net/ethoc.c b/drivers/net/ethoc.c index 209742304e20..f6be5aeaf94c 100644 --- a/drivers/net/ethoc.c +++ b/drivers/net/ethoc.c @@ -755,7 +755,7 @@ static void ethoc_set_multicast_list(struct net_device *dev) { struct ethoc *priv = netdev_priv(dev); u32 mode = ethoc_read(priv, MODER); - struct dev_mc_list *mc; + struct netdev_hw_addr *ha; u32 hash[2] = { 0, 0 }; /* set loopback mode if requested */ @@ -783,8 +783,8 @@ static void ethoc_set_multicast_list(struct net_device *dev) hash[0] = 0xffffffff; hash[1] = 0xffffffff; } else { - netdev_for_each_mc_addr(mc, dev) { - u32 crc = ether_crc(ETH_ALEN, mc->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + u32 crc = ether_crc(ETH_ALEN, ha->addr); int bit = (crc >> 26) & 0x3f; hash[bit >> 5] |= 1 << (bit & 0x1f); } diff --git a/drivers/net/ewrk3.c b/drivers/net/ewrk3.c index a2bade586886..11ba70f49971 100644 --- a/drivers/net/ewrk3.c +++ b/drivers/net/ewrk3.c @@ -1169,7 +1169,7 @@ static void set_multicast_list(struct net_device *dev) static void SetMulticastFilter(struct net_device *dev) { struct ewrk3_private *lp = netdev_priv(dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; u_long iobase = dev->base_addr; int i; char *addrs, bit, byte; @@ -1213,8 +1213,8 @@ static void SetMulticastFilter(struct net_device *dev) } /* Update table */ - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; if ((*addrs & 0x01) == 1) { /* multicast address? */ crc = ether_crc_le(ETH_ALEN, addrs); hashcode = crc & ((1 << 9) - 1); /* hashcode is 9 LSb of CRC */ diff --git a/drivers/net/fealnx.c b/drivers/net/fealnx.c index 9d5ad08a119f..e8a2705237bf 100644 --- a/drivers/net/fealnx.c +++ b/drivers/net/fealnx.c @@ -1792,12 +1792,12 @@ static void __set_rx_mode(struct net_device *dev) memset(mc_filter, 0xff, sizeof(mc_filter)); rx_mode = CR_W_AB | CR_W_AM; } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { unsigned int bit; - bit = (ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26) ^ 0x3F; + bit = (ether_crc(ETH_ALEN, ha->addr) >> 26) ^ 0x3F; mc_filter[bit >> 5] |= (1 << bit); } rx_mode = CR_W_AB | CR_W_AM; diff --git a/drivers/net/fec.c b/drivers/net/fec.c index 848eb1968abf..2b1651aee13f 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -954,7 +954,7 @@ fec_enet_close(struct net_device *dev) static void set_multicast_list(struct net_device *dev) { struct fec_enet_private *fep = netdev_priv(dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; unsigned int i, bit, data, crc, tmp; unsigned char hash; @@ -984,16 +984,16 @@ static void set_multicast_list(struct net_device *dev) writel(0, fep->hwp + FEC_GRP_HASH_TABLE_HIGH); writel(0, fep->hwp + FEC_GRP_HASH_TABLE_LOW); - netdev_for_each_mc_addr(dmi, dev) { + netdev_for_each_mc_addr(ha, dev) { /* Only support group multicast for now */ - if (!(dmi->dmi_addr[0] & 1)) + if (!(ha->addr[0] & 1)) continue; /* calculate crc32 value of mac address */ crc = 0xffffffff; - for (i = 0; i < dmi->dmi_addrlen; i++) { - data = dmi->dmi_addr[i]; + for (i = 0; i < dev->addr_len; i++) { + data = ha->addr[i]; for (bit = 0; bit < 8; bit++, data >>= 1) { crc = (crc >> 1) ^ (((crc ^ data) & 1) ? CRC32_POLY : 0); diff --git a/drivers/net/fec_mpc52xx.c b/drivers/net/fec_mpc52xx.c index 704155e2bdd1..667ba1391b9d 100644 --- a/drivers/net/fec_mpc52xx.c +++ b/drivers/net/fec_mpc52xx.c @@ -574,12 +574,12 @@ static void mpc52xx_fec_set_multicast_list(struct net_device *dev) out_be32(&fec->gaddr2, 0xffffffff); } else { u32 crc; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; u32 gaddr1 = 0x00000000; u32 gaddr2 = 0x00000000; - netdev_for_each_mc_addr(dmi, dev) { - crc = ether_crc_le(6, dmi->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, dev) { + crc = ether_crc_le(6, ha->addr) >> 26; if (crc >= 32) gaddr1 |= 1 << (crc-32); else diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c index ca05e5662029..6a2b64f0a7db 100644 --- a/drivers/net/forcedeth.c +++ b/drivers/net/forcedeth.c @@ -3103,12 +3103,14 @@ static void nv_set_multicast(struct net_device *dev) if (dev->flags & IFF_ALLMULTI) { alwaysOn[0] = alwaysOn[1] = alwaysOff[0] = alwaysOff[1] = 0; } else { - struct dev_mc_list *walk; + struct netdev_hw_addr *ha; - netdev_for_each_mc_addr(walk, dev) { + netdev_for_each_mc_addr(ha, dev) { + unsigned char *addr = ha->addr; u32 a, b; - a = le32_to_cpu(*(__le32 *) walk->dmi_addr); - b = le16_to_cpu(*(__le16 *) (&walk->dmi_addr[4])); + + a = le32_to_cpu(*(__le32 *) addr); + b = le16_to_cpu(*(__le16 *) (&addr[4])); alwaysOn[0] &= a; alwaysOff[0] &= ~a; alwaysOn[1] &= b; diff --git a/drivers/net/fs_enet/mac-fcc.c b/drivers/net/fs_enet/mac-fcc.c index cf4f674f9e2e..b3bad7c15d02 100644 --- a/drivers/net/fs_enet/mac-fcc.c +++ b/drivers/net/fs_enet/mac-fcc.c @@ -231,12 +231,12 @@ static void set_multicast_finish(struct net_device *dev) static void set_multicast_list(struct net_device *dev) { - struct dev_mc_list *pmc; + struct netdev_hw_addr *ha; if ((dev->flags & IFF_PROMISC) == 0) { set_multicast_start(dev); - netdev_for_each_mc_addr(pmc, dev) - set_multicast_one(dev, pmc->dmi_addr); + netdev_for_each_mc_addr(ha, dev) + set_multicast_one(dev, ha->addr); set_multicast_finish(dev); } else set_promiscuous_mode(dev); diff --git a/drivers/net/fs_enet/mac-fec.c b/drivers/net/fs_enet/mac-fec.c index cd2c6cca5f24..75974c6d201b 100644 --- a/drivers/net/fs_enet/mac-fec.c +++ b/drivers/net/fs_enet/mac-fec.c @@ -232,12 +232,12 @@ static void set_multicast_finish(struct net_device *dev) static void set_multicast_list(struct net_device *dev) { - struct dev_mc_list *pmc; + struct netdev_hw_addr *ha; if ((dev->flags & IFF_PROMISC) == 0) { set_multicast_start(dev); - netdev_for_each_mc_addr(pmc, dev) - set_multicast_one(dev, pmc->dmi_addr); + netdev_for_each_mc_addr(ha, dev) + set_multicast_one(dev, ha->addr); set_multicast_finish(dev); } else set_promiscuous_mode(dev); diff --git a/drivers/net/fs_enet/mac-scc.c b/drivers/net/fs_enet/mac-scc.c index c490a466cae1..0ab6a346a193 100644 --- a/drivers/net/fs_enet/mac-scc.c +++ b/drivers/net/fs_enet/mac-scc.c @@ -224,12 +224,12 @@ static void set_multicast_finish(struct net_device *dev) static void set_multicast_list(struct net_device *dev) { - struct dev_mc_list *pmc; + struct netdev_hw_addr *ha; if ((dev->flags & IFF_PROMISC) == 0) { set_multicast_start(dev); - netdev_for_each_mc_addr(pmc, dev) - set_multicast_one(dev, pmc->dmi_addr); + netdev_for_each_mc_addr(ha, dev) + set_multicast_one(dev, ha->addr); set_multicast_finish(dev); } else set_promiscuous_mode(dev); diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c index b6715553cf17..fdd26c2b1a2f 100644 --- a/drivers/net/gianfar.c +++ b/drivers/net/gianfar.c @@ -2797,7 +2797,7 @@ static void adjust_link(struct net_device *dev) * whenever dev->flags is changed */ static void gfar_set_multi(struct net_device *dev) { - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; struct gfar_private *priv = netdev_priv(dev); struct gfar __iomem *regs = priv->gfargrp[0].regs; u32 tempval; @@ -2870,13 +2870,12 @@ static void gfar_set_multi(struct net_device *dev) return; /* Parse the list, and set the appropriate bits */ - netdev_for_each_mc_addr(mc_ptr, dev) { + netdev_for_each_mc_addr(ha, dev) { if (idx < em_num) { - gfar_set_mac_for_addr(dev, idx, - mc_ptr->dmi_addr); + gfar_set_mac_for_addr(dev, idx, ha->addr); idx++; } else - gfar_set_hash_for_addr(dev, mc_ptr->dmi_addr); + gfar_set_hash_for_addr(dev, ha->addr); } } diff --git a/drivers/net/greth.c b/drivers/net/greth.c index c5e0d28a6de9..fbe6ab6b919b 100644 --- a/drivers/net/greth.c +++ b/drivers/net/greth.c @@ -988,7 +988,7 @@ static u32 greth_hash_get_index(__u8 *addr) static void greth_set_hash_filter(struct net_device *dev) { - struct dev_mc_list *curr; + struct netdev_hw_addr *ha; struct greth_private *greth = netdev_priv(dev); struct greth_regs *regs = (struct greth_regs *) greth->regs; u32 mc_filter[2]; @@ -996,8 +996,8 @@ static void greth_set_hash_filter(struct net_device *dev) mc_filter[0] = mc_filter[1] = 0; - netdev_for_each_mc_addr(curr, dev) { - bitnr = greth_hash_get_index(curr->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + bitnr = greth_hash_get_index(ha->addr); mc_filter[bitnr >> 5] |= 1 << (bitnr & 31); } diff --git a/drivers/net/hamachi.c b/drivers/net/hamachi.c index 373546dd0831..2bfcca6d180c 100644 --- a/drivers/net/hamachi.c +++ b/drivers/net/hamachi.c @@ -1858,12 +1858,12 @@ static void set_rx_mode(struct net_device *dev) /* Too many to match, or accept all multicasts. */ writew(0x000B, ioaddr + AddrMode); } else if (!netdev_mc_empty(dev)) { /* Must use the CAM filter. */ - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; int i = 0; - netdev_for_each_mc_addr(mclist, dev) { - writel(*(u32*)(mclist->dmi_addr), ioaddr + 0x100 + i*8); - writel(0x20000 | (*(u16*)&mclist->dmi_addr[4]), + netdev_for_each_mc_addr(ha, dev) { + writel(*(u32 *)(ha->addr), ioaddr + 0x100 + i*8); + writel(0x20000 | (*(u16 *)&ha->addr[4]), ioaddr + 0x104 + i*8); i++; } diff --git a/drivers/net/hp100.c b/drivers/net/hp100.c index b766a69bf0ca..86b2b4332341 100644 --- a/drivers/net/hp100.c +++ b/drivers/net/hp100.c @@ -2100,15 +2100,15 @@ static void hp100_set_multicast_list(struct net_device *dev) } else { int i, idx; u_char *addrs; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; memset(&lp->hash_bytes, 0x00, 8); #ifdef HP100_DEBUG printk("hp100: %s: computing hash filter - mc_count = %i\n", dev->name, netdev_mc_count(dev)); #endif - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; if ((*addrs & 0x01) == 0x01) { /* multicast address? */ #ifdef HP100_DEBUG printk("hp100: %s: multicast = %pM, ", diff --git a/drivers/net/ibm_newemac/core.c b/drivers/net/ibm_newemac/core.c index d8533a4ef825..40c78507ef16 100644 --- a/drivers/net/ibm_newemac/core.c +++ b/drivers/net/ibm_newemac/core.c @@ -388,18 +388,19 @@ static void emac_hash_mc(struct emac_instance *dev) const int regs = EMAC_XAHT_REGS(dev); u32 *gaht_base = emac_gaht_base(dev); u32 gaht_temp[regs]; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int i; DBG(dev, "hash_mc %d" NL, netdev_mc_count(dev->ndev)); memset(gaht_temp, 0, sizeof (gaht_temp)); - netdev_for_each_mc_addr(dmi, dev->ndev) { + netdev_for_each_mc_addr(ha, dev->ndev) { int slot, reg, mask; - DBG2(dev, "mc %pM" NL, dmi->dmi_addr); + DBG2(dev, "mc %pM" NL, ha->addr); - slot = EMAC_XAHT_CRC_TO_SLOT(dev, ether_crc(ETH_ALEN, dmi->dmi_addr)); + slot = EMAC_XAHT_CRC_TO_SLOT(dev, + ether_crc(ETH_ALEN, ha->addr)); reg = EMAC_XAHT_SLOT_TO_REG(dev, slot); mask = EMAC_XAHT_SLOT_TO_MASK(dev, slot); @@ -1176,7 +1177,7 @@ static int emac_open(struct net_device *ndev) netif_carrier_on(dev->ndev); /* Required for Pause packet support in EMAC */ - dev_mc_add(ndev, default_mcast_addr, sizeof(default_mcast_addr), 1); + dev_mc_add_global(ndev, default_mcast_addr); emac_configure(dev); mal_poll_add(dev->mal, &dev->commac); diff --git a/drivers/net/ibmlana.c b/drivers/net/ibmlana.c index b5d0f4e973f7..76949e08ee84 100644 --- a/drivers/net/ibmlana.c +++ b/drivers/net/ibmlana.c @@ -385,7 +385,7 @@ static void InitBoard(struct net_device *dev) int camcnt; camentry_t cams[16]; u32 cammask; - struct dev_mc_list *mcptr; + struct netdev_hw_addr *ha; u16 rcrval; /* reset the SONIC */ @@ -420,8 +420,8 @@ static void InitBoard(struct net_device *dev) /* start putting the multicast addresses into the CAM list. Stop if it is full. */ - netdev_for_each_mc_addr(mcptr, dev) { - putcam(cams, &camcnt, mcptr->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + putcam(cams, &camcnt, ha->addr); if (camcnt == 16) break; } @@ -479,7 +479,7 @@ static void InitBoard(struct net_device *dev) /* if still multicast addresses left or ALLMULTI is set, set the multicast enable bit */ - if ((dev->flags & IFF_ALLMULTI) || (mcptr != NULL)) + if ((dev->flags & IFF_ALLMULTI) || netdev_mc_count(dev) > camcnt) rcrval |= RCREG_AMC; /* promiscous mode ? */ diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index 0bc777bac9b4..f468590ed454 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -1072,7 +1072,7 @@ static void ibmveth_set_multicast_list(struct net_device *netdev) ibmveth_error_printk("h_multicast_ctrl rc=%ld when entering promisc mode\n", lpar_rc); } } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; /* clear the filter table & disable filtering */ lpar_rc = h_multicast_ctrl(adapter->vdev->unit_address, IbmVethMcastEnableRecv | @@ -1083,10 +1083,10 @@ static void ibmveth_set_multicast_list(struct net_device *netdev) ibmveth_error_printk("h_multicast_ctrl rc=%ld when attempting to clear filter table\n", lpar_rc); } /* add the addresses to the filter table */ - netdev_for_each_mc_addr(mclist, netdev) { + netdev_for_each_mc_addr(ha, netdev) { // add the multicast address to the filter table unsigned long mcast_addr = 0; - memcpy(((char *)&mcast_addr)+2, mclist->dmi_addr, 6); + memcpy(((char *)&mcast_addr)+2, ha->addr, 6); lpar_rc = h_multicast_ctrl(adapter->vdev->unit_address, IbmVethMcastAddFilter, mcast_addr); diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c index ea875709f053..78cc742e233f 100644 --- a/drivers/net/igb/igb_main.c +++ b/drivers/net/igb/igb_main.c @@ -2864,7 +2864,7 @@ static int igb_write_mc_addr_list(struct net_device *netdev) { struct igb_adapter *adapter = netdev_priv(netdev); struct e1000_hw *hw = &adapter->hw; - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u8 *mta_list; int i; @@ -2881,8 +2881,8 @@ static int igb_write_mc_addr_list(struct net_device *netdev) /* The shared function expects a packed array of only addresses. */ i = 0; - netdev_for_each_mc_addr(mc_ptr, netdev) - memcpy(mta_list + (i++ * ETH_ALEN), mc_ptr->dmi_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, netdev) + memcpy(mta_list + (i++ * ETH_ALEN), ha->addr, ETH_ALEN); igb_update_mc_addr_list(hw, mta_list, i); kfree(mta_list); diff --git a/drivers/net/igbvf/netdev.c b/drivers/net/igbvf/netdev.c index a77afd8a14bb..ea8abf5c1ef2 100644 --- a/drivers/net/igbvf/netdev.c +++ b/drivers/net/igbvf/netdev.c @@ -1399,7 +1399,7 @@ static void igbvf_set_multi(struct net_device *netdev) { struct igbvf_adapter *adapter = netdev_priv(netdev); struct e1000_hw *hw = &adapter->hw; - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u8 *mta_list = NULL; int i; @@ -1414,8 +1414,8 @@ static void igbvf_set_multi(struct net_device *netdev) /* prepare a packed array of only addresses. */ i = 0; - netdev_for_each_mc_addr(mc_ptr, netdev) - memcpy(mta_list + (i++ * ETH_ALEN), mc_ptr->dmi_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, netdev) + memcpy(mta_list + (i++ * ETH_ALEN), ha->addr, ETH_ALEN); hw->mac.ops.update_mc_addr_list(hw, mta_list, i, 0, 0); kfree(mta_list); diff --git a/drivers/net/ioc3-eth.c b/drivers/net/ioc3-eth.c index 70871b9b045a..57d873da9789 100644 --- a/drivers/net/ioc3-eth.c +++ b/drivers/net/ioc3-eth.c @@ -1664,7 +1664,7 @@ static int ioc3_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) static void ioc3_set_multicast_list(struct net_device *dev) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; struct ioc3_private *ip = netdev_priv(dev); struct ioc3 *ioc3 = ip->regs; u64 ehar = 0; @@ -1688,8 +1688,8 @@ static void ioc3_set_multicast_list(struct net_device *dev) ip->ehar_h = 0xffffffff; ip->ehar_l = 0xffffffff; } else { - netdev_for_each_mc_addr(dmi, dev) { - char *addr = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + char *addr = ha->addr; if (!(*addr & 1)) continue; diff --git a/drivers/net/ipg.c b/drivers/net/ipg.c index 0d7ad3f2d0f3..67cfc7d9d895 100644 --- a/drivers/net/ipg.c +++ b/drivers/net/ipg.c @@ -569,7 +569,7 @@ static int ipg_config_autoneg(struct net_device *dev) static void ipg_nic_set_multicast_list(struct net_device *dev) { void __iomem *ioaddr = ipg_ioaddr(dev); - struct dev_mc_list *mc_list_ptr; + struct netdev_hw_addr *ha; unsigned int hashindex; u32 hashtable[2]; u8 receivemode; @@ -608,9 +608,9 @@ static void ipg_nic_set_multicast_list(struct net_device *dev) hashtable[1] = 0x00000000; /* Cycle through all multicast addresses to filter. */ - netdev_for_each_mc_addr(mc_list_ptr, dev) { + netdev_for_each_mc_addr(ha, dev) { /* Calculate CRC result for each multicast address. */ - hashindex = crc32_le(0xffffffff, mc_list_ptr->dmi_addr, + hashindex = crc32_le(0xffffffff, ha->addr, ETH_ALEN); /* Use only the least significant 6 bits. */ diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c index e6e972d9b7ca..cd65b8629bcc 100644 --- a/drivers/net/iseries_veth.c +++ b/drivers/net/iseries_veth.c @@ -961,15 +961,15 @@ static void veth_set_multicast_list(struct net_device *dev) (netdev_mc_count(dev) > VETH_MAX_MCAST)) { port->promiscuous = 1; } else { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; port->promiscuous = 0; /* Update table */ port->num_mcast = 0; - netdev_for_each_mc_addr(dmi, dev) { - u8 *addr = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + u8 *addr = ha->addr; u64 xaddr = 0; if (addr[0] & 0x01) {/* multicast address? */ diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c index c9fef65cb98b..912dd1d5772c 100644 --- a/drivers/net/ixgb/ixgb_main.c +++ b/drivers/net/ixgb/ixgb_main.c @@ -1058,7 +1058,7 @@ ixgb_set_multi(struct net_device *netdev) { struct ixgb_adapter *adapter = netdev_priv(netdev); struct ixgb_hw *hw = &adapter->hw; - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u32 rctl; int i; @@ -1089,9 +1089,9 @@ ixgb_set_multi(struct net_device *netdev) IXGB_WRITE_REG(hw, RCTL, rctl); i = 0; - netdev_for_each_mc_addr(mc_ptr, netdev) + netdev_for_each_mc_addr(ha, netdev) memcpy(&mta[i++ * IXGB_ETH_LENGTH_OF_ADDRESS], - mc_ptr->dmi_addr, IXGB_ETH_LENGTH_OF_ADDRESS); + ha->addr, IXGB_ETH_LENGTH_OF_ADDRESS); ixgb_mc_addr_list_update(hw, mta, netdev_mc_count(netdev), 0); } diff --git a/drivers/net/ixgbe/ixgbe_common.c b/drivers/net/ixgbe/ixgbe_common.c index 4d1c3a429457..6eb5814ca7da 100644 --- a/drivers/net/ixgbe/ixgbe_common.c +++ b/drivers/net/ixgbe/ixgbe_common.c @@ -1494,7 +1494,7 @@ static void ixgbe_set_mta(struct ixgbe_hw *hw, u8 *mc_addr) s32 ixgbe_update_mc_addr_list_generic(struct ixgbe_hw *hw, struct net_device *netdev) { - struct dev_addr_list *dmi; + struct netdev_hw_addr *ha; u32 i; /* @@ -1510,9 +1510,9 @@ s32 ixgbe_update_mc_addr_list_generic(struct ixgbe_hw *hw, IXGBE_WRITE_REG(hw, IXGBE_MTA(i), 0); /* Add the new addresses */ - netdev_for_each_mc_addr(dmi, netdev) { + netdev_for_each_mc_addr(ha, netdev) { hw_dbg(hw, " Adding the multicast addresses:\n"); - ixgbe_set_mta(hw, dmi->dmi_addr); + ixgbe_set_mta(hw, ha->addr); } /* Enable mta */ diff --git a/drivers/net/ixgbevf/vf.c b/drivers/net/ixgbevf/vf.c index f457c52b5ed4..852e9c4fd934 100644 --- a/drivers/net/ixgbevf/vf.c +++ b/drivers/net/ixgbevf/vf.c @@ -259,7 +259,7 @@ static s32 ixgbevf_set_rar_vf(struct ixgbe_hw *hw, u32 index, u8 *addr, static s32 ixgbevf_update_mc_addr_list_vf(struct ixgbe_hw *hw, struct net_device *netdev) { - struct dev_addr_list *dmi; + struct netdev_hw_addr *ha; struct ixgbe_mbx_info *mbx = &hw->mbx; u32 msgbuf[IXGBE_VFMAILBOX_SIZE]; u16 *vector_list = (u16 *)&msgbuf[1]; @@ -281,10 +281,10 @@ static s32 ixgbevf_update_mc_addr_list_vf(struct ixgbe_hw *hw, msgbuf[0] |= cnt << IXGBE_VT_MSGINFO_SHIFT; i = 0; - netdev_for_each_mc_addr(dmi, netdev) { + netdev_for_each_mc_addr(ha, netdev) { if (i == cnt) break; - vector_list[i++] = ixgbevf_mta_vector(hw, dmi->dmi_addr); + vector_list[i++] = ixgbevf_mta_vector(hw, ha->addr); } mbx->ops.write_posted(hw, msgbuf, IXGBE_VFMAILBOX_SIZE); diff --git a/drivers/net/jme.c b/drivers/net/jme.c index c0b59a555384..10e816d2caff 100644 --- a/drivers/net/jme.c +++ b/drivers/net/jme.c @@ -2009,12 +2009,12 @@ jme_set_multi(struct net_device *netdev) } else if (netdev->flags & IFF_ALLMULTI) { jme->reg_rxmcs |= RXMCS_ALLMULFRAME; } else if (netdev->flags & IFF_MULTICAST) { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; int bit_nr; jme->reg_rxmcs |= RXMCS_MULFRAME | RXMCS_MULFILTERED; - netdev_for_each_mc_addr(mclist, netdev) { - bit_nr = ether_crc(ETH_ALEN, mclist->dmi_addr) & 0x3F; + netdev_for_each_mc_addr(ha, netdev) { + bit_nr = ether_crc(ETH_ALEN, ha->addr) & 0x3F; mc_hash[bit_nr >> 5] |= 1 << (bit_nr & 0x1F); } diff --git a/drivers/net/korina.c b/drivers/net/korina.c index edaedc7aa03f..26bf1b76b997 100644 --- a/drivers/net/korina.c +++ b/drivers/net/korina.c @@ -482,7 +482,7 @@ static void korina_multicast_list(struct net_device *dev) { struct korina_private *lp = netdev_priv(dev); unsigned long flags; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; u32 recognise = ETH_ARC_AB; /* always accept broadcasts */ int i; @@ -502,8 +502,8 @@ static void korina_multicast_list(struct net_device *dev) for (i = 0; i < 4; i++) hash_table[i] = 0; - netdev_for_each_mc_addr(dmi, dev) { - char *addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + char *addrs = ha->addr; if (!(*addrs & 1)) continue; diff --git a/drivers/net/ks8851.c b/drivers/net/ks8851.c index 66be4e449f02..4dcd61f81ec2 100644 --- a/drivers/net/ks8851.c +++ b/drivers/net/ks8851.c @@ -956,13 +956,13 @@ static void ks8851_set_rx_mode(struct net_device *dev) rxctrl.rxcr1 = (RXCR1_RXME | RXCR1_RXAE | RXCR1_RXPAFMA | RXCR1_RXMAFMA); } else if (dev->flags & IFF_MULTICAST && !netdev_mc_empty(dev)) { - struct dev_mc_list *mcptr; + struct netdev_hw_addr *ha; u32 crc; /* accept some multicast */ - netdev_for_each_mc_addr(mcptr, dev) { - crc = ether_crc(ETH_ALEN, mcptr->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + crc = ether_crc(ETH_ALEN, ha->addr); crc >>= (32 - 6); /* get top six bits */ rxctrl.mchash[crc >> 4] |= (1 << (crc & 0xf)); diff --git a/drivers/net/ks8851_mll.c b/drivers/net/ks8851_mll.c index ee3fe30b2ad7..70a3d98f4bd7 100644 --- a/drivers/net/ks8851_mll.c +++ b/drivers/net/ks8851_mll.c @@ -362,7 +362,6 @@ static u8 KS_DEFAULT_MAC_ADDRESS[] = { 0x00, 0x10, 0xA1, 0x86, 0x95, 0x11 }; #define MAX_MCAST_LST 32 #define HW_MCAST_SIZE 8 -#define MAC_ADDR_LEN 6 /** * union ks_tx_hdr - tx header data @@ -450,7 +449,7 @@ struct ks_net { u16 promiscuous; u16 all_mcast; u16 mcast_lst_size; - u8 mcast_lst[MAX_MCAST_LST][MAC_ADDR_LEN]; + u8 mcast_lst[MAX_MCAST_LST][ETH_ALEN]; u8 mcast_bits[HW_MCAST_SIZE]; u8 mac_addr[6]; u8 fid; @@ -1170,7 +1169,7 @@ static void ks_set_mcast(struct ks_net *ks, u16 mcast) static void ks_set_rx_mode(struct net_device *netdev) { struct ks_net *ks = netdev_priv(netdev); - struct dev_mc_list *ptr; + struct netdev_hw_addr *ha; /* Turn on/off promiscuous mode. */ if ((netdev->flags & IFF_PROMISC) == IFF_PROMISC) @@ -1187,13 +1186,12 @@ static void ks_set_rx_mode(struct net_device *netdev) if (netdev_mc_count(netdev) <= MAX_MCAST_LST) { int i = 0; - netdev_for_each_mc_addr(ptr, netdev) { - if (!(*ptr->dmi_addr & 1)) + netdev_for_each_mc_addr(ha, netdev) { + if (!(*ha->addr & 1)) continue; if (i >= MAX_MCAST_LST) break; - memcpy(ks->mcast_lst[i++], ptr->dmi_addr, - MAC_ADDR_LEN); + memcpy(ks->mcast_lst[i++], ha->addr, ETH_ALEN); } ks->mcast_lst_size = (u8)i; ks_set_grpaddr(ks); diff --git a/drivers/net/ksz884x.c b/drivers/net/ksz884x.c index b843bf7d8c17..348769521615 100644 --- a/drivers/net/ksz884x.c +++ b/drivers/net/ksz884x.c @@ -5763,7 +5763,7 @@ static void netdev_set_rx_mode(struct net_device *dev) struct dev_priv *priv = netdev_priv(dev); struct dev_info *hw_priv = priv->adapter; struct ksz_hw *hw = &hw_priv->hw; - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; int multicast = (dev->flags & IFF_ALLMULTI); dev_set_promiscuous(dev, priv, hw, (dev->flags & IFF_PROMISC)); @@ -5780,7 +5780,7 @@ static void netdev_set_rx_mode(struct net_device *dev) int i = 0; /* List too big to support so turn on all multicast mode. */ - if (dev->mc_count > MAX_MULTICAST_LIST) { + if (netdev_mc_count(dev) > MAX_MULTICAST_LIST) { if (MAX_MULTICAST_LIST != hw->multi_list_size) { hw->multi_list_size = MAX_MULTICAST_LIST; ++hw->all_multi; @@ -5789,13 +5789,12 @@ static void netdev_set_rx_mode(struct net_device *dev) return; } - netdev_for_each_mc_addr(mc_ptr, dev) { - if (!(*mc_ptr->dmi_addr & 1)) + netdev_for_each_mc_addr(ha, dev) { + if (!(*ha->addr & 1)) continue; if (i >= MAX_MULTICAST_LIST) break; - memcpy(hw->multi_list[i++], mc_ptr->dmi_addr, - MAC_ADDR_LEN); + memcpy(hw->multi_list[i++], ha->addr, MAC_ADDR_LEN); } hw->multi_list_size = (u8) i; hw_set_grp_addr(hw); diff --git a/drivers/net/lib82596.c b/drivers/net/lib82596.c index 443c39a3732f..fddaf921885e 100644 --- a/drivers/net/lib82596.c +++ b/drivers/net/lib82596.c @@ -1388,7 +1388,7 @@ static void set_multicast_list(struct net_device *dev) } if (!netdev_mc_empty(dev)) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; unsigned char *cp; struct mc_cmd *cmd; @@ -1396,10 +1396,10 @@ static void set_multicast_list(struct net_device *dev) cmd->cmd.command = SWAP16(CmdMulticastList); cmd->mc_cnt = SWAP16(netdev_mc_count(dev) * 6); cp = cmd->mc_addrs; - netdev_for_each_mc_addr(dmi, dev) { + netdev_for_each_mc_addr(ha, dev) { if (!cnt--) break; - memcpy(cp, dmi->dmi_addr, 6); + memcpy(cp, ha->addr, 6); if (i596_debug > 1) DEB(DEB_MULTI, printk(KERN_DEBUG diff --git a/drivers/net/lib8390.c b/drivers/net/lib8390.c index 56f66f485400..526dc9cbc3c6 100644 --- a/drivers/net/lib8390.c +++ b/drivers/net/lib8390.c @@ -905,10 +905,10 @@ static struct net_device_stats *__ei_get_stats(struct net_device *dev) static inline void make_mc_bits(u8 *bits, struct net_device *dev) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; - netdev_for_each_mc_addr(dmi, dev) { - u32 crc = ether_crc(ETH_ALEN, dmi->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + u32 crc = ether_crc(ETH_ALEN, ha->addr); /* * The 8390 uses the 6 most significant bits of the * CRC to index the multicast table. diff --git a/drivers/net/ll_temac_main.c b/drivers/net/ll_temac_main.c index a18e3485476e..30474d6b15c3 100644 --- a/drivers/net/ll_temac_main.c +++ b/drivers/net/ll_temac_main.c @@ -250,20 +250,20 @@ static void temac_set_multicast_list(struct net_device *ndev) temac_indirect_out32(lp, XTE_AFM_OFFSET, XTE_AFM_EPPRM_MASK); dev_info(&ndev->dev, "Promiscuous mode enabled.\n"); } else if (!netdev_mc_empty(ndev)) { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; i = 0; - netdev_for_each_mc_addr(mclist, ndev) { + netdev_for_each_mc_addr(ha, ndev) { if (i >= MULTICAST_CAM_TABLE_NUM) break; - multi_addr_msw = ((mclist->dmi_addr[3] << 24) | - (mclist->dmi_addr[2] << 16) | - (mclist->dmi_addr[1] << 8) | - (mclist->dmi_addr[0])); + multi_addr_msw = ((ha->addr[3] << 24) | + (ha->addr[2] << 16) | + (ha->addr[1] << 8) | + (ha->addr[0])); temac_indirect_out32(lp, XTE_MAW0_OFFSET, multi_addr_msw); - multi_addr_lsw = ((mclist->dmi_addr[5] << 8) | - (mclist->dmi_addr[4]) | (i << 16)); + multi_addr_lsw = ((ha->addr[5] << 8) | + (ha->addr[4]) | (i << 16)); temac_indirect_out32(lp, XTE_MAW1_OFFSET, multi_addr_lsw); i++; diff --git a/drivers/net/lp486e.c b/drivers/net/lp486e.c index 3e3cc04defd0..72379c5439dc 100644 --- a/drivers/net/lp486e.c +++ b/drivers/net/lp486e.c @@ -1256,7 +1256,7 @@ static void set_multicast_list(struct net_device *dev) { dev->name, netdev_mc_count(dev)); if (!netdev_mc_empty(dev)) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *cp; cmd = kmalloc(sizeof(struct i596_cmd) + 2 + netdev_mc_count(dev) * 6, GFP_ATOMIC); @@ -1267,8 +1267,8 @@ static void set_multicast_list(struct net_device *dev) { cmd->command = CmdMulticastList; *((unsigned short *) (cmd + 1)) = netdev_mc_count(dev) * 6; cp = ((char *)(cmd + 1))+2; - netdev_for_each_mc_addr(dmi, dev) { - memcpy(cp, dmi->dmi_addr, 6); + netdev_for_each_mc_addr(ha, dev) { + memcpy(cp, ha->addr, 6); cp += 6; } if (i596_debug & LOG_SRCDST) diff --git a/drivers/net/macb.c b/drivers/net/macb.c index c8a18a6203c8..eab121945d7c 100644 --- a/drivers/net/macb.c +++ b/drivers/net/macb.c @@ -882,15 +882,15 @@ static int hash_get_index(__u8 *addr) */ static void macb_sethashtable(struct net_device *dev) { - struct dev_mc_list *curr; + struct netdev_hw_addr *ha; unsigned long mc_filter[2]; unsigned int bitnr; struct macb *bp = netdev_priv(dev); mc_filter[0] = mc_filter[1] = 0; - netdev_for_each_mc_addr(curr, dev) { - bitnr = hash_get_index(curr->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + bitnr = hash_get_index(ha->addr); mc_filter[bitnr >> 5] |= 1 << (bitnr & 31); } diff --git a/drivers/net/mace.c b/drivers/net/mace.c index ab5f0bf6d1ae..2328a7399dcb 100644 --- a/drivers/net/mace.c +++ b/drivers/net/mace.c @@ -598,7 +598,7 @@ static void mace_set_multicast(struct net_device *dev) mp->maccc |= PROM; } else { unsigned char multicast_filter[8]; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; if (dev->flags & IFF_ALLMULTI) { for (i = 0; i < 8; i++) @@ -606,8 +606,8 @@ static void mace_set_multicast(struct net_device *dev) } else { for (i = 0; i < 8; i++) multicast_filter[i] = 0; - netdev_for_each_mc_addr(dmi, dev) { - crc = ether_crc_le(6, dmi->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + crc = ether_crc_le(6, ha->addr); i = crc >> 26; /* bit number in multicast_filter */ multicast_filter[i >> 3] |= 1 << (i & 7); } diff --git a/drivers/net/macmace.c b/drivers/net/macmace.c index 13ba8f4afb7e..8a50c67e5928 100644 --- a/drivers/net/macmace.c +++ b/drivers/net/macmace.c @@ -508,7 +508,7 @@ static void mace_set_multicast(struct net_device *dev) mb->maccc |= PROM; } else { unsigned char multicast_filter[8]; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; if (dev->flags & IFF_ALLMULTI) { for (i = 0; i < 8; i++) { @@ -517,8 +517,8 @@ static void mace_set_multicast(struct net_device *dev) } else { for (i = 0; i < 8; i++) multicast_filter[i] = 0; - netdev_for_each_mc_addr(dmi, dev) { - crc = ether_crc_le(6, dmi->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + crc = ether_crc_le(6, ha->addr); /* bit number in multicast_filter */ i = crc >> 26; multicast_filter[i >> 3] |= 1 << (i & 7); diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c index 5eb96fe6ec5c..455464223b43 100644 --- a/drivers/net/mlx4/en_netdev.c +++ b/drivers/net/mlx4/en_netdev.c @@ -168,7 +168,7 @@ static void mlx4_en_clear_list(struct net_device *dev) static void mlx4_en_cache_mclist(struct net_device *dev) { struct mlx4_en_priv *priv = netdev_priv(dev); - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; char *mc_addrs; int mc_addrs_cnt = netdev_mc_count(dev); int i; @@ -179,8 +179,8 @@ static void mlx4_en_cache_mclist(struct net_device *dev) return; } i = 0; - netdev_for_each_mc_addr(mclist, dev) - memcpy(mc_addrs + i++ * ETH_ALEN, mclist->dmi_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, dev) + memcpy(mc_addrs + i++ * ETH_ALEN, ha->addr, ETH_ALEN); priv->mc_addrs = mc_addrs; priv->mc_addrs_cnt = mc_addrs_cnt; } diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index c97b6e4365a9..d5ebe43b0e65 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -1769,7 +1769,7 @@ static void mv643xx_eth_program_multicast_filter(struct net_device *dev) struct mv643xx_eth_private *mp = netdev_priv(dev); u32 *mc_spec; u32 *mc_other; - struct dev_addr_list *addr; + struct netdev_hw_addr *ha; int i; if (dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) { @@ -1794,8 +1794,8 @@ oom: memset(mc_spec, 0, 0x100); memset(mc_other, 0, 0x100); - netdev_for_each_mc_addr(addr, dev) { - u8 *a = addr->da_addr; + netdev_for_each_mc_addr(ha, dev) { + u8 *a = ha->addr; u32 *table; int entry; diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c index e84dd3ee9c5a..6d33adf988dc 100644 --- a/drivers/net/myri10ge/myri10ge.c +++ b/drivers/net/myri10ge/myri10ge.c @@ -3001,7 +3001,7 @@ static void myri10ge_set_multicast_list(struct net_device *dev) { struct myri10ge_priv *mgp = netdev_priv(dev); struct myri10ge_cmd cmd; - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; __be32 data[2] = { 0, 0 }; int err; @@ -3038,8 +3038,8 @@ static void myri10ge_set_multicast_list(struct net_device *dev) } /* Walk the multicast list, and add each address */ - netdev_for_each_mc_addr(mc_list, dev) { - memcpy(data, &mc_list->dmi_addr, 6); + netdev_for_each_mc_addr(ha, dev) { + memcpy(data, &ha->addr, 6); cmd.data0 = ntohl(data[0]); cmd.data1 = ntohl(data[1]); err = myri10ge_send_cmd(mgp, MXGEFW_JOIN_MULTICAST_GROUP, @@ -3047,7 +3047,7 @@ static void myri10ge_set_multicast_list(struct net_device *dev) if (err != 0) { netdev_err(dev, "Failed MXGEFW_JOIN_MULTICAST_GROUP, error status:%d %pM\n", - err, mc_list->dmi_addr); + err, ha->addr); goto abort; } } diff --git a/drivers/net/natsemi.c b/drivers/net/natsemi.c index e52038783245..9250bf6573ec 100644 --- a/drivers/net/natsemi.c +++ b/drivers/net/natsemi.c @@ -2493,12 +2493,12 @@ static void __set_rx_mode(struct net_device *dev) rx_mode = RxFilterEnable | AcceptBroadcast | AcceptAllMulticast | AcceptMyPhys; } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; int i; memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { - int b = (ether_crc(ETH_ALEN, mclist->dmi_addr) >> 23) & 0x1ff; + netdev_for_each_mc_addr(ha, dev) { + int b = (ether_crc(ETH_ALEN, ha->addr) >> 23) & 0x1ff; mc_filter[b/8] |= (1 << (b & 0x07)); } rx_mode = RxFilterEnable | AcceptBroadcast diff --git a/drivers/net/netxen/netxen_nic_hw.c b/drivers/net/netxen/netxen_nic_hw.c index a945591298a8..781ca893ee0a 100644 --- a/drivers/net/netxen/netxen_nic_hw.c +++ b/drivers/net/netxen/netxen_nic_hw.c @@ -537,7 +537,7 @@ netxen_nic_set_mcast_addr(struct netxen_adapter *adapter, void netxen_p2_nic_set_multi(struct net_device *netdev) { struct netxen_adapter *adapter = netdev_priv(netdev); - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u8 null_addr[6]; int i; @@ -571,8 +571,8 @@ void netxen_p2_nic_set_multi(struct net_device *netdev) netxen_nic_enable_mcast_filter(adapter); i = 0; - netdev_for_each_mc_addr(mc_ptr, netdev) - netxen_nic_set_mcast_addr(adapter, i++, mc_ptr->dmi_addr); + netdev_for_each_mc_addr(ha, netdev) + netxen_nic_set_mcast_addr(adapter, i++, ha->addr); /* Clear out remaining addresses */ while (i < adapter->max_mc_count) @@ -680,7 +680,7 @@ static int nx_p3_nic_add_mac(struct netxen_adapter *adapter, void netxen_p3_nic_set_multi(struct net_device *netdev) { struct netxen_adapter *adapter = netdev_priv(netdev); - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u8 bcast_addr[ETH_ALEN] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }; u32 mode = VPORT_MISS_MODE_DROP; LIST_HEAD(del_list); @@ -707,8 +707,8 @@ void netxen_p3_nic_set_multi(struct net_device *netdev) } if (!netdev_mc_empty(netdev)) { - netdev_for_each_mc_addr(mc_ptr, netdev) - nx_p3_nic_add_mac(adapter, mc_ptr->dmi_addr, &del_list); + netdev_for_each_mc_addr(ha, netdev) + nx_p3_nic_add_mac(adapter, ha->addr, &del_list); } send_fw_cmd: diff --git a/drivers/net/ni52.c b/drivers/net/ni52.c index 05c29c2cef2a..a76fabe26292 100644 --- a/drivers/net/ni52.c +++ b/drivers/net/ni52.c @@ -596,7 +596,7 @@ static int init586(struct net_device *dev) struct iasetup_cmd_struct __iomem *ias_cmd; struct tdr_cmd_struct __iomem *tdr_cmd; struct mcsetup_cmd_struct __iomem *mc_cmd; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int num_addrs = netdev_mc_count(dev); ptr = p->scb + 1; @@ -725,8 +725,8 @@ static int init586(struct net_device *dev) writew(num_addrs * 6, &mc_cmd->mc_cnt); i = 0; - netdev_for_each_mc_addr(dmi, dev) - memcpy_toio(mc_cmd->mc_list[i++], dmi->dmi_addr, 6); + netdev_for_each_mc_addr(ha, dev) + memcpy_toio(mc_cmd->mc_list[i++], ha->addr, 6); writew(make16(mc_cmd), &p->scb->cbl_offset); writeb(CUC_START, &p->scb->cmd_cuc); diff --git a/drivers/net/niu.c b/drivers/net/niu.c index 0678f3106cbc..7b52c466cf48 100644 --- a/drivers/net/niu.c +++ b/drivers/net/niu.c @@ -6313,7 +6313,6 @@ 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, }; @@ -6365,8 +6364,8 @@ static void niu_set_rx_mode(struct net_device *dev) for (i = 0; i < 16; i++) hash[i] = 0xffff; } else if (!netdev_mc_empty(dev)) { - netdev_for_each_mc_addr(addr, dev) { - u32 crc = ether_crc_le(ETH_ALEN, addr->da_addr); + netdev_for_each_mc_addr(ha, dev) { + u32 crc = ether_crc_le(ETH_ALEN, ha->addr); crc >>= 24; hash[crc >> 4] |= (1 << (15 - (crc & 0xf))); diff --git a/drivers/net/octeon/octeon_mgmt.c b/drivers/net/octeon/octeon_mgmt.c index be368e5cbf75..ee894ed35f7f 100644 --- a/drivers/net/octeon/octeon_mgmt.c +++ b/drivers/net/octeon/octeon_mgmt.c @@ -474,7 +474,7 @@ static void octeon_mgmt_set_rx_filtering(struct net_device *netdev) unsigned int cam_mode = 1; /* 1 - Accept on CAM match */ unsigned int multicast_mode = 1; /* 1 - Reject all multicast. */ struct octeon_mgmt_cam_state cam_state; - struct dev_addr_list *list; + struct netdev_hw_addr *ha; struct list_head *pos; int available_cam_entries; @@ -510,8 +510,8 @@ static void octeon_mgmt_set_rx_filtering(struct net_device *netdev) } } if (multicast_mode == 0) { - netdev_for_each_mc_addr(list, netdev) - octeon_mgmt_cam_state_add(&cam_state, list->da_addr); + netdev_for_each_mc_addr(ha, netdev) + octeon_mgmt_cam_state_add(&cam_state, ha->addr); } diff --git a/drivers/net/pci-skeleton.c b/drivers/net/pci-skeleton.c index 36785853a149..dc3b4c7914fd 100644 --- a/drivers/net/pci-skeleton.c +++ b/drivers/net/pci-skeleton.c @@ -1813,12 +1813,12 @@ static void netdrv_set_rx_mode(struct net_device *dev) rx_mode = AcceptBroadcast | AcceptMulticast | AcceptMyPhys; mc_filter[1] = mc_filter[0] = 0xffffffff; } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; rx_mode = AcceptBroadcast | AcceptMulticast | AcceptMyPhys; mc_filter[1] = mc_filter[0] = 0; - netdev_for_each_mc_addr(mclist, dev) { - int bit_nr = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, dev) { + int bit_nr = ether_crc(ETH_ALEN, ha->addr) >> 26; mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 31); } diff --git a/drivers/net/pcmcia/axnet_cs.c b/drivers/net/pcmcia/axnet_cs.c index 09291e60d309..333cb3344d54 100644 --- a/drivers/net/pcmcia/axnet_cs.c +++ b/drivers/net/pcmcia/axnet_cs.c @@ -1623,11 +1623,11 @@ static struct net_device_stats *get_stats(struct net_device *dev) static inline void make_mc_bits(u8 *bits, struct net_device *dev) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; u32 crc; - netdev_for_each_mc_addr(dmi, dev) { - crc = ether_crc(ETH_ALEN, dmi->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + crc = ether_crc(ETH_ALEN, ha->addr); /* * The 8390 uses the 6 most significant bits of the * CRC to index the multicast table. diff --git a/drivers/net/pcmcia/fmvj18x_cs.c b/drivers/net/pcmcia/fmvj18x_cs.c index b9dc80b9d04a..6734f7d6da98 100644 --- a/drivers/net/pcmcia/fmvj18x_cs.c +++ b/drivers/net/pcmcia/fmvj18x_cs.c @@ -1196,11 +1196,11 @@ static void set_rx_mode(struct net_device *dev) memset(mc_filter, 0x00, sizeof(mc_filter)); outb(1, ioaddr + RX_MODE); /* Ignore almost all multicasts. */ } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { - unsigned int bit = ether_crc_le(ETH_ALEN, mclist->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, dev) { + unsigned int bit = ether_crc_le(ETH_ALEN, ha->addr) >> 26; mc_filter[bit >> 3] |= (1 << (bit & 7)); } outb(2, ioaddr + RX_MODE); /* Use normal mode. */ diff --git a/drivers/net/pcmcia/nmclan_cs.c b/drivers/net/pcmcia/nmclan_cs.c index c717b143f11a..c516c1996354 100644 --- a/drivers/net/pcmcia/nmclan_cs.c +++ b/drivers/net/pcmcia/nmclan_cs.c @@ -1475,7 +1475,7 @@ static void set_multicast_list(struct net_device *dev) { mace_private *lp = netdev_priv(dev); int adr[ETHER_ADDR_LEN] = {0}; /* Ethernet address */ - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; #ifdef PCMCIA_DEBUG { @@ -1495,8 +1495,8 @@ static void set_multicast_list(struct net_device *dev) if (num_addrs > 0) { /* Calculate multicast logical address filter */ memset(lp->multicast_ladrf, 0, MACE_LADRF_LEN); - netdev_for_each_mc_addr(dmi, dev) { - memcpy(adr, dmi->dmi_addr, ETHER_ADDR_LEN); + netdev_for_each_mc_addr(ha, dev) { + memcpy(adr, ha->addr, ETHER_ADDR_LEN); BuildLAF(lp->multicast_ladrf, adr); } } diff --git a/drivers/net/pcmcia/smc91c92_cs.c b/drivers/net/pcmcia/smc91c92_cs.c index 5adc662c4bfb..f45c626003a4 100644 --- a/drivers/net/pcmcia/smc91c92_cs.c +++ b/drivers/net/pcmcia/smc91c92_cs.c @@ -1616,12 +1616,12 @@ static void set_rx_mode(struct net_device *dev) rx_cfg_setting = RxStripCRC | RxEnable | RxAllMulti; else { if (!netdev_mc_empty(dev)) { - struct dev_mc_list *mc_addr; + struct netdev_hw_addr *ha; - netdev_for_each_mc_addr(mc_addr, dev) { - u_int position = ether_crc(6, mc_addr->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + u_int position = ether_crc(6, ha->addr); #ifndef final_version /* Verify multicast address. */ - if ((mc_addr->dmi_addr[0] & 1) == 0) + if ((ha->addr[0] & 1) == 0) continue; #endif multicast_table[position >> 29] |= 1 << ((position >> 26) & 7); diff --git a/drivers/net/pcmcia/xirc2ps_cs.c b/drivers/net/pcmcia/xirc2ps_cs.c index 4d1802e457be..656be931207a 100644 --- a/drivers/net/pcmcia/xirc2ps_cs.c +++ b/drivers/net/pcmcia/xirc2ps_cs.c @@ -1398,7 +1398,7 @@ static void set_addresses(struct net_device *dev) { unsigned int ioaddr = dev->base_addr; local_info_t *lp = netdev_priv(dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; struct set_address_info sa_info; int i; @@ -1413,10 +1413,10 @@ static void set_addresses(struct net_device *dev) set_address(&sa_info, dev->dev_addr); i = 0; - netdev_for_each_mc_addr(dmi, dev) { + netdev_for_each_mc_addr(ha, dev) { if (i++ == 9) break; - set_address(&sa_info, dmi->dmi_addr); + set_address(&sa_info, ha->addr); } while (i++ < 9) set_address(&sa_info, dev->dev_addr); diff --git a/drivers/net/pcnet32.c b/drivers/net/pcnet32.c index 084d78dd1637..a2254f749a9a 100644 --- a/drivers/net/pcnet32.c +++ b/drivers/net/pcnet32.c @@ -2590,7 +2590,7 @@ static void pcnet32_load_multicast(struct net_device *dev) struct pcnet32_private *lp = netdev_priv(dev); volatile struct pcnet32_init_block *ib = lp->init_block; volatile __le16 *mcast_table = (__le16 *)ib->filter; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; unsigned long ioaddr = dev->base_addr; char *addrs; int i; @@ -2611,8 +2611,8 @@ static void pcnet32_load_multicast(struct net_device *dev) ib->filter[1] = 0; /* Add addresses */ - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; /* multicast address? */ if (!(*addrs & 1)) diff --git a/drivers/net/ps3_gelic_net.c b/drivers/net/ps3_gelic_net.c index a602da6418c8..fed5df9c5ea1 100644 --- a/drivers/net/ps3_gelic_net.c +++ b/drivers/net/ps3_gelic_net.c @@ -546,7 +546,7 @@ out: void gelic_net_set_multi(struct net_device *netdev) { struct gelic_card *card = netdev_card(netdev); - struct dev_mc_list *mc; + struct netdev_hw_addr *ha; unsigned int i; uint8_t *p; u64 addr; @@ -580,9 +580,9 @@ void gelic_net_set_multi(struct net_device *netdev) } /* set multicast addresses */ - netdev_for_each_mc_addr(mc, netdev) { + netdev_for_each_mc_addr(ha, netdev) { addr = 0; - p = mc->dmi_addr; + p = ha->addr; for (i = 0; i < ETH_ALEN; i++) { addr <<= 8; addr |= *p++; diff --git a/drivers/net/qlcnic/qlcnic_hw.c b/drivers/net/qlcnic/qlcnic_hw.c index 14c999ab4222..9a1daa4ebe22 100644 --- a/drivers/net/qlcnic/qlcnic_hw.c +++ b/drivers/net/qlcnic/qlcnic_hw.c @@ -415,7 +415,7 @@ static int qlcnic_nic_add_mac(struct qlcnic_adapter *adapter, u8 *addr) void qlcnic_set_multi(struct net_device *netdev) { struct qlcnic_adapter *adapter = netdev_priv(netdev); - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; u8 bcast_addr[ETH_ALEN] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }; u32 mode = VPORT_MISS_MODE_DROP; @@ -434,8 +434,8 @@ void qlcnic_set_multi(struct net_device *netdev) } if (!netdev_mc_empty(netdev)) { - netdev_for_each_mc_addr(mc_ptr, netdev) { - qlcnic_nic_add_mac(adapter, mc_ptr->dmi_addr); + netdev_for_each_mc_addr(ha, netdev) { + qlcnic_nic_add_mac(adapter, ha->addr); } } diff --git a/drivers/net/qlge/qlge_main.c b/drivers/net/qlge/qlge_main.c index fd34f266c0a8..76df96813a7b 100644 --- a/drivers/net/qlge/qlge_main.c +++ b/drivers/net/qlge/qlge_main.c @@ -4207,7 +4207,7 @@ static struct net_device_stats *qlge_get_stats(struct net_device static void qlge_set_multicast_list(struct net_device *ndev) { struct ql_adapter *qdev = (struct ql_adapter *)netdev_priv(ndev); - struct dev_mc_list *mc_ptr; + struct netdev_hw_addr *ha; int i, status; status = ql_sem_spinlock(qdev, SEM_RT_IDX_MASK); @@ -4271,8 +4271,8 @@ static void qlge_set_multicast_list(struct net_device *ndev) if (status) goto exit; i = 0; - netdev_for_each_mc_addr(mc_ptr, ndev) { - if (ql_set_mac_addr_reg(qdev, (u8 *) mc_ptr->dmi_addr, + netdev_for_each_mc_addr(ha, ndev) { + if (ql_set_mac_addr_reg(qdev, (u8 *) ha->addr, MAC_ADDR_TYPE_MULTI_MAC, i)) { netif_err(qdev, hw, qdev->ndev, "Failed to loadmulticast address.\n"); diff --git a/drivers/net/r6040.c b/drivers/net/r6040.c index 15d5373dc8f3..f5a0e963e688 100644 --- a/drivers/net/r6040.c +++ b/drivers/net/r6040.c @@ -938,7 +938,7 @@ static void r6040_multicast_list(struct net_device *dev) u16 *adrp; u16 reg; unsigned long flags; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int i; /* MAC Address */ @@ -973,8 +973,8 @@ static void r6040_multicast_list(struct net_device *dev) for (i = 0; i < 4; i++) hash_table[i] = 0; - netdev_for_each_mc_addr(dmi, dev) { - char *addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + char *addrs = ha->addr; if (!(*addrs & 1)) continue; @@ -994,9 +994,9 @@ static void r6040_multicast_list(struct net_device *dev) } /* Multicast Address 1~4 case */ i = 0; - netdev_for_each_mc_addr(dmi, dev) { + netdev_for_each_mc_addr(ha, dev) { if (i < MCAST_MAX) { - adrp = (u16 *) dmi->dmi_addr; + adrp = (u16 *) ha->addr; iowrite16(adrp[0], ioaddr + MID_1L + 8 * i); iowrite16(adrp[1], ioaddr + MID_1M + 8 * i); iowrite16(adrp[2], ioaddr + MID_1H + 8 * i); diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index f7ffa5d8ffeb..64cd250f642d 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -4785,12 +4785,12 @@ static void rtl_set_rx_mode(struct net_device *dev) rx_mode = AcceptBroadcast | AcceptMulticast | AcceptMyPhys; mc_filter[1] = mc_filter[0] = 0xffffffff; } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; rx_mode = AcceptBroadcast | AcceptMyPhys; mc_filter[1] = mc_filter[0] = 0; - netdev_for_each_mc_addr(mclist, dev) { - int bit_nr = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, dev) { + int bit_nr = ether_crc(ETH_ALEN, ha->addr) >> 26; mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 31); rx_mode |= AcceptMulticast; } diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c index 2eb7f8a0d926..ad5a6a873b29 100644 --- a/drivers/net/s2io.c +++ b/drivers/net/s2io.c @@ -4964,7 +4964,7 @@ static struct net_device_stats *s2io_get_stats(struct net_device *dev) static void s2io_set_multicast(struct net_device *dev) { int i, j, prev_cnt; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; struct s2io_nic *sp = netdev_priv(dev); struct XENA_dev_config __iomem *bar0 = sp->bar0; u64 val64 = 0, multi_mac = 0x010203040506ULL, mask = @@ -5093,12 +5093,12 @@ static void s2io_set_multicast(struct net_device *dev) /* Create the new Rx filter list and update the same in H/W. */ i = 0; - netdev_for_each_mc_addr(mclist, dev) { - memcpy(sp->usr_addrs[i].addr, mclist->dmi_addr, + netdev_for_each_mc_addr(ha, dev) { + memcpy(sp->usr_addrs[i].addr, ha->addr, ETH_ALEN); mac_addr = 0; for (j = 0; j < ETH_ALEN; j++) { - mac_addr |= mclist->dmi_addr[j]; + mac_addr |= ha->addr[j]; mac_addr <<= 8; } mac_addr >>= 8; diff --git a/drivers/net/sb1250-mac.c b/drivers/net/sb1250-mac.c index 9944e5d662c0..f7de960ca06f 100644 --- a/drivers/net/sb1250-mac.c +++ b/drivers/net/sb1250-mac.c @@ -2112,7 +2112,7 @@ static void sbmac_setmulti(struct sbmac_softc *sc) uint64_t reg; void __iomem *port; int idx; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; struct net_device *dev = sc->sbm_dev; /* @@ -2161,10 +2161,10 @@ static void sbmac_setmulti(struct sbmac_softc *sc) * XXX if the table overflows */ idx = 1; /* skip station address */ - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { if (idx == MAC_ADDR_COUNT) break; - reg = sbmac_addr2reg(mclist->dmi_addr); + reg = sbmac_addr2reg(ha->addr); port = sc->sbm_base + R_MAC_ADDR_BASE+(idx * sizeof(uint64_t)); __raw_writeq(reg, port); idx++; diff --git a/drivers/net/sc92031.c b/drivers/net/sc92031.c index d87c4787fffa..1b3260588933 100644 --- a/drivers/net/sc92031.c +++ b/drivers/net/sc92031.c @@ -433,13 +433,13 @@ static void _sc92031_set_mar(struct net_device *dev) (dev->flags & IFF_ALLMULTI)) mar0 = mar1 = 0xffffffff; else if (dev->flags & IFF_MULTICAST) { - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; - netdev_for_each_mc_addr(mc_list, dev) { + netdev_for_each_mc_addr(ha, dev) { u32 crc; unsigned bit = 0; - crc = ~ether_crc(ETH_ALEN, mc_list->dmi_addr); + crc = ~ether_crc(ETH_ALEN, ha->addr); crc >>= 24; if (crc & 0x01) bit |= 0x02; diff --git a/drivers/net/sfc/efx.c b/drivers/net/sfc/efx.c index 88f2fb193abe..e07b82b266d9 100644 --- a/drivers/net/sfc/efx.c +++ b/drivers/net/sfc/efx.c @@ -1602,7 +1602,7 @@ static int efx_set_mac_address(struct net_device *net_dev, void *data) static void efx_set_multicast_list(struct net_device *net_dev) { struct efx_nic *efx = netdev_priv(net_dev); - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; union efx_multicast_hash *mc_hash = &efx->multicast_hash; u32 crc; int bit; @@ -1614,8 +1614,8 @@ static void efx_set_multicast_list(struct net_device *net_dev) memset(mc_hash, 0xff, sizeof(*mc_hash)); } else { memset(mc_hash, 0x00, sizeof(*mc_hash)); - netdev_for_each_mc_addr(mc_list, net_dev) { - crc = ether_crc_le(ETH_ALEN, mc_list->dmi_addr); + netdev_for_each_mc_addr(ha, net_dev) { + crc = ether_crc_le(ETH_ALEN, ha->addr); bit = crc & (EFX_MCAST_HASH_ENTRIES - 1); set_bit_le(bit, mc_hash->byte); } diff --git a/drivers/net/sis190.c b/drivers/net/sis190.c index 760d9e83a465..6a05b93ae6c9 100644 --- a/drivers/net/sis190.c +++ b/drivers/net/sis190.c @@ -848,13 +848,13 @@ static void sis190_set_rx_mode(struct net_device *dev) rx_mode = AcceptBroadcast | AcceptMulticast | AcceptMyPhys; mc_filter[1] = mc_filter[0] = 0xffffffff; } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; rx_mode = AcceptBroadcast | AcceptMyPhys; mc_filter[1] = mc_filter[0] = 0; - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { int bit_nr = - ether_crc(ETH_ALEN, mclist->dmi_addr) & 0x3f; + ether_crc(ETH_ALEN, ha->addr) & 0x3f; mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 31); rx_mode |= AcceptMulticast; } diff --git a/drivers/net/sis900.c b/drivers/net/sis900.c index bc7ff411aad5..6293592635be 100644 --- a/drivers/net/sis900.c +++ b/drivers/net/sis900.c @@ -2298,12 +2298,14 @@ static void set_rx_mode(struct net_device *net_dev) /* Accept Broadcast packet, destination address matchs our * MAC address, use Receive Filter to reject unwanted MCAST * packets */ - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; rx_mode = RFAAB; - netdev_for_each_mc_addr(mclist, net_dev) { - unsigned int bit_nr = - sis900_mcast_bitnr(mclist->dmi_addr, sis_priv->chipset_rev); + netdev_for_each_mc_addr(ha, net_dev) { + unsigned int bit_nr; + + bit_nr = sis900_mcast_bitnr(ha->addr, + sis_priv->chipset_rev); mc_filter[bit_nr >> 4] |= (1 << (bit_nr & 0xf)); } } diff --git a/drivers/net/skfp/skfddi.c b/drivers/net/skfp/skfddi.c index 1921a54ea995..7e5b9f310fb2 100644 --- a/drivers/net/skfp/skfddi.c +++ b/drivers/net/skfp/skfddi.c @@ -852,7 +852,7 @@ static void skfp_ctl_set_multicast_list(struct net_device *dev) static void skfp_ctl_set_multicast_list_wo_lock(struct net_device *dev) { struct s_smc *smc = netdev_priv(dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; /* Enable promiscuous mode, if necessary */ if (dev->flags & IFF_PROMISC) { @@ -876,13 +876,13 @@ static void skfp_ctl_set_multicast_list_wo_lock(struct net_device *dev) /* use exact filtering */ // point to first multicast addr - netdev_for_each_mc_addr(dmi, dev) { - mac_add_multicast(smc, - (struct fddi_addr *)dmi->dmi_addr, - 1); + netdev_for_each_mc_addr(ha, dev) { + mac_add_multicast(smc, + (struct fddi_addr *)ha->addr, + 1); pr_debug(KERN_INFO "ENABLE MC ADDRESS: %pMF\n", - dmi->dmi_addr); + ha->addr); } } else { // more MC addresses than HW supports diff --git a/drivers/net/skge.c b/drivers/net/skge.c index bd8c5e8413b1..de5ef3877baf 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -2917,7 +2917,7 @@ static void genesis_set_multicast(struct net_device *dev) struct skge_port *skge = netdev_priv(dev); struct skge_hw *hw = skge->hw; int port = skge->port; - struct dev_mc_list *list; + struct netdev_hw_addr *ha; u32 mode; u8 filter[8]; @@ -2937,8 +2937,8 @@ static void genesis_set_multicast(struct net_device *dev) skge->flow_status == FLOW_STAT_SYMMETRIC) genesis_add_filter(filter, pause_mc_addr); - netdev_for_each_mc_addr(list, dev) - genesis_add_filter(filter, list->dmi_addr); + netdev_for_each_mc_addr(ha, dev) + genesis_add_filter(filter, ha->addr); } xm_write32(hw, port, XM_MODE, mode); @@ -2956,7 +2956,7 @@ static void yukon_set_multicast(struct net_device *dev) struct skge_port *skge = netdev_priv(dev); struct skge_hw *hw = skge->hw; int port = skge->port; - struct dev_mc_list *list; + struct netdev_hw_addr *ha; int rx_pause = (skge->flow_status == FLOW_STAT_REM_SEND || skge->flow_status == FLOW_STAT_SYMMETRIC); u16 reg; @@ -2979,8 +2979,8 @@ static void yukon_set_multicast(struct net_device *dev) if (rx_pause) yukon_add_filter(filter, pause_mc_addr); - netdev_for_each_mc_addr(list, dev) - yukon_add_filter(filter, list->dmi_addr); + netdev_for_each_mc_addr(ha, dev) + yukon_add_filter(filter, ha->addr); } diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index f1c0ec308bfe..5b97edb7a35f 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -3627,7 +3627,7 @@ static void sky2_set_multicast(struct net_device *dev) struct sky2_port *sky2 = netdev_priv(dev); struct sky2_hw *hw = sky2->hw; unsigned port = sky2->port; - struct dev_mc_list *list; + struct netdev_hw_addr *ha; u16 reg; u8 filter[8]; int rx_pause; @@ -3651,8 +3651,8 @@ static void sky2_set_multicast(struct net_device *dev) if (rx_pause) sky2_add_filter(filter, pause_mc_addr); - netdev_for_each_mc_addr(list, dev) - sky2_add_filter(filter, list->dmi_addr); + netdev_for_each_mc_addr(ha, dev) + sky2_add_filter(filter, ha->addr); } gma_write16(hw, port, GM_MC_ADDR_H1, diff --git a/drivers/net/smc911x.c b/drivers/net/smc911x.c index 0f97b5af91e4..6278734c735f 100644 --- a/drivers/net/smc911x.c +++ b/drivers/net/smc911x.c @@ -1341,7 +1341,7 @@ static void smc911x_set_multicast_list(struct net_device *dev) * within that register. */ else if (!netdev_mc_empty(dev)) { - struct dev_mc_list *cur_addr; + struct netdev_hw_addr *ha; /* Set the Hash perfec mode */ mcr |= MAC_CR_HPFILT_; @@ -1349,19 +1349,16 @@ static void smc911x_set_multicast_list(struct net_device *dev) /* start with a table of all zeros: reject all */ memset(multicast_table, 0, sizeof(multicast_table)); - netdev_for_each_mc_addr(cur_addr, dev) { + netdev_for_each_mc_addr(ha, dev) { u32 position; - /* do we have a pointer here? */ - if (!cur_addr) - break; /* make sure this is a multicast address - shouldn't this be a given if we have it here ? */ - if (!(*cur_addr->dmi_addr & 1)) - continue; + if (!(*ha->addr & 1)) + continue; /* upper 6 bits are used as hash index */ - position = ether_crc(ETH_ALEN, cur_addr->dmi_addr)>>26; + position = ether_crc(ETH_ALEN, ha->addr)>>26; multicast_table[position>>5] |= 1 << (position&0x1f); } diff --git a/drivers/net/smc9194.c b/drivers/net/smc9194.c index ee1487791d55..9e5e9e792e86 100644 --- a/drivers/net/smc9194.c +++ b/drivers/net/smc9194.c @@ -417,7 +417,7 @@ static void smc_shutdown( int ioaddr ) /* - . Function: smc_setmulticast( int ioaddr, int count, dev_mc_list * adds ) + . Function: smc_setmulticast( int ioaddr, struct net_device *dev ) . Purpose: . This sets the internal hardware table to filter out unwanted multicast . packets before they take up memory. @@ -438,26 +438,23 @@ static void smc_setmulticast(int ioaddr, struct net_device *dev) { int i; unsigned char multicast_table[ 8 ]; - struct dev_mc_list *cur_addr; + struct netdev_hw_addr *ha; /* table for flipping the order of 3 bits */ unsigned char invert3[] = { 0, 4, 2, 6, 1, 5, 3, 7 }; /* start with a table of all zeros: reject all */ memset( multicast_table, 0, sizeof( multicast_table ) ); - netdev_for_each_mc_addr(cur_addr, dev) { + netdev_for_each_mc_addr(ha, dev) { int position; - /* do we have a pointer here? */ - if ( !cur_addr ) - break; /* make sure this is a multicast address - shouldn't this be a given if we have it here ? */ - if ( !( *cur_addr->dmi_addr & 1 ) ) + if (!(*ha->addr & 1)) continue; /* only use the low order bits */ - position = ether_crc_le(6, cur_addr->dmi_addr) & 0x3f; + position = ether_crc_le(6, ha->addr) & 0x3f; /* do some messy swapping to put the bit in the right spot */ multicast_table[invert3[position&7]] |= diff --git a/drivers/net/smc91x.c b/drivers/net/smc91x.c index 2e8133809dc0..20c756577631 100644 --- a/drivers/net/smc91x.c +++ b/drivers/net/smc91x.c @@ -1413,7 +1413,7 @@ static void smc_set_multicast_list(struct net_device *dev) * within that register. */ else if (!netdev_mc_empty(dev)) { - struct dev_mc_list *cur_addr; + struct netdev_hw_addr *ha; /* table for flipping the order of 3 bits */ static const unsigned char invert3[] = {0, 4, 2, 6, 1, 5, 3, 7}; @@ -1421,16 +1421,16 @@ static void smc_set_multicast_list(struct net_device *dev) /* start with a table of all zeros: reject all */ memset(multicast_table, 0, sizeof(multicast_table)); - netdev_for_each_mc_addr(cur_addr, dev) { + netdev_for_each_mc_addr(ha, dev) { int position; /* make sure this is a multicast address - shouldn't this be a given if we have it here ? */ - if (!(*cur_addr->dmi_addr & 1)) + if (!(*ha->addr & 1)) continue; /* only use the low order bits */ - position = crc32_le(~0, cur_addr->dmi_addr, 6) & 0x3f; + position = crc32_le(~0, ha->addr, 6) & 0x3f; /* do some messy swapping to put the bit in the right spot */ multicast_table[invert3[position&7]] |= diff --git a/drivers/net/smsc911x.c b/drivers/net/smsc911x.c index 4fd1d8b38788..d6e3a12d65d3 100644 --- a/drivers/net/smsc911x.c +++ b/drivers/net/smsc911x.c @@ -1383,13 +1383,13 @@ static void smsc911x_set_multicast_list(struct net_device *dev) /* Enabling specific multicast addresses */ unsigned int hash_high = 0; unsigned int hash_low = 0; - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; pdata->set_bits_mask = MAC_CR_HPFILT_; pdata->clear_bits_mask = (MAC_CR_PRMS_ | MAC_CR_MCPAS_); - netdev_for_each_mc_addr(mc_list, dev) { - unsigned int bitnum = smsc911x_hash(mc_list->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + unsigned int bitnum = smsc911x_hash(ha->addr); unsigned int mask = 0x01 << (bitnum & 0x1F); if (bitnum & 0x20) diff --git a/drivers/net/smsc9420.c b/drivers/net/smsc9420.c index 34fa10d8ad40..5409ec1a5fd7 100644 --- a/drivers/net/smsc9420.c +++ b/drivers/net/smsc9420.c @@ -1063,12 +1063,12 @@ static void smsc9420_set_multicast_list(struct net_device *dev) mac_cr |= MAC_CR_MCPAS_; mac_cr &= (~MAC_CR_HPFILT_); } else if (!netdev_mc_empty(dev)) { - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; u32 hash_lo = 0, hash_hi = 0; smsc_dbg(HW, "Multicast filter enabled"); - netdev_for_each_mc_addr(mc_list, dev) { - u32 bit_num = smsc9420_hash(mc_list->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + u32 bit_num = smsc9420_hash(ha->addr); u32 mask = 1 << (bit_num & 0x1F); if (bit_num & 0x20) diff --git a/drivers/net/sonic.c b/drivers/net/sonic.c index 287c251075e5..e5d67327d707 100644 --- a/drivers/net/sonic.c +++ b/drivers/net/sonic.c @@ -531,7 +531,7 @@ static void sonic_multicast_list(struct net_device *dev) { struct sonic_local *lp = netdev_priv(dev); unsigned int rcr; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; unsigned char *addr; int i; @@ -550,8 +550,8 @@ static void sonic_multicast_list(struct net_device *dev) netdev_mc_count(dev)); sonic_set_cam_enable(dev, 1); /* always enable our own address */ i = 1; - netdev_for_each_mc_addr(dmi, dev) { - addr = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addr = ha->addr; sonic_cda_put(dev, i, SONIC_CD_CAP0, addr[1] << 8 | addr[0]); sonic_cda_put(dev, i, SONIC_CD_CAP1, addr[3] << 8 | addr[2]); sonic_cda_put(dev, i, SONIC_CD_CAP2, addr[5] << 8 | addr[4]); diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 5ba9d989f8fc..c04c8f9a2d96 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -625,7 +625,7 @@ spider_net_get_multicast_hash(struct net_device *netdev, __u8 *addr) static void spider_net_set_multi(struct net_device *netdev) { - struct dev_mc_list *mc; + struct netdev_hw_addr *ha; u8 hash; int i; u32 reg; @@ -646,8 +646,8 @@ spider_net_set_multi(struct net_device *netdev) hash = spider_net_get_multicast_hash(netdev, netdev->broadcast); */ set_bit(0xfd, bitmask); - netdev_for_each_mc_addr(mc, netdev) { - hash = spider_net_get_multicast_hash(netdev, mc->dmi_addr); + netdev_for_each_mc_addr(ha, netdev) { + hash = spider_net_get_multicast_hash(netdev, ha->addr); set_bit(hash, bitmask); } diff --git a/drivers/net/starfire.c b/drivers/net/starfire.c index 6dfa69899019..8a6d27cdc0bd 100644 --- a/drivers/net/starfire.c +++ b/drivers/net/starfire.c @@ -1766,7 +1766,7 @@ static void set_rx_mode(struct net_device *dev) struct netdev_private *np = netdev_priv(dev); void __iomem *ioaddr = np->base; u32 rx_mode = MinVLANPrio; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; int i; #ifdef VLAN_SUPPORT @@ -1804,8 +1804,8 @@ static void set_rx_mode(struct net_device *dev) /* Use the 16 element perfect filter, skip first two entries. */ void __iomem *filter_addr = ioaddr + PerfFilterTable + 2 * 16; __be16 *eaddrs; - netdev_for_each_mc_addr(mclist, dev) { - eaddrs = (__be16 *)mclist->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + eaddrs = (__be16 *) ha->addr; writew(be16_to_cpu(eaddrs[2]), filter_addr); filter_addr += 4; writew(be16_to_cpu(eaddrs[1]), filter_addr); filter_addr += 4; writew(be16_to_cpu(eaddrs[0]), filter_addr); filter_addr += 8; @@ -1825,10 +1825,10 @@ static void set_rx_mode(struct net_device *dev) __le16 mc_filter[32] __attribute__ ((aligned(sizeof(long)))); /* Multicast hash filter */ memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { /* The chip uses the upper 9 CRC bits as index into the hash table */ - int bit_nr = ether_crc_le(ETH_ALEN, mclist->dmi_addr) >> 23; + int bit_nr = ether_crc_le(ETH_ALEN, ha->addr) >> 23; __le32 *fptr = (__le32 *) &mc_filter[(bit_nr >> 4) & ~1]; *fptr |= cpu_to_le32(1 << (bit_nr & 31)); diff --git a/drivers/net/stmmac/dwmac100.c b/drivers/net/stmmac/dwmac100.c index a183de29c39a..c0a1c9df3ac7 100644 --- a/drivers/net/stmmac/dwmac100.c +++ b/drivers/net/stmmac/dwmac100.c @@ -316,7 +316,7 @@ static void dwmac100_set_filter(struct net_device *dev) MAC_CONTROL_HO | MAC_CONTROL_HP); } else { u32 mc_filter[2]; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; /* Perfect filter mode for physical address and Hash filter for multicast */ @@ -325,11 +325,11 @@ static void dwmac100_set_filter(struct net_device *dev) MAC_CONTROL_IF | MAC_CONTROL_HO); memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { /* The upper 6 bits of the calculated CRC are used to * index the contens of the hash table */ int bit_nr = - ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26; + ether_crc(ETH_ALEN, ha->addr) >> 26; /* The most significant bit determines the register to * use (H/L) while the other 5 bits determine the bit * within the register. */ diff --git a/drivers/net/stmmac/dwmac1000_core.c b/drivers/net/stmmac/dwmac1000_core.c index a6538ae4694c..bf73fda6adfb 100644 --- a/drivers/net/stmmac/dwmac1000_core.c +++ b/drivers/net/stmmac/dwmac1000_core.c @@ -94,17 +94,17 @@ static void dwmac1000_set_filter(struct net_device *dev) writel(0xffffffff, ioaddr + GMAC_HASH_LOW); } else if (!netdev_mc_empty(dev)) { u32 mc_filter[2]; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; /* Hash filter for multicast */ value = GMAC_FRAME_FILTER_HMC; memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { /* The upper 6 bits of the calculated CRC are used to index the contens of the hash table */ int bit_nr = - bitrev32(~crc32_le(~0, mclist->dmi_addr, 6)) >> 26; + bitrev32(~crc32_le(~0, ha->addr, 6)) >> 26; /* The most significant bit determines the register to * use (H/L) while the other 5 bits determine the bit * within the register. */ diff --git a/drivers/net/sun3_82586.c b/drivers/net/sun3_82586.c index 2f6a760e5f21..2cfa065c39c0 100644 --- a/drivers/net/sun3_82586.c +++ b/drivers/net/sun3_82586.c @@ -413,7 +413,7 @@ static int init586(struct net_device *dev) volatile struct iasetup_cmd_struct *ias_cmd; volatile struct tdr_cmd_struct *tdr_cmd; volatile struct mcsetup_cmd_struct *mc_cmd; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int num_addrs=netdev_mc_count(dev); ptr = (void *) ((char *)p->scb + sizeof(struct scb_struct)); @@ -537,9 +537,9 @@ static int init586(struct net_device *dev) mc_cmd->mc_cnt = swab16(num_addrs * 6); i = 0; - netdev_for_each_mc_addr(dmi, dev) + netdev_for_each_mc_addr(ha, dev) memcpy((char *) mc_cmd->mc_list[i++], - dmi->dmi_addr, ETH_ALEN); + ha->addr, ETH_ALEN); p->scb->cbl_offset = make16(mc_cmd); p->scb->cmd_cuc = CUC_START; diff --git a/drivers/net/sunbmac.c b/drivers/net/sunbmac.c index a0bd361d5eca..4dd159b4116b 100644 --- a/drivers/net/sunbmac.c +++ b/drivers/net/sunbmac.c @@ -999,7 +999,7 @@ static void bigmac_set_multicast(struct net_device *dev) { struct bigmac *bp = netdev_priv(dev); void __iomem *bregs = bp->bregs; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *addrs; int i; u32 tmp, crc; @@ -1028,8 +1028,8 @@ static void bigmac_set_multicast(struct net_device *dev) for (i = 0; i < 4; i++) hash_table[i] = 0; - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; if (!(*addrs & 1)) continue; diff --git a/drivers/net/sundance.c b/drivers/net/sundance.c index a855934dfc3b..47e8cce10d1e 100644 --- a/drivers/net/sundance.c +++ b/drivers/net/sundance.c @@ -1523,13 +1523,13 @@ static void set_rx_mode(struct net_device *dev) memset(mc_filter, 0xff, sizeof(mc_filter)); rx_mode = AcceptBroadcast | AcceptMulticast | AcceptMyPhys; } else if (!netdev_mc_empty(dev)) { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; int bit; int index; int crc; memset (mc_filter, 0, sizeof (mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { - crc = ether_crc_le (ETH_ALEN, mclist->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + crc = ether_crc_le(ETH_ALEN, ha->addr); for (index=0, bit=0; bit < 6; bit++, crc <<= 1) if (crc & 0x80000000) index |= 1 << bit; mc_filter[index/16] |= (1 << (index % 16)); diff --git a/drivers/net/sungem.c b/drivers/net/sungem.c index 70196bc5fe61..58a27541ae39 100644 --- a/drivers/net/sungem.c +++ b/drivers/net/sungem.c @@ -1846,12 +1846,12 @@ static u32 gem_setup_multicast(struct gem *gp) } else { u16 hash_table[16]; u32 crc; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int i; memset(hash_table, 0, sizeof(hash_table)); - netdev_for_each_mc_addr(dmi, gp->dev) { - char *addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, gp->dev) { + char *addrs = ha->addr; if (!(*addrs & 1)) continue; diff --git a/drivers/net/sunhme.c b/drivers/net/sunhme.c index b17dbb11bd67..20deb14e98ff 100644 --- a/drivers/net/sunhme.c +++ b/drivers/net/sunhme.c @@ -1523,13 +1523,13 @@ static int happy_meal_init(struct happy_meal *hp) hme_write32(hp, bregs + BMAC_HTABLE3, 0xffff); } else if ((hp->dev->flags & IFF_PROMISC) == 0) { u16 hash_table[4]; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *addrs; u32 crc; memset(hash_table, 0, sizeof(hash_table)); - netdev_for_each_mc_addr(dmi, hp->dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, hp->dev) { + addrs = ha->addr; if (!(*addrs & 1)) continue; @@ -2362,7 +2362,7 @@ static void happy_meal_set_multicast(struct net_device *dev) { struct happy_meal *hp = netdev_priv(dev); void __iomem *bregs = hp->bigmacregs; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *addrs; u32 crc; @@ -2380,8 +2380,8 @@ static void happy_meal_set_multicast(struct net_device *dev) u16 hash_table[4]; memset(hash_table, 0, sizeof(hash_table)); - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; if (!(*addrs & 1)) continue; diff --git a/drivers/net/sunlance.c b/drivers/net/sunlance.c index d7c73f478ef5..674570584bd5 100644 --- a/drivers/net/sunlance.c +++ b/drivers/net/sunlance.c @@ -1170,7 +1170,7 @@ static int lance_start_xmit(struct sk_buff *skb, struct net_device *dev) static void lance_load_multicast(struct net_device *dev) { struct lance_private *lp = netdev_priv(dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; char *addrs; u32 crc; u32 val; @@ -1195,8 +1195,8 @@ static void lance_load_multicast(struct net_device *dev) return; /* Add addresses */ - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; /* multicast address? */ if (!(*addrs & 1)) diff --git a/drivers/net/sunqe.c b/drivers/net/sunqe.c index be637dce944c..239f09772191 100644 --- a/drivers/net/sunqe.c +++ b/drivers/net/sunqe.c @@ -627,7 +627,7 @@ static int qe_start_xmit(struct sk_buff *skb, struct net_device *dev) static void qe_set_multicast(struct net_device *dev) { struct sunqe *qep = netdev_priv(dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; u8 new_mconfig = qep->mconfig; char *addrs; int i; @@ -651,8 +651,8 @@ static void qe_set_multicast(struct net_device *dev) u8 *hbytes = (unsigned char *) &hash_table[0]; memset(hash_table, 0, sizeof(hash_table)); - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; if (!(*addrs & 1)) continue; diff --git a/drivers/net/sunvnet.c b/drivers/net/sunvnet.c index 6b1b7cea7f6b..6cf8b06be5cd 100644 --- a/drivers/net/sunvnet.c +++ b/drivers/net/sunvnet.c @@ -763,12 +763,12 @@ static struct vnet_mcast_entry *__vnet_mc_find(struct vnet *vp, u8 *addr) static void __update_mc_list(struct vnet *vp, struct net_device *dev) { - struct dev_addr_list *p; + struct netdev_hw_addr *ha; - netdev_for_each_mc_addr(p, dev) { + netdev_for_each_mc_addr(ha, dev) { struct vnet_mcast_entry *m; - m = __vnet_mc_find(vp, p->dmi_addr); + m = __vnet_mc_find(vp, ha->addr); if (m) { m->hit = 1; continue; @@ -778,7 +778,7 @@ static void __update_mc_list(struct vnet *vp, struct net_device *dev) m = kzalloc(sizeof(*m), GFP_ATOMIC); if (!m) continue; - memcpy(m->addr, p->dmi_addr, ETH_ALEN); + memcpy(m->addr, ha->addr, ETH_ALEN); m->hit = 1; m->next = vp->mcast_list; diff --git a/drivers/net/tc35815.c b/drivers/net/tc35815.c index 49bd84c0d583..36149ddace46 100644 --- a/drivers/net/tc35815.c +++ b/drivers/net/tc35815.c @@ -1954,16 +1954,16 @@ tc35815_set_multicast_list(struct net_device *dev) /* Disable promiscuous mode, use normal mode. */ tc_writel(CAM_CompEn | CAM_BroadAcc | CAM_GroupAcc, &tr->CAM_Ctl); } else if (!netdev_mc_empty(dev)) { - struct dev_mc_list *cur_addr; + struct netdev_hw_addr *ha; int i; int ena_bits = CAM_Ena_Bit(CAM_ENTRY_SOURCE); tc_writel(0, &tr->CAM_Ctl); /* Walk the address list, and load the filter */ i = 0; - netdev_for_each_mc_addr(cur_addr, dev) { + netdev_for_each_mc_addr(ha, dev) { /* entry 0,1 is reserved. */ - tc35815_set_cam_entry(dev, i + 2, cur_addr->dmi_addr); + tc35815_set_cam_entry(dev, i + 2, ha->addr); ena_bits |= CAM_Ena_Bit(i + 2); i++; } diff --git a/drivers/net/tehuti.c b/drivers/net/tehuti.c index 068a47174fca..a38aede5c8d6 100644 --- a/drivers/net/tehuti.c +++ b/drivers/net/tehuti.c @@ -808,7 +808,7 @@ static void bdx_setmulti(struct net_device *ndev) WRITE_REG(priv, regRX_MCST_HASH0 + i * 4, ~0); } else if (!netdev_mc_empty(ndev)) { u8 hash; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; u32 reg, val; /* set IMF to deny all multicast frames */ @@ -825,10 +825,10 @@ static void bdx_setmulti(struct net_device *ndev) * into RX_MAC_MCST regs. we skip this phase now and accept ALL * multicast frames throu IMF */ /* accept the rest of addresses throu IMF */ - netdev_for_each_mc_addr(mclist, ndev) { + netdev_for_each_mc_addr(ha, ndev) { hash = 0; for (i = 0; i < ETH_ALEN; i++) - hash ^= mclist->dmi_addr[i]; + hash ^= ha->addr[i]; reg = regRX_MCST_HASH0 + ((hash >> 5) << 2); val = READ_REG(priv, reg); val |= (1 << (hash % 32)); diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 22cf1c446de3..aff11f84f525 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -9457,14 +9457,14 @@ static void __tg3_set_rx_mode(struct net_device *dev) tg3_set_multi (tp, 0); } else { /* Accept one or more multicast(s). */ - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; u32 mc_filter[4] = { 0, }; u32 regidx; u32 bit; u32 crc; - netdev_for_each_mc_addr(mclist, dev) { - crc = calc_crc (mclist->dmi_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, dev) { + crc = calc_crc(ha->addr, ETH_ALEN); bit = ~crc & 0x7f; regidx = (bit & 0x60) >> 5; bit &= 0x1f; diff --git a/drivers/net/tlan.c b/drivers/net/tlan.c index 05ea30a94e89..8ffec22b74bf 100644 --- a/drivers/net/tlan.c +++ b/drivers/net/tlan.c @@ -1314,7 +1314,7 @@ static struct net_device_stats *TLan_GetStats( struct net_device *dev ) static void TLan_SetMulticastList( struct net_device *dev ) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; u32 hash1 = 0; u32 hash2 = 0; int i; @@ -1336,12 +1336,12 @@ static void TLan_SetMulticastList( struct net_device *dev ) TLan_DioWrite32( dev->base_addr, TLAN_HASH_2, 0xFFFFFFFF ); } else { i = 0; - netdev_for_each_mc_addr(dmi, dev) { + netdev_for_each_mc_addr(ha, dev) { if ( i < 3 ) { TLan_SetMac( dev, i + 1, - (char *) &dmi->dmi_addr ); + (char *) &ha->addr); } else { - offset = TLan_HashFunc( (u8 *) &dmi->dmi_addr ); + offset = TLan_HashFunc((u8 *)&ha->addr); if ( offset < 32 ) hash1 |= ( 1 << offset ); else diff --git a/drivers/net/tokenring/3c359.c b/drivers/net/tokenring/3c359.c index 8c54d89e60d7..4673e38c52a9 100644 --- a/drivers/net/tokenring/3c359.c +++ b/drivers/net/tokenring/3c359.c @@ -1390,7 +1390,7 @@ static int xl_close(struct net_device *dev) static void xl_set_rx_mode(struct net_device *dev) { struct xl_private *xl_priv = netdev_priv(dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; unsigned char dev_mc_address[4] ; u16 options ; @@ -1407,11 +1407,11 @@ static void xl_set_rx_mode(struct net_device *dev) dev_mc_address[0] = dev_mc_address[1] = dev_mc_address[2] = dev_mc_address[3] = 0 ; - netdev_for_each_mc_addr(dmi, dev) { - dev_mc_address[0] |= dmi->dmi_addr[2] ; - dev_mc_address[1] |= dmi->dmi_addr[3] ; - dev_mc_address[2] |= dmi->dmi_addr[4] ; - dev_mc_address[3] |= dmi->dmi_addr[5] ; + netdev_for_each_mc_addr(ha, dev) { + dev_mc_address[0] |= ha->addr[2]; + dev_mc_address[1] |= ha->addr[3]; + dev_mc_address[2] |= ha->addr[4]; + dev_mc_address[3] |= ha->addr[5]; } if (memcmp(xl_priv->xl_functional_addr,dev_mc_address,4) != 0) { /* Options have changed, run the command */ diff --git a/drivers/net/tokenring/ibmtr.c b/drivers/net/tokenring/ibmtr.c index 1a0967246e2f..eebdaae24328 100644 --- a/drivers/net/tokenring/ibmtr.c +++ b/drivers/net/tokenring/ibmtr.c @@ -986,7 +986,7 @@ static void open_sap(unsigned char type, struct net_device *dev) static void tok_set_multicast_list(struct net_device *dev) { struct tok_info *ti = netdev_priv(dev); - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; unsigned char address[4]; int i; @@ -995,11 +995,11 @@ static void tok_set_multicast_list(struct net_device *dev) /*BMS ifconfig tr down or hot unplug a PCMCIA card ??hownowbrowncow*/ if (/*BMSHELPdev->start == 0 ||*/ ti->open_status != OPEN) return; address[0] = address[1] = address[2] = address[3] = 0; - netdev_for_each_mc_addr(mclist, dev) { - address[0] |= mclist->dmi_addr[2]; - address[1] |= mclist->dmi_addr[3]; - address[2] |= mclist->dmi_addr[4]; - address[3] |= mclist->dmi_addr[5]; + netdev_for_each_mc_addr(ha, dev) { + address[0] |= ha->addr[2]; + address[1] |= ha->addr[3]; + address[2] |= ha->addr[4]; + address[3] |= ha->addr[5]; } SET_PAGE(ti->srb_page); for (i = 0; i < sizeof(struct srb_set_funct_addr); i++) diff --git a/drivers/net/tokenring/lanstreamer.c b/drivers/net/tokenring/lanstreamer.c index 01c780f25e98..88c893100c2b 100644 --- a/drivers/net/tokenring/lanstreamer.c +++ b/drivers/net/tokenring/lanstreamer.c @@ -1266,7 +1266,7 @@ static void streamer_set_rx_mode(struct net_device *dev) netdev_priv(dev); __u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio; __u8 options = 0; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; unsigned char dev_mc_address[5]; writel(streamer_priv->srb, streamer_mmio + LAPA); @@ -1302,11 +1302,11 @@ static void streamer_set_rx_mode(struct net_device *dev) writel(streamer_priv->srb,streamer_mmio+LAPA); dev_mc_address[0] = dev_mc_address[1] = dev_mc_address[2] = dev_mc_address[3] = 0 ; - netdev_for_each_mc_addr(dmi, dev) { - dev_mc_address[0] |= dmi->dmi_addr[2] ; - dev_mc_address[1] |= dmi->dmi_addr[3] ; - dev_mc_address[2] |= dmi->dmi_addr[4] ; - dev_mc_address[3] |= dmi->dmi_addr[5] ; + netdev_for_each_mc_addr(ha, dev) { + dev_mc_address[0] |= ha->addr[2]; + dev_mc_address[1] |= ha->addr[3]; + dev_mc_address[2] |= ha->addr[4]; + dev_mc_address[3] |= ha->addr[5]; } writew(htons(SRB_SET_FUNC_ADDRESS << 8),streamer_mmio+LAPDINC); diff --git a/drivers/net/tokenring/olympic.c b/drivers/net/tokenring/olympic.c index 3decaf4b6cf2..3d2fbe60b46e 100644 --- a/drivers/net/tokenring/olympic.c +++ b/drivers/net/tokenring/olympic.c @@ -1139,7 +1139,7 @@ static void olympic_set_rx_mode(struct net_device *dev) u8 __iomem *olympic_mmio = olympic_priv->olympic_mmio ; u8 options = 0; u8 __iomem *srb; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; unsigned char dev_mc_address[4] ; writel(olympic_priv->srb,olympic_mmio+LAPA); @@ -1177,11 +1177,11 @@ static void olympic_set_rx_mode(struct net_device *dev) dev_mc_address[0] = dev_mc_address[1] = dev_mc_address[2] = dev_mc_address[3] = 0 ; - netdev_for_each_mc_addr(dmi, dev) { - dev_mc_address[0] |= dmi->dmi_addr[2] ; - dev_mc_address[1] |= dmi->dmi_addr[3] ; - dev_mc_address[2] |= dmi->dmi_addr[4] ; - dev_mc_address[3] |= dmi->dmi_addr[5] ; + netdev_for_each_mc_addr(ha, dev) { + dev_mc_address[0] |= ha->addr[2]; + dev_mc_address[1] |= ha->addr[3]; + dev_mc_address[2] |= ha->addr[4]; + dev_mc_address[3] |= ha->addr[5]; } writeb(SRB_SET_FUNC_ADDRESS,srb+0); diff --git a/drivers/net/tokenring/tms380tr.c b/drivers/net/tokenring/tms380tr.c index b545e20d2898..9fa2c36ff904 100644 --- a/drivers/net/tokenring/tms380tr.c +++ b/drivers/net/tokenring/tms380tr.c @@ -1212,17 +1212,17 @@ static void tms380tr_set_multicast_list(struct net_device *dev) } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { ((char *)(&tp->ocpl.FunctAddr))[0] |= - mclist->dmi_addr[2]; + ha->addr[2]; ((char *)(&tp->ocpl.FunctAddr))[1] |= - mclist->dmi_addr[3]; + ha->addr[3]; ((char *)(&tp->ocpl.FunctAddr))[2] |= - mclist->dmi_addr[4]; + ha->addr[4]; ((char *)(&tp->ocpl.FunctAddr))[3] |= - mclist->dmi_addr[5]; + ha->addr[5]; } } tms380tr_exec_cmd(dev, OC_SET_FUNCT_ADDR); diff --git a/drivers/net/tsi108_eth.c b/drivers/net/tsi108_eth.c index 80333a4d3287..1366541c30a2 100644 --- a/drivers/net/tsi108_eth.c +++ b/drivers/net/tsi108_eth.c @@ -1186,15 +1186,15 @@ static void tsi108_set_rx_mode(struct net_device *dev) if (dev->flags & IFF_ALLMULTI || !netdev_mc_empty(dev)) { int i; - struct dev_mc_list *mc; + struct netdev_hw_addr *ha; rxcfg |= TSI108_EC_RXCFG_MFE | TSI108_EC_RXCFG_MC_HASH; memset(data->mc_hash, 0, sizeof(data->mc_hash)); - netdev_for_each_mc_addr(mc, dev) { + netdev_for_each_mc_addr(ha, dev) { u32 hash, crc; - crc = ether_crc(6, mc->dmi_addr); + crc = ether_crc(6, ha->addr); hash = crc >> 23; __set_bit(hash, &data->mc_hash[0]); } diff --git a/drivers/net/tulip/de2104x.c b/drivers/net/tulip/de2104x.c index cb429723b2c8..a68b91764627 100644 --- a/drivers/net/tulip/de2104x.c +++ b/drivers/net/tulip/de2104x.c @@ -670,15 +670,15 @@ static void build_setup_frame_hash(u16 *setup_frm, struct net_device *dev) { struct de_private *de = netdev_priv(dev); u16 hash_table[32]; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; int i; u16 *eaddrs; memset(hash_table, 0, sizeof(hash_table)); set_bit_le(255, hash_table); /* Broadcast entry */ /* This should work on big-endian machines as well. */ - netdev_for_each_mc_addr(mclist, dev) { - int index = ether_crc_le(ETH_ALEN, mclist->dmi_addr) & 0x1ff; + netdev_for_each_mc_addr(ha, dev) { + int index = ether_crc_le(ETH_ALEN, ha->addr) & 0x1ff; set_bit_le(index, hash_table); } @@ -699,13 +699,13 @@ static void build_setup_frame_hash(u16 *setup_frm, struct net_device *dev) static void build_setup_frame_perfect(u16 *setup_frm, struct net_device *dev) { struct de_private *de = netdev_priv(dev); - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; u16 *eaddrs; /* We have <= 14 addresses so we can use the wonderful 16 address perfect filtering of the Tulip. */ - netdev_for_each_mc_addr(mclist, dev) { - eaddrs = (u16 *)mclist->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + eaddrs = (u16 *) ha->addr; *setup_frm++ = *eaddrs; *setup_frm++ = *eaddrs++; *setup_frm++ = *eaddrs; *setup_frm++ = *eaddrs++; *setup_frm++ = *eaddrs; *setup_frm++ = *eaddrs++; diff --git a/drivers/net/tulip/de4x5.c b/drivers/net/tulip/de4x5.c index 55ade8949a68..cac4bcc83187 100644 --- a/drivers/net/tulip/de4x5.c +++ b/drivers/net/tulip/de4x5.c @@ -1951,7 +1951,7 @@ static void SetMulticastFilter(struct net_device *dev) { struct de4x5_private *lp = netdev_priv(dev); - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; u_long iobase = dev->base_addr; int i, bit, byte; u16 hashcode; @@ -1966,8 +1966,8 @@ SetMulticastFilter(struct net_device *dev) if ((dev->flags & IFF_ALLMULTI) || (netdev_mc_count(dev) > 14)) { omr |= OMR_PM; /* Pass all multicasts */ } else if (lp->setup_f == HASH_PERF) { /* Hash Filtering */ - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; if ((*addrs & 0x01) == 1) { /* multicast address? */ crc = ether_crc_le(ETH_ALEN, addrs); hashcode = crc & HASH_BITS; /* hashcode is 9 LSb of CRC */ @@ -1983,8 +1983,8 @@ SetMulticastFilter(struct net_device *dev) } } } else { /* Perfect filtering */ - netdev_for_each_mc_addr(dmi, dev) { - addrs = dmi->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrs = ha->addr; for (i=0; ibase_addr+0xc0; /* ID Table */ u32 hash_val; @@ -1478,8 +1478,8 @@ static void dm9132_id_table(struct DEVICE *dev) hash_table[3] = 0x8000; /* the multicast address in Hash Table : 64 bits */ - netdev_for_each_mc_addr(mcptr, dev) { - hash_val = cal_CRC((char *) mcptr->dmi_addr, 6, 0) & 0x3f; + netdev_for_each_mc_addr(ha, dev) { + hash_val = cal_CRC((char *) ha->addr, 6, 0) & 0x3f; hash_table[hash_val / 16] |= (u16) 1 << (hash_val % 16); } @@ -1497,7 +1497,7 @@ static void dm9132_id_table(struct DEVICE *dev) static void send_filter_frame(struct DEVICE *dev) { struct dmfe_board_info *db = netdev_priv(dev); - struct dev_mc_list *mcptr; + struct netdev_hw_addr *ha; struct tx_desc *txptr; u16 * addrptr; u32 * suptr; @@ -1520,8 +1520,8 @@ static void send_filter_frame(struct DEVICE *dev) *suptr++ = 0xffff; /* fit the multicast address */ - netdev_for_each_mc_addr(mcptr, dev) { - addrptr = (u16 *) mcptr->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrptr = (u16 *) ha->addr; *suptr++ = addrptr[0]; *suptr++ = addrptr[1]; *suptr++ = addrptr[2]; diff --git a/drivers/net/tulip/tulip_core.c b/drivers/net/tulip/tulip_core.c index 7f544ef2f5fc..c4b7cd726b69 100644 --- a/drivers/net/tulip/tulip_core.c +++ b/drivers/net/tulip/tulip_core.c @@ -990,15 +990,15 @@ static void build_setup_frame_hash(u16 *setup_frm, struct net_device *dev) { struct tulip_private *tp = netdev_priv(dev); u16 hash_table[32]; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; int i; u16 *eaddrs; memset(hash_table, 0, sizeof(hash_table)); set_bit_le(255, hash_table); /* Broadcast entry */ /* This should work on big-endian machines as well. */ - netdev_for_each_mc_addr(mclist, dev) { - int index = ether_crc_le(ETH_ALEN, mclist->dmi_addr) & 0x1ff; + netdev_for_each_mc_addr(ha, dev) { + int index = ether_crc_le(ETH_ALEN, ha->addr) & 0x1ff; set_bit_le(index, hash_table); } @@ -1018,13 +1018,13 @@ static void build_setup_frame_hash(u16 *setup_frm, struct net_device *dev) static void build_setup_frame_perfect(u16 *setup_frm, struct net_device *dev) { struct tulip_private *tp = netdev_priv(dev); - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; u16 *eaddrs; /* We have <= 14 addresses so we can use the wonderful 16 address perfect filtering of the Tulip. */ - netdev_for_each_mc_addr(mclist, dev) { - eaddrs = (u16 *)mclist->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + eaddrs = (u16 *) ha->addr; *setup_frm++ = *eaddrs; *setup_frm++ = *eaddrs++; *setup_frm++ = *eaddrs; *setup_frm++ = *eaddrs++; *setup_frm++ = *eaddrs; *setup_frm++ = *eaddrs++; @@ -1061,7 +1061,7 @@ static void set_rx_mode(struct net_device *dev) } else if (tp->flags & MC_HASH_ONLY) { /* Some work-alikes have only a 64-entry hash filter table. */ /* Should verify correctness on big-endian/__powerpc__ */ - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; if (netdev_mc_count(dev) > 64) { /* Arbitrary non-effective limit. */ tp->csr6 |= AcceptAllMulticast; @@ -1069,18 +1069,21 @@ static void set_rx_mode(struct net_device *dev) } else { u32 mc_filter[2] = {0, 0}; /* Multicast hash filter */ int filterbit; - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { if (tp->flags & COMET_MAC_ADDR) - filterbit = ether_crc_le(ETH_ALEN, mclist->dmi_addr); + filterbit = ether_crc_le(ETH_ALEN, + ha->addr); else - filterbit = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26; + filterbit = ether_crc(ETH_ALEN, + ha->addr) >> 26; filterbit &= 0x3f; mc_filter[filterbit >> 5] |= 1 << (filterbit & 31); if (tulip_debug > 2) dev_info(&dev->dev, "Added filter for %pM %08x bit %d\n", - mclist->dmi_addr, - ether_crc(ETH_ALEN, mclist->dmi_addr), filterbit); + ha->addr, + ether_crc(ETH_ALEN, ha->addr), + filterbit); } if (mc_filter[0] == tp->mc_filter[0] && mc_filter[1] == tp->mc_filter[1]) diff --git a/drivers/net/tulip/uli526x.c b/drivers/net/tulip/uli526x.c index 0ab05af237e5..b79d908fe34e 100644 --- a/drivers/net/tulip/uli526x.c +++ b/drivers/net/tulip/uli526x.c @@ -1392,7 +1392,7 @@ static void update_cr6(u32 cr6_data, unsigned long ioaddr) static void send_filter_frame(struct net_device *dev, int mc_cnt) { struct uli526x_board_info *db = netdev_priv(dev); - struct dev_mc_list *mcptr; + struct netdev_hw_addr *ha; struct tx_desc *txptr; u16 * addrptr; u32 * suptr; @@ -1415,8 +1415,8 @@ static void send_filter_frame(struct net_device *dev, int mc_cnt) *suptr++ = 0xffff << FLT_SHIFT; /* fit the multicast address */ - netdev_for_each_mc_addr(mcptr, dev) { - addrptr = (u16 *) mcptr->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addrptr = (u16 *) ha->addr; *suptr++ = addrptr[0] << FLT_SHIFT; *suptr++ = addrptr[1] << FLT_SHIFT; *suptr++ = addrptr[2] << FLT_SHIFT; diff --git a/drivers/net/tulip/winbond-840.c b/drivers/net/tulip/winbond-840.c index 304f43866c44..3e3822c98a0d 100644 --- a/drivers/net/tulip/winbond-840.c +++ b/drivers/net/tulip/winbond-840.c @@ -1367,13 +1367,15 @@ static u32 __set_rx_mode(struct net_device *dev) memset(mc_filter, 0xff, sizeof(mc_filter)); rx_mode = RxAcceptBroadcast | AcceptMulticast | AcceptMyPhys; } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { - int filterbit = (ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26) ^ 0x3F; - filterbit &= 0x3f; - mc_filter[filterbit >> 5] |= 1 << (filterbit & 31); + netdev_for_each_mc_addr(ha, dev) { + int filbit; + + filbit = (ether_crc(ETH_ALEN, ha->addr) >> 26) ^ 0x3F; + filbit &= 0x3f; + mc_filter[filbit >> 5] |= 1 << (filbit & 31); } rx_mode = RxAcceptBroadcast | AcceptMulticast | AcceptMyPhys; } diff --git a/drivers/net/typhoon.c b/drivers/net/typhoon.c index cd24e5f2b2a2..3053f85aa066 100644 --- a/drivers/net/typhoon.c +++ b/drivers/net/typhoon.c @@ -921,11 +921,11 @@ typhoon_set_rx_mode(struct net_device *dev) /* Too many to match, or accept all multicasts. */ filter |= TYPHOON_RX_FILTER_ALL_MCAST; } else if (!netdev_mc_empty(dev)) { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { - int bit = ether_crc(ETH_ALEN, mclist->dmi_addr) & 0x3f; + netdev_for_each_mc_addr(ha, dev) { + int bit = ether_crc(ETH_ALEN, ha->addr) & 0x3f; mc_filter[bit >> 5] |= 1 << (bit & 0x1f); } diff --git a/drivers/net/ucc_geth.c b/drivers/net/ucc_geth.c index 75743a76bbe8..081f76bff341 100644 --- a/drivers/net/ucc_geth.c +++ b/drivers/net/ucc_geth.c @@ -1999,7 +1999,7 @@ static void ucc_geth_memclean(struct ucc_geth_private *ugeth) static void ucc_geth_set_multi(struct net_device *dev) { struct ucc_geth_private *ugeth; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; struct ucc_fast __iomem *uf_regs; struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt; @@ -2028,16 +2028,16 @@ static void ucc_geth_set_multi(struct net_device *dev) out_be32(&p_82xx_addr_filt->gaddr_h, 0x0); out_be32(&p_82xx_addr_filt->gaddr_l, 0x0); - netdev_for_each_mc_addr(dmi, dev) { + netdev_for_each_mc_addr(ha, dev) { /* Only support group multicast for now. */ - if (!(dmi->dmi_addr[0] & 1)) + if (!(ha->addr[0] & 1)) continue; /* Ask CPM to run CRC and set bit in * filter mask. */ - hw_add_addr_in_hash(ugeth, dmi->dmi_addr); + hw_add_addr_in_hash(ugeth, ha->addr); } } } diff --git a/drivers/net/usb/asix.c b/drivers/net/usb/asix.c index 9e05639435f2..763364f0972e 100644 --- a/drivers/net/usb/asix.c +++ b/drivers/net/usb/asix.c @@ -557,16 +557,14 @@ static void asix_set_multicast(struct net_device *net) * for our 8 byte filter buffer * to avoid allocating memory that * is tricky to free later */ - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; u32 crc_bits; memset(data->multi_filter, 0, AX_MCAST_FILTER_SIZE); /* Build the multicast hash filter. */ - netdev_for_each_mc_addr(mc_list, net) { - crc_bits = - ether_crc(ETH_ALEN, - mc_list->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, net) { + crc_bits = ether_crc(ETH_ALEN, ha->addr) >> 26; data->multi_filter[crc_bits >> 3] |= 1 << (crc_bits & 7); } @@ -793,16 +791,14 @@ static void ax88172_set_multicast(struct net_device *net) * for our 8 byte filter buffer * to avoid allocating memory that * is tricky to free later */ - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; u32 crc_bits; memset(data->multi_filter, 0, AX_MCAST_FILTER_SIZE); /* Build the multicast hash filter. */ - netdev_for_each_mc_addr(mc_list, net) { - crc_bits = - ether_crc(ETH_ALEN, - mc_list->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, net) { + crc_bits = ether_crc(ETH_ALEN, ha->addr) >> 26; data->multi_filter[crc_bits >> 3] |= 1 << (crc_bits & 7); } diff --git a/drivers/net/usb/catc.c b/drivers/net/usb/catc.c index 96f1ebe0d348..0c48ff97f47c 100644 --- a/drivers/net/usb/catc.c +++ b/drivers/net/usb/catc.c @@ -629,7 +629,7 @@ static void catc_multicast(unsigned char *addr, u8 *multicast) static void catc_set_multicast_list(struct net_device *netdev) { struct catc *catc = netdev_priv(netdev); - struct dev_mc_list *mc; + struct netdev_hw_addr *ha; u8 broadcast[6]; u8 rx = RxEnable | RxPolarity | RxMultiCast; @@ -647,8 +647,8 @@ static void catc_set_multicast_list(struct net_device *netdev) if (netdev->flags & IFF_ALLMULTI) { memset(catc->multicast, 0xff, 64); } else { - netdev_for_each_mc_addr(mc, netdev) { - u32 crc = ether_crc_le(6, mc->dmi_addr); + netdev_for_each_mc_addr(ha, netdev) { + u32 crc = ether_crc_le(6, ha->addr); if (!catc->is_f5u011) { catc->multicast[(crc >> 3) & 0x3f] |= 1 << (crc & 7); } else { diff --git a/drivers/net/usb/dm9601.c b/drivers/net/usb/dm9601.c index 269339769f47..4eb1fb31ff0b 100644 --- a/drivers/net/usb/dm9601.c +++ b/drivers/net/usb/dm9601.c @@ -386,10 +386,10 @@ static void dm9601_set_multicast(struct net_device *net) netdev_mc_count(net) > DM_MAX_MCAST) { rx_ctl |= 0x04; } else if (!netdev_mc_empty(net)) { - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; - netdev_for_each_mc_addr(mc_list, net) { - u32 crc = ether_crc(ETH_ALEN, mc_list->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, net) { + u32 crc = ether_crc(ETH_ALEN, ha->addr) >> 26; hashes[crc >> 3] |= 1 << (crc & 0x7); } } diff --git a/drivers/net/usb/mcs7830.c b/drivers/net/usb/mcs7830.c index 70978219e98a..0dc92c8ba4b2 100644 --- a/drivers/net/usb/mcs7830.c +++ b/drivers/net/usb/mcs7830.c @@ -452,12 +452,12 @@ static void mcs7830_data_set_multicast(struct net_device *net) * for our 8 byte filter buffer * to avoid allocating memory that * is tricky to free later */ - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; u32 crc_bits; /* Build the multicast hash filter. */ - netdev_for_each_mc_addr(mc_list, net) { - crc_bits = ether_crc(ETH_ALEN, mc_list->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, net) { + crc_bits = ether_crc(ETH_ALEN, ha->addr) >> 26; data->multi_filter[crc_bits >> 3] |= 1 << (crc_bits & 7); } } diff --git a/drivers/net/usb/smsc75xx.c b/drivers/net/usb/smsc75xx.c index 300e3e764fa2..b8b00d06ea79 100644 --- a/drivers/net/usb/smsc75xx.c +++ b/drivers/net/usb/smsc75xx.c @@ -444,14 +444,14 @@ static void smsc75xx_set_multicast(struct net_device *netdev) netif_dbg(dev, drv, dev->net, "receive all multicast enabled"); pdata->rfe_ctl |= RFE_CTL_AM | RFE_CTL_DPF; } else if (!netdev_mc_empty(dev->net)) { - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; netif_dbg(dev, drv, dev->net, "receive multicast hash filter"); pdata->rfe_ctl |= RFE_CTL_MHF | RFE_CTL_DPF; - netdev_for_each_mc_addr(mc_list, netdev) { - u32 bitnum = smsc75xx_hash(mc_list->dmi_addr); + netdev_for_each_mc_addr(ha, netdev) { + u32 bitnum = smsc75xx_hash(ha->addr); pdata->multicast_hash_table[bitnum / 32] |= (1 << (bitnum % 32)); } diff --git a/drivers/net/usb/smsc95xx.c b/drivers/net/usb/smsc95xx.c index 73f9a31cf94d..ccd55ca3a06a 100644 --- a/drivers/net/usb/smsc95xx.c +++ b/drivers/net/usb/smsc95xx.c @@ -384,13 +384,13 @@ static void smsc95xx_set_multicast(struct net_device *netdev) pdata->mac_cr |= MAC_CR_MCPAS_; pdata->mac_cr &= ~(MAC_CR_PRMS_ | MAC_CR_HPFILT_); } else if (!netdev_mc_empty(dev->net)) { - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; pdata->mac_cr |= MAC_CR_HPFILT_; pdata->mac_cr &= ~(MAC_CR_PRMS_ | MAC_CR_MCPAS_); - netdev_for_each_mc_addr(mc_list, netdev) { - u32 bitnum = smsc95xx_hash(mc_list->dmi_addr); + netdev_for_each_mc_addr(ha, netdev) { + u32 bitnum = smsc95xx_hash(ha->addr); u32 mask = 0x01 << (bitnum & 0x1F); if (bitnum & 0x20) hash_hi |= mask; diff --git a/drivers/net/via-rhine.c b/drivers/net/via-rhine.c index 50f881aa3939..d9133c62a2f5 100644 --- a/drivers/net/via-rhine.c +++ b/drivers/net/via-rhine.c @@ -1704,11 +1704,11 @@ static void rhine_set_rx_mode(struct net_device *dev) iowrite32(0xffffffff, ioaddr + MulticastFilter1); rx_mode = 0x0C; } else { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { - int bit_nr = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, dev) { + int bit_nr = ether_crc(ETH_ALEN, ha->addr) >> 26; mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 31); } diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c index 3a486f3bad3d..91f3b841288c 100644 --- a/drivers/net/via-velocity.c +++ b/drivers/net/via-velocity.c @@ -1126,7 +1126,7 @@ static void velocity_set_multi(struct net_device *dev) struct mac_regs __iomem *regs = vptr->mac_regs; u8 rx_mode; int i; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; if (dev->flags & IFF_PROMISC) { /* Set promiscuous. */ writel(0xffffffff, ®s->MARCAM[0]); @@ -1142,8 +1142,8 @@ static void velocity_set_multi(struct net_device *dev) mac_get_cam_mask(regs, vptr->mCAMmask); i = 0; - netdev_for_each_mc_addr(mclist, dev) { - mac_set_cam(regs, i + offset, mclist->dmi_addr); + netdev_for_each_mc_addr(ha, dev) { + mac_set_cam(regs, i + offset, ha->addr); vptr->mCAMmask[(offset + i) / 8] |= 1 << ((offset + i) & 7); i++; } diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 3f5be35dea25..ecec9a8527f3 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -721,7 +721,6 @@ static void virtnet_set_rx_mode(struct net_device *dev) struct scatterlist sg[2]; u8 promisc, allmulti; struct virtio_net_ctrl_mac *mac_data; - struct dev_addr_list *addr; struct netdev_hw_addr *ha; int uc_count; int mc_count; @@ -778,8 +777,8 @@ static void virtnet_set_rx_mode(struct net_device *dev) mac_data->entries = mc_count; i = 0; - netdev_for_each_mc_addr(addr, dev) - memcpy(&mac_data->macs[i++][0], addr->da_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, dev) + memcpy(&mac_data->macs[i++][0], ha->addr, ETH_ALEN); sg_set_buf(&sg[1], mac_data, sizeof(mac_data->entries) + (mc_count * ETH_ALEN)); diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c index cff3485d9673..58dfa367bf86 100644 --- a/drivers/net/vmxnet3/vmxnet3_drv.c +++ b/drivers/net/vmxnet3/vmxnet3_drv.c @@ -1675,11 +1675,11 @@ vmxnet3_copy_mc(struct net_device *netdev) /* We may be called with BH disabled */ buf = kmalloc(sz, GFP_ATOMIC); if (buf) { - struct dev_mc_list *mc; + struct netdev_hw_addr *ha; int i = 0; - netdev_for_each_mc_addr(mc, netdev) - memcpy(buf + i++ * ETH_ALEN, mc->dmi_addr, + netdev_for_each_mc_addr(ha, netdev) + memcpy(buf + i++ * ETH_ALEN, ha->addr, ETH_ALEN); } } diff --git a/drivers/net/vxge/vxge-main.c b/drivers/net/vxge/vxge-main.c index e5f2d3ee0df3..37836a10d093 100644 --- a/drivers/net/vxge/vxge-main.c +++ b/drivers/net/vxge/vxge-main.c @@ -1117,7 +1117,7 @@ vxge_tx_term(void *dtrh, enum vxge_hw_txdl_state state, void *userdata) */ static void vxge_set_multicast(struct net_device *dev) { - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; struct vxgedev *vdev; int i, mcast_cnt = 0; struct __vxge_hw_device *hldev; @@ -1217,8 +1217,8 @@ static void vxge_set_multicast(struct net_device *dev) } /* Add new ones */ - netdev_for_each_mc_addr(mclist, dev) { - memcpy(mac_info.macaddr, mclist->dmi_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, dev) { + memcpy(mac_info.macaddr, ha->addr, ETH_ALEN); for (vpath_idx = 0; vpath_idx < vdev->no_of_vpath; vpath_idx++) { mac_info.vpath_no = vpath_idx; diff --git a/drivers/net/wireless/adm8211.c b/drivers/net/wireless/adm8211.c index 547912e6843f..2277998b7264 100644 --- a/drivers/net/wireless/adm8211.c +++ b/drivers/net/wireless/adm8211.c @@ -1317,21 +1317,19 @@ static void adm8211_bss_info_changed(struct ieee80211_hw *dev, } static u64 adm8211_prepare_multicast(struct ieee80211_hw *hw, - int mc_count, struct dev_addr_list *mclist) + struct netdev_hw_addr_list *mc_list) { - unsigned int bit_nr, i; + unsigned int bit_nr; u32 mc_filter[2]; + struct netdev_hw_addr *ha; mc_filter[1] = mc_filter[0] = 0; - for (i = 0; i < mc_count; i++) { - if (!mclist) - break; - bit_nr = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26; + netdev_hw_addr_list_for_each(ha, mc_list) { + bit_nr = ether_crc(ETH_ALEN, ha->addr) >> 26; bit_nr &= 0x3F; mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 31); - mclist = mclist->next; } return mc_filter[0] | ((u64)(mc_filter[1]) << 32); diff --git a/drivers/net/wireless/ath/ar9170/main.c b/drivers/net/wireless/ath/ar9170/main.c index 257c734733d1..b43d4b006d7e 100644 --- a/drivers/net/wireless/ath/ar9170/main.c +++ b/drivers/net/wireless/ath/ar9170/main.c @@ -2045,21 +2045,17 @@ out: return err; } -static u64 ar9170_op_prepare_multicast(struct ieee80211_hw *hw, int mc_count, - struct dev_addr_list *mclist) +static u64 ar9170_op_prepare_multicast(struct ieee80211_hw *hw, + struct netdev_hw_addr_list *mc_list) { u64 mchash; - int i; + struct netdev_hw_addr *ha; /* always get broadcast frames */ mchash = 1ULL << (0xff >> 2); - for (i = 0; i < mc_count; i++) { - if (WARN_ON(!mclist)) - break; - mchash |= 1ULL << (mclist->dmi_addr[5] >> 2); - mclist = mclist->next; - } + netdev_hw_addr_list_for_each(ha, mc_list) + mchash |= 1ULL << (ha->addr[5] >> 2); return mchash; } diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c index b142a78ed1e5..53a2340f52bc 100644 --- a/drivers/net/wireless/ath/ath5k/base.c +++ b/drivers/net/wireless/ath/ath5k/base.c @@ -230,7 +230,7 @@ static void ath5k_remove_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif); static int ath5k_config(struct ieee80211_hw *hw, u32 changed); static u64 ath5k_prepare_multicast(struct ieee80211_hw *hw, - int mc_count, struct dev_addr_list *mc_list); + struct netdev_hw_addr_list *mc_list); static void ath5k_configure_filter(struct ieee80211_hw *hw, unsigned int changed_flags, unsigned int *new_flags, @@ -2999,22 +2999,20 @@ unlock: } static u64 ath5k_prepare_multicast(struct ieee80211_hw *hw, - int mc_count, struct dev_addr_list *mclist) + struct netdev_hw_addr_list *mc_list) { u32 mfilt[2], val; - int i; u8 pos; + struct netdev_hw_addr *ha; mfilt[0] = 0; mfilt[1] = 1; - for (i = 0; i < mc_count; i++) { - if (!mclist) - break; + netdev_hw_addr_list_for_each(ha, mc_list) { /* calculate XOR of eight 6-bit values */ - val = get_unaligned_le32(mclist->dmi_addr + 0); + val = get_unaligned_le32(ha->addr + 0); pos = (val >> 18) ^ (val >> 12) ^ (val >> 6) ^ val; - val = get_unaligned_le32(mclist->dmi_addr + 3); + val = get_unaligned_le32(ha->addr + 3); pos ^= (val >> 18) ^ (val >> 12) ^ (val >> 6) ^ val; pos &= 0x3f; mfilt[pos / 32] |= (1 << (pos % 32)); @@ -3022,8 +3020,7 @@ static u64 ath5k_prepare_multicast(struct ieee80211_hw *hw, * but not sure, needs testing, if we do use this we'd * neet to inform below to not reset the mcast */ /* ath5k_hw_set_mcast_filterindex(ah, - * mclist->dmi_addr[5]); */ - mclist = mclist->next; + * ha->addr[5]); */ } return ((u64)(mfilt[1]) << 32) | mfilt[0]; diff --git a/drivers/net/wireless/libertas/main.c b/drivers/net/wireless/libertas/main.c index 3c889f43d909..ff7b0d0cca56 100644 --- a/drivers/net/wireless/libertas/main.c +++ b/drivers/net/wireless/libertas/main.c @@ -318,7 +318,7 @@ static int lbs_add_mcast_addrs(struct cmd_ds_mac_multicast_adr *cmd, struct net_device *dev, int nr_addrs) { int i = nr_addrs; - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; int cnt; if ((dev->flags & (IFF_UP|IFF_MULTICAST)) != (IFF_UP|IFF_MULTICAST)) @@ -326,19 +326,19 @@ static int lbs_add_mcast_addrs(struct cmd_ds_mac_multicast_adr *cmd, netif_addr_lock_bh(dev); cnt = netdev_mc_count(dev); - netdev_for_each_mc_addr(mc_list, dev) { - if (mac_in_list(cmd->maclist, nr_addrs, mc_list->dmi_addr)) { + netdev_for_each_mc_addr(ha, dev) { + if (mac_in_list(cmd->maclist, nr_addrs, ha->addr)) { lbs_deb_net("mcast address %s:%pM skipped\n", dev->name, - mc_list->dmi_addr); + ha->addr); cnt--; continue; } if (i == MRVDRV_MAX_MULTICAST_LIST_SIZE) break; - memcpy(&cmd->maclist[6*i], mc_list->dmi_addr, ETH_ALEN); + memcpy(&cmd->maclist[6*i], ha->addr, ETH_ALEN); lbs_deb_net("mcast address %s:%pM added to filter\n", dev->name, - mc_list->dmi_addr); + ha->addr); i++; cnt--; } diff --git a/drivers/net/wireless/libertas_tf/main.c b/drivers/net/wireless/libertas_tf/main.c index 6ab30033c26c..c20eef6b95c2 100644 --- a/drivers/net/wireless/libertas_tf/main.c +++ b/drivers/net/wireless/libertas_tf/main.c @@ -367,22 +367,20 @@ static int lbtf_op_config(struct ieee80211_hw *hw, u32 changed) } static u64 lbtf_op_prepare_multicast(struct ieee80211_hw *hw, - int mc_count, struct dev_addr_list *mclist) + struct netdev_hw_addr_list *mc_list) { struct lbtf_private *priv = hw->priv; int i; + struct netdev_hw_addr *ha; + int mc_count = netdev_hw_addr_list_count(mc_list); if (!mc_count || mc_count > MRVDRV_MAX_MULTICAST_LIST_SIZE) return mc_count; priv->nr_of_multicastmacaddr = mc_count; - for (i = 0; i < mc_count; i++) { - if (!mclist) - break; - memcpy(&priv->multicastlist[i], mclist->da_addr, - ETH_ALEN); - mclist = mclist->next; - } + i = 0; + netdev_hw_addr_list_for_each(ha, mc_list) + memcpy(&priv->multicastlist[i++], ha->addr, ETH_ALEN); return mc_count; } diff --git a/drivers/net/wireless/mwl8k.c b/drivers/net/wireless/mwl8k.c index ac65e13eb0de..6599fd15e675 100644 --- a/drivers/net/wireless/mwl8k.c +++ b/drivers/net/wireless/mwl8k.c @@ -1938,11 +1938,15 @@ struct mwl8k_cmd_mac_multicast_adr { static struct mwl8k_cmd_pkt * __mwl8k_cmd_mac_multicast_adr(struct ieee80211_hw *hw, int allmulti, - int mc_count, struct dev_addr_list *mclist) + struct netdev_hw_addr_list *mc_list) { struct mwl8k_priv *priv = hw->priv; struct mwl8k_cmd_mac_multicast_adr *cmd; int size; + int mc_count = 0; + + if (mc_list) + mc_count = netdev_hw_addr_list_count(mc_list); if (allmulti || mc_count > priv->num_mcaddrs) { allmulti = 1; @@ -1963,17 +1967,13 @@ __mwl8k_cmd_mac_multicast_adr(struct ieee80211_hw *hw, int allmulti, if (allmulti) { cmd->action |= cpu_to_le16(MWL8K_ENABLE_RX_ALL_MULTICAST); } else if (mc_count) { - int i; + struct netdev_hw_addr *ha; + int i = 0; cmd->action |= cpu_to_le16(MWL8K_ENABLE_RX_MULTICAST); cmd->numaddr = cpu_to_le16(mc_count); - for (i = 0; i < mc_count && mclist; i++) { - if (mclist->da_addrlen != ETH_ALEN) { - kfree(cmd); - return NULL; - } - memcpy(cmd->addr[i], mclist->da_addr, ETH_ALEN); - mclist = mclist->next; + netdev_hw_addr_list_for_each(ha, mc_list) { + memcpy(cmd->addr[i], ha->addr, ETH_ALEN); } } @@ -3552,7 +3552,7 @@ mwl8k_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif, } static u64 mwl8k_prepare_multicast(struct ieee80211_hw *hw, - int mc_count, struct dev_addr_list *mclist) + struct netdev_hw_addr_list *mc_list) { struct mwl8k_cmd_pkt *cmd; @@ -3563,7 +3563,7 @@ static u64 mwl8k_prepare_multicast(struct ieee80211_hw *hw, * we'll end up throwing this packet away and creating a new * one in mwl8k_configure_filter(). */ - cmd = __mwl8k_cmd_mac_multicast_adr(hw, 0, mc_count, mclist); + cmd = __mwl8k_cmd_mac_multicast_adr(hw, 0, mc_list); return (unsigned long)cmd; } @@ -3686,7 +3686,7 @@ static void mwl8k_configure_filter(struct ieee80211_hw *hw, */ if (*total_flags & FIF_ALLMULTI) { kfree(cmd); - cmd = __mwl8k_cmd_mac_multicast_adr(hw, 1, 0, NULL); + cmd = __mwl8k_cmd_mac_multicast_adr(hw, 1, NULL); } if (cmd != NULL) { diff --git a/drivers/net/wireless/orinoco/hw.c b/drivers/net/wireless/orinoco/hw.c index 883b8f868626..9f657afaa3e5 100644 --- a/drivers/net/wireless/orinoco/hw.c +++ b/drivers/net/wireless/orinoco/hw.c @@ -1056,14 +1056,14 @@ int __orinoco_hw_set_multicast_list(struct orinoco_private *priv, * group address if either we want to multicast, or if we were * multicasting and want to stop */ if (!promisc && (mc_count || priv->mc_count)) { - struct dev_mc_list *p; + struct netdev_hw_addr *ha; struct hermes_multicast mclist; int i = 0; - netdev_for_each_mc_addr(p, dev) { + netdev_for_each_mc_addr(ha, dev) { if (i == mc_count) break; - memcpy(mclist.addr[i++], p->dmi_addr, ETH_ALEN); + memcpy(mclist.addr[i++], ha->addr, ETH_ALEN); } err = hermes_write_ltv(hw, USER_BAP, diff --git a/drivers/net/wireless/orinoco/hw.h b/drivers/net/wireless/orinoco/hw.h index 9799a1d14a63..97af71e79950 100644 --- a/drivers/net/wireless/orinoco/hw.h +++ b/drivers/net/wireless/orinoco/hw.h @@ -22,7 +22,6 @@ /* Forward declarations */ struct orinoco_private; -struct dev_addr_list; int determine_fw_capabilities(struct orinoco_private *priv, char *fw_name, size_t fw_name_len, u32 *hw_ver); diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c index 4f5bdb528ef7..54680a3a5acb 100644 --- a/drivers/net/wireless/ray_cs.c +++ b/drivers/net/wireless/ray_cs.c @@ -1875,17 +1875,17 @@ static void ray_update_multi_list(struct net_device *dev, int all) writeb(0xff, &pccs->var); local->num_multi = 0xff; } else { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int i = 0; /* Copy the kernel's list of MC addresses to card */ - netdev_for_each_mc_addr(dmi, dev) { - memcpy_toio(p, dmi->dmi_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, dev) { + memcpy_toio(p, ha->addr, ETH_ALEN); dev_dbg(&link->dev, "ray_update_multi add addr %02x%02x%02x%02x%02x%02x\n", - dmi->dmi_addr[0], dmi->dmi_addr[1], - dmi->dmi_addr[2], dmi->dmi_addr[3], - dmi->dmi_addr[4], dmi->dmi_addr[5]); + ha->addr[0], ha->addr[1], + ha->addr[2], ha->addr[3], + ha->addr[4], ha->addr[5]); p += ETH_ALEN; i++; } diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c index aceb95ef7274..8ab9f094747b 100644 --- a/drivers/net/wireless/rndis_wlan.c +++ b/drivers/net/wireless/rndis_wlan.c @@ -1545,7 +1545,7 @@ static int remove_key(struct usbnet *usbdev, int index, const u8 *bssid) static void set_multicast_list(struct usbnet *usbdev) { struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev); - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; __le32 filter, basefilter; int ret; char *mc_addrs = NULL; @@ -1584,9 +1584,9 @@ static void set_multicast_list(struct usbnet *usbdev) return; } - netdev_for_each_mc_addr(mclist, usbdev->net) + netdev_for_each_mc_addr(ha, usbdev->net) memcpy(mc_addrs + i++ * ETH_ALEN, - mclist->dmi_addr, ETH_ALEN); + ha->addr, ETH_ALEN); } netif_addr_unlock_bh(usbdev->net); diff --git a/drivers/net/wireless/rtl818x/rtl8180_dev.c b/drivers/net/wireless/rtl818x/rtl8180_dev.c index 2b928ecf47bd..fb8a62f2b9b4 100644 --- a/drivers/net/wireless/rtl818x/rtl8180_dev.c +++ b/drivers/net/wireless/rtl818x/rtl8180_dev.c @@ -723,10 +723,10 @@ static void rtl8180_bss_info_changed(struct ieee80211_hw *dev, priv->rf->conf_erp(dev, info); } -static u64 rtl8180_prepare_multicast(struct ieee80211_hw *dev, int mc_count, - struct dev_addr_list *mc_list) +static u64 rtl8180_prepare_multicast(struct ieee80211_hw *dev, + struct netdev_hw_addr_list *mc_list) { - return mc_count; + return netdev_hw_addr_list_count(mc_list); } static void rtl8180_configure_filter(struct ieee80211_hw *dev, diff --git a/drivers/net/wireless/rtl818x/rtl8187_dev.c b/drivers/net/wireless/rtl818x/rtl8187_dev.c index 0fb850e0c656..441d817ed489 100644 --- a/drivers/net/wireless/rtl818x/rtl8187_dev.c +++ b/drivers/net/wireless/rtl818x/rtl8187_dev.c @@ -1193,9 +1193,9 @@ static void rtl8187_bss_info_changed(struct ieee80211_hw *dev, } static u64 rtl8187_prepare_multicast(struct ieee80211_hw *dev, - int mc_count, struct dev_addr_list *mc_list) + struct netdev_hw_addr_list *mc_list) { - return mc_count; + return netdev_hw_addr_list_count(mc_list); } static void rtl8187_configure_filter(struct ieee80211_hw *dev, diff --git a/drivers/net/wireless/wl12xx/wl1271_main.c b/drivers/net/wireless/wl12xx/wl1271_main.c index 3daba6c0c77f..6036d0206fec 100644 --- a/drivers/net/wireless/wl12xx/wl1271_main.c +++ b/drivers/net/wireless/wl12xx/wl1271_main.c @@ -1266,11 +1266,11 @@ struct wl1271_filter_params { u8 mc_list[ACX_MC_ADDRESS_GROUP_MAX][ETH_ALEN]; }; -static u64 wl1271_op_prepare_multicast(struct ieee80211_hw *hw, int mc_count, - struct dev_addr_list *mc_list) +static u64 wl1271_op_prepare_multicast(struct ieee80211_hw *hw, + struct netdev_hw_addr_list *mc_list) { struct wl1271_filter_params *fp; - int i; + struct netdev_hw_addr *ha; fp = kzalloc(sizeof(*fp), GFP_ATOMIC); if (!fp) { @@ -1279,21 +1279,16 @@ static u64 wl1271_op_prepare_multicast(struct ieee80211_hw *hw, int mc_count, } /* update multicast filtering parameters */ - fp->enabled = true; - if (mc_count > ACX_MC_ADDRESS_GROUP_MAX) { - mc_count = 0; - fp->enabled = false; - } - fp->mc_list_length = 0; - for (i = 0; i < mc_count; i++) { - if (mc_list->da_addrlen == ETH_ALEN) { + if (netdev_hw_addr_list_count(mc_list) > ACX_MC_ADDRESS_GROUP_MAX) { + fp->enabled = false; + } else { + fp->enabled = true; + netdev_hw_addr_list_for_each(ha, mc_list) { memcpy(fp->mc_list[fp->mc_list_length], - mc_list->da_addr, ETH_ALEN); + ha->addr, ETH_ALEN); fp->mc_list_length++; - } else - wl1271_warning("Unknown mc address length."); - mc_list = mc_list->next; + } } return (u64)(unsigned long)fp; diff --git a/drivers/net/wireless/zd1201.c b/drivers/net/wireless/zd1201.c index 6917286edcae..74530b2d672c 100644 --- a/drivers/net/wireless/zd1201.c +++ b/drivers/net/wireless/zd1201.c @@ -875,7 +875,7 @@ static struct iw_statistics *zd1201_get_wireless_stats(struct net_device *dev) static void zd1201_set_multicast(struct net_device *dev) { struct zd1201 *zd = netdev_priv(dev); - struct dev_mc_list *mc; + struct netdev_hw_addr *ha; unsigned char reqbuf[ETH_ALEN*ZD1201_MAXMULTI]; int i; @@ -883,8 +883,8 @@ static void zd1201_set_multicast(struct net_device *dev) return; i = 0; - netdev_for_each_mc_addr(mc, dev) - memcpy(reqbuf + i++ * ETH_ALEN, mc->dmi_addr, ETH_ALEN); + netdev_for_each_mc_addr(ha, dev) + memcpy(reqbuf + i++ * ETH_ALEN, ha->addr, ETH_ALEN); zd1201_setconfig(zd, ZD1201_RID_CNFGROUPADDRESS, reqbuf, netdev_mc_count(dev) * ETH_ALEN, 0); } diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c index 00e09e26c826..6d95e4d74d7f 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.c +++ b/drivers/net/wireless/zd1211rw/zd_mac.c @@ -947,20 +947,17 @@ static void set_rx_filter_handler(struct work_struct *work) } static u64 zd_op_prepare_multicast(struct ieee80211_hw *hw, - int mc_count, struct dev_addr_list *mclist) + struct netdev_hw_addr_list *mc_list) { struct zd_mac *mac = zd_hw_mac(hw); struct zd_mc_hash hash; - int i; + struct netdev_hw_addr *ha; zd_mc_clear(&hash); - for (i = 0; i < mc_count; i++) { - if (!mclist) - break; - dev_dbg_f(zd_mac_dev(mac), "mc addr %pM\n", mclist->dmi_addr); - zd_mc_add_addr(&hash, mclist->dmi_addr); - mclist = mclist->next; + netdev_hw_addr_list_for_each(ha, mc_list) { + dev_dbg_f(zd_mac_dev(mac), "mc addr %pM\n", ha->addr); + zd_mc_add_addr(&hash, ha->addr); } return hash.low | ((u64)hash.high << 32); diff --git a/drivers/net/yellowfin.c b/drivers/net/yellowfin.c index 7d4107f5eeb0..34c91cf5d839 100644 --- a/drivers/net/yellowfin.c +++ b/drivers/net/yellowfin.c @@ -1300,25 +1300,25 @@ static void set_rx_mode(struct net_device *dev) /* Too many to filter well, or accept all multicasts. */ iowrite16(0x000B, ioaddr + AddrMode); } else if (!netdev_mc_empty(dev)) { /* Must use the multicast hash table. */ - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; u16 hash_table[4]; int i; memset(hash_table, 0, sizeof(hash_table)); - netdev_for_each_mc_addr(mclist, dev) { + netdev_for_each_mc_addr(ha, dev) { unsigned int bit; /* Due to a bug in the early chip versions, multiple filter slots must be set for each address. */ if (yp->drv_flags & HasMulticastBug) { - bit = (ether_crc_le(3, mclist->dmi_addr) >> 3) & 0x3f; + bit = (ether_crc_le(3, ha->addr) >> 3) & 0x3f; hash_table[bit >> 4] |= (1 << bit); - bit = (ether_crc_le(4, mclist->dmi_addr) >> 3) & 0x3f; + bit = (ether_crc_le(4, ha->addr) >> 3) & 0x3f; hash_table[bit >> 4] |= (1 << bit); - bit = (ether_crc_le(5, mclist->dmi_addr) >> 3) & 0x3f; + bit = (ether_crc_le(5, ha->addr) >> 3) & 0x3f; hash_table[bit >> 4] |= (1 << bit); } - bit = (ether_crc_le(6, mclist->dmi_addr) >> 3) & 0x3f; + bit = (ether_crc_le(6, ha->addr) >> 3) & 0x3f; hash_table[bit >> 4] |= (1 << bit); } /* Copy the hash table to the chip. */ diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index 7576ad5a833a..945f3e0a9f06 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -607,7 +607,6 @@ static int qeth_l2_set_mac_address(struct net_device *dev, void *p) 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) @@ -619,8 +618,8 @@ static void qeth_l2_set_multicast_list(struct net_device *dev) return; qeth_l2_del_all_mc(card); spin_lock_bh(&card->mclock); - netdev_for_each_mc_addr(dm, dev) - qeth_l2_add_mc(card, dm->da_addr, 0); + netdev_for_each_mc_addr(ha, dev) + qeth_l2_add_mc(card, ha->addr, 0); netdev_for_each_uc_addr(ha, dev) qeth_l2_add_mc(card, ha->addr, 1); diff --git a/drivers/scsi/fcoe/fcoe.c b/drivers/scsi/fcoe/fcoe.c index de33e38a4059..37de40e01684 100644 --- a/drivers/scsi/fcoe/fcoe.c +++ b/drivers/scsi/fcoe/fcoe.c @@ -311,7 +311,7 @@ static int fcoe_interface_setup(struct fcoe_interface *fcoe, dev_uc_add(netdev, flogi_maddr); if (fip->spma) dev_uc_add(netdev, fip->ctl_src_addr); - dev_mc_add(netdev, FIP_ALL_ENODE_MACS, ETH_ALEN, 0); + dev_mc_add(netdev, FIP_ALL_ENODE_MACS); /* * setup the receive function from ethernet driver @@ -397,7 +397,7 @@ void fcoe_interface_cleanup(struct fcoe_interface *fcoe) dev_uc_del(netdev, flogi_maddr); if (fip->spma) dev_uc_del(netdev, fip->ctl_src_addr); - dev_mc_delete(netdev, FIP_ALL_ENODE_MACS, ETH_ALEN, 0); + dev_mc_del(netdev, FIP_ALL_ENODE_MACS); /* Tell the LLD we are done w/ FCoE */ ops = netdev->netdev_ops; diff --git a/drivers/staging/arlan/arlan-main.c b/drivers/staging/arlan/arlan-main.c index 88fdd53cf5d3..80284522c42b 100644 --- a/drivers/staging/arlan/arlan-main.c +++ b/drivers/staging/arlan/arlan-main.c @@ -1458,7 +1458,7 @@ static void arlan_rx_interrupt(struct net_device *dev, u_char rxStatus, u_short !netdev_mc_empty(dev)) { char hw_dst_addr[6]; - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int i; memcpy_fromio(hw_dst_addr, arlan->ultimateDestAddress, 6); @@ -1469,12 +1469,13 @@ static void arlan_rx_interrupt(struct net_device *dev, u_char rxStatus, u_short printk(KERN_ERR "%s mcast 0x0100 \n", dev->name); else if (hw_dst_addr[1] == 0x40) printk(KERN_ERR "%s m/bcast 0x0140 \n", dev->name); - netdev_for_each_mc_entry(dmi, dev) { + netdev_for_each_mc_entry(ha, dev) { if (arlan_debug & ARLAN_DEBUG_HEADER_DUMP) printk(KERN_ERR "%s mcl %pM\n", - dev->name, dmi->dmi_addr); + dev->name, + ha->addr); for (i = 0; i < 6; i++) - if (dmi->dmi_addr[i] != hw_dst_addr[i]) + if (ha->addr[i] != hw_dst_addr[i]) break; if (i == 6) break; diff --git a/drivers/staging/et131x/et131x_netdev.c b/drivers/staging/et131x/et131x_netdev.c index 40f8954dde47..2fb89cddef16 100644 --- a/drivers/staging/et131x/et131x_netdev.c +++ b/drivers/staging/et131x/et131x_netdev.c @@ -405,7 +405,7 @@ void et131x_multicast(struct net_device *netdev) struct et131x_adapter *adapter = netdev_priv(netdev); uint32_t PacketFilter = 0; unsigned long flags; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; int i; spin_lock_irqsave(&adapter->Lock, flags); @@ -450,10 +450,10 @@ void et131x_multicast(struct net_device *netdev) /* Set values in the private adapter struct */ i = 0; - netdev_for_each_mc_addr(mclist, netdev) { + netdev_for_each_mc_addr(ha, netdev) { if (i == NIC_MAX_MCAST_LIST) break; - memcpy(adapter->MCList[i++], mclist->dmi_addr, ETH_ALEN); + memcpy(adapter->MCList[i++], ha->addr, ETH_ALEN); } adapter->MCAddressCount = i; diff --git a/drivers/staging/slicoss/slicoss.c b/drivers/staging/slicoss/slicoss.c index 7daeced317c4..bebf0fd2af85 100644 --- a/drivers/staging/slicoss/slicoss.c +++ b/drivers/staging/slicoss/slicoss.c @@ -1367,12 +1367,12 @@ static void slic_mcast_set_list(struct net_device *dev) struct adapter *adapter = netdev_priv(dev); int status = STATUS_SUCCESS; char *addresses; - struct dev_mc_list *mc_list; + struct netdev_hw_addr *ha; ASSERT(adapter); - netdev_for_each_mc_addr(mc_list, dev) { - addresses = (char *) &mc_list->dmi_addr; + netdev_for_each_mc_addr(ha, dev) { + addresses = (char *) &ha->addr; status = slic_mcast_add_list(adapter, addresses); if (status != STATUS_SUCCESS) break; diff --git a/drivers/staging/vt6655/device_main.c b/drivers/staging/vt6655/device_main.c index 1d643653a7ed..b698de40a60e 100644 --- a/drivers/staging/vt6655/device_main.c +++ b/drivers/staging/vt6655/device_main.c @@ -3079,7 +3079,7 @@ static void device_set_multi(struct net_device *dev) { PSMgmtObject pMgmt = pDevice->pMgmt; u32 mc_filter[2]; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; VNSvInPortB(pDevice->PortOffset + MAC_REG_RCR, &(pDevice->byRxMode)); @@ -3099,8 +3099,8 @@ static void device_set_multi(struct net_device *dev) { } else { memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { - int bit_nr = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, dev) { + int bit_nr = ether_crc(ETH_ALEN, ha->addr) >> 26; mc_filter[bit_nr >> 5] |= cpu_to_le32(1 << (bit_nr & 31)); } MACvSelectPage1(pDevice->PortOffset); diff --git a/drivers/staging/vt6656/main_usb.c b/drivers/staging/vt6656/main_usb.c index a8e1adbc9592..49270db98fbb 100644 --- a/drivers/staging/vt6656/main_usb.c +++ b/drivers/staging/vt6656/main_usb.c @@ -1596,7 +1596,7 @@ static void device_set_multi(struct net_device *dev) { PSMgmtObject pMgmt = &(pDevice->sMgmtObj); u32 mc_filter[2]; int ii; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; BYTE pbyData[8] = {0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff}; BYTE byTmpMode = 0; int rc; @@ -1632,8 +1632,8 @@ static void device_set_multi(struct net_device *dev) { } else { memset(mc_filter, 0, sizeof(mc_filter)); - netdev_for_each_mc_addr(mclist, dev) { - int bit_nr = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26; + netdev_for_each_mc_addr(ha, dev) { + int bit_nr = ether_crc(ETH_ALEN, ha->addr) >> 26; mc_filter[bit_nr >> 5] |= cpu_to_le32(1 << (bit_nr & 31)); } for (ii = 0; ii < 4; ii++) { diff --git a/drivers/staging/wavelan/wavelan.c b/drivers/staging/wavelan/wavelan.c index 54ca63196fdd..f44ef351647b 100644 --- a/drivers/staging/wavelan/wavelan.c +++ b/drivers/staging/wavelan/wavelan.c @@ -3419,7 +3419,7 @@ static void wv_82586_config(struct net_device * dev) ac_cfg_t cfg; /* Configure action */ ac_ias_t ias; /* IA-setup action */ ac_mcs_t mcs; /* Multicast setup */ - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; #ifdef DEBUG_CONFIG_TRACE printk(KERN_DEBUG "%s: ->wv_82586_config()\n", dev->name); @@ -3531,16 +3531,16 @@ static void wv_82586_config(struct net_device * dev) /* Any address to set? */ if (lp->mc_count) { - netdev_for_each_mc_addr(dmi, dev) - outsw(PIOP1(ioaddr), (u16 *) dmi->dmi_addr, + netdev_for_each_mc_addr(ha, dev) + outsw(PIOP1(ioaddr), (u16 *) ha->addr, WAVELAN_ADDR_SIZE >> 1); #ifdef DEBUG_CONFIG_INFO printk(KERN_DEBUG "%s: wv_82586_config(): set %d multicast addresses:\n", dev->name, lp->mc_count); - netdev_for_each_mc_addr(dmi, dev) - printk(KERN_DEBUG " %pM\n", dmi->dmi_addr); + netdev_for_each_mc_addr(ha, dev) + printk(KERN_DEBUG " %pM\n", ha->addr); #endif } diff --git a/drivers/staging/wavelan/wavelan_cs.c b/drivers/staging/wavelan/wavelan_cs.c index 04f691d127b4..a90132a204e6 100644 --- a/drivers/staging/wavelan/wavelan_cs.c +++ b/drivers/staging/wavelan/wavelan_cs.c @@ -3591,20 +3591,20 @@ wv_82593_config(struct net_device * dev) /* If roaming is enabled, join the "Beacon Request" multicast group... */ /* But only if it's not in there already! */ if(do_roaming) - dev_mc_add(dev,WAVELAN_BEACON_ADDRESS, WAVELAN_ADDR_SIZE, 1); + dev_mc_add(dev, WAVELAN_BEACON_ADDRESS); #endif /* WAVELAN_ROAMING */ /* If any multicast address to set */ if(lp->mc_count) { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int addrs_len = WAVELAN_ADDR_SIZE * lp->mc_count; #ifdef DEBUG_CONFIG_INFO printk(KERN_DEBUG "%s: wv_hw_config(): set %d multicast addresses:\n", dev->name, lp->mc_count); - netdev_for_each_mc_addr(dmi, dev) - printk(KERN_DEBUG " %pM\n", dmi->dmi_addr); + netdev_for_each_mc_addr(ha, dev) + printk(KERN_DEBUG " %pM\n", ha->addr); #endif /* Initialize adapter's ethernet multicast addresses */ @@ -3612,8 +3612,8 @@ wv_82593_config(struct net_device * dev) outb(((TX_BASE >> 8) & PIORH_MASK) | PIORH_SEL_TX, PIORH(base)); outb(addrs_len & 0xff, PIOP(base)); /* byte count lsb */ outb((addrs_len >> 8), PIOP(base)); /* byte count msb */ - netdev_for_each_mc_addr(dmi, dev) - outsb(PIOP(base), dmi->dmi_addr, dmi->dmi_addrlen); + netdev_for_each_mc_addr(ha, dev) + outsb(PIOP(base), ha->addr, dev->addr_len); /* reset transmit DMA pointer */ hacr_write_slow(base, HACR_PWR_STAT | HACR_TX_DMA_RESET); diff --git a/drivers/staging/winbond/wbusb.c b/drivers/staging/winbond/wbusb.c index 3482eec18651..5d9499bba9cc 100644 --- a/drivers/staging/winbond/wbusb.c +++ b/drivers/staging/winbond/wbusb.c @@ -92,10 +92,10 @@ static int wbsoft_get_stats(struct ieee80211_hw *hw, return 0; } -static u64 wbsoft_prepare_multicast(struct ieee80211_hw *hw, int mc_count, - struct dev_addr_list *mc_list) +static u64 wbsoft_prepare_multicast(struct ieee80211_hw *hw, + struct netdev_hw_addr_list *mc_list) { - return mc_count; + return netdev_hw_addr_list_count(mc_list); } static void wbsoft_configure_filter(struct ieee80211_hw *dev, diff --git a/drivers/staging/wlags49_h2/wl_netdev.c b/drivers/staging/wlags49_h2/wl_netdev.c index fa082d90fcad..d7532e89f5fc 100644 --- a/drivers/staging/wlags49_h2/wl_netdev.c +++ b/drivers/staging/wlags49_h2/wl_netdev.c @@ -1049,7 +1049,7 @@ void wl_multicast( struct net_device *dev ) //;?seems reasonable that even an AP-only driver could afford this small additional footprint int x; - struct dev_mc_list *mclist; + struct netdev_hw_addr *ha; struct wl_private *lp = wl_priv(dev); unsigned long flags; /*------------------------------------------------------------------------*/ @@ -1072,9 +1072,9 @@ void wl_multicast( struct net_device *dev ) DBG_PRINT( " mc_count: %d\n", netdev_mc_count(dev)); - netdev_for_each_mc_addr(mclist, dev) - DBG_PRINT( " %s (%d)\n", DbgHwAddr(mclist->dmi_addr), - mclist->dmi_addrlen ); + netdev_for_each_mc_addr(ha, dev) + DBG_PRINT(" %s (%d)\n", DbgHwAddr(ha->addr), + dev->addr_len); } #endif /* DBG */ @@ -1119,9 +1119,9 @@ void wl_multicast( struct net_device *dev ) lp->ltvRecord.typ = CFG_GROUP_ADDR; x = 0; - netdev_for_each_mc_addr(mclist, dev) + netdev_for_each_mc_addr(ha, dev) memcpy(&(lp->ltvRecord.u.u8[x++ * ETH_ALEN]), - mclist->dmi_addr, ETH_ALEN); + ha->addr, ETH_ALEN); DBG_PRINT( "Setting multicast list\n" ); hcf_put_info( &( lp->hcfCtx ), (LTVP)&( lp->ltvRecord )); } else { diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 60f0c83192fe..a343a21ba8b9 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -228,25 +228,6 @@ struct netif_rx_stats { DECLARE_PER_CPU(struct netif_rx_stats, netdev_rx_stat); -struct dev_addr_list { - struct dev_addr_list *next; - u8 da_addr[MAX_ADDR_LEN]; - u8 da_addrlen; - u8 da_synced; - int da_users; - int da_gusers; -}; - -/* - * We tag multicasts with these structures. - */ - -#define dev_mc_list dev_addr_list -#define dmi_addr da_addr -#define dmi_addrlen da_addrlen -#define dmi_users da_users -#define dmi_gusers da_gusers - struct netdev_hw_addr { struct list_head list; unsigned char addr[MAX_ADDR_LEN]; @@ -255,8 +236,10 @@ struct netdev_hw_addr { #define NETDEV_HW_ADDR_T_SAN 2 #define NETDEV_HW_ADDR_T_SLAVE 3 #define NETDEV_HW_ADDR_T_UNICAST 4 +#define NETDEV_HW_ADDR_T_MULTICAST 5 int refcount; bool synced; + bool global_use; struct rcu_head rcu_head; }; @@ -265,16 +248,20 @@ struct netdev_hw_addr_list { int count; }; -#define netdev_uc_count(dev) ((dev)->uc.count) -#define netdev_uc_empty(dev) ((dev)->uc.count == 0) -#define netdev_for_each_uc_addr(ha, dev) \ - list_for_each_entry(ha, &dev->uc.list, list) +#define netdev_hw_addr_list_count(l) ((l)->count) +#define netdev_hw_addr_list_empty(l) (netdev_hw_addr_list_count(l) == 0) +#define netdev_hw_addr_list_for_each(ha, l) \ + list_for_each_entry(ha, &(l)->list, list) -#define netdev_mc_count(dev) ((dev)->mc_count) -#define netdev_mc_empty(dev) (netdev_mc_count(dev) == 0) +#define netdev_uc_count(dev) netdev_hw_addr_list_count(&(dev)->uc) +#define netdev_uc_empty(dev) netdev_hw_addr_list_empty(&(dev)->uc) +#define netdev_for_each_uc_addr(ha, dev) \ + netdev_hw_addr_list_for_each(ha, &(dev)->uc) +#define netdev_mc_count(dev) netdev_hw_addr_list_count(&(dev)->mc) +#define netdev_mc_empty(dev) netdev_hw_addr_list_empty(&(dev)->mc) #define netdev_for_each_mc_addr(mclist, dev) \ - for (mclist = dev->mc_list; mclist; mclist = mclist->next) + netdev_hw_addr_list_for_each(ha, &(dev)->mc) struct hh_cache { struct hh_cache *hh_next; /* Next entry */ @@ -862,12 +849,10 @@ struct net_device { unsigned char addr_len; /* hardware address length */ unsigned short dev_id; /* for shared network cards */ - 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 */ - int mc_count; /* Number of installed mcasts */ + struct netdev_hw_addr_list uc; /* Unicast mac addresses */ + struct netdev_hw_addr_list mc; /* Multicast mac addresses */ + int uc_promisc; unsigned int promiscuity; unsigned int allmulti; @@ -1980,6 +1965,22 @@ extern struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, extern int register_netdev(struct net_device *dev); extern void unregister_netdev(struct net_device *dev); +/* General hardware address lists handling functions */ +extern 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); +extern 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); +extern int __hw_addr_sync(struct netdev_hw_addr_list *to_list, + struct netdev_hw_addr_list *from_list, + int addr_len); +extern void __hw_addr_unsync(struct netdev_hw_addr_list *to_list, + struct netdev_hw_addr_list *from_list, + int addr_len); +extern void __hw_addr_flush(struct netdev_hw_addr_list *list); +extern void __hw_addr_init(struct netdev_hw_addr_list *list); + /* Functions used for device addresses handling */ extern int dev_addr_add(struct net_device *dev, unsigned char *addr, unsigned char addr_type); @@ -2002,18 +2003,19 @@ extern void dev_uc_unsync(struct net_device *to, struct net_device *from); extern void dev_uc_flush(struct net_device *dev); extern void dev_uc_init(struct net_device *dev); +/* Functions used for multicast addresses handling */ +extern int dev_mc_add(struct net_device *dev, unsigned char *addr); +extern int dev_mc_add_global(struct net_device *dev, unsigned char *addr); +extern int dev_mc_del(struct net_device *dev, unsigned char *addr); +extern int dev_mc_del_global(struct net_device *dev, unsigned char *addr); +extern int dev_mc_sync(struct net_device *to, struct net_device *from); +extern void dev_mc_unsync(struct net_device *to, struct net_device *from); +extern void dev_mc_flush(struct net_device *dev); +extern void dev_mc_init(struct net_device *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_mc_delete(struct net_device *dev, void *addr, int alen, int all); -extern int dev_mc_add(struct net_device *dev, void *addr, int alen, int newonly); -extern int dev_mc_sync(struct net_device *to, struct net_device *from); -extern void dev_mc_unsync(struct net_device *to, struct net_device *from); -extern void dev_addr_discard(struct net_device *dev); -extern int __dev_addr_delete(struct dev_addr_list **list, int *count, void *addr, int alen, int all); -extern int __dev_addr_add(struct dev_addr_list **list, int *count, void *addr, int alen, int newonly); -extern int __dev_addr_sync(struct dev_addr_list **to, int *to_count, struct dev_addr_list **from, int *from_count); -extern void __dev_addr_unsync(struct dev_addr_list **to, int *to_count, struct dev_addr_list **from, int *from_count); extern int dev_set_promiscuity(struct net_device *dev, int inc); extern int dev_set_allmulti(struct net_device *dev, int inc); extern void netdev_state_change(struct net_device *dev); diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 1a8f50af49a0..20823d04e03c 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1626,7 +1626,7 @@ struct ieee80211_ops { struct ieee80211_bss_conf *info, u32 changed); u64 (*prepare_multicast)(struct ieee80211_hw *hw, - int mc_count, struct dev_addr_list *mc_list); + struct netdev_hw_addr_list *mc_list); void (*configure_filter)(struct ieee80211_hw *hw, unsigned int changed_flags, unsigned int *total_flags, diff --git a/net/802/garp.c b/net/802/garp.c index 1dcb0660c49d..78cff9ec2cb4 100644 --- a/net/802/garp.c +++ b/net/802/garp.c @@ -575,7 +575,7 @@ int garp_init_applicant(struct net_device *dev, struct garp_application *appl) if (!app) goto err2; - err = dev_mc_add(dev, appl->proto.group_address, ETH_ALEN, 0); + err = dev_mc_add(dev, appl->proto.group_address); if (err < 0) goto err3; @@ -615,7 +615,7 @@ void garp_uninit_applicant(struct net_device *dev, struct garp_application *appl garp_pdu_queue(app); garp_queue_xmit(app); - dev_mc_delete(dev, appl->proto.group_address, ETH_ALEN, 0); + dev_mc_del(dev, appl->proto.group_address); kfree(app); garp_release_port(dev); } diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 9fc4da56fb1d..1d15a60b23af 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -781,7 +781,7 @@ static int atif_ioctl(int cmd, void __user *arg) atrtr_create(&rtdef, dev); } } - dev_mc_add(dev, aarp_mcast, 6, 1); + dev_mc_add_global(dev, aarp_mcast); return 0; case SIOCGIFADDR: diff --git a/net/bluetooth/bnep/netdev.c b/net/bluetooth/bnep/netdev.c index 326ab453edb7..260a9507e542 100644 --- a/net/bluetooth/bnep/netdev.c +++ b/net/bluetooth/bnep/netdev.c @@ -87,7 +87,7 @@ static void bnep_net_set_mc_list(struct net_device *dev) memcpy(__skb_put(skb, ETH_ALEN), dev->broadcast, ETH_ALEN); r->len = htons(ETH_ALEN * 2); } else { - struct dev_mc_list *dmi; + struct netdev_hw_addr *ha; int i, len = skb->len; if (dev->flags & IFF_BROADCAST) { @@ -98,11 +98,11 @@ static void bnep_net_set_mc_list(struct net_device *dev) /* FIXME: We should group addresses here. */ i = 0; - netdev_for_each_mc_addr(dmi, dev) { + netdev_for_each_mc_addr(ha, dev) { if (i == BNEP_MAX_MULTICAST_FILTERS) break; - memcpy(__skb_put(skb, ETH_ALEN), dmi->dmi_addr, ETH_ALEN); - memcpy(__skb_put(skb, ETH_ALEN), dmi->dmi_addr, ETH_ALEN); + memcpy(__skb_put(skb, ETH_ALEN), ha->addr, ETH_ALEN); + memcpy(__skb_put(skb, ETH_ALEN), ha->addr, ETH_ALEN); } r->len = htons(skb->len - len); } diff --git a/net/core/Makefile b/net/core/Makefile index 0a899f1aadb9..51c3eec850ef 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -7,9 +7,8 @@ obj-y := sock.o request_sock.o skbuff.o iovec.o datagram.o stream.o scm.o \ obj-$(CONFIG_SYSCTL) += sysctl_net_core.o -obj-y += dev.o ethtool.o dev_mcast.o dst.o netevent.o \ - neighbour.o rtnetlink.o utils.o link_watch.o filter.o \ - dev_addr_lists.o +obj-y += dev.o ethtool.o dev_addr_lists.o dst.o netevent.o \ + neighbour.o rtnetlink.o utils.o link_watch.o filter.o obj-$(CONFIG_XFRM) += flow.o obj-y += net-sysfs.o diff --git a/net/core/dev.c b/net/core/dev.c index 949c62dba719..2a9b7dd0bb6e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3968,140 +3968,6 @@ void dev_set_rx_mode(struct net_device *dev) netif_addr_unlock_bh(dev); } -/* multicast addresses handling functions */ - -int __dev_addr_delete(struct dev_addr_list **list, int *count, - void *addr, int alen, int glbl) -{ - struct dev_addr_list *da; - - for (; (da = *list) != NULL; list = &da->next) { - if (memcmp(da->da_addr, addr, da->da_addrlen) == 0 && - alen == da->da_addrlen) { - if (glbl) { - int old_glbl = da->da_gusers; - da->da_gusers = 0; - if (old_glbl == 0) - break; - } - if (--da->da_users) - return 0; - - *list = da->next; - kfree(da); - (*count)--; - return 0; - } - } - return -ENOENT; -} - -int __dev_addr_add(struct dev_addr_list **list, int *count, - void *addr, int alen, int glbl) -{ - struct dev_addr_list *da; - - for (da = *list; da != NULL; da = da->next) { - if (memcmp(da->da_addr, addr, da->da_addrlen) == 0 && - da->da_addrlen == alen) { - if (glbl) { - int old_glbl = da->da_gusers; - da->da_gusers = 1; - if (old_glbl) - return 0; - } - da->da_users++; - return 0; - } - } - - da = kzalloc(sizeof(*da), GFP_ATOMIC); - if (da == NULL) - return -ENOMEM; - memcpy(da->da_addr, addr, alen); - da->da_addrlen = alen; - da->da_users = 1; - da->da_gusers = glbl ? 1 : 0; - da->next = *list; - *list = da; - (*count)++; - return 0; -} - - -int __dev_addr_sync(struct dev_addr_list **to, int *to_count, - struct dev_addr_list **from, int *from_count) -{ - struct dev_addr_list *da, *next; - int err = 0; - - da = *from; - while (da != NULL) { - next = da->next; - if (!da->da_synced) { - err = __dev_addr_add(to, to_count, - da->da_addr, da->da_addrlen, 0); - if (err < 0) - break; - da->da_synced = 1; - da->da_users++; - } else if (da->da_users == 1) { - __dev_addr_delete(to, to_count, - da->da_addr, da->da_addrlen, 0); - __dev_addr_delete(from, from_count, - da->da_addr, da->da_addrlen, 0); - } - da = next; - } - return err; -} -EXPORT_SYMBOL_GPL(__dev_addr_sync); - -void __dev_addr_unsync(struct dev_addr_list **to, int *to_count, - struct dev_addr_list **from, int *from_count) -{ - struct dev_addr_list *da, *next; - - da = *from; - while (da != NULL) { - next = da->next; - if (da->da_synced) { - __dev_addr_delete(to, to_count, - da->da_addr, da->da_addrlen, 0); - da->da_synced = 0; - __dev_addr_delete(from, from_count, - da->da_addr, da->da_addrlen, 0); - } - da = next; - } -} -EXPORT_SYMBOL_GPL(__dev_addr_unsync); - -static void __dev_addr_discard(struct dev_addr_list **list) -{ - struct dev_addr_list *tmp; - - while (*list != NULL) { - tmp = *list; - *list = tmp->next; - if (tmp->da_users > tmp->da_gusers) - printk("__dev_addr_discard: address leakage! " - "da_users=%d\n", tmp->da_users); - kfree(tmp); - } -} - -void dev_addr_discard(struct net_device *dev) -{ - netif_addr_lock_bh(dev); - - __dev_addr_discard(&dev->mc_list); - netdev_mc_count(dev) = 0; - - netif_addr_unlock_bh(dev); -} -EXPORT_SYMBOL(dev_addr_discard); - /** * dev_get_flags - get flags reported to userspace * @dev: device @@ -4412,8 +4278,7 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, unsigned int cmd) return -EINVAL; if (!netif_device_present(dev)) return -ENODEV; - return dev_mc_add(dev, ifr->ifr_hwaddr.sa_data, - dev->addr_len, 1); + return dev_mc_add_global(dev, ifr->ifr_hwaddr.sa_data); case SIOCDELMULTI: if ((!ops->ndo_set_multicast_list && !ops->ndo_set_rx_mode) || @@ -4421,8 +4286,7 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, unsigned int cmd) return -EINVAL; if (!netif_device_present(dev)) return -ENODEV; - return dev_mc_delete(dev, ifr->ifr_hwaddr.sa_data, - dev->addr_len, 1); + return dev_mc_del_global(dev, ifr->ifr_hwaddr.sa_data); case SIOCSIFTXQLEN: if (ifr->ifr_qlen < 0) @@ -4730,7 +4594,7 @@ static void rollback_registered_many(struct list_head *head) * Flush the unicast and multicast chains */ dev_uc_flush(dev); - dev_addr_discard(dev); + dev_mc_flush(dev); if (dev->netdev_ops->ndo_uninit) dev->netdev_ops->ndo_uninit(dev); @@ -5310,6 +5174,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, if (dev_addr_init(dev)) goto free_rx; + dev_mc_init(dev); dev_uc_init(dev); dev_net_set(dev, &init_net); @@ -5545,7 +5410,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char * Flush the unicast and multicast chains */ dev_uc_flush(dev); - dev_addr_discard(dev); + dev_mc_flush(dev); netdev_unregister_kobject(dev); diff --git a/net/core/dev_addr_lists.c b/net/core/dev_addr_lists.c index 7e52b6d18add..37d5975e18a3 100644 --- a/net/core/dev_addr_lists.c +++ b/net/core/dev_addr_lists.c @@ -19,8 +19,9 @@ * General list handling functions */ -static int __hw_addr_add(struct netdev_hw_addr_list *list, unsigned char *addr, - int addr_len, unsigned char addr_type) +static int __hw_addr_add_ex(struct netdev_hw_addr_list *list, + unsigned char *addr, int addr_len, + unsigned char addr_type, bool global) { struct netdev_hw_addr *ha; int alloc_size; @@ -31,6 +32,13 @@ static int __hw_addr_add(struct netdev_hw_addr_list *list, unsigned char *addr, list_for_each_entry(ha, &list->list, list) { if (!memcmp(ha->addr, addr, addr_len) && ha->type == addr_type) { + if (global) { + /* check if addr is already used as global */ + if (ha->global_use) + return 0; + else + ha->global_use = true; + } ha->refcount++; return 0; } @@ -46,12 +54,19 @@ static int __hw_addr_add(struct netdev_hw_addr_list *list, unsigned char *addr, memcpy(ha->addr, addr, addr_len); ha->type = addr_type; ha->refcount = 1; + ha->global_use = global; ha->synced = false; list_add_tail_rcu(&ha->list, &list->list); list->count++; return 0; } +static int __hw_addr_add(struct netdev_hw_addr_list *list, unsigned char *addr, + int addr_len, unsigned char addr_type) +{ + return __hw_addr_add_ex(list, addr, addr_len, addr_type, false); +} + static void ha_rcu_free(struct rcu_head *head) { struct netdev_hw_addr *ha; @@ -60,14 +75,21 @@ static void ha_rcu_free(struct rcu_head *head) kfree(ha); } -static int __hw_addr_del(struct netdev_hw_addr_list *list, unsigned char *addr, - int addr_len, unsigned char addr_type) +static int __hw_addr_del_ex(struct netdev_hw_addr_list *list, + unsigned char *addr, int addr_len, + unsigned char addr_type, bool global) { struct netdev_hw_addr *ha; list_for_each_entry(ha, &list->list, list) { if (!memcmp(ha->addr, addr, addr_len) && (ha->type == addr_type || !addr_type)) { + if (global) { + if (!ha->global_use) + break; + else + ha->global_use = false; + } if (--ha->refcount) return 0; list_del_rcu(&ha->list); @@ -79,10 +101,15 @@ static int __hw_addr_del(struct netdev_hw_addr_list *list, unsigned char *addr, return -ENOENT; } -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) +static int __hw_addr_del(struct netdev_hw_addr_list *list, unsigned char *addr, + int addr_len, unsigned char addr_type) +{ + return __hw_addr_del_ex(list, addr, addr_len, addr_type, false); +} + +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; @@ -105,11 +132,11 @@ unroll: } return err; } +EXPORT_SYMBOL(__hw_addr_add_multiple); -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) +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; @@ -119,10 +146,11 @@ static void __hw_addr_del_multiple(struct netdev_hw_addr_list *to_list, __hw_addr_del(to_list, ha->addr, addr_len, addr_type); } } +EXPORT_SYMBOL(__hw_addr_del_multiple); -static int __hw_addr_sync(struct netdev_hw_addr_list *to_list, - struct netdev_hw_addr_list *from_list, - int addr_len) +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; @@ -142,10 +170,11 @@ static int __hw_addr_sync(struct netdev_hw_addr_list *to_list, } return err; } +EXPORT_SYMBOL(__hw_addr_sync); -static void __hw_addr_unsync(struct netdev_hw_addr_list *to_list, - struct netdev_hw_addr_list *from_list, - int addr_len) +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; @@ -159,8 +188,9 @@ static void __hw_addr_unsync(struct netdev_hw_addr_list *to_list, } } } +EXPORT_SYMBOL(__hw_addr_unsync); -static void __hw_addr_flush(struct netdev_hw_addr_list *list) +void __hw_addr_flush(struct netdev_hw_addr_list *list) { struct netdev_hw_addr *ha, *tmp; @@ -170,12 +200,14 @@ static void __hw_addr_flush(struct netdev_hw_addr_list *list) } list->count = 0; } +EXPORT_SYMBOL(__hw_addr_flush); -static void __hw_addr_init(struct netdev_hw_addr_list *list) +void __hw_addr_init(struct netdev_hw_addr_list *list) { INIT_LIST_HEAD(&list->list); list->count = 0; } +EXPORT_SYMBOL(__hw_addr_init); /* * Device addresses handling functions @@ -475,4 +507,235 @@ EXPORT_SYMBOL(dev_uc_init); * Multicast list handling functions */ -/* To be filled here */ +static int __dev_mc_add(struct net_device *dev, unsigned char *addr, + bool global) +{ + int err; + + netif_addr_lock_bh(dev); + err = __hw_addr_add_ex(&dev->mc, addr, dev->addr_len, + NETDEV_HW_ADDR_T_MULTICAST, global); + if (!err) + __dev_set_rx_mode(dev); + netif_addr_unlock_bh(dev); + return err; +} +/** + * dev_mc_add - Add a multicast address + * @dev: device + * @addr: address to add + * + * Add a multicast address to the device or increase + * the reference count if it already exists. + */ +int dev_mc_add(struct net_device *dev, unsigned char *addr) +{ + return __dev_mc_add(dev, addr, false); +} +EXPORT_SYMBOL(dev_mc_add); + +/** + * dev_mc_add_global - Add a global multicast address + * @dev: device + * @addr: address to add + * + * Add a global multicast address to the device. + */ +int dev_mc_add_global(struct net_device *dev, unsigned char *addr) +{ + return __dev_mc_add(dev, addr, true); +} +EXPORT_SYMBOL(dev_mc_add_global); + +static int __dev_mc_del(struct net_device *dev, unsigned char *addr, + bool global) +{ + int err; + + netif_addr_lock_bh(dev); + err = __hw_addr_del_ex(&dev->mc, addr, dev->addr_len, + NETDEV_HW_ADDR_T_MULTICAST, global); + if (!err) + __dev_set_rx_mode(dev); + netif_addr_unlock_bh(dev); + return err; +} + +/** + * dev_mc_del - Delete a multicast address. + * @dev: device + * @addr: address to delete + * + * Release reference to a multicast address and remove it + * from the device if the reference count drops to zero. + */ +int dev_mc_del(struct net_device *dev, unsigned char *addr) +{ + return __dev_mc_del(dev, addr, false); +} +EXPORT_SYMBOL(dev_mc_del); + +/** + * dev_mc_del_global - Delete a global multicast address. + * @dev: device + * @addr: address to delete + * + * Release reference to a multicast address and remove it + * from the device if the reference count drops to zero. + */ +int dev_mc_del_global(struct net_device *dev, unsigned char *addr) +{ + return __dev_mc_del(dev, addr, true); +} +EXPORT_SYMBOL(dev_mc_del_global); + +/** + * dev_mc_sync - Synchronize device's unicast list to another device + * @to: destination device + * @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. + * + * This function is intended to be called from the dev->set_multicast_list + * or dev->set_rx_mode function of layered software devices. + */ +int dev_mc_sync(struct net_device *to, struct net_device *from) +{ + int err = 0; + + if (to->addr_len != from->addr_len) + return -EINVAL; + + netif_addr_lock_bh(to); + err = __hw_addr_sync(&to->mc, &from->mc, to->addr_len); + if (!err) + __dev_set_rx_mode(to); + netif_addr_unlock_bh(to); + return err; +} +EXPORT_SYMBOL(dev_mc_sync); + +/** + * dev_mc_unsync - Remove synchronized addresses from the destination device + * @to: destination device + * @from: source device + * + * Remove all addresses that were added to the destination device by + * dev_mc_sync(). This function is intended to be called from the + * dev->stop function of layered software devices. + */ +void dev_mc_unsync(struct net_device *to, struct net_device *from) +{ + if (to->addr_len != from->addr_len) + return; + + netif_addr_lock_bh(from); + netif_addr_lock(to); + __hw_addr_unsync(&to->mc, &from->mc, to->addr_len); + __dev_set_rx_mode(to); + netif_addr_unlock(to); + netif_addr_unlock_bh(from); +} +EXPORT_SYMBOL(dev_mc_unsync); + +/** + * dev_mc_flush - Flush multicast addresses + * @dev: device + * + * Flush multicast addresses. + */ +void dev_mc_flush(struct net_device *dev) +{ + netif_addr_lock_bh(dev); + __hw_addr_flush(&dev->mc); + netif_addr_unlock_bh(dev); +} +EXPORT_SYMBOL(dev_mc_flush); + +/** + * dev_mc_flush - Init multicast address list + * @dev: device + * + * Init multicast address list. + */ +void dev_mc_init(struct net_device *dev) +{ + __hw_addr_init(&dev->mc); +} +EXPORT_SYMBOL(dev_mc_init); + +#ifdef CONFIG_PROC_FS +#include +#include + +static int dev_mc_seq_show(struct seq_file *seq, void *v) +{ + struct netdev_hw_addr *ha; + struct net_device *dev = v; + + if (v == SEQ_START_TOKEN) + return 0; + + netif_addr_lock_bh(dev); + netdev_for_each_mc_addr(ha, dev) { + int i; + + seq_printf(seq, "%-4d %-15s %-5d %-5d ", dev->ifindex, + dev->name, ha->refcount, ha->global_use); + + for (i = 0; i < dev->addr_len; i++) + seq_printf(seq, "%02x", ha->addr[i]); + + seq_putc(seq, '\n'); + } + netif_addr_unlock_bh(dev); + return 0; +} + +static const struct seq_operations dev_mc_seq_ops = { + .start = dev_seq_start, + .next = dev_seq_next, + .stop = dev_seq_stop, + .show = dev_mc_seq_show, +}; + +static int dev_mc_seq_open(struct inode *inode, struct file *file) +{ + return seq_open_net(inode, file, &dev_mc_seq_ops, + sizeof(struct seq_net_private)); +} + +static const struct file_operations dev_mc_seq_fops = { + .owner = THIS_MODULE, + .open = dev_mc_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_net, +}; + +#endif + +static int __net_init dev_mc_net_init(struct net *net) +{ + if (!proc_net_fops_create(net, "dev_mcast", 0, &dev_mc_seq_fops)) + return -ENOMEM; + return 0; +} + +static void __net_exit dev_mc_net_exit(struct net *net) +{ + proc_net_remove(net, "dev_mcast"); +} + +static struct pernet_operations __net_initdata dev_mc_net_ops = { + .init = dev_mc_net_init, + .exit = dev_mc_net_exit, +}; + +void __init dev_mcast_init(void) +{ + register_pernet_subsys(&dev_mc_net_ops); +} + diff --git a/net/core/dev_mcast.c b/net/core/dev_mcast.c deleted file mode 100644 index 3dc295beb483..000000000000 --- a/net/core/dev_mcast.c +++ /dev/null @@ -1,232 +0,0 @@ -/* - * Linux NET3: Multicast List maintenance. - * - * Authors: - * Tim Kordas - * Richard Underwood - * - * Stir fried together from the IP multicast and CAP patches above - * Alan Cox - * - * Fixes: - * Alan Cox : Update the device on a real delete - * rather than any time but... - * Alan Cox : IFF_ALLMULTI support. - * Alan Cox : New format set_multicast_list() calls. - * Gleb Natapov : Remove dev_mc_lock. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - - -/* - * Device multicast list maintenance. - * - * This is used both by IP and by the user level maintenance functions. - * Unlike BSD we maintain a usage count on a given multicast address so - * that a casual user application can add/delete multicasts used by - * protocols without doing damage to the protocols when it deletes the - * entries. It also helps IP as it tracks overlapping maps. - * - * Device mc lists are changed by bh at least if IPv6 is enabled, - * so that it must be bh protected. - * - * We block accesses to device mc filters with netif_tx_lock. - */ - -/* - * Delete a device level multicast - */ - -int dev_mc_delete(struct net_device *dev, void *addr, int alen, int glbl) -{ - int err; - - netif_addr_lock_bh(dev); - err = __dev_addr_delete(&dev->mc_list, &dev->mc_count, - addr, alen, glbl); - if (!err) { - /* - * We have altered the list, so the card - * loaded filter is now wrong. Fix it - */ - - __dev_set_rx_mode(dev); - } - netif_addr_unlock_bh(dev); - return err; -} - -/* - * Add a device level multicast - */ - -int dev_mc_add(struct net_device *dev, void *addr, int alen, int glbl) -{ - int err; - - netif_addr_lock_bh(dev); - if (alen != dev->addr_len) - err = -EINVAL; - else - err = __dev_addr_add(&dev->mc_list, &dev->mc_count, addr, alen, glbl); - if (!err) - __dev_set_rx_mode(dev); - netif_addr_unlock_bh(dev); - return err; -} - -/** - * dev_mc_sync - Synchronize device's multicast list to another device - * @to: destination device - * @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. - * - * This function is intended to be called from the dev->set_multicast_list - * or dev->set_rx_mode function of layered software devices. - */ -int dev_mc_sync(struct net_device *to, struct net_device *from) -{ - int err = 0; - - netif_addr_lock_bh(to); - err = __dev_addr_sync(&to->mc_list, &to->mc_count, - &from->mc_list, &from->mc_count); - if (!err) - __dev_set_rx_mode(to); - netif_addr_unlock_bh(to); - - return err; -} -EXPORT_SYMBOL(dev_mc_sync); - - -/** - * dev_mc_unsync - Remove synchronized addresses from the destination - * device - * @to: destination device - * @from: source device - * - * Remove all addresses that were added to the destination device by - * dev_mc_sync(). This function is intended to be called from the - * dev->stop function of layered software devices. - */ -void dev_mc_unsync(struct net_device *to, struct net_device *from) -{ - netif_addr_lock_bh(from); - netif_addr_lock(to); - - __dev_addr_unsync(&to->mc_list, &to->mc_count, - &from->mc_list, &from->mc_count); - __dev_set_rx_mode(to); - - netif_addr_unlock(to); - netif_addr_unlock_bh(from); -} -EXPORT_SYMBOL(dev_mc_unsync); - -#ifdef CONFIG_PROC_FS -static int dev_mc_seq_show(struct seq_file *seq, void *v) -{ - struct dev_addr_list *m; - struct net_device *dev = v; - - if (v == SEQ_START_TOKEN) - return 0; - - netif_addr_lock_bh(dev); - for (m = dev->mc_list; m; m = m->next) { - int i; - - seq_printf(seq, "%-4d %-15s %-5d %-5d ", dev->ifindex, - dev->name, m->dmi_users, m->dmi_gusers); - - for (i = 0; i < m->dmi_addrlen; i++) - seq_printf(seq, "%02x", m->dmi_addr[i]); - - seq_putc(seq, '\n'); - } - netif_addr_unlock_bh(dev); - return 0; -} - -static const struct seq_operations dev_mc_seq_ops = { - .start = dev_seq_start, - .next = dev_seq_next, - .stop = dev_seq_stop, - .show = dev_mc_seq_show, -}; - -static int dev_mc_seq_open(struct inode *inode, struct file *file) -{ - return seq_open_net(inode, file, &dev_mc_seq_ops, - sizeof(struct seq_net_private)); -} - -static const struct file_operations dev_mc_seq_fops = { - .owner = THIS_MODULE, - .open = dev_mc_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_net, -}; - -#endif - -static int __net_init dev_mc_net_init(struct net *net) -{ - if (!proc_net_fops_create(net, "dev_mcast", 0, &dev_mc_seq_fops)) - return -ENOMEM; - return 0; -} - -static void __net_exit dev_mc_net_exit(struct net *net) -{ - proc_net_remove(net, "dev_mcast"); -} - -static struct pernet_operations __net_initdata dev_mc_net_ops = { - .init = dev_mc_net_init, - .exit = dev_mc_net_exit, -}; - -void __init dev_mcast_init(void) -{ - register_pernet_subsys(&dev_mc_net_ops); -} - -EXPORT_SYMBOL(dev_mc_add); -EXPORT_SYMBOL(dev_mc_delete); diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index 238af093495b..f3e4734d207f 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -349,7 +349,7 @@ static void dn_dev_del_ifa(struct dn_dev *dn_db, struct dn_ifaddr **ifap, int de if (dn_db->dev->type == ARPHRD_ETHER) { if (ifa1->ifa_local != dn_eth2dn(dev->dev_addr)) { dn_dn2eth(mac_addr, ifa1->ifa_local); - dev_mc_delete(dev, mac_addr, ETH_ALEN, 0); + dev_mc_del(dev, mac_addr); } } @@ -380,7 +380,7 @@ static int dn_dev_insert_ifa(struct dn_dev *dn_db, struct dn_ifaddr *ifa) if (dev->type == ARPHRD_ETHER) { if (ifa->ifa_local != dn_eth2dn(dev->dev_addr)) { dn_dn2eth(mac_addr, ifa->ifa_local); - dev_mc_add(dev, mac_addr, ETH_ALEN, 0); + dev_mc_add(dev, mac_addr); } } @@ -1000,9 +1000,9 @@ static int dn_eth_up(struct net_device *dev) struct dn_dev *dn_db = dev->dn_ptr; if (dn_db->parms.forwarding == 0) - dev_mc_add(dev, dn_rt_all_end_mcast, ETH_ALEN, 0); + dev_mc_add(dev, dn_rt_all_end_mcast); else - dev_mc_add(dev, dn_rt_all_rt_mcast, ETH_ALEN, 0); + dev_mc_add(dev, dn_rt_all_rt_mcast); dn_db->use_long = 1; @@ -1014,9 +1014,9 @@ static void dn_eth_down(struct net_device *dev) struct dn_dev *dn_db = dev->dn_ptr; if (dn_db->parms.forwarding == 0) - dev_mc_delete(dev, dn_rt_all_end_mcast, ETH_ALEN, 0); + dev_mc_del(dev, dn_rt_all_end_mcast); else - dev_mc_delete(dev, dn_rt_all_rt_mcast, ETH_ALEN, 0); + dev_mc_del(dev, dn_rt_all_rt_mcast); } static void dn_dev_set_timer(struct net_device *dev); diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 63bf298ca109..51824c42b775 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -997,7 +997,7 @@ static void ip_mc_filter_add(struct in_device *in_dev, __be32 addr) --ANK */ if (arp_mc_map(addr, buf, dev, 0) == 0) - dev_mc_add(dev, buf, dev->addr_len, 0); + dev_mc_add(dev, buf); } /* @@ -1010,7 +1010,7 @@ static void ip_mc_filter_del(struct in_device *in_dev, __be32 addr) struct net_device *dev = in_dev->dev; if (arp_mc_map(addr, buf, dev, 0) == 0) - dev_mc_delete(dev, buf, dev->addr_len, 0); + dev_mc_del(dev, buf); } #ifdef CONFIG_IP_MULTICAST diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 0886f96c736b..a2208b7b313d 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -87,7 +87,7 @@ clusterip_config_entry_put(struct clusterip_config *c) list_del(&c->list); write_unlock_bh(&clusterip_lock); - dev_mc_delete(c->dev, c->clustermac, ETH_ALEN, 0); + dev_mc_del(c->dev, c->clustermac); dev_put(c->dev); /* In case anyone still accesses the file, the open/close @@ -396,7 +396,7 @@ static bool clusterip_tg_check(const struct xt_tgchk_param *par) dev_put(dev); return false; } - dev_mc_add(config->dev,config->clustermac, ETH_ALEN, 0); + dev_mc_add(config->dev, config->clustermac); } } cipinfo->config = config; diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index bcd971915969..37d1868c0064 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -714,7 +714,7 @@ static void igmp6_group_added(struct ifmcaddr6 *mc) if (!(mc->mca_flags&MAF_LOADED)) { mc->mca_flags |= MAF_LOADED; if (ndisc_mc_map(&mc->mca_addr, buf, dev, 0) == 0) - dev_mc_add(dev, buf, dev->addr_len, 0); + dev_mc_add(dev, buf); } spin_unlock_bh(&mc->mca_lock); @@ -740,7 +740,7 @@ static void igmp6_group_dropped(struct ifmcaddr6 *mc) if (mc->mca_flags&MAF_LOADED) { mc->mca_flags &= ~MAF_LOADED; if (ndisc_mc_map(&mc->mca_addr, buf, dev, 0) == 0) - dev_mc_delete(dev, buf, dev->addr_len, 0); + dev_mc_del(dev, buf); } if (mc->mca_flags & MAF_NOREPORT) diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index c3d844093a2f..9179196da264 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -84,16 +84,14 @@ static inline void drv_bss_info_changed(struct ieee80211_local *local, } static inline u64 drv_prepare_multicast(struct ieee80211_local *local, - int mc_count, - struct dev_addr_list *mc_list) + struct netdev_hw_addr_list *mc_list) { u64 ret = 0; if (local->ops->prepare_multicast) - ret = local->ops->prepare_multicast(&local->hw, mc_count, - mc_list); + ret = local->ops->prepare_multicast(&local->hw, mc_list); - trace_drv_prepare_multicast(local, mc_count, ret); + trace_drv_prepare_multicast(local, mc_list->count, ret); return ret; } diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index ab369e2a5282..7fdacf9408b1 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -646,8 +646,7 @@ struct ieee80211_local { struct work_struct recalc_smps; /* aggregated multicast list */ - struct dev_addr_list *mc_list; - int mc_count; + struct netdev_hw_addr_list mc_list; bool tim_in_locked_section; /* see ieee80211_beacon_get() */ diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index b4ec59a8dc03..00f3a93c6b04 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -412,8 +412,7 @@ static int ieee80211_stop(struct net_device *dev) netif_addr_lock_bh(dev); spin_lock_bh(&local->filter_lock); - __dev_addr_unsync(&local->mc_list, &local->mc_count, - &dev->mc_list, &dev->mc_count); + __hw_addr_unsync(&local->mc_list, &dev->mc, dev->addr_len); spin_unlock_bh(&local->filter_lock); netif_addr_unlock_bh(dev); @@ -596,8 +595,7 @@ static void ieee80211_set_multicast_list(struct net_device *dev) sdata->flags ^= IEEE80211_SDATA_PROMISC; } spin_lock_bh(&local->filter_lock); - __dev_addr_sync(&local->mc_list, &local->mc_count, - &dev->mc_list, &dev->mc_count); + __hw_addr_sync(&local->mc_list, &dev->mc, dev->addr_len); spin_unlock_bh(&local->filter_lock); ieee80211_queue_work(&local->hw, &local->reconfig_filter); } diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 06c33b68d8e5..84ad249a4e2e 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -71,7 +71,7 @@ void ieee80211_configure_filter(struct ieee80211_local *local) spin_lock_bh(&local->filter_lock); changed_flags = local->filter_flags ^ new_flags; - mc = drv_prepare_multicast(local, local->mc_count, local->mc_list); + mc = drv_prepare_multicast(local, &local->mc_list); spin_unlock_bh(&local->filter_lock); /* be a bit nasty */ diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 48c1e0ae565f..b0f037cc899c 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1691,9 +1691,9 @@ static int packet_dev_mc(struct net_device *dev, struct packet_mclist *i, if (i->alen != dev->addr_len) return -EINVAL; if (what > 0) - return dev_mc_add(dev, i->addr, i->alen, 0); + return dev_mc_add(dev, i->addr); else - return dev_mc_delete(dev, i->addr, i->alen, 0); + return dev_mc_del(dev, i->addr); break; case PACKET_MR_PROMISC: return dev_set_promiscuity(dev, what); -- cgit v1.2.3 From 5a6d234e73d7d021c74e1aa349b3b37b81372c66 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 5 Apr 2010 14:37:19 -0700 Subject: rps: fixed missed rps_unlock Fix spin_unlock_irq which needs to be rps_unlock. Signed-off-by: Tom Herbert Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 2a9b7dd0bb6e..74f77ca03349 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3120,7 +3120,7 @@ static int process_backlog(struct napi_struct *napi, int quota) skb = __skb_dequeue(&queue->input_pkt_queue); if (!skb) { __napi_complete(napi); - spin_unlock_irq(&queue->input_pkt_queue.lock); + rps_unlock(queue); break; } rps_unlock(queue); -- cgit v1.2.3 From e4008276fddd10445ff06707694a938cb7f35ed4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 5 Apr 2010 15:42:39 -0700 Subject: net: Add a missing local_irq_enable() As noticed by Changli Gao, we must call local_irq_enable() after rps_unlock() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 74f77ca03349..b98ddc62a55d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3121,6 +3121,7 @@ static int process_backlog(struct napi_struct *napi, int quota) if (!skb) { __napi_complete(napi); rps_unlock(queue); + local_irq_enable(); break; } rps_unlock(queue); -- cgit v1.2.3 From 7a161ea92471087a1579239d7a58dd06eaa5601c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 8 Apr 2010 21:26:13 +0000 Subject: net: Dont use netdev_warn() Dont use netdev_warn() in dev_cap_txqueue() and get_rps_cpu() so that we can catch following warnings without crash. bond0.2240 received packet on queue 6, but number of RX queues is 1 bond0.2240 received packet on queue 11, but number of RX queues is 1 Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index a10a21619ae3..0eb79e35671f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1987,9 +1987,9 @@ static inline u16 dev_cap_txqueue(struct net_device *dev, u16 queue_index) { if (unlikely(queue_index >= dev->real_num_tx_queues)) { if (net_ratelimit()) { - netdev_warn(dev, "selects TX queue %d, but " - "real number of TX queues is %d\n", - queue_index, dev->real_num_tx_queues); + pr_warning("%s selects TX queue %d, but " + "real number of TX queues is %d\n", + dev->name, queue_index, dev->real_num_tx_queues); } return 0; } @@ -2223,9 +2223,9 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb) u16 index = skb_get_rx_queue(skb); if (unlikely(index >= dev->num_rx_queues)) { if (net_ratelimit()) { - netdev_warn(dev, "received packet on queue " - "%u, but number of RX queues is %u\n", - index, dev->num_rx_queues); + pr_warning("%s received packet on queue " + "%u, but number of RX queues is %u\n", + dev->name, index, dev->num_rx_queues); } goto done; } -- cgit v1.2.3 From b6c6712a42ca3f9fa7f4a3d7c40e3a9dd1fd9e03 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 8 Apr 2010 23:03:29 +0000 Subject: net: sk_dst_cache RCUification With latest CONFIG_PROVE_RCU stuff, I felt more comfortable to make this work. sk->sk_dst_cache is currently protected by a rwlock (sk_dst_lock) This rwlock is readlocked for a very small amount of time, and dst entries are already freed after RCU grace period. This calls for RCU again :) This patch converts sk_dst_lock to a spinlock, and use RCU for readers. __sk_dst_get() is supposed to be called with rcu_read_lock() or if socket locked by user, so use appropriate rcu_dereference_check() condition (rcu_read_lock_held() || sock_owned_by_user(sk)) This patch avoids two atomic ops per tx packet on UDP connected sockets, for example, and permits sk_dst_lock to be much less dirtied. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/dst.h | 15 --------------- include/net/ip6_route.h | 4 ++-- include/net/sock.h | 47 ++++++++++++++++++++++++++++++----------------- net/core/dev.c | 2 +- net/core/sock.c | 8 ++++---- net/dccp/timer.c | 4 ++-- net/decnet/af_decnet.c | 6 +++--- net/ipv4/af_inet.c | 2 +- net/ipv4/tcp_input.c | 4 ++-- net/ipv4/tcp_timer.c | 4 ++-- net/ipv6/ipv6_sockglue.c | 25 +++++++++++++------------ 11 files changed, 60 insertions(+), 61 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/net/dst.h b/include/net/dst.h index ce078cda6b74..aac5a5fcfda9 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -225,21 +225,6 @@ static inline void dst_confirm(struct dst_entry *dst) neigh_confirm(dst->neighbour); } -static inline void dst_negative_advice(struct dst_entry **dst_p, - struct sock *sk) -{ - struct dst_entry * dst = *dst_p; - if (dst && dst->ops->negative_advice) { - *dst_p = dst->ops->negative_advice(dst); - - if (dst != *dst_p) { - extern void sk_reset_txq(struct sock *sk); - - sk_reset_txq(sk); - } - } -} - static inline void dst_link_failure(struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 68f67836e146..278312c95f96 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -152,9 +152,9 @@ static inline void __ip6_dst_store(struct sock *sk, struct dst_entry *dst, static inline void ip6_dst_store(struct sock *sk, struct dst_entry *dst, struct in6_addr *daddr, struct in6_addr *saddr) { - write_lock(&sk->sk_dst_lock); + spin_lock(&sk->sk_dst_lock); __ip6_dst_store(sk, dst, daddr, saddr); - write_unlock(&sk->sk_dst_lock); + spin_unlock(&sk->sk_dst_lock); } static inline int ipv6_unicast_destination(struct sk_buff *skb) diff --git a/include/net/sock.h b/include/net/sock.h index b4603cd54fcd..56df440a950b 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -262,7 +262,7 @@ struct sock { #ifdef CONFIG_XFRM struct xfrm_policy *sk_policy[2]; #endif - rwlock_t sk_dst_lock; + spinlock_t sk_dst_lock; atomic_t sk_rmem_alloc; atomic_t sk_wmem_alloc; atomic_t sk_omem_alloc; @@ -1192,7 +1192,8 @@ extern unsigned long sock_i_ino(struct sock *sk); static inline struct dst_entry * __sk_dst_get(struct sock *sk) { - return sk->sk_dst_cache; + return rcu_dereference_check(sk->sk_dst_cache, rcu_read_lock_held() || + sock_owned_by_user(sk)); } static inline struct dst_entry * @@ -1200,50 +1201,62 @@ sk_dst_get(struct sock *sk) { struct dst_entry *dst; - read_lock(&sk->sk_dst_lock); - dst = sk->sk_dst_cache; + rcu_read_lock(); + dst = rcu_dereference(sk->sk_dst_cache); if (dst) dst_hold(dst); - read_unlock(&sk->sk_dst_lock); + rcu_read_unlock(); return dst; } +extern void sk_reset_txq(struct sock *sk); + +static inline void dst_negative_advice(struct sock *sk) +{ + struct dst_entry *ndst, *dst = __sk_dst_get(sk); + + if (dst && dst->ops->negative_advice) { + ndst = dst->ops->negative_advice(dst); + + if (ndst != dst) { + rcu_assign_pointer(sk->sk_dst_cache, ndst); + sk_reset_txq(sk); + } + } +} + static inline void __sk_dst_set(struct sock *sk, struct dst_entry *dst) { struct dst_entry *old_dst; sk_tx_queue_clear(sk); - old_dst = sk->sk_dst_cache; - sk->sk_dst_cache = dst; + old_dst = rcu_dereference_check(sk->sk_dst_cache, + lockdep_is_held(&sk->sk_dst_lock)); + rcu_assign_pointer(sk->sk_dst_cache, dst); dst_release(old_dst); } static inline void sk_dst_set(struct sock *sk, struct dst_entry *dst) { - write_lock(&sk->sk_dst_lock); + spin_lock(&sk->sk_dst_lock); __sk_dst_set(sk, dst); - write_unlock(&sk->sk_dst_lock); + spin_unlock(&sk->sk_dst_lock); } static inline void __sk_dst_reset(struct sock *sk) { - struct dst_entry *old_dst; - - sk_tx_queue_clear(sk); - old_dst = sk->sk_dst_cache; - sk->sk_dst_cache = NULL; - dst_release(old_dst); + __sk_dst_set(sk, NULL); } static inline void sk_dst_reset(struct sock *sk) { - write_lock(&sk->sk_dst_lock); + spin_lock(&sk->sk_dst_lock); __sk_dst_reset(sk); - write_unlock(&sk->sk_dst_lock); + spin_unlock(&sk->sk_dst_lock); } extern struct dst_entry *__sk_dst_check(struct sock *sk, u32 cookie); diff --git a/net/core/dev.c b/net/core/dev.c index 0eb79e35671f..ca4cdef74a1b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2015,7 +2015,7 @@ static struct netdev_queue *dev_pick_tx(struct net_device *dev, if (dev->real_num_tx_queues > 1) queue_index = skb_tx_hash(dev, skb); - if (sk && sk->sk_dst_cache) + if (sk && rcu_dereference_check(sk->sk_dst_cache, 1)) sk_tx_queue_set(sk, queue_index); } } diff --git a/net/core/sock.c b/net/core/sock.c index c5812bbc2cc9..7effa1e689df 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -364,11 +364,11 @@ EXPORT_SYMBOL(sk_reset_txq); struct dst_entry *__sk_dst_check(struct sock *sk, u32 cookie) { - struct dst_entry *dst = sk->sk_dst_cache; + struct dst_entry *dst = __sk_dst_get(sk); if (dst && dst->obsolete && dst->ops->check(dst, cookie) == NULL) { sk_tx_queue_clear(sk); - sk->sk_dst_cache = NULL; + rcu_assign_pointer(sk->sk_dst_cache, NULL); dst_release(dst); return NULL; } @@ -1157,7 +1157,7 @@ struct sock *sk_clone(const struct sock *sk, const gfp_t priority) skb_queue_head_init(&newsk->sk_async_wait_queue); #endif - rwlock_init(&newsk->sk_dst_lock); + spin_lock_init(&newsk->sk_dst_lock); rwlock_init(&newsk->sk_callback_lock); lockdep_set_class_and_name(&newsk->sk_callback_lock, af_callback_keys + newsk->sk_family, @@ -1898,7 +1898,7 @@ void sock_init_data(struct socket *sock, struct sock *sk) } else sk->sk_sleep = NULL; - rwlock_init(&sk->sk_dst_lock); + spin_lock_init(&sk->sk_dst_lock); rwlock_init(&sk->sk_callback_lock); lockdep_set_class_and_name(&sk->sk_callback_lock, af_callback_keys + sk->sk_family, diff --git a/net/dccp/timer.c b/net/dccp/timer.c index bbfeb5eae46a..1a9aa05d4dc4 100644 --- a/net/dccp/timer.c +++ b/net/dccp/timer.c @@ -38,7 +38,7 @@ static int dccp_write_timeout(struct sock *sk) if (sk->sk_state == DCCP_REQUESTING || sk->sk_state == DCCP_PARTOPEN) { if (icsk->icsk_retransmits != 0) - dst_negative_advice(&sk->sk_dst_cache, sk); + dst_negative_advice(sk); retry_until = icsk->icsk_syn_retries ? : sysctl_dccp_request_retries; } else { @@ -63,7 +63,7 @@ static int dccp_write_timeout(struct sock *sk) Golden words :-). */ - dst_negative_advice(&sk->sk_dst_cache, sk); + dst_negative_advice(sk); } retry_until = sysctl_dccp_retries2; diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index 2b494fac9468..55e3b6b0061a 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -446,7 +446,7 @@ static void dn_destruct(struct sock *sk) skb_queue_purge(&scp->other_xmit_queue); skb_queue_purge(&scp->other_receive_queue); - dst_release(xchg(&sk->sk_dst_cache, NULL)); + dst_release(rcu_dereference_check(sk->sk_dst_cache, 1)); } static int dn_memory_pressure; @@ -1105,7 +1105,7 @@ static int dn_accept(struct socket *sock, struct socket *newsock, int flags) release_sock(sk); dst = skb_dst(skb); - dst_release(xchg(&newsk->sk_dst_cache, dst)); + sk_dst_set(newsk, dst); skb_dst_set(skb, NULL); DN_SK(newsk)->state = DN_CR; @@ -1956,7 +1956,7 @@ static int dn_sendmsg(struct kiocb *iocb, struct socket *sock, } if ((flags & MSG_TRYHARD) && sk->sk_dst_cache) - dst_negative_advice(&sk->sk_dst_cache, sk); + dst_negative_advice(sk); mss = scp->segsize_rem; fctype = scp->services_rem & NSP_FC_MASK; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index a0beb32beaa3..193dcd6ed64f 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -154,7 +154,7 @@ void inet_sock_destruct(struct sock *sk) WARN_ON(sk->sk_forward_alloc); kfree(inet->opt); - dst_release(sk->sk_dst_cache); + dst_release(rcu_dereference_check(sk->sk_dst_cache, 1)); sk_refcnt_debug_dec(sk); } EXPORT_SYMBOL(inet_sock_destruct); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 4000b10610b7..ae3ec15fb630 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3710,7 +3710,7 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) } if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) - dst_confirm(sk->sk_dst_cache); + dst_confirm(__sk_dst_get(sk)); return 1; @@ -5833,7 +5833,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, if (tp->snd_una == tp->write_seq) { tcp_set_state(sk, TCP_FIN_WAIT2); sk->sk_shutdown |= SEND_SHUTDOWN; - dst_confirm(sk->sk_dst_cache); + dst_confirm(__sk_dst_get(sk)); if (!sock_flag(sk, SOCK_DEAD)) /* Wake up lingering close() */ diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 8a0ab2977f1f..c732be00606b 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -172,14 +172,14 @@ static int tcp_write_timeout(struct sock *sk) if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) { if (icsk->icsk_retransmits) - dst_negative_advice(&sk->sk_dst_cache, sk); + dst_negative_advice(sk); retry_until = icsk->icsk_syn_retries ? : sysctl_tcp_syn_retries; } else { if (retransmits_timed_out(sk, sysctl_tcp_retries1)) { /* Black hole detection */ tcp_mtu_probing(icsk, sk); - dst_negative_advice(&sk->sk_dst_cache, sk); + dst_negative_advice(sk); } retry_until = sysctl_tcp_retries2; diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 33f60fca7aa7..1160400e9dbd 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -114,9 +114,9 @@ struct ipv6_txoptions *ipv6_update_options(struct sock *sk, } opt = xchg(&inet6_sk(sk)->opt, opt); } else { - write_lock(&sk->sk_dst_lock); + spin_lock(&sk->sk_dst_lock); opt = xchg(&inet6_sk(sk)->opt, opt); - write_unlock(&sk->sk_dst_lock); + spin_unlock(&sk->sk_dst_lock); } sk_dst_reset(sk); @@ -971,14 +971,13 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, case IPV6_MTU: { struct dst_entry *dst; + val = 0; - lock_sock(sk); - dst = sk_dst_get(sk); - if (dst) { + rcu_read_lock(); + dst = __sk_dst_get(sk); + if (dst) val = dst_mtu(dst); - dst_release(dst); - } - release_sock(sk); + rcu_read_unlock(); if (!val) return -ENOTCONN; break; @@ -1066,12 +1065,14 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, else val = np->mcast_hops; - dst = sk_dst_get(sk); - if (dst) { - if (val < 0) + if (val < 0) { + rcu_read_lock(); + dst = __sk_dst_get(sk); + if (dst) val = ip6_dst_hoplimit(dst); - dst_release(dst); + rcu_read_unlock(); } + if (val < 0) val = sock_net(sk)->ipv6.devconf_all->hop_limit; break; -- cgit v1.2.3 From acbbc07145b919248c410e1852b953d385be5c97 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 11 Apr 2010 06:56:11 +0000 Subject: net: uninline skb_bond_should_drop() skb_bond_should_drop() is too big to be inlined. This patch reduces kernel text size, and its compilation time as well (shrinking include/linux/netdevice.h) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 48 ++++------------------------------------------ net/core/dev.c | 49 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 53 insertions(+), 44 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d1a21b576a40..470f7c951afb 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2089,54 +2089,14 @@ static inline void netif_set_gso_max_size(struct net_device *dev, dev->gso_max_size = size; } -static inline void skb_bond_set_mac_by_master(struct sk_buff *skb, - struct net_device *master) -{ - if (skb->pkt_type == PACKET_HOST) { - u16 *dest = (u16 *) eth_hdr(skb)->h_dest; - - memcpy(dest, master->dev_addr, ETH_ALEN); - } -} +extern int __skb_bond_should_drop(struct sk_buff *skb, + struct net_device *master); -/* On bonding slaves other than the currently active slave, suppress - * duplicates except for 802.3ad ETH_P_SLOW, alb non-mcast/bcast, and - * ARP on active-backup slaves with arp_validate enabled. - */ static inline int skb_bond_should_drop(struct sk_buff *skb, struct net_device *master) { - if (master) { - struct net_device *dev = skb->dev; - - if (master->priv_flags & IFF_MASTER_ARPMON) - dev->last_rx = jiffies; - - if ((master->priv_flags & IFF_MASTER_ALB) && master->br_port) { - /* Do address unmangle. The local destination address - * will be always the one master has. Provides the right - * functionality in a bridge. - */ - skb_bond_set_mac_by_master(skb, master); - } - - if (dev->priv_flags & IFF_SLAVE_INACTIVE) { - if ((dev->priv_flags & IFF_SLAVE_NEEDARP) && - skb->protocol == __cpu_to_be16(ETH_P_ARP)) - return 0; - - if (master->priv_flags & IFF_MASTER_ALB) { - if (skb->pkt_type != PACKET_BROADCAST && - skb->pkt_type != PACKET_MULTICAST) - return 0; - } - if (master->priv_flags & IFF_MASTER_8023AD && - skb->protocol == __cpu_to_be16(ETH_P_SLOW)) - return 0; - - return 1; - } - } + if (master) + return __skb_bond_should_drop(skb, master); return 0; } diff --git a/net/core/dev.c b/net/core/dev.c index ca4cdef74a1b..876b1112d5ba 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2646,6 +2646,55 @@ void netif_nit_deliver(struct sk_buff *skb) rcu_read_unlock(); } +static inline void skb_bond_set_mac_by_master(struct sk_buff *skb, + struct net_device *master) +{ + if (skb->pkt_type == PACKET_HOST) { + u16 *dest = (u16 *) eth_hdr(skb)->h_dest; + + memcpy(dest, master->dev_addr, ETH_ALEN); + } +} + +/* On bonding slaves other than the currently active slave, suppress + * duplicates except for 802.3ad ETH_P_SLOW, alb non-mcast/bcast, and + * ARP on active-backup slaves with arp_validate enabled. + */ +int __skb_bond_should_drop(struct sk_buff *skb, struct net_device *master) +{ + struct net_device *dev = skb->dev; + + if (master->priv_flags & IFF_MASTER_ARPMON) + dev->last_rx = jiffies; + + if ((master->priv_flags & IFF_MASTER_ALB) && master->br_port) { + /* Do address unmangle. The local destination address + * will be always the one master has. Provides the right + * functionality in a bridge. + */ + skb_bond_set_mac_by_master(skb, master); + } + + if (dev->priv_flags & IFF_SLAVE_INACTIVE) { + if ((dev->priv_flags & IFF_SLAVE_NEEDARP) && + skb->protocol == __cpu_to_be16(ETH_P_ARP)) + return 0; + + if (master->priv_flags & IFF_MASTER_ALB) { + if (skb->pkt_type != PACKET_BROADCAST && + skb->pkt_type != PACKET_MULTICAST) + return 0; + } + if (master->priv_flags & IFF_MASTER_8023AD && + skb->protocol == __cpu_to_be16(ETH_P_SLOW)) + return 0; + + return 1; + } + return 0; +} +EXPORT_SYMBOL(__skb_bond_should_drop); + static int __netif_receive_skb(struct sk_buff *skb) { struct packet_type *ptype, *pt_prev; -- cgit v1.2.3 From b0e28f1effd1d840b36e961edc1def81e01b1ca1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 15 Apr 2010 00:14:07 -0700 Subject: net: netif_rx() must disable preemption Eric Paris reported netif_rx() is calling smp_processor_id() from preemptible context, in particular when caller is ip_dev_loopback_xmit(). RPS commit added this smp_processor_id() call, this patch makes sure preemption is disabled. rps_get_cpus() wants rcu_read_lock() anyway, we can dot it a bit earlier. Reported-by: Eric Paris Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 876b1112d5ba..e8041eb76ac1 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2206,6 +2206,7 @@ DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; /* * get_rps_cpu is called from netif_receive_skb and returns the target * CPU from the RPS map of the receiving queue for a given skb. + * rcu_read_lock must be held on entry. */ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb) { @@ -2217,8 +2218,6 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb) u8 ip_proto; u32 addr1, addr2, ports, ihl; - rcu_read_lock(); - if (skb_rx_queue_recorded(skb)) { u16 index = skb_get_rx_queue(skb); if (unlikely(index >= dev->num_rx_queues)) { @@ -2296,7 +2295,6 @@ got_hash: } done: - rcu_read_unlock(); return cpu; } @@ -2392,7 +2390,7 @@ enqueue: int netif_rx(struct sk_buff *skb) { - int cpu; + int ret; /* if netpoll wants it, pretend we never saw it */ if (netpoll_rx(skb)) @@ -2402,14 +2400,21 @@ int netif_rx(struct sk_buff *skb) net_timestamp(skb); #ifdef CONFIG_RPS - cpu = get_rps_cpu(skb->dev, skb); - if (cpu < 0) - cpu = smp_processor_id(); + { + int cpu; + + rcu_read_lock(); + cpu = get_rps_cpu(skb->dev, skb); + if (cpu < 0) + cpu = smp_processor_id(); + ret = enqueue_to_backlog(skb, cpu); + rcu_read_unlock(); + } #else - cpu = smp_processor_id(); + ret = enqueue_to_backlog(skb, get_cpu()); + put_cpu(); #endif - - return enqueue_to_backlog(skb, cpu); + return ret; } EXPORT_SYMBOL(netif_rx); -- cgit v1.2.3 From fec5e652e58fa6017b2c9e06466cb2a6538de5b4 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Fri, 16 Apr 2010 16:01:27 -0700 Subject: rfs: Receive Flow Steering This patch implements receive flow steering (RFS). RFS steers received packets for layer 3 and 4 processing to the CPU where the application for the corresponding flow is running. RFS is an extension of Receive Packet Steering (RPS). The basic idea of RFS is that when an application calls recvmsg (or sendmsg) the application's running CPU is stored in a hash table that is indexed by the connection's rxhash which is stored in the socket structure. The rxhash is passed in skb's received on the connection from netif_receive_skb. For each received packet, the associated rxhash is used to look up the CPU in the hash table, if a valid CPU is set then the packet is steered to that CPU using the RPS mechanisms. The convolution of the simple approach is that it would potentially allow OOO packets. If threads are thrashing around CPUs or multiple threads are trying to read from the same sockets, a quickly changing CPU value in the hash table could cause rampant OOO packets-- we consider this a non-starter. To avoid OOO packets, this solution implements two types of hash tables: rps_sock_flow_table and rps_dev_flow_table. rps_sock_table is a global hash table. Each entry is just a CPU number and it is populated in recvmsg and sendmsg as described above. This table contains the "desired" CPUs for flows. rps_dev_flow_table is specific to each device queue. Each entry contains a CPU and a tail queue counter. The CPU is the "current" CPU for a matching flow. The tail queue counter holds the value of a tail queue counter for the associated CPU's backlog queue at the time of last enqueue for a flow matching the entry. Each backlog queue has a queue head counter which is incremented on dequeue, and so a queue tail counter is computed as queue head count + queue length. When a packet is enqueued on a backlog queue, the current value of the queue tail counter is saved in the hash entry of the rps_dev_flow_table. And now the trick: when selecting the CPU for RPS (get_rps_cpu) the rps_sock_flow table and the rps_dev_flow table for the RX queue are consulted. When the desired CPU for the flow (found in the rps_sock_flow table) does not match the current CPU (found in the rps_dev_flow table), the current CPU is changed to the desired CPU if one of the following is true: - The current CPU is unset (equal to RPS_NO_CPU) - Current CPU is offline - The current CPU's queue head counter >= queue tail counter in the rps_dev_flow table. This checks if the queue tail has advanced beyond the last packet that was enqueued using this table entry. This guarantees that all packets queued using this entry have been dequeued, thus preserving in order delivery. Making each queue have its own rps_dev_flow table has two advantages: 1) the tail queue counters will be written on each receive, so keeping the table local to interrupting CPU s good for locality. 2) this allows lockless access to the table-- the CPU number and queue tail counter need to be accessed together under mutual exclusion from netif_receive_skb, we assume that this is only called from device napi_poll which is non-reentrant. This patch implements RFS for TCP and connected UDP sockets. It should be usable for other flow oriented protocols. There are two configuration parameters for RFS. The "rps_flow_entries" kernel init parameter sets the number of entries in the rps_sock_flow_table, the per rxqueue sysfs entry "rps_flow_cnt" contains the number of entries in the rps_dev_flow table for the rxqueue. Both are rounded to power of two. The obvious benefit of RFS (over just RPS) is that it achieves CPU locality between the receive processing for a flow and the applications processing; this can result in increased performance (higher pps, lower latency). The benefits of RFS are dependent on cache hierarchy, application load, and other factors. On simple benchmarks, we don't necessarily see improvement and sometimes see degradation. However, for more complex benchmarks and for applications where cache pressure is much higher this technique seems to perform very well. Below are some benchmark results which show the potential benfit of this patch. The netperf test has 500 instances of netperf TCP_RR test with 1 byte req. and resp. The RPC test is an request/response test similar in structure to netperf RR test ith 100 threads on each host, but does more work in userspace that netperf. e1000e on 8 core Intel No RFS or RPS 104K tps at 30% CPU No RFS (best RPS config): 290K tps at 63% CPU RFS 303K tps at 61% CPU RPC test tps CPU% 50/90/99% usec latency Latency StdDev No RFS/RPS 103K 48% 757/900/3185 4472.35 RPS only: 174K 73% 415/993/2468 491.66 RFS 223K 73% 379/651/1382 315.61 Signed-off-by: Tom Herbert Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 69 +++++++++++++++++++++++++++- include/net/inet_sock.h | 38 ++++++++++++++++ net/core/dev.c | 111 +++++++++++++++++++++++++++++++++++++-------- net/core/net-sysfs.c | 94 ++++++++++++++++++++++++++++++++++++-- net/core/sysctl_net_core.c | 68 +++++++++++++++++++++++++++ net/ipv4/af_inet.c | 29 ++++++++++-- net/ipv4/tcp_ipv4.c | 2 + net/ipv4/udp.c | 7 ++- 8 files changed, 389 insertions(+), 29 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 55c2086e1f06..649a0252686e 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -530,14 +530,73 @@ struct rps_map { }; #define RPS_MAP_SIZE(_num) (sizeof(struct rps_map) + (_num * sizeof(u16))) +/* + * The rps_dev_flow structure contains the mapping of a flow to a CPU and the + * tail pointer for that CPU's input queue at the time of last enqueue. + */ +struct rps_dev_flow { + u16 cpu; + u16 fill; + unsigned int last_qtail; +}; + +/* + * The rps_dev_flow_table structure contains a table of flow mappings. + */ +struct rps_dev_flow_table { + unsigned int mask; + struct rcu_head rcu; + struct work_struct free_work; + struct rps_dev_flow flows[0]; +}; +#define RPS_DEV_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_dev_flow_table) + \ + (_num * sizeof(struct rps_dev_flow))) + +/* + * The rps_sock_flow_table contains mappings of flows to the last CPU + * on which they were processed by the application (set in recvmsg). + */ +struct rps_sock_flow_table { + unsigned int mask; + u16 ents[0]; +}; +#define RPS_SOCK_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_sock_flow_table) + \ + (_num * sizeof(u16))) + +#define RPS_NO_CPU 0xffff + +static inline void rps_record_sock_flow(struct rps_sock_flow_table *table, + u32 hash) +{ + if (table && hash) { + unsigned int cpu, index = hash & table->mask; + + /* We only give a hint, preemption can change cpu under us */ + cpu = raw_smp_processor_id(); + + if (table->ents[index] != cpu) + table->ents[index] = cpu; + } +} + +static inline void rps_reset_sock_flow(struct rps_sock_flow_table *table, + u32 hash) +{ + if (table && hash) + table->ents[hash & table->mask] = RPS_NO_CPU; +} + +extern struct rps_sock_flow_table *rps_sock_flow_table; + /* This structure contains an instance of an RX queue. */ struct netdev_rx_queue { struct rps_map *rps_map; + struct rps_dev_flow_table *rps_flow_table; struct kobject kobj; struct netdev_rx_queue *first; atomic_t count; } ____cacheline_aligned_in_smp; -#endif +#endif /* CONFIG_RPS */ /* * This structure defines the management hooks for network devices. @@ -1333,11 +1392,19 @@ struct softnet_data { /* Elements below can be accessed between CPUs for RPS */ #ifdef CONFIG_RPS struct call_single_data csd ____cacheline_aligned_in_smp; + unsigned int input_queue_head; #endif struct sk_buff_head input_pkt_queue; struct napi_struct backlog; }; +static inline void incr_input_queue_head(struct softnet_data *queue) +{ +#ifdef CONFIG_RPS + queue->input_queue_head++; +#endif +} + DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data); #define HAVE_NETIF_QUEUE diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index 83fd34437cf1..b487bc1b99ab 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -21,6 +21,7 @@ #include #include #include +#include #include #include @@ -101,6 +102,7 @@ struct rtable; * @uc_ttl - Unicast TTL * @inet_sport - Source port * @inet_id - ID counter for DF pkts + * @rxhash - flow hash received from netif layer * @tos - TOS * @mc_ttl - Multicasting TTL * @is_icsk - is this an inet_connection_sock? @@ -124,6 +126,9 @@ struct inet_sock { __u16 cmsg_flags; __be16 inet_sport; __u16 inet_id; +#ifdef CONFIG_RPS + __u32 rxhash; +#endif struct ip_options *opt; __u8 tos; @@ -219,4 +224,37 @@ static inline __u8 inet_sk_flowi_flags(const struct sock *sk) return inet_sk(sk)->transparent ? FLOWI_FLAG_ANYSRC : 0; } +static inline void inet_rps_record_flow(const struct sock *sk) +{ +#ifdef CONFIG_RPS + struct rps_sock_flow_table *sock_flow_table; + + rcu_read_lock(); + sock_flow_table = rcu_dereference(rps_sock_flow_table); + rps_record_sock_flow(sock_flow_table, inet_sk(sk)->rxhash); + rcu_read_unlock(); +#endif +} + +static inline void inet_rps_reset_flow(const struct sock *sk) +{ +#ifdef CONFIG_RPS + struct rps_sock_flow_table *sock_flow_table; + + rcu_read_lock(); + sock_flow_table = rcu_dereference(rps_sock_flow_table); + rps_reset_sock_flow(sock_flow_table, inet_sk(sk)->rxhash); + rcu_read_unlock(); +#endif +} + +static inline void inet_rps_save_rxhash(const struct sock *sk, u32 rxhash) +{ +#ifdef CONFIG_RPS + if (unlikely(inet_sk(sk)->rxhash != rxhash)) { + inet_rps_reset_flow(sk); + inet_sk(sk)->rxhash = rxhash; + } +#endif +} #endif /* _INET_SOCK_H */ diff --git a/net/core/dev.c b/net/core/dev.c index e8041eb76ac1..d7107ac835fa 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2203,19 +2203,28 @@ int weight_p __read_mostly = 64; /* old backlog weight */ DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; #ifdef CONFIG_RPS + +/* One global table that all flow-based protocols share. */ +struct rps_sock_flow_table *rps_sock_flow_table; +EXPORT_SYMBOL(rps_sock_flow_table); + /* * get_rps_cpu is called from netif_receive_skb and returns the target * CPU from the RPS map of the receiving queue for a given skb. * rcu_read_lock must be held on entry. */ -static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb) +static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, + struct rps_dev_flow **rflowp) { struct ipv6hdr *ip6; struct iphdr *ip; struct netdev_rx_queue *rxqueue; struct rps_map *map; + struct rps_dev_flow_table *flow_table; + struct rps_sock_flow_table *sock_flow_table; int cpu = -1; u8 ip_proto; + u16 tcpu; u32 addr1, addr2, ports, ihl; if (skb_rx_queue_recorded(skb)) { @@ -2232,7 +2241,7 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb) } else rxqueue = dev->_rx; - if (!rxqueue->rps_map) + if (!rxqueue->rps_map && !rxqueue->rps_flow_table) goto done; if (skb->rxhash) @@ -2284,9 +2293,48 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb) skb->rxhash = 1; got_hash: + flow_table = rcu_dereference(rxqueue->rps_flow_table); + sock_flow_table = rcu_dereference(rps_sock_flow_table); + if (flow_table && sock_flow_table) { + u16 next_cpu; + struct rps_dev_flow *rflow; + + rflow = &flow_table->flows[skb->rxhash & flow_table->mask]; + tcpu = rflow->cpu; + + next_cpu = sock_flow_table->ents[skb->rxhash & + sock_flow_table->mask]; + + /* + * If the desired CPU (where last recvmsg was done) is + * different from current CPU (one in the rx-queue flow + * table entry), switch if one of the following holds: + * - Current CPU is unset (equal to RPS_NO_CPU). + * - Current CPU is offline. + * - The current CPU's queue tail has advanced beyond the + * last packet that was enqueued using this table entry. + * This guarantees that all previous packets for the flow + * have been dequeued, thus preserving in order delivery. + */ + if (unlikely(tcpu != next_cpu) && + (tcpu == RPS_NO_CPU || !cpu_online(tcpu) || + ((int)(per_cpu(softnet_data, tcpu).input_queue_head - + rflow->last_qtail)) >= 0)) { + tcpu = rflow->cpu = next_cpu; + if (tcpu != RPS_NO_CPU) + rflow->last_qtail = per_cpu(softnet_data, + tcpu).input_queue_head; + } + if (tcpu != RPS_NO_CPU && cpu_online(tcpu)) { + *rflowp = rflow; + cpu = tcpu; + goto done; + } + } + map = rcu_dereference(rxqueue->rps_map); if (map) { - u16 tcpu = map->cpus[((u64) skb->rxhash * map->len) >> 32]; + tcpu = map->cpus[((u64) skb->rxhash * map->len) >> 32]; if (cpu_online(tcpu)) { cpu = tcpu; @@ -2320,13 +2368,14 @@ static void trigger_softirq(void *data) __napi_schedule(&queue->backlog); __get_cpu_var(netdev_rx_stat).received_rps++; } -#endif /* CONFIG_SMP */ +#endif /* CONFIG_RPS */ /* * enqueue_to_backlog is called to queue an skb to a per CPU backlog * queue (may be a remote CPU queue). */ -static int enqueue_to_backlog(struct sk_buff *skb, int cpu) +static int enqueue_to_backlog(struct sk_buff *skb, int cpu, + unsigned int *qtail) { struct softnet_data *queue; unsigned long flags; @@ -2341,6 +2390,10 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu) if (queue->input_pkt_queue.qlen) { enqueue: __skb_queue_tail(&queue->input_pkt_queue, skb); +#ifdef CONFIG_RPS + *qtail = queue->input_queue_head + + queue->input_pkt_queue.qlen; +#endif rps_unlock(queue); local_irq_restore(flags); return NET_RX_SUCCESS; @@ -2355,11 +2408,10 @@ enqueue: cpu_set(cpu, rcpus->mask[rcpus->select]); __raise_softirq_irqoff(NET_RX_SOFTIRQ); - } else - __napi_schedule(&queue->backlog); -#else - __napi_schedule(&queue->backlog); + goto enqueue; + } #endif + __napi_schedule(&queue->backlog); } goto enqueue; } @@ -2401,18 +2453,25 @@ int netif_rx(struct sk_buff *skb) #ifdef CONFIG_RPS { + struct rps_dev_flow voidflow, *rflow = &voidflow; int cpu; rcu_read_lock(); - cpu = get_rps_cpu(skb->dev, skb); + + cpu = get_rps_cpu(skb->dev, skb, &rflow); if (cpu < 0) cpu = smp_processor_id(); - ret = enqueue_to_backlog(skb, cpu); + + ret = enqueue_to_backlog(skb, cpu, &rflow->last_qtail); + rcu_read_unlock(); } #else - ret = enqueue_to_backlog(skb, get_cpu()); - put_cpu(); + { + unsigned int qtail; + ret = enqueue_to_backlog(skb, get_cpu(), &qtail); + put_cpu(); + } #endif return ret; } @@ -2830,14 +2889,22 @@ out: int netif_receive_skb(struct sk_buff *skb) { #ifdef CONFIG_RPS - int cpu; + struct rps_dev_flow voidflow, *rflow = &voidflow; + int cpu, ret; + + rcu_read_lock(); - cpu = get_rps_cpu(skb->dev, skb); + cpu = get_rps_cpu(skb->dev, skb, &rflow); - if (cpu < 0) - return __netif_receive_skb(skb); - else - return enqueue_to_backlog(skb, cpu); + if (cpu >= 0) { + ret = enqueue_to_backlog(skb, cpu, &rflow->last_qtail); + rcu_read_unlock(); + } else { + rcu_read_unlock(); + ret = __netif_receive_skb(skb); + } + + return ret; #else return __netif_receive_skb(skb); #endif @@ -2856,6 +2923,7 @@ static void flush_backlog(void *arg) if (skb->dev == dev) { __skb_unlink(skb, &queue->input_pkt_queue); kfree_skb(skb); + incr_input_queue_head(queue); } rps_unlock(queue); } @@ -3179,6 +3247,7 @@ static int process_backlog(struct napi_struct *napi, int quota) local_irq_enable(); break; } + incr_input_queue_head(queue); rps_unlock(queue); local_irq_enable(); @@ -5542,8 +5611,10 @@ static int dev_cpu_callback(struct notifier_block *nfb, local_irq_enable(); /* Process offline CPU's input_pkt_queue */ - while ((skb = __skb_dequeue(&oldsd->input_pkt_queue))) + while ((skb = __skb_dequeue(&oldsd->input_pkt_queue))) { netif_rx(skb); + incr_input_queue_head(oldsd); + } return NOTIFY_OK; } diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 96ed6905b823..143052a22b9b 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include "net-sysfs.h" @@ -601,22 +602,109 @@ ssize_t store_rps_map(struct netdev_rx_queue *queue, return len; } +static ssize_t show_rps_dev_flow_table_cnt(struct netdev_rx_queue *queue, + struct rx_queue_attribute *attr, + char *buf) +{ + struct rps_dev_flow_table *flow_table; + unsigned int val = 0; + + rcu_read_lock(); + flow_table = rcu_dereference(queue->rps_flow_table); + if (flow_table) + val = flow_table->mask + 1; + rcu_read_unlock(); + + return sprintf(buf, "%u\n", val); +} + +static void rps_dev_flow_table_release_work(struct work_struct *work) +{ + struct rps_dev_flow_table *table = container_of(work, + struct rps_dev_flow_table, free_work); + + vfree(table); +} + +static void rps_dev_flow_table_release(struct rcu_head *rcu) +{ + struct rps_dev_flow_table *table = container_of(rcu, + struct rps_dev_flow_table, rcu); + + INIT_WORK(&table->free_work, rps_dev_flow_table_release_work); + schedule_work(&table->free_work); +} + +ssize_t store_rps_dev_flow_table_cnt(struct netdev_rx_queue *queue, + struct rx_queue_attribute *attr, + const char *buf, size_t len) +{ + unsigned int count; + char *endp; + struct rps_dev_flow_table *table, *old_table; + static DEFINE_SPINLOCK(rps_dev_flow_lock); + + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + + count = simple_strtoul(buf, &endp, 0); + if (endp == buf) + return -EINVAL; + + if (count) { + int i; + + if (count > 1<<30) { + /* Enforce a limit to prevent overflow */ + return -EINVAL; + } + count = roundup_pow_of_two(count); + table = vmalloc(RPS_DEV_FLOW_TABLE_SIZE(count)); + if (!table) + return -ENOMEM; + + table->mask = count - 1; + for (i = 0; i < count; i++) + table->flows[i].cpu = RPS_NO_CPU; + } else + table = NULL; + + spin_lock(&rps_dev_flow_lock); + old_table = queue->rps_flow_table; + rcu_assign_pointer(queue->rps_flow_table, table); + spin_unlock(&rps_dev_flow_lock); + + if (old_table) + call_rcu(&old_table->rcu, rps_dev_flow_table_release); + + return len; +} + static struct rx_queue_attribute rps_cpus_attribute = __ATTR(rps_cpus, S_IRUGO | S_IWUSR, show_rps_map, store_rps_map); + +static struct rx_queue_attribute rps_dev_flow_table_cnt_attribute = + __ATTR(rps_flow_cnt, S_IRUGO | S_IWUSR, + show_rps_dev_flow_table_cnt, store_rps_dev_flow_table_cnt); + static struct attribute *rx_queue_default_attrs[] = { &rps_cpus_attribute.attr, + &rps_dev_flow_table_cnt_attribute.attr, NULL }; static void rx_queue_release(struct kobject *kobj) { struct netdev_rx_queue *queue = to_rx_queue(kobj); - struct rps_map *map = queue->rps_map; struct netdev_rx_queue *first = queue->first; - if (map) - call_rcu(&map->rcu, rps_map_release); + if (queue->rps_map) + call_rcu(&queue->rps_map->rcu, rps_map_release); + + if (queue->rps_flow_table) + call_rcu(&queue->rps_flow_table->rcu, + rps_dev_flow_table_release); if (atomic_dec_and_test(&first->count)) kfree(first); diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index b7b6b8208f75..dcc7d25996ab 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -11,12 +11,72 @@ #include #include #include +#include #include #include #include #include +#ifdef CONFIG_RPS +static int rps_sock_flow_sysctl(ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + unsigned int orig_size, size; + int ret, i; + ctl_table tmp = { + .data = &size, + .maxlen = sizeof(size), + .mode = table->mode + }; + struct rps_sock_flow_table *orig_sock_table, *sock_table; + static DEFINE_MUTEX(sock_flow_mutex); + + mutex_lock(&sock_flow_mutex); + + orig_sock_table = rps_sock_flow_table; + size = orig_size = orig_sock_table ? orig_sock_table->mask + 1 : 0; + + ret = proc_dointvec(&tmp, write, buffer, lenp, ppos); + + if (write) { + if (size) { + if (size > 1<<30) { + /* Enforce limit to prevent overflow */ + mutex_unlock(&sock_flow_mutex); + return -EINVAL; + } + size = roundup_pow_of_two(size); + if (size != orig_size) { + sock_table = + vmalloc(RPS_SOCK_FLOW_TABLE_SIZE(size)); + if (!sock_table) { + mutex_unlock(&sock_flow_mutex); + return -ENOMEM; + } + + sock_table->mask = size - 1; + } else + sock_table = orig_sock_table; + + for (i = 0; i < size; i++) + sock_table->ents[i] = RPS_NO_CPU; + } else + sock_table = NULL; + + if (sock_table != orig_sock_table) { + rcu_assign_pointer(rps_sock_flow_table, sock_table); + synchronize_rcu(); + vfree(orig_sock_table); + } + } + + mutex_unlock(&sock_flow_mutex); + + return ret; +} +#endif /* CONFIG_RPS */ + static struct ctl_table net_core_table[] = { #ifdef CONFIG_NET { @@ -82,6 +142,14 @@ static struct ctl_table net_core_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, +#ifdef CONFIG_RPS + { + .procname = "rps_sock_flow_entries", + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = rps_sock_flow_sysctl + }, +#endif #endif /* CONFIG_NET */ { .procname = "netdev_budget", diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 193dcd6ed64f..c5376c725503 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -419,6 +419,8 @@ int inet_release(struct socket *sock) if (sk) { long timeout; + inet_rps_reset_flow(sk); + /* Applications forget to leave groups before exiting */ ip_mc_drop_socket(sk); @@ -720,6 +722,8 @@ int inet_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, { struct sock *sk = sock->sk; + inet_rps_record_flow(sk); + /* We may need to bind the socket. */ if (!inet_sk(sk)->inet_num && inet_autobind(sk)) return -EAGAIN; @@ -728,12 +732,13 @@ int inet_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, } EXPORT_SYMBOL(inet_sendmsg); - static ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset, size_t size, int flags) { struct sock *sk = sock->sk; + inet_rps_record_flow(sk); + /* We may need to bind the socket. */ if (!inet_sk(sk)->inet_num && inet_autobind(sk)) return -EAGAIN; @@ -743,6 +748,22 @@ static ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset, return sock_no_sendpage(sock, page, offset, size, flags); } +int inet_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, + size_t size, int flags) +{ + struct sock *sk = sock->sk; + int addr_len = 0; + int err; + + inet_rps_record_flow(sk); + + err = sk->sk_prot->recvmsg(iocb, sk, msg, size, flags & MSG_DONTWAIT, + flags & ~MSG_DONTWAIT, &addr_len); + if (err >= 0) + msg->msg_namelen = addr_len; + return err; +} +EXPORT_SYMBOL(inet_recvmsg); int inet_shutdown(struct socket *sock, int how) { @@ -872,7 +893,7 @@ const struct proto_ops inet_stream_ops = { .setsockopt = sock_common_setsockopt, .getsockopt = sock_common_getsockopt, .sendmsg = tcp_sendmsg, - .recvmsg = sock_common_recvmsg, + .recvmsg = inet_recvmsg, .mmap = sock_no_mmap, .sendpage = tcp_sendpage, .splice_read = tcp_splice_read, @@ -899,7 +920,7 @@ const struct proto_ops inet_dgram_ops = { .setsockopt = sock_common_setsockopt, .getsockopt = sock_common_getsockopt, .sendmsg = inet_sendmsg, - .recvmsg = sock_common_recvmsg, + .recvmsg = inet_recvmsg, .mmap = sock_no_mmap, .sendpage = inet_sendpage, #ifdef CONFIG_COMPAT @@ -929,7 +950,7 @@ static const struct proto_ops inet_sockraw_ops = { .setsockopt = sock_common_setsockopt, .getsockopt = sock_common_getsockopt, .sendmsg = inet_sendmsg, - .recvmsg = sock_common_recvmsg, + .recvmsg = inet_recvmsg, .mmap = sock_no_mmap, .sendpage = inet_sendpage, #ifdef CONFIG_COMPAT diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index a24995cdc4b6..ad08392a738c 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1672,6 +1672,8 @@ process: skb->dev = NULL; + inet_rps_save_rxhash(sk, skb->rxhash); + bh_lock_sock_nested(sk); ret = 0; if (!sock_owned_by_user(sk)) { diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 8fef859db35d..666b963496ff 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1217,6 +1217,7 @@ int udp_disconnect(struct sock *sk, int flags) sk->sk_state = TCP_CLOSE; inet->inet_daddr = 0; inet->inet_dport = 0; + inet_rps_save_rxhash(sk, 0); sk->sk_bound_dev_if = 0; if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK)) inet_reset_saddr(sk); @@ -1258,8 +1259,12 @@ EXPORT_SYMBOL(udp_lib_unhash); static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) { - int rc = sock_queue_rcv_skb(sk, skb); + int rc; + + if (inet_sk(sk)->inet_daddr) + inet_rps_save_rxhash(sk, skb->rxhash); + rc = sock_queue_rcv_skb(sk, skb); if (rc < 0) { int is_udplite = IS_UDPLITE(sk); -- cgit v1.2.3 From 8770acf0494ae06de6abd34f951a436f8f15d1de Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 17 Apr 2010 00:54:36 -0700 Subject: rps: rps_sock_flow_table is mostly read Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index d7107ac835fa..7abf9590e3c5 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2205,7 +2205,7 @@ DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; #ifdef CONFIG_RPS /* One global table that all flow-based protocols share. */ -struct rps_sock_flow_table *rps_sock_flow_table; +struct rps_sock_flow_table *rps_sock_flow_table __read_mostly; EXPORT_SYMBOL(rps_sock_flow_table); /* -- cgit v1.2.3 From 9958da0501fced47c1ac5c5a3a7731c87e45472c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 17 Apr 2010 04:17:02 +0000 Subject: net: remove time limit in process_backlog() - There is no point to enforce a time limit in process_backlog(), since other napi instances dont follow same rule. We can exit after only one packet processed... The normal quota of 64 packets per napi instance should be the norm, and net_rx_action() already has its own time limit. Note : /proc/net/core/dev_weight can be used to tune this 64 default value. - Use DEFINE_PER_CPU_ALIGNED for softnet_data definition. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 7abf9590e3c5..8092f01713fb 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -264,7 +264,7 @@ static RAW_NOTIFIER_HEAD(netdev_chain); * queue in the local softnet handler. */ -DEFINE_PER_CPU(struct softnet_data, softnet_data); +DEFINE_PER_CPU_ALIGNED(struct softnet_data, softnet_data); EXPORT_PER_CPU_SYMBOL(softnet_data); #ifdef CONFIG_LOCKDEP @@ -3232,7 +3232,6 @@ static int process_backlog(struct napi_struct *napi, int quota) { int work = 0; struct softnet_data *queue = &__get_cpu_var(softnet_data); - unsigned long start_time = jiffies; napi->weight = weight_p; do { @@ -3252,7 +3251,7 @@ static int process_backlog(struct napi_struct *napi, int quota) local_irq_enable(); __netif_receive_skb(skb); - } while (++work < quota && jiffies == start_time); + } while (++work < quota); return work; } -- cgit v1.2.3 From fc6055a5ba31e2c14e36e8939f9bf2b6d586a7f5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 16 Apr 2010 12:18:22 +0000 Subject: net: Introduce skb_orphan_try() Transmitted skb might be attached to a socket and a destructor, for memory accounting purposes. Traditionally, this destructor is called at tx completion time, when skb is freed. When tx completion is performed by another cpu than the sender, this forces some cache lines to change ownership. XPS was an attempt to give tx completion to initial cpu. David idea is to call destructor right before giving skb to device (call to ndo_start_xmit()). Because device queues are usually small, orphaning skb before tx completion is not a big deal. Some drivers already do this, we could do it in upper level. There is one known exception to this early orphaning, called tx timestamping. It needs to keep a reference to socket until device can give a hardware or software timestamp. This patch adds a skb_orphan_try() helper, to centralize all exceptions to early orphaning in one spot, and use it in dev_hard_start_xmit(). "tbench 16" results on a Nehalem machine (2 X5570 @ 2.93GHz) before: Throughput 4428.9 MB/sec 16 procs after: Throughput 4448.14 MB/sec 16 procs UDP should get even better results, its destructor being more complex, since SOCK_USE_WRITE_QUEUE is not set (four atomic ops instead of one) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 27 +++++++++++++-------------- 1 file changed, 13 insertions(+), 14 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 8092f01713fb..8eb50e2292fb 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1880,6 +1880,17 @@ static int dev_gso_segment(struct sk_buff *skb) return 0; } +/* + * Try to orphan skb early, right before transmission by the device. + * We cannot orphan skb if tx timestamp is requested, since + * drivers need to call skb_tstamp_tx() to send the timestamp. + */ +static inline void skb_orphan_try(struct sk_buff *skb) +{ + if (!skb_tx(skb)->flags) + skb_orphan(skb); +} + int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, struct netdev_queue *txq) { @@ -1904,23 +1915,10 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, if (dev->priv_flags & IFF_XMIT_DST_RELEASE) skb_dst_drop(skb); + skb_orphan_try(skb); rc = ops->ndo_start_xmit(skb, dev); if (rc == NETDEV_TX_OK) txq_trans_update(txq); - /* - * TODO: if skb_orphan() was called by - * dev->hard_start_xmit() (for example, the unmodified - * igb driver does that; bnx2 doesn't), then - * skb_tx_software_timestamp() will be unable to send - * back the time stamp. - * - * How can this be prevented? Always create another - * reference to the socket before calling - * dev->hard_start_xmit()? Prevent that skb_orphan() - * does anything in dev->hard_start_xmit() by clearing - * the skb destructor before the call and restoring it - * afterwards, then doing the skb_orphan() ourselves? - */ return rc; } @@ -1938,6 +1936,7 @@ gso: if (dev->priv_flags & IFF_XMIT_DST_RELEASE) skb_dst_drop(nskb); + skb_orphan_try(nskb); rc = ops->ndo_start_xmit(nskb, dev); if (unlikely(rc != NETDEV_TX_OK)) { if (rc & ~NETDEV_TX_MASK) -- cgit v1.2.3 From 88751275b8e867d756e4f86ae92afe0232de129f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 19 Apr 2010 05:07:33 +0000 Subject: rps: shortcut net_rps_action() net_rps_action() is a bit expensive on NR_CPUS=64..4096 kernels, even if RPS is not active. Tom Herbert used two bitmasks to hold information needed to send IPI, but a single LIFO list seems more appropriate. Move all RPS logic into net_rps_action() to cleanup net_rx_action() code (remove two ifdefs) Move rps_remote_softirq_cpus into softnet_data to share its first cache line, filling an existing hole. In a future patch, we could call net_rps_action() from process_backlog() to make sure we send IPI before handling this cpu backlog. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 9 ++++-- net/core/dev.c | 79 +++++++++++++++++++---------------------------- 2 files changed, 38 insertions(+), 50 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 649a0252686e..83ab3da149ad 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1381,17 +1381,20 @@ static inline int unregister_gifconf(unsigned int family) } /* - * Incoming packets are placed on per-cpu queues so that - * no locking is needed. + * Incoming packets are placed on per-cpu queues */ struct softnet_data { struct Qdisc *output_queue; struct list_head poll_list; struct sk_buff *completion_queue; - /* Elements below can be accessed between CPUs for RPS */ #ifdef CONFIG_RPS + struct softnet_data *rps_ipi_list; + + /* Elements below can be accessed between CPUs for RPS */ struct call_single_data csd ____cacheline_aligned_in_smp; + struct softnet_data *rps_ipi_next; + unsigned int cpu; unsigned int input_queue_head; #endif struct sk_buff_head input_pkt_queue; diff --git a/net/core/dev.c b/net/core/dev.c index 8eb50e2292fb..05a2b294906b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2345,21 +2345,6 @@ done: return cpu; } -/* - * This structure holds the per-CPU mask of CPUs for which IPIs are scheduled - * to be sent to kick remote softirq processing. There are two masks since - * the sending of IPIs must be done with interrupts enabled. The select field - * indicates the current mask that enqueue_backlog uses to schedule IPIs. - * select is flipped before net_rps_action is called while still under lock, - * net_rps_action then uses the non-selected mask to send the IPIs and clears - * it without conflicting with enqueue_backlog operation. - */ -struct rps_remote_softirq_cpus { - cpumask_t mask[2]; - int select; -}; -static DEFINE_PER_CPU(struct rps_remote_softirq_cpus, rps_remote_softirq_cpus); - /* Called from hardirq (IPI) context */ static void trigger_softirq(void *data) { @@ -2402,10 +2387,12 @@ enqueue: if (napi_schedule_prep(&queue->backlog)) { #ifdef CONFIG_RPS if (cpu != smp_processor_id()) { - struct rps_remote_softirq_cpus *rcpus = - &__get_cpu_var(rps_remote_softirq_cpus); + struct softnet_data *myqueue; + + myqueue = &__get_cpu_var(softnet_data); + queue->rps_ipi_next = myqueue->rps_ipi_list; + myqueue->rps_ipi_list = queue; - cpu_set(cpu, rcpus->mask[rcpus->select]); __raise_softirq_irqoff(NET_RX_SOFTIRQ); goto enqueue; } @@ -2910,7 +2897,9 @@ int netif_receive_skb(struct sk_buff *skb) } EXPORT_SYMBOL(netif_receive_skb); -/* Network device is going away, flush any packets still pending */ +/* Network device is going away, flush any packets still pending + * Called with irqs disabled. + */ static void flush_backlog(void *arg) { struct net_device *dev = arg; @@ -3338,24 +3327,33 @@ void netif_napi_del(struct napi_struct *napi) } EXPORT_SYMBOL(netif_napi_del); -#ifdef CONFIG_RPS /* - * net_rps_action sends any pending IPI's for rps. This is only called from - * softirq and interrupts must be enabled. + * net_rps_action sends any pending IPI's for rps. + * Note: called with local irq disabled, but exits with local irq enabled. */ -static void net_rps_action(cpumask_t *mask) +static void net_rps_action(void) { - int cpu; +#ifdef CONFIG_RPS + struct softnet_data *locqueue = &__get_cpu_var(softnet_data); + struct softnet_data *remqueue = locqueue->rps_ipi_list; - /* Send pending IPI's to kick RPS processing on remote cpus. */ - for_each_cpu_mask_nr(cpu, *mask) { - struct softnet_data *queue = &per_cpu(softnet_data, cpu); - if (cpu_online(cpu)) - __smp_call_function_single(cpu, &queue->csd, 0); - } - cpus_clear(*mask); -} + if (remqueue) { + locqueue->rps_ipi_list = NULL; + + local_irq_enable(); + + /* Send pending IPI's to kick RPS processing on remote cpus. */ + while (remqueue) { + struct softnet_data *next = remqueue->rps_ipi_next; + if (cpu_online(remqueue->cpu)) + __smp_call_function_single(remqueue->cpu, + &remqueue->csd, 0); + remqueue = next; + } + } else #endif + local_irq_enable(); +} static void net_rx_action(struct softirq_action *h) { @@ -3363,10 +3361,6 @@ static void net_rx_action(struct softirq_action *h) unsigned long time_limit = jiffies + 2; int budget = netdev_budget; void *have; -#ifdef CONFIG_RPS - int select; - struct rps_remote_softirq_cpus *rcpus; -#endif local_irq_disable(); @@ -3429,17 +3423,7 @@ static void net_rx_action(struct softirq_action *h) netpoll_poll_unlock(have); } out: -#ifdef CONFIG_RPS - rcpus = &__get_cpu_var(rps_remote_softirq_cpus); - select = rcpus->select; - rcpus->select ^= 1; - - local_irq_enable(); - - net_rps_action(&rcpus->mask[select]); -#else - local_irq_enable(); -#endif + net_rps_action(); #ifdef CONFIG_NET_DMA /* @@ -5839,6 +5823,7 @@ static int __init net_dev_init(void) queue->csd.func = trigger_softirq; queue->csd.info = queue; queue->csd.flags = 0; + queue->cpu = i; #endif queue->backlog.poll = process_backlog; -- cgit v1.2.3 From e36fa2f7e92f25aab2e3d787dcfe3590817f19d3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 19 Apr 2010 21:17:14 +0000 Subject: rps: cleanups struct softnet_data holds many queues, so consistent use "sd" name instead of "queue" is better. Adds a rps_ipi_queued() helper to cleanup enqueue_to_backlog() Adds a _and_irq_disable suffix to net_rps_action() name, as David suggested. incr_input_queue_head() becomes input_queue_head_incr() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 +- net/core/dev.c | 149 +++++++++++++++++++++++++--------------------- 2 files changed, 82 insertions(+), 71 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 83ab3da149ad..3c5ed5f5274e 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1401,10 +1401,10 @@ struct softnet_data { struct napi_struct backlog; }; -static inline void incr_input_queue_head(struct softnet_data *queue) +static inline void input_queue_head_incr(struct softnet_data *sd) { #ifdef CONFIG_RPS - queue->input_queue_head++; + sd->input_queue_head++; #endif } diff --git a/net/core/dev.c b/net/core/dev.c index 05a2b294906b..7f5755b0a57c 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -208,17 +208,17 @@ static inline struct hlist_head *dev_index_hash(struct net *net, int ifindex) return &net->dev_index_head[ifindex & (NETDEV_HASHENTRIES - 1)]; } -static inline void rps_lock(struct softnet_data *queue) +static inline void rps_lock(struct softnet_data *sd) { #ifdef CONFIG_RPS - spin_lock(&queue->input_pkt_queue.lock); + spin_lock(&sd->input_pkt_queue.lock); #endif } -static inline void rps_unlock(struct softnet_data *queue) +static inline void rps_unlock(struct softnet_data *sd) { #ifdef CONFIG_RPS - spin_unlock(&queue->input_pkt_queue.lock); + spin_unlock(&sd->input_pkt_queue.lock); #endif } @@ -2346,14 +2346,37 @@ done: } /* Called from hardirq (IPI) context */ -static void trigger_softirq(void *data) +static void rps_trigger_softirq(void *data) { - struct softnet_data *queue = data; - __napi_schedule(&queue->backlog); + struct softnet_data *sd = data; + + __napi_schedule(&sd->backlog); __get_cpu_var(netdev_rx_stat).received_rps++; } + #endif /* CONFIG_RPS */ +/* + * Check if this softnet_data structure is another cpu one + * If yes, queue it to our IPI list and return 1 + * If no, return 0 + */ +static int rps_ipi_queued(struct softnet_data *sd) +{ +#ifdef CONFIG_RPS + struct softnet_data *mysd = &__get_cpu_var(softnet_data); + + if (sd != mysd) { + sd->rps_ipi_next = mysd->rps_ipi_list; + mysd->rps_ipi_list = sd; + + __raise_softirq_irqoff(NET_RX_SOFTIRQ); + return 1; + } +#endif /* CONFIG_RPS */ + return 0; +} + /* * enqueue_to_backlog is called to queue an skb to a per CPU backlog * queue (may be a remote CPU queue). @@ -2361,48 +2384,36 @@ static void trigger_softirq(void *data) static int enqueue_to_backlog(struct sk_buff *skb, int cpu, unsigned int *qtail) { - struct softnet_data *queue; + struct softnet_data *sd; unsigned long flags; - queue = &per_cpu(softnet_data, cpu); + sd = &per_cpu(softnet_data, cpu); local_irq_save(flags); __get_cpu_var(netdev_rx_stat).total++; - rps_lock(queue); - if (queue->input_pkt_queue.qlen <= netdev_max_backlog) { - if (queue->input_pkt_queue.qlen) { + rps_lock(sd); + if (sd->input_pkt_queue.qlen <= netdev_max_backlog) { + if (sd->input_pkt_queue.qlen) { enqueue: - __skb_queue_tail(&queue->input_pkt_queue, skb); + __skb_queue_tail(&sd->input_pkt_queue, skb); #ifdef CONFIG_RPS - *qtail = queue->input_queue_head + - queue->input_pkt_queue.qlen; + *qtail = sd->input_queue_head + sd->input_pkt_queue.qlen; #endif - rps_unlock(queue); + rps_unlock(sd); local_irq_restore(flags); return NET_RX_SUCCESS; } /* Schedule NAPI for backlog device */ - if (napi_schedule_prep(&queue->backlog)) { -#ifdef CONFIG_RPS - if (cpu != smp_processor_id()) { - struct softnet_data *myqueue; - - myqueue = &__get_cpu_var(softnet_data); - queue->rps_ipi_next = myqueue->rps_ipi_list; - myqueue->rps_ipi_list = queue; - - __raise_softirq_irqoff(NET_RX_SOFTIRQ); - goto enqueue; - } -#endif - __napi_schedule(&queue->backlog); + if (napi_schedule_prep(&sd->backlog)) { + if (!rps_ipi_queued(sd)) + __napi_schedule(&sd->backlog); } goto enqueue; } - rps_unlock(queue); + rps_unlock(sd); __get_cpu_var(netdev_rx_stat).dropped++; local_irq_restore(flags); @@ -2903,17 +2914,17 @@ EXPORT_SYMBOL(netif_receive_skb); static void flush_backlog(void *arg) { struct net_device *dev = arg; - struct softnet_data *queue = &__get_cpu_var(softnet_data); + struct softnet_data *sd = &__get_cpu_var(softnet_data); struct sk_buff *skb, *tmp; - rps_lock(queue); - skb_queue_walk_safe(&queue->input_pkt_queue, skb, tmp) + rps_lock(sd); + skb_queue_walk_safe(&sd->input_pkt_queue, skb, tmp) if (skb->dev == dev) { - __skb_unlink(skb, &queue->input_pkt_queue); + __skb_unlink(skb, &sd->input_pkt_queue); kfree_skb(skb); - incr_input_queue_head(queue); + input_queue_head_incr(sd); } - rps_unlock(queue); + rps_unlock(sd); } static int napi_gro_complete(struct sk_buff *skb) @@ -3219,23 +3230,23 @@ EXPORT_SYMBOL(napi_gro_frags); static int process_backlog(struct napi_struct *napi, int quota) { int work = 0; - struct softnet_data *queue = &__get_cpu_var(softnet_data); + struct softnet_data *sd = &__get_cpu_var(softnet_data); napi->weight = weight_p; do { struct sk_buff *skb; local_irq_disable(); - rps_lock(queue); - skb = __skb_dequeue(&queue->input_pkt_queue); + rps_lock(sd); + skb = __skb_dequeue(&sd->input_pkt_queue); if (!skb) { __napi_complete(napi); - rps_unlock(queue); + rps_unlock(sd); local_irq_enable(); break; } - incr_input_queue_head(queue); - rps_unlock(queue); + input_queue_head_incr(sd); + rps_unlock(sd); local_irq_enable(); __netif_receive_skb(skb); @@ -3331,24 +3342,25 @@ EXPORT_SYMBOL(netif_napi_del); * net_rps_action sends any pending IPI's for rps. * Note: called with local irq disabled, but exits with local irq enabled. */ -static void net_rps_action(void) +static void net_rps_action_and_irq_disable(void) { #ifdef CONFIG_RPS - struct softnet_data *locqueue = &__get_cpu_var(softnet_data); - struct softnet_data *remqueue = locqueue->rps_ipi_list; + struct softnet_data *sd = &__get_cpu_var(softnet_data); + struct softnet_data *remsd = sd->rps_ipi_list; - if (remqueue) { - locqueue->rps_ipi_list = NULL; + if (remsd) { + sd->rps_ipi_list = NULL; local_irq_enable(); /* Send pending IPI's to kick RPS processing on remote cpus. */ - while (remqueue) { - struct softnet_data *next = remqueue->rps_ipi_next; - if (cpu_online(remqueue->cpu)) - __smp_call_function_single(remqueue->cpu, - &remqueue->csd, 0); - remqueue = next; + while (remsd) { + struct softnet_data *next = remsd->rps_ipi_next; + + if (cpu_online(remsd->cpu)) + __smp_call_function_single(remsd->cpu, + &remsd->csd, 0); + remsd = next; } } else #endif @@ -3423,7 +3435,7 @@ static void net_rx_action(struct softirq_action *h) netpoll_poll_unlock(have); } out: - net_rps_action(); + net_rps_action_and_irq_disable(); #ifdef CONFIG_NET_DMA /* @@ -5595,7 +5607,7 @@ static int dev_cpu_callback(struct notifier_block *nfb, /* Process offline CPU's input_pkt_queue */ while ((skb = __skb_dequeue(&oldsd->input_pkt_queue))) { netif_rx(skb); - incr_input_queue_head(oldsd); + input_queue_head_incr(oldsd); } return NOTIFY_OK; @@ -5812,24 +5824,23 @@ static int __init net_dev_init(void) */ for_each_possible_cpu(i) { - struct softnet_data *queue; + struct softnet_data *sd = &per_cpu(softnet_data, i); - queue = &per_cpu(softnet_data, i); - skb_queue_head_init(&queue->input_pkt_queue); - queue->completion_queue = NULL; - INIT_LIST_HEAD(&queue->poll_list); + skb_queue_head_init(&sd->input_pkt_queue); + sd->completion_queue = NULL; + INIT_LIST_HEAD(&sd->poll_list); #ifdef CONFIG_RPS - queue->csd.func = trigger_softirq; - queue->csd.info = queue; - queue->csd.flags = 0; - queue->cpu = i; + sd->csd.func = rps_trigger_softirq; + sd->csd.info = sd; + sd->csd.flags = 0; + sd->cpu = i; #endif - queue->backlog.poll = process_backlog; - queue->backlog.weight = weight_p; - queue->backlog.gro_list = NULL; - queue->backlog.gro_count = 0; + sd->backlog.poll = process_backlog; + sd->backlog.weight = weight_p; + sd->backlog.gro_list = NULL; + sd->backlog.gro_count = 0; } dev_boot_phase = 0; -- cgit v1.2.3 From b249dcb82d327e419d3cb45773b146ebb5faf419 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 19 Apr 2010 21:56:38 +0000 Subject: rps: consistent rxhash In case we compute a software skb->rxhash, we can generate a consistent hash : Its value will be the same in both flow directions. This helps some workloads, like conntracking, since the same state needs to be accessed in both directions. tbench + RFS + this patch gives better results than tbench with default kernel configuration (no RPS, no RFS) Also fixed some sparse warnings. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 25 ++++++++++++++++++------- 1 file changed, 18 insertions(+), 7 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 7f5755b0a57c..0d78e0454a6d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1974,7 +1974,7 @@ u16 skb_tx_hash(const struct net_device *dev, const struct sk_buff *skb) if (skb->sk && skb->sk->sk_hash) hash = skb->sk->sk_hash; else - hash = skb->protocol; + hash = (__force u16) skb->protocol; hash = jhash_1word(hash, hashrnd); @@ -2253,8 +2253,8 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, ip = (struct iphdr *) skb->data; ip_proto = ip->protocol; - addr1 = ip->saddr; - addr2 = ip->daddr; + addr1 = (__force u32) ip->saddr; + addr2 = (__force u32) ip->daddr; ihl = ip->ihl; break; case __constant_htons(ETH_P_IPV6): @@ -2263,8 +2263,8 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, ip6 = (struct ipv6hdr *) skb->data; ip_proto = ip6->nexthdr; - addr1 = ip6->saddr.s6_addr32[3]; - addr2 = ip6->daddr.s6_addr32[3]; + addr1 = (__force u32) ip6->saddr.s6_addr32[3]; + addr2 = (__force u32) ip6->daddr.s6_addr32[3]; ihl = (40 >> 2); break; default: @@ -2279,14 +2279,25 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, case IPPROTO_AH: case IPPROTO_SCTP: case IPPROTO_UDPLITE: - if (pskb_may_pull(skb, (ihl * 4) + 4)) - ports = *((u32 *) (skb->data + (ihl * 4))); + if (pskb_may_pull(skb, (ihl * 4) + 4)) { + __be16 *hports = (__be16 *) (skb->data + (ihl * 4)); + u32 sport, dport; + + sport = (__force u16) hports[0]; + dport = (__force u16) hports[1]; + if (dport < sport) + swap(sport, dport); + ports = (sport << 16) + dport; + } break; default: break; } + /* get a consistent hash (same value on both flow directions) */ + if (addr2 < addr1) + swap(addr1, addr2); skb->rxhash = jhash_3words(addr1, addr2, ports, hashrnd); if (!skb->rxhash) skb->rxhash = 1; -- cgit v1.2.3 From ab9304717f7624c41927f442e6b6d418b2d8b3e4 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 20 Apr 2010 01:45:37 -0700 Subject: net: emphasize rtnl lock required in call_netdevice_notifiers Since netdev_chain is guarded by rtnl_lock, ASSERT_RTNL should be present here to make sure that all callers of call_netdevice_notifiers does the locking properly. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- net/core/dev.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 0d78e0454a6d..b31d5d69a467 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1435,6 +1435,7 @@ EXPORT_SYMBOL(unregister_netdevice_notifier); int call_netdevice_notifiers(unsigned long val, struct net_device *dev) { + ASSERT_RTNL(); return raw_notifier_call_chain(&netdev_chain, val, dev); } -- cgit v1.2.3 From 9a20e3197e7f6097897c6d1f18335a326ee06299 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 20 Apr 2010 20:08:36 +0000 Subject: net: Introduce skb_orphan_try() At this point, skb->destructor is not the original one (stored in DEV_GSO_CB(skb)->destructor) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index e904c476b112..9bf1cccb067e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1937,7 +1937,6 @@ gso: if (dev->priv_flags & IFF_XMIT_DST_RELEASE) skb_dst_drop(nskb); - skb_orphan_try(nskb); rc = ops->ndo_start_xmit(nskb, dev); if (unlikely(rc != NETDEV_TX_OK)) { if (rc & ~NETDEV_TX_MASK) -- cgit v1.2.3 From e326bed2f47d0365da5a8faaf8ee93ed2d86325b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 22 Apr 2010 00:22:45 -0700 Subject: rps: immediate send IPI in process_backlog() If some skb are queued to our backlog, we are delaying IPI sending at the end of net_rx_action(), increasing latencies. This defeats the queueing, since we want to quickly dispatch packets to the pool of worker cpus, then eventually deeply process our packets. It's better to send IPI before processing our packets in upper layers, from process_backlog(). Change the _and_disable_irq suffix to _and_enable_irq(), since we enable local irq in net_rps_action(), sorry for the confusion. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 76 ++++++++++++++++++++++++++++++++-------------------------- 1 file changed, 42 insertions(+), 34 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 9bf1cccb067e..3ba774b6091c 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3242,11 +3242,48 @@ gro_result_t napi_gro_frags(struct napi_struct *napi) } EXPORT_SYMBOL(napi_gro_frags); +/* + * net_rps_action sends any pending IPI's for rps. + * Note: called with local irq disabled, but exits with local irq enabled. + */ +static void net_rps_action_and_irq_enable(struct softnet_data *sd) +{ +#ifdef CONFIG_RPS + struct softnet_data *remsd = sd->rps_ipi_list; + + if (remsd) { + sd->rps_ipi_list = NULL; + + local_irq_enable(); + + /* Send pending IPI's to kick RPS processing on remote cpus. */ + while (remsd) { + struct softnet_data *next = remsd->rps_ipi_next; + + if (cpu_online(remsd->cpu)) + __smp_call_function_single(remsd->cpu, + &remsd->csd, 0); + remsd = next; + } + } else +#endif + local_irq_enable(); +} + static int process_backlog(struct napi_struct *napi, int quota) { int work = 0; struct softnet_data *sd = &__get_cpu_var(softnet_data); +#ifdef CONFIG_RPS + /* Check if we have pending ipi, its better to send them now, + * not waiting net_rx_action() end. + */ + if (sd->rps_ipi_list) { + local_irq_disable(); + net_rps_action_and_irq_enable(sd); + } +#endif napi->weight = weight_p; do { struct sk_buff *skb; @@ -3353,45 +3390,16 @@ void netif_napi_del(struct napi_struct *napi) } EXPORT_SYMBOL(netif_napi_del); -/* - * net_rps_action sends any pending IPI's for rps. - * Note: called with local irq disabled, but exits with local irq enabled. - */ -static void net_rps_action_and_irq_disable(void) -{ -#ifdef CONFIG_RPS - struct softnet_data *sd = &__get_cpu_var(softnet_data); - struct softnet_data *remsd = sd->rps_ipi_list; - - if (remsd) { - sd->rps_ipi_list = NULL; - - local_irq_enable(); - - /* Send pending IPI's to kick RPS processing on remote cpus. */ - while (remsd) { - struct softnet_data *next = remsd->rps_ipi_next; - - if (cpu_online(remsd->cpu)) - __smp_call_function_single(remsd->cpu, - &remsd->csd, 0); - remsd = next; - } - } else -#endif - local_irq_enable(); -} - static void net_rx_action(struct softirq_action *h) { - struct list_head *list = &__get_cpu_var(softnet_data).poll_list; + struct softnet_data *sd = &__get_cpu_var(softnet_data); unsigned long time_limit = jiffies + 2; int budget = netdev_budget; void *have; local_irq_disable(); - while (!list_empty(list)) { + while (!list_empty(&sd->poll_list)) { struct napi_struct *n; int work, weight; @@ -3409,7 +3417,7 @@ static void net_rx_action(struct softirq_action *h) * entries to the tail of this list, and only ->poll() * calls can remove this head entry from the list. */ - n = list_first_entry(list, struct napi_struct, poll_list); + n = list_first_entry(&sd->poll_list, struct napi_struct, poll_list); have = netpoll_poll_lock(n); @@ -3444,13 +3452,13 @@ static void net_rx_action(struct softirq_action *h) napi_complete(n); local_irq_disable(); } else - list_move_tail(&n->poll_list, list); + list_move_tail(&n->poll_list, &sd->poll_list); } netpoll_poll_unlock(have); } out: - net_rps_action_and_irq_disable(); + net_rps_action_and_irq_enable(sd); #ifdef CONFIG_NET_DMA /* -- cgit v1.2.3 From 9ccb8975940c4ee51161152e37058e3d9e06c62f Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 22 Apr 2010 01:02:07 -0700 Subject: net: Orphan and de-dst skbs earlier in xmit path. This way GSO packets don't get handled differently. With help from Eric Dumazet. Signed-off-by: David S. Miller Signed-off-by: Eric Dumazet --- net/core/dev.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 3ba774b6091c..a4a7c36917d1 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1902,13 +1902,6 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, if (!list_empty(&ptype_all)) dev_queue_xmit_nit(skb, dev); - if (netif_needs_gso(dev, skb)) { - if (unlikely(dev_gso_segment(skb))) - goto out_kfree_skb; - if (skb->next) - goto gso; - } - /* * If device doesnt need skb->dst, release it right now while * its hot in this cpu cache @@ -1917,6 +1910,14 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, skb_dst_drop(skb); skb_orphan_try(skb); + + if (netif_needs_gso(dev, skb)) { + if (unlikely(dev_gso_segment(skb))) + goto out_kfree_skb; + if (skb->next) + goto gso; + } + rc = ops->ndo_start_xmit(skb, dev); if (rc == NETDEV_TX_OK) txq_trans_update(txq); -- cgit v1.2.3 From 8c52d509e84bbf26cffb8b6e75b399689af67885 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Sat, 24 Apr 2010 22:50:10 -0700 Subject: rps: optimize rps_get_cpu() optimize rps_get_cpu(). don't initialize ports when we can get the ports. one memory access for ports than two. Signed-off-by: Changli Gao Signed-off-by: David S. Miller --- net/core/dev.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index a4a7c36917d1..4d43f1a80f74 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2229,7 +2229,11 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, int cpu = -1; u8 ip_proto; u16 tcpu; - u32 addr1, addr2, ports, ihl; + u32 addr1, addr2, ihl; + union { + u32 v32; + u16 v16[2]; + } ports; if (skb_rx_queue_recorded(skb)) { u16 index = skb_get_rx_queue(skb); @@ -2275,7 +2279,6 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, default: goto done; } - ports = 0; switch (ip_proto) { case IPPROTO_TCP: case IPPROTO_UDP: @@ -2285,25 +2288,20 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, case IPPROTO_SCTP: case IPPROTO_UDPLITE: if (pskb_may_pull(skb, (ihl * 4) + 4)) { - __be16 *hports = (__be16 *) (skb->data + (ihl * 4)); - u32 sport, dport; - - sport = (__force u16) hports[0]; - dport = (__force u16) hports[1]; - if (dport < sport) - swap(sport, dport); - ports = (sport << 16) + dport; + ports.v32 = * (__force u32 *) (skb->data + (ihl * 4)); + if (ports.v16[1] < ports.v16[0]) + swap(ports.v16[0], ports.v16[1]); + break; } - break; - default: + ports.v32 = 0; break; } /* get a consistent hash (same value on both flow directions) */ if (addr2 < addr1) swap(addr1, addr2); - skb->rxhash = jhash_3words(addr1, addr2, ports, hashrnd); + skb->rxhash = jhash_3words(addr1, addr2, ports.v32, hashrnd); if (!skb->rxhash) skb->rxhash = 1; -- cgit v1.2.3 From a9cbd588fdb71ea415754c885e2f9f03e6bf1ba0 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Mon, 26 Apr 2010 23:06:24 +0000 Subject: net: reimplement softnet_data.output_queue as a FIFO queue reimplement softnet_data.output_queue as a FIFO queue to keep the fairness among the qdiscs rescheduled. Signed-off-by: Changli Gao Acked-by: Eric Dumazet ---- include/linux/netdevice.h | 1 + net/core/dev.c | 22 ++++++++++++---------- 2 files changed, 13 insertions(+), 10 deletions(-) Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 + net/core/dev.c | 22 ++++++++++++---------- 2 files changed, 13 insertions(+), 10 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 3c5ed5f5274e..c04ca246395d 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1385,6 +1385,7 @@ static inline int unregister_gifconf(unsigned int family) */ struct softnet_data { struct Qdisc *output_queue; + struct Qdisc **output_queue_tailp; struct list_head poll_list; struct sk_buff *completion_queue; diff --git a/net/core/dev.c b/net/core/dev.c index 4d43f1a80f74..3d314919a2cf 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1557,8 +1557,9 @@ static inline void __netif_reschedule(struct Qdisc *q) local_irq_save(flags); sd = &__get_cpu_var(softnet_data); - q->next_sched = sd->output_queue; - sd->output_queue = q; + q->next_sched = NULL; + *sd->output_queue_tailp = q; + sd->output_queue_tailp = &q->next_sched; raise_softirq_irqoff(NET_TX_SOFTIRQ); local_irq_restore(flags); } @@ -2529,6 +2530,7 @@ static void net_tx_action(struct softirq_action *h) local_irq_disable(); head = sd->output_queue; sd->output_queue = NULL; + sd->output_queue_tailp = &sd->output_queue; local_irq_enable(); while (head) { @@ -5594,7 +5596,6 @@ static int dev_cpu_callback(struct notifier_block *nfb, void *ocpu) { struct sk_buff **list_skb; - struct Qdisc **list_net; struct sk_buff *skb; unsigned int cpu, oldcpu = (unsigned long)ocpu; struct softnet_data *sd, *oldsd; @@ -5615,13 +5616,13 @@ static int dev_cpu_callback(struct notifier_block *nfb, *list_skb = oldsd->completion_queue; oldsd->completion_queue = NULL; - /* Find end of our output_queue. */ - list_net = &sd->output_queue; - while (*list_net) - list_net = &(*list_net)->next_sched; /* Append output queue from offline CPU. */ - *list_net = oldsd->output_queue; - oldsd->output_queue = NULL; + if (oldsd->output_queue) { + *sd->output_queue_tailp = oldsd->output_queue; + sd->output_queue_tailp = oldsd->output_queue_tailp; + oldsd->output_queue = NULL; + oldsd->output_queue_tailp = &oldsd->output_queue; + } raise_softirq_irqoff(NET_TX_SOFTIRQ); local_irq_enable(); @@ -5851,7 +5852,8 @@ static int __init net_dev_init(void) skb_queue_head_init(&sd->input_pkt_queue); sd->completion_queue = NULL; INIT_LIST_HEAD(&sd->poll_list); - + sd->output_queue = NULL; + sd->output_queue_tailp = &sd->output_queue; #ifdef CONFIG_RPS sd->csd.func = rps_trigger_softirq; sd->csd.info = sd; -- cgit v1.2.3 From 6e7676c1a76aed6e957611d8d7a9e5592e23aeba Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 27 Apr 2010 15:07:33 -0700 Subject: net: batch skb dequeueing from softnet input_pkt_queue batch skb dequeueing from softnet input_pkt_queue to reduce potential lock contention when RPS is enabled. Note: in the worst case, the number of packets in a softnet_data may be double of netdev_max_backlog. Signed-off-by: Changli Gao Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 6 +++-- net/core/dev.c | 57 ++++++++++++++++++++++++++++++++--------------- 2 files changed, 43 insertions(+), 20 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c04ca246395d..40d4c20d034b 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1388,6 +1388,7 @@ struct softnet_data { struct Qdisc **output_queue_tailp; struct list_head poll_list; struct sk_buff *completion_queue; + struct sk_buff_head process_queue; #ifdef CONFIG_RPS struct softnet_data *rps_ipi_list; @@ -1402,10 +1403,11 @@ struct softnet_data { struct napi_struct backlog; }; -static inline void input_queue_head_incr(struct softnet_data *sd) +static inline void input_queue_head_add(struct softnet_data *sd, + unsigned int len) { #ifdef CONFIG_RPS - sd->input_queue_head++; + sd->input_queue_head += len; #endif } diff --git a/net/core/dev.c b/net/core/dev.c index 3d314919a2cf..100dcbd29739 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2408,12 +2408,13 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, __get_cpu_var(netdev_rx_stat).total++; rps_lock(sd); - if (sd->input_pkt_queue.qlen <= netdev_max_backlog) { - if (sd->input_pkt_queue.qlen) { + if (skb_queue_len(&sd->input_pkt_queue) <= netdev_max_backlog) { + if (skb_queue_len(&sd->input_pkt_queue)) { enqueue: __skb_queue_tail(&sd->input_pkt_queue, skb); #ifdef CONFIG_RPS - *qtail = sd->input_queue_head + sd->input_pkt_queue.qlen; + *qtail = sd->input_queue_head + + skb_queue_len(&sd->input_pkt_queue); #endif rps_unlock(sd); local_irq_restore(flags); @@ -2934,13 +2935,21 @@ static void flush_backlog(void *arg) struct sk_buff *skb, *tmp; rps_lock(sd); - skb_queue_walk_safe(&sd->input_pkt_queue, skb, tmp) + skb_queue_walk_safe(&sd->input_pkt_queue, skb, tmp) { if (skb->dev == dev) { __skb_unlink(skb, &sd->input_pkt_queue); kfree_skb(skb); - input_queue_head_incr(sd); + input_queue_head_add(sd, 1); } + } rps_unlock(sd); + + skb_queue_walk_safe(&sd->process_queue, skb, tmp) { + if (skb->dev == dev) { + __skb_unlink(skb, &sd->process_queue); + kfree_skb(skb); + } + } } static int napi_gro_complete(struct sk_buff *skb) @@ -3286,24 +3295,33 @@ static int process_backlog(struct napi_struct *napi, int quota) } #endif napi->weight = weight_p; - do { + local_irq_disable(); + while (work < quota) { struct sk_buff *skb; + unsigned int qlen; + + while ((skb = __skb_dequeue(&sd->process_queue))) { + local_irq_enable(); + __netif_receive_skb(skb); + if (++work >= quota) + return work; + local_irq_disable(); + } - local_irq_disable(); rps_lock(sd); - skb = __skb_dequeue(&sd->input_pkt_queue); - if (!skb) { + qlen = skb_queue_len(&sd->input_pkt_queue); + if (qlen) { + input_queue_head_add(sd, qlen); + skb_queue_splice_tail_init(&sd->input_pkt_queue, + &sd->process_queue); + } + if (qlen < quota - work) { __napi_complete(napi); - rps_unlock(sd); - local_irq_enable(); - break; + quota = work + qlen; } - input_queue_head_incr(sd); rps_unlock(sd); - local_irq_enable(); - - __netif_receive_skb(skb); - } while (++work < quota); + } + local_irq_enable(); return work; } @@ -5630,8 +5648,10 @@ static int dev_cpu_callback(struct notifier_block *nfb, /* Process offline CPU's input_pkt_queue */ while ((skb = __skb_dequeue(&oldsd->input_pkt_queue))) { netif_rx(skb); - input_queue_head_incr(oldsd); + input_queue_head_add(oldsd, 1); } + while ((skb = __skb_dequeue(&oldsd->process_queue))) + netif_rx(skb); return NOTIFY_OK; } @@ -5850,6 +5870,7 @@ static int __init net_dev_init(void) struct softnet_data *sd = &per_cpu(softnet_data, i); skb_queue_head_init(&sd->input_pkt_queue); + skb_queue_head_init(&sd->process_queue); sd->completion_queue = NULL; INIT_LIST_HEAD(&sd->poll_list); sd->output_queue = NULL; -- cgit v1.2.3 From dee42870a423ad485129f43cddfe7275479f11d8 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Sun, 2 May 2010 05:42:16 +0000 Subject: net: fix softnet_stat Per cpu variable softnet_data.total was shared between IRQ and SoftIRQ context without any protection. And enqueue_to_backlog should update the netdev_rx_stat of the target CPU. This patch renames softnet_data.total to softnet_data.processed: the number of packets processed in uppper levels(IP stacks). softnet_stat data is moved into softnet_data. Signed-off-by: Changli Gao ---- include/linux/netdevice.h | 17 +++++++---------- net/core/dev.c | 26 ++++++++++++-------------- net/sched/sch_generic.c | 2 +- 3 files changed, 20 insertions(+), 25 deletions(-) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 17 +++++++---------- net/core/dev.c | 26 ++++++++++++-------------- net/sched/sch_generic.c | 2 +- 3 files changed, 20 insertions(+), 25 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 40d4c20d034b..c39938f8a8d8 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -218,16 +218,6 @@ struct neighbour; struct neigh_parms; struct sk_buff; -struct netif_rx_stats { - unsigned total; - unsigned dropped; - unsigned time_squeeze; - unsigned cpu_collision; - unsigned received_rps; -}; - -DECLARE_PER_CPU(struct netif_rx_stats, netdev_rx_stat); - struct netdev_hw_addr { struct list_head list; unsigned char addr[MAX_ADDR_LEN]; @@ -1390,6 +1380,12 @@ struct softnet_data { struct sk_buff *completion_queue; struct sk_buff_head process_queue; + /* stats */ + unsigned processed; + unsigned time_squeeze; + unsigned cpu_collision; + unsigned received_rps; + #ifdef CONFIG_RPS struct softnet_data *rps_ipi_list; @@ -1399,6 +1395,7 @@ struct softnet_data { unsigned int cpu; unsigned int input_queue_head; #endif + unsigned dropped; struct sk_buff_head input_pkt_queue; struct napi_struct backlog; }; diff --git a/net/core/dev.c b/net/core/dev.c index 100dcbd29739..36d53be4fca6 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2205,8 +2205,6 @@ int netdev_max_backlog __read_mostly = 1000; int netdev_budget __read_mostly = 300; int weight_p __read_mostly = 64; /* old backlog weight */ -DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, }; - #ifdef CONFIG_RPS /* One global table that all flow-based protocols share. */ @@ -2366,7 +2364,7 @@ static void rps_trigger_softirq(void *data) struct softnet_data *sd = data; __napi_schedule(&sd->backlog); - __get_cpu_var(netdev_rx_stat).received_rps++; + sd->received_rps++; } #endif /* CONFIG_RPS */ @@ -2405,7 +2403,6 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, sd = &per_cpu(softnet_data, cpu); local_irq_save(flags); - __get_cpu_var(netdev_rx_stat).total++; rps_lock(sd); if (skb_queue_len(&sd->input_pkt_queue) <= netdev_max_backlog) { @@ -2429,9 +2426,9 @@ enqueue: goto enqueue; } + sd->dropped++; rps_unlock(sd); - __get_cpu_var(netdev_rx_stat).dropped++; local_irq_restore(flags); kfree_skb(skb); @@ -2806,7 +2803,7 @@ static int __netif_receive_skb(struct sk_buff *skb) skb->dev = master; } - __get_cpu_var(netdev_rx_stat).total++; + __get_cpu_var(softnet_data).processed++; skb_reset_network_header(skb); skb_reset_transport_header(skb); @@ -3490,7 +3487,7 @@ out: return; softnet_break: - __get_cpu_var(netdev_rx_stat).time_squeeze++; + sd->time_squeeze++; __raise_softirq_irqoff(NET_RX_SOFTIRQ); goto out; } @@ -3691,17 +3688,17 @@ static int dev_seq_show(struct seq_file *seq, void *v) return 0; } -static struct netif_rx_stats *softnet_get_online(loff_t *pos) +static struct softnet_data *softnet_get_online(loff_t *pos) { - struct netif_rx_stats *rc = NULL; + struct softnet_data *sd = NULL; while (*pos < nr_cpu_ids) if (cpu_online(*pos)) { - rc = &per_cpu(netdev_rx_stat, *pos); + sd = &per_cpu(softnet_data, *pos); break; } else ++*pos; - return rc; + return sd; } static void *softnet_seq_start(struct seq_file *seq, loff_t *pos) @@ -3721,12 +3718,12 @@ static void softnet_seq_stop(struct seq_file *seq, void *v) static int softnet_seq_show(struct seq_file *seq, void *v) { - struct netif_rx_stats *s = v; + struct softnet_data *sd = v; seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x %08x %08x\n", - s->total, s->dropped, s->time_squeeze, 0, + sd->processed, sd->dropped, sd->time_squeeze, 0, 0, 0, 0, 0, /* was fastroute */ - s->cpu_collision, s->received_rps); + sd->cpu_collision, sd->received_rps); return 0; } @@ -5869,6 +5866,7 @@ static int __init net_dev_init(void) for_each_possible_cpu(i) { struct softnet_data *sd = &per_cpu(softnet_data, i); + memset(sd, 0, sizeof(*sd)); skb_queue_head_init(&sd->input_pkt_queue); skb_queue_head_init(&sd->process_queue); sd->completion_queue = NULL; diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index aeddabfb8e4e..a969b111bd76 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -94,7 +94,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb, * Another cpu is holding lock, requeue & delay xmits for * some time. */ - __get_cpu_var(netdev_rx_stat).cpu_collision++; + __get_cpu_var(softnet_data).cpu_collision++; ret = dev_requeue_skb(skb, q); } -- cgit v1.2.3 From eecfd7c4e36ff532d895885971d01d049bd3e014 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 May 2010 22:07:48 -0700 Subject: rps: Various optimizations Introduce ____napi_schedule() helper for callers in irq disabled contexts. rps_trigger_softirq() becomes a leaf function. Use container_of() in process_backlog() instead of accessing per_cpu address. Use a custom inlined version of __napi_complete() in process_backlog() to avoid one locked instruction : only current cpu owns and manipulates this napi, and NAPI_STATE_SCHED is the only possible flag set on backlog. we can use a plain write instead of clear_bit(), and we dont need an smp_mb() memory barrier, since RPS is on, backlog is protected by a spinlock. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 28 ++++++++++++++++++++++------ 1 file changed, 22 insertions(+), 6 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 36d53be4fca6..32611c8f1219 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2205,6 +2205,14 @@ int netdev_max_backlog __read_mostly = 1000; int netdev_budget __read_mostly = 300; int weight_p __read_mostly = 64; /* old backlog weight */ +/* Called with irq disabled */ +static inline void ____napi_schedule(struct softnet_data *sd, + struct napi_struct *napi) +{ + list_add_tail(&napi->poll_list, &sd->poll_list); + __raise_softirq_irqoff(NET_RX_SOFTIRQ); +} + #ifdef CONFIG_RPS /* One global table that all flow-based protocols share. */ @@ -2363,7 +2371,7 @@ static void rps_trigger_softirq(void *data) { struct softnet_data *sd = data; - __napi_schedule(&sd->backlog); + ____napi_schedule(sd, &sd->backlog); sd->received_rps++; } @@ -2421,7 +2429,7 @@ enqueue: /* Schedule NAPI for backlog device */ if (napi_schedule_prep(&sd->backlog)) { if (!rps_ipi_queued(sd)) - __napi_schedule(&sd->backlog); + ____napi_schedule(sd, &sd->backlog); } goto enqueue; } @@ -3280,7 +3288,7 @@ static void net_rps_action_and_irq_enable(struct softnet_data *sd) static int process_backlog(struct napi_struct *napi, int quota) { int work = 0; - struct softnet_data *sd = &__get_cpu_var(softnet_data); + struct softnet_data *sd = container_of(napi, struct softnet_data, backlog); #ifdef CONFIG_RPS /* Check if we have pending ipi, its better to send them now, @@ -3313,7 +3321,16 @@ static int process_backlog(struct napi_struct *napi, int quota) &sd->process_queue); } if (qlen < quota - work) { - __napi_complete(napi); + /* + * Inline a custom version of __napi_complete(). + * only current cpu owns and manipulates this napi, + * and NAPI_STATE_SCHED is the only possible flag set on backlog. + * we can use a plain write instead of clear_bit(), + * and we dont need an smp_mb() memory barrier. + */ + list_del(&napi->poll_list); + napi->state = 0; + quota = work + qlen; } rps_unlock(sd); @@ -3334,8 +3351,7 @@ void __napi_schedule(struct napi_struct *n) unsigned long flags; local_irq_save(flags); - list_add_tail(&n->poll_list, &__get_cpu_var(softnet_data).poll_list); - __raise_softirq_irqoff(NET_RX_SOFTIRQ); + ____napi_schedule(&__get_cpu_var(softnet_data), n); local_irq_restore(flags); } EXPORT_SYMBOL(__napi_schedule); -- cgit v1.2.3 From a14462f1bd4d3962994f518459102000438665aa Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 6 May 2010 01:33:53 +0000 Subject: net: adjust handle_macvlan to pass port struct to hook Now there's null check here and also again in the hook. Looking at bridge bits which are simmilar, port structure is rcu_dereferenced right away in handle_bridge and passed to hook. Looks nicer. Signed-off-by: Jiri Pirko Acked-by: Patrick McHardy Signed-off-by: David S. Miller --- drivers/net/macvlan.c | 8 ++------ include/linux/if_macvlan.h | 3 ++- net/core/dev.c | 10 +++++++--- 3 files changed, 11 insertions(+), 10 deletions(-) (limited to 'net/core/dev.c') diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 9a939d828b47..1b78c0057a8d 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -145,19 +145,15 @@ static void macvlan_broadcast(struct sk_buff *skb, } /* called under rcu_read_lock() from netif_receive_skb */ -static struct sk_buff *macvlan_handle_frame(struct sk_buff *skb) +static struct sk_buff *macvlan_handle_frame(struct macvlan_port *port, + struct sk_buff *skb) { const struct ethhdr *eth = eth_hdr(skb); - const struct macvlan_port *port; const struct macvlan_dev *vlan; const struct macvlan_dev *src; struct net_device *dev; unsigned int len; - port = rcu_dereference(skb->dev->macvlan_port); - if (port == NULL) - return skb; - if (is_multicast_ether_addr(eth->h_dest)) { src = macvlan_hash_lookup(port, eth->h_source); if (!src) diff --git a/include/linux/if_macvlan.h b/include/linux/if_macvlan.h index b78a712247da..9ea047aca795 100644 --- a/include/linux/if_macvlan.h +++ b/include/linux/if_macvlan.h @@ -85,6 +85,7 @@ extern netdev_tx_t macvlan_start_xmit(struct sk_buff *skb, struct net_device *dev); -extern struct sk_buff *(*macvlan_handle_frame_hook)(struct sk_buff *); +extern struct sk_buff *(*macvlan_handle_frame_hook)(struct macvlan_port *, + struct sk_buff *); #endif /* _LINUX_IF_MACVLAN_H */ diff --git a/net/core/dev.c b/net/core/dev.c index 3daee30a7c82..5cbba0927a8e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2612,7 +2612,8 @@ static inline struct sk_buff *handle_bridge(struct sk_buff *skb, #endif #if defined(CONFIG_MACVLAN) || defined(CONFIG_MACVLAN_MODULE) -struct sk_buff *(*macvlan_handle_frame_hook)(struct sk_buff *skb) __read_mostly; +struct sk_buff *(*macvlan_handle_frame_hook)(struct macvlan_port *p, + struct sk_buff *skb) __read_mostly; EXPORT_SYMBOL_GPL(macvlan_handle_frame_hook); static inline struct sk_buff *handle_macvlan(struct sk_buff *skb, @@ -2620,14 +2621,17 @@ static inline struct sk_buff *handle_macvlan(struct sk_buff *skb, int *ret, struct net_device *orig_dev) { - if (skb->dev->macvlan_port == NULL) + struct macvlan_port *port; + + port = rcu_dereference(skb->dev->macvlan_port); + if (!port) return skb; if (*pt_prev) { *ret = deliver_skb(skb, *pt_prev, orig_dev); *pt_prev = NULL; } - return macvlan_handle_frame_hook(skb); + return macvlan_handle_frame_hook(port, skb); } #else #define handle_macvlan(skb, pt_prev, ret, orig_dev) (skb) -- cgit v1.2.3 From 3b098e2d7c693796cc4dffb07caa249fc0f70771 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 15 May 2010 23:57:10 -0700 Subject: net: Consistent skb timestamping With RPS inclusion, skb timestamping is not consistent in RX path. If netif_receive_skb() is used, its deferred after RPS dispatch. If netif_rx() is used, its done before RPS dispatch. This can give strange tcpdump timestamps results. I think timestamping should be done as soon as possible in the receive path, to get meaningful values (ie timestamps taken at the time packet was delivered by NIC driver to our stack), even if NAPI already can defer timestamping a bit (RPS can help to reduce the gap) Tom Herbert prefer to sample timestamps after RPS dispatch. In case sampling is expensive (HPET/acpi_pm on x86), this makes sense. Let admins switch from one mode to another, using a new sysctl, /proc/sys/net/core/netdev_tstamp_prequeue Its default value (1), means timestamps are taken as soon as possible, before backlog queueing, giving accurate timestamps. Setting a 0 value permits to sample timestamps when processing backlog, after RPS dispatch, to lower the load of the pre-RPS cpu. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- Documentation/sysctl/net.txt | 10 +++++++++ include/linux/netdevice.h | 1 + net/core/dev.c | 50 +++++++++++++++++++++++++++----------------- net/core/sysctl_net_core.c | 7 +++++++ 4 files changed, 49 insertions(+), 19 deletions(-) (limited to 'net/core/dev.c') diff --git a/Documentation/sysctl/net.txt b/Documentation/sysctl/net.txt index df38ef046f8d..cbd05ffc606b 100644 --- a/Documentation/sysctl/net.txt +++ b/Documentation/sysctl/net.txt @@ -84,6 +84,16 @@ netdev_max_backlog Maximum number of packets, queued on the INPUT side, when the interface receives packets faster than kernel can process them. +netdev_tstamp_prequeue +---------------------- + +If set to 0, RX packet timestamps can be sampled after RPS processing, when +the target CPU processes packets. It might give some delay on timestamps, but +permit to distribute the load on several cpus. + +If set to 1 (default), timestamps are sampled as soon as possible, before +queueing. + optmem_max ---------- diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 69022d47d6f2..c1b2341897c2 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2100,6 +2100,7 @@ extern const struct net_device_stats *dev_get_stats(struct net_device *dev); extern void dev_txq_stats_fold(const struct net_device *dev, struct net_device_stats *stats); extern int netdev_max_backlog; +extern int netdev_tstamp_prequeue; extern int weight_p; extern int netdev_set_master(struct net_device *dev, struct net_device *master); extern int skb_checksum_help(struct sk_buff *skb); diff --git a/net/core/dev.c b/net/core/dev.c index 5cbba0927a8e..988e42912e72 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1454,7 +1454,7 @@ void net_disable_timestamp(void) } EXPORT_SYMBOL(net_disable_timestamp); -static inline void net_timestamp(struct sk_buff *skb) +static inline void net_timestamp_set(struct sk_buff *skb) { if (atomic_read(&netstamp_needed)) __net_timestamp(skb); @@ -1462,6 +1462,12 @@ static inline void net_timestamp(struct sk_buff *skb) skb->tstamp.tv64 = 0; } +static inline void net_timestamp_check(struct sk_buff *skb) +{ + if (!skb->tstamp.tv64 && atomic_read(&netstamp_needed)) + __net_timestamp(skb); +} + /** * dev_forward_skb - loopback an skb to another netif * @@ -1508,9 +1514,9 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev) #ifdef CONFIG_NET_CLS_ACT if (!(skb->tstamp.tv64 && (G_TC_FROM(skb->tc_verd) & AT_INGRESS))) - net_timestamp(skb); + net_timestamp_set(skb); #else - net_timestamp(skb); + net_timestamp_set(skb); #endif rcu_read_lock(); @@ -2201,6 +2207,7 @@ EXPORT_SYMBOL(dev_queue_xmit); =======================================================================*/ int netdev_max_backlog __read_mostly = 1000; +int netdev_tstamp_prequeue __read_mostly = 1; int netdev_budget __read_mostly = 300; int weight_p __read_mostly = 64; /* old backlog weight */ @@ -2465,8 +2472,8 @@ int netif_rx(struct sk_buff *skb) if (netpoll_rx(skb)) return NET_RX_DROP; - if (!skb->tstamp.tv64) - net_timestamp(skb); + if (netdev_tstamp_prequeue) + net_timestamp_check(skb); #ifdef CONFIG_RPS { @@ -2791,8 +2798,8 @@ static int __netif_receive_skb(struct sk_buff *skb) int ret = NET_RX_DROP; __be16 type; - if (!skb->tstamp.tv64) - net_timestamp(skb); + if (!netdev_tstamp_prequeue) + net_timestamp_check(skb); if (vlan_tx_tag_present(skb) && vlan_hwaccel_do_receive(skb)) return NET_RX_SUCCESS; @@ -2910,23 +2917,28 @@ out: */ int netif_receive_skb(struct sk_buff *skb) { + if (netdev_tstamp_prequeue) + net_timestamp_check(skb); + #ifdef CONFIG_RPS - struct rps_dev_flow voidflow, *rflow = &voidflow; - int cpu, ret; + { + struct rps_dev_flow voidflow, *rflow = &voidflow; + int cpu, ret; - rcu_read_lock(); + rcu_read_lock(); + + cpu = get_rps_cpu(skb->dev, skb, &rflow); - cpu = get_rps_cpu(skb->dev, skb, &rflow); + if (cpu >= 0) { + ret = enqueue_to_backlog(skb, cpu, &rflow->last_qtail); + rcu_read_unlock(); + } else { + rcu_read_unlock(); + ret = __netif_receive_skb(skb); + } - if (cpu >= 0) { - ret = enqueue_to_backlog(skb, cpu, &rflow->last_qtail); - rcu_read_unlock(); - } else { - rcu_read_unlock(); - ret = __netif_receive_skb(skb); + return ret; } - - return ret; #else return __netif_receive_skb(skb); #endif diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index dcc7d25996ab..01eee5d984be 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -121,6 +121,13 @@ static struct ctl_table net_core_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "netdev_tstamp_prequeue", + .data = &netdev_tstamp_prequeue, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { .procname = "message_cost", .data = &net_ratelimit_state.interval, -- cgit v1.2.3 From ebda37c27d0c768947e9b058332d7ea798210cf8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 May 2010 23:51:21 +0000 Subject: rps: avoid one atomic in enqueue_to_backlog If CONFIG_SMP=y, then we own a queue spinlock, we can avoid the atomic test_and_set_bit() from napi_schedule_prep(). We now have same number of atomic ops per netif_rx() calls than with pre-RPS kernel. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 988e42912e72..cdcb9cbedf41 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2432,8 +2432,10 @@ enqueue: return NET_RX_SUCCESS; } - /* Schedule NAPI for backlog device */ - if (napi_schedule_prep(&sd->backlog)) { + /* Schedule NAPI for backlog device + * We can use non atomic operation since we own the queue lock + */ + if (!__test_and_set_bit(NAPI_STATE_SCHED, &sd->backlog.state)) { if (!rps_ipi_queued(sd)) ____napi_schedule(sd, &sd->backlog); } -- cgit v1.2.3 From 7fee226ad2397b635e2fd565a59ca3ae08a164cd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 11 May 2010 23:19:48 +0000 Subject: net: add a noref bit on skb dst Use low order bit of skb->_skb_dst to tell dst is not refcounted. Change _skb_dst to _skb_refdst to make sure all uses are catched. skb_dst() returns the dst, regardless of noref bit set or not, but with a lockdep check to make sure a noref dst is not given if current user is not rcu protected. New skb_dst_set_noref() helper to set an notrefcounted dst on a skb. (with lockdep check) skb_dst_drop() drops a reference only if skb dst was refcounted. skb_dst_force() helper is used to force a refcount on dst, when skb is queued and not anymore RCU protected. Use skb_dst_force() in __sk_add_backlog(), __dev_xmit_skb() if !IFF_XMIT_DST_RELEASE or skb enqueued on qdisc queue, in sock_queue_rcv_skb(), in __nf_queue(). Use skb_dst_force() in dev_requeue_skb(). Note: dst_use_noref() still dirties dst, we might transform it later to do one dirtying per jiffies. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/skbuff.h | 58 ++++++++++++++++++++++++++++++++++++++++++++---- include/net/dst.h | 48 ++++++++++++++++++++++++++++++++++++--- include/net/sock.h | 13 ++++++----- net/core/dev.c | 3 +++ net/core/skbuff.c | 2 +- net/core/sock.c | 6 +++++ net/ipv4/icmp.c | 6 ++--- net/ipv4/ip_options.c | 9 ++++---- net/ipv4/netfilter.c | 6 ++--- net/ipv4/route.c | 2 +- net/netfilter/nf_queue.c | 2 ++ net/sched/sch_generic.c | 4 +++- 12 files changed, 134 insertions(+), 25 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index c9525bce80f6..7cdfb4d52847 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -264,7 +264,7 @@ typedef unsigned char *sk_buff_data_t; * @transport_header: Transport layer header * @network_header: Network layer header * @mac_header: Link layer header - * @_skb_dst: destination entry + * @_skb_refdst: destination entry (with norefcount bit) * @sp: the security path, used for xfrm * @cb: Control buffer. Free for use by every layer. Put private vars here * @len: Length of actual data @@ -328,7 +328,7 @@ struct sk_buff { */ char cb[48] __aligned(8); - unsigned long _skb_dst; + unsigned long _skb_refdst; #ifdef CONFIG_XFRM struct sec_path *sp; #endif @@ -419,14 +419,64 @@ struct sk_buff { #include +/* + * skb might have a dst pointer attached, refcounted or not. + * _skb_refdst low order bit is set if refcount was _not_ taken + */ +#define SKB_DST_NOREF 1UL +#define SKB_DST_PTRMASK ~(SKB_DST_NOREF) + +/** + * skb_dst - returns skb dst_entry + * @skb: buffer + * + * Returns skb dst_entry, regardless of reference taken or not. + */ static inline struct dst_entry *skb_dst(const struct sk_buff *skb) { - return (struct dst_entry *)skb->_skb_dst; + /* If refdst was not refcounted, check we still are in a + * rcu_read_lock section + */ + WARN_ON((skb->_skb_refdst & SKB_DST_NOREF) && + !rcu_read_lock_held() && + !rcu_read_lock_bh_held()); + return (struct dst_entry *)(skb->_skb_refdst & SKB_DST_PTRMASK); } +/** + * skb_dst_set - sets skb dst + * @skb: buffer + * @dst: dst entry + * + * Sets skb dst, assuming a reference was taken on dst and should + * be released by skb_dst_drop() + */ static inline void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst) { - skb->_skb_dst = (unsigned long)dst; + skb->_skb_refdst = (unsigned long)dst; +} + +/** + * skb_dst_set_noref - sets skb dst, without a reference + * @skb: buffer + * @dst: dst entry + * + * Sets skb dst, assuming a reference was not taken on dst + * skb_dst_drop() should not dst_release() this dst + */ +static inline void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst) +{ + WARN_ON(!rcu_read_lock_held() && !rcu_read_lock_bh_held()); + skb->_skb_refdst = (unsigned long)dst | SKB_DST_NOREF; +} + +/** + * skb_dst_is_noref - Test if skb dst isnt refcounted + * @skb: buffer + */ +static inline bool skb_dst_is_noref(const struct sk_buff *skb) +{ + return (skb->_skb_refdst & SKB_DST_NOREF) && skb_dst(skb); } static inline struct rtable *skb_rtable(const struct sk_buff *skb) diff --git a/include/net/dst.h b/include/net/dst.h index aac5a5fcfda9..27207a13f2a6 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -168,6 +168,12 @@ static inline void dst_use(struct dst_entry *dst, unsigned long time) dst->lastuse = time; } +static inline void dst_use_noref(struct dst_entry *dst, unsigned long time) +{ + dst->__use++; + dst->lastuse = time; +} + static inline struct dst_entry * dst_clone(struct dst_entry * dst) { @@ -177,11 +183,47 @@ struct dst_entry * dst_clone(struct dst_entry * dst) } extern void dst_release(struct dst_entry *dst); + +static inline void refdst_drop(unsigned long refdst) +{ + if (!(refdst & SKB_DST_NOREF)) + dst_release((struct dst_entry *)(refdst & SKB_DST_PTRMASK)); +} + +/** + * skb_dst_drop - drops skb dst + * @skb: buffer + * + * Drops dst reference count if a reference was taken. + */ static inline void skb_dst_drop(struct sk_buff *skb) { - if (skb->_skb_dst) - dst_release(skb_dst(skb)); - skb->_skb_dst = 0UL; + if (skb->_skb_refdst) { + refdst_drop(skb->_skb_refdst); + skb->_skb_refdst = 0UL; + } +} + +static inline void skb_dst_copy(struct sk_buff *nskb, const struct sk_buff *oskb) +{ + nskb->_skb_refdst = oskb->_skb_refdst; + if (!(nskb->_skb_refdst & SKB_DST_NOREF)) + dst_clone(skb_dst(nskb)); +} + +/** + * skb_dst_force - makes sure skb dst is refcounted + * @skb: buffer + * + * If dst is not yet refcounted, let's do it + */ +static inline void skb_dst_force(struct sk_buff *skb) +{ + if (skb_dst_is_noref(skb)) { + WARN_ON(!rcu_read_lock_held()); + skb->_skb_refdst &= ~SKB_DST_NOREF; + dst_clone(skb_dst(skb)); + } } /* Children define the path of the packet through the diff --git a/include/net/sock.h b/include/net/sock.h index aed16eb9db4b..5697caf8cc76 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -600,12 +600,15 @@ static inline int sk_stream_memory_free(struct sock *sk) /* OOB backlog add */ static inline void __sk_add_backlog(struct sock *sk, struct sk_buff *skb) { - if (!sk->sk_backlog.tail) { - sk->sk_backlog.head = sk->sk_backlog.tail = skb; - } else { + /* dont let skb dst not refcounted, we are going to leave rcu lock */ + skb_dst_force(skb); + + if (!sk->sk_backlog.tail) + sk->sk_backlog.head = skb; + else sk->sk_backlog.tail->next = skb; - sk->sk_backlog.tail = skb; - } + + sk->sk_backlog.tail = skb; skb->next = NULL; } diff --git a/net/core/dev.c b/net/core/dev.c index cdcb9cbedf41..6c820650b80f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2052,6 +2052,8 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, * waiting to be sent out; and the qdisc is not running - * xmit the skb directly. */ + if (!(dev->priv_flags & IFF_XMIT_DST_RELEASE)) + skb_dst_force(skb); __qdisc_update_bstats(q, skb->len); if (sch_direct_xmit(skb, q, dev, txq, root_lock)) __qdisc_run(q); @@ -2060,6 +2062,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, rc = NET_XMIT_SUCCESS; } else { + skb_dst_force(skb); rc = qdisc_enqueue_root(skb, q); qdisc_run(q); } diff --git a/net/core/skbuff.c b/net/core/skbuff.c index a9b0e1f77806..c543dd252433 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -520,7 +520,7 @@ 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; - skb_dst_set(new, dst_clone(skb_dst(old))); + skb_dst_copy(new, old); new->rxhash = old->rxhash; #ifdef CONFIG_XFRM new->sp = secpath_get(old->sp); diff --git a/net/core/sock.c b/net/core/sock.c index 63530a03b8c2..bf88a167c8f2 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -307,6 +307,11 @@ int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) */ skb_len = skb->len; + /* we escape from rcu protected region, make sure we dont leak + * a norefcounted dst + */ + skb_dst_force(skb); + spin_lock_irqsave(&list->lock, flags); skb->dropcount = atomic_read(&sk->sk_drops); __skb_queue_tail(list, skb); @@ -1536,6 +1541,7 @@ static void __release_sock(struct sock *sk) do { struct sk_buff *next = skb->next; + WARN_ON_ONCE(skb_dst_is_noref(skb)); skb->next = NULL; sk_backlog_rcv(sk, skb); diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index f3d339f728b0..d65e9215bcd7 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -587,20 +587,20 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) err = __ip_route_output_key(net, &rt2, &fl); else { struct flowi fl2 = {}; - struct dst_entry *odst; + unsigned long orefdst; fl2.fl4_dst = fl.fl4_src; if (ip_route_output_key(net, &rt2, &fl2)) goto relookup_failed; /* Ugh! */ - odst = skb_dst(skb_in); + orefdst = skb_in->_skb_refdst; /* save old refdst */ 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_dst_set(skb_in, odst); + skb_in->_skb_refdst = orefdst; /* restore old refdst */ } if (err) diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index 4c09a31fd140..3244133c24f6 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -601,6 +601,7 @@ int ip_options_rcv_srr(struct sk_buff *skb) unsigned char *optptr = skb_network_header(skb) + opt->srr; struct rtable *rt = skb_rtable(skb); struct rtable *rt2; + unsigned long orefdst; int err; if (!opt->srr) @@ -624,16 +625,16 @@ int ip_options_rcv_srr(struct sk_buff *skb) } memcpy(&nexthop, &optptr[srrptr-1], 4); - rt = skb_rtable(skb); + orefdst = skb->_skb_refdst; 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_set(skb, &rt->u.dst); + skb_dst_drop(skb); + skb->_skb_refdst = orefdst; return -EINVAL; } - ip_rt_put(rt); + refdst_drop(orefdst); if (rt2->rt_type != RTN_LOCAL) break; /* Superfast 8) loopback forward */ diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index 82fb43c5c59e..07de855e2175 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -17,7 +17,7 @@ int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type) const struct iphdr *iph = ip_hdr(skb); struct rtable *rt; struct flowi fl = {}; - struct dst_entry *odst; + unsigned long orefdst; unsigned int hh_len; unsigned int type; @@ -51,14 +51,14 @@ 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(skb); + orefdst = skb->_skb_refdst; if (ip_route_input(skb, iph->daddr, iph->saddr, RT_TOS(iph->tos), rt->u.dst.dev) != 0) { dst_release(&rt->u.dst); return -1; } dst_release(&rt->u.dst); - dst_release(odst); + refdst_drop(orefdst); } if (skb_dst(skb)->error) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index dea3f9264250..705eccfb4769 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -3033,7 +3033,7 @@ int ip_rt_dump(struct sk_buff *skb, struct netlink_callback *cb) continue; if (rt_is_expired(rt)) continue; - skb_dst_set(skb, dst_clone(&rt->u.dst)); + skb_dst_set_noref(skb, &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) { diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index 0b1103c0b1f3..78b3cf9c519c 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -9,6 +9,7 @@ #include #include #include +#include #include "nf_internals.h" @@ -170,6 +171,7 @@ static int __nf_queue(struct sk_buff *skb, dev_hold(physoutdev); } #endif + skb_dst_force(skb); afinfo->saveroute(skb, entry); status = qh->outfn(entry, queuenum); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index a969b111bd76..a63029ef3edd 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -26,6 +26,7 @@ #include #include #include +#include /* Main transmission queue. */ @@ -40,6 +41,7 @@ static inline int dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q) { + skb_dst_force(skb); q->gso_skb = skb; q->qstats.requeues++; q->q.qlen++; /* it's still part of the queue */ @@ -179,7 +181,7 @@ static inline int qdisc_restart(struct Qdisc *q) skb = dequeue_skb(q); if (unlikely(!skb)) return 0; - + WARN_ON_ONCE(skb_dst_is_noref(skb)); root_lock = qdisc_lock(q); dev = qdisc_dev(q); txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb)); -- cgit v1.2.3