diff options
Diffstat (limited to 'net/core')
| -rw-r--r-- | net/core/dev.c | 81 | ||||
| -rw-r--r-- | net/core/flow.c | 2 | ||||
| -rw-r--r-- | net/core/link_watch.c | 166 | ||||
| -rw-r--r-- | net/core/net-sysfs.c | 8 | ||||
| -rw-r--r-- | net/core/netpoll.c | 1 | ||||
| -rw-r--r-- | net/core/pktgen.c | 1 | ||||
| -rw-r--r-- | net/core/skbuff.c | 5 |
7 files changed, 193 insertions, 71 deletions
diff --git a/net/core/dev.c b/net/core/dev.c index 4317c1be4d3f..5a7f20f78574 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -116,6 +116,7 @@ #include <linux/dmaengine.h> #include <linux/err.h> #include <linux/ctype.h> +#include <linux/if_arp.h> /* * The list of packet types we will receive (as opposed to discard) @@ -217,6 +218,73 @@ extern void netdev_unregister_sysfs(struct net_device *); #define netdev_unregister_sysfs(dev) do { } while(0) #endif +#ifdef CONFIG_DEBUG_LOCK_ALLOC +/* + * register_netdevice() inits dev->_xmit_lock and sets lockdep class + * according to dev->type + */ +static const unsigned short netdev_lock_type[] = + {ARPHRD_NETROM, ARPHRD_ETHER, ARPHRD_EETHER, ARPHRD_AX25, + ARPHRD_PRONET, ARPHRD_CHAOS, ARPHRD_IEEE802, ARPHRD_ARCNET, + ARPHRD_APPLETLK, ARPHRD_DLCI, ARPHRD_ATM, ARPHRD_METRICOM, + ARPHRD_IEEE1394, ARPHRD_EUI64, ARPHRD_INFINIBAND, ARPHRD_SLIP, + ARPHRD_CSLIP, ARPHRD_SLIP6, ARPHRD_CSLIP6, ARPHRD_RSRVD, + ARPHRD_ADAPT, ARPHRD_ROSE, ARPHRD_X25, ARPHRD_HWX25, + ARPHRD_PPP, ARPHRD_CISCO, ARPHRD_LAPB, ARPHRD_DDCMP, + ARPHRD_RAWHDLC, ARPHRD_TUNNEL, ARPHRD_TUNNEL6, ARPHRD_FRAD, + ARPHRD_SKIP, ARPHRD_LOOPBACK, ARPHRD_LOCALTLK, ARPHRD_FDDI, + ARPHRD_BIF, ARPHRD_SIT, ARPHRD_IPDDP, ARPHRD_IPGRE, + ARPHRD_PIMREG, ARPHRD_HIPPI, ARPHRD_ASH, ARPHRD_ECONET, + ARPHRD_IRDA, ARPHRD_FCPP, ARPHRD_FCAL, ARPHRD_FCPL, + ARPHRD_FCFABRIC, ARPHRD_IEEE802_TR, ARPHRD_IEEE80211, + ARPHRD_IEEE80211_PRISM, ARPHRD_IEEE80211_RADIOTAP, ARPHRD_VOID, + ARPHRD_NONE}; + +static const char *netdev_lock_name[] = + {"_xmit_NETROM", "_xmit_ETHER", "_xmit_EETHER", "_xmit_AX25", + "_xmit_PRONET", "_xmit_CHAOS", "_xmit_IEEE802", "_xmit_ARCNET", + "_xmit_APPLETLK", "_xmit_DLCI", "_xmit_ATM", "_xmit_METRICOM", + "_xmit_IEEE1394", "_xmit_EUI64", "_xmit_INFINIBAND", "_xmit_SLIP", + "_xmit_CSLIP", "_xmit_SLIP6", "_xmit_CSLIP6", "_xmit_RSRVD", + "_xmit_ADAPT", "_xmit_ROSE", "_xmit_X25", "_xmit_HWX25", + "_xmit_PPP", "_xmit_CISCO", "_xmit_LAPB", "_xmit_DDCMP", + "_xmit_RAWHDLC", "_xmit_TUNNEL", "_xmit_TUNNEL6", "_xmit_FRAD", + "_xmit_SKIP", "_xmit_LOOPBACK", "_xmit_LOCALTLK", "_xmit_FDDI", + "_xmit_BIF", "_xmit_SIT", "_xmit_IPDDP", "_xmit_IPGRE", + "_xmit_PIMREG", "_xmit_HIPPI", "_xmit_ASH", "_xmit_ECONET", + "_xmit_IRDA", "_xmit_FCPP", "_xmit_FCAL", "_xmit_FCPL", + "_xmit_FCFABRIC", "_xmit_IEEE802_TR", "_xmit_IEEE80211", + "_xmit_IEEE80211_PRISM", "_xmit_IEEE80211_RADIOTAP", "_xmit_VOID", + "_xmit_NONE"}; + +static struct lock_class_key netdev_xmit_lock_key[ARRAY_SIZE(netdev_lock_type)]; + +static inline unsigned short netdev_lock_pos(unsigned short dev_type) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(netdev_lock_type); i++) + if (netdev_lock_type[i] == dev_type) + return i; + /* the last key is used by default */ + return ARRAY_SIZE(netdev_lock_type) - 1; +} + +static inline void netdev_set_lockdep_class(spinlock_t *lock, + unsigned short dev_type) +{ + int i; + + i = netdev_lock_pos(dev_type); + lockdep_set_class_and_name(lock, &netdev_xmit_lock_key[i], + netdev_lock_name[i]); +} +#else +static inline void netdev_set_lockdep_class(spinlock_t *lock, + unsigned short dev_type) +{ +} +#endif /******************************************************************************* @@ -3001,6 +3069,7 @@ int register_netdevice(struct net_device *dev) spin_lock_init(&dev->queue_lock); spin_lock_init(&dev->_xmit_lock); + netdev_set_lockdep_class(&dev->_xmit_lock, dev->type); dev->xmit_lock_owner = -1; spin_lock_init(&dev->ingress_lock); @@ -3245,7 +3314,6 @@ void netdev_run_todo(void) continue; } - netdev_unregister_sysfs(dev); dev->reg_state = NETREG_UNREGISTERED; netdev_wait_allrefs(dev); @@ -3256,11 +3324,11 @@ void netdev_run_todo(void) BUG_TRAP(!dev->ip6_ptr); BUG_TRAP(!dev->dn_ptr); - /* It must be the very last action, - * after this 'dev' may point to freed up memory. - */ if (dev->destructor) dev->destructor(dev); + + /* Free network device */ + kobject_put(&dev->dev.kobj); } out: @@ -3411,6 +3479,9 @@ void unregister_netdevice(struct net_device *dev) /* Notifier chain MUST detach us from master device. */ BUG_TRAP(!dev->master); + /* Remove entries from sysfs */ + netdev_unregister_sysfs(dev); + /* Finish processing unregister after unlock */ net_set_todo(dev); @@ -3450,7 +3521,7 @@ static int dev_cpu_callback(struct notifier_block *nfb, unsigned int cpu, oldcpu = (unsigned long)ocpu; struct softnet_data *sd, *oldsd; - if (action != CPU_DEAD) + if (action != CPU_DEAD && action != CPU_DEAD_FROZEN) return NOTIFY_OK; local_irq_disable(); diff --git a/net/core/flow.c b/net/core/flow.c index 5d25697920b1..051430545a05 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -338,7 +338,7 @@ static int flow_cache_cpu(struct notifier_block *nfb, unsigned long action, void *hcpu) { - if (action == CPU_DEAD) + if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) __flow_cache_shrink((unsigned long)hcpu, 0); return NOTIFY_OK; } diff --git a/net/core/link_watch.c b/net/core/link_watch.c index e3c26a9ccad6..a5e372b9ec4d 100644 --- a/net/core/link_watch.c +++ b/net/core/link_watch.c @@ -19,7 +19,6 @@ #include <linux/rtnetlink.h> #include <linux/jiffies.h> #include <linux/spinlock.h> -#include <linux/list.h> #include <linux/slab.h> #include <linux/workqueue.h> #include <linux/bitops.h> @@ -27,8 +26,7 @@ enum lw_bits { - LW_RUNNING = 0, - LW_SE_USED + LW_URGENT = 0, }; static unsigned long linkwatch_flags; @@ -37,17 +35,9 @@ static unsigned long linkwatch_nextevent; static void linkwatch_event(struct work_struct *dummy); static DECLARE_DELAYED_WORK(linkwatch_work, linkwatch_event); -static LIST_HEAD(lweventlist); +static struct net_device *lweventlist; static DEFINE_SPINLOCK(lweventlist_lock); -struct lw_event { - struct list_head list; - struct net_device *dev; -}; - -/* Avoid kmalloc() for most systems */ -static struct lw_event singleevent; - static unsigned char default_operstate(const struct net_device *dev) { if (!netif_carrier_ok(dev)) @@ -87,25 +77,102 @@ static void rfc2863_policy(struct net_device *dev) } -/* Must be called with the rtnl semaphore held */ -void linkwatch_run_queue(void) +static int linkwatch_urgent_event(struct net_device *dev) { - struct list_head head, *n, *next; + return netif_running(dev) && netif_carrier_ok(dev) && + dev->qdisc != dev->qdisc_sleeping; +} + + +static void linkwatch_add_event(struct net_device *dev) +{ + unsigned long flags; + + spin_lock_irqsave(&lweventlist_lock, flags); + dev->link_watch_next = lweventlist; + lweventlist = dev; + spin_unlock_irqrestore(&lweventlist_lock, flags); +} + + +static void linkwatch_schedule_work(int urgent) +{ + unsigned long delay = linkwatch_nextevent - jiffies; + + if (test_bit(LW_URGENT, &linkwatch_flags)) + return; + + /* Minimise down-time: drop delay for up event. */ + if (urgent) { + if (test_and_set_bit(LW_URGENT, &linkwatch_flags)) + return; + delay = 0; + } + + /* If we wrap around we'll delay it by at most HZ. */ + if (delay > HZ) + delay = 0; + + /* + * This is true if we've scheduled it immeditately or if we don't + * need an immediate execution and it's already pending. + */ + if (schedule_delayed_work(&linkwatch_work, delay) == !delay) + return; + + /* Don't bother if there is nothing urgent. */ + if (!test_bit(LW_URGENT, &linkwatch_flags)) + return; + + /* It's already running which is good enough. */ + if (!cancel_delayed_work(&linkwatch_work)) + return; + + /* Otherwise we reschedule it again for immediate exection. */ + schedule_delayed_work(&linkwatch_work, 0); +} + + +static void __linkwatch_run_queue(int urgent_only) +{ + struct net_device *next; + + /* + * Limit the number of linkwatch events to one + * per second so that a runaway driver does not + * cause a storm of messages on the netlink + * socket. This limit does not apply to up events + * while the device qdisc is down. + */ + if (!urgent_only) + linkwatch_nextevent = jiffies + HZ; + /* Limit wrap-around effect on delay. */ + else if (time_after(linkwatch_nextevent, jiffies + HZ)) + linkwatch_nextevent = jiffies; + + clear_bit(LW_URGENT, &linkwatch_flags); spin_lock_irq(&lweventlist_lock); - list_replace_init(&lweventlist, &head); + next = lweventlist; + lweventlist = NULL; spin_unlock_irq(&lweventlist_lock); - list_for_each_safe(n, next, &head) { - struct lw_event *event = list_entry(n, struct lw_event, list); - struct net_device *dev = event->dev; + while (next) { + struct net_device *dev = next; - if (event == &singleevent) { - clear_bit(LW_SE_USED, &linkwatch_flags); - } else { - kfree(event); + next = dev->link_watch_next; + + if (urgent_only && !linkwatch_urgent_event(dev)) { + linkwatch_add_event(dev); + continue; } + /* + * Make sure the above read is complete since it can be + * rewritten as soon as we clear the bit below. + */ + smp_mb__before_clear_bit(); + /* We are about to handle this device, * so new events can be accepted */ @@ -124,58 +191,39 @@ void linkwatch_run_queue(void) dev_put(dev); } + + if (lweventlist) + linkwatch_schedule_work(0); } -static void linkwatch_event(struct work_struct *dummy) +/* Must be called with the rtnl semaphore held */ +void linkwatch_run_queue(void) { - /* Limit the number of linkwatch events to one - * per second so that a runaway driver does not - * cause a storm of messages on the netlink - * socket - */ - linkwatch_nextevent = jiffies + HZ; - clear_bit(LW_RUNNING, &linkwatch_flags); + __linkwatch_run_queue(0); +} + +static void linkwatch_event(struct work_struct *dummy) +{ rtnl_lock(); - linkwatch_run_queue(); + __linkwatch_run_queue(time_after(linkwatch_nextevent, jiffies)); rtnl_unlock(); } void linkwatch_fire_event(struct net_device *dev) { - if (!test_and_set_bit(__LINK_STATE_LINKWATCH_PENDING, &dev->state)) { - unsigned long flags; - struct lw_event *event; - - if (test_and_set_bit(LW_SE_USED, &linkwatch_flags)) { - event = kmalloc(sizeof(struct lw_event), GFP_ATOMIC); - - if (unlikely(event == NULL)) { - clear_bit(__LINK_STATE_LINKWATCH_PENDING, &dev->state); - return; - } - } else { - event = &singleevent; - } + int urgent = linkwatch_urgent_event(dev); + if (!test_and_set_bit(__LINK_STATE_LINKWATCH_PENDING, &dev->state)) { dev_hold(dev); - event->dev = dev; - - spin_lock_irqsave(&lweventlist_lock, flags); - list_add_tail(&event->list, &lweventlist); - spin_unlock_irqrestore(&lweventlist_lock, flags); - if (!test_and_set_bit(LW_RUNNING, &linkwatch_flags)) { - unsigned long delay = linkwatch_nextevent - jiffies; + linkwatch_add_event(dev); + } else if (!urgent) + return; - /* If we wrap around we'll delay it by at most HZ. */ - if (delay > HZ) - delay = 0; - schedule_delayed_work(&linkwatch_work, delay); - } - } + linkwatch_schedule_work(urgent); } EXPORT_SYMBOL(linkwatch_fire_event); diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index b21307b15b82..5c19b0646d7a 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -456,9 +456,15 @@ static struct class net_class = { #endif }; +/* Delete sysfs entries but hold kobject reference until after all + * netdev references are gone. + */ void netdev_unregister_sysfs(struct net_device * net) { - device_del(&(net->dev)); + struct device *dev = &(net->dev); + + kobject_get(&dev->kobj); + device_del(dev); } /* Create sysfs entries for network device. */ diff --git a/net/core/netpoll.c b/net/core/netpoll.c index b316435b0e2a..758dafe284c0 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -9,7 +9,6 @@ * Copyright (C) 2002 Red Hat, Inc. */ -#include <linux/smp_lock.h> #include <linux/netdevice.h> #include <linux/etherdevice.h> #include <linux/string.h> diff --git a/net/core/pktgen.c b/net/core/pktgen.c index b92a322872a8..9cd3a1cb60ef 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -117,7 +117,6 @@ #include <linux/module.h> #include <linux/moduleparam.h> #include <linux/kernel.h> -#include <linux/smp_lock.h> #include <linux/mutex.h> #include <linux/sched.h> #include <linux/slab.h> diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 142257307fa2..7c6a34e21eee 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -644,11 +644,10 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, /* Copy only real data... and, alas, header. This should be * optimized for the cases when header is void. */ - memcpy(data + nhead, skb->head, #ifdef NET_SKBUFF_DATA_USES_OFFSET - skb->tail); + memcpy(data + nhead, skb->head, skb->tail); #else - skb->tail - skb->head); + memcpy(data + nhead, skb->head, skb->tail - skb->head); #endif memcpy(data + size, skb_end_pointer(skb), sizeof(struct skb_shared_info)); |