From 19b05f811341aaef3e5e22d2832aa2d8e0bad5ab Mon Sep 17 00:00:00 2001 From: david decotigny Date: Wed, 16 Nov 2011 12:15:08 +0000 Subject: net-sysfs: fixed minor sparse warning This commit fixes following warning: net/core/net-sysfs.c:921:6: warning: symbol 'numa_node' shadows an earlier one include/linux/topology.h:222:1: originally declared here Signed-off-by: David Decotigny Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'net/core/net-sysfs.c') diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index c71c434a4c05..a64382f201b8 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -901,7 +901,7 @@ static ssize_t store_xps_map(struct netdev_queue *queue, struct xps_map *map, *new_map; struct xps_dev_maps *dev_maps, *new_dev_maps; int nonempty = 0; - int numa_node = -2; + int numa_node_id = -2; if (!capable(CAP_NET_ADMIN)) return -EPERM; @@ -944,10 +944,10 @@ static ssize_t store_xps_map(struct netdev_queue *queue, need_set = cpumask_test_cpu(cpu, mask) && cpu_online(cpu); #ifdef CONFIG_NUMA if (need_set) { - if (numa_node == -2) - numa_node = cpu_to_node(cpu); - else if (numa_node != cpu_to_node(cpu)) - numa_node = -1; + if (numa_node_id == -2) + numa_node_id = cpu_to_node(cpu); + else if (numa_node_id != cpu_to_node(cpu)) + numa_node_id = -1; } #endif if (need_set && pos >= map_len) { @@ -997,7 +997,7 @@ static ssize_t store_xps_map(struct netdev_queue *queue, if (dev_maps) kfree_rcu(dev_maps, rcu); - netdev_queue_numa_node_write(queue, (numa_node >= 0) ? numa_node : + netdev_queue_numa_node_write(queue, (numa_node_id >= 0) ? numa_node_id : NUMA_NO_NODE); mutex_unlock(&xps_map_mutex); -- cgit v1.2.3 From ccf5ff69fbbd8d877377f5786369cf5aa78a15fc Mon Sep 17 00:00:00 2001 From: david decotigny Date: Wed, 16 Nov 2011 12:15:10 +0000 Subject: net: new counter for tx_timeout errors in sysfs This adds the /sys/class/net/DEV/queues/Q/tx_timeout attribute containing the total number of timeout events on the given queue. It is always available with CONFIG_SYSFS, independently of CONFIG_RPS/XPS. Credits to Stephen Hemminger for a preliminary version of this patch. Tested: without CONFIG_SYSFS (compilation only) with sysfs and without CONFIG_RPS & CONFIG_XPS with sysfs and without CONFIG_RPS with sysfs and without CONFIG_XPS with defaults Signed-off-by: David Decotigny Signed-off-by: David S. Miller --- include/linux/netdevice.h | 12 ++++++++++-- net/core/net-sysfs.c | 37 +++++++++++++++++++++++++++++++------ net/sched/sch_generic.c | 1 + 3 files changed, 42 insertions(+), 8 deletions(-) (limited to 'net/core/net-sysfs.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 31da3bbe7b1b..4d5698aa828b 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -532,7 +532,7 @@ struct netdev_queue { struct Qdisc *qdisc; unsigned long state; struct Qdisc *qdisc_sleeping; -#if defined(CONFIG_RPS) || defined(CONFIG_XPS) +#ifdef CONFIG_SYSFS struct kobject kobj; #endif #if defined(CONFIG_XPS) && defined(CONFIG_NUMA) @@ -547,6 +547,12 @@ struct netdev_queue { * please use this field instead of dev->trans_start */ unsigned long trans_start; + + /* + * Number of TX timeouts for this queue + * (/sys/class/net/DEV/Q/trans_timeout) + */ + unsigned long trans_timeout; } ____cacheline_aligned_in_smp; static inline int netdev_queue_numa_node_read(const struct netdev_queue *q) @@ -1109,9 +1115,11 @@ struct net_device { unsigned char broadcast[MAX_ADDR_LEN]; /* hw bcast add */ -#if defined(CONFIG_RPS) || defined(CONFIG_XPS) +#ifdef CONFIG_SYSFS struct kset *queues_kset; +#endif +#ifdef CONFIG_RPS struct netdev_rx_queue *_rx; /* Number of RX queues allocated at register_netdev() time */ diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index a64382f201b8..602b1419998c 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -780,7 +780,7 @@ net_rx_queue_update_kobjects(struct net_device *net, int old_num, int new_num) #endif } -#ifdef CONFIG_XPS +#ifdef CONFIG_SYSFS /* * netdev_queue sysfs structures and functions. */ @@ -826,6 +826,23 @@ static const struct sysfs_ops netdev_queue_sysfs_ops = { .store = netdev_queue_attr_store, }; +static ssize_t show_trans_timeout(struct netdev_queue *queue, + struct netdev_queue_attribute *attribute, + char *buf) +{ + unsigned long trans_timeout; + + spin_lock_irq(&queue->_xmit_lock); + trans_timeout = queue->trans_timeout; + spin_unlock_irq(&queue->_xmit_lock); + + return sprintf(buf, "%lu", trans_timeout); +} + +static struct netdev_queue_attribute queue_trans_timeout = + __ATTR(tx_timeout, S_IRUGO, show_trans_timeout, NULL); + +#ifdef CONFIG_XPS static inline unsigned int get_netdev_queue_index(struct netdev_queue *queue) { struct net_device *dev = queue->dev; @@ -1020,12 +1037,17 @@ error: static struct netdev_queue_attribute xps_cpus_attribute = __ATTR(xps_cpus, S_IRUGO | S_IWUSR, show_xps_map, store_xps_map); +#endif /* CONFIG_XPS */ static struct attribute *netdev_queue_default_attrs[] = { + &queue_trans_timeout.attr, +#ifdef CONFIG_XPS &xps_cpus_attribute.attr, +#endif NULL }; +#ifdef CONFIG_XPS static void netdev_queue_release(struct kobject *kobj) { struct netdev_queue *queue = to_netdev_queue(kobj); @@ -1076,10 +1098,13 @@ static void netdev_queue_release(struct kobject *kobj) memset(kobj, 0, sizeof(*kobj)); dev_put(queue->dev); } +#endif /* CONFIG_XPS */ static struct kobj_type netdev_queue_ktype = { .sysfs_ops = &netdev_queue_sysfs_ops, +#ifdef CONFIG_XPS .release = netdev_queue_release, +#endif .default_attrs = netdev_queue_default_attrs, }; @@ -1102,12 +1127,12 @@ static int netdev_queue_add_kobject(struct net_device *net, int index) return error; } -#endif /* CONFIG_XPS */ +#endif /* CONFIG_SYSFS */ int netdev_queue_update_kobjects(struct net_device *net, int old_num, int new_num) { -#ifdef CONFIG_XPS +#ifdef CONFIG_SYSFS int i; int error = 0; @@ -1125,14 +1150,14 @@ netdev_queue_update_kobjects(struct net_device *net, int old_num, int new_num) return error; #else return 0; -#endif +#endif /* CONFIG_SYSFS */ } static int register_queue_kobjects(struct net_device *net) { int error = 0, txq = 0, rxq = 0, real_rx = 0, real_tx = 0; -#if defined(CONFIG_RPS) || defined(CONFIG_XPS) +#ifdef CONFIG_SYSFS net->queues_kset = kset_create_and_add("queues", NULL, &net->dev.kobj); if (!net->queues_kset) @@ -1173,7 +1198,7 @@ static void remove_queue_kobjects(struct net_device *net) net_rx_queue_update_kobjects(net, real_rx, 0); netdev_queue_update_kobjects(net, real_tx, 0); -#if defined(CONFIG_RPS) || defined(CONFIG_XPS) +#ifdef CONFIG_SYSFS kset_unregister(net->queues_kset); #endif } diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 69fca2798804..79ac1458c2ba 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -246,6 +246,7 @@ static void dev_watchdog(unsigned long arg) time_after(jiffies, (trans_start + dev->watchdog_timeo))) { some_queue_timedout = 1; + txq->trans_timeout++; break; } } -- cgit v1.2.3 From adc9300e78e6091a7eaa1821213836379d4dbaa8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 17 Nov 2011 03:13:26 +0000 Subject: net: use jump_label to shortcut RPS if not setup Most machines dont use RPS/RFS, and pay a fair amount of instructions in netif_receive_skb() / netif_rx() / get_rps_cpu() just to discover RPS/RFS is not setup. Add a jump_label named rps_needed. If no device rps_map or global rps_sock_flow_table is setup, netif_receive_skb() / netif_rx() do a single instruction instead of many ones, including conditional jumps. jmp +0 (if CONFIG_JUMP_LABEL=y) Signed-off-by: Eric Dumazet CC: Tom Herbert Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 +++++ net/core/dev.c | 21 +++++++++------------ net/core/net-sysfs.c | 7 +++++-- net/core/sysctl_net_core.c | 9 +++++++-- 4 files changed, 26 insertions(+), 16 deletions(-) (limited to 'net/core/net-sysfs.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 4d5698aa828b..0bbe030fc014 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -214,6 +214,11 @@ enum { #include #include +#ifdef CONFIG_RPS +#include +extern struct jump_label_key rps_needed; +#endif + struct neighbour; struct neigh_parms; struct sk_buff; diff --git a/net/core/dev.c b/net/core/dev.c index 26c49d55e79d..f78959996148 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2711,6 +2711,8 @@ EXPORT_SYMBOL(__skb_get_rxhash); struct rps_sock_flow_table __rcu *rps_sock_flow_table __read_mostly; EXPORT_SYMBOL(rps_sock_flow_table); +struct jump_label_key rps_needed __read_mostly; + static struct rps_dev_flow * set_rps_cpu(struct net_device *dev, struct sk_buff *skb, struct rps_dev_flow *rflow, u16 next_cpu) @@ -2994,7 +2996,7 @@ int netif_rx(struct sk_buff *skb) trace_netif_rx(skb); #ifdef CONFIG_RPS - { + if (static_branch(&rps_needed)) { struct rps_dev_flow voidflow, *rflow = &voidflow; int cpu; @@ -3009,14 +3011,13 @@ int netif_rx(struct sk_buff *skb) rcu_read_unlock(); preempt_enable(); - } -#else + } else +#endif { unsigned int qtail; ret = enqueue_to_backlog(skb, get_cpu(), &qtail); put_cpu(); } -#endif return ret; } EXPORT_SYMBOL(netif_rx); @@ -3359,7 +3360,7 @@ int netif_receive_skb(struct sk_buff *skb) return NET_RX_SUCCESS; #ifdef CONFIG_RPS - { + if (static_branch(&rps_needed)) { struct rps_dev_flow voidflow, *rflow = &voidflow; int cpu, ret; @@ -3370,16 +3371,12 @@ int netif_receive_skb(struct sk_buff *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; + rcu_read_unlock(); } -#else - return __netif_receive_skb(skb); #endif + return __netif_receive_skb(skb); } EXPORT_SYMBOL(netif_receive_skb); diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 602b1419998c..db6c2f83633f 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -606,9 +606,12 @@ static ssize_t store_rps_map(struct netdev_rx_queue *queue, rcu_assign_pointer(queue->rps_map, map); spin_unlock(&rps_map_lock); - if (old_map) + if (map) + jump_label_inc(&rps_needed); + if (old_map) { kfree_rcu(old_map, rcu); - + jump_label_dec(&rps_needed); + } free_cpumask_var(mask); return len; } diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 77a65f031488..d05559d4d9cd 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -68,8 +68,13 @@ static int rps_sock_flow_sysctl(ctl_table *table, int write, if (sock_table != orig_sock_table) { rcu_assign_pointer(rps_sock_flow_table, sock_table); - synchronize_rcu(); - vfree(orig_sock_table); + if (sock_table) + jump_label_inc(&rps_needed); + if (orig_sock_table) { + jump_label_dec(&rps_needed); + synchronize_rcu(); + vfree(orig_sock_table); + } } } -- cgit v1.2.3 From 927fbec13e40648d3c87cbb1daaac5b1fb9c8775 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 28 Nov 2011 16:33:02 +0000 Subject: xps: Add xps_queue_release function This patch moves the xps specific parts in netdev_queue_release into its own function which netdev_queue_release can call. This allows netdev_queue_release to be more generic (for adding new attributes to tx queues). Signed-off-by: Tom Herbert Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 89 +++++++++++++++++++++++++++------------------------- 1 file changed, 47 insertions(+), 42 deletions(-) (limited to 'net/core/net-sysfs.c') diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index db6c2f83633f..b17c14a0fce9 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -910,6 +910,52 @@ static DEFINE_MUTEX(xps_map_mutex); #define xmap_dereference(P) \ rcu_dereference_protected((P), lockdep_is_held(&xps_map_mutex)) +static void xps_queue_release(struct netdev_queue *queue) +{ + struct net_device *dev = queue->dev; + struct xps_dev_maps *dev_maps; + struct xps_map *map; + unsigned long index; + int i, pos, nonempty = 0; + + index = get_netdev_queue_index(queue); + + mutex_lock(&xps_map_mutex); + dev_maps = xmap_dereference(dev->xps_maps); + + if (dev_maps) { + for_each_possible_cpu(i) { + map = xmap_dereference(dev_maps->cpu_map[i]); + if (!map) + continue; + + for (pos = 0; pos < map->len; pos++) + if (map->queues[pos] == index) + break; + + if (pos < map->len) { + if (map->len > 1) + map->queues[pos] = + map->queues[--map->len]; + else { + RCU_INIT_POINTER(dev_maps->cpu_map[i], + NULL); + kfree_rcu(map, rcu); + map = NULL; + } + } + if (map) + nonempty = 1; + } + + if (!nonempty) { + RCU_INIT_POINTER(dev->xps_maps, NULL); + kfree_rcu(dev_maps, rcu); + } + } + mutex_unlock(&xps_map_mutex); +} + static ssize_t store_xps_map(struct netdev_queue *queue, struct netdev_queue_attribute *attribute, const char *buf, size_t len) @@ -1054,49 +1100,8 @@ static struct attribute *netdev_queue_default_attrs[] = { static void netdev_queue_release(struct kobject *kobj) { struct netdev_queue *queue = to_netdev_queue(kobj); - struct net_device *dev = queue->dev; - struct xps_dev_maps *dev_maps; - struct xps_map *map; - unsigned long index; - int i, pos, nonempty = 0; - - index = get_netdev_queue_index(queue); - - mutex_lock(&xps_map_mutex); - dev_maps = xmap_dereference(dev->xps_maps); - if (dev_maps) { - for_each_possible_cpu(i) { - map = xmap_dereference(dev_maps->cpu_map[i]); - if (!map) - continue; - - for (pos = 0; pos < map->len; pos++) - if (map->queues[pos] == index) - break; - - if (pos < map->len) { - if (map->len > 1) - map->queues[pos] = - map->queues[--map->len]; - else { - RCU_INIT_POINTER(dev_maps->cpu_map[i], - NULL); - kfree_rcu(map, rcu); - map = NULL; - } - } - if (map) - nonempty = 1; - } - - if (!nonempty) { - RCU_INIT_POINTER(dev->xps_maps, NULL); - kfree_rcu(dev_maps, rcu); - } - } - - mutex_unlock(&xps_map_mutex); + xps_queue_release(queue); memset(kobj, 0, sizeof(*kobj)); dev_put(queue->dev); -- cgit v1.2.3 From 114cf5802165ee93e3ab461c9c505cd94a08b800 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 28 Nov 2011 16:33:09 +0000 Subject: bql: Byte queue limits Networking stack support for byte queue limits, uses dynamic queue limits library. Byte queue limits are maintained per transmit queue, and a dql structure has been added to netdev_queue structure for this purpose. Configuration of bql is in the tx- sysfs directory for the queue under the byte_queue_limits directory. Configuration includes: limit_min, bql minimum limit limit_max, bql maximum limit hold_time, bql slack hold time Also under the directory are: limit, current byte limit inflight, current number of bytes on the queue Signed-off-by: Tom Herbert Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 32 ++++++++++- net/Kconfig | 6 ++ net/core/dev.c | 3 + net/core/net-sysfs.c | 140 +++++++++++++++++++++++++++++++++++++++++++--- 4 files changed, 172 insertions(+), 9 deletions(-) (limited to 'net/core/net-sysfs.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 9b24cc7a54d1..97edb3215a5a 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -43,6 +43,7 @@ #include #include #include +#include #include #include @@ -541,7 +542,6 @@ struct netdev_queue { */ struct net_device *dev; struct Qdisc *qdisc; - unsigned long state; struct Qdisc *qdisc_sleeping; #ifdef CONFIG_SYSFS struct kobject kobj; @@ -564,6 +564,12 @@ struct netdev_queue { * (/sys/class/net/DEV/Q/trans_timeout) */ unsigned long trans_timeout; + + unsigned long state; + +#ifdef CONFIG_BQL + struct dql dql; +#endif } ____cacheline_aligned_in_smp; static inline int netdev_queue_numa_node_read(const struct netdev_queue *q) @@ -1862,6 +1868,15 @@ static inline int netif_xmit_frozen_or_stopped(const struct netdev_queue *dev_qu static inline void netdev_tx_sent_queue(struct netdev_queue *dev_queue, unsigned int bytes) { +#ifdef CONFIG_BQL + dql_queued(&dev_queue->dql, bytes); + if (unlikely(dql_avail(&dev_queue->dql) < 0)) { + set_bit(__QUEUE_STATE_STACK_XOFF, &dev_queue->state); + if (unlikely(dql_avail(&dev_queue->dql) >= 0)) + clear_bit(__QUEUE_STATE_STACK_XOFF, + &dev_queue->state); + } +#endif } static inline void netdev_sent_queue(struct net_device *dev, unsigned int bytes) @@ -1872,6 +1887,18 @@ static inline void netdev_sent_queue(struct net_device *dev, unsigned int bytes) static inline void netdev_tx_completed_queue(struct netdev_queue *dev_queue, unsigned pkts, unsigned bytes) { +#ifdef CONFIG_BQL + if (likely(bytes)) { + dql_completed(&dev_queue->dql, bytes); + if (unlikely(test_bit(__QUEUE_STATE_STACK_XOFF, + &dev_queue->state) && + dql_avail(&dev_queue->dql) >= 0)) { + if (test_and_clear_bit(__QUEUE_STATE_STACK_XOFF, + &dev_queue->state)) + netif_schedule_queue(dev_queue); + } + } +#endif } static inline void netdev_completed_queue(struct net_device *dev, @@ -1882,6 +1909,9 @@ static inline void netdev_completed_queue(struct net_device *dev, static inline void netdev_tx_reset_queue(struct netdev_queue *q) { +#ifdef CONFIG_BQL + dql_reset(&q->dql); +#endif } static inline void netdev_reset_queue(struct net_device *dev_queue) diff --git a/net/Kconfig b/net/Kconfig index 63d2c5dc36ff..2d998735c4d8 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -239,6 +239,12 @@ config NETPRIO_CGROUP Cgroup subsystem for use in assigning processes to network priorities on a per-interface basis +config BQL + boolean + depends on SYSFS + select DQL + default y + config HAVE_BPF_JIT bool diff --git a/net/core/dev.c b/net/core/dev.c index cb8f753b4238..91a599122074 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5470,6 +5470,9 @@ static void netdev_init_one_queue(struct net_device *dev, queue->xmit_lock_owner = -1; netdev_queue_numa_node_write(queue, NUMA_NO_NODE); queue->dev = dev; +#ifdef CONFIG_BQL + dql_init(&queue->dql, HZ); +#endif } static int netif_alloc_netdev_queues(struct net_device *dev) diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index b17c14a0fce9..3bf72b638d34 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include "net-sysfs.h" @@ -845,6 +846,116 @@ static ssize_t show_trans_timeout(struct netdev_queue *queue, static struct netdev_queue_attribute queue_trans_timeout = __ATTR(tx_timeout, S_IRUGO, show_trans_timeout, NULL); +#ifdef CONFIG_BQL +/* + * Byte queue limits sysfs structures and functions. + */ +static ssize_t bql_show(char *buf, unsigned int value) +{ + return sprintf(buf, "%u\n", value); +} + +static ssize_t bql_set(const char *buf, const size_t count, + unsigned int *pvalue) +{ + unsigned int value; + int err; + + if (!strcmp(buf, "max") || !strcmp(buf, "max\n")) + value = DQL_MAX_LIMIT; + else { + err = kstrtouint(buf, 10, &value); + if (err < 0) + return err; + if (value > DQL_MAX_LIMIT) + return -EINVAL; + } + + *pvalue = value; + + return count; +} + +static ssize_t bql_show_hold_time(struct netdev_queue *queue, + struct netdev_queue_attribute *attr, + char *buf) +{ + struct dql *dql = &queue->dql; + + return sprintf(buf, "%u\n", jiffies_to_msecs(dql->slack_hold_time)); +} + +static ssize_t bql_set_hold_time(struct netdev_queue *queue, + struct netdev_queue_attribute *attribute, + const char *buf, size_t len) +{ + struct dql *dql = &queue->dql; + unsigned value; + int err; + + err = kstrtouint(buf, 10, &value); + if (err < 0) + return err; + + dql->slack_hold_time = msecs_to_jiffies(value); + + return len; +} + +static struct netdev_queue_attribute bql_hold_time_attribute = + __ATTR(hold_time, S_IRUGO | S_IWUSR, bql_show_hold_time, + bql_set_hold_time); + +static ssize_t bql_show_inflight(struct netdev_queue *queue, + struct netdev_queue_attribute *attr, + char *buf) +{ + struct dql *dql = &queue->dql; + + return sprintf(buf, "%u\n", dql->num_queued - dql->num_completed); +} + +static struct netdev_queue_attribute bql_inflight_attribute = + __ATTR(inflight, S_IRUGO | S_IWUSR, bql_show_inflight, NULL); + +#define BQL_ATTR(NAME, FIELD) \ +static ssize_t bql_show_ ## NAME(struct netdev_queue *queue, \ + struct netdev_queue_attribute *attr, \ + char *buf) \ +{ \ + return bql_show(buf, queue->dql.FIELD); \ +} \ + \ +static ssize_t bql_set_ ## NAME(struct netdev_queue *queue, \ + struct netdev_queue_attribute *attr, \ + const char *buf, size_t len) \ +{ \ + return bql_set(buf, len, &queue->dql.FIELD); \ +} \ + \ +static struct netdev_queue_attribute bql_ ## NAME ## _attribute = \ + __ATTR(NAME, S_IRUGO | S_IWUSR, bql_show_ ## NAME, \ + bql_set_ ## NAME); + +BQL_ATTR(limit, limit) +BQL_ATTR(limit_max, max_limit) +BQL_ATTR(limit_min, min_limit) + +static struct attribute *dql_attrs[] = { + &bql_limit_attribute.attr, + &bql_limit_max_attribute.attr, + &bql_limit_min_attribute.attr, + &bql_hold_time_attribute.attr, + &bql_inflight_attribute.attr, + NULL +}; + +static struct attribute_group dql_group = { + .name = "byte_queue_limits", + .attrs = dql_attrs, +}; +#endif /* CONFIG_BQL */ + #ifdef CONFIG_XPS static inline unsigned int get_netdev_queue_index(struct netdev_queue *queue) { @@ -1096,17 +1207,17 @@ static struct attribute *netdev_queue_default_attrs[] = { NULL }; -#ifdef CONFIG_XPS static void netdev_queue_release(struct kobject *kobj) { struct netdev_queue *queue = to_netdev_queue(kobj); +#ifdef CONFIG_XPS xps_queue_release(queue); +#endif memset(kobj, 0, sizeof(*kobj)); dev_put(queue->dev); } -#endif /* CONFIG_XPS */ static struct kobj_type netdev_queue_ktype = { .sysfs_ops = &netdev_queue_sysfs_ops, @@ -1125,14 +1236,21 @@ static int netdev_queue_add_kobject(struct net_device *net, int index) kobj->kset = net->queues_kset; error = kobject_init_and_add(kobj, &netdev_queue_ktype, NULL, "tx-%u", index); - if (error) { - kobject_put(kobj); - return error; - } + if (error) + goto exit; + +#ifdef CONFIG_BQL + error = sysfs_create_group(kobj, &dql_group); + if (error) + goto exit; +#endif kobject_uevent(kobj, KOBJ_ADD); dev_hold(queue->dev); + return 0; +exit: + kobject_put(kobj); return error; } #endif /* CONFIG_SYSFS */ @@ -1152,8 +1270,14 @@ netdev_queue_update_kobjects(struct net_device *net, int old_num, int new_num) } } - while (--i >= new_num) - kobject_put(&net->_tx[i].kobj); + while (--i >= new_num) { + struct netdev_queue *queue = net->_tx + i; + +#ifdef CONFIG_BQL + sysfs_remove_group(&queue->kobj, &dql_group); +#endif + kobject_put(&queue->kobj); + } return error; #else -- cgit v1.2.3 From b474ae77609b725098d5a7cc8f69c1c528710d53 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 4 Dec 2011 12:38:00 +0000 Subject: bql: fix CONFIG_XPS=n build netdev_queue_release() should be called even if CONFIG_XPS=n to properly release device reference. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'net/core/net-sysfs.c') diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 3bf72b638d34..9d134636f87c 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -1221,9 +1221,7 @@ static void netdev_queue_release(struct kobject *kobj) static struct kobj_type netdev_queue_ktype = { .sysfs_ops = &netdev_queue_sysfs_ops, -#ifdef CONFIG_XPS .release = netdev_queue_release, -#endif .default_attrs = netdev_queue_default_attrs, }; -- cgit v1.2.3 From a0a129f8b6cff54ab479324a54aefdab5db4f240 Mon Sep 17 00:00:00 2001 From: Xi Wang Date: Thu, 22 Dec 2011 13:35:22 +0000 Subject: rps: fix insufficient bounds checking in store_rps_dev_flow_table_cnt() Setting a large rps_flow_cnt like (1 << 30) on 32-bit platform will cause a kernel oops due to insufficient bounds checking. 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)); Note that the macro RPS_DEV_FLOW_TABLE_SIZE(count) is defined as: ... + (count * sizeof(struct rps_dev_flow)) where sizeof(struct rps_dev_flow) is 8. (1 << 30) * 8 will overflow 32 bits. This patch replaces the magic number (1 << 30) with a symbolic bound. Suggested-by: Eric Dumazet Signed-off-by: Xi Wang Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'net/core/net-sysfs.c') diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index c71c434a4c05..385aefe53648 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -665,11 +665,14 @@ static ssize_t store_rps_dev_flow_table_cnt(struct netdev_rx_queue *queue, if (count) { int i; - if (count > 1<<30) { + if (count > INT_MAX) + return -EINVAL; + count = roundup_pow_of_two(count); + if (count > (ULONG_MAX - sizeof(struct rps_dev_flow_table)) + / sizeof(struct rps_dev_flow)) { /* 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; -- cgit v1.2.3 From 60b778ce519625102d3f72a2071ea72a05e990ce Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 24 Dec 2011 06:56:49 +0000 Subject: rfs: better sizing of dev_flow_table Aim of this patch is to provide full range of rps_flow_cnt on 64bit arches. Theorical limit on number of flows is 2^32 Fix some buggy RPS/RFS macros as well. Signed-off-by: Eric Dumazet CC: Tom Herbert CC: Xi Wang CC: Laurent Chavey Signed-off-by: David S. Miller --- include/linux/netdevice.h | 8 ++++---- net/core/net-sysfs.c | 44 +++++++++++++++++++++++++++----------------- 2 files changed, 31 insertions(+), 21 deletions(-) (limited to 'net/core/net-sysfs.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 603730804da5..a776a675c0e5 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -597,7 +597,7 @@ struct rps_map { struct rcu_head rcu; u16 cpus[0]; }; -#define RPS_MAP_SIZE(_num) (sizeof(struct rps_map) + (_num * sizeof(u16))) +#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, the @@ -621,7 +621,7 @@ struct rps_dev_flow_table { 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))) + ((_num) * sizeof(struct rps_dev_flow))) /* * The rps_sock_flow_table contains mappings of flows to the last CPU @@ -632,7 +632,7 @@ struct rps_sock_flow_table { u16 ents[0]; }; #define RPS_SOCK_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_sock_flow_table) + \ - (_num * sizeof(u16))) + ((_num) * sizeof(u16))) #define RPS_NO_CPU 0xffff @@ -684,7 +684,7 @@ struct xps_map { struct rcu_head rcu; u16 queues[0]; }; -#define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + (_num * sizeof(u16))) +#define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + ((_num) * sizeof(u16))) #define XPS_MIN_MAP_ALLOC ((L1_CACHE_BYTES - sizeof(struct xps_map)) \ / sizeof(u16)) diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 4b4d0b0a3543..abf4393a77b3 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -622,15 +622,15 @@ static ssize_t show_rps_dev_flow_table_cnt(struct netdev_rx_queue *queue, char *buf) { struct rps_dev_flow_table *flow_table; - unsigned int val = 0; + unsigned long val = 0; rcu_read_lock(); flow_table = rcu_dereference(queue->rps_flow_table); if (flow_table) - val = flow_table->mask + 1; + val = (unsigned long)flow_table->mask + 1; rcu_read_unlock(); - return sprintf(buf, "%u\n", val); + return sprintf(buf, "%lu\n", val); } static void rps_dev_flow_table_release_work(struct work_struct *work) @@ -654,36 +654,46 @@ static 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; + unsigned long mask, count; struct rps_dev_flow_table *table, *old_table; static DEFINE_SPINLOCK(rps_dev_flow_lock); + int rc; if (!capable(CAP_NET_ADMIN)) return -EPERM; - count = simple_strtoul(buf, &endp, 0); - if (endp == buf) - return -EINVAL; + rc = kstrtoul(buf, 0, &count); + if (rc < 0) + return rc; if (count) { - int i; - - if (count > INT_MAX) + mask = count - 1; + /* mask = roundup_pow_of_two(count) - 1; + * without overflows... + */ + while ((mask | (mask >> 1)) != mask) + mask |= (mask >> 1); + /* On 64 bit arches, must check mask fits in table->mask (u32), + * and on 32bit arches, must check RPS_DEV_FLOW_TABLE_SIZE(mask + 1) + * doesnt overflow. + */ +#if BITS_PER_LONG > 32 + if (mask > (unsigned long)(u32)mask) return -EINVAL; - count = roundup_pow_of_two(count); - if (count > (ULONG_MAX - sizeof(struct rps_dev_flow_table)) +#else + if (mask > (ULONG_MAX - RPS_DEV_FLOW_TABLE_SIZE(1)) / sizeof(struct rps_dev_flow)) { /* Enforce a limit to prevent overflow */ return -EINVAL; } - table = vmalloc(RPS_DEV_FLOW_TABLE_SIZE(count)); +#endif + table = vmalloc(RPS_DEV_FLOW_TABLE_SIZE(mask + 1)); if (!table) return -ENOMEM; - table->mask = count - 1; - for (i = 0; i < count; i++) - table->flows[i].cpu = RPS_NO_CPU; + table->mask = mask; + for (count = 0; count <= mask; count++) + table->flows[count].cpu = RPS_NO_CPU; } else table = NULL; -- cgit v1.2.3 From cf778b00e96df6d64f8e21b8395d1f8a859ecdc7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 12 Jan 2012 04:41:32 +0000 Subject: net: reintroduce missing rcu_assign_pointer() calls commit a9b3cd7f32 (rcu: convert uses of rcu_assign_pointer(x, NULL) to RCU_INIT_POINTER) did a lot of incorrect changes, since it did a complete conversion of rcu_assign_pointer(x, y) to RCU_INIT_POINTER(x, y). We miss needed barriers, even on x86, when y is not NULL. Signed-off-by: Eric Dumazet CC: Stephen Hemminger CC: Paul E. McKenney Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 6 +++--- net/core/netpoll.c | 2 +- net/decnet/dn_dev.c | 4 ++-- net/ipv4/devinet.c | 2 +- net/ipv4/fib_trie.c | 10 +++++----- net/ipv4/igmp.c | 8 ++++---- net/ipv4/ipip.c | 8 ++++---- net/ipv4/ipmr.c | 2 +- net/ipv6/addrconf.c | 2 +- net/ipv6/ip6_tunnel.c | 8 ++++---- net/ipv6/raw.c | 2 +- net/ipv6/sit.c | 10 +++++----- net/mac80211/agg-rx.c | 2 +- net/mac80211/cfg.c | 4 ++-- net/mac80211/ibss.c | 2 +- net/mac80211/sta_info.c | 6 +++--- net/netfilter/nf_conntrack_core.c | 2 +- net/netfilter/nf_conntrack_ecache.c | 4 ++-- net/netfilter/nf_conntrack_extend.c | 2 +- net/netfilter/nf_conntrack_helper.c | 2 +- net/netfilter/nf_conntrack_netlink.c | 2 +- net/netfilter/nf_log.c | 6 +++--- net/netfilter/nf_queue.c | 2 +- net/netfilter/nfnetlink.c | 4 ++-- net/netlabel/netlabel_domainhash.c | 4 ++-- net/netlabel/netlabel_unlabeled.c | 6 ++---- net/phonet/af_phonet.c | 2 +- net/phonet/pn_dev.c | 2 +- net/phonet/socket.c | 2 +- net/socket.c | 2 +- net/sunrpc/auth_gss/auth_gss.c | 2 +- net/xfrm/xfrm_user.c | 2 +- 32 files changed, 61 insertions(+), 63 deletions(-) (limited to 'net/core/net-sysfs.c') diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index abf4393a77b3..f3dbd4f596a4 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -1177,9 +1177,9 @@ static ssize_t store_xps_map(struct netdev_queue *queue, nonempty = 1; } - if (nonempty) - RCU_INIT_POINTER(dev->xps_maps, new_dev_maps); - else { + if (nonempty) { + rcu_assign_pointer(dev->xps_maps, new_dev_maps); + } else { kfree(new_dev_maps); RCU_INIT_POINTER(dev->xps_maps, NULL); } diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 0d38808a2305..556b08298669 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -765,7 +765,7 @@ int __netpoll_setup(struct netpoll *np) } /* last thing to do is link it to the net device structure */ - RCU_INIT_POINTER(ndev->npinfo, npinfo); + rcu_assign_pointer(ndev->npinfo, npinfo); return 0; diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index 2ab16e12520c..74d321a60e7b 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -388,7 +388,7 @@ static int dn_dev_insert_ifa(struct dn_dev *dn_db, struct dn_ifaddr *ifa) } ifa->ifa_next = dn_db->ifa_list; - RCU_INIT_POINTER(dn_db->ifa_list, ifa); + rcu_assign_pointer(dn_db->ifa_list, ifa); dn_ifaddr_notify(RTM_NEWADDR, ifa); blocking_notifier_call_chain(&dnaddr_chain, NETDEV_UP, ifa); @@ -1093,7 +1093,7 @@ static struct dn_dev *dn_dev_create(struct net_device *dev, int *err) memcpy(&dn_db->parms, p, sizeof(struct dn_dev_parms)); - RCU_INIT_POINTER(dev->dn_ptr, dn_db); + rcu_assign_pointer(dev->dn_ptr, dn_db); dn_db->dev = dev; init_timer(&dn_db->timer); diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 65f01dc47565..e41c40f48cfe 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -258,7 +258,7 @@ static struct in_device *inetdev_init(struct net_device *dev) ip_mc_up(in_dev); /* we can receive as soon as ip_ptr is set -- do this last */ - RCU_INIT_POINTER(dev->ip_ptr, in_dev); + rcu_assign_pointer(dev->ip_ptr, in_dev); out: return in_dev; out_kfree: diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index d04b13ae18fe..2b555a5521e0 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -205,7 +205,7 @@ static inline struct tnode *node_parent_rcu(const struct rt_trie_node *node) return (struct tnode *)(parent & ~NODE_TYPE_MASK); } -/* Same as RCU_INIT_POINTER +/* Same as rcu_assign_pointer * but that macro() assumes that value is a pointer. */ static inline void node_set_parent(struct rt_trie_node *node, struct tnode *ptr) @@ -529,7 +529,7 @@ static void tnode_put_child_reorg(struct tnode *tn, int i, struct rt_trie_node * if (n) node_set_parent(n, tn); - RCU_INIT_POINTER(tn->child[i], n); + rcu_assign_pointer(tn->child[i], n); } #define MAX_WORK 10 @@ -1015,7 +1015,7 @@ static void trie_rebalance(struct trie *t, struct tnode *tn) tp = node_parent((struct rt_trie_node *) tn); if (!tp) - RCU_INIT_POINTER(t->trie, (struct rt_trie_node *)tn); + rcu_assign_pointer(t->trie, (struct rt_trie_node *)tn); tnode_free_flush(); if (!tp) @@ -1027,7 +1027,7 @@ static void trie_rebalance(struct trie *t, struct tnode *tn) if (IS_TNODE(tn)) tn = (struct tnode *)resize(t, (struct tnode *)tn); - RCU_INIT_POINTER(t->trie, (struct rt_trie_node *)tn); + rcu_assign_pointer(t->trie, (struct rt_trie_node *)tn); tnode_free_flush(); } @@ -1164,7 +1164,7 @@ static struct list_head *fib_insert_node(struct trie *t, u32 key, int plen) put_child(t, (struct tnode *)tp, cindex, (struct rt_trie_node *)tn); } else { - RCU_INIT_POINTER(t->trie, (struct rt_trie_node *)tn); + rcu_assign_pointer(t->trie, (struct rt_trie_node *)tn); tp = tn; } } diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 5104bc0bbdbe..450e5d21ed2a 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1249,7 +1249,7 @@ void ip_mc_inc_group(struct in_device *in_dev, __be32 addr) im->next_rcu = in_dev->mc_list; in_dev->mc_count++; - RCU_INIT_POINTER(in_dev->mc_list, im); + rcu_assign_pointer(in_dev->mc_list, im); #ifdef CONFIG_IP_MULTICAST igmpv3_del_delrec(in_dev, im->multiaddr); @@ -1821,7 +1821,7 @@ int ip_mc_join_group(struct sock *sk , struct ip_mreqn *imr) iml->next_rcu = inet->mc_list; iml->sflist = NULL; iml->sfmode = MCAST_EXCLUDE; - RCU_INIT_POINTER(inet->mc_list, iml); + rcu_assign_pointer(inet->mc_list, iml); ip_mc_inc_group(in_dev, addr); err = 0; done: @@ -2008,7 +2008,7 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct atomic_sub(IP_SFLSIZE(psl->sl_max), &sk->sk_omem_alloc); kfree_rcu(psl, rcu); } - RCU_INIT_POINTER(pmc->sflist, newpsl); + rcu_assign_pointer(pmc->sflist, newpsl); psl = newpsl; } rv = 1; /* > 0 for insert logic below if sl_count is 0 */ @@ -2111,7 +2111,7 @@ int ip_mc_msfilter(struct sock *sk, struct ip_msfilter *msf, int ifindex) } else (void) ip_mc_del_src(in_dev, &msf->imsf_multiaddr, pmc->sfmode, 0, NULL, 0); - RCU_INIT_POINTER(pmc->sflist, newpsl); + rcu_assign_pointer(pmc->sflist, newpsl); pmc->sfmode = msf->imsf_fmode; err = 0; done: diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 413ed1ba7a5a..22a199315309 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -231,7 +231,7 @@ static void ipip_tunnel_unlink(struct ipip_net *ipn, struct ip_tunnel *t) (iter = rtnl_dereference(*tp)) != NULL; tp = &iter->next) { if (t == iter) { - RCU_INIT_POINTER(*tp, t->next); + rcu_assign_pointer(*tp, t->next); break; } } @@ -241,8 +241,8 @@ static void ipip_tunnel_link(struct ipip_net *ipn, struct ip_tunnel *t) { struct ip_tunnel __rcu **tp = ipip_bucket(ipn, t); - RCU_INIT_POINTER(t->next, rtnl_dereference(*tp)); - RCU_INIT_POINTER(*tp, t); + rcu_assign_pointer(t->next, rtnl_dereference(*tp)); + rcu_assign_pointer(*tp, t); } static struct ip_tunnel * ipip_tunnel_locate(struct net *net, @@ -792,7 +792,7 @@ static int __net_init ipip_fb_tunnel_init(struct net_device *dev) return -ENOMEM; dev_hold(dev); - RCU_INIT_POINTER(ipn->tunnels_wc[0], tunnel); + rcu_assign_pointer(ipn->tunnels_wc[0], tunnel); return 0; } diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 8e54490ee3f4..7bc2db6db8d4 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1225,7 +1225,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi ret = ip_ra_control(sk, 1, mrtsock_destruct); if (ret == 0) { - RCU_INIT_POINTER(mrt->mroute_sk, sk); + rcu_assign_pointer(mrt->mroute_sk, sk); IPV4_DEVCONF_ALL(net, MC_FORWARDING)++; } rtnl_unlock(); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 0ba0866230c9..a225d5ee3c2f 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -429,7 +429,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev) ndev->tstamp = jiffies; addrconf_sysctl_register(ndev); /* protected by rtnl_lock */ - RCU_INIT_POINTER(dev->ip6_ptr, ndev); + rcu_assign_pointer(dev->ip6_ptr, ndev); /* Join all-node multicast group */ ipv6_dev_mc_inc(dev, &in6addr_linklocal_allnodes); diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index e1f7761815f3..aa21da6a09cd 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -218,8 +218,8 @@ ip6_tnl_link(struct ip6_tnl_net *ip6n, struct ip6_tnl *t) { struct ip6_tnl __rcu **tp = ip6_tnl_bucket(ip6n, &t->parms); - RCU_INIT_POINTER(t->next , rtnl_dereference(*tp)); - RCU_INIT_POINTER(*tp, t); + rcu_assign_pointer(t->next , rtnl_dereference(*tp)); + rcu_assign_pointer(*tp, t); } /** @@ -237,7 +237,7 @@ ip6_tnl_unlink(struct ip6_tnl_net *ip6n, struct ip6_tnl *t) (iter = rtnl_dereference(*tp)) != NULL; tp = &iter->next) { if (t == iter) { - RCU_INIT_POINTER(*tp, t->next); + rcu_assign_pointer(*tp, t->next); break; } } @@ -1450,7 +1450,7 @@ static int __net_init ip6_fb_tnl_dev_init(struct net_device *dev) t->parms.proto = IPPROTO_IPV6; dev_hold(dev); - RCU_INIT_POINTER(ip6n->tnls_wc[0], t); + rcu_assign_pointer(ip6n->tnls_wc[0], t); return 0; } diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index a4894f4f1944..d02f7e4dd611 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -131,7 +131,7 @@ static mh_filter_t __rcu *mh_filter __read_mostly; int rawv6_mh_filter_register(mh_filter_t filter) { - RCU_INIT_POINTER(mh_filter, filter); + rcu_assign_pointer(mh_filter, filter); return 0; } EXPORT_SYMBOL(rawv6_mh_filter_register); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 3b6dac956bb0..133768e52912 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -182,7 +182,7 @@ static void ipip6_tunnel_unlink(struct sit_net *sitn, struct ip_tunnel *t) (iter = rtnl_dereference(*tp)) != NULL; tp = &iter->next) { if (t == iter) { - RCU_INIT_POINTER(*tp, t->next); + rcu_assign_pointer(*tp, t->next); break; } } @@ -192,8 +192,8 @@ static void ipip6_tunnel_link(struct sit_net *sitn, struct ip_tunnel *t) { struct ip_tunnel __rcu **tp = ipip6_bucket(sitn, t); - RCU_INIT_POINTER(t->next, rtnl_dereference(*tp)); - RCU_INIT_POINTER(*tp, t); + rcu_assign_pointer(t->next, rtnl_dereference(*tp)); + rcu_assign_pointer(*tp, t); } static void ipip6_tunnel_clone_6rd(struct net_device *dev, struct sit_net *sitn) @@ -393,7 +393,7 @@ ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg) p->addr = a->addr; p->flags = a->flags; t->prl_count++; - RCU_INIT_POINTER(t->prl, p); + rcu_assign_pointer(t->prl, p); out: return err; } @@ -1177,7 +1177,7 @@ static int __net_init ipip6_fb_tunnel_init(struct net_device *dev) if (!dev->tstats) return -ENOMEM; dev_hold(dev); - RCU_INIT_POINTER(sitn->tunnels_wc[0], tunnel); + rcu_assign_pointer(sitn->tunnels_wc[0], tunnel); return 0; } diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c index 96debba2c407..1068f668ac4e 100644 --- a/net/mac80211/agg-rx.c +++ b/net/mac80211/agg-rx.c @@ -332,7 +332,7 @@ void ieee80211_process_addba_request(struct ieee80211_local *local, status = WLAN_STATUS_SUCCESS; /* activate it for RX */ - RCU_INIT_POINTER(sta->ampdu_mlme.tid_rx[tid], tid_agg_rx); + rcu_assign_pointer(sta->ampdu_mlme.tid_rx[tid], tid_agg_rx); if (timeout) mod_timer(&tid_agg_rx->session_timer, TU_TO_EXP_TIME(timeout)); diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 850bb96bd680..e60df48fa4d4 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -616,7 +616,7 @@ static int ieee80211_config_beacon(struct ieee80211_sub_if_data *sdata, sdata->vif.bss_conf.dtim_period = new->dtim_period; - RCU_INIT_POINTER(sdata->u.ap.beacon, new); + rcu_assign_pointer(sdata->u.ap.beacon, new); synchronize_rcu(); @@ -1033,7 +1033,7 @@ static int ieee80211_change_station(struct wiphy *wiphy, return -EBUSY; } - RCU_INIT_POINTER(vlansdata->u.vlan.sta, sta); + rcu_assign_pointer(vlansdata->u.vlan.sta, sta); } sta->sdata = vlansdata; diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c index f8a32bf98216..b3d76b756cd5 100644 --- a/net/mac80211/ibss.c +++ b/net/mac80211/ibss.c @@ -207,7 +207,7 @@ static void __ieee80211_sta_join_ibss(struct ieee80211_sub_if_data *sdata, *pos++ = 0; /* U-APSD no in use */ } - RCU_INIT_POINTER(ifibss->presp, skb); + rcu_assign_pointer(ifibss->presp, skb); sdata->vif.bss_conf.beacon_int = beacon_int; sdata->vif.bss_conf.basic_rates = basic_rates; diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index b197136aea2c..3c428d4839c7 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -73,7 +73,7 @@ static int sta_info_hash_del(struct ieee80211_local *local, if (!s) return -ENOENT; if (s == sta) { - RCU_INIT_POINTER(local->sta_hash[STA_HASH(sta->sta.addr)], + rcu_assign_pointer(local->sta_hash[STA_HASH(sta->sta.addr)], s->hnext); return 0; } @@ -83,7 +83,7 @@ static int sta_info_hash_del(struct ieee80211_local *local, s = rcu_dereference_protected(s->hnext, lockdep_is_held(&local->sta_mtx)); if (rcu_access_pointer(s->hnext)) { - RCU_INIT_POINTER(s->hnext, sta->hnext); + rcu_assign_pointer(s->hnext, sta->hnext); return 0; } @@ -226,7 +226,7 @@ static void sta_info_hash_add(struct ieee80211_local *local, { lockdep_assert_held(&local->sta_mtx); sta->hnext = local->sta_hash[STA_HASH(sta->sta.addr)]; - RCU_INIT_POINTER(local->sta_hash[STA_HASH(sta->sta.addr)], sta); + rcu_assign_pointer(local->sta_hash[STA_HASH(sta->sta.addr)], sta); } static void sta_unblock(struct work_struct *wk) diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index e875f8902db3..76613f5a55c0 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -777,7 +777,7 @@ init_conntrack(struct net *net, struct nf_conn *tmpl, if (exp->helper) { help = nf_ct_helper_ext_add(ct, GFP_ATOMIC); if (help) - RCU_INIT_POINTER(help->helper, exp->helper); + rcu_assign_pointer(help->helper, exp->helper); } #ifdef CONFIG_NF_CONNTRACK_MARK diff --git a/net/netfilter/nf_conntrack_ecache.c b/net/netfilter/nf_conntrack_ecache.c index b62c4148b921..14af6329bdda 100644 --- a/net/netfilter/nf_conntrack_ecache.c +++ b/net/netfilter/nf_conntrack_ecache.c @@ -91,7 +91,7 @@ int nf_conntrack_register_notifier(struct net *net, ret = -EBUSY; goto out_unlock; } - RCU_INIT_POINTER(net->ct.nf_conntrack_event_cb, new); + rcu_assign_pointer(net->ct.nf_conntrack_event_cb, new); mutex_unlock(&nf_ct_ecache_mutex); return ret; @@ -128,7 +128,7 @@ int nf_ct_expect_register_notifier(struct net *net, ret = -EBUSY; goto out_unlock; } - RCU_INIT_POINTER(net->ct.nf_expect_event_cb, new); + rcu_assign_pointer(net->ct.nf_expect_event_cb, new); mutex_unlock(&nf_ct_ecache_mutex); return ret; diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c index 4605c947dcc4..641ff5f96718 100644 --- a/net/netfilter/nf_conntrack_extend.c +++ b/net/netfilter/nf_conntrack_extend.c @@ -169,7 +169,7 @@ int nf_ct_extend_register(struct nf_ct_ext_type *type) before updating alloc_size */ type->alloc_size = ALIGN(sizeof(struct nf_ct_ext), type->align) + type->len; - RCU_INIT_POINTER(nf_ct_ext_types[type->id], type); + rcu_assign_pointer(nf_ct_ext_types[type->id], type); update_alloc_size(type); out: mutex_unlock(&nf_ct_ext_type_mutex); diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index c9e0de08aa87..299fec91f741 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -157,7 +157,7 @@ int __nf_ct_try_assign_helper(struct nf_conn *ct, struct nf_conn *tmpl, memset(&help->help, 0, sizeof(help->help)); } - RCU_INIT_POINTER(help->helper, helper); + rcu_assign_pointer(help->helper, helper); out: return ret; } diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index e07dc3ae930e..2a4834b83332 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1172,7 +1172,7 @@ ctnetlink_change_helper(struct nf_conn *ct, const struct nlattr * const cda[]) return -EOPNOTSUPP; } - RCU_INIT_POINTER(help->helper, helper); + rcu_assign_pointer(help->helper, helper); return 0; } diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index ce0c406f58a8..957374a234d4 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -55,7 +55,7 @@ int nf_log_register(u_int8_t pf, struct nf_logger *logger) llog = rcu_dereference_protected(nf_loggers[pf], lockdep_is_held(&nf_log_mutex)); if (llog == NULL) - RCU_INIT_POINTER(nf_loggers[pf], logger); + rcu_assign_pointer(nf_loggers[pf], logger); } mutex_unlock(&nf_log_mutex); @@ -92,7 +92,7 @@ int nf_log_bind_pf(u_int8_t pf, const struct nf_logger *logger) mutex_unlock(&nf_log_mutex); return -ENOENT; } - RCU_INIT_POINTER(nf_loggers[pf], logger); + rcu_assign_pointer(nf_loggers[pf], logger); mutex_unlock(&nf_log_mutex); return 0; } @@ -250,7 +250,7 @@ static int nf_log_proc_dostring(ctl_table *table, int write, mutex_unlock(&nf_log_mutex); return -ENOENT; } - RCU_INIT_POINTER(nf_loggers[tindex], logger); + rcu_assign_pointer(nf_loggers[tindex], logger); mutex_unlock(&nf_log_mutex); } else { mutex_lock(&nf_log_mutex); diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index 99ffd2885088..b3a7db678b8d 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -40,7 +40,7 @@ int nf_register_queue_handler(u_int8_t pf, const struct nf_queue_handler *qh) else if (old) ret = -EBUSY; else { - RCU_INIT_POINTER(queue_handler[pf], qh); + rcu_assign_pointer(queue_handler[pf], qh); ret = 0; } mutex_unlock(&queue_handler_mutex); diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index c879c1a2370e..b4f8d849480c 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -59,7 +59,7 @@ int nfnetlink_subsys_register(const struct nfnetlink_subsystem *n) nfnl_unlock(); return -EBUSY; } - RCU_INIT_POINTER(subsys_table[n->subsys_id], n); + rcu_assign_pointer(subsys_table[n->subsys_id], n); nfnl_unlock(); return 0; @@ -210,7 +210,7 @@ static int __net_init nfnetlink_net_init(struct net *net) if (!nfnl) return -ENOMEM; net->nfnl_stash = nfnl; - RCU_INIT_POINTER(net->nfnl, nfnl); + rcu_assign_pointer(net->nfnl, nfnl); return 0; } diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c index 38204112b9f4..d8d424337550 100644 --- a/net/netlabel/netlabel_domainhash.c +++ b/net/netlabel/netlabel_domainhash.c @@ -282,7 +282,7 @@ int __init netlbl_domhsh_init(u32 size) INIT_LIST_HEAD(&hsh_tbl->tbl[iter]); spin_lock(&netlbl_domhsh_lock); - RCU_INIT_POINTER(netlbl_domhsh, hsh_tbl); + rcu_assign_pointer(netlbl_domhsh, hsh_tbl); spin_unlock(&netlbl_domhsh_lock); return 0; @@ -330,7 +330,7 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry, &rcu_dereference(netlbl_domhsh)->tbl[bkt]); } else { INIT_LIST_HEAD(&entry->list); - RCU_INIT_POINTER(netlbl_domhsh_def, entry); + rcu_assign_pointer(netlbl_domhsh_def, entry); } if (entry->type == NETLBL_NLTYPE_ADDRSELECT) { diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c index 4b5fa0fe78fd..e7ff694f1049 100644 --- a/net/netlabel/netlabel_unlabeled.c +++ b/net/netlabel/netlabel_unlabeled.c @@ -354,7 +354,7 @@ static struct netlbl_unlhsh_iface *netlbl_unlhsh_add_iface(int ifindex) INIT_LIST_HEAD(&iface->list); if (netlbl_unlhsh_rcu_deref(netlbl_unlhsh_def) != NULL) goto add_iface_failure; - RCU_INIT_POINTER(netlbl_unlhsh_def, iface); + rcu_assign_pointer(netlbl_unlhsh_def, iface); } spin_unlock(&netlbl_unlhsh_lock); @@ -1447,11 +1447,9 @@ int __init netlbl_unlabel_init(u32 size) for (iter = 0; iter < hsh_tbl->size; iter++) INIT_LIST_HEAD(&hsh_tbl->tbl[iter]); - rcu_read_lock(); spin_lock(&netlbl_unlhsh_lock); - RCU_INIT_POINTER(netlbl_unlhsh, hsh_tbl); + rcu_assign_pointer(netlbl_unlhsh, hsh_tbl); spin_unlock(&netlbl_unlhsh_lock); - rcu_read_unlock(); register_netdevice_notifier(&netlbl_unlhsh_netdev_notifier); diff --git a/net/phonet/af_phonet.c b/net/phonet/af_phonet.c index bf10ea8fbbf9..d65f699fbf34 100644 --- a/net/phonet/af_phonet.c +++ b/net/phonet/af_phonet.c @@ -480,7 +480,7 @@ int __init_or_module phonet_proto_register(unsigned int protocol, if (proto_tab[protocol]) err = -EBUSY; else - RCU_INIT_POINTER(proto_tab[protocol], pp); + rcu_assign_pointer(proto_tab[protocol], pp); mutex_unlock(&proto_tab_lock); return err; diff --git a/net/phonet/pn_dev.c b/net/phonet/pn_dev.c index c5827614376b..9b9a85ecc4c7 100644 --- a/net/phonet/pn_dev.c +++ b/net/phonet/pn_dev.c @@ -390,7 +390,7 @@ int phonet_route_add(struct net_device *dev, u8 daddr) daddr = daddr >> 2; mutex_lock(&routes->lock); if (routes->table[daddr] == NULL) { - RCU_INIT_POINTER(routes->table[daddr], dev); + rcu_assign_pointer(routes->table[daddr], dev); dev_hold(dev); err = 0; } diff --git a/net/phonet/socket.c b/net/phonet/socket.c index 3f8d0b1603b9..4c7eff30dfa9 100644 --- a/net/phonet/socket.c +++ b/net/phonet/socket.c @@ -680,7 +680,7 @@ int pn_sock_bind_res(struct sock *sk, u8 res) mutex_lock(&resource_mutex); if (pnres.sk[res] == NULL) { sock_hold(sk); - RCU_INIT_POINTER(pnres.sk[res], sk); + rcu_assign_pointer(pnres.sk[res], sk); ret = 0; } mutex_unlock(&resource_mutex); diff --git a/net/socket.c b/net/socket.c index e56162cd65b0..28a96af484b4 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2492,7 +2492,7 @@ int sock_register(const struct net_proto_family *ops) lockdep_is_held(&net_family_lock))) err = -EEXIST; else { - RCU_INIT_POINTER(net_families[ops->family], ops); + rcu_assign_pointer(net_families[ops->family], ops); err = 0; } spin_unlock(&net_family_lock); diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index afb56553dfe7..db0efdec4e45 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -122,7 +122,7 @@ gss_cred_set_ctx(struct rpc_cred *cred, struct gss_cl_ctx *ctx) if (!test_bit(RPCAUTH_CRED_NEW, &cred->cr_flags)) return; gss_get_ctx(ctx); - RCU_INIT_POINTER(gss_cred->gc_ctx, ctx); + rcu_assign_pointer(gss_cred->gc_ctx, ctx); set_bit(RPCAUTH_CRED_UPTODATE, &cred->cr_flags); smp_mb__before_clear_bit(); clear_bit(RPCAUTH_CRED_NEW, &cred->cr_flags); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index e0d747a2e803..637f11a1e4df 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2927,7 +2927,7 @@ static int __net_init xfrm_user_net_init(struct net *net) if (nlsk == NULL) return -ENOMEM; net->xfrm.nlsk_stash = nlsk; /* Don't set to NULL */ - RCU_INIT_POINTER(net->xfrm.nlsk, nlsk); + rcu_assign_pointer(net->xfrm.nlsk, nlsk); return 0; } -- cgit v1.2.3 From 795d9a2538b205d9329f34798ec96503a07b7919 Mon Sep 17 00:00:00 2001 From: Hiroaki SHIMODA Date: Sat, 14 Jan 2012 07:10:21 +0000 Subject: bql: Fix inconsistency between file mode and attr method. There is no store() method for inflight attribute in the tx-/byte_queue_limits sysfs directory. So remove S_IWUSR bit. Signed-off-by: Hiroaki SHIMODA Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/core/net-sysfs.c') diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index f3dbd4f596a4..a1727cda03d7 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -929,7 +929,7 @@ static ssize_t bql_show_inflight(struct netdev_queue *queue, } static struct netdev_queue_attribute bql_inflight_attribute = - __ATTR(inflight, S_IRUGO | S_IWUSR, bql_show_inflight, NULL); + __ATTR(inflight, S_IRUGO, bql_show_inflight, NULL); #define BQL_ATTR(NAME, FIELD) \ static ssize_t bql_show_ ## NAME(struct netdev_queue *queue, \ -- cgit v1.2.3