aboutsummaryrefslogtreecommitdiff
path: root/net/sched/sch_generic.c
diff options
context:
space:
mode:
Diffstat (limited to 'net/sched/sch_generic.c')
-rw-r--r--net/sched/sch_generic.c270
1 files changed, 191 insertions, 79 deletions
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 5173c1e1b19..e1543b03e39 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -24,7 +24,15 @@
#include <linux/init.h>
#include <linux/rcupdate.h>
#include <linux/list.h>
+#include <linux/slab.h>
+#include <linux/if_vlan.h>
+#include <net/sch_generic.h>
#include <net/pkt_sched.h>
+#include <net/dst.h>
+
+/* Qdisc to use by default */
+const struct Qdisc_ops *default_qdisc_ops = &pfifo_fast_ops;
+EXPORT_SYMBOL(default_qdisc_ops);
/* Main transmission queue. */
@@ -39,6 +47,7 @@
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 */
@@ -50,21 +59,19 @@ static inline int dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q)
static inline struct sk_buff *dequeue_skb(struct Qdisc *q)
{
struct sk_buff *skb = q->gso_skb;
+ const struct netdev_queue *txq = q->dev_queue;
if (unlikely(skb)) {
- struct net_device *dev = qdisc_dev(q);
- struct netdev_queue *txq;
-
/* check the reason of requeuing without tx lock first */
- txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
- if (!netif_tx_queue_stopped(txq) &&
- !netif_tx_queue_frozen(txq)) {
+ txq = netdev_get_tx_queue(txq->dev, skb_get_queue_mapping(skb));
+ if (!netif_xmit_frozen_or_stopped(txq)) {
q->gso_skb = NULL;
q->q.qlen--;
} else
skb = NULL;
} else {
- skb = q->dequeue(q);
+ if (!(q->flags & TCQ_F_ONETXQUEUE) || !netif_xmit_frozen_or_stopped(txq))
+ skb = q->dequeue(q);
}
return skb;
@@ -84,16 +91,15 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
* deadloop is detected. Return OK to try the next skb.
*/
kfree_skb(skb);
- if (net_ratelimit())
- printk(KERN_WARNING "Dead loop on netdevice %s, "
- "fix it urgently!\n", dev_queue->dev->name);
+ net_warn_ratelimited("Dead loop on netdevice %s, fix it urgently!\n",
+ dev_queue->dev->name);
ret = qdisc_qlen(q);
} else {
/*
* Another cpu is holding lock, requeue & delay xmits for
* some time.
*/
- __get_cpu_var(netdev_rx_stat).cpu_collision++;
+ __this_cpu_inc(softnet_data.cpu_collision);
ret = dev_requeue_skb(skb, q);
}
@@ -119,7 +125,7 @@ int sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q,
spin_unlock(root_lock);
HARD_TX_LOCK(dev, txq, smp_processor_id());
- if (!netif_tx_queue_stopped(txq) && !netif_tx_queue_frozen(txq))
+ if (!netif_xmit_frozen_or_stopped(txq))
ret = dev_hard_start_xmit(skb, dev, txq);
HARD_TX_UNLOCK(dev, txq);
@@ -134,15 +140,14 @@ int sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q,
ret = handle_dev_cpu_collision(skb, txq, q);
} else {
/* Driver returned NETDEV_TX_BUSY - requeue skb */
- if (unlikely (ret != NETDEV_TX_BUSY && net_ratelimit()))
- printk(KERN_WARNING "BUG %s code %d qlen %d\n",
- dev->name, ret, q->q.qlen);
+ if (unlikely(ret != NETDEV_TX_BUSY))
+ net_warn_ratelimited("BUG %s code %d qlen %d\n",
+ dev->name, ret, q->q.qlen);
ret = dev_requeue_skb(skb, q);
}
- if (ret && (netif_tx_queue_stopped(txq) ||
- netif_tx_queue_frozen(txq)))
+ if (ret && netif_xmit_frozen_or_stopped(txq))
ret = 0;
return ret;
@@ -178,7 +183,7 @@ 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));
@@ -188,34 +193,38 @@ static inline int qdisc_restart(struct Qdisc *q)
void __qdisc_run(struct Qdisc *q)
{
- unsigned long start_time = jiffies;
+ int quota = weight_p;
while (qdisc_restart(q)) {
/*
- * Postpone processing if
- * 1. another process needs the CPU;
- * 2. we've been doing it for too long.
+ * Ordered by possible occurrence: Postpone processing if
+ * 1. we've exceeded packet quota
+ * 2. another process needs the CPU;
*/
- if (need_resched() || jiffies != start_time) {
+ if (--quota <= 0 || need_resched()) {
__netif_schedule(q);
break;
}
}
- clear_bit(__QDISC_STATE_RUNNING, &q->state);
+ qdisc_run_end(q);
}
unsigned long dev_trans_start(struct net_device *dev)
{
- unsigned long val, res = dev->trans_start;
+ unsigned long val, res;
unsigned int i;
+ if (is_vlan_dev(dev))
+ dev = vlan_dev_real_dev(dev);
+ res = dev->trans_start;
for (i = 0; i < dev->num_tx_queues; i++) {
val = netdev_get_tx_queue(dev, i)->trans_start;
if (val && time_after(val, res))
res = val;
}
dev->trans_start = res;
+
return res;
}
EXPORT_SYMBOL(dev_trans_start);
@@ -241,18 +250,18 @@ static void dev_watchdog(unsigned long arg)
* old device drivers set dev->trans_start
*/
trans_start = txq->trans_start ? : dev->trans_start;
- if (netif_tx_queue_stopped(txq) &&
+ if (netif_xmit_stopped(txq) &&
time_after(jiffies, (trans_start +
dev->watchdog_timeo))) {
some_queue_timedout = 1;
+ txq->trans_timeout++;
break;
}
}
if (some_queue_timedout) {
- char drivername[64];
WARN_ONCE(1, KERN_INFO "NETDEV WATCHDOG: %s (%s): transmit queue %u timed out\n",
- dev->name, netdev_drivername(dev, drivername, 64), i);
+ dev->name, netdev_drivername(dev), i);
dev->netdev_ops->ndo_tx_timeout(dev);
}
if (!mod_timer(&dev->watchdog_timer,
@@ -301,6 +310,7 @@ void netif_carrier_on(struct net_device *dev)
if (test_and_clear_bit(__LINK_STATE_NOCARRIER, &dev->state)) {
if (dev->reg_state == NETREG_UNINITIALIZED)
return;
+ atomic_inc(&dev->carrier_changes);
linkwatch_fire_event(dev);
if (netif_running(dev))
__netdev_watchdog_up(dev);
@@ -319,6 +329,7 @@ void netif_carrier_off(struct net_device *dev)
if (!test_and_set_bit(__LINK_STATE_NOCARRIER, &dev->state)) {
if (dev->reg_state == NETREG_UNINITIALIZED)
return;
+ atomic_inc(&dev->carrier_changes);
linkwatch_fire_event(dev);
}
}
@@ -329,13 +340,13 @@ EXPORT_SYMBOL(netif_carrier_off);
cheaper.
*/
-static int noop_enqueue(struct sk_buff *skb, struct Qdisc * qdisc)
+static int noop_enqueue(struct sk_buff *skb, struct Qdisc *qdisc)
{
kfree_skb(skb);
return NET_XMIT_CN;
}
-static struct sk_buff *noop_dequeue(struct Qdisc * qdisc)
+static struct sk_buff *noop_dequeue(struct Qdisc *qdisc)
{
return NULL;
}
@@ -362,6 +373,7 @@ struct Qdisc noop_qdisc = {
.list = LIST_HEAD_INIT(noop_qdisc.list),
.q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock),
.dev_queue = &noop_netdev_queue,
+ .busylock = __SPIN_LOCK_UNLOCKED(noop_qdisc.busylock),
};
EXPORT_SYMBOL(noop_qdisc);
@@ -388,11 +400,13 @@ static struct Qdisc noqueue_qdisc = {
.list = LIST_HEAD_INIT(noqueue_qdisc.list),
.q.lock = __SPIN_LOCK_UNLOCKED(noqueue_qdisc.q.lock),
.dev_queue = &noqueue_netdev_queue,
+ .busylock = __SPIN_LOCK_UNLOCKED(noqueue_qdisc.busylock),
};
-static const u8 prio2band[TC_PRIO_MAX+1] =
- { 1, 2, 2, 2, 1, 2, 0, 0 , 1, 1, 1, 1, 1, 1, 1, 1 };
+static const u8 prio2band[TC_PRIO_MAX + 1] = {
+ 1, 2, 2, 2, 1, 2, 0, 0 , 1, 1, 1, 1, 1, 1, 1, 1
+};
/* 3-band FIFO queue: old style, but should be a bit faster than
generic prio+fifo combination.
@@ -424,7 +438,7 @@ static inline struct sk_buff_head *band2list(struct pfifo_fast_priv *priv,
return priv->q + band;
}
-static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc* qdisc)
+static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc *qdisc)
{
if (skb_queue_len(&qdisc->q) < qdisc_dev(qdisc)->tx_queue_len) {
int band = prio2band[skb->priority & TC_PRIO_MAX];
@@ -439,7 +453,7 @@ static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc* qdisc)
return qdisc_drop(skb, qdisc);
}
-static struct sk_buff *pfifo_fast_dequeue(struct Qdisc* qdisc)
+static struct sk_buff *pfifo_fast_dequeue(struct Qdisc *qdisc)
{
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
int band = bitmap2band[priv->bitmap];
@@ -458,7 +472,7 @@ static struct sk_buff *pfifo_fast_dequeue(struct Qdisc* qdisc)
return NULL;
}
-static struct sk_buff *pfifo_fast_peek(struct Qdisc* qdisc)
+static struct sk_buff *pfifo_fast_peek(struct Qdisc *qdisc)
{
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
int band = bitmap2band[priv->bitmap];
@@ -472,7 +486,7 @@ static struct sk_buff *pfifo_fast_peek(struct Qdisc* qdisc)
return NULL;
}
-static void pfifo_fast_reset(struct Qdisc* qdisc)
+static void pfifo_fast_reset(struct Qdisc *qdisc)
{
int prio;
struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
@@ -489,8 +503,9 @@ static int pfifo_fast_dump(struct Qdisc *qdisc, struct sk_buff *skb)
{
struct tc_prio_qopt opt = { .bands = PFIFO_FAST_BANDS };
- memcpy(&opt.priomap, prio2band, TC_PRIO_MAX+1);
- NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
+ memcpy(&opt.priomap, prio2band, TC_PRIO_MAX + 1);
+ if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt))
+ goto nla_put_failure;
return skb->len;
nla_put_failure:
@@ -505,6 +520,8 @@ static int pfifo_fast_init(struct Qdisc *qdisc, struct nlattr *opt)
for (prio = 0; prio < PFIFO_FAST_BANDS; prio++)
skb_queue_head_init(band2list(priv, prio));
+ /* Can by-pass the queue discipline */
+ qdisc->flags |= TCQ_F_CAN_BYPASS;
return 0;
}
@@ -520,31 +537,45 @@ struct Qdisc_ops pfifo_fast_ops __read_mostly = {
.owner = THIS_MODULE,
};
+static struct lock_class_key qdisc_tx_busylock;
+
struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
- struct Qdisc_ops *ops)
+ const struct Qdisc_ops *ops)
{
void *p;
struct Qdisc *sch;
- unsigned int size;
+ unsigned int size = QDISC_ALIGN(sizeof(*sch)) + ops->priv_size;
int err = -ENOBUFS;
+ struct net_device *dev = dev_queue->dev;
- /* ensure that the Qdisc and the private data are 32-byte aligned */
- size = QDISC_ALIGN(sizeof(*sch));
- size += ops->priv_size + (QDISC_ALIGNTO - 1);
+ p = kzalloc_node(size, GFP_KERNEL,
+ netdev_queue_numa_node_read(dev_queue));
- p = kzalloc(size, GFP_KERNEL);
if (!p)
goto errout;
sch = (struct Qdisc *) QDISC_ALIGN((unsigned long) p);
- sch->padded = (char *) sch - (char *) p;
-
+ /* if we got non aligned memory, ask more and do alignment ourself */
+ if (sch != p) {
+ kfree(p);
+ p = kzalloc_node(size + QDISC_ALIGNTO - 1, GFP_KERNEL,
+ netdev_queue_numa_node_read(dev_queue));
+ if (!p)
+ goto errout;
+ sch = (struct Qdisc *) QDISC_ALIGN((unsigned long) p);
+ sch->padded = (char *) sch - (char *) p;
+ }
INIT_LIST_HEAD(&sch->list);
skb_queue_head_init(&sch->q);
+
+ spin_lock_init(&sch->busylock);
+ lockdep_set_class(&sch->busylock,
+ dev->qdisc_tx_busylock ?: &qdisc_tx_busylock);
+
sch->ops = ops;
sch->enqueue = ops->enqueue;
sch->dequeue = ops->dequeue;
sch->dev_queue = dev_queue;
- dev_hold(qdisc_dev(sch));
+ dev_hold(dev);
atomic_set(&sch->refcnt, 1);
return sch;
@@ -552,13 +583,15 @@ errout:
return ERR_PTR(err);
}
-struct Qdisc * qdisc_create_dflt(struct net_device *dev,
- struct netdev_queue *dev_queue,
- struct Qdisc_ops *ops,
- unsigned int parentid)
+struct Qdisc *qdisc_create_dflt(struct netdev_queue *dev_queue,
+ const struct Qdisc_ops *ops,
+ unsigned int parentid)
{
struct Qdisc *sch;
+ if (!try_module_get(ops->owner))
+ goto errout;
+
sch = qdisc_alloc(dev_queue, ops);
if (IS_ERR(sch))
goto errout;
@@ -590,6 +623,13 @@ void qdisc_reset(struct Qdisc *qdisc)
}
EXPORT_SYMBOL(qdisc_reset);
+static void qdisc_rcu_free(struct rcu_head *head)
+{
+ struct Qdisc *qdisc = container_of(head, struct Qdisc, rcu_head);
+
+ kfree((char *) qdisc - qdisc->padded);
+}
+
void qdisc_destroy(struct Qdisc *qdisc)
{
const struct Qdisc_ops *ops = qdisc->ops;
@@ -601,7 +641,7 @@ void qdisc_destroy(struct Qdisc *qdisc)
#ifdef CONFIG_NET_SCHED
qdisc_list_del(qdisc);
- qdisc_put_stab(qdisc->stab);
+ qdisc_put_stab(rtnl_dereference(qdisc->stab));
#endif
gen_kill_estimator(&qdisc->bstats, &qdisc->rate_est);
if (ops->reset)
@@ -613,7 +653,11 @@ void qdisc_destroy(struct Qdisc *qdisc)
dev_put(qdisc_dev(qdisc));
kfree_skb(qdisc->gso_skb);
- kfree((char *) qdisc - qdisc->padded);
+ /*
+ * gen_estimator est_timer() might access qdisc->q.lock,
+ * wait a RCU grace period before freeing qdisc.
+ */
+ call_rcu(&qdisc->rcu_head, qdisc_rcu_free);
}
EXPORT_SYMBOL(qdisc_destroy);
@@ -641,25 +685,23 @@ struct Qdisc *dev_graft_qdisc(struct netdev_queue *dev_queue,
return oqdisc;
}
+EXPORT_SYMBOL(dev_graft_qdisc);
static void attach_one_default_qdisc(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_unused)
{
- struct Qdisc *qdisc;
+ struct Qdisc *qdisc = &noqueue_qdisc;
if (dev->tx_queue_len) {
- qdisc = qdisc_create_dflt(dev, dev_queue,
- &pfifo_fast_ops, TC_H_ROOT);
+ qdisc = qdisc_create_dflt(dev_queue,
+ default_qdisc_ops, TC_H_ROOT);
if (!qdisc) {
- printk(KERN_INFO "%s: activation failed\n", dev->name);
+ netdev_info(dev, "activation failed\n");
return;
}
-
- /* Can by-pass the queue discipline for default qdisc */
- qdisc->flags |= TCQ_F_CAN_BYPASS;
- } else {
- qdisc = &noqueue_qdisc;
+ if (!netif_is_multiqueue(dev))
+ qdisc->flags |= TCQ_F_ONETXQUEUE;
}
dev_queue->qdisc_sleeping = qdisc;
}
@@ -676,10 +718,10 @@ static void attach_default_qdiscs(struct net_device *dev)
dev->qdisc = txq->qdisc_sleeping;
atomic_inc(&dev->qdisc->refcnt);
} else {
- qdisc = qdisc_create_dflt(dev, txq, &mq_qdisc_ops, TC_H_ROOT);
+ qdisc = qdisc_create_dflt(txq, &mq_qdisc_ops, TC_H_ROOT);
if (qdisc) {
- qdisc->ops->attach(qdisc);
dev->qdisc = qdisc;
+ qdisc->ops->attach(qdisc);
}
}
}
@@ -706,9 +748,8 @@ void dev_activate(struct net_device *dev)
int need_watchdog;
/* No queueing discipline is attached to device;
- create default one i.e. pfifo_fast for devices,
- which need queueing and noqueue_qdisc for
- virtual interfaces
+ * create default one for devices, which need queueing
+ * and noqueue_qdisc for virtual interfaces
*/
if (dev->qdisc == &noop_qdisc)
@@ -720,13 +761,15 @@ void dev_activate(struct net_device *dev)
need_watchdog = 0;
netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog);
- transition_one_qdisc(dev, &dev->rx_queue, NULL);
+ if (dev_ingress_queue(dev))
+ transition_one_qdisc(dev, dev_ingress_queue(dev), NULL);
if (need_watchdog) {
dev->trans_start = jiffies;
dev_watchdog_up(dev);
}
}
+EXPORT_SYMBOL(dev_activate);
static void dev_deactivate_queue(struct net_device *dev,
struct netdev_queue *dev_queue,
@@ -765,7 +808,7 @@ static bool some_qdisc_is_busy(struct net_device *dev)
spin_lock_bh(root_lock);
- val = (test_bit(__QDISC_STATE_RUNNING, &q->state) ||
+ val = (qdisc_is_running(q) ||
test_bit(__QDISC_STATE_SCHED, &q->state));
spin_unlock_bh(root_lock);
@@ -776,21 +819,52 @@ static bool some_qdisc_is_busy(struct net_device *dev)
return false;
}
-void dev_deactivate(struct net_device *dev)
+/**
+ * dev_deactivate_many - deactivate transmissions on several devices
+ * @head: list of devices to deactivate
+ *
+ * This function returns only when all outstanding transmissions
+ * have completed, unless all devices are in dismantle phase.
+ */
+void dev_deactivate_many(struct list_head *head)
{
- netdev_for_each_tx_queue(dev, dev_deactivate_queue, &noop_qdisc);
- dev_deactivate_queue(dev, &dev->rx_queue, &noop_qdisc);
+ struct net_device *dev;
+ bool sync_needed = false;
- dev_watchdog_down(dev);
+ list_for_each_entry(dev, head, close_list) {
+ netdev_for_each_tx_queue(dev, dev_deactivate_queue,
+ &noop_qdisc);
+ if (dev_ingress_queue(dev))
+ dev_deactivate_queue(dev, dev_ingress_queue(dev),
+ &noop_qdisc);
- /* Wait for outstanding qdisc-less dev_queue_xmit calls. */
- synchronize_rcu();
+ dev_watchdog_down(dev);
+ sync_needed |= !dev->dismantle;
+ }
+
+ /* Wait for outstanding qdisc-less dev_queue_xmit calls.
+ * This is avoided if all devices are in dismantle phase :
+ * Caller will call synchronize_net() for us
+ */
+ if (sync_needed)
+ synchronize_net();
/* Wait for outstanding qdisc_run calls. */
- while (some_qdisc_is_busy(dev))
- yield();
+ list_for_each_entry(dev, head, close_list)
+ while (some_qdisc_is_busy(dev))
+ yield();
}
+void dev_deactivate(struct net_device *dev)
+{
+ LIST_HEAD(single);
+
+ list_add(&dev->close_list, &single);
+ dev_deactivate_many(&single);
+ list_del(&single);
+}
+EXPORT_SYMBOL(dev_deactivate);
+
static void dev_init_scheduler_queue(struct net_device *dev,
struct netdev_queue *dev_queue,
void *_qdisc)
@@ -805,7 +879,8 @@ void dev_init_scheduler(struct net_device *dev)
{
dev->qdisc = &noop_qdisc;
netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc);
- dev_init_scheduler_queue(dev, &dev->rx_queue, &noop_qdisc);
+ if (dev_ingress_queue(dev))
+ dev_init_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc);
setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev);
}
@@ -828,9 +903,46 @@ static void shutdown_scheduler_queue(struct net_device *dev,
void dev_shutdown(struct net_device *dev)
{
netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc);
- shutdown_scheduler_queue(dev, &dev->rx_queue, &noop_qdisc);
+ if (dev_ingress_queue(dev))
+ shutdown_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc);
qdisc_destroy(dev->qdisc);
dev->qdisc = &noop_qdisc;
WARN_ON(timer_pending(&dev->watchdog_timer));
}
+
+void psched_ratecfg_precompute(struct psched_ratecfg *r,
+ const struct tc_ratespec *conf,
+ u64 rate64)
+{
+ memset(r, 0, sizeof(*r));
+ r->overhead = conf->overhead;
+ r->rate_bytes_ps = max_t(u64, conf->rate, rate64);
+ r->linklayer = (conf->linklayer & TC_LINKLAYER_MASK);
+ r->mult = 1;
+ /*
+ * The deal here is to replace a divide by a reciprocal one
+ * in fast path (a reciprocal divide is a multiply and a shift)
+ *
+ * Normal formula would be :
+ * time_in_ns = (NSEC_PER_SEC * len) / rate_bps
+ *
+ * We compute mult/shift to use instead :
+ * time_in_ns = (len * mult) >> shift;
+ *
+ * We try to get the highest possible mult value for accuracy,
+ * but have to make sure no overflows will ever happen.
+ */
+ if (r->rate_bytes_ps > 0) {
+ u64 factor = NSEC_PER_SEC;
+
+ for (;;) {
+ r->mult = div64_u64(factor, r->rate_bytes_ps);
+ if (r->mult & (1U << 31) || factor & (1ULL << 63))
+ break;
+ factor <<= 1;
+ r->shift++;
+ }
+ }
+}
+EXPORT_SYMBOL(psched_ratecfg_precompute);
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-09 06:55:56 +0000