#include <linux/kmod.h>
#include <linux/list.h>
#include <linux/hrtimer.h>
+#include <linux/lockdep.h>
+#include <linux/slab.h>
#include <net/net_namespace.h>
#include <net/sock.h>
#include <net/netlink.h>
#include <net/pkt_sched.h>
-static int qdisc_notify(struct sk_buff *oskb, struct nlmsghdr *n, u32 clid,
+static int qdisc_notify(struct net *net, struct sk_buff *oskb,
+ struct nlmsghdr *n, u32 clid,
struct Qdisc *old, struct Qdisc *new);
-static int tclass_notify(struct sk_buff *oskb, struct nlmsghdr *n,
- struct Qdisc *q, unsigned long cl, int event);
+static int tclass_notify(struct net *net, struct sk_buff *oskb,
+ struct nlmsghdr *n, struct Qdisc *q,
+ unsigned long cl, int event);
/*
Auxiliary routines:
- ---requeue
+ ---peek
- requeues once dequeued packet. It is used for non-standard or
- just buggy devices, which can defer output even if netif_queue_stopped()=0.
+ like dequeue but without removing a packet from the queue
---reset
if (qops->enqueue == NULL)
qops->enqueue = noop_qdisc_ops.enqueue;
- if (qops->requeue == NULL)
- qops->requeue = noop_qdisc_ops.requeue;
+ if (qops->peek == NULL) {
+ if (qops->dequeue == NULL) {
+ qops->peek = noop_qdisc_ops.peek;
+ } else {
+ rc = -EINVAL;
+ goto out;
+ }
+ }
if (qops->dequeue == NULL)
qops->dequeue = noop_qdisc_ops.dequeue;
(root qdisc, all its children, children of children etc.)
*/
-struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
+static struct Qdisc *qdisc_match_from_root(struct Qdisc *root, u32 handle)
{
- struct netdev_queue *dev_queue = &dev->tx_queue;
struct Qdisc *q;
- list_for_each_entry(q, &dev_queue->qdisc_list, list) {
+ if (!(root->flags & TCQ_F_BUILTIN) &&
+ root->handle == handle)
+ return root;
+
+ list_for_each_entry(q, &root->list, list) {
if (q->handle == handle)
return q;
}
return NULL;
}
+static void qdisc_list_add(struct Qdisc *q)
+{
+ if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS))
+ list_add_tail(&q->list, &qdisc_dev(q)->qdisc->list);
+}
+
+void qdisc_list_del(struct Qdisc *q)
+{
+ if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS))
+ list_del(&q->list);
+}
+EXPORT_SYMBOL(qdisc_list_del);
+
+struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
+{
+ struct Qdisc *q;
+
+ q = qdisc_match_from_root(dev->qdisc, handle);
+ if (q)
+ goto out;
+
+ q = qdisc_match_from_root(dev->rx_queue.qdisc_sleeping, handle);
+out:
+ return q;
+}
+
static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid)
{
unsigned long cl;
}
EXPORT_SYMBOL(qdisc_put_rtab);
+static LIST_HEAD(qdisc_stab_list);
+static DEFINE_SPINLOCK(qdisc_stab_lock);
+
+static const struct nla_policy stab_policy[TCA_STAB_MAX + 1] = {
+ [TCA_STAB_BASE] = { .len = sizeof(struct tc_sizespec) },
+ [TCA_STAB_DATA] = { .type = NLA_BINARY },
+};
+
+static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt)
+{
+ struct nlattr *tb[TCA_STAB_MAX + 1];
+ struct qdisc_size_table *stab;
+ struct tc_sizespec *s;
+ unsigned int tsize = 0;
+ u16 *tab = NULL;
+ int err;
+
+ err = nla_parse_nested(tb, TCA_STAB_MAX, opt, stab_policy);
+ if (err < 0)
+ return ERR_PTR(err);
+ if (!tb[TCA_STAB_BASE])
+ return ERR_PTR(-EINVAL);
+
+ s = nla_data(tb[TCA_STAB_BASE]);
+
+ if (s->tsize > 0) {
+ if (!tb[TCA_STAB_DATA])
+ return ERR_PTR(-EINVAL);
+ tab = nla_data(tb[TCA_STAB_DATA]);
+ tsize = nla_len(tb[TCA_STAB_DATA]) / sizeof(u16);
+ }
+
+ if (!s || tsize != s->tsize || (!tab && tsize > 0))
+ return ERR_PTR(-EINVAL);
+
+ spin_lock(&qdisc_stab_lock);
+
+ list_for_each_entry(stab, &qdisc_stab_list, list) {
+ if (memcmp(&stab->szopts, s, sizeof(*s)))
+ continue;
+ if (tsize > 0 && memcmp(stab->data, tab, tsize * sizeof(u16)))
+ continue;
+ stab->refcnt++;
+ spin_unlock(&qdisc_stab_lock);
+ return stab;
+ }
+
+ spin_unlock(&qdisc_stab_lock);
+
+ stab = kmalloc(sizeof(*stab) + tsize * sizeof(u16), GFP_KERNEL);
+ if (!stab)
+ return ERR_PTR(-ENOMEM);
+
+ stab->refcnt = 1;
+ stab->szopts = *s;
+ if (tsize > 0)
+ memcpy(stab->data, tab, tsize * sizeof(u16));
+
+ spin_lock(&qdisc_stab_lock);
+ list_add_tail(&stab->list, &qdisc_stab_list);
+ spin_unlock(&qdisc_stab_lock);
+
+ return stab;
+}
+
+void qdisc_put_stab(struct qdisc_size_table *tab)
+{
+ if (!tab)
+ return;
+
+ spin_lock(&qdisc_stab_lock);
+
+ if (--tab->refcnt == 0) {
+ list_del(&tab->list);
+ kfree(tab);
+ }
+
+ spin_unlock(&qdisc_stab_lock);
+}
+EXPORT_SYMBOL(qdisc_put_stab);
+
+static int qdisc_dump_stab(struct sk_buff *skb, struct qdisc_size_table *stab)
+{
+ struct nlattr *nest;
+
+ nest = nla_nest_start(skb, TCA_STAB);
+ if (nest == NULL)
+ goto nla_put_failure;
+ NLA_PUT(skb, TCA_STAB_BASE, sizeof(stab->szopts), &stab->szopts);
+ nla_nest_end(skb, nest);
+
+ return skb->len;
+
+nla_put_failure:
+ return -1;
+}
+
+void qdisc_calculate_pkt_len(struct sk_buff *skb, struct qdisc_size_table *stab)
+{
+ int pkt_len, slot;
+
+ pkt_len = skb->len + stab->szopts.overhead;
+ if (unlikely(!stab->szopts.tsize))
+ goto out;
+
+ slot = pkt_len + stab->szopts.cell_align;
+ if (unlikely(slot < 0))
+ slot = 0;
+
+ slot >>= stab->szopts.cell_log;
+ if (likely(slot < stab->szopts.tsize))
+ pkt_len = stab->data[slot];
+ else
+ pkt_len = stab->data[stab->szopts.tsize - 1] *
+ (slot / stab->szopts.tsize) +
+ stab->data[slot % stab->szopts.tsize];
+
+ pkt_len <<= stab->szopts.size_log;
+out:
+ if (unlikely(pkt_len < 1))
+ pkt_len = 1;
+ qdisc_skb_cb(skb)->pkt_len = pkt_len;
+}
+EXPORT_SYMBOL(qdisc_calculate_pkt_len);
+
+void qdisc_warn_nonwc(char *txt, struct Qdisc *qdisc)
+{
+ if (!(qdisc->flags & TCQ_F_WARN_NONWC)) {
+ printk(KERN_WARNING
+ "%s: %s qdisc %X: is non-work-conserving?\n",
+ txt, qdisc->ops->id, qdisc->handle >> 16);
+ qdisc->flags |= TCQ_F_WARN_NONWC;
+ }
+}
+EXPORT_SYMBOL(qdisc_warn_nonwc);
+
static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer)
{
struct qdisc_watchdog *wd = container_of(timer, struct qdisc_watchdog,
timer);
- struct net_device *dev = qdisc_dev(wd->qdisc);
wd->qdisc->flags &= ~TCQ_F_THROTTLED;
- smp_wmb();
- netif_schedule(dev);
+ __netif_schedule(qdisc_root(wd->qdisc));
return HRTIMER_NORESTART;
}
{
ktime_t time;
+ if (test_bit(__QDISC_STATE_DEACTIVATED,
+ &qdisc_root_sleeping(wd->qdisc)->state))
+ return;
+
wd->qdisc->flags |= TCQ_F_THROTTLED;
time = ktime_set(0, 0);
- time = ktime_add_ns(time, PSCHED_US2NS(expires));
+ time = ktime_add_ns(time, PSCHED_TICKS2NS(expires));
hrtimer_start(&wd->timer, time, HRTIMER_MODE_ABS);
}
EXPORT_SYMBOL(qdisc_watchdog_schedule);
}
EXPORT_SYMBOL(qdisc_watchdog_cancel);
-struct hlist_head *qdisc_class_hash_alloc(unsigned int n)
+static struct hlist_head *qdisc_class_hash_alloc(unsigned int n)
{
unsigned int size = n * sizeof(struct hlist_head), i;
struct hlist_head *h;
return i>0 ? autohandle : 0;
}
-/* Attach toplevel qdisc to device dev */
-
-static struct Qdisc *
-dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
-{
- struct netdev_queue *dev_queue;
- struct Qdisc *oqdisc;
-
- if (dev->flags & IFF_UP)
- dev_deactivate(dev);
-
- qdisc_lock_tree(dev);
- if (qdisc && qdisc->flags&TCQ_F_INGRESS) {
- oqdisc = dev->qdisc_ingress;
- /* Prune old scheduler */
- if (oqdisc && atomic_read(&oqdisc->refcnt) <= 1) {
- /* delete */
- qdisc_reset(oqdisc);
- dev->qdisc_ingress = NULL;
- } else { /* new */
- dev->qdisc_ingress = qdisc;
- }
-
- } else {
- dev_queue = &dev->tx_queue;
- oqdisc = dev_queue->qdisc_sleeping;
-
- /* Prune old scheduler */
- if (oqdisc && atomic_read(&oqdisc->refcnt) <= 1)
- qdisc_reset(oqdisc);
-
- /* ... and graft new one */
- if (qdisc == NULL)
- qdisc = &noop_qdisc;
- dev_queue->qdisc_sleeping = qdisc;
- dev_queue->qdisc = &noop_qdisc;
- }
-
- qdisc_unlock_tree(dev);
-
- if (dev->flags & IFF_UP)
- dev_activate(dev);
-
- return oqdisc;
-}
-
void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n)
{
const struct Qdisc_class_ops *cops;
}
EXPORT_SYMBOL(qdisc_tree_decrease_qlen);
-/* Graft qdisc "new" to class "classid" of qdisc "parent" or
- to device "dev".
+static void notify_and_destroy(struct net *net, struct sk_buff *skb,
+ struct nlmsghdr *n, u32 clid,
+ struct Qdisc *old, struct Qdisc *new)
+{
+ if (new || old)
+ qdisc_notify(net, skb, n, clid, old, new);
- Old qdisc is not destroyed but returned in *old.
+ if (old)
+ qdisc_destroy(old);
+}
+
+/* Graft qdisc "new" to class "classid" of qdisc "parent" or
+ * to device "dev".
+ *
+ * When appropriate send a netlink notification using 'skb'
+ * and "n".
+ *
+ * On success, destroy old qdisc.
*/
static int qdisc_graft(struct net_device *dev, struct Qdisc *parent,
- u32 classid,
- struct Qdisc *new, struct Qdisc **old)
+ struct sk_buff *skb, struct nlmsghdr *n, u32 classid,
+ struct Qdisc *new, struct Qdisc *old)
{
+ struct Qdisc *q = old;
+ struct net *net = dev_net(dev);
int err = 0;
- struct Qdisc *q = *old;
-
if (parent == NULL) {
- if (q && q->flags&TCQ_F_INGRESS) {
- *old = dev_graft_qdisc(dev, q);
+ unsigned int i, num_q, ingress;
+
+ ingress = 0;
+ num_q = dev->num_tx_queues;
+ if ((q && q->flags & TCQ_F_INGRESS) ||
+ (new && new->flags & TCQ_F_INGRESS)) {
+ num_q = 1;
+ ingress = 1;
+ }
+
+ if (dev->flags & IFF_UP)
+ dev_deactivate(dev);
+
+ if (new && new->ops->attach) {
+ new->ops->attach(new);
+ num_q = 0;
+ }
+
+ for (i = 0; i < num_q; i++) {
+ struct netdev_queue *dev_queue = &dev->rx_queue;
+
+ if (!ingress)
+ dev_queue = netdev_get_tx_queue(dev, i);
+
+ old = dev_graft_qdisc(dev_queue, new);
+ if (new && i > 0)
+ atomic_inc(&new->refcnt);
+
+ if (!ingress)
+ qdisc_destroy(old);
+ }
+
+ if (!ingress) {
+ notify_and_destroy(net, skb, n, classid,
+ dev->qdisc, new);
+ if (new && !new->ops->attach)
+ atomic_inc(&new->refcnt);
+ dev->qdisc = new ? : &noop_qdisc;
} else {
- *old = dev_graft_qdisc(dev, new);
+ notify_and_destroy(net, skb, n, classid, old, new);
}
+
+ if (dev->flags & IFF_UP)
+ dev_activate(dev);
} else {
const struct Qdisc_class_ops *cops = parent->ops->cl_ops;
- err = -EINVAL;
-
- if (cops) {
+ err = -EOPNOTSUPP;
+ if (cops && cops->graft) {
unsigned long cl = cops->get(parent, classid);
if (cl) {
- err = cops->graft(parent, cl, new, old);
+ err = cops->graft(parent, cl, new, &old);
cops->put(parent, cl);
- }
+ } else
+ err = -ENOENT;
}
+ if (!err)
+ notify_and_destroy(net, skb, n, classid, old, new);
}
return err;
}
+/* lockdep annotation is needed for ingress; egress gets it only for name */
+static struct lock_class_key qdisc_tx_lock;
+static struct lock_class_key qdisc_rx_lock;
+
/*
Allocate and initialize new qdisc.
static struct Qdisc *
qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
- u32 parent, u32 handle, struct nlattr **tca, int *errp)
+ struct Qdisc *p, u32 parent, u32 handle,
+ struct nlattr **tca, int *errp)
{
int err;
struct nlattr *kind = tca[TCA_KIND];
struct Qdisc *sch;
struct Qdisc_ops *ops;
+ struct qdisc_size_table *stab;
ops = qdisc_lookup_ops(kind);
-#ifdef CONFIG_KMOD
+#ifdef CONFIG_MODULES
if (ops == NULL && kind != NULL) {
char name[IFNAMSIZ];
if (nla_strlcpy(name, kind, IFNAMSIZ) < IFNAMSIZ) {
sch->parent = parent;
- sch->stats_lock = &dev_queue->lock;
if (handle == TC_H_INGRESS) {
sch->flags |= TCQ_F_INGRESS;
handle = TC_H_MAKE(TC_H_INGRESS, 0);
+ lockdep_set_class(qdisc_lock(sch), &qdisc_rx_lock);
} else {
if (handle == 0) {
handle = qdisc_alloc_handle(dev);
if (handle == 0)
goto err_out3;
}
+ lockdep_set_class(qdisc_lock(sch), &qdisc_tx_lock);
}
sch->handle = handle;
if (!ops->init || (err = ops->init(sch, tca[TCA_OPTIONS])) == 0) {
+ if (tca[TCA_STAB]) {
+ stab = qdisc_get_stab(tca[TCA_STAB]);
+ if (IS_ERR(stab)) {
+ err = PTR_ERR(stab);
+ goto err_out4;
+ }
+ sch->stab = stab;
+ }
if (tca[TCA_RATE]) {
+ spinlock_t *root_lock;
+
+ err = -EOPNOTSUPP;
+ if (sch->flags & TCQ_F_MQROOT)
+ goto err_out4;
+
+ if ((sch->parent != TC_H_ROOT) &&
+ !(sch->flags & TCQ_F_INGRESS) &&
+ (!p || !(p->flags & TCQ_F_MQROOT)))
+ root_lock = qdisc_root_sleeping_lock(sch);
+ else
+ root_lock = qdisc_lock(sch);
+
err = gen_new_estimator(&sch->bstats, &sch->rate_est,
- sch->stats_lock,
- tca[TCA_RATE]);
- if (err) {
- /*
- * Any broken qdiscs that would require
- * a ops->reset() here? The qdisc was never
- * in action so it shouldn't be necessary.
- */
- if (ops->destroy)
- ops->destroy(sch);
- goto err_out3;
- }
+ root_lock, tca[TCA_RATE]);
+ if (err)
+ goto err_out4;
}
- qdisc_lock_tree(dev);
- list_add_tail(&sch->list, &dev_queue->qdisc_list);
- qdisc_unlock_tree(dev);
+
+ qdisc_list_add(sch);
return sch;
}
err_out:
*errp = err;
return NULL;
+
+err_out4:
+ /*
+ * Any broken qdiscs that would require a ops->reset() here?
+ * The qdisc was never in action so it shouldn't be necessary.
+ */
+ qdisc_put_stab(sch->stab);
+ if (ops->destroy)
+ ops->destroy(sch);
+ goto err_out3;
}
static int qdisc_change(struct Qdisc *sch, struct nlattr **tca)
{
- if (tca[TCA_OPTIONS]) {
- int err;
+ struct qdisc_size_table *stab = NULL;
+ int err = 0;
+ if (tca[TCA_OPTIONS]) {
if (sch->ops->change == NULL)
return -EINVAL;
err = sch->ops->change(sch, tca[TCA_OPTIONS]);
if (err)
return err;
}
- if (tca[TCA_RATE])
+
+ if (tca[TCA_STAB]) {
+ stab = qdisc_get_stab(tca[TCA_STAB]);
+ if (IS_ERR(stab))
+ return PTR_ERR(stab);
+ }
+
+ qdisc_put_stab(sch->stab);
+ sch->stab = stab;
+
+ if (tca[TCA_RATE]) {
+ /* NB: ignores errors from replace_estimator
+ because change can't be undone. */
+ if (sch->flags & TCQ_F_MQROOT)
+ goto out;
gen_replace_estimator(&sch->bstats, &sch->rate_est,
- sch->stats_lock, tca[TCA_RATE]);
+ qdisc_root_sleeping_lock(sch),
+ tca[TCA_RATE]);
+ }
+out:
return 0;
}
struct Qdisc *p = NULL;
int err;
- if (net != &init_net)
- return -EINVAL;
-
- if ((dev = __dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
+ if ((dev = __dev_get_by_index(net, tcm->tcm_ifindex)) == NULL)
return -ENODEV;
err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL);
return -ENOENT;
q = qdisc_leaf(p, clid);
} else { /* ingress */
- q = dev->qdisc_ingress;
+ q = dev->rx_queue.qdisc_sleeping;
}
} else {
- struct netdev_queue *dev_queue = &dev->tx_queue;
- q = dev_queue->qdisc_sleeping;
+ q = dev->qdisc;
}
if (!q)
return -ENOENT;
return -EINVAL;
if (q->handle == 0)
return -ENOENT;
- if ((err = qdisc_graft(dev, p, clid, NULL, &q)) != 0)
+ if ((err = qdisc_graft(dev, p, skb, n, clid, NULL, q)) != 0)
return err;
- if (q) {
- qdisc_notify(skb, n, clid, q, NULL);
- qdisc_lock_tree(dev);
- qdisc_destroy(q);
- qdisc_unlock_tree(dev);
- }
} else {
- qdisc_notify(skb, n, clid, NULL, q);
+ qdisc_notify(net, skb, n, clid, NULL, q);
}
return 0;
}
struct Qdisc *q, *p;
int err;
- if (net != &init_net)
- return -EINVAL;
-
replay:
/* Reinit, just in case something touches this. */
tcm = NLMSG_DATA(n);
clid = tcm->tcm_parent;
q = p = NULL;
- if ((dev = __dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
+ if ((dev = __dev_get_by_index(net, tcm->tcm_ifindex)) == NULL)
return -ENODEV;
err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL);
return -ENOENT;
q = qdisc_leaf(p, clid);
} else { /*ingress */
- q = dev->qdisc_ingress;
+ q = dev->rx_queue.qdisc_sleeping;
}
} else {
- struct netdev_queue *dev_queue = &dev->tx_queue;
- q = dev_queue->qdisc_sleeping;
+ q = dev->qdisc;
}
/* It may be default qdisc, ignore it */
return -EINVAL;
err = qdisc_change(q, tca);
if (err == 0)
- qdisc_notify(skb, n, clid, NULL, q);
+ qdisc_notify(net, skb, n, clid, NULL, q);
return err;
create_n_graft:
if (!(n->nlmsg_flags&NLM_F_CREATE))
return -ENOENT;
if (clid == TC_H_INGRESS)
- q = qdisc_create(dev, &dev->rx_queue,
+ q = qdisc_create(dev, &dev->rx_queue, p,
tcm->tcm_parent, tcm->tcm_parent,
tca, &err);
- else
- q = qdisc_create(dev, &dev->tx_queue,
+ else {
+ struct netdev_queue *dev_queue;
+
+ if (p && p->ops->cl_ops && p->ops->cl_ops->select_queue)
+ dev_queue = p->ops->cl_ops->select_queue(p, tcm);
+ else if (p)
+ dev_queue = p->dev_queue;
+ else
+ dev_queue = netdev_get_tx_queue(dev, 0);
+
+ q = qdisc_create(dev, dev_queue, p,
tcm->tcm_parent, tcm->tcm_handle,
tca, &err);
+ }
if (q == NULL) {
if (err == -EAGAIN)
goto replay;
}
graft:
- if (1) {
- struct Qdisc *old_q = NULL;
- err = qdisc_graft(dev, p, clid, q, &old_q);
- if (err) {
- if (q) {
- qdisc_lock_tree(dev);
- qdisc_destroy(q);
- qdisc_unlock_tree(dev);
- }
- return err;
- }
- qdisc_notify(skb, n, clid, old_q, q);
- if (old_q) {
- qdisc_lock_tree(dev);
- qdisc_destroy(old_q);
- qdisc_unlock_tree(dev);
- }
+ err = qdisc_graft(dev, p, skb, n, clid, q, NULL);
+ if (err) {
+ if (q)
+ qdisc_destroy(q);
+ return err;
}
+
return 0;
}
goto nla_put_failure;
q->qstats.qlen = q->q.qlen;
- if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
- TCA_XSTATS, q->stats_lock, &d) < 0)
+ if (q->stab && qdisc_dump_stab(skb, q->stab) < 0)
+ goto nla_put_failure;
+
+ if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS,
+ qdisc_root_sleeping_lock(q), &d) < 0)
goto nla_put_failure;
if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0)
goto nla_put_failure;
if (gnet_stats_copy_basic(&d, &q->bstats) < 0 ||
- gnet_stats_copy_rate_est(&d, &q->rate_est) < 0 ||
+ gnet_stats_copy_rate_est(&d, &q->bstats, &q->rate_est) < 0 ||
gnet_stats_copy_queue(&d, &q->qstats) < 0)
goto nla_put_failure;
return -1;
}
-static int qdisc_notify(struct sk_buff *oskb, struct nlmsghdr *n,
- u32 clid, struct Qdisc *old, struct Qdisc *new)
+static int qdisc_notify(struct net *net, struct sk_buff *oskb,
+ struct nlmsghdr *n, u32 clid,
+ struct Qdisc *old, struct Qdisc *new)
{
struct sk_buff *skb;
u32 pid = oskb ? NETLINK_CB(oskb).pid : 0;
}
if (skb->len)
- return rtnetlink_send(skb, &init_net, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
+ return rtnetlink_send(skb, net, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
err_out:
kfree_skb(skb);
return -EINVAL;
}
+static bool tc_qdisc_dump_ignore(struct Qdisc *q)
+{
+ return (q->flags & TCQ_F_BUILTIN) ? true : false;
+}
+
+static int tc_dump_qdisc_root(struct Qdisc *root, struct sk_buff *skb,
+ struct netlink_callback *cb,
+ int *q_idx_p, int s_q_idx)
+{
+ int ret = 0, q_idx = *q_idx_p;
+ struct Qdisc *q;
+
+ if (!root)
+ return 0;
+
+ q = root;
+ if (q_idx < s_q_idx) {
+ q_idx++;
+ } else {
+ if (!tc_qdisc_dump_ignore(q) &&
+ tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).pid,
+ cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC) <= 0)
+ goto done;
+ q_idx++;
+ }
+ list_for_each_entry(q, &root->list, list) {
+ if (q_idx < s_q_idx) {
+ q_idx++;
+ continue;
+ }
+ if (!tc_qdisc_dump_ignore(q) &&
+ tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).pid,
+ cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC) <= 0)
+ goto done;
+ q_idx++;
+ }
+
+out:
+ *q_idx_p = q_idx;
+ return ret;
+done:
+ ret = -1;
+ goto out;
+}
+
static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
{
struct net *net = sock_net(skb->sk);
int idx, q_idx;
int s_idx, s_q_idx;
struct net_device *dev;
- struct Qdisc *q;
-
- if (net != &init_net)
- return 0;
s_idx = cb->args[0];
s_q_idx = q_idx = cb->args[1];
- read_lock(&dev_base_lock);
+
+ rcu_read_lock();
idx = 0;
- for_each_netdev(&init_net, dev) {
+ for_each_netdev_rcu(net, dev) {
struct netdev_queue *dev_queue;
+
if (idx < s_idx)
goto cont;
if (idx > s_idx)
s_q_idx = 0;
q_idx = 0;
- dev_queue = &dev->tx_queue;
- list_for_each_entry(q, &dev_queue->qdisc_list, list) {
- if (q_idx < s_q_idx) {
- q_idx++;
- continue;
- }
- if (tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).pid,
- cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC) <= 0)
- goto done;
- q_idx++;
- }
+
+ if (tc_dump_qdisc_root(dev->qdisc, skb, cb, &q_idx, s_q_idx) < 0)
+ goto done;
+
+ dev_queue = &dev->rx_queue;
+ if (tc_dump_qdisc_root(dev_queue->qdisc_sleeping, skb, cb, &q_idx, s_q_idx) < 0)
+ goto done;
+
cont:
idx++;
}
done:
- read_unlock(&dev_base_lock);
+ rcu_read_unlock();
cb->args[0] = idx;
cb->args[1] = q_idx;
static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
{
struct net *net = sock_net(skb->sk);
- struct netdev_queue *dev_queue;
struct tcmsg *tcm = NLMSG_DATA(n);
struct nlattr *tca[TCA_MAX + 1];
struct net_device *dev;
u32 qid = TC_H_MAJ(clid);
int err;
- if (net != &init_net)
- return -EINVAL;
-
- if ((dev = __dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
+ if ((dev = __dev_get_by_index(net, tcm->tcm_ifindex)) == NULL)
return -ENODEV;
err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL);
/* Step 1. Determine qdisc handle X:0 */
- dev_queue = &dev->tx_queue;
if (pid != TC_H_ROOT) {
u32 qid1 = TC_H_MAJ(pid);
} else if (qid1) {
qid = qid1;
} else if (qid == 0)
- qid = dev_queue->qdisc_sleeping->handle;
+ qid = dev->qdisc->handle;
/* Now qid is genuine qdisc handle consistent
both with parent and child.
pid = TC_H_MAKE(qid, pid);
} else {
if (qid == 0)
- qid = dev_queue->qdisc_sleeping->handle;
+ qid = dev->qdisc->handle;
}
/* OK. Locate qdisc */
goto out;
break;
case RTM_DELTCLASS:
- err = cops->delete(q, cl);
+ err = -EOPNOTSUPP;
+ if (cops->delete)
+ err = cops->delete(q, cl);
if (err == 0)
- tclass_notify(skb, n, q, cl, RTM_DELTCLASS);
+ tclass_notify(net, skb, n, q, cl, RTM_DELTCLASS);
goto out;
case RTM_GETTCLASS:
- err = tclass_notify(skb, n, q, cl, RTM_NEWTCLASS);
+ err = tclass_notify(net, skb, n, q, cl, RTM_NEWTCLASS);
goto out;
default:
err = -EINVAL;
}
new_cl = cl;
- err = cops->change(q, clid, pid, tca, &new_cl);
+ err = -EOPNOTSUPP;
+ if (cops->change)
+ err = cops->change(q, clid, pid, tca, &new_cl);
if (err == 0)
- tclass_notify(skb, n, q, new_cl, RTM_NEWTCLASS);
+ tclass_notify(net, skb, n, q, new_cl, RTM_NEWTCLASS);
out:
if (cl)
nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*tcm), flags);
tcm = NLMSG_DATA(nlh);
tcm->tcm_family = AF_UNSPEC;
+ tcm->tcm__pad1 = 0;
+ tcm->tcm__pad2 = 0;
tcm->tcm_ifindex = qdisc_dev(q)->ifindex;
tcm->tcm_parent = q->handle;
tcm->tcm_handle = q->handle;
if (cl_ops->dump && cl_ops->dump(q, cl, skb, tcm) < 0)
goto nla_put_failure;
- if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
- TCA_XSTATS, q->stats_lock, &d) < 0)
+ if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS,
+ qdisc_root_sleeping_lock(q), &d) < 0)
goto nla_put_failure;
if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0)
return -1;
}
-static int tclass_notify(struct sk_buff *oskb, struct nlmsghdr *n,
- struct Qdisc *q, unsigned long cl, int event)
+static int tclass_notify(struct net *net, struct sk_buff *oskb,
+ struct nlmsghdr *n, struct Qdisc *q,
+ unsigned long cl, int event)
{
struct sk_buff *skb;
u32 pid = oskb ? NETLINK_CB(oskb).pid : 0;
return -EINVAL;
}
- return rtnetlink_send(skb, &init_net, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
+ return rtnetlink_send(skb, net, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
}
struct qdisc_dump_args
a->cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWTCLASS);
}
+static int tc_dump_tclass_qdisc(struct Qdisc *q, struct sk_buff *skb,
+ struct tcmsg *tcm, struct netlink_callback *cb,
+ int *t_p, int s_t)
+{
+ struct qdisc_dump_args arg;
+
+ if (tc_qdisc_dump_ignore(q) ||
+ *t_p < s_t || !q->ops->cl_ops ||
+ (tcm->tcm_parent &&
+ TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
+ (*t_p)++;
+ return 0;
+ }
+ if (*t_p > s_t)
+ memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0]));
+ arg.w.fn = qdisc_class_dump;
+ arg.skb = skb;
+ arg.cb = cb;
+ arg.w.stop = 0;
+ arg.w.skip = cb->args[1];
+ arg.w.count = 0;
+ q->ops->cl_ops->walk(q, &arg.w);
+ cb->args[1] = arg.w.count;
+ if (arg.w.stop)
+ return -1;
+ (*t_p)++;
+ return 0;
+}
+
+static int tc_dump_tclass_root(struct Qdisc *root, struct sk_buff *skb,
+ struct tcmsg *tcm, struct netlink_callback *cb,
+ int *t_p, int s_t)
+{
+ struct Qdisc *q;
+
+ if (!root)
+ return 0;
+
+ if (tc_dump_tclass_qdisc(root, skb, tcm, cb, t_p, s_t) < 0)
+ return -1;
+
+ list_for_each_entry(q, &root->list, list) {
+ if (tc_dump_tclass_qdisc(q, skb, tcm, cb, t_p, s_t) < 0)
+ return -1;
+ }
+
+ return 0;
+}
+
static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
{
+ struct tcmsg *tcm = (struct tcmsg*)NLMSG_DATA(cb->nlh);
struct net *net = sock_net(skb->sk);
struct netdev_queue *dev_queue;
- int t;
- int s_t;
struct net_device *dev;
- struct Qdisc *q;
- struct tcmsg *tcm = (struct tcmsg*)NLMSG_DATA(cb->nlh);
- struct qdisc_dump_args arg;
-
- if (net != &init_net)
- return 0;
+ int t, s_t;
if (cb->nlh->nlmsg_len < NLMSG_LENGTH(sizeof(*tcm)))
return 0;
- if ((dev = dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
+ if ((dev = dev_get_by_index(net, tcm->tcm_ifindex)) == NULL)
return 0;
s_t = cb->args[0];
t = 0;
- dev_queue = &dev->tx_queue;
- list_for_each_entry(q, &dev_queue->qdisc_list, list) {
- if (t < s_t || !q->ops->cl_ops ||
- (tcm->tcm_parent &&
- TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
- t++;
- continue;
- }
- if (t > s_t)
- memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0]));
- arg.w.fn = qdisc_class_dump;
- arg.skb = skb;
- arg.cb = cb;
- arg.w.stop = 0;
- arg.w.skip = cb->args[1];
- arg.w.count = 0;
- q->ops->cl_ops->walk(q, &arg.w);
- cb->args[1] = arg.w.count;
- if (arg.w.stop)
- break;
- t++;
- }
+ if (tc_dump_tclass_root(dev->qdisc, skb, tcm, cb, &t, s_t) < 0)
+ goto done;
+
+ dev_queue = &dev->rx_queue;
+ if (tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, &t, s_t) < 0)
+ goto done;
+done:
cb->args[0] = t;
dev_put(dev);
tp = otp;
if (verd++ >= MAX_REC_LOOP) {
- printk("rule prio %u protocol %02x reclassify loop, "
- "packet dropped\n",
- tp->prio&0xffff, ntohs(tp->protocol));
+ if (net_ratelimit())
+ printk(KERN_NOTICE
+ "%s: packet reclassify loop"
+ " rule prio %u protocol %02x\n",
+ tp->q->ops->id,
+ tp->prio & 0xffff, ntohs(tp->protocol));
return TC_ACT_SHOT;
}
skb->tc_verd = SET_TC_VERD(skb->tc_verd, verd);
hrtimer_get_res(CLOCK_MONOTONIC, &ts);
seq_printf(seq, "%08x %08x %08x %08x\n",
- (u32)NSEC_PER_USEC, (u32)PSCHED_US2NS(1),
+ (u32)NSEC_PER_USEC, (u32)PSCHED_TICKS2NS(1),
1000000,
(u32)NSEC_PER_SEC/(u32)ktime_to_ns(timespec_to_ktime(ts)));
static int psched_open(struct inode *inode, struct file *file)
{
- return single_open(file, psched_show, PDE(inode)->data);
+ return single_open(file, psched_show, NULL);
}
static const struct file_operations psched_fops = {
.llseek = seq_lseek,
.release = single_release,
};
+
+static int __net_init psched_net_init(struct net *net)
+{
+ struct proc_dir_entry *e;
+
+ e = proc_net_fops_create(net, "psched", 0, &psched_fops);
+ if (e == NULL)
+ return -ENOMEM;
+
+ return 0;
+}
+
+static void __net_exit psched_net_exit(struct net *net)
+{
+ proc_net_remove(net, "psched");
+}
+#else
+static int __net_init psched_net_init(struct net *net)
+{
+ return 0;
+}
+
+static void __net_exit psched_net_exit(struct net *net)
+{
+}
#endif
+static struct pernet_operations psched_net_ops = {
+ .init = psched_net_init,
+ .exit = psched_net_exit,
+};
+
static int __init pktsched_init(void)
{
+ int err;
+
+ err = register_pernet_subsys(&psched_net_ops);
+ if (err) {
+ printk(KERN_ERR "pktsched_init: "
+ "cannot initialize per netns operations\n");
+ return err;
+ }
+
register_qdisc(&pfifo_qdisc_ops);
register_qdisc(&bfifo_qdisc_ops);
- proc_net_fops_create(&init_net, "psched", 0, &psched_fops);
+ register_qdisc(&pfifo_head_drop_qdisc_ops);
+ register_qdisc(&mq_qdisc_ops);
rtnl_register(PF_UNSPEC, RTM_NEWQDISC, tc_modify_qdisc, NULL);
rtnl_register(PF_UNSPEC, RTM_DELQDISC, tc_get_qdisc, NULL);
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff