aboutsummaryrefslogtreecommitdiff
path: root/net/sched/sch_api.c
diff options
context:
space:
mode:
Diffstat (limited to 'net/sched/sch_api.c')
-rw-r--r--net/sched/sch_api.c128
1 files changed, 116 insertions, 12 deletions
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 281c1bded1f..58bed7599db 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -135,7 +135,7 @@ static DEFINE_RWLOCK(qdisc_mod_lock);
static struct Qdisc_ops *qdisc_base;
-/* Register/uregister queueing discipline */
+/* Register/unregister queueing discipline */
int register_qdisc(struct Qdisc_ops *qops)
{
@@ -200,6 +200,58 @@ int unregister_qdisc(struct Qdisc_ops *qops)
}
EXPORT_SYMBOL(unregister_qdisc);
+/* Get default qdisc if not otherwise specified */
+void qdisc_get_default(char *name, size_t len)
+{
+ read_lock(&qdisc_mod_lock);
+ strlcpy(name, default_qdisc_ops->id, len);
+ read_unlock(&qdisc_mod_lock);
+}
+
+static struct Qdisc_ops *qdisc_lookup_default(const char *name)
+{
+ struct Qdisc_ops *q = NULL;
+
+ for (q = qdisc_base; q; q = q->next) {
+ if (!strcmp(name, q->id)) {
+ if (!try_module_get(q->owner))
+ q = NULL;
+ break;
+ }
+ }
+
+ return q;
+}
+
+/* Set new default qdisc to use */
+int qdisc_set_default(const char *name)
+{
+ const struct Qdisc_ops *ops;
+
+ if (!capable(CAP_NET_ADMIN))
+ return -EPERM;
+
+ write_lock(&qdisc_mod_lock);
+ ops = qdisc_lookup_default(name);
+ if (!ops) {
+ /* Not found, drop lock and try to load module */
+ write_unlock(&qdisc_mod_lock);
+ request_module("sch_%s", name);
+ write_lock(&qdisc_mod_lock);
+
+ ops = qdisc_lookup_default(name);
+ }
+
+ if (ops) {
+ /* Set new default */
+ module_put(default_qdisc_ops->owner);
+ default_qdisc_ops = ops;
+ }
+ write_unlock(&qdisc_mod_lock);
+
+ return ops ? 0 : -ENOENT;
+}
+
/* We know handle. Find qdisc among all qdisc's attached to device
(root qdisc, all its children, children of children etc.)
*/
@@ -219,11 +271,16 @@ static struct Qdisc *qdisc_match_from_root(struct Qdisc *root, u32 handle)
return NULL;
}
-static void qdisc_list_add(struct Qdisc *q)
+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);
+ if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) {
+ struct Qdisc *root = qdisc_dev(q)->qdisc;
+
+ WARN_ON_ONCE(root == &noop_qdisc);
+ list_add_tail(&q->list, &root->list);
+ }
}
+EXPORT_SYMBOL(qdisc_list_add);
void qdisc_list_del(struct Qdisc *q)
{
@@ -285,6 +342,45 @@ static struct Qdisc_ops *qdisc_lookup_ops(struct nlattr *kind)
return q;
}
+/* The linklayer setting were not transferred from iproute2, in older
+ * versions, and the rate tables lookup systems have been dropped in
+ * the kernel. To keep backward compatible with older iproute2 tc
+ * utils, we detect the linklayer setting by detecting if the rate
+ * table were modified.
+ *
+ * For linklayer ATM table entries, the rate table will be aligned to
+ * 48 bytes, thus some table entries will contain the same value. The
+ * mpu (min packet unit) is also encoded into the old rate table, thus
+ * starting from the mpu, we find low and high table entries for
+ * mapping this cell. If these entries contain the same value, when
+ * the rate tables have been modified for linklayer ATM.
+ *
+ * This is done by rounding mpu to the nearest 48 bytes cell/entry,
+ * and then roundup to the next cell, calc the table entry one below,
+ * and compare.
+ */
+static __u8 __detect_linklayer(struct tc_ratespec *r, __u32 *rtab)
+{
+ int low = roundup(r->mpu, 48);
+ int high = roundup(low+1, 48);
+ int cell_low = low >> r->cell_log;
+ int cell_high = (high >> r->cell_log) - 1;
+
+ /* rtab is too inaccurate at rates > 100Mbit/s */
+ if ((r->rate > (100000000/8)) || (rtab[0] == 0)) {
+ pr_debug("TC linklayer: Giving up ATM detection\n");
+ return TC_LINKLAYER_ETHERNET;
+ }
+
+ if ((cell_high > cell_low) && (cell_high < 256)
+ && (rtab[cell_low] == rtab[cell_high])) {
+ pr_debug("TC linklayer: Detected ATM, low(%d)=high(%d)=%u\n",
+ cell_low, cell_high, rtab[cell_high]);
+ return TC_LINKLAYER_ATM;
+ }
+ return TC_LINKLAYER_ETHERNET;
+}
+
static struct qdisc_rate_table *qdisc_rtab_list;
struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, struct nlattr *tab)
@@ -308,6 +404,8 @@ struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, struct nlattr *ta
rtab->rate = *r;
rtab->refcnt = 1;
memcpy(rtab->data, nla_data(tab), 1024);
+ if (r->linklayer == TC_LINKLAYER_UNAWARE)
+ r->linklayer = __detect_linklayer(r, rtab->data);
rtab->next = qdisc_rtab_list;
qdisc_rtab_list = rtab;
}
@@ -465,7 +563,7 @@ out:
}
EXPORT_SYMBOL(__qdisc_calculate_pkt_len);
-void qdisc_warn_nonwc(char *txt, struct Qdisc *qdisc)
+void qdisc_warn_nonwc(const char *txt, struct Qdisc *qdisc)
{
if (!(qdisc->flags & TCQ_F_WARN_NONWC)) {
pr_warn("%s: %s qdisc %X: is non-work-conserving?\n",
@@ -644,9 +742,11 @@ void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n)
const struct Qdisc_class_ops *cops;
unsigned long cl;
u32 parentid;
+ int drops;
if (n == 0)
return;
+ drops = max_t(int, n, 0);
while ((parentid = sch->parent)) {
if (TC_H_MAJ(parentid) == TC_H_MAJ(TC_H_INGRESS))
return;
@@ -663,6 +763,7 @@ void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n)
cops->put(sch, cl);
}
sch->q.qlen -= n;
+ sch->qstats.drops += drops;
}
}
EXPORT_SYMBOL(qdisc_tree_decrease_qlen);
@@ -983,7 +1084,8 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n)
struct Qdisc *p = NULL;
int err;
- if ((n->nlmsg_type != RTM_GETQDISC) && !capable(CAP_NET_ADMIN))
+ if ((n->nlmsg_type != RTM_GETQDISC) &&
+ !netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN))
return -EPERM;
err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL);
@@ -1050,7 +1152,7 @@ static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n)
struct Qdisc *q, *p;
int err;
- if (!capable(CAP_NET_ADMIN))
+ if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN))
return -EPERM;
replay:
@@ -1203,6 +1305,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
struct gnet_dump d;
struct qdisc_size_table *stab;
+ cond_resched();
nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags);
if (!nlh)
goto out_nlmsg_trim;
@@ -1334,9 +1437,9 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
s_idx = cb->args[0];
s_q_idx = q_idx = cb->args[1];
- rcu_read_lock();
idx = 0;
- for_each_netdev_rcu(net, dev) {
+ ASSERT_RTNL();
+ for_each_netdev(net, dev) {
struct netdev_queue *dev_queue;
if (idx < s_idx)
@@ -1359,8 +1462,6 @@ cont:
}
done:
- rcu_read_unlock();
-
cb->args[0] = idx;
cb->args[1] = q_idx;
@@ -1390,7 +1491,8 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n)
u32 qid;
int err;
- if ((n->nlmsg_type != RTM_GETTCLASS) && !capable(CAP_NET_ADMIN))
+ if ((n->nlmsg_type != RTM_GETTCLASS) &&
+ !netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN))
return -EPERM;
err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL);
@@ -1517,6 +1619,7 @@ static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q,
struct gnet_dump d;
const struct Qdisc_class_ops *cl_ops = q->ops->cl_ops;
+ cond_resched();
nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags);
if (!nlh)
goto out_nlmsg_trim;
@@ -1813,6 +1916,7 @@ static int __init pktsched_init(void)
return err;
}
+ register_qdisc(&pfifo_fast_ops);
register_qdisc(&pfifo_qdisc_ops);
register_qdisc(&bfifo_qdisc_ops);
register_qdisc(&pfifo_head_drop_qdisc_ops);
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-10 09:54:09 +0000