From: Jarek Poplawski Date: Fri, 22 Aug 2008 10:24:05 +0000 (-0700) Subject: pkt_sched: Fix qdisc list locking X-Git-Tag: v2.6.27-rc5~42^2~2 X-Git-Url: https://err.no/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=f6e0b239a2657ea8cb67f0d83d0bfdbfd19a481b;p=linux-2.6 pkt_sched: Fix qdisc list locking Since some qdiscs call qdisc_tree_decrease_qlen() (so qdisc_lookup()) without rtnl_lock(), adding and deleting from a qdisc list needs additional locking. This patch adds global spinlock qdisc_list_lock and wrapper functions for modifying the list. It is considered as a temporary solution until hfsc_dequeue(), netem_dequeue() and tbf_dequeue() (or qdisc_tree_decrease_qlen()) are redone. With feedback from Herbert Xu and David S. Miller. Signed-off-by: Jarek Poplawski Acked-by: Herbert Xu Signed-off-by: David S. Miller --- diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index 853fe83d9f..b786a5b092 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -78,6 +78,7 @@ extern struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops, extern int register_qdisc(struct Qdisc_ops *qops); extern int unregister_qdisc(struct Qdisc_ops *qops); +extern void qdisc_list_del(struct Qdisc *q); extern struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle); extern struct Qdisc *qdisc_lookup_class(struct net_device *dev, u32 handle); extern struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 45f442d7de..e7fb9e0d21 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -199,19 +199,53 @@ struct Qdisc *qdisc_match_from_root(struct Qdisc *root, u32 handle) return NULL; } +/* + * This lock is needed until some qdiscs stop calling qdisc_tree_decrease_qlen() + * without rtnl_lock(); currently hfsc_dequeue(), netem_dequeue(), tbf_dequeue() + */ +static DEFINE_SPINLOCK(qdisc_list_lock); + +static void qdisc_list_add(struct Qdisc *q) +{ + if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { + spin_lock_bh(&qdisc_list_lock); + list_add_tail(&q->list, &qdisc_root_sleeping(q)->list); + spin_unlock_bh(&qdisc_list_lock); + } +} + +void qdisc_list_del(struct Qdisc *q) +{ + if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { + spin_lock_bh(&qdisc_list_lock); + list_del(&q->list); + spin_unlock_bh(&qdisc_list_lock); + } +} +EXPORT_SYMBOL(qdisc_list_del); + struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) { unsigned int i; + struct Qdisc *q; + + spin_lock_bh(&qdisc_list_lock); for (i = 0; i < dev->num_tx_queues; i++) { struct netdev_queue *txq = netdev_get_tx_queue(dev, i); - struct Qdisc *q, *txq_root = txq->qdisc_sleeping; + struct Qdisc *txq_root = txq->qdisc_sleeping; q = qdisc_match_from_root(txq_root, handle); if (q) - return q; + goto unlock; } - return qdisc_match_from_root(dev->rx_queue.qdisc_sleeping, handle); + + q = qdisc_match_from_root(dev->rx_queue.qdisc_sleeping, handle); + +unlock: + spin_unlock_bh(&qdisc_list_lock); + + return q; } static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid) @@ -810,8 +844,8 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, goto err_out3; } } - if ((parent != TC_H_ROOT) && !(sch->flags & TCQ_F_INGRESS)) - list_add_tail(&sch->list, &dev_queue->qdisc_sleeping->list); + + qdisc_list_add(sch); return sch; } diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index c3ed4d44fc..5f0ade7806 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -526,10 +526,9 @@ void qdisc_destroy(struct Qdisc *qdisc) !atomic_dec_and_test(&qdisc->refcnt)) return; - if (qdisc->parent) - list_del(&qdisc->list); - #ifdef CONFIG_NET_SCHED + qdisc_list_del(qdisc); + qdisc_put_stab(qdisc->stab); #endif gen_kill_estimator(&qdisc->bstats, &qdisc->rate_est);