1 files changed, 130 insertions, 143 deletions

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 151208704ed2..d5517719af4e 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -36,8 +36,23 @@ struct qdisc_rate_table {
 enum qdisc_state_t {
 	__QDISC_STATE_SCHED,
 	__QDISC_STATE_DEACTIVATED,
+	__QDISC_STATE_MISSED,
+	__QDISC_STATE_DRAINING,
 };
 
+enum qdisc_state2_t {
+	/* Only for !TCQ_F_NOLOCK qdisc. Never access it directly.
+	 * Use qdisc_run_begin/end() or qdisc_is_running() instead.
+	 */
+	__QDISC_STATE2_RUNNING,
+};
+
+#define QDISC_STATE_MISSED	BIT(__QDISC_STATE_MISSED)
+#define QDISC_STATE_DRAINING	BIT(__QDISC_STATE_DRAINING)
+
+#define QDISC_STATE_NON_EMPTY	(QDISC_STATE_MISSED | \
+					QDISC_STATE_DRAINING)
+
 struct qdisc_size_table {
 	struct rcu_head		rcu;
 	struct list_head	list;
@@ -89,9 +104,9 @@ struct Qdisc {
 	struct netdev_queue	*dev_queue;
 
 	struct net_rate_estimator __rcu *rate_est;
-	struct gnet_stats_basic_cpu __percpu *cpu_bstats;
+	struct gnet_stats_basic_sync __percpu *cpu_bstats;
 	struct gnet_stats_queue	__percpu *cpu_qstats;
-	int			padded;
+	int			pad;
 	refcount_t		refcnt;
 
 	/*
@@ -99,19 +114,20 @@ struct Qdisc {
 	 */
 	struct sk_buff_head	gso_skb ____cacheline_aligned_in_smp;
 	struct qdisc_skb_head	q;
-	struct gnet_stats_basic_packed bstats;
-	seqcount_t		running;
+	struct gnet_stats_basic_sync bstats;
 	struct gnet_stats_queue	qstats;
 	unsigned long		state;
+	unsigned long		state2; /* must be written under qdisc spinlock */
 	struct Qdisc            *next_sched;
 	struct sk_buff_head	skb_bad_txq;
 
 	spinlock_t		busylock ____cacheline_aligned_in_smp;
 	spinlock_t		seqlock;
 
-	/* for NOLOCK qdisc, true if there are no enqueued skbs */
-	bool			empty;
 	struct rcu_head		rcu;
+	netdevice_tracker	dev_tracker;
+	/* private data */
+	long privdata[] ____cacheline_aligned;
 };
 
 static inline void qdisc_refcount_inc(struct Qdisc *qdisc)
@@ -134,11 +150,20 @@ static inline struct Qdisc *qdisc_refcount_inc_nz(struct Qdisc *qdisc)
 	return NULL;
 }
 
+/* For !TCQ_F_NOLOCK qdisc: callers must either call this within a qdisc
+ * root_lock section, or provide their own memory barriers -- ordering
+ * against qdisc_run_begin/end() atomic bit operations.
+ */
 static inline bool qdisc_is_running(struct Qdisc *qdisc)
 {
 	if (qdisc->flags & TCQ_F_NOLOCK)
 		return spin_is_locked(&qdisc->seqlock);
-	return (raw_read_seqcount(&qdisc->running) & 1) ? true : false;
+	return test_bit(__QDISC_STATE2_RUNNING, &qdisc->state2);
+}
+
+static inline bool nolock_qdisc_is_empty(const struct Qdisc *qdisc)
+{
+	return !(READ_ONCE(qdisc->state) & QDISC_STATE_NON_EMPTY);
 }
 
 static inline bool qdisc_is_percpu_stats(const struct Qdisc *q)
@@ -149,32 +174,53 @@ static inline bool qdisc_is_percpu_stats(const struct Qdisc *q)
 static inline bool qdisc_is_empty(const struct Qdisc *qdisc)
 {
 	if (qdisc_is_percpu_stats(qdisc))
-		return READ_ONCE(qdisc->empty);
+		return nolock_qdisc_is_empty(qdisc);
 	return !READ_ONCE(qdisc->q.qlen);
 }
 
+/* For !TCQ_F_NOLOCK qdisc, qdisc_run_begin/end() must be invoked with
+ * the qdisc root lock acquired.
+ */
 static inline bool qdisc_run_begin(struct Qdisc *qdisc)
 {
 	if (qdisc->flags & TCQ_F_NOLOCK) {
-		if (!spin_trylock(&qdisc->seqlock))
+		if (spin_trylock(&qdisc->seqlock))
+			return true;
+
+		/* No need to insist if the MISSED flag was already set.
+		 * Note that test_and_set_bit() also gives us memory ordering
+		 * guarantees wrt potential earlier enqueue() and below
+		 * spin_trylock(), both of which are necessary to prevent races
+		 */
+		if (test_and_set_bit(__QDISC_STATE_MISSED, &qdisc->state))
 			return false;
-		WRITE_ONCE(qdisc->empty, false);
-	} else if (qdisc_is_running(qdisc)) {
-		return false;
+
+		/* Try to take the lock again to make sure that we will either
+		 * grab it or the CPU that still has it will see MISSED set
+		 * when testing it in qdisc_run_end()
+		 */
+		return spin_trylock(&qdisc->seqlock);
 	}
-	/* Variant of write_seqcount_begin() telling lockdep a trylock
-	 * was attempted.
-	 */
-	raw_write_seqcount_begin(&qdisc->running);
-	seqcount_acquire(&qdisc->running.dep_map, 0, 1, _RET_IP_);
-	return true;
+	return !__test_and_set_bit(__QDISC_STATE2_RUNNING, &qdisc->state2);
 }
 
 static inline void qdisc_run_end(struct Qdisc *qdisc)
 {
-	write_seqcount_end(&qdisc->running);
-	if (qdisc->flags & TCQ_F_NOLOCK)
+	if (qdisc->flags & TCQ_F_NOLOCK) {
 		spin_unlock(&qdisc->seqlock);
+
+		/* spin_unlock() only has store-release semantic. The unlock
+		 * and test_bit() ordering is a store-load ordering, so a full
+		 * memory barrier is needed here.
+		 */
+		smp_mb();
+
+		if (unlikely(test_bit(__QDISC_STATE_MISSED,
+				      &qdisc->state)))
+			__netif_schedule(qdisc);
+	} else {
+		__clear_bit(__QDISC_STATE2_RUNNING, &qdisc->state2);
+	}
 }
 
 static inline bool qdisc_may_bulk(const struct Qdisc *qdisc)
@@ -207,7 +253,8 @@ struct Qdisc_class_ops {
 	int			(*change)(struct Qdisc *, u32, u32,
 					struct nlattr **, unsigned long *,
 					struct netlink_ext_ack *);
-	int			(*delete)(struct Qdisc *, unsigned long);
+	int			(*delete)(struct Qdisc *, unsigned long,
+					  struct netlink_ext_ack *);
 	void			(*walk)(struct Qdisc *, struct qdisc_walker * arg);
 
 	/* Filter manipulation */
@@ -254,6 +301,8 @@ struct Qdisc_ops {
 					  struct netlink_ext_ack *extack);
 	void			(*attach)(struct Qdisc *sch);
 	int			(*change_tx_queue_len)(struct Qdisc *, unsigned int);
+	void			(*change_real_num_tx)(struct Qdisc *sch,
+						      unsigned int new_real_tx);
 
 	int			(*dump)(struct Qdisc *, struct sk_buff *);
 	int			(*dump_stats)(struct Qdisc *, struct gnet_dump *);
@@ -277,11 +326,6 @@ struct tcf_result {
 		};
 		const struct tcf_proto *goto_tp;
 
-		/* used in the skb_tc_reinsert function */
-		struct {
-			bool		ingress;
-			struct gnet_stats_queue *qstats;
-		};
 	};
 };
 
@@ -303,7 +347,7 @@ struct tcf_proto_ops {
 	int			(*change)(struct net *net, struct sk_buff *,
 					struct tcf_proto*, unsigned long,
 					u32 handle, struct nlattr **,
-					void **, bool, bool,
+					void **, u32,
 					struct netlink_ext_ack *);
 	int			(*delete)(struct tcf_proto *tp, void *arg,
 					  bool *last, bool rtnl_held,
@@ -330,6 +374,10 @@ struct tcf_proto_ops {
 	int			(*dump)(struct net*, struct tcf_proto*, void *,
 					struct sk_buff *skb, struct tcmsg*,
 					bool);
+	int			(*terse_dump)(struct net *net,
+					      struct tcf_proto *tp, void *fh,
+					      struct sk_buff *skb,
+					      struct tcmsg *t, bool rtnl_held);
 	int			(*tmplt_dump)(struct sk_buff *skb,
 					      struct net *net,
 					      void *tmplt_priv);
@@ -407,6 +455,7 @@ struct tcf_block {
 	struct mutex lock;
 	struct list_head chain_list;
 	u32 index; /* block index for shared blocks */
+	u32 classid; /* which class this block belongs to */
 	refcount_t refcnt;
 	struct net *net;
 	struct Qdisc *q;
@@ -426,7 +475,6 @@ struct tcf_block {
 	struct mutex proto_destroy_lock; /* Lock for proto_destroy hashtable. */
 };
 
-#ifdef CONFIG_PROVE_LOCKING
 static inline bool lockdep_tcf_chain_is_locked(struct tcf_chain *chain)
 {
 	return lockdep_is_held(&chain->filter_chain_lock);
@@ -436,17 +484,6 @@ static inline bool lockdep_tcf_proto_is_locked(struct tcf_proto *tp)
 {
 	return lockdep_is_held(&tp->lock);
 }
-#else
-static inline bool lockdep_tcf_chain_is_locked(struct tcf_block *chain)
-{
-	return true;
-}
-
-static inline bool lockdep_tcf_proto_is_locked(struct tcf_proto *tp)
-{
-	return true;
-}
-#endif /* #ifdef CONFIG_PROVE_LOCKING */
 
 #define tcf_chain_dereference(p, chain)					\
 	rcu_dereference_protected(p, lockdep_tcf_chain_is_locked(chain))
@@ -458,15 +495,10 @@ static inline void qdisc_cb_private_validate(const struct sk_buff *skb, int sz)
 {
 	struct qdisc_skb_cb *qcb;
 
-	BUILD_BUG_ON(sizeof(skb->cb) < offsetof(struct qdisc_skb_cb, data) + sz);
+	BUILD_BUG_ON(sizeof(skb->cb) < sizeof(*qcb));
 	BUILD_BUG_ON(sizeof(qcb->data) < sz);
 }
 
-static inline int qdisc_qlen_cpu(const struct Qdisc *q)
-{
-	return this_cpu_ptr(q->cpu_qstats)->qlen;
-}
-
 static inline int qdisc_qlen(const struct Qdisc *q)
 {
 	return q->q.qlen;
@@ -514,25 +546,6 @@ static inline struct Qdisc *qdisc_root_sleeping(const struct Qdisc *qdisc)
 	return qdisc->dev_queue->qdisc_sleeping;
 }
 
-/* The qdisc root lock is a mechanism by which to top level
- * of a qdisc tree can be locked from any qdisc node in the
- * forest.  This allows changing the configuration of some
- * aspect of the qdisc tree while blocking out asynchronous
- * qdisc access in the packet processing paths.
- *
- * It is only legal to do this when the root will not change
- * on us.  Otherwise we'll potentially lock the wrong qdisc
- * root.  This is enforced by holding the RTNL semaphore, which
- * all users of this lock accessor must do.
- */
-static inline spinlock_t *qdisc_root_lock(const struct Qdisc *qdisc)
-{
-	struct Qdisc *root = qdisc_root(qdisc);
-
-	ASSERT_RTNL();
-	return qdisc_lock(root);
-}
-
 static inline spinlock_t *qdisc_root_sleeping_lock(const struct Qdisc *qdisc)
 {
 	struct Qdisc *root = qdisc_root_sleeping(qdisc);
@@ -541,27 +554,25 @@ static inline spinlock_t *qdisc_root_sleeping_lock(const struct Qdisc *qdisc)
 	return qdisc_lock(root);
 }
 
-static inline seqcount_t *qdisc_root_sleeping_running(const struct Qdisc *qdisc)
-{
-	struct Qdisc *root = qdisc_root_sleeping(qdisc);
-
-	ASSERT_RTNL();
-	return &root->running;
-}
-
 static inline struct net_device *qdisc_dev(const struct Qdisc *qdisc)
 {
 	return qdisc->dev_queue->dev;
 }
 
-static inline void sch_tree_lock(const struct Qdisc *q)
+static inline void sch_tree_lock(struct Qdisc *q)
 {
-	spin_lock_bh(qdisc_root_sleeping_lock(q));
+	if (q->flags & TCQ_F_MQROOT)
+		spin_lock_bh(qdisc_lock(q));
+	else
+		spin_lock_bh(qdisc_root_sleeping_lock(q));
 }
 
-static inline void sch_tree_unlock(const struct Qdisc *q)
+static inline void sch_tree_unlock(struct Qdisc *q)
 {
-	spin_unlock_bh(qdisc_root_sleeping_lock(q));
+	if (q->flags & TCQ_F_MQROOT)
+		spin_unlock_bh(qdisc_lock(q));
+	else
+		spin_unlock_bh(qdisc_root_sleeping_lock(q));
 }
 
 extern struct Qdisc noop_qdisc;
@@ -629,6 +640,8 @@ void qdisc_class_hash_grow(struct Qdisc *, struct Qdisc_class_hash *);
 void qdisc_class_hash_destroy(struct Qdisc_class_hash *);
 
 int dev_qdisc_change_tx_queue_len(struct net_device *dev);
+void dev_qdisc_change_real_num_tx(struct net_device *dev,
+				  unsigned int new_real_tx);
 void dev_init_scheduler(struct net_device *dev);
 void dev_shutdown(struct net_device *dev);
 void dev_activate(struct net_device *dev);
@@ -664,6 +677,9 @@ qdisc_offload_graft_helper(struct net_device *dev, struct Qdisc *sch,
 {
 }
 #endif
+void qdisc_offload_query_caps(struct net_device *dev,
+			      enum tc_setup_type type,
+			      void *caps, size_t caps_len);
 struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
 			  const struct Qdisc_ops *ops,
 			  struct netlink_ext_ack *extack);
@@ -675,22 +691,6 @@ void __qdisc_calculate_pkt_len(struct sk_buff *skb,
 			       const struct qdisc_size_table *stab);
 int skb_do_redirect(struct sk_buff *);
 
-static inline void skb_reset_tc(struct sk_buff *skb)
-{
-#ifdef CONFIG_NET_CLS_ACT
-	skb->tc_redirected = 0;
-#endif
-}
-
-static inline bool skb_is_tc_redirected(const struct sk_buff *skb)
-{
-#ifdef CONFIG_NET_CLS_ACT
-	return skb->tc_redirected;
-#else
-	return false;
-#endif
-}
-
 static inline bool skb_at_tc_ingress(const struct sk_buff *skb)
 {
 #ifdef CONFIG_NET_CLS_ACT
@@ -726,11 +726,6 @@ static inline void qdisc_reset_all_tx_gt(struct net_device *dev, unsigned int i)
 	}
 }
 
-static inline void qdisc_reset_all_tx(struct net_device *dev)
-{
-	qdisc_reset_all_tx_gt(dev, 0);
-}
-
 /* Are all TX queues of the device empty?  */
 static inline bool qdisc_all_tx_empty(const struct net_device *dev)
 {
@@ -811,14 +806,16 @@ static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 	return sch->enqueue(skb, sch, to_free);
 }
 
-static inline void _bstats_update(struct gnet_stats_basic_packed *bstats,
+static inline void _bstats_update(struct gnet_stats_basic_sync *bstats,
 				  __u64 bytes, __u32 packets)
 {
-	bstats->bytes += bytes;
-	bstats->packets += packets;
+	u64_stats_update_begin(&bstats->syncp);
+	u64_stats_add(&bstats->bytes, bytes);
+	u64_stats_add(&bstats->packets, packets);
+	u64_stats_update_end(&bstats->syncp);
 }
 
-static inline void bstats_update(struct gnet_stats_basic_packed *bstats,
+static inline void bstats_update(struct gnet_stats_basic_sync *bstats,
 				 const struct sk_buff *skb)
 {
 	_bstats_update(bstats,
@@ -826,26 +823,10 @@ static inline void bstats_update(struct gnet_stats_basic_packed *bstats,
 		       skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1);
 }
 
-static inline void _bstats_cpu_update(struct gnet_stats_basic_cpu *bstats,
-				      __u64 bytes, __u32 packets)
-{
-	u64_stats_update_begin(&bstats->syncp);
-	_bstats_update(&bstats->bstats, bytes, packets);
-	u64_stats_update_end(&bstats->syncp);
-}
-
-static inline void bstats_cpu_update(struct gnet_stats_basic_cpu *bstats,
-				     const struct sk_buff *skb)
-{
-	u64_stats_update_begin(&bstats->syncp);
-	bstats_update(&bstats->bstats, skb);
-	u64_stats_update_end(&bstats->syncp);
-}
-
 static inline void qdisc_bstats_cpu_update(struct Qdisc *sch,
 					   const struct sk_buff *skb)
 {
-	bstats_cpu_update(this_cpu_ptr(sch->cpu_bstats), skb);
+	bstats_update(this_cpu_ptr(sch->cpu_bstats), skb);
 }
 
 static inline void qdisc_bstats_update(struct Qdisc *sch,
@@ -934,10 +915,9 @@ static inline void qdisc_qstats_qlen_backlog(struct Qdisc *sch,  __u32 *qlen,
 					     __u32 *backlog)
 {
 	struct gnet_stats_queue qstats = { 0 };
-	__u32 len = qdisc_qlen_sum(sch);
 
-	__gnet_stats_copy_queue(&qstats, sch->cpu_qstats, &sch->qstats, len);
-	*qlen = qstats.qlen;
+	gnet_stats_add_queue(&qstats, sch->cpu_qstats, &sch->qstats);
+	*qlen = qstats.qlen + qdisc_qlen(sch);
 	*backlog = qstats.backlog;
 }
 
@@ -958,13 +938,6 @@ static inline void qdisc_purge_queue(struct Qdisc *sch)
 	qdisc_tree_reduce_backlog(sch, qlen, backlog);
 }
 
-static inline void qdisc_skb_head_init(struct qdisc_skb_head *qh)
-{
-	qh->head = NULL;
-	qh->tail = NULL;
-	qh->qlen = 0;
-}
-
 static inline void __qdisc_enqueue_tail(struct sk_buff *skb,
 					struct qdisc_skb_head *qh)
 {
@@ -1062,12 +1035,6 @@ static inline unsigned int __qdisc_queue_drop_head(struct Qdisc *sch,
 	return 0;
 }
 
-static inline unsigned int qdisc_queue_drop_head(struct Qdisc *sch,
-						 struct sk_buff **to_free)
-{
-	return __qdisc_queue_drop_head(sch, &sch->q, to_free);
-}
-
 static inline struct sk_buff *qdisc_peek_head(struct Qdisc *sch)
 {
 	const struct qdisc_skb_head *qh = &sch->q;
@@ -1161,7 +1128,6 @@ static inline void __qdisc_reset_queue(struct qdisc_skb_head *qh)
 static inline void qdisc_reset_queue(struct Qdisc *sch)
 {
 	__qdisc_reset_queue(&sch->q);
-	sch->qstats.backlog = 0;
 }
 
 static inline struct Qdisc *qdisc_replace(struct Qdisc *sch, struct Qdisc *new,
@@ -1173,7 +1139,7 @@ static inline struct Qdisc *qdisc_replace(struct Qdisc *sch, struct Qdisc *new,
 	old = *pold;
 	*pold = new;
 	if (old != NULL)
-		qdisc_tree_flush_backlog(old);
+		qdisc_purge_queue(old);
 	sch_tree_unlock(sch);
 
 	return old;
@@ -1230,6 +1196,7 @@ struct psched_ratecfg {
 	u64	rate_bytes_ps; /* bytes per second */
 	u32	mult;
 	u16	overhead;
+	u16	mpu;
 	u8	linklayer;
 	u8	shift;
 };
@@ -1239,6 +1206,9 @@ static inline u64 psched_l2t_ns(const struct psched_ratecfg *r,
 {
 	len += r->overhead;
 
+	if (len < r->mpu)
+		len = r->mpu;
+
 	if (unlikely(r->linklayer == TC_LINKLAYER_ATM))
 		return ((u64)(DIV_ROUND_UP(len,48)*53) * r->mult) >> r->shift;
 
@@ -1261,23 +1231,39 @@ static inline void psched_ratecfg_getrate(struct tc_ratespec *res,
 	res->rate = min_t(u64, r->rate_bytes_ps, ~0U);
 
 	res->overhead = r->overhead;
+	res->mpu = r->mpu;
 	res->linklayer = (r->linklayer & TC_LINKLAYER_MASK);
 }
 
+struct psched_pktrate {
+	u64	rate_pkts_ps; /* packets per second */
+	u32	mult;
+	u8	shift;
+};
+
+static inline u64 psched_pkt2t_ns(const struct psched_pktrate *r,
+				  unsigned int pkt_num)
+{
+	return ((u64)pkt_num * r->mult) >> r->shift;
+}
+
+void psched_ppscfg_precompute(struct psched_pktrate *r, u64 pktrate64);
+
 /* Mini Qdisc serves for specific needs of ingress/clsact Qdisc.
  * The fast path only needs to access filter list and to update stats
  */
 struct mini_Qdisc {
 	struct tcf_proto *filter_list;
-	struct gnet_stats_basic_cpu __percpu *cpu_bstats;
+	struct tcf_block *block;
+	struct gnet_stats_basic_sync __percpu *cpu_bstats;
 	struct gnet_stats_queue	__percpu *cpu_qstats;
-	struct rcu_head rcu;
+	unsigned long rcu_state;
 };
 
 static inline void mini_qdisc_bstats_cpu_update(struct mini_Qdisc *miniq,
 						const struct sk_buff *skb)
 {
-	bstats_cpu_update(this_cpu_ptr(miniq->cpu_bstats), skb);
+	bstats_update(this_cpu_ptr(miniq->cpu_bstats), skb);
 }
 
 static inline void mini_qdisc_qstats_cpu_drop(struct mini_Qdisc *miniq)
@@ -1295,10 +1281,11 @@ void mini_qdisc_pair_swap(struct mini_Qdisc_pair *miniqp,
 			  struct tcf_proto *tp_head);
 void mini_qdisc_pair_init(struct mini_Qdisc_pair *miniqp, struct Qdisc *qdisc,
 			  struct mini_Qdisc __rcu **p_miniq);
+void mini_qdisc_pair_block_init(struct mini_Qdisc_pair *miniqp,
+				struct tcf_block *block);
 
-static inline int skb_tc_reinsert(struct sk_buff *skb, struct tcf_result *res)
-{
-	return res->ingress ? netif_receive_skb(skb) : dev_queue_xmit(skb);
-}
+void mq_change_real_num_tx(struct Qdisc *sch, unsigned int new_real_tx);
+
+int sch_frag_xmit_hook(struct sk_buff *skb, int (*xmit)(struct sk_buff *skb));
 
 #endif