net: sched: Remove Qdisc::running sequence counter

The Qdisc::running sequence counter has two uses:

1. Reliably reading qdisc's tc statistics while the qdisc is running
(a seqcount read/retry loop at gnet_stats_add_basic()).

2. As a flag, indicating whether the qdisc in question is running
(without any retry loops).

For the first usage, the Qdisc::running sequence counter write section,
qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what
is actually needed: the raw qdisc's bstats update. A u64_stats sync
point was thus introduced (in previous commits) inside the bstats
structure itself. A local u64_stats write section is then started and
stopped for the bstats updates.

Use that u64_stats sync point mechanism for the bstats read/retry loop
at gnet_stats_add_basic().

For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag,
accessed with atomic bitops, is sufficient. Using a bit flag instead of
a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads
to the SMP barriers implicitly added through raw_read_seqcount() and
write_seqcount_begin/end() getting removed. All call sites have been
surveyed though, and no required ordering was identified.

Now that the qdisc->running sequence counter is no longer used, remove
it.

Note, using u64_stats implies no sequence counter protection for 64-bit
architectures. This can lead to the qdisc tc statistics "packets" vs.
"bytes" values getting out of sync on rare occasions. The individual
values will still be valid.

Signed-off-by: Ahmed S. Darwish <[e-mail]>
Signed-off-by: Sebastian Andrzej Siewior <[e-mail]>
Signed-off-by: David S. Miller <[e-mail]>

include/linux/netdevice.h

···
  *	@sfp_bus:	attached &struct sfp_bus structure.
  *
  *	@qdisc_tx_busylock: lockdep class annotating Qdisc->busylock spinlock
- *	@qdisc_running_key: lockdep class annotating Qdisc->running seqcount
  *
  *	@proto_down:	protocol port state information can be sent to the
  *			switch driver and used to set the phys state of the
···
 	struct phy_device	*phydev;
 	struct sfp_bus		*sfp_bus;
 	struct lock_class_key	*qdisc_tx_busylock;
-	struct lock_class_key	*qdisc_running_key;
 	bool			proto_down;
 	unsigned		wol_enabled:1;
 	unsigned		threaded:1;
···
 #define netdev_lockdep_set_classes(dev)				\
 {								\
 	static struct lock_class_key qdisc_tx_busylock_key;	\
-	static struct lock_class_key qdisc_running_key;		\
 	static struct lock_class_key qdisc_xmit_lock_key;	\
 	static struct lock_class_key dev_addr_list_lock_key;	\
 	unsigned int i;						\
 								\
 	(dev)->qdisc_tx_busylock = &qdisc_tx_busylock_key;	\
-	(dev)->qdisc_running_key = &qdisc_running_key;		\
 	lockdep_set_class(&(dev)->addr_list_lock,		\
 			  &dev_addr_list_lock_key);		\
 	for (i = 0; i < (dev)->num_tx_queues; i++)		\

include/net/gen_stats.h

···
 				 spinlock_t *lock, struct gnet_dump *d,
 				 int padattr);
 
-int gnet_stats_copy_basic(const seqcount_t *running,
-			  struct gnet_dump *d,
+int gnet_stats_copy_basic(struct gnet_dump *d,
 			  struct gnet_stats_basic_sync __percpu *cpu,
-			  struct gnet_stats_basic_sync *b);
-void gnet_stats_add_basic(const seqcount_t *running,
-			  struct gnet_stats_basic_sync *bstats,
+			  struct gnet_stats_basic_sync *b, bool running);
+void gnet_stats_add_basic(struct gnet_stats_basic_sync *bstats,
 			  struct gnet_stats_basic_sync __percpu *cpu,
-			  struct gnet_stats_basic_sync *b);
-int gnet_stats_copy_basic_hw(const seqcount_t *running,
-			     struct gnet_dump *d,
+			  struct gnet_stats_basic_sync *b, bool running);
+int gnet_stats_copy_basic_hw(struct gnet_dump *d,
 			     struct gnet_stats_basic_sync __percpu *cpu,
-			     struct gnet_stats_basic_sync *b);
+			     struct gnet_stats_basic_sync *b, bool running);
 int gnet_stats_copy_rate_est(struct gnet_dump *d,
 			     struct net_rate_estimator __rcu **ptr);
 int gnet_stats_copy_queue(struct gnet_dump *d,
···
 		      struct gnet_stats_basic_sync __percpu *cpu_bstats,
 		      struct net_rate_estimator __rcu **rate_est,
 		      spinlock_t *lock,
-		      seqcount_t *running, struct nlattr *opt);
+		      bool running, struct nlattr *opt);
 void gen_kill_estimator(struct net_rate_estimator __rcu **ptr);
 int gen_replace_estimator(struct gnet_stats_basic_sync *bstats,
 			  struct gnet_stats_basic_sync __percpu *cpu_bstats,
 			  struct net_rate_estimator __rcu **ptr,
 			  spinlock_t *lock,
-			  seqcount_t *running, struct nlattr *opt);
+			  bool running, struct nlattr *opt);
 bool gen_estimator_active(struct net_rate_estimator __rcu **ptr);
 bool gen_estimator_read(struct net_rate_estimator __rcu **ptr,
 			struct gnet_stats_rate_est64 *sample);

include/net/sch_generic.h

···
 	__QDISC_STATE_DEACTIVATED,
 	__QDISC_STATE_MISSED,
 	__QDISC_STATE_DRAINING,
+	/* Only for !TCQ_F_NOLOCK qdisc. Never access it directly.
+	 * Use qdisc_run_begin/end() or qdisc_is_running() instead.
+	 */
+	__QDISC_STATE_RUNNING,
 };
 
 #define QDISC_STATE_MISSED	BIT(__QDISC_STATE_MISSED)
···
 	struct sk_buff_head	gso_skb ____cacheline_aligned_in_smp;
 	struct qdisc_skb_head	q;
 	struct gnet_stats_basic_sync bstats;
-	seqcount_t		running;
 	struct gnet_stats_queue	qstats;
 	unsigned long		state;
 	struct Qdisc            *next_sched;
···
 	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_STATE_RUNNING, &qdisc->state);
 }
 
 static inline bool nolock_qdisc_is_empty(const struct Qdisc *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) {
···
 		 * after it releases the lock at the end of qdisc_run_end().
 		 */
 		return spin_trylock(&qdisc->seqlock);
-	} else if (qdisc_is_running(qdisc)) {
-		return false;
 	}
-	/* 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_STATE_RUNNING, &qdisc->state);
 }
 
 static inline void qdisc_run_end(struct Qdisc *qdisc)
···
 				      &qdisc->state)))
 			__netif_schedule(qdisc);
 	} else {
-		write_seqcount_end(&qdisc->running);
+		clear_bit(__QDISC_STATE_RUNNING, &qdisc->state);
 	}
 }
 
···
 
 	ASSERT_RTNL();
 	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)

net/core/gen_estimator.c

···
 struct net_rate_estimator {
 	struct gnet_stats_basic_sync	*bstats;
 	spinlock_t		*stats_lock;
-	seqcount_t		*running;
+	bool			running;
 	struct gnet_stats_basic_sync __percpu *cpu_bstats;
 	u8			ewma_log;
 	u8			intvl_log; /* period : (250ms << intvl_log) */
···
 	if (e->stats_lock)
 		spin_lock(e->stats_lock);
 
-	gnet_stats_add_basic(e->running, b, e->cpu_bstats, e->bstats);
+	gnet_stats_add_basic(b, e->cpu_bstats, e->bstats, e->running);
 
 	if (e->stats_lock)
 		spin_unlock(e->stats_lock);
···
  * @cpu_bstats: bstats per cpu
  * @rate_est: rate estimator statistics
  * @lock: lock for statistics and control path
- * @running: qdisc running seqcount
+ * @running: true if @bstats represents a running qdisc, thus @bstats'
+ *           internal values might change during basic reads. Only used
+ *           if @bstats_cpu is NULL
  * @opt: rate estimator configuration TLV
  *
  * Creates a new rate estimator with &bstats as source and &rate_est
···
 		      struct gnet_stats_basic_sync __percpu *cpu_bstats,
 		      struct net_rate_estimator __rcu **rate_est,
 		      spinlock_t *lock,
-		      seqcount_t *running,
+		      bool running,
 		      struct nlattr *opt)
 {
 	struct gnet_estimator *parm = nla_data(opt);
···
  * @cpu_bstats: bstats per cpu
  * @rate_est: rate estimator statistics
  * @lock: lock for statistics and control path
- * @running: qdisc running seqcount (might be NULL)
+ * @running: true if @bstats represents a running qdisc, thus @bstats'
+ *           internal values might change during basic reads. Only used
+ *           if @cpu_bstats is NULL
  * @opt: rate estimator configuration TLV
  *
  * Replaces the configuration of a rate estimator by calling
···
 			  struct gnet_stats_basic_sync __percpu *cpu_bstats,
 			  struct net_rate_estimator __rcu **rate_est,
 			  spinlock_t *lock,
-			  seqcount_t *running, struct nlattr *opt)
+			  bool running, struct nlattr *opt)
 {
 	return gen_new_estimator(bstats, cpu_bstats, rate_est,
 				 lock, running, opt);

net/core/gen_stats.c

···
 	_bstats_update(bstats, t_bytes, t_packets);
 }
 
-void gnet_stats_add_basic(const seqcount_t *running,
-			  struct gnet_stats_basic_sync *bstats,
+void gnet_stats_add_basic(struct gnet_stats_basic_sync *bstats,
 			  struct gnet_stats_basic_sync __percpu *cpu,
-			  struct gnet_stats_basic_sync *b)
+			  struct gnet_stats_basic_sync *b, bool running)
 {
-	unsigned int seq;
+	unsigned int start;
 	u64 bytes = 0;
 	u64 packets = 0;
+
+	WARN_ON_ONCE((cpu || running) && !in_task());
 
 	if (cpu) {
 		gnet_stats_add_basic_cpu(bstats, cpu);
···
 	}
 	do {
 		if (running)
-			seq = read_seqcount_begin(running);
+			start = u64_stats_fetch_begin_irq(&b->syncp);
 		bytes = u64_stats_read(&b->bytes);
 		packets = u64_stats_read(&b->packets);
-	} while (running && read_seqcount_retry(running, seq));
+	} while (running && u64_stats_fetch_retry_irq(&b->syncp, start));
 
 	_bstats_update(bstats, bytes, packets);
 }
 EXPORT_SYMBOL(gnet_stats_add_basic);
 
 static int
-___gnet_stats_copy_basic(const seqcount_t *running,
-			 struct gnet_dump *d,
+___gnet_stats_copy_basic(struct gnet_dump *d,
 			 struct gnet_stats_basic_sync __percpu *cpu,
 			 struct gnet_stats_basic_sync *b,
-			 int type)
+			 int type, bool running)
 {
 	struct gnet_stats_basic_sync bstats;
 	u64 bstats_bytes, bstats_packets;
 
 	gnet_stats_basic_sync_init(&bstats);
-	gnet_stats_add_basic(running, &bstats, cpu, b);
+	gnet_stats_add_basic(&bstats, cpu, b, running);
 
 	bstats_bytes = u64_stats_read(&bstats.bytes);
 	bstats_packets = u64_stats_read(&bstats.packets);
···
 
 /**
  * gnet_stats_copy_basic - copy basic statistics into statistic TLV
- * @running: seqcount_t pointer
  * @d: dumping handle
  * @cpu: copy statistic per cpu
  * @b: basic statistics
+ * @running: true if @b represents a running qdisc, thus @b's
+ *           internal values might change during basic reads.
+ *           Only used if @cpu is NULL
+ *
+ * Context: task; must not be run from IRQ or BH contexts
  *
  * Appends the basic statistics to the top level TLV created by
  * gnet_stats_start_copy().
···
  * if the room in the socket buffer was not sufficient.
  */
 int
-gnet_stats_copy_basic(const seqcount_t *running,
-		      struct gnet_dump *d,
+gnet_stats_copy_basic(struct gnet_dump *d,
 		      struct gnet_stats_basic_sync __percpu *cpu,
-		      struct gnet_stats_basic_sync *b)
+		      struct gnet_stats_basic_sync *b,
+		      bool running)
 {
-	return ___gnet_stats_copy_basic(running, d, cpu, b,
-					TCA_STATS_BASIC);
+	return ___gnet_stats_copy_basic(d, cpu, b, TCA_STATS_BASIC, running);
 }
 EXPORT_SYMBOL(gnet_stats_copy_basic);
 
 /**
  * gnet_stats_copy_basic_hw - copy basic hw statistics into statistic TLV
- * @running: seqcount_t pointer
  * @d: dumping handle
  * @cpu: copy statistic per cpu
  * @b: basic statistics
+ * @running: true if @b represents a running qdisc, thus @b's
+ *           internal values might change during basic reads.
+ *           Only used if @cpu is NULL
+ *
+ * Context: task; must not be run from IRQ or BH contexts
  *
  * Appends the basic statistics to the top level TLV created by
  * gnet_stats_start_copy().
···
  * if the room in the socket buffer was not sufficient.
  */
 int
-gnet_stats_copy_basic_hw(const seqcount_t *running,
-			 struct gnet_dump *d,
+gnet_stats_copy_basic_hw(struct gnet_dump *d,
 			 struct gnet_stats_basic_sync __percpu *cpu,
-			 struct gnet_stats_basic_sync *b)
+			 struct gnet_stats_basic_sync *b,
+			 bool running)
 {
-	return ___gnet_stats_copy_basic(running, d, cpu, b,
-					TCA_STATS_BASIC_HW);
+	return ___gnet_stats_copy_basic(d, cpu, b, TCA_STATS_BASIC_HW, running);
 }
 EXPORT_SYMBOL(gnet_stats_copy_basic_hw);
 

net/sched/act_api.c

···
 	if (est) {
 		err = gen_new_estimator(&p->tcfa_bstats, p->cpu_bstats,
 					&p->tcfa_rate_est,
-					&p->tcfa_lock, NULL, est);
+					&p->tcfa_lock, false, est);
 		if (err)
 			goto err4;
 	}
···
 	if (err < 0)
 		goto errout;
 
-	if (gnet_stats_copy_basic(NULL, &d, p->cpu_bstats, &p->tcfa_bstats) < 0 ||
-	    gnet_stats_copy_basic_hw(NULL, &d, p->cpu_bstats_hw,
-				     &p->tcfa_bstats_hw) < 0 ||
+	if (gnet_stats_copy_basic(&d, p->cpu_bstats,
+				  &p->tcfa_bstats, false) < 0 ||
+	    gnet_stats_copy_basic_hw(&d, p->cpu_bstats_hw,
+				     &p->tcfa_bstats_hw, false) < 0 ||
 	    gnet_stats_copy_rate_est(&d, &p->tcfa_rate_est) < 0 ||
 	    gnet_stats_copy_queue(&d, p->cpu_qstats,
 				  &p->tcfa_qstats,

net/sched/act_police.c

···
 					    police->common.cpu_bstats,
 					    &police->tcf_rate_est,
 					    &police->tcf_lock,
-					    NULL, est);
+					    false, est);
 		if (err)
 			goto failure;
 	} else if (tb[TCA_POLICE_AVRATE] &&

net/sched/sch_api.c

···
 		cpu_qstats = q->cpu_qstats;
 	}
 
-	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(q),
-				  &d, cpu_bstats, &q->bstats) < 0 ||
+	if (gnet_stats_copy_basic(&d, cpu_bstats, &q->bstats, true) < 0 ||
 	    gnet_stats_copy_rate_est(&d, &q->rate_est) < 0 ||
 	    gnet_stats_copy_queue(&d, cpu_qstats, &q->qstats, qlen) < 0)
 		goto nla_put_failure;
···
 		rcu_assign_pointer(sch->stab, stab);
 	}
 	if (tca[TCA_RATE]) {
-		seqcount_t *running;
-
 		err = -EOPNOTSUPP;
 		if (sch->flags & TCQ_F_MQROOT) {
 			NL_SET_ERR_MSG(extack, "Cannot attach rate estimator to a multi-queue root qdisc");
 			goto err_out4;
 		}
 
-		if (sch->parent != TC_H_ROOT &&
-		    !(sch->flags & TCQ_F_INGRESS) &&
-		    (!p || !(p->flags & TCQ_F_MQROOT)))
-			running = qdisc_root_sleeping_running(sch);
-		else
-			running = &sch->running;
-
 		err = gen_new_estimator(&sch->bstats,
 					sch->cpu_bstats,
 					&sch->rate_est,
 					NULL,
-					running,
+					true,
 					tca[TCA_RATE]);
 		if (err) {
 			NL_SET_ERR_MSG(extack, "Failed to generate new estimator");
···
 				      sch->cpu_bstats,
 				      &sch->rate_est,
 				      NULL,
-				      qdisc_root_sleeping_running(sch),
+				      true,
 				      tca[TCA_RATE]);
 	}
 out:

net/sched/sch_atm.c

···
 {
 	struct atm_flow_data *flow = (struct atm_flow_data *)arg;
 
-	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-				  d, NULL, &flow->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, NULL, &flow->bstats, true) < 0 ||
 	    gnet_stats_copy_queue(d, NULL, &flow->qstats, flow->q->q.qlen) < 0)
 		return -1;
 

net/sched/sch_cbq.c

···
 	if (cl->undertime != PSCHED_PASTPERFECT)
 		cl->xstats.undertime = cl->undertime - q->now;
 
-	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-				  d, NULL, &cl->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, NULL, &cl->bstats, true) < 0 ||
 	    gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 ||
 	    gnet_stats_copy_queue(d, NULL, &cl->qstats, qlen) < 0)
 		return -1;
···
 			err = gen_replace_estimator(&cl->bstats, NULL,
 						    &cl->rate_est,
 						    NULL,
-						    qdisc_root_sleeping_running(sch),
+						    true,
 						    tca[TCA_RATE]);
 			if (err) {
 				NL_SET_ERR_MSG(extack, "Failed to replace specified rate estimator");
···
 
 	if (tca[TCA_RATE]) {
 		err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est,
-					NULL,
-					qdisc_root_sleeping_running(sch),
-					tca[TCA_RATE]);
+					NULL, true, tca[TCA_RATE]);
 		if (err) {
 			NL_SET_ERR_MSG(extack, "Couldn't create new estimator");
 			tcf_block_put(cl->block);

net/sched/sch_drr.c

···
 		if (tca[TCA_RATE]) {
 			err = gen_replace_estimator(&cl->bstats, NULL,
 						    &cl->rate_est,
-						    NULL,
-						    qdisc_root_sleeping_running(sch),
+						    NULL, true,
 						    tca[TCA_RATE]);
 			if (err) {
 				NL_SET_ERR_MSG(extack, "Failed to replace estimator");
···
 
 	if (tca[TCA_RATE]) {
 		err = gen_replace_estimator(&cl->bstats, NULL, &cl->rate_est,
-					    NULL,
-					    qdisc_root_sleeping_running(sch),
-					    tca[TCA_RATE]);
+					    NULL, true, tca[TCA_RATE]);
 		if (err) {
 			NL_SET_ERR_MSG(extack, "Failed to replace estimator");
 			qdisc_put(cl->qdisc);
···
 	if (qlen)
 		xstats.deficit = cl->deficit;
 
-	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-				  d, NULL, &cl->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, NULL, &cl->bstats, true) < 0 ||
 	    gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 ||
 	    gnet_stats_copy_queue(d, cl_q->cpu_qstats, &cl_q->qstats, qlen) < 0)
 		return -1;

net/sched/sch_ets.c

···
 	struct ets_class *cl = ets_class_from_arg(sch, arg);
 	struct Qdisc *cl_q = cl->qdisc;
 
-	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-				  d, NULL, &cl_q->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats, true) < 0 ||
 	    qdisc_qstats_copy(d, cl_q) < 0)
 		return -1;
 

net/sched/sch_generic.c

···
 
 /*
  * Transmit possibly several skbs, and handle the return status as
- * required. Owning running seqcount bit guarantees that
- * only one CPU can execute this function.
+ * required. Owning qdisc running bit guarantees that only one CPU
+ * can execute this function.
  *
  * Returns to the caller:
  *				false  - hardware queue frozen backoff
···
 	.ops		=	&noop_qdisc_ops,
 	.q.lock		=	__SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock),
 	.dev_queue	=	&noop_netdev_queue,
-	.running	=	SEQCNT_ZERO(noop_qdisc.running),
 	.busylock	=	__SPIN_LOCK_UNLOCKED(noop_qdisc.busylock),
 	.gso_skb = {
 		.next = (struct sk_buff *)&noop_qdisc.gso_skb,
···
 EXPORT_SYMBOL(pfifo_fast_ops);
 
 static struct lock_class_key qdisc_tx_busylock;
-static struct lock_class_key qdisc_running_key;
 
 struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
 			  const struct Qdisc_ops *ops,
···
 	spin_lock_init(&sch->seqlock);
 	lockdep_set_class(&sch->seqlock,
 			  dev->qdisc_tx_busylock ?: &qdisc_tx_busylock);
-
-	seqcount_init(&sch->running);
-	lockdep_set_class(&sch->running,
-			  dev->qdisc_running_key ?: &qdisc_running_key);
 
 	sch->ops = ops;
 	sch->flags = ops->static_flags;

net/sched/sch_hfsc.c

···
 			err = gen_replace_estimator(&cl->bstats, NULL,
 						    &cl->rate_est,
 						    NULL,
-						    qdisc_root_sleeping_running(sch),
+						    true,
 						    tca[TCA_RATE]);
 			if (err)
 				return err;
···
 
 	if (tca[TCA_RATE]) {
 		err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est,
-					NULL,
-					qdisc_root_sleeping_running(sch),
-					tca[TCA_RATE]);
+					NULL, true, tca[TCA_RATE]);
 		if (err) {
 			tcf_block_put(cl->block);
 			kfree(cl);
···
 	xstats.work    = cl->cl_total;
 	xstats.rtwork  = cl->cl_cumul;
 
-	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), d, NULL, &cl->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, NULL, &cl->bstats, true) < 0 ||
 	    gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 ||
 	    gnet_stats_copy_queue(d, NULL, &cl->qstats, qlen) < 0)
 		return -1;

net/sched/sch_htb.c

···
 		}
 	}
 
-	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-				  d, NULL, &cl->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, NULL, &cl->bstats, true) < 0 ||
 	    gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 ||
 	    gnet_stats_copy_queue(d, NULL, &qs, qlen) < 0)
 		return -1;
···
 			err = gen_new_estimator(&cl->bstats, NULL,
 						&cl->rate_est,
 						NULL,
-						qdisc_root_sleeping_running(sch),
+						true,
 						tca[TCA_RATE] ? : &est.nla);
 			if (err)
 				goto err_block_put;
···
 			err = gen_replace_estimator(&cl->bstats, NULL,
 						    &cl->rate_est,
 						    NULL,
-						    qdisc_root_sleeping_running(sch),
+						    true,
 						    tca[TCA_RATE]);
 			if (err)
 				return err;

net/sched/sch_mq.c

···
 		qdisc = netdev_get_tx_queue(dev, ntx)->qdisc_sleeping;
 		spin_lock_bh(qdisc_lock(qdisc));
 
-		gnet_stats_add_basic(NULL, &sch->bstats, qdisc->cpu_bstats,
-				     &qdisc->bstats);
+		gnet_stats_add_basic(&sch->bstats, qdisc->cpu_bstats,
+				     &qdisc->bstats, false);
 		gnet_stats_add_queue(&sch->qstats, qdisc->cpu_qstats,
 				     &qdisc->qstats);
 		sch->q.qlen += qdisc_qlen(qdisc);
···
 	struct netdev_queue *dev_queue = mq_queue_get(sch, cl);
 
 	sch = dev_queue->qdisc_sleeping;
-	if (gnet_stats_copy_basic(&sch->running, d, sch->cpu_bstats,
-				  &sch->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, sch->cpu_bstats, &sch->bstats, true) < 0 ||
 	    qdisc_qstats_copy(d, sch) < 0)
 		return -1;
 	return 0;

net/sched/sch_mqprio.c

···
 		qdisc = netdev_get_tx_queue(dev, ntx)->qdisc_sleeping;
 		spin_lock_bh(qdisc_lock(qdisc));
 
-		gnet_stats_add_basic(NULL, &sch->bstats, qdisc->cpu_bstats,
-				     &qdisc->bstats);
+		gnet_stats_add_basic(&sch->bstats, qdisc->cpu_bstats,
+				     &qdisc->bstats, false);
 		gnet_stats_add_queue(&sch->qstats, qdisc->cpu_qstats,
 				     &qdisc->qstats);
 		sch->q.qlen += qdisc_qlen(qdisc);
···
 
 			spin_lock_bh(qdisc_lock(qdisc));
 
-			gnet_stats_add_basic(NULL, &bstats, qdisc->cpu_bstats,
-					     &qdisc->bstats);
+			gnet_stats_add_basic(&bstats, qdisc->cpu_bstats,
+					     &qdisc->bstats, false);
 			gnet_stats_add_queue(&qstats, qdisc->cpu_qstats,
 					     &qdisc->qstats);
 			sch->q.qlen += qdisc_qlen(qdisc);
···
 		/* Reclaim root sleeping lock before completing stats */
 		if (d->lock)
 			spin_lock_bh(d->lock);
-		if (gnet_stats_copy_basic(NULL, d, NULL, &bstats) < 0 ||
+		if (gnet_stats_copy_basic(d, NULL, &bstats, false) < 0 ||
 		    gnet_stats_copy_queue(d, NULL, &qstats, qlen) < 0)
 			return -1;
 	} else {
 		struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl);
 
 		sch = dev_queue->qdisc_sleeping;
-		if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), d,
-					  sch->cpu_bstats, &sch->bstats) < 0 ||
+		if (gnet_stats_copy_basic(d, sch->cpu_bstats,
+					  &sch->bstats, true) < 0 ||
 		    qdisc_qstats_copy(d, sch) < 0)
 			return -1;
 	}

net/sched/sch_multiq.c

···
 	struct Qdisc *cl_q;
 
 	cl_q = q->queues[cl - 1];
-	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-				  d, cl_q->cpu_bstats, &cl_q->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, cl_q->cpu_bstats, &cl_q->bstats, true) < 0 ||
 	    qdisc_qstats_copy(d, cl_q) < 0)
 		return -1;
 

net/sched/sch_prio.c

···
 	struct Qdisc *cl_q;
 
 	cl_q = q->queues[cl - 1];
-	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-				  d, cl_q->cpu_bstats, &cl_q->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, cl_q->cpu_bstats,
+				  &cl_q->bstats, true) < 0 ||
 	    qdisc_qstats_copy(d, cl_q) < 0)
 		return -1;
 

net/sched/sch_qfq.c

···
 			err = gen_replace_estimator(&cl->bstats, NULL,
 						    &cl->rate_est,
 						    NULL,
-						    qdisc_root_sleeping_running(sch),
+						    true,
 						    tca[TCA_RATE]);
 			if (err)
 				return err;
···
 		err = gen_new_estimator(&cl->bstats, NULL,
 					&cl->rate_est,
 					NULL,
-					qdisc_root_sleeping_running(sch),
+					true,
 					tca[TCA_RATE]);
 		if (err)
 			goto destroy_class;
···
 	xstats.weight = cl->agg->class_weight;
 	xstats.lmax = cl->agg->lmax;
 
-	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-				  d, NULL, &cl->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, NULL, &cl->bstats, true) < 0 ||
 	    gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 ||
 	    qdisc_qstats_copy(d, cl->qdisc) < 0)
 		return -1;

net/sched/sch_taprio.c

···
 	struct netdev_queue *dev_queue = taprio_queue_get(sch, cl);
 
 	sch = dev_queue->qdisc_sleeping;
-	if (gnet_stats_copy_basic(&sch->running, d, NULL, &sch->bstats) < 0 ||
+	if (gnet_stats_copy_basic(d, NULL, &sch->bstats, true) < 0 ||
 	    qdisc_qstats_copy(d, sch) < 0)
 		return -1;
 	return 0;