Skip to contentSkip to conversation

net: sched: fix stats accounting for child NOLOCK qdiscs

PATCH In review Paolo Abeni 2 participants 4 messages Mar 28, 2019 – Apr 1, 2019

Subsystems (4)
Networking Network drivers Traffic control (tc) CBS/ETF/taprio qdiscs
lore
Share & tools
lore

Paolo Abeni

author maintainer
#0 lore raw

cover net: sched: fix stats accounting for child NOLOCK qdiscs

Currently, stats accounting for NOLOCK qdisc enslaved to classful (lock) qdiscs is buggy. Per CPU values are ignored in most places, as a result, stats dump in the above scenario always report 0 length backlog and parent backlog len is not updated correctly on NOLOCK qdisc removal.

The first patch address stats dumping, and the second one child qdisc removal. I'm targeting the net tree as this is a bugfix, but it could be moved to net-next due to the relatively large diffstat.

Paolo Abeni (2):
net: sched: introduce and use qstats read helpers
net: sched: introduce and use qdisc tree flush/purge helpers

 include/net/sch_generic.h | 44 ++++++++++++++++++++++++++++++++-------
 net/sched/sch_cbq.c       | 10 ++++-----
 net/sched/sch_drr.c       | 16 ++++----------
 net/sched/sch_hfsc.c      | 19 +++++------------
 net/sched/sch_htb.c       | 22 ++++++--------------
 net/sched/sch_mq.c        |  2 +-
 net/sched/sch_mqprio.c    |  3 +--
 net/sched/sch_multiq.c    | 10 ++++-----
 net/sched/sch_prio.c      | 10 +++------
 net/sched/sch_qfq.c       | 14 ++-----------
 net/sched/sch_red.c       |  3 +--
 net/sched/sch_sfb.c       |  3 +--
 net/sched/sch_taprio.c    |  2 +-
 net/sched/sch_tbf.c       |  3 +--
 14 files changed, 71 insertions(+), 90 deletions(-)
Signature · 1 line
2.20.1
author maintainer
#1 lore raw

PATCH 1/2 net: sched: introduce and use qstats read helpers

Classful qdiscs can't access directly the child qdiscs backlog
length: if such qdisc is NOLOCK, per CPU values should be accounted instead.

Most qdiscs no not respect the above. As a result, qstats fetching for most classful qdisc is currently incorrect: if the child qdisc is NOLOCK, it always reports 0 len backlog.

This change introduces a pair of helpers to safely fetch both backlog and qlen and use them in stats class dumping functions, fixing the above issue and cleaning a bit the code.

DRR needs also to access the child qdisc queue length, so it needs custom handling.

Fixes c5ad119fb6c0 ("net: sched: pfifo_fast use skb_array")
Signed-off-by Paolo Abeni <[email protected]>
---
 include/net/sch_generic.h | 18 ++++++++++++++++++
 net/sched/sch_cbq.c       |  4 +++-
 net/sched/sch_drr.c       |  5 +++--
 net/sched/sch_hfsc.c      |  5 +++--
 net/sched/sch_htb.c       |  7 +++----
 net/sched/sch_mq.c        |  2 +-
 net/sched/sch_mqprio.c    |  3 +--
 net/sched/sch_multiq.c    |  2 +-
 net/sched/sch_prio.c      |  2 +-
 net/sched/sch_qfq.c       |  3 +--
 net/sched/sch_taprio.c    |  2 +-
 11 files changed, 36 insertions(+), 17 deletions(-)
include/net/sch_generic.h +18 -0
index 7d1a0483a17b..43e4e17aa938 100644
@@ -923,6 +923,24 @@ static inline void qdisc_qstats_overlimit(struct Qdisc *sch)
 	sch->qstats.overlimits++;
 }
 
+static inline int qdisc_qstats_copy(struct gnet_dump *d, struct Qdisc *sch)
+{
+	__u32 qlen = qdisc_qlen_sum(sch);
+
+	return gnet_stats_copy_queue(d, sch->cpu_qstats, &sch->qstats, qlen);
+}
+
+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;
+	*backlog = qstats.backlog;
+}
+
 static inline void qdisc_skb_head_init(struct qdisc_skb_head *qh)
 {
 	qh->head = NULL;
net/sched/sch_cbq.c +3 -1
index 4dc05409e3fb..651879c1b655 100644
@@ -1358,9 +1358,11 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
 {
 	struct cbq_sched_data *q = qdisc_priv(sch);
 	struct cbq_class *cl = (struct cbq_class *)arg;
+	__u32 qlen;
 
 	cl->xstats.avgidle = cl->avgidle;
 	cl->xstats.undertime = 0;
+	qdisc_qstats_qlen_backlog(cl->q, &qlen, &cl->qstats.backlog);
 
 	if (cl->undertime != PSCHED_PASTPERFECT)
 		cl->xstats.undertime = cl->undertime - q->now;

@@ -1368,7 +1370,7 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
 	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
 				  d, NULL, &cl->bstats) < 0 ||
 	    gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 ||
-	    gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->q->q.qlen) < 0)
+	    gnet_stats_copy_queue(d, NULL, &cl->qstats, qlen) < 0)
 		return -1;
 
 	return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats));
net/sched/sch_drr.c +3 -2
index 09b800991065..8a181591b0ea 100644
@@ -269,7 +269,8 @@ static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg,
 				struct gnet_dump *d)
 {
 	struct drr_class *cl = (struct drr_class *)arg;
-	__u32 qlen = cl->qdisc->q.qlen;
+	__u32 qlen = qdisc_qlen_sum(cl->qdisc);
+	struct Qdisc *cl_q = cl->qdisc;
 	struct tc_drr_stats xstats;
 
 	memset(&xstats, 0, sizeof(xstats));

@@ -279,7 +280,7 @@ static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg,
 	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
 				  d, NULL, &cl->bstats) < 0 ||
 	    gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 ||
-	    gnet_stats_copy_queue(d, NULL, &cl->qdisc->qstats, qlen) < 0)
+	    gnet_stats_copy_queue(d, cl_q->cpu_qstats, &cl_q->qstats, qlen) < 0)
 		return -1;
 
 	return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
net/sched/sch_hfsc.c +3 -2
index 24cc220a3218..a946a419d717 100644
@@ -1328,8 +1328,9 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
 {
 	struct hfsc_class *cl = (struct hfsc_class *)arg;
 	struct tc_hfsc_stats xstats;
+	__u32 qlen;
 
-	cl->qstats.backlog = cl->qdisc->qstats.backlog;
+	qdisc_qstats_qlen_backlog(cl->qdisc, &qlen, &cl->qstats.backlog);
 	xstats.level   = cl->level;
 	xstats.period  = cl->cl_vtperiod;
 	xstats.work    = cl->cl_total;

@@ -1337,7 +1338,7 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
 
 	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), d, NULL, &cl->bstats) < 0 ||
 	    gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 ||
-	    gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->qdisc->q.qlen) < 0)
+	    gnet_stats_copy_queue(d, NULL, &cl->qstats, qlen) < 0)
 		return -1;
 
 	return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
net/sched/sch_htb.c +3 -4
index 30f9da7e1076..ed92836f528a 100644
@@ -1127,10 +1127,9 @@ htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d)
 	};
 	__u32 qlen = 0;
 
-	if (!cl->level && cl->leaf.q) {
-		qlen = cl->leaf.q->q.qlen;
-		qs.backlog = cl->leaf.q->qstats.backlog;
-	}
+	if (!cl->level && cl->leaf.q)
+		qdisc_qstats_qlen_backlog(cl->leaf.q, &qlen, &qs.backlog);
+
 	cl->xstats.tokens = clamp_t(s64, PSCHED_NS2TICKS(cl->tokens),
 				    INT_MIN, INT_MAX);
 	cl->xstats.ctokens = clamp_t(s64, PSCHED_NS2TICKS(cl->ctokens),
net/sched/sch_mq.c +1 -1
index 203659bc3906..3a3312467692 100644
@@ -249,7 +249,7 @@ static int mq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
 
 	sch = dev_queue->qdisc_sleeping;
 	if (gnet_stats_copy_basic(&sch->running, d, NULL, &sch->bstats) < 0 ||
-	    gnet_stats_copy_queue(d, NULL, &sch->qstats, sch->q.qlen) < 0)
+	    qdisc_qstats_copy(d, sch) < 0)
 		return -1;
 	return 0;
 }
net/sched/sch_mqprio.c +1 -2
index d364e63c396d..ea0dc112b38d 100644
@@ -561,8 +561,7 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
 		sch = dev_queue->qdisc_sleeping;
 		if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
 					  d, NULL, &sch->bstats) < 0 ||
-		    gnet_stats_copy_queue(d, NULL,
-					  &sch->qstats, sch->q.qlen) < 0)
+		    qdisc_qstats_copy(d, sch) < 0)
 			return -1;
 	}
 	return 0;
net/sched/sch_multiq.c +1 -1
index 7410ce4d0321..53c918a11378 100644
@@ -344,7 +344,7 @@ static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
 	cl_q = q->queues[cl - 1];
 	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
 				  d, NULL, &cl_q->bstats) < 0 ||
-	    gnet_stats_copy_queue(d, NULL, &cl_q->qstats, cl_q->q.qlen) < 0)
+	    qdisc_qstats_copy(d, cl_q) < 0)
 		return -1;
 
 	return 0;
net/sched/sch_prio.c +1 -1
index 847141cd900f..dfb06d5bfacc 100644
@@ -365,7 +365,7 @@ static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
 	cl_q = q->queues[cl - 1];
 	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
 				  d, NULL, &cl_q->bstats) < 0 ||
-	    gnet_stats_copy_queue(d, NULL, &cl_q->qstats, cl_q->q.qlen) < 0)
+	    qdisc_qstats_copy(d, cl_q) < 0)
 		return -1;
 
 	return 0;
net/sched/sch_qfq.c +1 -2
index 29f5c4a24688..9fbda3ec5861 100644
@@ -655,8 +655,7 @@ static int qfq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
 	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
 				  d, NULL, &cl->bstats) < 0 ||
 	    gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 ||
-	    gnet_stats_copy_queue(d, NULL,
-				  &cl->qdisc->qstats, cl->qdisc->q.qlen) < 0)
+	    qdisc_qstats_copy(d, cl->qdisc) < 0)
 		return -1;
 
 	return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
net/sched/sch_taprio.c +1 -1
index 206e4dbed12f..c7041999eb5d 100644
@@ -895,7 +895,7 @@ static int taprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
 
 	sch = dev_queue->qdisc_sleeping;
 	if (gnet_stats_copy_basic(&sch->running, d, NULL, &sch->bstats) < 0 ||
-	    gnet_stats_copy_queue(d, NULL, &sch->qstats, sch->q.qlen) < 0)
+	    qdisc_qstats_copy(d, sch) < 0)
 		return -1;
 	return 0;
 }
Signature · 1 line
2.20.1
author maintainer
#2 lore raw

PATCH 2/2 net: sched: introduce and use qdisc tree flush/purge helpers

In reply to Paolo Abeni

The same code to flush qdisc tree and purge the qdisc queue is duplicated in many places and in most cases it does not respect NOLOCK qdisc: the global backlog len is used and the per CPU values are ignored.

This change addresses the above, factoring-out the relevant code and using the helpers introduced by the previous patch to fetch the correct backlog len.

Fixes c5ad119fb6c0 ("net: sched: pfifo_fast use skb_array")
Signed-off-by Paolo Abeni <[email protected]>
---
 include/net/sch_generic.h | 26 +++++++++++++++++++-------
 net/sched/sch_cbq.c       |  6 +-----
 net/sched/sch_drr.c       | 11 +----------
 net/sched/sch_hfsc.c      | 14 ++------------
 net/sched/sch_htb.c       | 15 +++------------
 net/sched/sch_multiq.c    |  8 +++-----
 net/sched/sch_prio.c      |  8 ++------
 net/sched/sch_qfq.c       | 11 +----------
 net/sched/sch_red.c       |  3 +--
 net/sched/sch_sfb.c       |  3 +--
 net/sched/sch_tbf.c       |  3 +--
 11 files changed, 35 insertions(+), 73 deletions(-)
include/net/sch_generic.h +19 -7
index 43e4e17aa938..a2b38b3deeca 100644
@@ -941,6 +941,23 @@ static inline void qdisc_qstats_qlen_backlog(struct Qdisc *sch,  __u32 *qlen,
 	*backlog = qstats.backlog;
 }
 
+static inline void qdisc_tree_flush_backlog(struct Qdisc *sch)
+{
+	__u32 qlen, backlog;
+
+	qdisc_qstats_qlen_backlog(sch, &qlen, &backlog);
+	qdisc_tree_reduce_backlog(sch, qlen, backlog);
+}
+
+static inline void qdisc_purge_queue(struct Qdisc *sch)
+{
+	__u32 qlen, backlog;
+
+	qdisc_qstats_qlen_backlog(sch, &qlen, &backlog);
+	qdisc_reset(sch);
+	qdisc_tree_reduce_backlog(sch, qlen, backlog);
+}
+
 static inline void qdisc_skb_head_init(struct qdisc_skb_head *qh)
 {
 	qh->head = NULL;

@@ -1124,13 +1141,8 @@ static inline struct Qdisc *qdisc_replace(struct Qdisc *sch, struct Qdisc *new,
 	sch_tree_lock(sch);
 	old = *pold;
 	*pold = new;
-	if (old != NULL) {
-		unsigned int qlen = old->q.qlen;
-		unsigned int backlog = old->qstats.backlog;
-
-		qdisc_reset(old);
-		qdisc_tree_reduce_backlog(old, qlen, backlog);
-	}
+	if (old != NULL)
+		qdisc_tree_flush_backlog(old);
 	sch_tree_unlock(sch);
 
 	return old;
net/sched/sch_cbq.c +1 -5
index 651879c1b655..114b9048ea7e 100644
@@ -1667,17 +1667,13 @@ static int cbq_delete(struct Qdisc *sch, unsigned long arg)
 {
 	struct cbq_sched_data *q = qdisc_priv(sch);
 	struct cbq_class *cl = (struct cbq_class *)arg;
-	unsigned int qlen, backlog;
 
 	if (cl->filters || cl->children || cl == &q->link)
 		return -EBUSY;
 
 	sch_tree_lock(sch);
 
-	qlen = cl->q->q.qlen;
-	backlog = cl->q->qstats.backlog;
-	qdisc_reset(cl->q);
-	qdisc_tree_reduce_backlog(cl->q, qlen, backlog);
+	qdisc_purge_queue(cl->q);
 
 	if (cl->next_alive)
 		cbq_deactivate_class(cl);
net/sched/sch_drr.c +1 -10
index 8a181591b0ea..430df9a55ec4 100644
@@ -50,15 +50,6 @@ static struct drr_class *drr_find_class(struct Qdisc *sch, u32 classid)
 	return container_of(clc, struct drr_class, common);
 }
 
-static void drr_purge_queue(struct drr_class *cl)
-{
-	unsigned int len = cl->qdisc->q.qlen;
-	unsigned int backlog = cl->qdisc->qstats.backlog;
-
-	qdisc_reset(cl->qdisc);
-	qdisc_tree_reduce_backlog(cl->qdisc, len, backlog);
-}
-
 static const struct nla_policy drr_policy[TCA_DRR_MAX + 1] = {
 	[TCA_DRR_QUANTUM]	= { .type = NLA_U32 },
 };

@@ -167,7 +158,7 @@ static int drr_delete_class(struct Qdisc *sch, unsigned long arg)
 
 	sch_tree_lock(sch);
 
-	drr_purge_queue(cl);
+	qdisc_purge_queue(cl->qdisc);
 	qdisc_class_hash_remove(&q->clhash, &cl->common);
 
 	sch_tree_unlock(sch);
net/sched/sch_hfsc.c +2 -12
index a946a419d717..d2ab463f22ae 100644
@@ -844,16 +844,6 @@ qdisc_peek_len(struct Qdisc *sch)
 	return len;
 }
 
-static void
-hfsc_purge_queue(struct Qdisc *sch, struct hfsc_class *cl)
-{
-	unsigned int len = cl->qdisc->q.qlen;
-	unsigned int backlog = cl->qdisc->qstats.backlog;
-
-	qdisc_reset(cl->qdisc);
-	qdisc_tree_reduce_backlog(cl->qdisc, len, backlog);
-}
-
 static void
 hfsc_adjust_levels(struct hfsc_class *cl)
 {

@@ -1076,7 +1066,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	qdisc_class_hash_insert(&q->clhash, &cl->cl_common);
 	list_add_tail(&cl->siblings, &parent->children);
 	if (parent->level == 0)
-		hfsc_purge_queue(sch, parent);
+		qdisc_purge_queue(parent->qdisc);
 	hfsc_adjust_levels(parent);
 	sch_tree_unlock(sch);
 

@@ -1112,7 +1102,7 @@ hfsc_delete_class(struct Qdisc *sch, unsigned long arg)
 	list_del(&cl->siblings);
 	hfsc_adjust_levels(cl->cl_parent);
 
-	hfsc_purge_queue(sch, cl);
+	qdisc_purge_queue(cl->qdisc);
 	qdisc_class_hash_remove(&q->clhash, &cl->cl_common);
 
 	sch_tree_unlock(sch);
net/sched/sch_htb.c +3 -12
index ed92836f528a..2f9883b196e8 100644
@@ -1269,13 +1269,8 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg)
 
 	sch_tree_lock(sch);
 
-	if (!cl->level) {
-		unsigned int qlen = cl->leaf.q->q.qlen;
-		unsigned int backlog = cl->leaf.q->qstats.backlog;
-
-		qdisc_reset(cl->leaf.q);
-		qdisc_tree_reduce_backlog(cl->leaf.q, qlen, backlog);
-	}
+	if (!cl->level)
+		qdisc_purge_queue(cl->leaf.q);
 
 	/* delete from hash and active; remainder in destroy_class */
 	qdisc_class_hash_remove(&q->clhash, &cl->common);

@@ -1403,12 +1398,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 					  classid, NULL);
 		sch_tree_lock(sch);
 		if (parent && !parent->level) {
-			unsigned int qlen = parent->leaf.q->q.qlen;
-			unsigned int backlog = parent->leaf.q->qstats.backlog;
-
 			/* turn parent into inner node */
-			qdisc_reset(parent->leaf.q);
-			qdisc_tree_reduce_backlog(parent->leaf.q, qlen, backlog);
+			qdisc_purge_queue(parent->leaf.q);
 			qdisc_put(parent->leaf.q);
 			if (parent->prio_activity)
 				htb_deactivate(q, parent);
net/sched/sch_multiq.c +3 -5
index 53c918a11378..35b03ae08e0f 100644
@@ -201,9 +201,9 @@ static int multiq_tune(struct Qdisc *sch, struct nlattr *opt,
 	for (i = q->bands; i < q->max_bands; i++) {
 		if (q->queues[i] != &noop_qdisc) {
 			struct Qdisc *child = q->queues[i];
+
 			q->queues[i] = &noop_qdisc;
-			qdisc_tree_reduce_backlog(child, child->q.qlen,
-						  child->qstats.backlog);
+			qdisc_tree_flush_backlog(child);
 			qdisc_put(child);
 		}
 	}

@@ -225,9 +225,7 @@ static int multiq_tune(struct Qdisc *sch, struct nlattr *opt,
 					qdisc_hash_add(child, true);
 
 				if (old != &noop_qdisc) {
-					qdisc_tree_reduce_backlog(old,
-								  old->q.qlen,
-								  old->qstats.backlog);
+					qdisc_tree_flush_backlog(old);
 					qdisc_put(old);
 				}
 				sch_tree_unlock(sch);
net/sched/sch_prio.c +2 -6
index dfb06d5bfacc..d519b21535b3 100644
@@ -216,12 +216,8 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt,
 	q->bands = qopt->bands;
 	memcpy(q->prio2band, qopt->priomap, TC_PRIO_MAX+1);
 
-	for (i = q->bands; i < oldbands; i++) {
-		struct Qdisc *child = q->queues[i];
-
-		qdisc_tree_reduce_backlog(child, child->q.qlen,
-					  child->qstats.backlog);
-	}
+	for (i = q->bands; i < oldbands; i++)
+		qdisc_tree_flush_backlog(q->queues[i]);
 
 	for (i = oldbands; i < q->bands; i++) {
 		q->queues[i] = queues[i];
net/sched/sch_qfq.c +1 -10
index 9fbda3ec5861..1589364b54da 100644
@@ -217,15 +217,6 @@ static struct qfq_class *qfq_find_class(struct Qdisc *sch, u32 classid)
 	return container_of(clc, struct qfq_class, common);
 }
 
-static void qfq_purge_queue(struct qfq_class *cl)
-{
-	unsigned int len = cl->qdisc->q.qlen;
-	unsigned int backlog = cl->qdisc->qstats.backlog;
-
-	qdisc_reset(cl->qdisc);
-	qdisc_tree_reduce_backlog(cl->qdisc, len, backlog);
-}
-
 static const struct nla_policy qfq_policy[TCA_QFQ_MAX + 1] = {
 	[TCA_QFQ_WEIGHT] = { .type = NLA_U32 },
 	[TCA_QFQ_LMAX] = { .type = NLA_U32 },

@@ -551,7 +542,7 @@ static int qfq_delete_class(struct Qdisc *sch, unsigned long arg)
 
 	sch_tree_lock(sch);
 
-	qfq_purge_queue(cl);
+	qdisc_purge_queue(cl->qdisc);
 	qdisc_class_hash_remove(&q->clhash, &cl->common);
 
 	sch_tree_unlock(sch);
net/sched/sch_red.c +1 -2
index 9df9942340ea..4e8c0abf6194 100644
@@ -233,8 +233,7 @@ static int red_change(struct Qdisc *sch, struct nlattr *opt,
 	q->flags = ctl->flags;
 	q->limit = ctl->limit;
 	if (child) {
-		qdisc_tree_reduce_backlog(q->qdisc, q->qdisc->q.qlen,
-					  q->qdisc->qstats.backlog);
+		qdisc_tree_flush_backlog(q->qdisc);
 		old_child = q->qdisc;
 		q->qdisc = child;
 	}
net/sched/sch_sfb.c +1 -2
index bab506b01a32..2419fdb75966 100644
@@ -521,8 +521,7 @@ static int sfb_change(struct Qdisc *sch, struct nlattr *opt,
 		qdisc_hash_add(child, true);
 	sch_tree_lock(sch);
 
-	qdisc_tree_reduce_backlog(q->qdisc, q->qdisc->q.qlen,
-				  q->qdisc->qstats.backlog);
+	qdisc_tree_flush_backlog(q->qdisc);
 	qdisc_put(q->qdisc);
 	q->qdisc = child;
net/sched/sch_tbf.c +1 -2
index 7f272a9070c5..f71578dbb9e3 100644
@@ -391,8 +391,7 @@ static int tbf_change(struct Qdisc *sch, struct nlattr *opt,
 
 	sch_tree_lock(sch);
 	if (child) {
-		qdisc_tree_reduce_backlog(q->qdisc, q->qdisc->q.qlen,
-					  q->qdisc->qstats.backlog);
+		qdisc_tree_flush_backlog(q->qdisc);
 		qdisc_put(q->qdisc);
 		q->qdisc = child;
 	}
Signature · 1 line
2.20.1
maintainer
#3 lore raw
In reply to Paolo Abeni

From: Paolo Abeni <[email protected]>
Date: Thu, 28 Mar 2019 16:53:11 +0100

Quoted 8 lines

Currently, stats accounting for NOLOCK qdisc enslaved to classful (lock) qdiscs is buggy. Per CPU values are ignored in most places, as a result, stats dump in the above scenario always report 0 length backlog and parent backlog len is not updated correctly on NOLOCK qdisc removal.

The first patch address stats dumping, and the second one child qdisc removal. I'm targeting the net tree as this is a bugfix, but it could be moved to net-next due to the relatively large diffstat.

Series applied, thanks Paolo.