From: "Toke Høiland-Jørgensen" <toke@toke.dk>
To: netdev@vger.kernel.org
Cc: cake@lists.bufferbloat.net,
"Toke Høiland-Jørgensen" <toke@redhat.com>,
"Pete Heist" <pete@heistp.net>
Subject: [Cake] [PATCH 2/4] sched: Fix detection of empty queues in child qdiscs
Date: Mon, 7 Jan 2019 20:47:31 +0100 [thread overview]
Message-ID: <20190107194733.31138-3-toke@toke.dk> (raw)
In-Reply-To: <20190107194733.31138-1-toke@toke.dk>
From: Toke Høiland-Jørgensen <toke@redhat.com>
Several qdiscs check on enqueue whether the packet was enqueued to a class
with an empty queue, in which case the class is activated. This is done by
checking if the qlen is exactly 1 after enqueue. However, if GSO splitting
is enabled in the child qdisc, a single packet can result in a qlen longer
than 1. This means the activation check fails, leading to a stalled queue.
Fix this by checking if the queue is empty *before* enqueue, and running
the activation logic if this was the case.
Reported-by: Pete Heist <pete@heistp.net>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
---
net/sched/sch_drr.c | 4 +++-
net/sched/sch_hfsc.c | 4 +++-
net/sched/sch_qfq.c | 4 +++-
3 files changed, 9 insertions(+), 3 deletions(-)
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index feaf47178653..09b800991065 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -354,6 +354,7 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch,
struct drr_sched *q = qdisc_priv(sch);
struct drr_class *cl;
int err = 0;
+ bool first;
cl = drr_classify(skb, sch, &err);
if (cl == NULL) {
@@ -363,6 +364,7 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch,
return err;
}
+ first = !cl->qdisc->q.qlen;
err = qdisc_enqueue(skb, cl->qdisc, to_free);
if (unlikely(err != NET_XMIT_SUCCESS)) {
if (net_xmit_drop_count(err)) {
@@ -372,7 +374,7 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch,
return err;
}
- if (cl->qdisc->q.qlen == 1) {
+ if (first) {
list_add_tail(&cl->alist, &q->active);
cl->deficit = cl->quantum;
}
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 6bb8f73a8473..24cc220a3218 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1542,6 +1542,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free)
unsigned int len = qdisc_pkt_len(skb);
struct hfsc_class *cl;
int uninitialized_var(err);
+ bool first;
cl = hfsc_classify(skb, sch, &err);
if (cl == NULL) {
@@ -1551,6 +1552,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free)
return err;
}
+ first = !cl->qdisc->q.qlen;
err = qdisc_enqueue(skb, cl->qdisc, to_free);
if (unlikely(err != NET_XMIT_SUCCESS)) {
if (net_xmit_drop_count(err)) {
@@ -1560,7 +1562,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free)
return err;
}
- if (cl->qdisc->q.qlen == 1) {
+ if (first) {
if (cl->cl_flags & HFSC_RSC)
init_ed(cl, len);
if (cl->cl_flags & HFSC_FSC)
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index 8d5e55d5bed2..29f5c4a24688 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -1215,6 +1215,7 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch,
struct qfq_class *cl;
struct qfq_aggregate *agg;
int err = 0;
+ bool first;
cl = qfq_classify(skb, sch, &err);
if (cl == NULL) {
@@ -1236,6 +1237,7 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch,
}
gso_segs = skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1;
+ first = !cl->qdisc->q.qlen;
err = qdisc_enqueue(skb, cl->qdisc, to_free);
if (unlikely(err != NET_XMIT_SUCCESS)) {
pr_debug("qfq_enqueue: enqueue failed %d\n", err);
@@ -1253,7 +1255,7 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch,
agg = cl->agg;
/* if the queue was not empty, then done here */
- if (cl->qdisc->q.qlen != 1) {
+ if (!first) {
if (unlikely(skb == cl->qdisc->ops->peek(cl->qdisc)) &&
list_first_entry(&agg->active, struct qfq_class, alist)
== cl && cl->deficit < len)
--
2.20.1
next prev parent reply other threads:[~2019-01-07 19:48 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-07 19:47 [Cake] [PATCH 0/4] sched: Fix qdisc interactions exposed by using sch_cake as a leaf qdisc Toke Høiland-Jørgensen
2019-01-07 19:47 ` [Cake] [PATCH 1/4] sched: Avoid dereferencing skb pointer after child enqueue Toke Høiland-Jørgensen
2019-01-09 1:38 ` Cong Wang
2019-01-09 8:14 ` Toke Høiland-Jørgensen
2019-01-10 18:16 ` Cong Wang
2019-01-07 19:47 ` Toke Høiland-Jørgensen [this message]
2019-01-07 19:57 ` [Cake] [PATCH 2/4] sched: Fix detection of empty queues in child qdiscs Dave Taht
2019-01-07 19:47 ` [Cake] [PATCH 3/4] sch_api: Allow reducing queue backlog by a negative value Toke Høiland-Jørgensen
2019-01-07 20:54 ` Eric Dumazet
2019-01-08 9:55 ` Toke Høiland-Jørgensen
2019-01-07 19:47 ` [Cake] [PATCH 4/4] sch_cake: Correctly update parent qlen when splitting GSO packets Toke Høiland-Jørgensen
2019-01-08 21:30 ` [Cake] [PATCH 0/4] sched: Fix qdisc interactions exposed by using sch_cake as a leaf qdisc David Miller
2019-01-09 8:37 ` Toke Høiland-Jørgensen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://lists.bufferbloat.net/postorius/lists/cake.lists.bufferbloat.net/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190107194733.31138-3-toke@toke.dk \
--to=toke@toke.dk \
--cc=cake@lists.bufferbloat.net \
--cc=netdev@vger.kernel.org \
--cc=pete@heistp.net \
--cc=toke@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox