netdev - [PATCH v2 net-next 14/14] net_sched: sch_skbprio: implement lockless skbprio

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Thu, 18 Apr 2024 07:32:48 +0000
From: Eric Dumazet <edumazet@...gle.com>
To: "David S . Miller" <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>, 
	Paolo Abeni <pabeni@...hat.com>
Cc: Jamal Hadi Salim <jhs@...atatu.com>, Simon Horman <horms@...nel.org>, 
	"Toke Høiland-Jørgensen" <toke@...hat.com>, Cong Wang <xiyou.wangcong@...il.com>, 
	Jiri Pirko <jiri@...nulli.us>, netdev@...r.kernel.org, eric.dumazet@...il.com, 
	Eric Dumazet <edumazet@...gle.com>
Subject: [PATCH v2 net-next 14/14] net_sched: sch_skbprio: implement lockless skbprio_dump()

Instead of relying on RTNL, skbprio_dump() can use READ_ONCE()
annotation, paired with WRITE_ONCE() one in skbprio_change().

Also add a READ_ONCE(sch->limit) in skbprio_enqueue().

Signed-off-by: Eric Dumazet <edumazet@...gle.com>
---
 net/sched/sch_skbprio.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/net/sched/sch_skbprio.c b/net/sched/sch_skbprio.c
index b4dd626c309c36725e6030a338d21d1fabcb6704..20ff7386b74bd89c00b50a8f0def91b6c5cce7f4 100644
--- a/net/sched/sch_skbprio.c
+++ b/net/sched/sch_skbprio.c
@@ -79,7 +79,9 @@ static int skbprio_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 	prio = min(skb->priority, max_priority);
 
 	qdisc = &q->qdiscs[prio];
-	if (sch->q.qlen < sch->limit) {
+
+	/* sch->limit can change under us from skbprio_change() */
+	if (sch->q.qlen < READ_ONCE(sch->limit)) {
 		__skb_queue_tail(qdisc, skb);
 		qdisc_qstats_backlog_inc(sch, skb);
 		q->qstats[prio].backlog += qdisc_pkt_len(skb);
@@ -172,7 +174,7 @@ static int skbprio_change(struct Qdisc *sch, struct nlattr *opt,
 	if (opt->nla_len != nla_attr_size(sizeof(*ctl)))
 		return -EINVAL;
 
-	sch->limit = ctl->limit;
+	WRITE_ONCE(sch->limit, ctl->limit);
 	return 0;
 }
 
@@ -200,7 +202,7 @@ static int skbprio_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	struct tc_skbprio_qopt opt;
 
-	opt.limit = sch->limit;
+	opt.limit = READ_ONCE(sch->limit);
 
 	if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt))
 		return -1;
-- 
2.44.0.683.g7961c838ac-goog