Netdev List
 help / color / mirror / Atom feed
From: Eric Dumazet <edumazet@google.com>
To: "David S . Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>,
	 Paolo Abeni <pabeni@redhat.com>
Cc: "Simon Horman" <horms@kernel.org>,
	"Jamal Hadi Salim" <jhs@mojatatu.com>,
	"Victor Nogueira" <victor@mojatatu.com>,
	"Jiri Pirko" <jiri@resnulli.us>,
	"Ido Schimmel" <idosch@nvidia.com>,
	"David Ahern" <dsahern@kernel.org>,
	"Toke Høiland-Jørgensen" <toke@toke.dk>,
	netdev@vger.kernel.org, eric.dumazet@gmail.com,
	"Eric Dumazet" <edumazet@google.com>
Subject: [PATCH net-next 3/3] net/sched: sch_hfsc: annotate data-races in hfsc_dump_class_stats()
Date: Wed, 13 May 2026 08:08:53 +0000	[thread overview]
Message-ID: <20260513080853.1383975-4-edumazet@google.com> (raw)
In-Reply-To: <20260513080853.1383975-1-edumazet@google.com>

hfsc_dump_class_stats() runs without qdisc spinlock being held.

Add READ_ONCE()/WRITE_ONCE() annotations around:

- cl->level
- cl->cl_vtperiod
- cl->cl_total
- cl->cl_cumul

Fixes: edb09eb17ed8 ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump")
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 net/sched/sch_hfsc.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 59409ee2d2ff9279d7439b744030c0e845386de0..c06d2761a1fb9dbc711b18d61cfa8c28a7de0bd8 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -715,7 +715,7 @@ init_vf(struct hfsc_class *cl, unsigned int len)
 			rtsc_min(&cl->cl_virtual, &cl->cl_fsc, cl->cl_vt, cl->cl_total);
 			cl->cl_vtadj = 0;
 
-			cl->cl_vtperiod++;  /* increment vt period */
+			WRITE_ONCE(cl->cl_vtperiod, cl->cl_vtperiod + 1);  /* increment vt period */
 			cl->cl_parentperiod = cl->cl_parent->cl_vtperiod;
 			if (cl->cl_parent->cl_nactive == 0)
 				cl->cl_parentperiod++;
@@ -757,7 +757,7 @@ update_vf(struct hfsc_class *cl, unsigned int len, u64 cur_time)
 		go_passive = 1;
 
 	for (; cl->cl_parent != NULL; cl = cl->cl_parent) {
-		cl->cl_total += len;
+		WRITE_ONCE(cl->cl_total, cl->cl_total + len);
 
 		if (!(cl->cl_flags & HFSC_FSC) || cl->cl_nactive == 0)
 			continue;
@@ -847,7 +847,7 @@ hfsc_adjust_levels(struct hfsc_class *cl)
 			if (p->level >= level)
 				level = p->level + 1;
 		}
-		cl->level = level;
+		WRITE_ONCE(cl->level, level);
 	} while ((cl = cl->cl_parent) != NULL);
 }
 
@@ -1338,10 +1338,10 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
 	__u32 qlen;
 
 	qdisc_qstats_qlen_backlog(cl->qdisc, &qlen, &cl->qstats.backlog);
-	xstats.level   = cl->level;
-	xstats.period  = cl->cl_vtperiod;
-	xstats.work    = cl->cl_total;
-	xstats.rtwork  = cl->cl_cumul;
+	xstats.level   = READ_ONCE(cl->level);
+	xstats.period  = READ_ONCE(cl->cl_vtperiod);
+	xstats.work    = READ_ONCE(cl->cl_total);
+	xstats.rtwork  = READ_ONCE(cl->cl_cumul);
 
 	if (gnet_stats_copy_basic(d, NULL, &cl->bstats, true) < 0 ||
 	    gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 ||
@@ -1452,15 +1452,15 @@ hfsc_change_qdisc(struct Qdisc *sch, struct nlattr *opt,
 static void
 hfsc_reset_class(struct hfsc_class *cl)
 {
-	cl->cl_total        = 0;
-	cl->cl_cumul        = 0;
+	WRITE_ONCE(cl->cl_total, 0);
+	WRITE_ONCE(cl->cl_cumul, 0);
 	cl->cl_d            = 0;
 	cl->cl_e            = 0;
 	cl->cl_vt           = 0;
 	cl->cl_vtadj        = 0;
 	cl->cl_cvtmin       = 0;
 	cl->cl_cvtoff       = 0;
-	cl->cl_vtperiod     = 0;
+	WRITE_ONCE(cl->cl_vtperiod, 0);
 	cl->cl_parentperiod = 0;
 	cl->cl_f            = 0;
 	cl->cl_myf          = 0;
@@ -1626,7 +1626,7 @@ hfsc_dequeue(struct Qdisc *sch)
 	bstats_update(&cl->bstats, skb);
 	update_vf(cl, qdisc_pkt_len(skb), cur_time);
 	if (realtime)
-		cl->cl_cumul += qdisc_pkt_len(skb);
+		WRITE_ONCE(cl->cl_cumul, cl->cl_cumul + qdisc_pkt_len(skb));
 
 	if (cl->cl_flags & HFSC_RSC) {
 		if (cl->qdisc->q.qlen != 0) {
-- 
2.54.0.563.g4f69b47b94-goog


  parent reply	other threads:[~2026-05-13  8:09 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-13  8:08 [PATCH net-next 0/3] net/sched: changes around qdisc_qstats_qlen_backlog() Eric Dumazet
2026-05-13  8:08 ` [PATCH net-next 1/3] net/sched: qdisc_qstats_qlen_backlog() runs locklessly Eric Dumazet
2026-05-13  8:08 ` [PATCH net-next 2/3] net: ioam6: no longer acquire qdisc spinlock while calling qdisc_qstats_qlen_backlog() Eric Dumazet
2026-05-13  8:08 ` Eric Dumazet [this message]
2026-05-13 10:08 ` [PATCH net-next 0/3] net/sched: changes around qdisc_qstats_qlen_backlog() 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

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260513080853.1383975-4-edumazet@google.com \
    --to=edumazet@google.com \
    --cc=davem@davemloft.net \
    --cc=dsahern@kernel.org \
    --cc=eric.dumazet@gmail.com \
    --cc=horms@kernel.org \
    --cc=idosch@nvidia.com \
    --cc=jhs@mojatatu.com \
    --cc=jiri@resnulli.us \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=toke@toke.dk \
    --cc=victor@mojatatu.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