From d35acc1be3480505b5931f17e4ea9b7617fea4d3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 11 Jun 2025 11:15:11 +0000 Subject: [PATCH 1/5] net_sched: prio: fix a race in prio_tune() Gerrard Tai reported a race condition in PRIO, whenever SFQ perturb timer fires at the wrong time. The race is as follows: CPU 0 CPU 1 [1]: lock root [2]: qdisc_tree_flush_backlog() [3]: unlock root | | [5]: lock root | [6]: rehash | [7]: qdisc_tree_reduce_backlog() | [4]: qdisc_put() This can be abused to underflow a parent's qlen. Calling qdisc_purge_queue() instead of qdisc_tree_flush_backlog() should fix the race, because all packets will be purged from the qdisc before releasing the lock. Fixes: 7b8e0b6e6599 ("net: sched: prio: delay destroying child qdiscs on change") Reported-by: Gerrard Tai Suggested-by: Gerrard Tai Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20250611111515.1983366-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_prio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index cc30f7a32f1a..9e2b9a490db2 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -211,7 +211,7 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt, memcpy(q->prio2band, qopt->priomap, TC_PRIO_MAX+1); for (i = q->bands; i < oldbands; i++) - qdisc_tree_flush_backlog(q->queues[i]); + qdisc_purge_queue(q->queues[i]); for (i = oldbands; i < q->bands; i++) { q->queues[i] = queues[i]; From 85a3e0ede38450ea3053b8c45d28cf55208409b8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 11 Jun 2025 11:15:12 +0000 Subject: [PATCH 2/5] net_sched: red: fix a race in __red_change() Gerrard Tai reported a race condition in RED, whenever SFQ perturb timer fires at the wrong time. The race is as follows: CPU 0 CPU 1 [1]: lock root [2]: qdisc_tree_flush_backlog() [3]: unlock root | | [5]: lock root | [6]: rehash | [7]: qdisc_tree_reduce_backlog() | [4]: qdisc_put() This can be abused to underflow a parent's qlen. Calling qdisc_purge_queue() instead of qdisc_tree_flush_backlog() should fix the race, because all packets will be purged from the qdisc before releasing the lock. Fixes: 0c8d13ac9607 ("net: sched: red: delay destroying child qdisc on replace") Reported-by: Gerrard Tai Suggested-by: Gerrard Tai Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20250611111515.1983366-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_red.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c index 1ba3e0bba54f..4696c893cf55 100644 --- a/net/sched/sch_red.c +++ b/net/sched/sch_red.c @@ -285,7 +285,7 @@ static int __red_change(struct Qdisc *sch, struct nlattr **tb, q->userbits = userbits; q->limit = ctl->limit; if (child) { - qdisc_tree_flush_backlog(q->qdisc); + qdisc_purge_queue(q->qdisc); old_child = q->qdisc; q->qdisc = child; } From 43eb466041216d25dedaef1c383ad7bd89929cbc Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 11 Jun 2025 11:15:13 +0000 Subject: [PATCH 3/5] net_sched: tbf: fix a race in tbf_change() Gerrard Tai reported a race condition in TBF, whenever SFQ perturb timer fires at the wrong time. The race is as follows: CPU 0 CPU 1 [1]: lock root [2]: qdisc_tree_flush_backlog() [3]: unlock root | | [5]: lock root | [6]: rehash | [7]: qdisc_tree_reduce_backlog() | [4]: qdisc_put() This can be abused to underflow a parent's qlen. Calling qdisc_purge_queue() instead of qdisc_tree_flush_backlog() should fix the race, because all packets will be purged from the qdisc before releasing the lock. Fixes: b05972f01e7d ("net: sched: tbf: don't call qdisc_put() while holding tree lock") Reported-by: Gerrard Tai Suggested-by: Gerrard Tai Signed-off-by: Eric Dumazet Cc: Zhengchao Shao Link: https://patch.msgid.link/20250611111515.1983366-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_tbf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index dc26b22d53c7..4c977f049670 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -452,7 +452,7 @@ static int tbf_change(struct Qdisc *sch, struct nlattr *opt, sch_tree_lock(sch); if (child) { - qdisc_tree_flush_backlog(q->qdisc); + qdisc_purge_queue(q->qdisc); old = q->qdisc; q->qdisc = child; } From d92adacdd8c2960be856e0b82acc5b7c5395fddb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 11 Jun 2025 11:15:14 +0000 Subject: [PATCH 4/5] net_sched: ets: fix a race in ets_qdisc_change() Gerrard Tai reported a race condition in ETS, whenever SFQ perturb timer fires at the wrong time. The race is as follows: CPU 0 CPU 1 [1]: lock root [2]: qdisc_tree_flush_backlog() [3]: unlock root | | [5]: lock root | [6]: rehash | [7]: qdisc_tree_reduce_backlog() | [4]: qdisc_put() This can be abused to underflow a parent's qlen. Calling qdisc_purge_queue() instead of qdisc_tree_flush_backlog() should fix the race, because all packets will be purged from the qdisc before releasing the lock. Fixes: b05972f01e7d ("net: sched: tbf: don't call qdisc_put() while holding tree lock") Reported-by: Gerrard Tai Suggested-by: Gerrard Tai Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20250611111515.1983366-5-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_ets.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/sch_ets.c b/net/sched/sch_ets.c index 2c069f0181c6..037f764822b9 100644 --- a/net/sched/sch_ets.c +++ b/net/sched/sch_ets.c @@ -661,7 +661,7 @@ static int ets_qdisc_change(struct Qdisc *sch, struct nlattr *opt, for (i = q->nbands; i < oldbands; i++) { if (i >= q->nstrict && q->classes[i].qdisc->q.qlen) list_del_init(&q->classes[i].alist); - qdisc_tree_flush_backlog(q->classes[i].qdisc); + qdisc_purge_queue(q->classes[i].qdisc); } WRITE_ONCE(q->nstrict, nstrict); memcpy(q->prio2band, priomap, sizeof(priomap)); From adcaa890c7a4a91a422168d8fb629183fff07b2f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 11 Jun 2025 11:15:15 +0000 Subject: [PATCH 5/5] net_sched: remove qdisc_tree_flush_backlog() This function is no longer used after the four prior fixes. Given all prior uses were wrong, it seems better to remove it. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20250611111515.1983366-6-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/sch_generic.h | 8 -------- 1 file changed, 8 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 629368ab2787..638948be4c50 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -973,14 +973,6 @@ 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;