linux/net/sched
Toke Høiland-Jørgensen 549e407569 sched: sch_cake: fix bulk flow accounting logic for host fairness
commit 546ea84d07 upstream.

In sch_cake, we keep track of the count of active bulk flows per host,
when running in dst/src host fairness mode, which is used as the
round-robin weight when iterating through flows. The count of active
bulk flows is updated whenever a flow changes state.

This has a peculiar interaction with the hash collision handling: when a
hash collision occurs (after the set-associative hashing), the state of
the hash bucket is simply updated to match the new packet that collided,
and if host fairness is enabled, that also means assigning new per-host
state to the flow. For this reason, the bulk flow counters of the
host(s) assigned to the flow are decremented, before new state is
assigned (and the counters, which may not belong to the same host
anymore, are incremented again).

Back when this code was introduced, the host fairness mode was always
enabled, so the decrement was unconditional. When the configuration
flags were introduced the *increment* was made conditional, but
the *decrement* was not. Which of course can lead to a spurious
decrement (and associated wrap-around to U16_MAX).

AFAICT, when host fairness is disabled, the decrement and wrap-around
happens as soon as a hash collision occurs (which is not that common in
itself, due to the set-associative hashing). However, in most cases this
is harmless, as the value is only used when host fairness mode is
enabled. So in order to trigger an array overflow, sch_cake has to first
be configured with host fairness disabled, and while running in this
mode, a hash collision has to occur to cause the overflow. Then, the
qdisc has to be reconfigured to enable host fairness, which leads to the
array out-of-bounds because the wrapped-around value is retained and
used as an array index. It seems that syzbot managed to trigger this,
which is quite impressive in its own right.

This patch fixes the issue by introducing the same conditional check on
decrement as is used on increment.

The original bug predates the upstreaming of cake, but the commit listed
in the Fixes tag touched that code, meaning that this patch won't apply
before that.

Fixes: 7126399299 ("sch_cake: Make the dual modes fairer")
Reported-by: syzbot+7fe7b81d602cc1e6b94d@syzkaller.appspotmail.com
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/20240903160846.20909-1-toke@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
History 2024-09-12 11:10:18 +02:00
..
act_api.c net/sched: act_api: fix possible infinite loop in tcf_idr_check_alloc() 2024-06-27 13:46:19 +02:00
act_bpf.c net: sched: act_bpf: simplify code logic in tcf_bpf_init() 2022-09-28 09:38:56 +01:00
act_connmark.c netfilter: conntrack: Fix data-races around ct mark 2022-11-18 15:21:00 +01:00
act_csum.c net: sched: act_csum: get rid of tcf_csum_walker and tcf_csum_search 2022-09-09 08:24:41 +01:00
act_ct.c sched: act_ct: take care of padding in struct zones_ht_key 2024-08-11 12:35:56 +02:00
act_ctinfo.c net/sched: act_ctinfo: use percpu stats 2023-02-22 12:59:52 +01:00
act_gact.c treewide: use prandom_u32_max() when possible, part 1 2022-10-11 17:42:55 -06:00
act_gate.c net: sched: act_gate: get rid of tcf_gate_walker and tcf_gate_search 2022-09-09 08:24:42 +01:00
act_ife.c net: sched: act_ife: get rid of tcf_ife_walker and tcf_ife_search 2022-09-09 08:24:42 +01:00
act_ipt.c net/sched: act_ipt: add sanity checks on skb before calling target 2023-07-19 16:22:01 +02:00
act_meta_mark.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
act_meta_skbprio.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
act_meta_skbtcindex.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
act_mirred.c net/sched: act_mirred: Add carrier check 2023-05-17 11:53:34 +02:00
act_mpls.c net/sched: act_mpls: fix action bind logic 2023-03-11 13:55:28 +01:00
act_nat.c net: sched: act_nat: get rid of tcf_nat_walker and tcf_nat_search 2022-09-09 08:24:42 +01:00
act_pedit.c net/sched: act_pedit: Add size check for TCA_PEDIT_PARMS_EX 2023-07-19 16:22:05 +02:00
act_police.c net: sched: act_police: fix sparse errors in tcf_police_dump() 2023-06-14 11:15:21 +02:00
act_sample.c net/sched: act_sample: fix action bind logic 2023-03-11 13:55:28 +01:00
act_simple.c net: sched: act_simple: get rid of tcf_simp_walker and tcf_simp_search 2022-09-09 08:24:42 +01:00
act_skbedit.c net: sched: act_skbedit: get rid of tcf_skbedit_walker and tcf_skbedit_search 2022-09-09 08:24:42 +01:00
act_skbmod.c net/sched: act_skbmod: prevent kernel-infoleak 2024-04-10 16:28:26 +02:00
act_tunnel_key.c net: sched: act_tunnel_key: get rid of tunnel_key_walker and tunnel_key_search 2022-09-09 08:24:42 +01:00
act_vlan.c net: sched: act_vlan: get rid of tcf_vlan_walker and tcf_vlan_search 2022-09-09 08:24:43 +01:00
cls_api.c net/sched: cls_api: Fix lockup on flushing explicitly created chain 2023-06-21 16:01:01 +02:00
cls_basic.c net: sched: use tc_cls_bind_class() in filter 2022-10-02 16:07:17 +01:00
cls_bpf.c net: sched: cls_bpf: Undo tcf_bind_filter in case of an error 2023-07-27 08:50:41 +02:00
cls_cgroup.c net_sched: refactor TC action init API 2021-08-02 10:24:38 +01:00
cls_flow.c net/sched: use tc_cls_stats_dump() in filter 2022-09-20 15:54:13 -07:00
cls_flower.c net/sched: flower: Ensure both minimum and maximum ports are specified 2023-07-23 13:49:26 +02:00
cls_fw.c net/sched: cls_fw: No longer copy tcf_result on update to avoid use-after-free 2023-08-11 12:08:15 +02:00
cls_matchall.c net: sched: cls_matchall: Undo tcf_bind_filter in case of failure after mall_set_parms 2023-07-27 08:50:41 +02:00
cls_route.c net/sched: cls_route: No longer copy tcf_result on update to avoid use-after-free 2023-08-11 12:08:16 +02:00
cls_u32.c net: sched: cls_u32: Fix allocation size in u32_init() 2023-11-08 14:10:57 +01:00
em_canid.c net: sched: kerneldoc fixes 2020-07-13 17:20:40 -07:00
em_cmp.c net: sched: fix misspellings using misspell-fixer tool 2020-11-10 17:00:28 -08:00
em_ipset.c sched: consistently handle layer3 header accesses in the presence of VLANs 2020-07-03 14:34:53 -07:00
em_ipt.c sched: consistently handle layer3 header accesses in the presence of VLANs 2020-07-03 14:34:53 -07:00
em_meta.c net: annotate data-races around sk->sk_lingertime 2023-09-13 09:42:33 +02:00
em_nbyte.c net: sched: Return the correct errno code 2021-02-06 11:15:28 -08:00
em_text.c net: sched: em_text: fix possible memory leak in em_text_destroy() 2024-01-10 17:10:23 +01:00
em_u32.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
ematch.c net_sched: reject TCF_EM_SIMPLE case for complex ematch module 2022-12-31 13:32:55 +01:00
Kconfig net/sched: Retire dsmark qdisc 2024-03-01 13:26:24 +01:00
Makefile net/sched: Retire dsmark qdisc 2024-03-01 13:26:24 +01:00
sch_api.c net/sched: unregister lockdep keys in qdisc_create/qdisc_alloc error path 2024-06-27 13:46:24 +02:00
sch_blackhole.c Revert "net: sched: Pass root lock to Qdisc_ops.enqueue" 2020-07-16 16:48:34 -07:00
sch_cake.c sched: sch_cake: fix bulk flow accounting logic for host fairness 2024-09-12 11:10:18 +02:00
sch_cbs.c net/sched: use tc_qdisc_stats_dump() in qdisc 2022-09-22 17:34:10 -07:00
sch_choke.c net: sched: choke: remove unused variables in struct choke_sched_data 2022-08-31 19:39:53 -07:00
sch_codel.c net: sched: remove redundant NULL check in change hook function 2022-09-01 08:06:45 +02:00
sch_drr.c net/sched: use tc_qdisc_stats_dump() in qdisc 2022-09-22 17:34:10 -07:00
sch_etf.c net: sched: etf: remove true check in etf_enable_offload() 2022-09-01 20:08:32 -07:00
sch_ets.c net/sched: use tc_qdisc_stats_dump() in qdisc 2022-09-22 17:34:10 -07:00
sch_fifo.c net_sched: fix NULL deref in fifo_set_limit() 2021-10-01 14:59:10 -07:00
sch_fq.c net/sched: sch_fq: fix integer overflow of "credit" 2023-05-11 23:03:26 +09:00
sch_fq_codel.c Revert "net: sched: fq_codel: remove redundant resource cleanup in fq_codel_init()" 2022-10-19 13:47:09 +01:00
sch_fq_pie.c net/sched: fq_pie: avoid stalls in fq_pie_timer() 2023-09-19 12:28:00 +02:00
sch_frag.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next 2021-12-31 14:35:40 +00:00
sch_generic.c net: don't dump stack on queue timeout 2024-08-29 17:30:19 +02:00
sch_gred.c net: sched: gred: prevent races when adding offloads to stats 2023-02-01 08:34:25 +01:00
sch_hfsc.c net/sched: sch_hfsc: upgrade 'rt' to 'sc' when it becomes a inner curve 2023-10-25 12:03:07 +02:00
sch_hhf.c net: sched: remove redundant NULL check in change hook function 2022-09-01 08:06:45 +02:00
sch_htb.c net/sched: fix false lockdep warning on qdisc root lock 2024-06-27 13:46:15 +02:00
sch_ingress.c net/sched: Reserve TC_H_INGRESS (TC_H_CLSACT) for ingress (clsact) Qdiscs 2023-06-09 10:34:03 +02:00
sch_mq.c net: sched: add rcu annotations around qdisc->qdisc_sleeping 2023-06-14 11:15:21 +02:00
sch_mqprio.c net/sched: mqprio: Add length check for TCA_MQPRIO_{MAX/MIN}_RATE64 2023-08-03 10:24:02 +02:00
sch_multiq.c net: sched: sch_multiq: fix possible OOB write in multiq_tune() 2024-06-21 14:35:33 +02:00
sch_netem.c sch/netem: fix use after free in netem_dequeue 2024-09-12 11:10:16 +02:00
sch_pie.c net: sched: add rcu annotations around qdisc->qdisc_sleeping 2023-06-14 11:15:21 +02:00
sch_plug.c net: sched: sch_qfq: Fix UAF in qfq_dequeue() 2023-09-19 12:28:02 +02:00
sch_prio.c net/sched: use tc_qdisc_stats_dump() in qdisc 2022-09-22 17:34:10 -07:00
sch_qfq.c net: sched: sch_qfq: Fix UAF in qfq_dequeue() 2023-09-19 12:28:02 +02:00
sch_red.c net: sched: add rcu annotations around qdisc->qdisc_sleeping 2023-06-14 11:15:21 +02:00
sch_sfb.c Networking fixes for 6.1-rc2, including fixes from netfilter 2022-10-20 17:24:59 -07:00
sch_sfq.c net: sched: add rcu annotations around qdisc->qdisc_sleeping 2023-06-14 11:15:21 +02:00
sch_skbprio.c net/sched: use tc_qdisc_stats_dump() in qdisc 2022-09-22 17:34:10 -07:00
sch_taprio.c net/sched: taprio: always validate TCA_TAPRIO_ATTR_PRIOMAP 2024-06-21 14:35:35 +02:00
sch_tbf.c net/sched: use tc_qdisc_stats_dump() in qdisc 2022-09-22 17:34:10 -07:00
sch_teql.c net: sched: add rcu annotations around qdisc->qdisc_sleeping 2023-06-14 11:15:21 +02:00