linux/net
William Liu 850226aef8 net/sched: Return NULL when htb_lookup_leaf encounters an empty rbtree
[ Upstream commit 0e1d5d9b5c ]

htb_lookup_leaf has a BUG_ON that can trigger with the following:

tc qdisc del dev lo root
tc qdisc add dev lo root handle 1: htb default 1
tc class add dev lo parent 1: classid 1:1 htb rate 64bit
tc qdisc add dev lo parent 1:1 handle 2: netem
tc qdisc add dev lo parent 2:1 handle 3: blackhole
ping -I lo -c1 -W0.001 127.0.0.1

The root cause is the following:

1. htb_dequeue calls htb_dequeue_tree which calls the dequeue handler on
   the selected leaf qdisc
2. netem_dequeue calls enqueue on the child qdisc
3. blackhole_enqueue drops the packet and returns a value that is not
   just NET_XMIT_SUCCESS
4. Because of this, netem_dequeue calls qdisc_tree_reduce_backlog, and
   since qlen is now 0, it calls htb_qlen_notify -> htb_deactivate ->
   htb_deactiviate_prios -> htb_remove_class_from_row -> htb_safe_rb_erase
5. As this is the only class in the selected hprio rbtree,
   __rb_change_child in __rb_erase_augmented sets the rb_root pointer to
   NULL
6. Because blackhole_dequeue returns NULL, netem_dequeue returns NULL,
   which causes htb_dequeue_tree to call htb_lookup_leaf with the same
   hprio rbtree, and fail the BUG_ON

The function graph for this scenario is shown here:
 0)               |  htb_enqueue() {
 0) + 13.635 us   |    netem_enqueue();
 0)   4.719 us    |    htb_activate_prios();
 0) # 2249.199 us |  }
 0)               |  htb_dequeue() {
 0)   2.355 us    |    htb_lookup_leaf();
 0)               |    netem_dequeue() {
 0) + 11.061 us   |      blackhole_enqueue();
 0)               |      qdisc_tree_reduce_backlog() {
 0)               |        qdisc_lookup_rcu() {
 0)   1.873 us    |          qdisc_match_from_root();
 0)   6.292 us    |        }
 0)   1.894 us    |        htb_search();
 0)               |        htb_qlen_notify() {
 0)   2.655 us    |          htb_deactivate_prios();
 0)   6.933 us    |        }
 0) + 25.227 us   |      }
 0)   1.983 us    |      blackhole_dequeue();
 0) + 86.553 us   |    }
 0) # 2932.761 us |    qdisc_warn_nonwc();
 0)               |    htb_lookup_leaf() {
 0)               |      BUG_ON();
 ------------------------------------------

The full original bug report can be seen here [1].

We can fix this just by returning NULL instead of the BUG_ON,
as htb_dequeue_tree returns NULL when htb_lookup_leaf returns
NULL.

[1] https://lore.kernel.org/netdev/pF5XOOIim0IuEfhI-SOxTgRvNoDwuux7UHKnE_Y5-zVd4wmGvNk2ceHjKb8ORnzw0cGwfmVu42g9dL7XyJLf1NEzaztboTWcm0Ogxuojoeo=@willsroot.io/

Fixes: 512bb43eb5 ("pkt_sched: sch_htb: Optimize WARN_ONs in htb_dequeue_tree() etc.")
Signed-off-by: William Liu <will@willsroot.io>
Signed-off-by: Savino Dicanosa <savy@syst3mfailure.io>
Link: https://patch.msgid.link/20250717022816.221364-1-will@willsroot.io
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
History 2025-08-28 16:24:02 +02:00
..
6lowpan 6lowpan: iphc: Fix an off-by-one check of array index 2021-07-22 16:19:03 +02:00
9p 9p/xen: fix release of IRQ 2024-12-14 19:51:21 +01:00
802 net: 802: LLC+SNAP OID:PID lookup on start of skb data 2025-01-23 17:15:49 +01:00
8021q net: vlan: fix VLAN 0 refcount imbalance of toggling filtering during runtime 2025-08-28 16:24:02 +02:00
appletalk net: appletalk: Fix device refcount leak in atrtr_create() 2025-07-17 18:30:53 +02:00
atm atm: clip: Fix NULL pointer dereference in vcc_sendmsg() 2025-07-17 18:30:47 +02:00
ax25 net: ax25: Fix deadlock caused by skb_recv_datagram in ax25_recvmsg 2022-06-22 14:22:01 +02:00
batman-adv batman-adv: Ignore own maximum aggregation size during RX 2025-04-10 14:31:55 +02:00
bluetooth Bluetooth: L2CAP: Fix attempting to adjust outgoing MTU 2025-08-28 16:24:02 +02:00
bpf bpf: Set run context for rawtp test_run callback 2024-07-05 09:14:06 +02:00
bpfilter bpfilter: Specify the log level for the kmsg message 2021-06-25 13:13:50 +02:00
bridge net: bridge: Do not offload IGMP/MLD messages 2025-08-28 16:24:02 +02:00
caif net: caif: Fix use-after-free in cfusbl_device_notify() 2023-03-17 08:48:54 +01:00
can can: bcm: add missing rcu read protection for procfs content 2025-06-04 14:38:06 +02:00
ceph libceph: fix race between delayed_work() and ceph_monc_stop() 2024-07-18 13:07:42 +02:00
core bpf, sockmap: Fix skb refcnt race after locking changes 2025-07-17 18:30:48 +02:00
dcb net: dcb: choose correct policy to parse DCB_ATTR_BCN 2023-08-11 15:13:53 +02:00
dccp net: fix data-races around sk->sk_forward_alloc 2025-01-23 17:16:04 +01:00
dns_resolver keys, dns: Fix size check of V1 server-list header 2024-01-25 14:52:46 -08:00
dsa net: dsa: tag_brcm: legacy: fix pskb_may_pull length 2025-06-27 11:05:17 +01:00
ethernet gro: remove rcu_read_lock/rcu_read_unlock from gro_complete handlers 2024-09-12 11:07:47 +02:00
ethtool net: ethtool: Don't call .cleanup_data when prepare_data fails 2025-05-02 07:43:52 +02:00
hsr net: hsr: fix fill_frame_info() regression vs VLAN packets 2025-03-13 12:50:13 +01:00
ieee802154 net: ieee802154: do not leave a dangling sk pointer in ieee802154_create() 2024-12-14 19:51:37 +01:00
ife net: sched: ife: fix potential use-after-free 2024-01-05 15:13:29 +01:00
ipv4 tcp: fix tcp_packet_delayed() for tcp_is_non_sack_preventing_reopen() behavior 2025-06-27 11:05:36 +01:00
ipv6 ipv6: mcast: Delay put pmc->idev in mld_del_delrec() 2025-08-28 16:24:02 +02:00
iucv s390/iucv: fix receive buffer virtual vs physical address confusion 2024-09-04 13:23:27 +02:00
kcm kcm: Serialise kcm_sendmsg() for the same socket. 2024-09-04 13:23:32 +02:00
key net: af_key: fix sadb_x_filter validation 2023-08-26 14:23:32 +02:00
l2tp genetlink: hold RCU in genlmsg_mcast() 2024-11-01 01:52:34 +01:00
l3mdev net: Add l3mdev index to flow struct and avoid oif reset for port devices 2024-10-17 15:11:57 +02:00
lapb net: lapb: Use list_for_each_entry() to simplify code in lapb_iface.c 2021-06-08 16:31:25 -07:00
llc llc: fix data loss when reading from a socket in llc_ui_recvmsg() 2025-06-04 14:38:06 +02:00
mac80211 wifi: mac80211: drop invalid source address OCB frames 2025-07-10 15:57:47 +02:00
mac802154 mac802154: check local interfaces before deleting sdata list 2025-01-23 17:16:00 +01:00
mctp net: mctp: Set SOCK_RCU_FREE 2025-05-02 07:44:10 +02:00
mpls mpls: Use rcu_dereference_rtnl() in mpls_route_input_rcu(). 2025-06-27 11:05:36 +01:00
mptcp net: make sock_inuse_add() available 2025-05-02 07:44:20 +02:00
ncsi net: ncsi: Fix GCPS 64-bit member variables 2025-06-27 11:05:12 +01:00
netfilter netfilter: nf_set_pipapo_avx2: fix initial map fill 2025-06-27 11:05:18 +01:00
netlabel calipso: unlock rcu before returning -EAFNOSUPPORT 2025-06-27 11:05:22 +01:00
netlink netlink: make sure we allow at least one dump skb 2025-07-17 18:30:50 +02:00
netrom netrom: check buffer length before accessing it 2025-01-09 13:28:46 +01:00
nfc NFC: nci: uart: Set tty->disc_data only in success path 2025-06-27 11:05:25 +01:00
nsh nsh: Restore skb->{protocol,data,mac_header} for outer header in nsh_gso_segment(). 2024-05-17 11:50:48 +02:00
openvswitch net: openvswitch: Fix the dead loop of MPLS parse 2025-06-27 11:05:14 +01:00
packet af_packet: fix soft lockup issue caused by tpacket_snd() 2025-08-28 16:23:59 +02:00
phonet phonet/pep: Move call to pn_skb_get_dst_sockaddr() earlier in pep_sock_accept() 2025-08-28 16:23:59 +02:00
psample psample: Require 'CAP_NET_ADMIN' when joining "packets" group 2023-12-13 18:36:37 +01:00
qrtr net: qrtr: Update packets cloning when broadcasting 2024-10-17 15:11:08 +02:00
rds net:rds: Fix possible deadlock in rds_message_put 2024-09-04 13:23:39 +02:00
rfkill net: rfkill: gpio: Add check for clk_enable() 2024-12-14 19:50:58 +01:00
rose rose: fix dangling neighbour pointers in rose_rt_device_down() 2025-07-10 15:57:46 +02:00
rxrpc rxrpc: Fix oops due to non-existence of prealloc backlog struct 2025-07-17 18:30:48 +02:00
sched net/sched: Return NULL when htb_lookup_leaf encounters an empty rbtree 2025-08-28 16:24:02 +02:00
sctp sctp: Do not wake readers in __sctp_write_space() 2025-06-27 11:05:31 +01:00
smc net/smc: fix data error when recvmsg with MSG_PEEK flag 2025-03-13 12:49:57 +01:00
strparser bpf: sockmap, strparser, and tls are reusing qdisc_skb_cb and colliding 2021-11-18 19:17:11 +01:00
sunrpc sunrpc: fix race in cache cleanup causing stale nextcheck time 2025-06-27 11:05:29 +01:00
switchdev net: make switchdev_bridge_port_{,unoffload} loosely coupled with the bridge 2021-08-04 12:35:07 +01:00
tipc tipc: Fix use-after-free in tipc_conn_close(). 2025-07-17 18:30:46 +02:00
tls ktls, sockmap: Fix missing uncharge operation 2025-06-27 11:05:13 +01:00
unix af_unix: Don't set -ECONNRESET for consumed OOB skb. 2025-07-10 15:57:38 +02:00
vmw_vsock vsock: Fix IOCTL_VM_SOCKETS_GET_LOCAL_CID to check also transport_local 2025-07-17 18:30:46 +02:00
wireless wifi: nl80211: reject cooked mode if it is set along with other flags 2025-03-13 12:51:03 +01:00
x25 net/x25: fix incorrect parameter validation in the x25_getsockopt() function 2024-03-26 18:21:23 -04:00
xdp xsk: fix OOB map writes when deleting elements 2024-12-14 19:51:33 +01:00
xfrm xfrm: Sanitize marks before insert 2025-06-04 14:38:05 +02:00
compat.c net: Return the correct errno code 2021-06-03 15:13:56 -07:00
devres.c net: devres: Correct a grammatical error 2021-06-11 12:55:28 -07:00
Kconfig Remove DECnet support from kernel 2023-06-21 15:59:15 +02:00
Makefile Remove DECnet support from kernel 2023-06-21 15:59:15 +02:00
socket.c net: explicitly clear the sk pointer, when pf->create fails 2024-10-17 15:12:00 +02:00
sysctl_net.c net: Ensure net namespace isolation of sysctls 2021-04-12 13:27:11 -07:00