kernel/linux.git/net/ipv4/udp.c, branch v6.18.22

udp: Fix wildcard bind conflict check when using hash2

2026-04-02T11:23:07+00:00

[ Upstream commit e537dd15d0d4ad989d56a1021290f0c674dd8b28 ] When binding a udp_sock to a local address and port, UDP uses two hashes (udptable->hash and udptable->hash2) for collision detection. The current code switches to "hash2" when hslot->count > 10. "hash2" is keyed by local address and local port. "hash" is keyed by local port only. The issue can be shown in the following bind sequence (pseudo code): bind(fd1, "[fd00::1]:8888") bind(fd2, "[fd00::2]:8888") bind(fd3, "[fd00::3]:8888") bind(fd4, "[fd00::4]:8888") bind(fd5, "[fd00::5]:8888") bind(fd6, "[fd00::6]:8888") bind(fd7, "[fd00::7]:8888") bind(fd8, "[fd00::8]:8888") bind(fd9, "[fd00::9]:8888") bind(fd10, "[fd00::10]:8888") /* Correctly return -EADDRINUSE because "hash" is used * instead of "hash2". udp_lib_lport_inuse() detects the * conflict. */ bind(fail_fd, "[::]:8888") /* After one more socket is bound to "[fd00::11]:8888", * hslot->count exceeds 10 and "hash2" is used instead. */ bind(fd11, "[fd00::11]:8888") bind(fail_fd, "[::]:8888") /* succeeds unexpectedly */ The same issue applies to the IPv4 wildcard address "0.0.0.0" and the IPv4-mapped wildcard address "::ffff:0.0.0.0". For example, if there are existing sockets bound to "192.168.1.[1-11]:8888", then binding "0.0.0.0:8888" or "[::ffff:0.0.0.0]:8888" can also miss the conflict when hslot->count > 10. TCP inet_csk_get_port() already has the correct check in inet_use_bhash2_on_bind(). Rename it to inet_use_hash2_on_bind() and move it to inet_hashtables.h so udp.c can reuse it in this fix. Fixes: 30fff9231fad ("udp: bind() optimisation") Reported-by: Andrew Onyshchuk Signed-off-by: Martin KaFai Lau Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260319181817.1901357-1-martin.lau@linux.dev Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

udp: Unhash auto-bound connected sk from 4-tuple hash table when disconnected.

2026-03-12T11:09:50+00:00

[ Upstream commit 6996a2d2d0a64808c19c98002aeb5d9d1b2df6a4 ] Let's say we bind() an UDP socket to the wildcard address with a non-zero port, connect() it to an address, and disconnect it from the address. bind() sets SOCK_BINDPORT_LOCK on sk->sk_userlocks (but not SOCK_BINDADDR_LOCK), and connect() calls udp_lib_hash4() to put the socket into the 4-tuple hash table. Then, __udp_disconnect() calls sk->sk_prot->rehash(sk). It computes a new hash based on the wildcard address and moves the socket to a new slot in the 4-tuple hash table, leaving a garbage in the chain that no packet hits. Let's remove such a socket from 4-tuple hash table when disconnected. Note that udp_sk(sk)->udp_portaddr_hash needs to be udpated after udp_hash4_dec(hslot2) in udp_unhash4(). Fixes: 78c91ae2c6de ("ipv4/udp: Add 4-tuple hash for connected socket") Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20260227035547.3321327-1-kuniyu@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

net: annotate data-races around sk->sk_{data_ready,write_space}

2026-03-12T11:09:49+00:00

[ Upstream commit 2ef2b20cf4e04ac8a6ba68493f8780776ff84300 ] skmsg (and probably other layers) are changing these pointers while other cpus might read them concurrently. Add corresponding READ_ONCE()/WRITE_ONCE() annotations for UDP, TCP and AF_UNIX. Fixes: 604326b41a6f ("bpf, sockmap: convert to generic sk_msg interface") Reported-by: syzbot+87f770387a9e5dc6b79b@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/699ee9fc.050a0220.1cd54b.0009.GAE@google.com/ Signed-off-by: Eric Dumazet Cc: Daniel Borkmann Cc: John Fastabend Cc: Jakub Sitnicki Cc: Willem de Bruijn Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260225131547.1085509-1-edumazet@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

udp: call skb_orphan() before skb_attempt_defer_free()

2026-01-17T15:35:28+00:00

[ Upstream commit e5c8eda39a9fc1547d1398d707aa06c1d080abdd ] Standard UDP receive path does not use skb->destructor. But skmsg layer does use it, since it calls skb_set_owner_sk_safe() from udp_read_skb(). This then triggers this warning in skb_attempt_defer_free(): DEBUG_NET_WARN_ON_ONCE(skb->destructor); We must call skb_orphan() to fix this issue. Fixes: 6471658dc66c ("udp: use skb_attempt_defer_free()") Reported-by: syzbot+3e68572cf2286ce5ebe9@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/695b83bd.050a0220.1c9965.002b.GAE@google.com/T/#u Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20260105093630.1976085-1-edumazet@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

udp: do not use skb_release_head_state() before skb_attempt_defer_free()

2025-10-16T14:03:07+00:00

Michal reported and bisected an issue after recent adoption of skb_attempt_defer_free() in UDP. The issue here is that skb_release_head_state() is called twice per skb, one time from skb_consume_udp(), then a second time from skb_defer_free_flush() and napi_consume_skb(). As Sabrina suggested, remove skb_release_head_state() call from skb_consume_udp(). Add a DEBUG_NET_WARN_ON_ONCE(skb_nfct(skb)) in skb_attempt_defer_free() Many thanks to Michal, Sabrina, Paolo and Florian for their help. Fixes: 6471658dc66c ("udp: use skb_attempt_defer_free()") Reported-and-bisected-by: Michal Kubecek Closes: https://lore.kernel.org/netdev/gpjh4lrotyephiqpuldtxxizrsg6job7cvhiqrw72saz2ubs3h@g6fgbvexgl3r/ Signed-off-by: Eric Dumazet Tested-by: Michal Kubecek Cc: Sabrina Dubroca Cc: Florian Westphal Reviewed-by: Sabrina Dubroca Link: https://patch.msgid.link/20251015052715.4140493-1-edumazet@google.com Signed-off-by: Paolo Abeni

udp: remove busylock and add per NUMA queues

2025-09-23T23:38:39+00:00

busylock was protecting UDP sockets against packet floods, but unfortunately was not protecting the host itself. Under stress, many cpus could spin while acquiring the busylock, and NIC had to drop packets. Or packets would be dropped in cpu backlog if RPS/RFS were in place. This patch replaces the busylock by intermediate lockless queues. (One queue per NUMA node). This means that fewer number of cpus have to acquire the UDP receive queue lock. Most of the cpus can either: - immediately drop the packet. - or queue it in their NUMA aware lockless queue. Then one of the cpu is chosen to process this lockless queue in a batch. The batch only contains packets that were cooked on the same NUMA node, thus with very limited latency impact. Tested: DDOS targeting a victim UDP socket, on a platform with 6 NUMA nodes (Intel(R) Xeon(R) 6985P-C) Before: nstat -n ; sleep 1 ; nstat | grep Udp Udp6InDatagrams 1004179 0.0 Udp6InErrors 3117 0.0 Udp6RcvbufErrors 3117 0.0 After: nstat -n ; sleep 1 ; nstat | grep Udp Udp6InDatagrams 1116633 0.0 Udp6InErrors 14197275 0.0 Udp6RcvbufErrors 14197275 0.0 We can see this host can now proces 14.2 M more packets per second while under attack, and the victim socket can receive 11 % more packets. I used a small bpftrace program measuring time (in us) spent in __udp_enqueue_schedule_skb(). Before: @udp_enqueue_us[398]: [0] 24901 |@@@ | [1] 63512 |@@@@@@@@@ | [2, 4) 344827 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@| [4, 8) 244673 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ | [8, 16) 54022 |@@@@@@@@ | [16, 32) 222134 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ | [32, 64) 232042 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ | [64, 128) 4219 | | [128, 256) 188 | | After: @udp_enqueue_us[398]: [0] 5608855 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@| [1] 1111277 |@@@@@@@@@@ | [2, 4) 501439 |@@@@ | [4, 8) 102921 | | [8, 16) 29895 | | [16, 32) 43500 | | [32, 64) 31552 | | [64, 128) 979 | | [128, 256) 13 | | Note that the remaining bottleneck for this platform is in udp_drops_inc() because we limited struct numa_drop_counters to only two nodes so far. Signed-off-by: Eric Dumazet Acked-by: Paolo Abeni Reviewed-by: Willem de Bruijn Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250922104240.2182559-1-edumazet@google.com Signed-off-by: Jakub Kicinski

net: ipv4: make udp_v4_early_demux explicitly return drop reason

2025-09-20T00:35:51+00:00

udp_v4_early_demux already returns drop reasons as it either returns 0 or ip_mc_validate_source, which itself returns drop reasons. Its return value is also already used as a drop reason itself. Makes this explicit by making it return drop reasons. Signed-off-by: Antoine Tenart Reviewed-by: David Ahern Link: https://patch.msgid.link/20250915091958.15382-2-atenart@kernel.org Signed-off-by: Jakub Kicinski

udp: use skb_attempt_defer_free()

2025-09-18T08:17:10+00:00

Move skb freeing from udp recvmsg() path to the cpu which allocated/received it, as TCP did in linux-5.17. This increases max thoughput by 20% to 30%, depending on number of BH producers. Signed-off-by: Eric Dumazet Acked-by: Willem de Bruijn Reviewed-by: David Ahern Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250916160951.541279-11-edumazet@google.com Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni

udp: make busylock per socket

2025-09-18T08:17:10+00:00

While having all spinlocks packed into an array was a space saver, this also caused NUMA imbalance and hash collisions. UDPv6 socket size becomes 1600 after this patch. Signed-off-by: Eric Dumazet Reviewed-by: Willem de Bruijn Reviewed-by: David Ahern Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250916160951.541279-10-edumazet@google.com Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni

udp: add udp_drops_inc() helper

2025-09-18T08:17:10+00:00

Generic sk_drops_inc() reads sk->sk_drop_counters. We know the precise location for UDP sockets. Move sk_drop_counters out of sock_read_rxtx so that sock_write_rxtx starts at a cache line boundary. Signed-off-by: Eric Dumazet Reviewed-by: Willem de Bruijn Reviewed-by: David Ahern Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250916160951.541279-9-edumazet@google.com Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni