linux/net/ipv4
Kuniyuki Iwashima 40276640be Revert "tcp: avoid the lookup process failing to get sk in ehash table"
[ Upstream commit 81b3ade5d2 ]

This reverts commit 3f4ca5fafc.

Commit 3f4ca5fafc ("tcp: avoid the lookup process failing to get sk in
ehash table") reversed the order in how a socket is inserted into ehash
to fix an issue that ehash-lookup could fail when reqsk/full sk/twsk are
swapped.  However, it introduced another lookup failure.

The full socket in ehash is allocated from a slab with SLAB_TYPESAFE_BY_RCU
and does not have SOCK_RCU_FREE, so the socket could be reused even while
it is being referenced on another CPU doing RCU lookup.

Let's say a socket is reused and inserted into the same hash bucket during
lookup.  After the blamed commit, a new socket is inserted at the end of
the list.  If that happens, we will skip sockets placed after the previous
position of the reused socket, resulting in ehash lookup failure.

As described in Documentation/RCU/rculist_nulls.rst, we should insert a
new socket at the head of the list to avoid such an issue.

This issue, the swap-lookup-failure, and another variant reported in [0]
can all be handled properly by adding a locked ehash lookup suggested by
Eric Dumazet [1].

However, this issue could occur for every packet, thus more likely than
the other two races, so let's revert the change for now.

Link: https://lore.kernel.org/netdev/20230606064306.9192-1-duanmuquan@baidu.com/ [0]
Link: https://lore.kernel.org/netdev/CANn89iK8snOz8TYOhhwfimC7ykYA78GA3Nyv8x06SZYa1nKdyA@mail.gmail.com/ [1]
Fixes: 3f4ca5fafc ("tcp: avoid the lookup process failing to get sk in ehash table")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20230717215918.15723-1-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
2023-07-27 08:47:02 +02:00
..
bpfilter
netfilter netfilter: tproxy: fix deadlock due to missing BH disable 2023-03-17 08:48:55 +01:00
af_inet.c tcp: deny tcp_disconnect() when threads are waiting 2023-06-09 10:32:17 +02:00
ah4.c Networking changes for 5.14. 2021-06-30 15:51:09 -07:00
arp.c ipv4: Invalidate neighbour for broadcast address upon address addition 2022-04-13 20:59:05 +02:00
bpf_tcp_ca.c bpf: Forbid bpf_ktime_get_coarse_ns and bpf_timer_* in tracing progs 2021-11-25 09:49:07 +01:00
cipso_ipv4.c cipso: Fix data-races around sysctl. 2022-07-21 21:24:21 +02:00
datagram.c udp: Update reuse->has_conns under reuseport_lock. 2022-10-29 10:12:56 +02:00
devinet.c net: Fix data-races around sysctl_devconf_inherit_init_net. 2022-08-31 17:16:44 +02:00
esp4_offload.c xfrm: Linearize the skb after offloading if needed. 2023-06-28 10:29:46 +02:00
esp4.c net: ipv4: Use kfree_sensitive instead of kfree 2023-07-27 08:47:01 +02:00
fib_frontend.c ipv4: Fix incorrect table ID in IOCTL path 2023-03-22 13:31:28 +01:00
fib_lookup.h ipv4: fix data races in fib_alias_hw_flags_set 2022-02-23 12:03:10 +01:00
fib_notifier.c
fib_rules.c ipv4: convert fib_num_tclassid_users to atomic_t 2021-12-08 09:04:49 +01:00
fib_semantics.c ipv4: prevent potential spectre v1 gadget in fib_metrics_match() 2023-02-01 08:27:27 +01:00
fib_trie.c ipv4: Fix error return code in fib_table_insert() 2022-12-02 17:41:07 +01:00
fou.c fou: remove sparse errors 2021-08-31 12:03:33 +01:00
gre_demux.c net: Remove the member netns_ok 2021-05-17 15:29:35 -07:00
gre_offload.c
icmp.c icmp: guard against too small mtu 2023-04-13 16:48:18 +02:00
igmp.c igmp: Fix data-races around sysctl_igmp_qrv. 2022-08-03 12:03:48 +02:00
inet_connection_sock.c tcp: deny tcp_disconnect() when threads are waiting 2023-06-09 10:32:17 +02:00
inet_diag.c inet_diag: fix kernel-infoleak for UDP sockets 2021-12-22 09:32:40 +01:00
inet_fragment.c inet: frags: annotate races around fqdir->dead and fqdir->high_thresh 2022-01-27 11:05:35 +01:00
inet_hashtables.c Revert "tcp: avoid the lookup process failing to get sk in ehash table" 2023-07-27 08:47:02 +02:00
inet_timewait_sock.c Revert "tcp: avoid the lookup process failing to get sk in ehash table" 2023-07-27 08:47:02 +02:00
inetpeer.c inetpeer: Fix data-races around sysctl. 2022-07-21 21:24:21 +02:00
ip_forward.c ip: Fix data-races around sysctl_ip_fwd_update_priority. 2022-07-29 17:25:13 +02:00
ip_fragment.c inet: frags: annotate races around fqdir->dead and fqdir->high_thresh 2022-01-27 11:05:35 +01:00
ip_gre.c erspan: do not use skb_mac_header() in ndo_start_xmit() 2023-03-30 12:47:48 +02:00
ip_input.c xfrm: fix "disable_policy" on ipv4 early demux 2022-12-02 17:41:02 +01:00
ip_options.c
ip_output.c ipv4: Fix potential uninit variable access bug in __ip_make_skb() 2023-05-11 23:00:31 +09:00
ip_sockglue.c ipv{4,6}/raw: fix output xfrm lookup wrt protocol 2023-06-05 09:21:26 +02:00
ip_tunnel_core.c tunnels: do not assume mac header is set in skb_tunnel_check_pmtu() 2022-07-07 17:53:29 +02:00
ip_tunnel.c net: tunnels: annotate lockless accesses to dev->needed_headroom 2023-03-22 13:31:26 +01:00
ip_vti.c ip_tunnel: use ndo_siocdevprivate 2021-07-27 20:11:44 +01:00
ipcomp.c Networking changes for 5.14. 2021-06-30 15:51:09 -07:00
ipconfig.c net: ipconfig: Don't override command-line hostnames or domains 2021-06-02 13:27:03 -07:00
ipip.c ip_tunnel: use ndo_siocdevprivate 2021-07-27 20:11:44 +01:00
ipmr_base.c
ipmr.c ipmr,ip6mr: acquire RTNL before calling ip[6]mr_free_table() on failure path 2022-02-16 12:56:29 +01:00
Kconfig tcp: configurable source port perturb table size 2022-12-02 17:41:11 +01:00
Makefile bpf: Clean up sockmap related Kconfigs 2021-02-26 12:28:03 -08:00
metrics.c ipv4: prevent potential spectre v1 gadget in ip_metrics_convert() 2023-02-01 08:27:27 +01:00
netfilter.c netfilter: Dissect flow after packet mangling 2021-04-18 22:04:16 +02:00
netlink.c
nexthop.c nh: fix scope used to find saddr when adding non gw nh 2022-11-03 23:59:19 +09:00
ping.c ping: fix address binding wrt vrf 2022-05-18 10:26:57 +02:00
proc.c ip: Fix data-races around sysctl_ip_default_ttl. 2022-07-29 17:25:09 +02:00
protocol.c net: Remove the member netns_ok 2021-05-17 15:29:35 -07:00
raw_diag.c net: Use nlmsg_unicast() instead of netlink_unicast() 2021-07-13 09:28:29 -07:00
raw.c ipv{4,6}/raw: fix output xfrm lookup wrt protocol 2023-06-05 09:21:26 +02:00
route.c ipv4: Fix data-races around sysctl_fib_multipath_hash_fields. 2022-07-29 17:25:21 +02:00
syncookies.c mptcp: remove MPTCP 'ifdef' in TCP SYN cookies 2023-01-12 11:58:52 +01:00
sysctl_net_ipv4.c tcp: restrict net.ipv4.tcp_app_win 2023-04-20 12:13:53 +02:00
tcp_bbr.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_bic.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_bpf.c net: deal with most data-races in sk_wait_event() 2023-05-24 17:36:42 +01:00
tcp_cdg.c tcp: cdg: allow tcp_cdg_release() to be called multiple times 2022-11-26 09:24:50 +01:00
tcp_cong.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_cubic.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_dctcp.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_dctcp.h
tcp_diag.c
tcp_fastopen.c tcp: Fix data-races around sysctl_tcp_fastopen_blackhole_timeout. 2022-07-29 17:25:19 +02:00
tcp_highspeed.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_htcp.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_hybla.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_illinois.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_input.c tcp: annotate data races in __tcp_oow_rate_limited() 2023-07-23 13:47:29 +02:00
tcp_ipv4.c tcp: annotate data-races around tcp_rsk(req)->ts_recent 2023-07-27 08:47:01 +02:00
tcp_lp.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_metrics.c tcp: Fix data-races around sysctl_tcp_no_ssthresh_metrics_save. 2022-08-03 12:03:45 +02:00
tcp_minisocks.c tcp: annotate data-races around tcp_rsk(req)->ts_recent 2023-07-27 08:47:01 +02:00
tcp_nv.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_offload.c net, gro: Set inner transport header offset in tcp/udp GRO hook 2021-08-02 10:20:56 +01:00
tcp_output.c tcp: annotate data-races around tcp_rsk(req)->ts_recent 2023-07-27 08:47:01 +02:00
tcp_rate.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_recovery.c tcp: Fix data-races around sysctl_tcp_recovery. 2022-07-29 17:25:22 +02:00
tcp_scalable.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_timer.c tcp: Fix a data-race around sysctl_tcp_thin_linear_timeouts. 2022-07-29 17:25:22 +02:00
tcp_ulp.c net/ulp: use consistent error code when blocking ULP 2023-01-24 07:22:48 +01:00
tcp_vegas.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_vegas.h
tcp_veno.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_westwood.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp_yeah.c tcp: add accessors to read/set tp->snd_cwnd 2022-06-14 18:36:11 +02:00
tcp.c tcp: Return user_mss for TCP_MAXSEG in CLOSE/LISTEN state if user_mss set 2023-06-09 10:32:17 +02:00
tunnel4.c net: Remove the member netns_ok 2021-05-17 15:29:35 -07:00
udp_bpf.c bpf, sockmap: Fix an infinite loop error when len is 0 in tcp_bpf_recvmsg_parser() 2023-03-17 08:48:54 +01:00
udp_diag.c net: Use nlmsg_unicast() instead of netlink_unicast() 2021-07-13 09:28:29 -07:00
udp_impl.h
udp_offload.c fou: remove sparse errors 2021-08-31 12:03:33 +01:00
udp_tunnel_core.c net/tunnel: wait until all sk_user_data reader finish before releasing the sock 2022-12-31 13:14:19 +01:00
udp_tunnel_nic.c udp_tunnel: Fix end of loop test in udp_tunnel_nic_unregister() 2022-03-02 11:47:59 +01:00
udp_tunnel_stub.c
udp.c tcp/udp: Call inet6_destroy_sock() in IPv6 sk->sk_destruct(). 2023-04-26 13:51:54 +02:00
udplite.c udplite: Fix NULL pointer dereference in __sk_mem_raise_allocated(). 2023-05-30 13:55:31 +01:00
xfrm4_input.c xfrm: fix inbound ipv4/udp/esp packets to UDPv6 dualstack sockets 2023-06-28 10:29:46 +02:00
xfrm4_output.c
xfrm4_policy.c
xfrm4_protocol.c net: xfrm: unexport __init-annotated xfrm4_protocol_init() 2022-06-14 18:36:18 +02:00
xfrm4_state.c
xfrm4_tunnel.c xfrm: remove description from xfrm_type struct 2021-06-09 09:38:52 +02:00