linux/net/ipv4
Joanne Koong d5a42de8bd net: Add a second bind table hashed by port and address
We currently have one tcp bind table (bhash) which hashes by port
number only. In the socket bind path, we check for bind conflicts by
traversing the specified port's inet_bind2_bucket while holding the
bucket's spinlock (see inet_csk_get_port() and inet_csk_bind_conflict()).

In instances where there are tons of sockets hashed to the same port
at different addresses, checking for a bind conflict is time-intensive
and can cause softirq cpu lockups, as well as stops new tcp connections
since __inet_inherit_port() also contests for the spinlock.

This patch proposes adding a second bind table, bhash2, that hashes by
port and ip address. Searching the bhash2 table leads to significantly
faster conflict resolution and less time holding the spinlock.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-05-20 18:16:24 -07:00
..
bpfilter
netfilter Merge git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next 2022-05-16 10:10:37 +01:00
af_inet.c ipv4: Avoid using RTO_ONLINK with ip_route_connect(). 2022-04-22 13:06:03 +01:00
ah4.c
arp.c arp: fix unused variable warnning when CONFIG_PROC_FS=n 2022-04-25 11:50:18 +01:00
bpf_tcp_ca.c bpf: reject program if a __user tagged memory accessed in kernel way 2022-01-27 12:03:46 -08:00
cipso_ipv4.c
datagram.c ipv4: Avoid using RTO_ONLINK with ip_route_connect(). 2022-04-22 13:06:03 +01:00
devinet.c memcg: accounting for objects allocated for new netdevice 2022-05-04 19:16:46 -07:00
esp4.c xfrm: free not used XFRM_ESP_NO_TRAILER flag 2022-05-06 08:24:20 +02:00
esp4_offload.c net: Fix esp GSO on inter address family tunnels. 2022-03-07 13:14:04 +01:00
fib_frontend.c ipv4: remove unnecessary type castings 2022-04-30 15:12:58 +01:00
fib_lookup.h Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2022-02-17 11:44:20 -08:00
fib_notifier.c
fib_rules.c ipv4: remove unnecessary type castings 2022-04-30 15:12:58 +01:00
fib_semantics.c ipv4: Use dscp_t in struct fib_rt_info 2022-04-11 17:37:50 -07:00
fib_trie.c ipv4: remove unnecessary type castings 2022-04-30 15:12:58 +01:00
fou.c fou: Remove XRFM from NET_FOU Kconfig 2022-04-12 14:56:33 -07:00
gre_demux.c
gre_offload.c gro: remove rcu_read_lock/rcu_read_unlock from gro_complete handlers 2021-11-24 17:21:42 -08:00
icmp.c ipv4: remove unnecessary type castings 2022-04-30 15:12:58 +01:00
igmp.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2022-05-05 13:03:18 -07:00
inet_connection_sock.c net: Add a second bind table hashed by port and address 2022-05-20 18:16:24 -07:00
inet_diag.c net: inet: Retire port only listening_hash 2022-05-12 16:52:18 -07:00
inet_fragment.c ipv4: remove unnecessary type castings 2022-04-30 15:12:58 +01:00
inet_hashtables.c net: Add a second bind table hashed by port and address 2022-05-20 18:16:24 -07:00
inet_timewait_sock.c Revert "tcp/dccp: get rid of inet_twsk_purge()" 2022-05-13 12:24:12 +01:00
inetpeer.c
ip_forward.c net: ip: add skb drop reasons to ip forwarding 2022-04-13 13:09:57 +01:00
ip_fragment.c net: ip: Handle delivery_time in ip defrag 2022-03-03 14:38:48 +00:00
ip_gre.c ip_gre: Make GRE and GRETAP devices always NETIF_F_LLTX 2022-05-02 10:30:33 +02:00
ip_input.c net-core: rx_otherhost_dropped to core_stats 2022-04-07 20:32:49 -07:00
ip_options.c ipv4: drop fragmentation code from ip_options_build() 2022-01-29 17:53:07 +00:00
ip_output.c net: Set skb->mono_delivery_time and clear it after sch_handle_ingress() 2022-03-03 14:38:48 +00:00
ip_sockglue.c ipv4: Exposing __ip_sock_set_tos() in ip.h 2021-11-20 14:11:00 +00:00
ip_tunnel.c net: Handle l3mdev in ip_tunnel_init_flow 2022-04-15 14:27:30 -07:00
ip_tunnel_core.c
ip_vti.c
ipcomp.c
ipconfig.c
ipip.c
ipmr.c ipv4: remove unnecessary type castings 2022-04-30 15:12:58 +01:00
ipmr_base.c
Kconfig fou: Remove XRFM from NET_FOU Kconfig 2022-04-12 14:56:33 -07:00
Makefile
metrics.c
netfilter.c netfilter: Use l3mdev flow key when re-routing mangled packets 2022-05-16 13:03:29 +02:00
netlink.c
nexthop.c nexthop: change nexthop_net_exit() to nexthop_net_exit_batch() 2022-02-08 20:41:33 -08:00
ping.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2022-05-12 16:15:30 -07:00
proc.c tcp: allocate tcp_death_row outside of struct netns_ipv4 2022-01-26 19:00:31 -08:00
protocol.c
raw.c net: SO_RCVMARK socket option for SO_MARK with recvmsg() 2022-04-28 13:08:15 -07:00
raw_diag.c
route.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2022-05-19 11:23:59 -07:00
syncookies.c tcp: make sure treq->af_specific is initialized 2022-04-25 12:10:11 +01:00
sysctl_net_ipv4.c net: sysctl: introduce sysctl SYSCTL_THREE 2022-05-03 10:15:06 +02:00
tcp.c net: Add a second bind table hashed by port and address 2022-05-20 18:16:24 -07:00
tcp_bbr.c net: allow gso_max_size to exceed 65536 2022-05-16 10:18:55 +01:00
tcp_bic.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_bpf.c net: remove noblock parameter from recvmsg() entities 2022-04-12 15:00:25 +02:00
tcp_cdg.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_cong.c tcp: Add tracepoint for tcp_set_ca_state 2022-04-07 20:33:15 -07:00
tcp_cubic.c tcp_cubic: make hystart_ack_delay() aware of BIG TCP 2022-05-16 10:18:56 +01:00
tcp_dctcp.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_dctcp.h
tcp_diag.c
tcp_fastopen.c
tcp_highspeed.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_htcp.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_hybla.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_illinois.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_input.c tcp: improve PRR loss recovery 2022-05-19 18:49:50 -07:00
tcp_ipv4.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2022-05-19 11:23:59 -07:00
tcp_lp.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_metrics.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_minisocks.c tcp: md5: incorrect tcp_header_len for incoming connections 2022-04-22 15:05:59 -07:00
tcp_nv.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_offload.c net: move gro definitions to include/net/gro.h 2021-11-16 13:16:54 +00:00
tcp_output.c net: allow gso_max_size to exceed 65536 2022-05-16 10:18:55 +01:00
tcp_rate.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2022-04-28 13:02:01 -07:00
tcp_recovery.c tcp: use tcp_skb_sent_after() instead in RACK 2022-04-30 13:56:46 +01:00
tcp_scalable.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_timer.c
tcp_ulp.c
tcp_vegas.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_vegas.h
tcp_veno.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_westwood.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tcp_yeah.c tcp: add accessors to read/set tp->snd_cwnd 2022-04-06 12:05:41 -07:00
tunnel4.c
udp.c inet: rename INET_MATCH() 2022-05-16 10:31:06 +01:00
udp_bpf.c net: remove noblock parameter from recvmsg() entities 2022-04-12 15:00:25 +02:00
udp_diag.c
udp_impl.h net: remove noblock parameter from recvmsg() entities 2022-04-12 15:00:25 +02:00
udp_offload.c gro: remove rcu_read_lock/rcu_read_unlock from gro_complete handlers 2021-11-24 17:21:42 -08:00
udp_tunnel_core.c
udp_tunnel_nic.c udp_tunnel: Fix end of loop test in udp_tunnel_nic_unregister() 2022-02-23 12:35:00 +00:00
udp_tunnel_stub.c
udplite.c
xfrm4_input.c
xfrm4_output.c
xfrm4_policy.c net: Add l3mdev index to flow struct and avoid oif reset for port devices 2022-03-15 20:20:02 -07:00
xfrm4_protocol.c
xfrm4_state.c
xfrm4_tunnel.c