linux/net/ipv6/inet6_hashtables.c

361 lines
10 KiB
C
Raw Permalink Normal View History

// SPDX-License-Identifier: GPL-2.0-or-later
/*
* INET An implementation of the TCP/IP protocol suite for the LINUX
* operating system. INET is implemented using the BSD Socket
* interface as the means of communication with the user level.
*
* Generic INET6 transport hashtables
*
* Authors: Lotsa people, from code originally in tcp, generalised here
* by Arnaldo Carvalho de Melo <acme@mandriva.com>
*/
#include <linux/module.h>
#include <linux/random.h>
#include <net/addrconf.h>
#include <net/hotdata.h>
#include <net/inet_connection_sock.h>
#include <net/inet_hashtables.h>
#include <net/inet6_hashtables.h>
#include <net/secure_seq.h>
#include <net/ip.h>
#include <net/sock_reuseport.h>
#include <net/tcp.h>
u32 inet6_ehashfn(const struct net *net,
const struct in6_addr *laddr, const u16 lport,
const struct in6_addr *faddr, const __be16 fport)
{
u32 lhash, fhash;
net_get_random_once(&inet6_ehash_secret, sizeof(inet6_ehash_secret));
net_get_random_once(&tcp_ipv6_hash_secret, sizeof(tcp_ipv6_hash_secret));
lhash = (__force u32)laddr->s6_addr32[3];
fhash = __ipv6_addr_jhash(faddr, tcp_ipv6_hash_secret);
return __inet6_ehashfn(lhash, lport, fhash, fport,
inet6_ehash_secret + net_hash_mix(net));
}
EXPORT_SYMBOL_GPL(inet6_ehashfn);
/*
* Sockets in TCP_CLOSE state are _always_ taken out of the hash, so
* we need not check it for TCP lookups anymore, thanks Alexey. -DaveM
*
* The sockhash lock must be held as a reader here.
*/
struct sock *__inet6_lookup_established(const struct net *net,
struct inet_hashinfo *hashinfo,
const struct in6_addr *saddr,
const __be16 sport,
const struct in6_addr *daddr,
const u16 hnum,
const int dif, const int sdif)
{
struct sock *sk;
const struct hlist_nulls_node *node;
const __portpair ports = INET_COMBINED_PORTS(sport, hnum);
/* Optimize here for direct hit, only listening connections can
* have wildcards anyways.
*/
unsigned int hash = inet6_ehashfn(net, daddr, hnum, saddr, sport);
unsigned int slot = hash & hashinfo->ehash_mask;
struct inet_ehash_bucket *head = &hashinfo->ehash[slot];
begin:
sk_nulls_for_each_rcu(sk, node, &head->chain) {
net: move inet_dport/inet_num in sock_common commit 68835aba4d9b (net: optimize INET input path further) moved some fields used for tcp/udp sockets lookup in the first cache line of struct sock_common. This patch moves inet_dport/inet_num as well, filling a 32bit hole on 64 bit arches and reducing number of cache line misses in lookups. Also change INET_MATCH()/INET_TW_MATCH() to perform the ports match before addresses match, as this check is more discriminant. Remove the hash check from MATCH() macros because we dont need to re validate the hash value after taking a refcount on socket, and use likely/unlikely compiler hints, as the sk_hash/hash check makes the following conditional tests 100% predicted by cpu. Introduce skc_addrpair/skc_portpair pair values to better document the alignment requirements of the port/addr pairs used in the various MATCH() macros, and remove some casts. The namespace check can also be done at last. This slightly improves TCP/UDP lookup times. IP/TCP early demux needs inet->rx_dst_ifindex and TCP needs inet->min_ttl, lets group them together in same cache line. With help from Ben Hutchings & Joe Perches. Idea of this patch came after Ling Ma proposal to move skc_hash to the beginning of struct sock_common, and should allow him to submit a final version of his patch. My tests show an improvement doing so. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Ben Hutchings <bhutchings@solarflare.com> Cc: Joe Perches <joe@perches.com> Cc: Ling Ma <ling.ma.program@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-30 09:49:27 +00:00
if (sk->sk_hash != hash)
continue;
if (!inet6_match(net, sk, saddr, daddr, ports, dif, sdif))
continue;
if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt)))
tcp/dccp: remove twchain TCP listener refactoring, part 3 : Our goal is to hash SYN_RECV sockets into main ehash for fast lookup, and parallel SYN processing. Current inet_ehash_bucket contains two chains, one for ESTABLISH (and friend states) sockets, another for TIME_WAIT sockets only. As the hash table is sized to get at most one socket per bucket, it makes little sense to have separate twchain, as it makes the lookup slightly more complicated, and doubles hash table memory usage. If we make sure all socket types have the lookup keys at the same offsets, we can use a generic and faster lookup. It turns out TIME_WAIT and ESTABLISHED sockets already have common lookup fields for IPv4. [ INET_TW_MATCH() is no longer needed ] I'll provide a follow-up to factorize IPv6 lookup as well, to remove INET6_TW_MATCH() This way, SYN_RECV pseudo sockets will be supported the same. A new sock_gen_put() helper is added, doing either a sock_put() or inet_twsk_put() [ and will support SYN_RECV later ]. Note this helper should only be called in real slow path, when rcu lookup found a socket that was moved to another identity (freed/reused immediately), but could eventually be used in other contexts, like sock_edemux() Before patch : dmesg | grep "TCP established" TCP established hash table entries: 524288 (order: 11, 8388608 bytes) After patch : TCP established hash table entries: 524288 (order: 10, 4194304 bytes) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 07:22:02 +00:00
goto out;
if (unlikely(!inet6_match(net, sk, saddr, daddr, ports, dif, sdif))) {
sock_gen_put(sk);
goto begin;
}
goto found;
}
if (get_nulls_value(node) != slot)
goto begin;
out:
tcp/dccp: remove twchain TCP listener refactoring, part 3 : Our goal is to hash SYN_RECV sockets into main ehash for fast lookup, and parallel SYN processing. Current inet_ehash_bucket contains two chains, one for ESTABLISH (and friend states) sockets, another for TIME_WAIT sockets only. As the hash table is sized to get at most one socket per bucket, it makes little sense to have separate twchain, as it makes the lookup slightly more complicated, and doubles hash table memory usage. If we make sure all socket types have the lookup keys at the same offsets, we can use a generic and faster lookup. It turns out TIME_WAIT and ESTABLISHED sockets already have common lookup fields for IPv4. [ INET_TW_MATCH() is no longer needed ] I'll provide a follow-up to factorize IPv6 lookup as well, to remove INET6_TW_MATCH() This way, SYN_RECV pseudo sockets will be supported the same. A new sock_gen_put() helper is added, doing either a sock_put() or inet_twsk_put() [ and will support SYN_RECV later ]. Note this helper should only be called in real slow path, when rcu lookup found a socket that was moved to another identity (freed/reused immediately), but could eventually be used in other contexts, like sock_edemux() Before patch : dmesg | grep "TCP established" TCP established hash table entries: 524288 (order: 11, 8388608 bytes) After patch : TCP established hash table entries: 524288 (order: 10, 4194304 bytes) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 07:22:02 +00:00
sk = NULL;
found:
return sk;
}
EXPORT_SYMBOL(__inet6_lookup_established);
static inline int compute_score(struct sock *sk, const struct net *net,
const unsigned short hnum,
const struct in6_addr *daddr,
const int dif, const int sdif)
{
int score = -1;
if (net_eq(sock_net(sk), net) && inet_sk(sk)->inet_num == hnum &&
sk->sk_family == PF_INET6) {
if (!ipv6_addr_equal(&sk->sk_v6_rcv_saddr, daddr))
return -1;
if (!inet_sk_bound_dev_eq(net, sk->sk_bound_dev_if, dif, sdif))
return -1;
score = sk->sk_bound_dev_if ? 2 : 1;
net: annotate accesses to sk->sk_incoming_cpu This socket field can be read and written by concurrent cpus. Use READ_ONCE() and WRITE_ONCE() annotations to document this, and avoid some compiler 'optimizations'. KCSAN reported : BUG: KCSAN: data-race in tcp_v4_rcv / tcp_v4_rcv write to 0xffff88812220763c of 4 bytes by interrupt on cpu 0: sk_incoming_cpu_update include/net/sock.h:953 [inline] tcp_v4_rcv+0x1b3c/0x1bb0 net/ipv4/tcp_ipv4.c:1934 ip_protocol_deliver_rcu+0x4d/0x420 net/ipv4/ip_input.c:204 ip_local_deliver_finish+0x110/0x140 net/ipv4/ip_input.c:231 NF_HOOK include/linux/netfilter.h:305 [inline] NF_HOOK include/linux/netfilter.h:299 [inline] ip_local_deliver+0x133/0x210 net/ipv4/ip_input.c:252 dst_input include/net/dst.h:442 [inline] ip_rcv_finish+0x121/0x160 net/ipv4/ip_input.c:413 NF_HOOK include/linux/netfilter.h:305 [inline] NF_HOOK include/linux/netfilter.h:299 [inline] ip_rcv+0x18f/0x1a0 net/ipv4/ip_input.c:523 __netif_receive_skb_one_core+0xa7/0xe0 net/core/dev.c:5010 __netif_receive_skb+0x37/0xf0 net/core/dev.c:5124 process_backlog+0x1d3/0x420 net/core/dev.c:5955 napi_poll net/core/dev.c:6392 [inline] net_rx_action+0x3ae/0xa90 net/core/dev.c:6460 __do_softirq+0x115/0x33f kernel/softirq.c:292 do_softirq_own_stack+0x2a/0x40 arch/x86/entry/entry_64.S:1082 do_softirq.part.0+0x6b/0x80 kernel/softirq.c:337 do_softirq kernel/softirq.c:329 [inline] __local_bh_enable_ip+0x76/0x80 kernel/softirq.c:189 read to 0xffff88812220763c of 4 bytes by interrupt on cpu 1: sk_incoming_cpu_update include/net/sock.h:952 [inline] tcp_v4_rcv+0x181a/0x1bb0 net/ipv4/tcp_ipv4.c:1934 ip_protocol_deliver_rcu+0x4d/0x420 net/ipv4/ip_input.c:204 ip_local_deliver_finish+0x110/0x140 net/ipv4/ip_input.c:231 NF_HOOK include/linux/netfilter.h:305 [inline] NF_HOOK include/linux/netfilter.h:299 [inline] ip_local_deliver+0x133/0x210 net/ipv4/ip_input.c:252 dst_input include/net/dst.h:442 [inline] ip_rcv_finish+0x121/0x160 net/ipv4/ip_input.c:413 NF_HOOK include/linux/netfilter.h:305 [inline] NF_HOOK include/linux/netfilter.h:299 [inline] ip_rcv+0x18f/0x1a0 net/ipv4/ip_input.c:523 __netif_receive_skb_one_core+0xa7/0xe0 net/core/dev.c:5010 __netif_receive_skb+0x37/0xf0 net/core/dev.c:5124 process_backlog+0x1d3/0x420 net/core/dev.c:5955 napi_poll net/core/dev.c:6392 [inline] net_rx_action+0x3ae/0xa90 net/core/dev.c:6460 __do_softirq+0x115/0x33f kernel/softirq.c:292 run_ksoftirqd+0x46/0x60 kernel/softirq.c:603 smpboot_thread_fn+0x37d/0x4a0 kernel/smpboot.c:165 Reported by Kernel Concurrency Sanitizer on: CPU: 1 PID: 16 Comm: ksoftirqd/1 Not tainted 5.4.0-rc3+ #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot <syzkaller@googlegroups.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-10-30 20:00:04 +00:00
if (READ_ONCE(sk->sk_incoming_cpu) == raw_smp_processor_id())
score++;
}
return score;
}
/**
* inet6_lookup_reuseport() - execute reuseport logic on AF_INET6 socket if necessary.
* @net: network namespace.
* @sk: AF_INET6 socket, must be in TCP_LISTEN state for TCP or TCP_CLOSE for UDP.
* @skb: context for a potential SK_REUSEPORT program.
* @doff: header offset.
* @saddr: source address.
* @sport: source port.
* @daddr: destination address.
* @hnum: destination port in host byte order.
* @ehashfn: hash function used to generate the fallback hash.
*
* Return: NULL if sk doesn't have SO_REUSEPORT set, otherwise a pointer to
* the selected sock or an error.
*/
struct sock *inet6_lookup_reuseport(const struct net *net, struct sock *sk,
struct sk_buff *skb, int doff,
const struct in6_addr *saddr,
__be16 sport,
const struct in6_addr *daddr,
unsigned short hnum,
inet6_ehashfn_t *ehashfn)
{
struct sock *reuse_sk = NULL;
u32 phash;
if (sk->sk_reuseport) {
phash = INDIRECT_CALL_INET(ehashfn, udp6_ehashfn, inet6_ehashfn,
net, daddr, hnum, saddr, sport);
reuse_sk = reuseport_select_sock(sk, phash, skb, doff);
}
return reuse_sk;
}
EXPORT_SYMBOL_GPL(inet6_lookup_reuseport);
/* called with rcu_read_lock() */
static struct sock *inet6_lhash2_lookup(const struct net *net,
struct inet_listen_hashbucket *ilb2,
struct sk_buff *skb, int doff,
const struct in6_addr *saddr,
const __be16 sport, const struct in6_addr *daddr,
const unsigned short hnum, const int dif, const int sdif)
{
struct sock *sk, *result = NULL;
net: inet: Retire port only listening_hash The listen sk is currently stored in two hash tables, listening_hash (hashed by port) and lhash2 (hashed by port and address). After commit 0ee58dad5b06 ("net: tcp6: prefer listeners bound to an address") and commit d9fbc7f6431f ("net: tcp: prefer listeners bound to an address"), the TCP-SYN lookup fast path does not use listening_hash. The commit 05c0b35709c5 ("tcp: seq_file: Replace listening_hash with lhash2") also moved the seq_file (/proc/net/tcp) iteration usage from listening_hash to lhash2. There are still a few listening_hash usages left. One of them is inet_reuseport_add_sock() which uses the listening_hash to search a listen sk during the listen() system call. This turns out to be very slow on use cases that listen on many different VIPs at a popular port (e.g. 443). [ On top of the slowness in adding to the tail in the IPv6 case ]. The latter patch has a selftest to demonstrate this case. This patch takes this chance to move all remaining listening_hash usages to lhash2 and then retire listening_hash. Since most changes need to be done together, it is hard to cut the listening_hash to lhash2 switch into small patches. The changes in this patch is highlighted here for the review purpose. 1. Because of the listening_hash removal, lhash2 can use the sk->sk_nulls_node instead of the icsk->icsk_listen_portaddr_node. This will also keep the sk_unhashed() check to work as is after stop adding sk to listening_hash. The union is removed from inet_listen_hashbucket because only nulls_head is needed. 2. icsk->icsk_listen_portaddr_node and its helpers are removed. 3. The current lhash2 users needs to iterate with sk_nulls_node instead of icsk_listen_portaddr_node. One case is in the inet[6]_lhash2_lookup(). Another case is the seq_file iterator in tcp_ipv4.c. One thing to note is sk_nulls_next() is needed because the old inet_lhash2_for_each_icsk_continue() does a "next" first before iterating. 4. Move the remaining listening_hash usage to lhash2 inet_reuseport_add_sock() which this series is trying to improve. inet_diag.c and mptcp_diag.c are the final two remaining use cases and is moved to lhash2 now also. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-05-12 00:06:05 +00:00
struct hlist_nulls_node *node;
int score, hiscore = 0;
net: inet: Retire port only listening_hash The listen sk is currently stored in two hash tables, listening_hash (hashed by port) and lhash2 (hashed by port and address). After commit 0ee58dad5b06 ("net: tcp6: prefer listeners bound to an address") and commit d9fbc7f6431f ("net: tcp: prefer listeners bound to an address"), the TCP-SYN lookup fast path does not use listening_hash. The commit 05c0b35709c5 ("tcp: seq_file: Replace listening_hash with lhash2") also moved the seq_file (/proc/net/tcp) iteration usage from listening_hash to lhash2. There are still a few listening_hash usages left. One of them is inet_reuseport_add_sock() which uses the listening_hash to search a listen sk during the listen() system call. This turns out to be very slow on use cases that listen on many different VIPs at a popular port (e.g. 443). [ On top of the slowness in adding to the tail in the IPv6 case ]. The latter patch has a selftest to demonstrate this case. This patch takes this chance to move all remaining listening_hash usages to lhash2 and then retire listening_hash. Since most changes need to be done together, it is hard to cut the listening_hash to lhash2 switch into small patches. The changes in this patch is highlighted here for the review purpose. 1. Because of the listening_hash removal, lhash2 can use the sk->sk_nulls_node instead of the icsk->icsk_listen_portaddr_node. This will also keep the sk_unhashed() check to work as is after stop adding sk to listening_hash. The union is removed from inet_listen_hashbucket because only nulls_head is needed. 2. icsk->icsk_listen_portaddr_node and its helpers are removed. 3. The current lhash2 users needs to iterate with sk_nulls_node instead of icsk_listen_portaddr_node. One case is in the inet[6]_lhash2_lookup(). Another case is the seq_file iterator in tcp_ipv4.c. One thing to note is sk_nulls_next() is needed because the old inet_lhash2_for_each_icsk_continue() does a "next" first before iterating. 4. Move the remaining listening_hash usage to lhash2 inet_reuseport_add_sock() which this series is trying to improve. inet_diag.c and mptcp_diag.c are the final two remaining use cases and is moved to lhash2 now also. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-05-12 00:06:05 +00:00
sk_nulls_for_each_rcu(sk, node, &ilb2->nulls_head) {
score = compute_score(sk, net, hnum, daddr, dif, sdif);
if (score > hiscore) {
result = inet6_lookup_reuseport(net, sk, skb, doff,
saddr, sport, daddr, hnum, inet6_ehashfn);
if (result)
return result;
result = sk;
hiscore = score;
}
}
return result;
}
struct sock *inet6_lookup_run_sk_lookup(const struct net *net,
int protocol,
struct sk_buff *skb, int doff,
const struct in6_addr *saddr,
const __be16 sport,
const struct in6_addr *daddr,
const u16 hnum, const int dif,
inet6_ehashfn_t *ehashfn)
{
struct sock *sk, *reuse_sk;
bool no_reuseport;
no_reuseport = bpf_sk_lookup_run_v6(net, protocol, saddr, sport,
daddr, hnum, dif, &sk);
if (no_reuseport || IS_ERR_OR_NULL(sk))
return sk;
reuse_sk = inet6_lookup_reuseport(net, sk, skb, doff,
saddr, sport, daddr, hnum, ehashfn);
if (reuse_sk)
sk = reuse_sk;
return sk;
}
EXPORT_SYMBOL_GPL(inet6_lookup_run_sk_lookup);
struct sock *inet6_lookup_listener(const struct net *net,
struct inet_hashinfo *hashinfo,
struct sk_buff *skb, int doff,
const struct in6_addr *saddr,
const __be16 sport, const struct in6_addr *daddr,
const unsigned short hnum, const int dif, const int sdif)
{
struct inet_listen_hashbucket *ilb2;
struct sock *result = NULL;
unsigned int hash2;
/* Lookup redirect from BPF */
if (static_branch_unlikely(&bpf_sk_lookup_enabled) &&
hashinfo == net->ipv4.tcp_death_row.hashinfo) {
result = inet6_lookup_run_sk_lookup(net, IPPROTO_TCP, skb, doff,
saddr, sport, daddr, hnum, dif,
inet6_ehashfn);
if (result)
goto done;
}
hash2 = ipv6_portaddr_hash(net, daddr, hnum);
ilb2 = inet_lhash2_bucket(hashinfo, hash2);
result = inet6_lhash2_lookup(net, ilb2, skb, doff,
saddr, sport, daddr, hnum,
dif, sdif);
if (result)
bpf: Enable BPF_PROG_TYPE_SK_REUSEPORT bpf prog in reuseport selection This patch allows a BPF_PROG_TYPE_SK_REUSEPORT bpf prog to select a SO_REUSEPORT sk from a BPF_MAP_TYPE_REUSEPORT_ARRAY introduced in the earlier patch. "bpf_run_sk_reuseport()" will return -ECONNREFUSED when the BPF_PROG_TYPE_SK_REUSEPORT prog returns SK_DROP. The callers, in inet[6]_hashtable.c and ipv[46]/udp.c, are modified to handle this case and return NULL immediately instead of continuing the sk search from its hashtable. It re-uses the existing SO_ATTACH_REUSEPORT_EBPF setsockopt to attach BPF_PROG_TYPE_SK_REUSEPORT. The "sk_reuseport_attach_bpf()" will check if the attaching bpf prog is in the new SK_REUSEPORT or the existing SOCKET_FILTER type and then check different things accordingly. One level of "__reuseport_attach_prog()" call is removed. The "sk_unhashed() && ..." and "sk->sk_reuseport_cb" tests are pushed back to "reuseport_attach_prog()" in sock_reuseport.c. sock_reuseport.c seems to have more knowledge on those test requirements than filter.c. In "reuseport_attach_prog()", after new_prog is attached to reuse->prog, the old_prog (if any) is also directly freed instead of returning the old_prog to the caller and asking the caller to free. The sysctl_optmem_max check is moved back to the "sk_reuseport_attach_filter()" and "sk_reuseport_attach_bpf()". As of other bpf prog types, the new BPF_PROG_TYPE_SK_REUSEPORT is only bounded by the usual "bpf_prog_charge_memlock()" during load time instead of bounded by both bpf_prog_charge_memlock and sysctl_optmem_max. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-08-08 08:01:26 +00:00
goto done;
/* Lookup lhash2 with in6addr_any */
hash2 = ipv6_portaddr_hash(net, &in6addr_any, hnum);
ilb2 = inet_lhash2_bucket(hashinfo, hash2);
bpf: Enable BPF_PROG_TYPE_SK_REUSEPORT bpf prog in reuseport selection This patch allows a BPF_PROG_TYPE_SK_REUSEPORT bpf prog to select a SO_REUSEPORT sk from a BPF_MAP_TYPE_REUSEPORT_ARRAY introduced in the earlier patch. "bpf_run_sk_reuseport()" will return -ECONNREFUSED when the BPF_PROG_TYPE_SK_REUSEPORT prog returns SK_DROP. The callers, in inet[6]_hashtable.c and ipv[46]/udp.c, are modified to handle this case and return NULL immediately instead of continuing the sk search from its hashtable. It re-uses the existing SO_ATTACH_REUSEPORT_EBPF setsockopt to attach BPF_PROG_TYPE_SK_REUSEPORT. The "sk_reuseport_attach_bpf()" will check if the attaching bpf prog is in the new SK_REUSEPORT or the existing SOCKET_FILTER type and then check different things accordingly. One level of "__reuseport_attach_prog()" call is removed. The "sk_unhashed() && ..." and "sk->sk_reuseport_cb" tests are pushed back to "reuseport_attach_prog()" in sock_reuseport.c. sock_reuseport.c seems to have more knowledge on those test requirements than filter.c. In "reuseport_attach_prog()", after new_prog is attached to reuse->prog, the old_prog (if any) is also directly freed instead of returning the old_prog to the caller and asking the caller to free. The sysctl_optmem_max check is moved back to the "sk_reuseport_attach_filter()" and "sk_reuseport_attach_bpf()". As of other bpf prog types, the new BPF_PROG_TYPE_SK_REUSEPORT is only bounded by the usual "bpf_prog_charge_memlock()" during load time instead of bounded by both bpf_prog_charge_memlock and sysctl_optmem_max. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-08-08 08:01:26 +00:00
result = inet6_lhash2_lookup(net, ilb2, skb, doff,
saddr, sport, &in6addr_any, hnum,
bpf: Enable BPF_PROG_TYPE_SK_REUSEPORT bpf prog in reuseport selection This patch allows a BPF_PROG_TYPE_SK_REUSEPORT bpf prog to select a SO_REUSEPORT sk from a BPF_MAP_TYPE_REUSEPORT_ARRAY introduced in the earlier patch. "bpf_run_sk_reuseport()" will return -ECONNREFUSED when the BPF_PROG_TYPE_SK_REUSEPORT prog returns SK_DROP. The callers, in inet[6]_hashtable.c and ipv[46]/udp.c, are modified to handle this case and return NULL immediately instead of continuing the sk search from its hashtable. It re-uses the existing SO_ATTACH_REUSEPORT_EBPF setsockopt to attach BPF_PROG_TYPE_SK_REUSEPORT. The "sk_reuseport_attach_bpf()" will check if the attaching bpf prog is in the new SK_REUSEPORT or the existing SOCKET_FILTER type and then check different things accordingly. One level of "__reuseport_attach_prog()" call is removed. The "sk_unhashed() && ..." and "sk->sk_reuseport_cb" tests are pushed back to "reuseport_attach_prog()" in sock_reuseport.c. sock_reuseport.c seems to have more knowledge on those test requirements than filter.c. In "reuseport_attach_prog()", after new_prog is attached to reuse->prog, the old_prog (if any) is also directly freed instead of returning the old_prog to the caller and asking the caller to free. The sysctl_optmem_max check is moved back to the "sk_reuseport_attach_filter()" and "sk_reuseport_attach_bpf()". As of other bpf prog types, the new BPF_PROG_TYPE_SK_REUSEPORT is only bounded by the usual "bpf_prog_charge_memlock()" during load time instead of bounded by both bpf_prog_charge_memlock and sysctl_optmem_max. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-08-08 08:01:26 +00:00
dif, sdif);
done:
if (IS_ERR(result))
bpf: Enable BPF_PROG_TYPE_SK_REUSEPORT bpf prog in reuseport selection This patch allows a BPF_PROG_TYPE_SK_REUSEPORT bpf prog to select a SO_REUSEPORT sk from a BPF_MAP_TYPE_REUSEPORT_ARRAY introduced in the earlier patch. "bpf_run_sk_reuseport()" will return -ECONNREFUSED when the BPF_PROG_TYPE_SK_REUSEPORT prog returns SK_DROP. The callers, in inet[6]_hashtable.c and ipv[46]/udp.c, are modified to handle this case and return NULL immediately instead of continuing the sk search from its hashtable. It re-uses the existing SO_ATTACH_REUSEPORT_EBPF setsockopt to attach BPF_PROG_TYPE_SK_REUSEPORT. The "sk_reuseport_attach_bpf()" will check if the attaching bpf prog is in the new SK_REUSEPORT or the existing SOCKET_FILTER type and then check different things accordingly. One level of "__reuseport_attach_prog()" call is removed. The "sk_unhashed() && ..." and "sk->sk_reuseport_cb" tests are pushed back to "reuseport_attach_prog()" in sock_reuseport.c. sock_reuseport.c seems to have more knowledge on those test requirements than filter.c. In "reuseport_attach_prog()", after new_prog is attached to reuse->prog, the old_prog (if any) is also directly freed instead of returning the old_prog to the caller and asking the caller to free. The sysctl_optmem_max check is moved back to the "sk_reuseport_attach_filter()" and "sk_reuseport_attach_bpf()". As of other bpf prog types, the new BPF_PROG_TYPE_SK_REUSEPORT is only bounded by the usual "bpf_prog_charge_memlock()" during load time instead of bounded by both bpf_prog_charge_memlock and sysctl_optmem_max. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-08-08 08:01:26 +00:00
return NULL;
return result;
}
EXPORT_SYMBOL_GPL(inet6_lookup_listener);
struct sock *inet6_lookup(const struct net *net,
struct inet_hashinfo *hashinfo,
struct sk_buff *skb, int doff,
const struct in6_addr *saddr, const __be16 sport,
const struct in6_addr *daddr, const __be16 dport,
const int dif)
{
struct sock *sk;
bool refcounted;
sk = __inet6_lookup(net, hashinfo, skb, doff, saddr, sport, daddr,
ntohs(dport), dif, 0, &refcounted);
if (sk && !refcounted && !refcount_inc_not_zero(&sk->sk_refcnt))
sk = NULL;
return sk;
}
EXPORT_SYMBOL_GPL(inet6_lookup);
static int __inet6_check_established(struct inet_timewait_death_row *death_row,
struct sock *sk, const __u16 lport,
struct inet_timewait_sock **twp)
{
struct inet_hashinfo *hinfo = death_row->hashinfo;
struct inet_sock *inet = inet_sk(sk);
const struct in6_addr *daddr = &sk->sk_v6_rcv_saddr;
const struct in6_addr *saddr = &sk->sk_v6_daddr;
const int dif = sk->sk_bound_dev_if;
struct net *net = sock_net(sk);
const int sdif = l3mdev_master_ifindex_by_index(net, dif);
const __portpair ports = INET_COMBINED_PORTS(inet->inet_dport, lport);
const unsigned int hash = inet6_ehashfn(net, daddr, lport, saddr,
inet->inet_dport);
struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash);
spinlock_t *lock = inet_ehash_lockp(hinfo, hash);
struct sock *sk2;
const struct hlist_nulls_node *node;
tcp/dccp: remove twchain TCP listener refactoring, part 3 : Our goal is to hash SYN_RECV sockets into main ehash for fast lookup, and parallel SYN processing. Current inet_ehash_bucket contains two chains, one for ESTABLISH (and friend states) sockets, another for TIME_WAIT sockets only. As the hash table is sized to get at most one socket per bucket, it makes little sense to have separate twchain, as it makes the lookup slightly more complicated, and doubles hash table memory usage. If we make sure all socket types have the lookup keys at the same offsets, we can use a generic and faster lookup. It turns out TIME_WAIT and ESTABLISHED sockets already have common lookup fields for IPv4. [ INET_TW_MATCH() is no longer needed ] I'll provide a follow-up to factorize IPv6 lookup as well, to remove INET6_TW_MATCH() This way, SYN_RECV pseudo sockets will be supported the same. A new sock_gen_put() helper is added, doing either a sock_put() or inet_twsk_put() [ and will support SYN_RECV later ]. Note this helper should only be called in real slow path, when rcu lookup found a socket that was moved to another identity (freed/reused immediately), but could eventually be used in other contexts, like sock_edemux() Before patch : dmesg | grep "TCP established" TCP established hash table entries: 524288 (order: 11, 8388608 bytes) After patch : TCP established hash table entries: 524288 (order: 10, 4194304 bytes) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 07:22:02 +00:00
struct inet_timewait_sock *tw = NULL;
spin_lock(lock);
tcp/dccp: remove twchain TCP listener refactoring, part 3 : Our goal is to hash SYN_RECV sockets into main ehash for fast lookup, and parallel SYN processing. Current inet_ehash_bucket contains two chains, one for ESTABLISH (and friend states) sockets, another for TIME_WAIT sockets only. As the hash table is sized to get at most one socket per bucket, it makes little sense to have separate twchain, as it makes the lookup slightly more complicated, and doubles hash table memory usage. If we make sure all socket types have the lookup keys at the same offsets, we can use a generic and faster lookup. It turns out TIME_WAIT and ESTABLISHED sockets already have common lookup fields for IPv4. [ INET_TW_MATCH() is no longer needed ] I'll provide a follow-up to factorize IPv6 lookup as well, to remove INET6_TW_MATCH() This way, SYN_RECV pseudo sockets will be supported the same. A new sock_gen_put() helper is added, doing either a sock_put() or inet_twsk_put() [ and will support SYN_RECV later ]. Note this helper should only be called in real slow path, when rcu lookup found a socket that was moved to another identity (freed/reused immediately), but could eventually be used in other contexts, like sock_edemux() Before patch : dmesg | grep "TCP established" TCP established hash table entries: 524288 (order: 11, 8388608 bytes) After patch : TCP established hash table entries: 524288 (order: 10, 4194304 bytes) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 07:22:02 +00:00
sk_nulls_for_each(sk2, node, &head->chain) {
net: move inet_dport/inet_num in sock_common commit 68835aba4d9b (net: optimize INET input path further) moved some fields used for tcp/udp sockets lookup in the first cache line of struct sock_common. This patch moves inet_dport/inet_num as well, filling a 32bit hole on 64 bit arches and reducing number of cache line misses in lookups. Also change INET_MATCH()/INET_TW_MATCH() to perform the ports match before addresses match, as this check is more discriminant. Remove the hash check from MATCH() macros because we dont need to re validate the hash value after taking a refcount on socket, and use likely/unlikely compiler hints, as the sk_hash/hash check makes the following conditional tests 100% predicted by cpu. Introduce skc_addrpair/skc_portpair pair values to better document the alignment requirements of the port/addr pairs used in the various MATCH() macros, and remove some casts. The namespace check can also be done at last. This slightly improves TCP/UDP lookup times. IP/TCP early demux needs inet->rx_dst_ifindex and TCP needs inet->min_ttl, lets group them together in same cache line. With help from Ben Hutchings & Joe Perches. Idea of this patch came after Ling Ma proposal to move skc_hash to the beginning of struct sock_common, and should allow him to submit a final version of his patch. My tests show an improvement doing so. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Ben Hutchings <bhutchings@solarflare.com> Cc: Joe Perches <joe@perches.com> Cc: Ling Ma <ling.ma.program@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-30 09:49:27 +00:00
if (sk2->sk_hash != hash)
continue;
if (likely(inet6_match(net, sk2, saddr, daddr, ports,
dif, sdif))) {
if (sk2->sk_state == TCP_TIME_WAIT) {
tcp/dccp: remove twchain TCP listener refactoring, part 3 : Our goal is to hash SYN_RECV sockets into main ehash for fast lookup, and parallel SYN processing. Current inet_ehash_bucket contains two chains, one for ESTABLISH (and friend states) sockets, another for TIME_WAIT sockets only. As the hash table is sized to get at most one socket per bucket, it makes little sense to have separate twchain, as it makes the lookup slightly more complicated, and doubles hash table memory usage. If we make sure all socket types have the lookup keys at the same offsets, we can use a generic and faster lookup. It turns out TIME_WAIT and ESTABLISHED sockets already have common lookup fields for IPv4. [ INET_TW_MATCH() is no longer needed ] I'll provide a follow-up to factorize IPv6 lookup as well, to remove INET6_TW_MATCH() This way, SYN_RECV pseudo sockets will be supported the same. A new sock_gen_put() helper is added, doing either a sock_put() or inet_twsk_put() [ and will support SYN_RECV later ]. Note this helper should only be called in real slow path, when rcu lookup found a socket that was moved to another identity (freed/reused immediately), but could eventually be used in other contexts, like sock_edemux() Before patch : dmesg | grep "TCP established" TCP established hash table entries: 524288 (order: 11, 8388608 bytes) After patch : TCP established hash table entries: 524288 (order: 10, 4194304 bytes) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 07:22:02 +00:00
tw = inet_twsk(sk2);
if (sk->sk_protocol == IPPROTO_TCP &&
tcp_twsk_unique(sk, sk2, twp))
break;
tcp/dccp: remove twchain TCP listener refactoring, part 3 : Our goal is to hash SYN_RECV sockets into main ehash for fast lookup, and parallel SYN processing. Current inet_ehash_bucket contains two chains, one for ESTABLISH (and friend states) sockets, another for TIME_WAIT sockets only. As the hash table is sized to get at most one socket per bucket, it makes little sense to have separate twchain, as it makes the lookup slightly more complicated, and doubles hash table memory usage. If we make sure all socket types have the lookup keys at the same offsets, we can use a generic and faster lookup. It turns out TIME_WAIT and ESTABLISHED sockets already have common lookup fields for IPv4. [ INET_TW_MATCH() is no longer needed ] I'll provide a follow-up to factorize IPv6 lookup as well, to remove INET6_TW_MATCH() This way, SYN_RECV pseudo sockets will be supported the same. A new sock_gen_put() helper is added, doing either a sock_put() or inet_twsk_put() [ and will support SYN_RECV later ]. Note this helper should only be called in real slow path, when rcu lookup found a socket that was moved to another identity (freed/reused immediately), but could eventually be used in other contexts, like sock_edemux() Before patch : dmesg | grep "TCP established" TCP established hash table entries: 524288 (order: 11, 8388608 bytes) After patch : TCP established hash table entries: 524288 (order: 10, 4194304 bytes) Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 07:22:02 +00:00
}
goto not_unique;
}
}
/* Must record num and sport now. Otherwise we will see
* in hash table socket with a funny identity.
*/
inet->inet_num = lport;
inet->inet_sport = htons(lport);
sk->sk_hash = hash;
WARN_ON(!sk_unhashed(sk));
__sk_nulls_add_node_rcu(sk, &head->chain);
if (tw) {
sk_nulls_del_node_init_rcu((struct sock *)tw);
__NET_INC_STATS(net, LINUX_MIB_TIMEWAITRECYCLED);
}
spin_unlock(lock);
sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
if (twp) {
*twp = tw;
} else if (tw) {
/* Silly. Should hash-dance instead... */
inet_twsk_deschedule_put(tw);
}
return 0;
not_unique:
spin_unlock(lock);
return -EADDRNOTAVAIL;
}
static u64 inet6_sk_port_offset(const struct sock *sk)
{
const struct inet_sock *inet = inet_sk(sk);
return secure_ipv6_port_ephemeral(sk->sk_v6_rcv_saddr.s6_addr32,
sk->sk_v6_daddr.s6_addr32,
inet->inet_dport);
}
int inet6_hash_connect(struct inet_timewait_death_row *death_row,
struct sock *sk)
{
u64 port_offset = 0;
if (!inet_sk(sk)->inet_num)
port_offset = inet6_sk_port_offset(sk);
return __inet_hash_connect(death_row, sk, port_offset,
__inet6_check_established);
}
EXPORT_SYMBOL_GPL(inet6_hash_connect);
int inet6_hash(struct sock *sk)
{
int err = 0;
tcp: Don't acquire inet_listen_hashbucket::lock with disabled BH. Commit 9652dc2eb9e40 ("tcp: relax listening_hash operations") removed the need to disable bottom half while acquiring listening_hash.lock. There are still two callers left which disable bottom half before the lock is acquired. On PREEMPT_RT the softirqs are preemptible and local_bh_disable() acts as a lock to ensure that resources, that are protected by disabling bottom halves, remain protected. This leads to a circular locking dependency if the lock acquired with disabled bottom halves is also acquired with enabled bottom halves followed by disabling bottom halves. This is the reverse locking order. It has been observed with inet_listen_hashbucket::lock: local_bh_disable() + spin_lock(&ilb->lock): inet_listen() inet_csk_listen_start() sk->sk_prot->hash() := inet_hash() local_bh_disable() __inet_hash() spin_lock(&ilb->lock); acquire(&ilb->lock); Reverse order: spin_lock(&ilb2->lock) + local_bh_disable(): tcp_seq_next() listening_get_next() spin_lock(&ilb2->lock); acquire(&ilb2->lock); tcp4_seq_show() get_tcp4_sock() sock_i_ino() read_lock_bh(&sk->sk_callback_lock); acquire(softirq_ctrl) // <---- whoops acquire(&sk->sk_callback_lock) Drop local_bh_disable() around __inet_hash() which acquires listening_hash->lock. Split inet_unhash() and acquire the listen_hashbucket lock without disabling bottom halves; the inet_ehash lock with disabled bottom halves. Reported-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Link: https://lkml.kernel.org/r/12d6f9879a97cd56c09fb53dee343cbb14f7f1f7.camel@gmx.de Link: https://lkml.kernel.org/r/X9CheYjuXWc75Spa@hirez.programming.kicks-ass.net Link: https://lore.kernel.org/r/YgQOebeZ10eNx1W6@linutronix.de Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-02-09 18:56:57 +00:00
if (sk->sk_state != TCP_CLOSE)
err = __inet_hash(sk, NULL);
return err;
}
EXPORT_SYMBOL_GPL(inet6_hash);